Developments and Innovation in Carbon Dioxide (CO2) Capture and Storage Technology: Volume 2: Carbon Dioxide (CO2) Storage and Utilisation (Woodhead Publishing Series in Energy)

i Developments and innovation in carbon dioxide (CO2) capture and storage technology © Woodhead Publishing Limited, 2...

Author: M.Mercedes Maroto-Valer

102 downloads 1798 Views 21MB Size Report

This content was uploaded by our users and we assume good faith they have the permission to share this book. If you own the copyright to this book and it is wrongfully on our website, we offer a simple DMCA procedure to remove your content from our site. Start by pressing the button below!
Report copyright / DMCA form

DOWNLOAD PDF

i

Developments and innovation in carbon dioxide (CO2) capture and storage technology

© Woodhead Publishing Limited, 2010

ii

Related titles: Advanced power plant materials, design and technology (ISBN 978-1-84569-515-6) Fossil-fuel power plants generate the majority of the world’s power, but many plants are ageing and cannot meet rising global energy demands and increasingly stringent emissions criteria. To ensure security and economy of supply, utilities are building a new generation of advanced power plant with increased output and environmental performance. This book initially reviews improved plant designs for efficiency and fuel flexibility, including combinedcycle technology and utilisation of lower-grade feedstocks. Coverage extends to advanced material and component use, and the incorporation of alternative energy conversion technology, such as hydrogen production. Environmental and emissions performance issues round off the book. Oxy-fuel combustion for power generation and carbon dioxide (CO2) capture (ISBN: 978-1-84569-671-9) Oxy-fuel combustion is a power generation and carbon dioxide (CO2) capture option for advanced power plant in which fuel is burnt in an oxygen-rich environment instead of in air. This allows for a reduction in NOx and SOx emissions as well as producing a high-purity carbon dioxide (CO2) flue gas stream. This high-purity CO2 stream allows for more efficient and economical capture, processing and sequestration. This book critically reviews the fundamental principles, processes and technology of oxy-fuel combustion, including advanced concepts for its implementation. Details of these and other Woodhead Publishing books can be obtained by:  visiting our web site at www.woodheadpublishing.com  contacting Customer Services (e-mail: [email protected]; fax: +44 (0) 1223 893694; tel.: +44 (0) 1223 891358 ext. 130; address: Woodhead Publishing Limited, Abington Hall, Granta Park, Great Abington, Cambridge CB21 6AH, UK) If you would like to receive information on forthcoming titles, please send your address details to: Francis Dodds (address, tel. and fax as above; e-mail: [email protected]). Please confirm which subject areas you are interested in.


iii

Woodhead Publishing Series in Energy: Number 16

Developments and innovation in carbon dioxide (CO2) capture and storage technology Volume 2: Carbon dioxide (CO2) storage and utilisation Edited by M. Mercedes Maroto-Valer

CRC Press Boca Raton Boston New York Washington, DC

Woodhead

publishing limited

Oxford Cambridge New Delhi


iv Published by Woodhead Publishing Limited, Abington Hall, Granta Park, Great Abington, Cambridge CB21 6AH, UK www.woodheadpublishing.com Woodhead Publishing India Private Limited, G-2, Vardaan House, 7/28 Ansari Road, Daryaganj, New Delhi – 110002, India www.woodheadpublishingindia.com Published in North America by CRC Press LLC, 6000 Broken Sound Parkway, NW, Suite 300, Boca Raton, FL 33487, USA First published 2010, Woodhead Publishing Limited and CRC Press LLC © Woodhead Publishing Limited, 2010 The authors have asserted their moral rights. This book contains information obtained from authentic and highly regarded sources. Reprinted material is quoted with permission, and sources are indicated. Reasonable efforts have been made to publish reliable data and information, but the author and the publishers cannot assume responsibility for the validity of all materials. Neither the author nor the publishers, nor anyone else associated with this publication, shall be liable for any loss, damage or liability directly or indirectly caused or alleged to be caused by this book. Neither this book nor any part may be reproduced or transmitted in any form or by any means, electronic or mechanical, including photocopying, microfilming and recording, or by any information storage or retrieval system, without permission in writing from Woodhead Publishing Limited. The consent of Woodhead Publishing Limited does not extend to copying for general distribution, for promotion, for creating new works, or for resale. Specific permission must be obtained in writing from Woodhead Publishing Limited for such copying. Trademark notice: Product or corporate names may be trademarks or registered trademarks, and are used only for identification and explanation, without intent to infringe. British Library Cataloguing in Publication Data A catalogue record for this book is available from the British Library. Library of Congress Cataloging in Publication Data A catalog record for this book is available from the Library of Congress. Woodhead Publishing ISBN 978-1-84569-797-6 (book) Woodhead Publishing ISBN 978-1-84569-958-1 (e-book) CRC Press ISBN 978-1-4398-3101-4 CRC Press order number: N10186 The publishers’ policy is to use permanent paper from mills that operate a sustainable forestry policy, and which has been manufactured from pulp which is processed using acidfree and elemental chlorine-free practices. Furthermore, the publishers ensure that the text paper and cover board used have met acceptable environmental accreditation standards. Cover image © BCS Creative, 88–90 North Sherwood Street, Nottingham NG1 4EE, UK, www.bcscreative.co.uk Typeset by Replika Press Pvt Ltd, India Printed by TJ International Limited, Padstow, Cornwall, UK


v

Contents

Contributor contact details

xiii

Woodhead Publishing Series in Energy

xvii

Foreword by Lord Oxburgh

xix

1

Overview of carbon dioxide (CO2) capture and storage technology

S. Bouzalakos and M. Mercedes Maroto-Valer, University of Nottingham, UK

1.1 1.2 1.3 1.4

Introduction Greenhouse gas emissions and global climate change Carbon management and stabilisation routes Development and innovation in carbon dioxide (CO2) capture and transport technology Development and innovation in carbon dioxide (CO2) storage and utilisation technology Future trends Sources of further information and advice Acknowledgements References

1.5 1.6 1.7 1.8 1.9

1

1 2 8 11 17 19 20 22 22

Part I Geological sequestration of carbon dioxide (CO2) 2

27

Screening and selection criteria, and characterisation techniques for the geological sequestration of carbon dioxide (CO2)

S. Bachu, Alberta Innovates – Technology Futures, Canada

2.1 2.2 2.3

Introduction Screening for storage suitability and site selection Site characterisation

27 28 43


vi

Contents

2.4 2.5 2.6 2.7

Estimation of carbon dioxide (CO2) storage capacity Future trends Sources of further information and advice References

47 51 52 53

3

Carbon dioxide (CO2) sequestration in deep saline aquifers and formations

57

R. J. Rosenbauer and B. Thomas, US Geological Survey, USA

3.1 3.2 3.3 3.4 3.5 3.6 3.7 3.8 3.9

Introduction Saline aquifers Trapping mechanisms Modeling of carbon dioxide (CO2) sequestration Carbon dioxide (CO2) sequestration pilot sites Future trends Conclusions Acknowledgements References

4

Carbon dioxide (CO2) sequestration in oil and gas reservoirs and use for enhanced oil recovery (EOR)

B. Vega and A.R. Kovscek, Stanford University, USA

4.1 4.2 4.3

Introduction Carbon dioxide (CO2) enhanced recovery mechanisms Co-optimization of enhanced oil recovery (EOR) and carbon storage Future trends: geologic storage in tight rocks Summary and conclusions Sources of further information and advice References

116 118 122 123 124

5

Carbon dioxide (CO2) sequestration in unmineable coal seams and use for enhanced coalbed methane recovery (ECBM)

127

M. Mazzotti and Ronny Pini, ETH Zurich, Switzerland, G. Storti, Politecnico di Milano, Italy, and L. Burlini, ETH Zurich, Switzerland

5.1 5.2 5.3 5.4 5.5 5.6

Introduction Storage in unmineable coal seams Enhanced coalbed methane recovery Competitive adsorption Swelling and permeability Mass transfer and enhanced coalbed methane (ECBM) modeling

4.4 4.5 4.6 4.7


57 58 64 74 80 86 88 88 88 104 104 109

127 128 129 131 139 148

Contents

5.7 5.8 5.9 5.10

Field tests Future trends Sources of further information and advice References

vii

151 155 158 159

Part II Maximising and verifying carbon dioxide (CO2) storage in underground reservoirs 6

Carbon dioxide (CO2) injection design to maximise underground reservoir storage and enhanced oil recovery (EOR)

R. Qi, T.C. LaForce and M.J. Blunt, Imperial College London, UK

6.1 6.2 6.3 6.4 6.5 6.6 6.7 6.8 6.9

Carbon storage in geological formations Experiments of capillary trapping Field-scale design of storage in aquifers Storage in oilfields Discussion and conclusions Future trends Sources of further information and advice Acknowledgements References

169 172 175 179 180 181 181 182 182

7

Capillary seals for trapping carbon dioxide (CO2) in underground reservoirs

185

T.A. Meckel, The University of Texas at Austin, USA

7.1 7.2

Introduction Calculations of anticipated capillary pressures and seal capacities Monte Carlo predictions of capillary pressure within a reservoir seal Discussion Conclusions Future trends Sources of further information and advice Acknowledgements References

193 195 198 199 199 200 200

8

Measurement and monitoring technologies for verification of carbon dioxide (CO2) storage in underground reservoirs

203

R.A. Chadwick, British Geological Survey, UK

8.1

Introduction

7.3 7.4 7.5 7.6 7.7 7.8 7.9

169

185 188

203


viii

Contents

8.2 8.3

Background to storage site monitoring Detection and measurement of carbon dioxide (CO2) in the subsurface Detection and measurement of carbon dioxide (CO2) leakage to surface Conclusions and future trends Sources of further information and advice References

8.4 8.5 8.6 8.7 9

Mathematical modeling of the long-term safety of carbon dioxide (CO2) storage in underground reservoirs

K. Pruess, J. Birkholzer and Q. Zhou, Lawrence Berkeley National Laboratory, University of California, USA

9.1 9.2 9.3 9.4 9.5 9.6

Introduction Coupled processes: a challenge for mathematical models Ilustrative modeling applications Conclusions Acknowledgements References

204 207 225 233 235 235

240

240 243 244 259 261 261

Part III Terrestrial and ocean sequestration of carbon dioxide (CO2) and environmental impacts 10

Terrestrial sequestration of carbon dioxide (CO2)

271

R. Lal, The Ohio State University, USA

10.1 10.2 10.3 10.4

Introduction The terrestrial pool and its role in the global carbon cycle Emissions from agricultural versus other activities Basic principles of carbon sequestration in terrestrial ecosystems 10.5 Potential of terrestrial sequestration 10.6 Challenges of terrestrial sequestration 10.7 Extrapolation 10.8 Soil and terrestrial carbon as indicators of climate change 10.9 Conclusions 10.10 References

271 273 276

11

279 290 291 295 296 297 298

Ocean sequestration of carbon dioxide (CO2)

304

D. Golomb and S. Pennell, University of Massachusetts Lowell, USA

11.1 11.2

Introduction History of carbon dioxide (CO2) deep ocean storage proposals

304


305

Contents

Legal constraints of deep ocean storage of carbon dioxide (CO2) 11.4 Sources of anthropogenic carbon dioxide (CO2) for ocean storage 11.5 Ocean structure 11.6 Properties of carbon dioxide (CO2) 11.7 Modeling of carbon dioxide (CO2) release 11.8 Injection of carbon dioxide, water and pulverized limestone (CO2/H2O/CaCO3) emulsion 11.9 Future trends 11.10 Conclusions 11.11 Sources of further information and advice 11.12 References

ix

11.3

12

307 308 309 311 312 313 318 320 320 321 324

Environmental risks and impacts of carbon dioxide (CO2) leakage in terrestrial ecosystems

M. D. Steven, K. L. Smith and J. J. Colls, University of Nottingham, UK

12.1 12.2 12.3 12.4 12.5 12.6 12.7 12.8

Introduction Leak scenarios Impacts of terrestrial leakage Atmospheric enrichment of carbon dioxide (CO2) Leak monitoring techniques Conclusions and future trends Sources of further information and advice References

324 325 327 332 334 336 338 338

13

Environmental risks and performance assessment of carbon dioxide (CO2) leakage in marine ecosystems

344

J. Blackford, S. Widdicombe and D. Lowe, Plymouth Marine Laboratory, UK, and B. Chen, Heriot Watt University, UK

13.1 13.2

Introduction The physical and chemical behaviour of carbon dioxide (CO2) in the marine system Marine ecosystem impacts of carbon dioxide (CO2) leakage Leak monitoring options Mitigation of leaks Future trends Sources of further information and advice References

13.3 13.4 13.5 13.6 13.7 13.8


344 346 358 365 366 366 367 368

x

Contents

Part IV Advanced concepts for carbon dioxide (CO2) storage and utilisation 14

Industrial utilization of carbon dioxide (CO2)

377

Introduction The conditions for using carbon dioxide (CO2) The carbon dioxide (CO2) sources and its value Technological uses of carbon dioxide (CO2) Biological enhanced utilization Carbon dioxide (CO2) conversion as ‘storage’ of excess electric energy or intermittent energies 14.7 Production of chemicals 14.8 Conclusions and future trends 14.9 Sources of further information and advice 14.10 References

377 378 380 381 384

M. Aresta and A. Dibenedetto, University of Bari, Italy

14.1 14.2 14.3 14.4 14.5 14.6

391 398 404 405 405

15

Biofixation of carbon dioxide (CO2) by microorganisms

B. Wang and C.Q. Lan, University of Ottawa, Canada

15.1 15.2 15.3

Introduction Basic principles and methods Carbon dioxide (CO2) fixation microorganisms: chemoautotrophs and photoautotrophs Carbon dioxide (CO2) fixation by microalgae Advantages and limitations Future trends References

411 412

Mineralisation of carbon dioxide (CO2)

433

15.4 15.5 15.6 15.7 16

411

414 418 426 427 428

R. Zevenhoven and J. Fagerlund, Åbo Akademi University, Finland

16.1 Introduction 16.2 Basic principles and methods 16.3 Technologies and potential applications 16.4 Related issues 16.5 Future trends 16.6 Sources of further information and advice 16.7 References Appendix: Energy efficiency of mineral carbonation processes


433 435 438 447 451 452 453 460

Contents

17

xi

Photocatalytic reduction of carbon dioxide (CO2)

463

Jeffrey C. S. Wu, Department of Chemical Engineering, National Taiwan University, Taiwan

17.1 17.2 17.3

463 465

17.4 17.5 17.6 17.7

Introduction Fundamentals of photocatalysis Renewable energy from photocatalytic reduction of carbon dioxide (CO2) Advantages and limitations of photocatalytic processes Future trends Sources of further information and advice References

470 495 495 497 497

Index

503


xii

xiii


(* = main contact)

Chapter 1

Chapter 3

Dr Steve Bouzalakos and Professor M. Mercedes Maroto-Valer* Centre for Innovation in Carbon Capture and Storage (CICCS) Faculty of Engineering University of Nottingham University Park Nottingham NG7 2RD UK

Robert J. Rosenbauer* and Burt Thomas US Geological Survey 345 Middlefield Road MS-999 Menlo Park CA 94025 USA

Email: mercedes.maroto-valer@ nottingham.ac.uk

Chapter 4

Chapter 2 Stefan Bachu Alberta Innovates – Technology Futures 250 Karl Clark Road NW Edmonton Alberta T6N 1E4 Canada

Email: [email protected]

B. Vega and A. R. Kovscek* Energy Resources Engineering Department Stanford University 367 Panama St. room 065 Stanford CA 94305-2220 USA Email: [email protected]

Email: Stefan.Bachu@albertainnovates. ca


xiv


Chapter 5

Chapter 7

Professor Dr Marco Mazzotti* and Ronny Pini Institute of Process Engineering ETH Zurich Sonneggstrasse 3 CH-8092 Zurich Switzerland

T. A. Meckel* Gulf Coast Carbon Center Bureau of Economic Geology John A. and Katherine G. Jackson School of Geosciences The University of Texas at Austin University Station, Box X Austin TX 78713-8924 USA


Professor Dr Giuseppe Storti Dipartimento di Chimica, Materiali e Ing. Chimica ‘Giulio Natta’ Politecnico di Milano, Sede Mancinelli Via Mancinelli 7 I-20131 Milano Italy L. Burlini Geological Institute ETH Zurich Leonhardstrasse 19 CH-8092 Zurich Switzerland


Chapter 8 Dr R. A. Chadwick British Geological Survey Kingsley Dunham Centre Keyworth Nottinghamshire NG12 5GG UK Email: [email protected]

Chapter 9

Chapter 6 Ran Qi, Tara C. LaForce and Martin J. Blunt* Department of Earth Science and Engineering Imperial College London London SW7 2AZ UK

Karsten Pruess*, Jens Birkholzer and Quanlin Zhou Earth Sciences Division Lawrence Berkeley National Laboratory University of California One Cyclotron Road Berkeley CA 94720 USA

Email: [email protected] Email: [email protected]



Chapter 10

Chapter 13

R. Lal Carbon Management and Sequestration Center The Ohio State University Columbus OH 42310 USA

Jeremy Blackford*, Stephen Widdicombe and David Lowe Plymouth Marine Laboratory Prospect Place Plymouth PL1 3DH UK



Chapter 11 Dr D. Golomb* and Dr S. Pennell University of Massachusetts Lowell 1 University Avenue Lowell MA 01854 USA Email: [email protected]

Chapter 12 Michael D Steven*, Karon L Smith and Jeremy J Colls University of Nottingham University Park Nottingham NG7 2RD UK

xv

Baixin Chen School of Engineering and Physical Sciences Heriot Watt University Edinburgh Scotland EH14 4AS UK

Chapter 14 Michele Aresta* and Angela Dibenedetto CIRCC and Department of Chemistry University of Bari Campus Universitario 70126 Bari Italy Email: [email protected]

Email: michael.steven@nottingham. ac.uk


xvi


Chapter 15

Chapter 17

Bei Wang and Christopher Q. Lan* Department of Chemical and Biological Engineering University of Ottawa Ottawa Ontario K1N 6N5 Canada

Professor Jeffrey C. S. Wu Department of Chemical Engineering National Taiwan University No. 1 Section 4 Roosevelt Road Taipei 10617 Taiwan (R.O.C.) Email: [email protected]


Chapter 16 R. Zevenhoven* and J. Fagerlund Thermal and Flow Engineering Laboratory Åbo Akademi University Biskopsgatan 8 FI-20500 Åbo/Turku Finland Email: [email protected]


xvii


1 Generating power at high efficiency: Combined cycle technology for sustainable energy production Eric Jeffs 2 Advanced separation techniques for nuclear fuel reprocessing and radioactive waste treatment Edited by Kenneth L. Nash and Gregg J. Lumetta 3 Bioalcohol production: Biochemical conversion of lignocellulosic biomass Edited by K.W. Waldron 4 Understanding and mitigating ageing in nuclear power plants: Materials and operational aspects of plant life management (PLiM) Edited by Philip G. Tipping 5 Advanced power plant materials, design and technology Edited by Dermot Roddy 6 Stand-alone and hybrid wind energy systems: Technology, energy storage and applications Edited by J.K. Kaldellis 7 Biodiesel science and technology: From soil to oil Jan C.J. Bart, Natale Palmeri and Stefano Cavallaro 8 Developments and innovation in carbon dioxide (CO2) capture and storage technology Volume 1: Carbon dioxide (CO2) capture, transport and industrial applications Edited by M. Mercedes Maroto-Valer 9 Geological repository systems for safe disposal of spent nuclear fuels and radioactive waste Edited by Joonhong Ahn and Michael J. Apted 10 Wind energy systems: Optimising design and construction for safe and reliable operation Edited by John D. Sørensen and Jens N. Sørensen © Woodhead Publishing Limited, 2010

xviii


11 Solid oxide fuel cell technology: Principles, performance and operations Kevin Huang and John Bannister Goodenough 12 Handbook of advanced radioactive waste conditioning technologies Edited by Michael I. Ojovan 13 Nuclear reactor safety systems Edited by Dan Gabriel Cacuci 14 Materials for energy efficiency and thermal comfort in buildings Edited by Matthew R. Hall 15 Handbook of biofuels production: Processes and technology Edited by Rafael Luque, Juan Campelo and James Clark 16 Developments and innovation in carbon dioxide (CO2) capture and storage technology Volume 2: Carbon dioxide (CO2) storage and utilisation Edited by M. Mercedes Maroto-Valer 17 Oxy-fuel combustion for power generation and carbon dioxide (CO2) capture Edited by Ligang Zheng


xix

Foreword

In an ideal world, we wouldn’t need carbon capture and storage. But in an ideal world the inhabitants would have been quicker to spot that large-scale burning of fossil fuel could interfere with their planet’s carbon cycle and have serious consequences. By building a world economy over the last 150 years that flourished on the cheap and accessible energy that was available from coal and oil, we short-circuited the natural cycle. We have transferred from the solid Earth to the atmosphere huge quantities of carbon that would not otherwise have seen the light of day for many millions of years. We now know to our cost that the carbon cycle couples into the processes that control the Earth’s climate and that we have triggered rapid climate change. True, the Earth’s climate has always changed, but most natural change has happened sufficiently slowly for plants and animals to migrate or adapt to the new conditions. What we are doing is too fast to allow this. Although fossil fuels are the main cause of the rise in greenhouse gases in the atmosphere, there are also significant contributions from deforestation and changes in land use. Life on Earth depends on the benign greenhouse effect of our atmosphere. It provides surface temperatures that we do not find on neighbouring planets and that allow water to exist as ice, liquid and vapour. By burning fossil fuels, we increase the atmospheric concentration of CO2 which, along with other greenhouse gases, increases the greenhouse effect and increases mean global temperature including in the oceans. The atmospheric consequence of warming the oceans is the same as that of turning up the gas under a pan of gently simmering water – movements become faster and more violent. In weather terms this means more extreme weather conditions – storms, floods, droughts. Continental ice masses, particularly Antarctica and Greenland, begin to melt faster and contribute to a rise in sea level beyond that expected from thermal expansion. None of this is particularly good news and, although it is denied by a few, the evidence for the human influence on observed climate change is overwhelming. The more fossil fuel we burn the greater the rise in atmospheric CO2 and the worse the perturbation of the Earth’s climate. The problem is


xx

Foreword

that virtually all the world energy is supplied by fossil fuels, and weaning ourselves off them will take decades until we make the transition to other energy sources. The best scientific forecasts suggest that massive reductions in emissions have to be achieved by 2050 if there is to be any hope of containing damage to the climate. This is a problem that has been created largely by the developed world which owes its prosperity to the use of cheap and abundant fossil fuel. Although estimates vary, around two thirds of the ‘excess’ atmospheric CO2 is attributable to Europe and the USA. Today, however, we have new players. Developing countries whose economies are growing fast have rising energy requirements and often the cheapest available source of energy is coal, the fuel that carries the heaviest carbon cost per unit energy produced. It is urgent therefore that we find a way of managing emissions while we make the move to a low-carbon economy. The best way of managing emissions is not to produce them in the first place. This means that improved efficiency and energy conservation are vital. However something has to be done about the ‘essential’ (for the moment) emissions that we cannot avoid. These come partly from vehicles that burn liquid fossil fuels and partly from a range of so-called fixed sources such as power stations, cement factories, oil refineries and a myriad of small local sources from office blocks to domestic houses. This is where carbon capture and storage (CCS) comes in. CCS is a group of technologies that are designed to capture and immobilise emissions from the larger fixed industrial sources. This involves separating the greenhouse gases from the other gases in the industrial exhaust streams and transporting them to a suitable site where they can be contained underground for many tens of thousands of years. This may sound relatively straightforward and indeed all the component technologies needed to achieve CCS have to a greater or lesser extent been demonstrated. The problem is that these technologies have their roots elsewhere and were developed with other ends in mind. It is only relatively recently that moves have been made to harness them together to achieve CCS. There is thus enormous scope for improvement of the systems. In reality there are three different sets of technologies required: the technology for separation of greenhouse gases from the exhaust gas stream at the point source, the pipeline or other means needed to transport the separated gas, and finally the technology for storing it. Most current concepts of storage involve pumping the gas underground into geological traps that have the demonstrated ability to retain gases for many tens of thousands of years. Of the three different activities, the first is likely to be the technically most challenging and the most expensive and currently appears likely to amount to between half and two thirds of the total cost of CCS. Overall, electricity generated in coal-fired power plant with its emissions reduced by


Foreword

xxi

90% or more through CCS might be expected to cost between 30 and 50 % more than at present. Although these costs would be very unwelcome in the developed world, they would not be unbearable; in the developing world, however, they would be very difficult to accept. This means that there is an overriding urgency to reduce the cost of CCS. It is to be expected that with time costs will come down as engineers and operators gain experience, but more than slow incremental improvement is needed before CCS becomes deployed worldwide on a scale that can be expected to influence climate change. It would be wrong to assume, however, that the challenges are solely technical. Because CCS is a new activity, a suitable regulatory framework has to be developed and because CCS was never contemplated when the existing wider regulatory framework was established, there will certainly be conflicts to be resolved. The framework will have to cover the legal obligations and rights of all parties and the basis for licensing of all aspects of the operations. Work has already begun on these problems in a number of countries and within the EU. From a business point of view too, there are substantial logistical challenges. All three main elements identified above involve major capital expenditure and have lead times of at least five years. For a reasonably cost-effective system they need to come on line together. Furthermore, they are the responsibility of different consenting authorities for some of which CCS may not be their highest priority. This title addresses a number of the important challenges faced by CCS. They are formidable, but the most important step is to recognize them early and to plan ways of dealing with them. No one should pretend that CCS is a complete answer to the problem of fossil fuels and climate change but, conversely, it is unthinkable that we can manage that problem without CCS. Lord Oxburgh House of Lords Westminster London SW1, UK Email: [email protected]


xxii

1


S. B o u z a l a k o s and M. M e r c e d e s M a r o t o - V a l e r, University of Nottingham, UK Abstract: Carbon dioxide (CO2) capture and storage (CCS) is considered one of the most promising strategies to reduce CO2 emissions while enabling the continued use of fossil fuels and without compromising the security of electricity supply. This chapter first states the global CO2 emissions from power generation and points out that climate change is a serious and urgent issue. The chapter then discusses carbon management options and puts CCS technology into perspective. An account is then given on current plans to deploy large-scale CCS demonstration projects around the world and obstacles that need to be overcome to achieve the current commercialisation target of 2020. Innovation in research, development and deployment is increasingly becoming an important driver of both the mature and developing CCS technologies. This is clearly perceptible throughout the chapters of this book. The chapter closes by offering an outlook of the future trends and recommendations of sources of further information on CCS. Key words: carbon dioxide, CO2 capture and storage, CCS, climate change, fossil fuels, power generation.

1.1

Introduction

Fossil-fuel derived energy presently dominates most aspects of modern human activities and our current way of life, and is projected to remain the main energy source for the foreseeable future. However, the combustion of fossil fuels in stationary and mobile power sources produces large amounts of greenhouse gas (GHG) emissions, including carbon dioxide (CO2) which accounts for approximately 57 % carbon dioxide-equivalent (CO2-eq) of the GHG emissions from fossil fuel use (IPCC, 2007). CO2-equivalent emission is the amount of CO2 emission that would cause the same time-integrated radiative forcing, over a given time horizon, as an emitted amount of a long-lived GHG or a mixture of GHGs such as, for example, a mixture with methane (CH4) and nitrous oxide (N2O). The equivalent CO2 emission is calculated by multiplying a given emission of GHG by its Global Warming Potential (GWP) for the given time horizon, and for a mix of GHGs the CO2-eq is calculated by summing the equivalent CO2 emissions of each gas. It should be noted that while equivalent CO2 emissions is a standard and useful metric for comparing emissions of different GHGs, it does not imply the same climate change responses (IPCC, 2007). 1 © Woodhead Publishing Limited, 2010

2

Developments and innovation in CCS technology

A recent study by McKinsey & Company (2008) states that approximately 47 % (approximately 2 GtCO2 in 2007) of total European CO2 emissions could be addressed by the application of CO2 capture and storage (CCS) technologies. This includes predominantly large stationary sources, with coal power stations accounting for 52 %. On a global scale, various recent reports estimate that CCS could potentially abate between 1.4 GtCO2 (Stern, 2006) and 4 GtCO2 (IEA, 2007) by 2030. With the increasing energy demand witnessed and projected, already soaring atmospheric CO2 emissions will continue to rise. As CO2 emissions have been unequivocally linked to global warming and climate change (IPCC, 2007), mitigation measures are a matter of urgency. A range of technologies, collectively termed CO2 capture and storage (CCS), have been identified as a critical option in the portfolio of solutions available to combat climate change, allowing for the reduction of CO2 emissions while enabling the continued use of fossil fuels (IPCC, 2005). CCS involves three main steps: capture, transportation and storage. Overall, the technologies are fairly mature and plans are underway for their largescale demonstration in the near future. Technological barriers are often a monetary concern, particularly for capture technologies that account for roughly two-thirds of the total cost of CCS. Given the fact that CCS is a relatively new activity for both power plant operators and governments, a suitable regulatory framework has to be put in place to facilitate the wider deployment of the technology. The development and use of CO2 capture technology could take place within existing regulatory frameworks for power stations; however, the main issues for regulation of CCS concern activities offshore, especially geological storage, and transportation.

1.2

Greenhouse gas emissions and global climate change

The greenhouse effect is necessary to sustain life on earth, and in its absence the average temperature on the planet would be around –18 °C. The major GHGs in terms of total emissions in 2004 were CO2 from fossil fuel use (56.6 % CO2-eq), CO2 from deforestation, decay of biomass, etc. (17.3 % CO2-eq); methane (CH4) (14.3 % CO2-eq); and nitrous oxide (N2O) (7.9 % CO2-eq) (IPCC, 2007). Carbon dioxide is the most important anthropogenic greenhouse gas, even though it is not as harmful as CH4 which is produced from fossil fuel combustion in smaller amounts (IPCC, 2007). Climate scientists have no doubt that the earth’s climate will warm in response to further release of man-made greenhouse gases into the atmosphere by intensifying the greenhouse effect. However, there are uncertainties about the extent of warming that will occur and what the regional impacts of this will be, precisely. To date, the most credible estimation of future climate states



3

comes from mathematical climate models based on physical approximations. Despite uncertainties, all climate models predict substantial climate warming under greenhouse gas increases (IPCC, 2007). Taking into account the range of human activities, power stations are the largest contributor of anthropogenic CO2 emissions with levels reaching approximately 0.17 GtCO2 in the UK in 2008 (BERR, 2009). This level of CO2 emissions is further emphasised by Fig. 1.1, which shows that approximately 29 % of CO2 emissions for 2006 in EU-15 countries were attributed to power generation. According to the International Energy Outlook 2008 (EIA, 2008), the total world energy-related CO2 emissions for 2005 were estimated at 28.1 GtCO2, and are projected to increase by an average of 1.7 % per annum from 2005–2030. Concentrations of atmospheric CO2 have been increasing from approximately 280 ppmv in the pre-industrial era (Fig. 1.2a) to 389.47 ppmv, as measured in April 2009 (Fig. 1.2b). The detrimental effects of increasing CO2 levels on global climate have been well documented, and it is clear that there is a need to reduce CO2 levels (Stocker and Schmittner, 1997; Palmer and Räisänen, 2002; Karl and Trenberth, 2003; Stern, 2006). According to the latest United Nations Intergovernmental Panel on Climate Change (IPCC) report (IPCC, 2007), climate change has been proven to be unequivocally linked to human activity from observations of increases in global average air and ocean temperatures, rising global average sea levels and widespread melting of sea-ice in the Arctic (Fig. 1.3).

Petroleum refining 3 %

Cement production 2 %

Iron and steel production 2 %

Other 8 % Public electricity and heat production 29 %

Commercial/ industrial 5 %

Residential 12 %

Manufacturing industries and construction 16 %

Road transportation 23 %

1.1 EU-15 CO2 emissions by source for 2006 (total emissions = 3.46 GtCO2) (EIA, 2008).


4

320 315


310


CO2 concentration (ppmv)

305 300 295 290 285 280 275 270 1000 1060

1120

1180 1240

1300

1360 1420

1480 1540 Year (a)

1600 1660

1720

1780 1840

1900

1960

1.2 (a) Average annual atmospheric CO2 concentrations from Antarctic ice and firn from 1010–1960 (Etheridge et al., 1996) (b) Average annual atmospheric CO2 concentrations based on direct measurements at Mauna Loa Observatory from 1960–2009 (Dr Pieter Tans, NOAA/ESRL, www.esrl.noaa.gov/gmd/ccgg/trends).

395 389.47 ppmv (April 2009)

390 385 380


CO2 concentration (ppmv)

375 370 365 360 355 350 345 340 335 330 325 320 315 310 1960

1965

1970

1975

1980

1985 Year (b)

1990

1995

2000

2005

2010


400

1.2 Continued

5


14.5

0.0

14.0

–0.5

13.5

Difference from 1961–1990 (mm)

(°c)

0.5

Temperature (°C)

6

(a)

50 0 –50 –100 –150

(b) 40

0

36

–4

(million km2)

(million km2)

4

32

1850

1900

Year (c)

1950

2000

1.3 Observed changes in (a) global average surface temperature; (b) global average sea level; and (c) Northern Hemisphere snow cover for March–April (IPCC, 2007).

Despite the increasing atmospheric CO2 concentrations mentioned in the previous paragraph, energy-related CO2 intensities, expressed as emissions per unit of economic output (Table 1.1), have been projected to improve (i.e., decline) from 2005–2030 as world economies strive to use energy more efficiently. Carbon dioxide intensity by non-OECD countries is projected to decline by an average of 2.6 % per year, from 529 metric tonnes per million dollars of GDP in 2005 to 274 metric tonnes per million dollars of GDP in 2030. For all OECD countries, average CO2 intensity in 2030 is projected to be 296 metric tonnes per million dollars of GDP. The average for the



7

Table 1.1 Carbon dioxide intensity by region and country, 1980–2030, in metric tonnes per million 2000 US dollars of gross domestic product (GDP) (EIA, 2008) Region History Projections

Average annual percent change

1980 1990 2005 2010 2015 2020 2025 2030 1990– 2005

2005– 2030

OECD United States Canada Mexico Europe Japan South Korea Australia/ New Zealand Non-OECD Europe/Eurasia Russia Other Asia China India Other Middle East Africa Central and South America Brazil Other Total world

732 916 867 394 674 482 942 694

565 701 679 441 508 353 729 679

461 544 607 381 383 358 670 633

411 483 563 337 343 316 580 558

379 439 521 312 318 297 521 500

347 399 486 288 290 284 464 449

319 366 453 266 264 273 424 404

296 339 422 247 241 262 396 365

–1.3 –1.7 –0.7 –1.0 –1.9 0.1 –0.6 –0.5

–1.8 –1.9 –1.4 –1.7 –1.8 –1.2 –2.1 –2.2

694 1019 900 1215 755 1959 295 400 450 398 317

711 1166 1060 1339 624 1242 333 352 854 448 310

529 804 836 762 498 693 287 360 903 421 305

440 615 649 573 411 552 221 313 827 362 290

388 531 554 504 363 478 189 299 747 327 262

344 469 494 440 322 421 165 270 679 292 234

306 410 432 385 289 373 148 246 605 255 209

274 368 392 342 261 334 135 224 539 220 187

–2.0 –2.4 –1.6 –3.7 –1.5 –3.8 –1.0 0.1 0.4 –0.4 –0.1

–2.6 –3.1 –3.0 –3.2 –2.5 –2.9 –3.0 –1.9 –2.0 –2.6 –1.9

212 403

211 219 224 398 379 342

208 303

192 267

175 234

162 205

0.2 –0.3

–1.2 –2.4

716

624 494 427

384

345

311

282

–1.6

–2.2

entire world is projected to fall from 494 metric tonnes per million dollars of GDP in 2005 to 282 metric tonnes in 2030 (EIA, 2008). The United Nations Framework Convention on Climate Change (UNFCCC) in 1994, through which the Kyoto Protocol entered into full force in 2005, commits nations to achieving a: ‘stabilisation of greenhouse gas concentrations in the atmosphere at a level that would prevent dangerous anthropogenic interference with the climate system.’ For instance, the Stern review comments that the worst impacts of climate change could be substantially reduced if greenhouse gas concentrations were to be stabilised between 450 and 550 ppm CO2-eq (Stern, 2006). In November 2008, the UK Climate Change Act became law in the UK, setting up a target of 80 % reduction over 1990 CO2 levels by 2050; making the UK the first country to set such a long-range and significant carbon reduction target into law (DECC, 2009). Further global commitments were discussed at the 15th Conference of the Parties (COP15) under the auspices of the United Nations Framework Convention on Climate


8


Change (UNFCCC), in Copenhagen, in December 2009, with the intention to create a legally binding, international treaty to replace the Kyoto Protocol that expires in 2012. Negotiations in the run-up to COP15 showed disagreement on how to tackle climate change and the expectations for a legally binding agreement were lowered (Climatico, 2010). An accord was reached that, although it has significant elements, is not legally binding. The key elements of the accord include the objective to keep the maximum temperature rise to below 2 °C, commitment to list developed countries emission reduction targets and mitigation actions for developing countries, and finance to kick start action in the developing world to fight climate change (http://unfccc. int/meetings/cop_15/items/5257.php). The success of COP15 will depend on the challenge to develop the Copenhagen Accord into a legally binding treaty in 2010 (COP16 in Mexico).

1.3

Carbon management and stabilisation routes

The management of increasing CO2 emissions typically revolves around three broad (but closely related) strategies as possible solutions, namely: (i) switching to a low-carbon economy, i.e., relying on renewable and/ or alternative sources of energy; (ii) increasing the efficiency and energy conservation of our current fossil-fuel energy generation; and (iii) applying CCS technologies to reduce CO2 emissions in order to bridge the gap presented by working to change from our current fossil-fuel dependency to a fully sustainable, low-carbon future. These strategies are discussed in more detail below. Fossil fuels (mainly coal) account for approximately 86 % of the overall world energy use (IEA, 2007; Orr, Jr, 2009), and are foreseen to remain the dominant energy source for the largest part of the 21st century (McKinsey & Company, 2008). Although there is significant concern about the increasing amount of CO2 that will be emitted (Bachu, 2008b; IPCC, 2007, 2005), alternative or renewable energy sources still have fundamental hurdles to overcome. For instance, there are many security and environmental issues still associated with nuclear energy generation, while on the other hand wind, solar, water, wave and geothermal power cannot currently provide sufficient amounts of base-load electricity generation to displace fossil-fuel power. Many of these technologies also rely on the availability of resources, which depends on the geographical location and attributes of a country (Martinot et al., 2007). Furthermore, the use of biomass leaves open the question of the correct technology to be implemented if heavy energy demand is to be met. Hydrogen is likely to be an important energy-carrier in the future (Edwards et al., 2008), but it requires reliable and high-capacity production that is independent of the decarbonisation of fossil fuels, as well as improved storage technologies, to achieve this role in a sustainable future.



9

For the time being, the reduction of CO2 emissions can be achieved by implementing efficient energy strategies. Innovative technologies for power generation, such as Integrated Gasification Combined Cycle (IGCC), may increase the efficiency of conversion of the fuel’s chemical energy from 28–32 % of the recent past to 52 %. Supercritical and ultra-supercritical coalfired power plant technology may also offer a major option for high-efficiency and low-emission power generation, with efficiency projected in the region of 50 % and approximately 30 % projected CO2 emissions reduction. Fuel flexibility can also contribute to the reduction of emissions. For instance, moving from coal to oil to liquefied natural gas (LNG), the amount of CO2 emitted per kWh goes down from 1 to 0.75 to 0.5 kg, respectively. However, even with increased efficiency and reduction in emissions, the rapid expansion of the worldwide demand for energy will ultimately produce a net increase in CO2 emissions. A net reduction in emissions would require a rigorous carbon management strategy to be applied worldwide. CO2 capture and storage (CCS) is a technically feasible strategy to reduce anthropogenic CO2 emissions from large point sources, and particularly fossil fuel-fired power plants, by up to 90 % (IPCC, 2005). One of the key features of this technology is that it allows for the continued use of fossil fuels, including coal which is relatively cheap and abundant, while simultaneously reducing CO2 emissions to the atmosphere (IRGC, 2008). Overall, CCS consists of three main steps: separating and capturing CO2 from other exhaust gases; compressing the CO2 to supercritical conditions in order to transport it to its storage location; and final isolation from the atmosphere by a variety of methods as illustrated in Fig. 1.4. The carbon mitigation potential of the different strategies mentioned above requires a fixed timeframe. For instance, Pacala and Socolow (2004) propose that a 50-year perspective could be long enough to allow changes in infrastructure and consumption patterns but short enough to be heavily influenced by decisions made today. Assuming the world continues on its current predicted path, i.e., business as usual (BAU), it is predicted that CO2 emissions will roughly double by 2054. Stabilising GHG concentrations in the region of 500 ± 50 ppm has been proposed as the target level to prevent the most damaging climate change. Avoiding the doubling of CO2 levels in the business as usual case, in order to reduce substantially the likelihood of the most dramatic consequences of climate change, would require a monumental effort. Committing to a CO2 emissions trajectory approximating a flat path requires an amount of CO2 emissions reduction in 2054 roughly equal to all CO2 emissions today (Fig. 1.5). To assess the potential of the various carbon mitigation strategies, Pacala and Socolow (2004) introduced the concept of stabilisation wedges (Fig. 1.5). The difference between currently predicted path and flat path from present to 2054 gives a triangle of emissions to be avoided, a total of nearly 200


10

Biomass

Cement/steel/ refineries, etc.

Gas Natural gas + CO2 capture

Oil

Mineral carbonation

Coal

CO2

Petrochemical plants + CO2 capture Future H2 use

Electricity generation

Industrial uses

Ocean storage (ship or pipeline)

1.4 Schematic diagram of possible CCS systems showing the sources for which CCS might be relevant, transport of CO 2 and storage options (IPCC, 2005).



Gas to domestic supply


Billions of tonnes of carbon emitted per year

14

11

7 wedges

C

re ur

nt

ly

pr

ed

ic

d te

pa

are needed to build the stabilisation triangle

th

Stabilisation triangle 1 wedge 1 ‘wedge’

7

Flat path 2004

Year

avoids 1 billion tonnes of carbon emissions per 2054 year by 2054

1.5 Stabilisation wedges concept for reducing carbon emissions by 2054 (Socolow et al., 2004).

GtC. The stabilisation triangle can be further divided into seven wedges of equal area each representing a reduction of 1 GtC/year by 2054. This is based on technologies that have the potential to contribute a full wedge to carbon mitigation. CCS technology prevents about 90 % of fossil carbon from reaching the atmosphere, so a wedge would be provided by the installation of CCS at 800 GW of base-load coal plants by 2054 or 1600 GW of baseload natural gas plants.

1.4

Development and innovation in carbon dioxide (CO2) capture and transport technology

Although many of the component technologies for CCS are fairly mature, there are no, as yet, fully integrated commercial applications (Fig. 1.6). There are, however, a number of pilot-scale CCS projects around the world demonstrating confidence in the technology (Table 1.2). The UK government launched a competition in 2007 to build one of the first commercial-scale CCS projects by 2014 (BERR, 2008; APGTF, 2009). As can be seen in Table 1.2, other world governments and energy corporations are focusing on similar incentives to facilitate widespread deployment of CCS technologies in the near future. A programme of 10–12 demonstrations has also been called for in the EU to be operational by 2015, in line with the target of commercialisation of CCS by 2020 (APGTF, 2009). China, which is overtaking the USA in CO2 emissions (approximately 6.0 GtCO2 in comparison to the 5.9 GtCO2 by the USA in 2006) from consumption and flaring of fossil fuels (EIA, 2008), is making significant


Concept

12

Stage of development Lab testing

Commercial refinements needed


First projects are coming online now

Commercial

Component technologies are mature; integrated platform to be proven

Several projects are operational (e.g., Weyburn (Canada)). EU has limited EOR potential

Post-combustion Membranes Chemical looping

Oxyfuel

CO2– EGR

Pre-combustion

Saline aquifers

Sleipner (Norway) field has been operational for around 10 years

Depleted oil and gas fields

Have been used for seasonal gas storage for decades

CO2–EOR Transport Transport offshore onshore

USA has existing CO2 pipeline network of more than 5000 km Capture Transport Storage

1.6 Stage of CCS component technologies (EGR = enhanced gas recovery) (McKinsey & Company, 2008).


Potential future breakthrough technologies

Demonstration


Project name

Location

Leader

Feedstock

Size (MW)

Capture process

CO2 fate

Start-up

Total Lacq Schwarze Pumpe AEP Alstom Mountaineer Callide-A Oxy Fuel GreenGen Williston Kimberlina NZEC AEP Alstom Northeastern Sargas Husnes Scottish & Southern Energy Ferrybridge Naturkraft Kårstø Fort Nelson ZeroGen Antelope Valley WA Parish UAE Project Appalachian Power Wallula Energy Resource Centre RWE npower Tilbury Tenaska HECA UK CCS Project Statoil Mongstad RWE Zero CO2 Boundary Dam Monash Energy

France Germany USA Australia China USA USA China USA Norway UK

Total Vattenfall AEP CS Energy GreenGen PCOR CES UK & China AEP Sargas SSE

Oil Coal Coal Coal Coal Coal Coal Coal Coal Coal Coal

35 30/300/1000 30 30 250/800 450 50 TBD 200 400 500

Oxy Oxy Post Oxy Pre Post Oxy TBD Post Post Post

Seq Seq/EOR Seq Seq Seq EOR Seq Seq EOR EOR Seq

2008 2008 2008 2009 2009 2009–15 2010 2010 2011 2011 2011–12

Norway Canada Australia USA USA UAE USA USA

Naturkraft PCOR ZeroGen Basin Electric NRG Energy Masdar AEP Wallula Energy

Gas Gas Coal Coal Coal Gas Coal Coal

420 Gas process 100 120 125 420 629 600–700

Post Pre Pre Post Post Pre Pre Pre

TBD Brine res Seq EOR EOR EOR TBD Seq

2011–12 2011 2012 2012 2012 2012 2012 2013

UK USA USA UK Norway Germany Canada Australia

RWE Tenaska HEI TBD Statoil RWE SaskPower Monash

Coal Coal Petcoke Coal Gas Coal Coal Coal

1600 600 390 300–400 630 CHP 450 100 60 k bpd

Post Post Post Post Post Pre Oxy Pre

Seq EOR EOR Seq Seq Seq EOR Seq

2013 2014 2014 2014 2014 2015 2015 2016

13

Notes: Seq = sequestration; EOR = enhanced oil recovery; TBD = to be decided; Brine res = brine reservoir; Gas process = gas processing; Pre = pre-combustion; Post = post-combustion; Oxy = oxyfuel combustion.


Table 1.2 Global CO2 capture and storage projects (MIT, 2008)

14


progress in CCS projects. GreenGen is a partnership between the Chinese government, Chinese energy companies and Peabody energy. It is planned to deploy incrementally to a 400 MW IGCC power plant with CCS by 2020. Near Zero Emission Coal (NZEC) is another CCS project between China and the European Union, and has the goal of deploying a coal-fuelled power plant with CCS by 2020. The Australian government in April 2009 formally launched the Global Carbon Capture and Storage Institute (GCCSI), a new initiative aimed at accelerating the worldwide commercial deployment of CCS technologies. The G8 countries have committed to the development of 20 large-scale CCS projects to be operational by 2020, with the GCCSI playing a vital role in developing the partnerships to make these projects a reality. The Department of Energy (DOE) in the USA has formed a nationwide network of Regional Carbon Sequestration Partnerships (RCSP) to help determine the best approaches for capturing and storing GHG. The RCSP initiative is currently in the development phase (2008–2017) to conduct large-volume carbon storage tests.

1.4.1 Carbon dioxide (CO2) capture and storage economics, regulation and planning A major difficulty in deploying CCS technology is the reluctance of corporations to invest given the absence of financial cost associated with greenhouse gas emissions, uncertainty over the future regulations governing coal-burning power plants and CO2 storage, and the need for additional research, development and demonstration (Gibbins and Chalmers, 2008). The absence of governmental regulations and policy frameworks creates additional uncertainty for companies considering investment in CCS (Bachu, 2008b). Volume 1, Chapter 3, deConinck, provides a regulatory analysis and outlook for CCS technologies. Commercial-scale CCS deployment will require a regime to manage risks as well as supporting policies to facilitate technology investment (IRGC, 2008). Public perception and support are also vital for actual implementation of CCS technologies. The main concerns society has over CCS are related to safety issues and the extent to which CCS provides a solution to climate change (Gough and Shackley, 2005; van Alphen et al., 2007). Estimating the cost of CCS technologies involves a high degree of uncertainty over how these costs may develop over time and in terms of potential variations in the technical requirements, scale and application of projects. McKinsey & Company (2008) have recently released a report in which CCS costs are predicted, based on a case-study approach. According to the main findings of the report, early commercial CCS projects, potentially around 2020, are estimated to cost 735–50 per tonne CO2 abated (Fig. 1.7). By far, CO2 capture is the most expensive component and may account for



15

Assumption

1 Capture

2 Transport

25–32

4–6

∑ CO2 capture rate of 90–92 % ∑ CCS efficiency penalty of 7–12 % points ∑ Same utilisation as non-CCS plant (86 %) ∑ CO2 compression at capture site ∑ Transport through onshore/offshore pipeline network of 200/300 km in supercritical state with no intermediate booster station ∑ Use of carbon steel (assumed sufficiently dry CO2) ∑ Injection depth of 1500 m in supercritical state 4–12 ∑ Use of carbon steel (assumed sufficiently dry CO2) ∑ Vertical well for onshore/directional for offshore

3 Storage

35–50*

Total *Ranges are rounded to 5 on totals.

1.7 Total cost of early commercial projects – reference case (7/tonne CO2 abated; ranges include on- and offshore) (McKinsey & Company, 2008).

up to two-thirds of the total cost of a CCS project (725–32 per tonne CO2 abated). Further information on economic analyses of CCS technologies is provided in Vol. 1, Chapter 2, Ogden, and further information on planning and economic modelling for CO2 capture and reduction is provided in Vol. 1, Chapter 4, Elkamel, Mirzaesmaeeli, Croiset and Douglas.

1.4.2 Carbon dioxide (CO2) capture processes and technologies in power plants There are three main technologies for carbon capture from fossil fuel power plants: ∑ after combustion (post-combustion); ∑ decarbonisation of the fuel before combustion (pre-combustion); and, ∑ burning the fuel in pure oxygen (oxyfuel combustion). Post- and pre-combustion processes include chemical and physical capture of CO2 by absorption (Vol. 1, Chapter 5, Desideri) and adsorption (Vol. 1, Chapter 6, Davidson), as well as CO2 separation by membranes (Vol. 1, Chapter 7, Basile, Gallucci, Morrone and Iulianelli) and gasification of fuels syngas/hydrogen for combustion and CO2 for capture (Vol. 1, Chapter 8, Higman). Under oxyfuel combustion conditions, fuel is burnt in pure oxygen rather than air, resulting in more complete combustion and producing a flue gas constituting approximately 90 % CO2 for easier separation (oxyfuel


16


combustion) (Vol. 1, Chapter 9, Mathieu). Advanced oxygen separation and generation systems (Vol. 1, Chapter 10, Kluiters, van den Brink and Haije), and chemical looping combustion systems (Vol. 1, Chapter 11, Anthony) are also being developed as promising alternatives to the more common post- and pre-combustion capture processes.

1.4.3 Carbon dioxide (CO2) compression, transport and injection processes and technologies Transport of pressurised supercritical CO2 from the point of capture to storage sites through an extensive pipeline network is considered to be the most cost-effective and reliable method for onshore CCS (Svensson et al., 2004). Although there may be a risk associated with potential leakage through infrastructure failure or third-party intrusion, there is a lot of experience from the petroleum industry in the USA and Canada in the transport of natural gas, hydrocarbon liquids and CO2 for enhanced oil recovery (EOR) that could be applicable to CO2 transport for CCS. Nevertheless, long-distance CO2 pipeline networks, both onshore and offshore, have technical challenges (e.g., safety and reliability) that need to be faced in order to minimise risks to the environment and human health. Further innovation is focused on pipeline materials, infrastructure and modelling studies to minimise the risk of failure and to yield a better understanding of the consequences and behaviour of a pipeline failure (Mazzoldi et al., 2008). Theses issues and further details on the various compression, transport and injection technologies applicable to CCS systems are respectively discussed in Vol. 1, Chapter 12, Aspelund; Chapter 13, Downie, Race and Seevam; and, Chapter 14, Solomon and Flach.

1.4.4 Industrial applications of carbon dioxide (CO2) capture and storage technologies Wider implementation of CCS is being encouraged in other industries responsible for significant contribution to global CO2 emissions, e.g., the cement and concrete industry (Vol. 1, Chapter 15, Ghoshal and Zeman), and the iron and steel industry (Vol. 1, Chapter 16, Birat). It is estimated that the cement industry is responsible for approximately 5 % of global CO2 emissions (IPCC, 2005). The reduction of CO2 emissions from cement production is currently being addressed by looking into post-combustion and oxygen combustion capture, and using the CO2 for accelerated curing of concrete products and cement-based waste stabilisation/solidification. The iron and steel industry is also responsible for another 6–7 % of global CO2 emissions (IPCC, 2005). Strategies to control CO2 emissions have focused on energy conservation measures. However, further development and incorporation of CCS systems need to be adopted to further reduce CO2 © Woodhead Publishing Limited, 2010


17

emissions. This involves post- and pre-combustion capture processes in the core of the blast furnace, with the potential of retrofitting existing steel mills from the 2020s onwards. New technologies may also be developed with CCS, and possibly without relying on CCS, through the use of hydrogen, electricity or biomass.

1.5

Development and innovation in carbon dioxide (CO2) storage and utilisation technology

1.5.1 Geological sequestration of carbon dioxide (CO2) Various options are possible for final storage of CO2. At present, injection into underground geological formations is the most promising and developed method (Holloway, 2005; IPCC, 2005; Bachu, 2008a,b), although these formations naturally need to be characterised and screened to ensure longterm sequestration (Vol. 2, Chapter 2, Bachu). There are three main types of proposed underground storage site: deep saline aquifers (Vol. 2, Chapter 3, Rosenbauer and Thomas); depleted oil/gas reservoirs and enhanced oil recovery (EOR) (Vol. 2, Chapter 4, Kovscek and Vega); and deep unmineable coal seams (Vol. 2, Chapter 5, Mazzotti, Pini, Storti and Burlini). Geological storage combines a number of engineering processes to ensure safe and long-term isolation of CO2 from the atmosphere. Deep saline aquifers are likely to be the most promising of other geological options, but there is still uncertainty regarding their capacity and geological/geochemical properties. To address the issue, innovative research is being focused to better understand the geochemical reactions between CO2, impurity gases, formation brine, host rocks and cap rocks. Depleted oil and gas reservoirs are frequently said to be the likely first category of geological formation to inject CO2 owing largely to the added benefit of EOR. It is estimated that 80 % of oil reservoirs worldwide might be suitable for CO2 injection for EOR. Enhanced Coal Bed Methane (ECBM) recovery is a technique under investigation for storing CO2 in unmineable coal seams with the added benefit of methane production.

1.5.2 Maximising and verifying carbon dioxide (CO2) storage in underground reservoirs Petrographical studies and the established body of knowledge concerning CO2 storage/migration mechanisms in geological media from the oil industry, combine to improve our understanding of the CO2 injection design approaches that can be adopted to maximise CO2 storage and/or EOR in underground reservoirs (Vol. 2, Chapter 6, Blunt, Qi and LaForce). Similarly, improved methods of sealing underground reservoirs for CO2 trapping have been


18


informed by our understanding of gas leakage mechanisms from geological media (Vol. 2, Chapter 7, Meckel). Improved measurement, monitoring and verification (MMV) (Vol. 2, Chapter 8, Chadwick) and modelling (Vol. 2, Chapter 9, Pruess, Birkholzer and Zhou) techniques are being developed to verify storage and to prove long-term storage security. The environmental impact and safety of CO2 storage is constantly being developed to quantify risk, and better understand and minimise potential leakage of CO2 into freshwater aquifers or back into the atmosphere.

1.5.3 Terrestrial and ocean sequestration of carbon dioxide (CO2) and environmental impacts Another option for CO2 storage is terrestrial sequestration which, along with CO2 emissions reduction, has such ancillary benefits as increased agronomic productivity through CO2 use by plants (Vol. 2, Chapter 10, Lal). While CO2 sequestration has been noted to present benefits to terrestrial ecosystems, up to a point, there remain risks of CO2 leakage from underground reservoirs, and increased levels of CO2 present greater impacts and risks, including the potential for phytotoxicity whereby affected plant species will no longer grow (Vol. 2, Chapter 12, Steven, Smith and Colls). Injection into the deep ocean has also been proposed (Vol. 2, Chapter 11, Golomb and Pennell) but is associated with many more risks to marine ecosystems (including risks of CO2 leakage from subsea reservoirs) that have not yet been adequately researched (Vol. 2, Chapter 13, Blackford, Widdicombe, Lowe and Chen).

1.5.4 Advanced concepts for carbon dioxide (CO2) storage and utilisation Alternative, less conventional, routes for CO2 storage and utilisation are receiving increased attention, including CO2 utilisation by industry, mineralisation/mineral carbonation, biofixation of CO2 by microorganisms and photocatalytic reduction of CO2. The majority of these technologies are still in the research phase, apart from mineralisation/mineral carbonation which is almost ready for pilot-scale demonstration within the next few years. The technological barriers that usually render the process financially non-viable are where further innovation and breakthroughs are needed if these technologies are to have any chance of being integrated into the CCS agenda in the near future. Industrial fixation in inorganic carbonates (mineral carbonation) could have an increasingly important role in future CCS operations, particularly if waste materials are put to use (Vol. 2, Chapter 16, Zevenhoven and



19

Fagerlund). Some industrial processes might also utilise and store small amounts of captured CO2 in manufactured products (Vol. 2, Chapter 14, Aresta and Dibenedetto). In biofixation of CO2, microorganisms use CO2 for cell growth, which can be achieved both in natural habitats as well as in controlled systems such as microalgal farming, while the potential for CO2 conversion to biofuels by microorganisms is also undergoing further research (Vol. 2, Chapter 15, Wang and Lan). The future development of photocatalytic reduction processes for CO2 presents the potential for a similar benefit through the conversion of CO2 to hydrocarbons using artificial light or sunlight, in a process similar to photosynthesis (Vol. 2, Chapter 17, Wu).

1.6

Future trends

As previously mentioned in this chapter, it is highly likely that coal and other fossil fuels will dominate worldwide power generation for the foreseeable future. In order to maintain current standards of living with the ever increasing demand for energy, CCS must play a vital role in reducing CO2 emissions to the atmosphere to mitigate the potential for further global warming and climate change until a low-carbon economy can be fully implemented. At present, however, CCS cannot justify itself on an economic basis alone owing to the high energy penalty and associated costs of installation, and it may remain unprofitable until policy issues are decided and technologies further developed. More research, development and deployment is clearly necessary to make CCS more realistic (in terms of economic costs) to occur on a large scale. Industries are increasingly focusing their attention on CCS technologies, with a large number of pilot-scale projects currently operational and many more being planned to begin within the next five years. International collaboration is now considered a key element in delivering CCS commercialisation targets. It is anticipated that CCS will be demonstrated on a large scale by around 2012, and by the 2020s there will be some tens of CCS plants and some hundreds by the 2030s around the world. Although in comparison to the total number of fossil-fuel fired power stations, these figures are quite low, these developments can be regarded as significant progress towards implementing effective global warming mitigation measures. Legal and regulatory frameworks are underway and are planned to be enforced from about 2010–2012; these will help to justify more robust investment decisions and to minimise barriers for deployment in the near future. For example, the European Commission has recently proposed a Directive on CCS (EC, 2008) to enable environmentally-safe capture and geological storage of CO2 in the EU as part of a major legislative package of measures to achieve the EU’s emissions targets, mitigate the potential effects of climate change and promote renewable energy up to 2020 and beyond.


20


Although this Directive provides a regulatory framework specifically for geological storage, additionally, CO2 capture is regulated under Commission Directive 96/61/EC concerning integrated pollution prevention and control (IPPC) and Commission Directive 85/337/EEC on the effects of certain public and private projects on the environment. The market for CCS is huge, and by about 2020 CCS should have the potential to develop into a significant and competitive option to simultaneously reduce CO2 emissions and promote economic development, energy security and air quality in many countries for decades to come. Compared to a business-as-usual case, CCS could reduce energy-related CO2 emissions by up to a half by 2050. This largely relies on the commercial value of carbon exceeding the cost of CCS technologies and eventually on other incentives for implementation. CCS could also present increased revenue opportunities in the future, particularly where CO2 can be sold to nearby oil field operators for EOR. Other future industrial applications are likely to develop, and business opportunities may include managing CCS applications for hydrogen production, fuel cell applications, emission trading and CO2 storage (WEC, 2007).

1.7

Sources of further information and advice

There is now a plethora of sources for information on CCS. This includes a range of reports, articles published in academic journals, international symposia, research organisations/institutions, professional bodies, popular science books, and the World Wide Web. The ones mentioned below are by no means exhaustive, but merely indicative of the variety available to anyone interested in this field. Each chapter of the book provides additional sources of information. Probably one of the most well-known publications on CCS is the Special Report by the IPCC in 2005. This was the first major reporting of a consortium of leading professionals in the form of an intergovernmental scientific body presenting the urgency to act upon the global warming phenomenon and the role of CCS. This was followed by another renowned publication, The Stern Review (Stern, 2006), on the economics of climate change. Two recent books presenting a holistic account of CCS technologies are available by Wilson and Gerard (2007) and Shackley and Gough (2006). An earlier account on the science and technology of CO2 mitigation is available by Halmann and Steinberg (1999). Marini (2007), Baines and Worden (2004) and Wanty and Seal II (2004) provide extensive information on geological sequestration of CO2. A series of reports are available from many leading professional and research organisations/institutions and government bodies/departments. Some recent examples are mentioned below and other reports have been previously



21

referenced in this chapter. The World Resources Institute (WRI) has released a publication providing guidelines for CO2 capture, transportation and storage (WRI, 2008). McKinsey & Company (2008) address the economics of CCS and predict the cost of CCS technologies in their early days of commercialisation. The International Energy Agency Greenhouse Gas Programme (IEA GHG) holds a large collection of technical reports on CCS and it is advisable to check their website for further information (www.ieagreen.org.uk). The Energy Information Administration (EIA) produces annual international energy outlook reports (EIA, 2008) providing vital information on global energy trends and predictions to 2030. Finally, the UK Advanced Power Generation Technology Forum (APGTF) produced an excellent document in April 2009 providing guidance and advice on research, development and deployment needs of carbon abatement technologies (CATs) for fossil fuels (APGTF, 2009). Academic peer-reviewed articles on CCS are becoming increasingly popular in many journal publications. Elsevier launched a journal in 2008 (International Journal on Greenhouse Gas Control) specifically targeting papers on CCS and related topics, and it has been well received by the scientific community. The Carbon Capture Journal is another recent magazine that aims to inform on developments in CCS and related government policy. Other journals popular amongst researchers include: Energy Conversion & Management (Elsevier); Fuel Processing Technology (Elsevier); Fuel (Elsevier); Energy Policy (Elsevier); Environmental Science & Technology (American Chemical Society); Energy & Fuels (American Chemical Society); Energy & Environmental Science (Royal Society of Chemistry) and Proceedings of ICE – Energy (Institution of Civil Engineers). Several symposia are available to exchange ideas and present the state of the art on research and developments in CCS. The bi-annual International Greenhouse Gas Technologies conference (GHGT) is probably the main venue attracting thousands of delegates. Other well-known venues include the Annual Conference on Carbon Capture and Sequestration (USA); the Platts Annual European Carbon Capture and Storage Conference; and the International Conference on Carbon Dioxide Utilisation (ICCDU). The above mentioned events are dedicated to CCS; however, there are many more conferences in related fields that cover themes on CCS. The main organisations/institutions and centres involved with CCS research and development include the: Centre for Innovation in Carbon Capture and Storage (CICCS) (UK); Scottish Centre for Carbon Storage (UK); UK Energy Research Centre (UKERC); British Geological Survey (BGS); Energy Technologies Institute (ETI) (UK); Princeton University Carbon Mitigation Initiative (USA); Columbia University Earth Engineering Centre (USA); Stanford University Global Climate & Energy Project (USA); US Geological Survey (USGS); Lawrence Berkeley National Laboratory (USA); Los Alamos


22


National Laboratory (USA); National Energy Technologies Laboratory (USA); Global CCS Institute (Australia); CO2 Cooperative Research Centre (CO2CRC) (Australia); Australian Commonwealth Scientific and Research Organisation (CSIRO); CO2NET; Carbon Sequestration Leadership Forum (CSLF); Research Institute of Innovative Technology for the Earth (RITE) (Japan); Bureau de Recherches Géologiques et Minières (BRGM) (France); SINTEF (Norway); Canadian Geological Survey; and Bellona (Norway/ Russia).

1.8

Acknowledgements

The financial support of the Centre for Innovation in Carbon Capture and Storage (CICCS, EP/F012098/1) is gratefully acknowledged by the authors.

1.9

References

APGTF (2009) Cleaner Fossil Power Generation in the 21st Century: A Technology Strategy for Carbon Capture and Storage. UK Advanced Power Generation Technology Forum (APGTF), April. Bachu S (2008a) CO2 storage in geological media: Role, means, status and barriers to deployment. Progress in Energy and Combustion Science, 34: 254–273. Bachu S (2008b) Legal and regulatory challenges in the implementation of CO2 geological storage: an Alberta and Canadian perspective. International Journal of Greenhouse Gas Control, 2: 259–273. Baines SJ and Worden RH (2004) Geological Storage of Carbon Dioxide. Special Publication 233, Geological Society, London, UK. BERR (2008) Towards Carbon Capture and Storage: A Consultation Document. Department for Business, Enterprise & Regulatory Reform (BERR), London, UK. BERR (2009) Energy Trends. Department of Energy and Climate Change, London, UK. Climatico (2010) Copenhagen De-briefing, An analysis of COP15 for long-term cooperation, available at: http://www.climaticoanalysis.org/post/copenhagen-de-briefing-an-analysisof-cop15-for-long-term-cooperation/ (accessed January 2010). DECC (2009) Climate Change Act 2008: Impact Assessment. Department of Energy and Climate Change (DECC), London, UK. Edwards PP, Kuznetsov VL, David WIF and Brandon NP (2008) Hydrogen and fuel cells: towards a sustainable energy future. Energy Policy, 36(12): 4356–4362. EC (2008) COM (2008) 18 final, Proposal for a Directive of the European Parliament and of the Council on the geological storage of carbon dioxide and amending Council Directives 85/337/EEC, 96/61/EC, Directives 2000/60/EC, 2001/80/EC, 2004/35/EC, 2006/12/EC and Regulation (EC) No 1013/2006. Brussels, 23.1.2008. EIA (2008) International Energy Outlook 2008. Energy Information Administration (EIA), US Department of Energy (DOE), DOE/EIA-0404(2008), Washington, DC. Etheridge DM, Steele LP, Langenfields RL, Francey RJ, Barnola J-M and Morgan VI (1996) Natural and anthropogenic changes in atmospheric CO2 over the last 1000



23

years from air in Antarctic ice and firn. Journal of Geophysical Research, 101(D2): 4115–4128. Gibbins J and Chalmers H (2008) Carbon capture and storage. Energy Policy, 36: 4317–4322. Gough C and Shackley S (2005) An Integrated Assessment of Carbon Dioxide Capture and Storage in the UK. Technical Report 47, Tyndall Centre for Climate Change Research, Manchester, UK. Halmann MM and Steinberg M (1999) Greenhouse Gas Carbon Dioxide Mitigation: Science and Technology. CRC Press, Boca Raton, FL. Holloway S (2005) Underground sequestration of carbon dioxide – a viable greenhouse gas mitigation option. Energy, 30: 2318–2333. IEA (2007) World Energy Outlook 2007. International Energy Agency, OECD/IEA, Paris, France. IPCC (2005) IPCC Special Report on Carbon Dioxide Capture and Storage, Working Group III of the Intergovernmental Panel on Climate Change, Metz B, Davidson O, de Coninck HC, Loos M and Meyer L A (eds), Cambridge University Press, Cambridge, UK. IPCC (2007) Climate Change 2007: Mitigation, Contribution of Working Group III to the Fourth Assessment Report of the Intergovernmental Panel on Climate Change, Metz B, Davidson O R, Bosch P R, Dave R, Meyer L A (eds), Cambridge University Press, Cambridge, UK. IRGC (2008) Policy Brief: Regulation of Carbon Capture and Storage. International Risk Governance Council (IRGC), Geneva, Switzerland. Karl TR and Trenberth KE (2003) Modern global climate change. Science, 302: 1719–1723. Marini L (2007) Geological Sequestration of Carbon Dioxide: Thermodynamics, Kinetics and Reaction Path Modelling. Elsevier, Amsterdam, the Netherlands. Martinot E, Dienst C, Weiliang L and Qimin C (2007) Renewable energy futures: Targets, scenarios and pathways. Annual Review of Environment and Resources, 32: 205–239. Mazzoldi A, Hill T and Colls JJ (2008) CO2 transportation for carbon capture and storage: Sublimation of carbon dioxide from a dry ice bank. International Journal of Greenhouse Gas Control, 2: 210–218. McKinsey & Company (2008) Carbon Capture and Storage: Assessing the Economics. McKinsey & Company, available at: http://www.mckinsey.com/clientservice/ccsi/pdf/ CCS_Assessing_the_Economics.pdf (accessed December, 2009). MIT (2008) Carbon capture and storage projects. Massachusetts Institute of Technology, Cambridge, MA, available at: http://sequestration.mit.edu/tools/projects/index.html (accessed December 2009). Orr Jr, FM (2009) CO2 capture and storage: are we ready? Energy & Environmental Science, 2: 449–458. Pacala S and Socolow R (2004) Stabilisation wedges: Solving the climate problem for the next 50 years with current technology. Science, 305: 968–972. Palmer TN and Räisänen J (2002) Quantifying the risk of extreme seasonal precipitation events in a changing climate. Nature, 415(6871): 512–514. Shackley S and Gough C (2006) Carbon Capture and its Storage: An Integrated Assessment. Ashgate Publishing, Aldershot, UK. Socolow R, Hotinski R, Greenblatt JB and Pacala S (2004) Solving the climate problem: technologies available to curb CO2 emissions. Environment, 46(10): 8–19.


24


Stern N (2006) The economics of climate change: The Stern review. Cambridge University Press, Cambridge, UK. Stocker TF and Schmittner A (1997) Influence of CO2 emission rates on the stability of the thermohaline circulation. Nature, 388(6645): 862–865. Svensson R, Odenberger M, Johnsson F and Strömberg L (2004) Transportation systems for CO2 – application to carbon capture and storage. Energy Conservation and Management, 45: 2343–2353. van Alphen K, van Voorst tot Voorst Q, Hekkert M and Smits REHM (2007) Societal acceptance of carbon capture and storage technologies. Energy Policy, 35: 4368– 4380. Wanty RB and Seal II RR (2004) Water–Rock Interaction: Proceedings of the Eleventh International Symposium on Water–Rock Interaction, WRI-11, Taylor & Francis, London, UK. WEC (2007) Carbon Capture and Storage: A WEC “Interim Balance”. World Energy Council (WEC) Clean Fossil Fuels Systems Committee. Wilson EJ and Gerard D (2007) Carbon Capture and Sequestration: Integrating Technology, Monitoring and Regulation. Wiley-Blackwell, Oxford, UK. WRI (2008) CCS Guidelines: Guidelines for Carbon Dioxide Capture, Transportation and Storage. World Resource Institute (WRI), Washington, DC.


2

Screening and selection criteria, and characterisation techniques for the geological sequestration of carbon dioxide (CO2) S. B a c h u, Alberta Innovates – Technology Futures, Canada Abstract: Potential sites for the geological sequestration of CO2 must be subject to a systematic screening process. The chapter covers fundamental site selection criteria (capacity and injectivity; safety and reliability; compatibility with other energy, mineral and water resources; regulatory and other societal requirements), site characterisation (geology; hydrogeological, pressure and geothermal regimes; land features; location, climate and access) as well as future planning (predicting the fate and effects of the injected CO2; site design, permitting, operating, monitoring and eventual abandonment). Estimates of storage capacity are critical to the site selection and characterisation process, and these are based on volumetric evaluations. Key words: CO2 storage, site selection, site characterisation, storage capacity, coal beds, hydrocarbon reservoirs, deep saline aquifers.

2.1

Introduction

Three storage media have been identified that have the potential to sequester carbon dioxide (CO2) in the short-to-medium term: uneconomic coal beds, oil and gas reservoirs and deep saline aquifers (IEA, 2004; Benson et al., 2005). Of these, CO2 storage in uneconomic coal beds has been identified as: (i) being an immature technology and (ii) having the smallest storage potential (Benson et al., 2005). Oil reservoirs possess, at depletion, smaller storage capacity than gas reservoirs or deep saline aquifers; however, they present the advantage that, if suitable for CO2 enhanced oil recovery (CO2–EOR), their storage capacity will increase and the cost of storage will decrease by producing additional oil. Gas reservoirs have significant CO2 storage capacity because of their very large recovery factor (between 80 and 90 %) and large size. However, it is believed that deep saline aquifers possess the largest CO2 storage capacity, besides having the advantage that they are present also in regions where there are no oil and gas reservoirs or where oil and gas reservoirs are still in production and are not yet available for CO2 storage. Along the various elements of the CO2 capture and storage (CCS) chain, site selection and characterisation is of critical importance because any 27 © Woodhead Publishing Limited, 2010

28


storage site must demonstrate that it can store the intended amount of CO2 at the rate that it is supplied to the site without posing any unacceptable risks. As identified in the IPCC Special Report on CCS (Benson et al., 2005), for sites that are properly selected, designed and operated, the expectation is that there will be no leakage or that, if there is leakage, it will be below an acceptable level from both the point of view of atmospheric greenhouse gas emissions and from a health, safety and environmental (HSE) point of view. Regulatory agencies, which will permit CO2 storage in geological media, and the public, which has to accept and support CCS as a climate-change mitigation strategy, need to be convinced that proposed CO2 storage sites meet these requirements, and this can be achieved only through a proper and transparent process of site selection and characterisation. Site screening is the process by which the potential for CO2 storage in a selected region, defined either by geology (e.g., sedimentary basin), jurisdiction (e.g., country, province or state) or any other criterion, is evaluated by assessing and comparing possible candidate storage sites. The aim is to identify the sites that meet CO2 storage requirements. Ranking of these sites according to various criteria allows identification of the best sites with respect to that set of criteria, enabling investment decisions into further site characterisation. Sites for CO 2 storage vary around the globe in their quality and characteristics, and there will be instances where sites of poorer quality will be used for storage because no other sites are available or because other sites are too far away and/or much more costly to develop and operate. However, use of poorer-quality storage sites means that additional measures may have to be taken, particularly with regard to ensuring their safety, in order to obtain regulatory approval. For this reason, it is important to be able to judge the quality of a storage site based on an established and accepted set of criteria. This chapter will present criteria by which prospective sites should be assessed in terms of their suitability for CO2 storage and ranked for site selection, together with methodology for estimating their storage capacity.

2.2

Screening for storage suitability and site selection

2.2.1 Selection of sites for carbon dioxide (CO2) storage Currently there is very little practical experience worldwide with the selection and characterisation of sites for the injection/disposal of buoyant fluids. Fifteen criteria for assessing and ranking sedimentary basins in terms of their suitability for CO2 storage were successively introduced by Bachu in a series of papers (Bachu and Gunter, 1999; Bachu, 2000, 2001, 2002, 2003), criteria that can be broadly divided into the following categories: 1. basin characteristics (e.g., size and depth);


Geological sequestration of carbon dioxide (CO2)

29

2. tectonic and geological; 3. hydrogeological and geothermal; 4. basin resources and industry maturity, i.e., the presence or absence of various media suitable for CO2 storage (oil and gas reservoirs, coal beds, salt domes and/or beds) and the degree of exploration and production associated with hydrocarbon resources; 5. CO 2 sources, accessibility and infrastructure, including climatic conditions; 6. economic, political and societal factors. In addition to these basin-scale criteria, Bachu (2003) introduced local-scale selection criteria such as: caprock integrity including well penetrations, in situ conditions, fate of the injected CO2, long-term site integrity and safety and local public acceptance. Building on previous work by Bachu (2003) and the Carbon Sequestration Leadership Forum (Bachu et al., 2007; CSLF, 2007), Kaldi and Gibson-Poole (2008) produced a similar system for screening of sedimentary basins for CO2 storage at the country scale, adding two more selection criteria: ‘Reservoir-seal pairs’ which refers to secondary geological barriers to upward CO2 flow, and ‘Coal rank’, which refers to the type and quality of coal for CO2 storage in coal seams. The IPCC Special Report on CO2 Capture and Storage (Benson et al., 2005) has reviewed the status of knowledge up to and including 2005 publications on this subject. Ideally, sites should be located in a stable geological environment thus avoiding potential compromising of the site in the future, but it is recognised that the fundamental storage criteria may be met also by sites in a less stable tectonic environment. Matching of CO2 sources with potential sinks is an important criterion for site selection, since potential sinks that are too far from CO2 sources will be stranded and will not be utilised due to the high cost of transportation. In 2008, the European Union produced a Best Practice Manual for Storage of CO2 in Saline Aquifers (Chadwick et al., 2008), building on the European experience gained between 1998 and 2006 through the SACS, SACS2 and CO2STORE projects in Europe. The authors compiled a table of positive (favourable) and cautionary (less favourable) qualitative indicators regarding reservoir and caprock properties to be used in the assessment of prospective CO2 storage sites. In the case of the five sites examined in the SACS and CO2STORE projects, various screening/selection criteria were used, such as size (capacity), injectivity, depth, structural enclosure, aquifer thickness, porosity, caprock quality, distance from the CO2 source, potential conflict with other industries (mainly hydrocarbon production) and land use, right of access and population density (for onshore sites). Societal risks associated with CO2 storage were also examined and used as a screening tool, including HSE risks, economic risks and risks related to public perception and trust.


30


Oldenburg (2008) presented a screening and ranking framework based on the potential for and effects of leakage from a CO2 storage site by evaluating three main characteristics: (i) the potential for primary containment of the CO2 within the target unit; (ii) the potential for secondary containment between the target unit and the shallow subsurface in case CO2 containment in the primary storage unit fails; and (iii) the potential for attenuation of the leaked CO2 in the shallow subsurface and at surface, in case CO2 leaks past secondary containment. Ramírez et al. (2009) presented a system for screening storage options in the Netherlands that was developed based on the opinion of a large group of international experts. Preliminary results show that aquifers have an average risk (mainly because of lack of detailed information and because seal capacity is not proven), and that storage in aquifers entails a higher cost than hydrocarbon reservoirs because of lack of infrastructure. Wilkinson et al. (2009) introduced three key priorities for storage site selection, design, construction and operations: (i) Safety, Health and Environment (SHE), which involves protection of personnel and the environment; (ii) Efficiency, which is defined by storing the greatest amount of CO2 in the minimum amount of pore space at the lowest unit cost; and (iii) Reliability, which is an operational concept and refers to maximising uptime and minimising downtime of the storage operation when CO2 storage will have to cease. In a series of two papers, Kopp et al. (2009a,b) have examined through dimensional analysis and numerical simulations the effect of parameters such as depth, temperature, absolute and relative permeability and capillary pressure on storage capacity in aquifers. Relative permeability exerts a great influence on storage capacity through both irreducible CO2 and brine saturations, and the shape of the relative permeability curves. Furthermore, storage capacity and injectivity are not completely independent of each other because both depend on permeability and mobility. Basically, higher storage capacity is achieved for lower injectivity. This counter-intuitive result is due to the fact that, in a homogeneous environment, poorer sweep efficiency is attained for higher permeability/injectivity, hence less pore space will be reached and less residual gas trapping and less dissolution will occur. The work to date suggests that screening for site selection is a two-stage process, by which the first assessment is performed at the basin and/or regional scale, and the more detailed selection is performed at the local and/ or site-specific scale. Basin and regional scale screening The criteria developed by Bachu (2003) with improvements by IPCC (Benson et al., 2005) and Kaldi and Gibson-Poole (2008) form the basis of site screening at these scales. Table 2.1 presents a set of eliminatory criteria, i.e.,



31

Table 2.1 Eliminatory suitability criteria for assessing sedimentary basins for CO 2 geological storage

Criterion

Not suitable

Suitable

1 Depth Less than 1000 m

Greater than 1000 m, with storage units deeper than 800 m

2 Reservoir-seal pairs Poor and stratigraphic sequences

Intermediate and excellent At least one major extensive, regional-scale competent seal

3 Pressure regime

Over-pressured

Hydrostatic or sub-hydrostatic

4 Seismicity (basin tectonic setting)

High and very high (subduction zones; syn-rift and strike- slip basins)

Very low to moderate (foreland, passive margin and cratonic basins)

5 Faulting and fracturing Extensive intensity

Limited to moderate

6 Surface areal extent

Less than 2500 km2

Greater than 2500 km2

7 Hydrogeology

Shallow, short flow systems, or compaction flow

Intermediate and regionalscale flow systems; topography and erosional flow

8 ‘Legal’ accessibility

Forbidden

Possible

a sedimentary basin or region that does not pass these criteria should not be considered for CO2 storage. The first three criteria (depth, seal and pressure regime) are critical criteria, in the sense that a basin or part thereof that does not satisfy all of these suitability criteria should automatically be deemed as being not suitable for CO2 storage because of the high risk of compromising the safety and security of storage. The next four criteria (seismicity, faulting and fracturing, size and hydrogeology) are essential criteria in the sense that there may be special cases, which have to be thoroughly documented and justified, where, if one of these criteria is not being met but all the others are, such a basin may still be considered for CO2 storage. Examples would be small intramontane basin with a single large CO2 source (e.g., a pulp mill) and with no other options for storage, such as in the Canadian Rocky Mountains, or sedimentary basins in California (an active tectonic zone) where hydrocarbon reservoirs are present. However, if more than one of the essential suitability criteria is not being met, then that basin or region should not be considered for CO2 storage. Finally, the last criterion (Legal Accessibility) is also a critical criterion (i.e., it is eliminatory by itself) but, unlike the others, it is not a physical characteristic of the basin but rather a designation resulting from a legislative or regulatory act that may change in the future. Examples are certain offshore North American basins where drilling and exploration are not permitted at this time.


32


Other suitability criteria for CO2 storage, listed in Table 2.2, fall in the desirable category. The first four criteria (1–4) refer to general CO2 storage characteristics of the basin. Storage sites within fold belts are less desirable because of faulting. However, there may be cases where storage sites may be found in the fold belts of mountain ranges, similarly to the very large gas reservoirs found in the thrust and fold belt of the Canadian Rocky Mountain foothills. Diagenetic processes usually lead to loss of porosity (hence of storage space) and permeability (hence of injectivity). Warm basins have lower storage efficacy and the stored CO2 is subject to stronger buoyancy forces because of lower CO2 density. Salt beds have the advantage that they form strong barriers to upward CO2 flow, and both salt domes and beds may offer the opportunity of CO2 storage in salt caverns. The next five criteria (5–9) refer to the potential for storage in specific media. Hydrocarbon potential and industry maturity refer to the potential for CO2 storage in oil and gas reservoirs, while the presence or absence of coal seams, their rank and their economic value refer to the potential for CO2 storage in coal beds. The next four criteria (10–13) are proxies for the economics of CO2 storage. Costs are higher deep offshore, in harsh climate conditions (e.g., in the Arctic), in places that are difficult to access (again, Table 2.2 Desirable characteristics of sedimentary basins or parts thereof suitable for CO2 storage

Criterion

Undesirable

1 Within fold belts Yes 2 Significant Present diagenesis 3 Geothermal Warm basin (gradients regime > 40 °C/km and/or high surface temperature) 4 Evaporites (salt) Absent 5 Hydrocarbon Absent or small potential 6 Industry maturity Immature 7 Coal seams Absent, very shallow or very deep (< 400 m or > 800 m depth) 8 Coal rank Lignite or anthracite 9 Coal value Economic 10 On/off shore Deep offshore 11 Climate Harsh 12 Accessibility Inaccessible or difficult 13 Infrastructure Absent or rudimentary 14 CO2 sources within Absent economic distance

Desirable No Absent Cold and moderate basins (gradients < 40 °C/km and low surface temperature) Domes and beds Medium to giant Mature At intermediate depth (400–800 m) Sub-bituminous and/or bituminous Uneconomic Shallow offshore and/or onshore Moderate Good Developed Present



33

Arctic, or intramontane basins, etc.) where infrastructure is lacking and has to be built or the terrain is difficult (e.g., in or across mountain ranges). The last criterion is also a proxy for economics, but the economic distance may change with developments in pipeline technology and construction, in shipping technology or with the construction of new CO2 sources. None of these criteria is eliminatory by itself, and it could be that a basin or region thereof may have several unfavourable characteristics and still be considered for CO2 storage. On the other hand, if too many criteria are unfavourable, then serious consideration should be given as to whether to proceed with CO2 geological storage in that basin or part thereof. Generally, Table 2.2 can be used to asses if a sedimentary basin has poor or good potential for CO2 storage before proceeding with storage capacity estimations and site selection. A basin or part thereof that meets most of the desirable criteria has good potential, while the reverse is true for basins or regions with poor CO2 storage potential. Local and site scale screening A storage site must pass the criteria for identification of sedimentary basins suitable for CO2 storage (Table 2.1), should posses desirable (or favourable) characteristics (Table 2.2) and must meet additional screening criteria that are specific and can be applied only at these scales. Local-scale site selection criteria can be grouped into the following broad categories: 1. capacity and injectivity, which is similar to capacity and deliverability in natural gas storage; 2. confinement, including avoidance or minimisation of risks to other resources, equity and life, as well as return of CO2 to the atmosphere; 3. legal and regulatory restrictions, including access; 4. economic, including costs, infrastructure, financing, etc., 5. societal attitudes. Although in the past capacity and injectivity were considered as separate criteria for site selection, more recent work indicates that they are not completely independent of each other, at least not during the active period of injection. Because of the link between them, they are considered hence as a single criterion. There is a significant difference between the first and the second criteria listed above. These criteria should be simultaneously met. However, if capacity and/or injectivity are less than those initially desired or estimated, operational adjustments can be made (e.g., storing less CO2 or injecting at a lower rate, with the balance of CO2 being diverted to another site, or just storing a smaller volume than that emitted). On the other hand, lack of confinement would automatically exclude a site from consideration. Also, the first criterion applies to the active period of CO2 injection, which


34


is in the order of decades, while the second one applies to a much longer period, up to three orders of magnitude longer than the injection period. Failure to properly assess site capacity and/or injectivity can and will be identified during the operational (injection) period, and, in the case that any of them is lacking, measures can immediately be taken, such as increasing well injectivity, drilling additional injection wells or moving to another site if there is insufficient capacity. Meeting the second criterion must be demonstrated prior to injection, based on site knowledge and predictions of the fate and effects of the injected CO2. Lack of confinement, with corresponding CO2 migration and/or leakage,1 may occur much later (years to centuries) after cessation of injection, in which case different remedial measures have to be taken that no longer affect the selection and operation of the site. Many other detailed site selection criteria derive from these two. Some of the screening criteria are outright eliminatory, such as unacceptable risk of leakage, or strong public opposition or lack of access. Some may be applicable at a particular moment in time, but may change in the future, e.g., economics of operation, or regulatory requirements or availability in the case of a hydrocarbon reservoir that is still producing. Very few of the site selection criteria listed below are eliminatory by themselves but, if a prospective storage site has too many unfavourable characteristics, then serious consideration should be given to rejecting it. The following are eliminatory criteria, i.e., sites falling in any of the following categories should generally be eliminated from further consideration. In discussing these criteria, one should be aware of the three-dimensionality of the problem, the storage unit being located at a certain depth in the subsurface but requiring land access at surface. These eliminatory criteria can be grouped into three broad categories: (i) lack of legal and/or physical access; (ii) potentially affecting other resources whose production and/ or utilisation has primacy over CO2 storage; and (iii) lacking security and safety. 1. Legally inaccessible. This is the case of potential sites located in protected or reserved areas, such as national parks, protected natural reserves, military areas and regions where drilling and exploration are prohibited (e.g., offshore in certain US and Canadian waters). 2. Legally unreachable. This is the case of sites for which right of access cannot be obtained for various reasons (mostly land access). In some 1

For clarification, migration in this context is defined as the lateral movement of a plume of injected CO2 within the same geological unit, e.g., aquifer, but out of the storage site as legally defined through the tenure and permitting process, while leakage is defined as cross-formational flow of CO2 out of the storage unit into overlying strata and possibly to shallow groundwater, soil and/or the atmosphere.



3.

4.

5. 6.

7.

8.

9.

35

jurisdictions, there are regulatory bodies that may decide and grant right of access for the greater public good. Legally unavailable. This would be the case of a prospective site, such as a hydrocarbon reservoir, in which there are third party equity interests that make the site unavailable for CO2 storage. Future legislation may deal with this situation through unitisation, like in the case of flooded oil reservoirs. Or it could be the case of a subsurface property owner, like in the case of freeholders in Canada, or land owners in the USA, who refuse access to/use of the subsurface storage site itself. Physically unavailable. This would be the case of a hydrocarbon reservoir still in production, and possibly of an aquifer used for mineral or geothermal energy production, unavailable for CO2 storage within the time period of interest. Located within high-density population areas. Sites located in areas of high population density should not be considered because of very likely public opposition. Potentially affecting other natural, energy and mineral resources and equity. Sites where CO2 storage may affect directly or indirectly other resources and/or equity should not be considered for CO2 storage. There will be cases when the interested parties may reach compensatory agreements, but generally such sites should be considered only as a last resort. Within the depth of protected groundwater. Sites at depths encompassed by the designation of protected groundwater should not be considered for CO2 storage, regardless of actual physical depth that may in some cases be greater than the recommended 800 m. Protected groundwater is defined as groundwater with salinity less than a certain threshold, which varies between 4000 and 10 000 ppm depending on jurisdiction. With increasing needs for water and increasing reliance on groundwater, probably groundwater with salinity less than 10 000 ppm should be protected for human, agricultural and industrial use. Located at shallow depth. Generally a depth of minimum 800 m has been considered as necessary for CO2 storage to maximize storage efficacy (amount of CO2 stored per unit of pore volume). Although this is not a ‘hard’ threshold, in the sense that some sites could be at a shallower depth, particularly depleted hydrocarbon reservoirs, the congruence of this and other criteria such as groundwater protection, and the general acceptance of this threshold depth, makes this an eliminatory criterion. Lacking at least one major, extensive, competent barrier to upward CO2 migration. This relates to the requirement of security and safety of storage, i.e., containment within the primary storage unit. A highly fractured region, with fractures reaching to the surface, will also fall into this category.


36


10. Located in an area of very high seismicity. This also relates to the security and safety of storage. 11. Located in overpressured strata. The risks of leakage and/or well blowout are higher in overpressured strata (approaching lithostatic) than in normally-pressured and sub-hydrostatic aquifers and/or reservoirs. 12. Lacking monitoring potential. Since it is assumed that regulatory requirements for site permitting, operation and abandonment will include monitoring of the fate and effects of the injected CO2, sites where monitoring is not possible or very poor most likely will not be approved, and hence should be avoided. This would be the case where an aquifer or reservoir is too thin or with low porosity such that the CO2 plume is not discernible on seismic, nor are wells available for monitoring. The last criterion does not fall specifically in any of the three broad categories mentioned initially; however, it may and likely will lead to the rejection of a proposed site. While the previous criteria were of an eliminatory nature (i.e., a site either passes all these criteria or is rejected), the following criteria are selection criteria in the sense that these are favourable characteristics that would make a site preferable to another, all other considerations being equal. Failure to meet a particular criterion will not eliminate a site, it will only reduce its ‘desirability’ or suitability. 1. Sufficient capacity and injectivity. It is desirable to have sufficient capacity for storing emissions at the supply rate for the entire period of time but, depending on economics and the expansion of the pipeline network for CO2 collection and distribution, smaller-capacity sites may be considered. It is very important to assess not just the ‘static’ storage capacity according to accepted methodology and guidelines and based on ultimately-available pore volume (e.g., see Bachu et al., 2007), but also the ‘dynamic’ storage capacity, i.e., the storage capacity that can be achieved during the active lifetime of the project by injecting CO2 at rates and pressures that meet safety and regulatory requirements (Birkholzer and Zhou, 2009; Mathias et al., 2009). This refers to maintaining maximum bottomhole injection pressure at injection wells and aquifer or reservoir pressure below any, or a combination, of: ∑ a fraction of the rock fracturing threshold (usually established by regulation, e.g., 90 % in Alberta and 75 % in British Columbia, Canada); ∑ fracture and/or fault opening pressure (for pre-existing fractures and faults); ∑ caprock fracturing pressure; ∑ caprock displacement pressure (the pressure at which the injected



37

CO2 intrudes into the caprock, related to capillary pressure and interfacial tension of the CO2/brine system at in situ conditions); ∑ initial reservoir pressure (in the case of storage in hydrocarbon reservoirs). It is important to note that the caprock displacement pressure may in some cases limit the injection pressure more significantly than the other geomechanical criteria like fault reactivation and fracturing threshold. This is because the interfacial tension (IFT) of water/CO2 systems is about half of water/CH4 systems (water/gas) (Li et al., 2006); thus capillary breakthrough may occur in systems that held natural gas for geological periods of time (e.g., in some cases of depleted gas reservoirs that are pressurised with CO2 back to the initial reservoir pressure). Furthermore, if the injected stream of CO2 contains impurities such as H2S, the capillary breakthrough pressure will be even lower because the IFT of H2S/water systems is lower than that of CO2/water systems by a factor of 2–3, and the IFT of systems containing CO2–H2S mixtures and water is proportional to the CO2 (or H2S) mole fraction in the injection stream (Shah et al., 2008). The dynamic storage capacity is most likely significantly less than the static storage capacity, and can be evaluated only through numerical modelling (simulation) of injection, including various injection strategies such as (i) increasing the number of injection wells, (ii) using directional and horizontal wells and (iii) pumping aquifer water while injecting CO2. Unacceptable pressure increases identified through modelling studies could render a potential site unable to store the intended volume (van der Meer and Egberts, 2008; Ghaderi et al., 2009), leading to its rejection or to changes in overall storage strategies (e.g., use of several sites instead of just one). Relative permeability affects the pressure buildup in an aquifer, hence the pressure-induced risk during the active phase of CO2 injection (Oruganti and Bryant, 2009). 2. Sufficient thickness. Thick aquifers or reservoirs are preferable to thin ones, not just because of assumed higher storage capacity but also because they allow various injection strategies (e.g., injection at the bottom and letting the plume of CO2 rise), can be assumed to lead to areally smaller plumes of CO2, increase CO2 dissolution through larger area of contact between CO2 and aquifer water, increase CO2 residual-gas trapping through exposing more pore volume to CO2, and increase the potential for monitoring of plume evolution using geophysical methods. A minimum thickness of 20 m is recommended. 3. Low temperature (as defined by low geothermal gradients and/or low surface temperatures). This increases storage efficacy by ensuring higher CO2 density, hence higher storage capacity for the same pore volume. It also increases storage security by decreasing the buoyancy force acting on CO2. © Woodhead Publishing Limited, 2010

38


4. Favourable hydrodynamic regime. Aquifers with long-range, regional-scale flow systems are preferable to those with intermediate or short systems because the former will allow a much longer travel time during which the injected CO2 will dissolve and/or be immobilised by residual-gas trapping. 5. Low number (density) of wells penetrating the storage area of influence. The presence of wells increases the potential for, and hence the risk of, leakage. The presence of wells constitutes a conundrum for the following reasons. A larger number of wells results in a better characterised storage unit, a greater degree of confidence in the assessment, and a better potential for monitoring through fluid sampling and/or seismic measurement. On the other hand, the potential for leakage increases with increasing number of wells. It is very important to have a clear understanding of what constitutes the area of influence. Injecting CO2 in a reservoir or aquifer has two effects: introduction of an immiscible buoyant fluid that will spread and migrate for a certain distance until immobilisation, and increasing reservoir or aquifer pressure which will propagate significantly faster and beyond the distance of the CO2 plume itself (Birkholzer and Zhou, 2009). This pressure wave may affect other resources and may induce brine leakage in wells and/or fractures that otherwise will not be reached by CO2. In this respect, the presence of wells and their potential for leakage should be considered for the area of influence as defined by pressure increase and not by CO2 reach. 6. Presence of a multilayered overlying system of aquifers/reservoirs and aquitards/caprock. This increases the safety and security of storage (secondary containment in case of leakage), and is particularly important in the case of sites with a significant number of well penetrations (e.g., see Nordbotten et al., 2005, 2008). 7. Potential for attenuation of leaked CO2 near and at surface (in shallow groundwater, soil and atmospheric). Sites with characteristics more favourable for CO2 attenuation and dispersion near and at surface as a result of topographic, climatic and/or vegetation conditions should be preferred to sites where CO2 will have a tendency to stagnate and accumulate. 8. Site accessibility and infrastructure. This affects the cost/economics of transportation to the storage site. This criterion includes any combination of location (onshore/offshore), distance from source, terrain and climate difficulty, right of access (e.g., pipeline corridors) and avoidance of populated or reserved areas that have to be bypassed. 9 Storage economics. This includes the cost of site facilities, wells and compression, and affects the cost/economics of the entire CO2 capture and storage operation.



39

Porosity, permeability and depth have been suggested in the past to be site selection criteria, but they should not explicitly be. Porosity and permeability are indirectly included in the criterion of capacity and injectivity. Depth is indirectly included in the criteria of groundwater protection and storage economics. Water salinity was also suggested by some to be a selection criterion, but it is implicitly taken into account by the criterion of groundwater and mineral resource protection. It could be seen that the first three site selection criteria refer to the efficacy of storage (capacity and injectivity), the next four criteria refer to the safety and security of storage, and the last two refer, in aggregate, to the cost and economics of storage. With regard to the latter category, conditions may change as infrastructure and technology develop, and as regulations and economic conditions change.

2.2.2 Carbon dioxide (CO2) storage in depleted hydrocarbon reservoirs and in enhanced oil or gas recovery Carbon dioxide can be stored in hydrocarbon reservoirs after abandonment (at depletion), or can be stored while hydrocarbons are still being produced, during enhanced recovery operations. The latter option provides the advantage that some of the CCS costs will be offset or, most likely, an economic profit will be realised, as a result of incremental oil production. Although 800 m is considered as the minimum depth for CO2 storage because of the high density of supercritical CO2, shallower hydrocarbon reservoirs should not be rejected a priori if they meet the requirements of capacity, confinement, risk minimisation and societal acceptance (injectivity has been demonstrated by producing the oil and/or gas). An important additional screening/selection criterion for oil and gas reservoirs is time of availability (i.e., time of depletion and abandonment2). Since these reservoirs are producing, they may not be available for CO2 storage within the desired time frame, thus being eliminated from consideration. The existence or absence of strong aquifer support is also an important factor in the selection of oil or gas reservoirs for CO2 storage because, in the case of strong support, aquifer water will invade the reservoir as oil and/or gas is being produced, thus reducing the reservoir storage capacity and also increasing its pressure (Bachu and Shaw, 2003). Similarly, oil reservoirs that have been water flooded (secondary recovery) and solvent or gas flooded (tertiary recovery) should not be considered for CO2 storage because the pore space is filled with 2

It should be noted here that reservoir depletion and abandonment are relative terms, dictated as much by geology and engineering as by economic conditions.


40


water and/or gas/solvent, and the pressure is most likely close to the initial reservoir pressure. Finally, reservoir size, expressed either by an estimate of CO2 storage capacity or by recoverable oil (or gas) in place (ROIP), which is given by the product of recovery factor (Rf) and original oil in place (OOIP), should be used as a screening criterion because it is uneconomic to develop the necessary transportation and storage infrastructure for reservoirs that do not have the required capacity. Carbon dioxide retention in oil or gas reservoirs during enhanced recovery represents a special case of CO2 storage. In these cases, sites must meet an additional set of criteria very specific to the respective process of enhanced oil or gas recovery. Use of CO2 in enhanced gas recovery has not been practised to date, and it is questionable because gas reservoirs have a high recovery factor (greater than 80 %) and because, at the low reservoir pressures attained at this stage of production, CO2 will expand quickly in the reservoir and will be produced together with the remaining gas (e.g., Pooladi-Darvish et al., 2008), hence necessitating additional separating facilities and recirculating the produced CO2. Thus, CO2 will most likely be stored in gas reservoirs after abandonment (depletion). The situation is different for oil reservoirs, which most often have much lower recovery factors (less than 30 %) and for which the value of the incremental oil produced justifies the additional costs of enhanced recovery (including CO2 separation). Holtz (2009) indicates that real-time CO2 trapping within oil reservoirs occurs as a mobile phase, residual phase and dissolved in the oil phase. Long-term trapping is achieved by dissolution of CO2 in the water phase and through mineral precipitation. Furthermore, Holtz (2009) identifies five successful gas displacement recovery processes where CO2 may be used: 1. miscible displacement; 2. immiscible displacement; 3. pressure maintenance, particularly for gas condensate reservoirs or oil reservoirs with a gas cap; 4. gas assisted gravity drainage, which usually results in the lowest water and oil saturations and highest saturation for stored CO2; 5. mixed gas EOR (e.g., mixtures of CO 2 , N 2 , solvents and light hydrocarbons). Not all oil reservoirs are suitable for CO2–EOR; thus, additional criteria must be applied for the identification and selection of oil reservoirs for CO2 flooding, notwithstanding the economics of such operations. This is because most CO2–EOR operations are based on the miscibility between oil and CO2, and their phase behaviour. Based on the experience with CO2–EOR in the USA, a series of authors have identified several criteria, reviewed in Shaw and Bachu (2002), for the identification of oil reservoirs technically suitable for CO2–EOR. Based on the previous analyses and on the characteristics of



41

more than 100 existing CO2–EOR operations in North America at the end of 2007 (Moritis, 2008), the criteria presented in Table 2.3 could be used for identifying oil reservoirs suitable for miscible CO2–EOR (the immiscible operations are too few to draw meaningful conclusions). These criteria refer only to miscible CO2–EOR, and not to the other gas displacement processes identified by Holtz (2009). In addition, reservoir pressure at the beginning of the CO2–EOR operation should be above the minimum miscibility pressure (MMP), i.e., the pressure at which CO2 and oil become miscible. The MMP depends on oil gravity and a few other characteristics, and is usually determined in the laboratory on a case-by-case basis, but it is always above CO2 critical pressure. In the absence of specific laboratory data, particularly when screening a large number of oil reservoirs, an empirical relationship can be applied to estimate MMP on the basis of the molecular weight of the C5+ components in reservoir oil and reservoir temperature (Núñez-López et al., 2008). The condition that reservoir pressure should be greater than the MMP for miscibility, while at the same time it should be less than the fracturing pressure Pf, implicitly introduces a screening criterion that MMP should be less than Pf. Bachu and Shaw (2005) have used an additional criterion for selecting oil reservoirs for CO2–EOR and CO2 storage, namely that of storage capacity greater than 1 Mt CO2 (which is a proxy for economics). Núñez-López et al. (2008) have developed similar methodology, based on the same principles, for screening of oil reservoirs suitable for CO2–EOR and estimating their incremental oil recovery and corresponding CO2 storage capacity, and have applied it to oil reservoirs along the US Gulf Coast, except that they start from reservoir size as the first screening criterion and consider only reservoirs with a cumulative production greater than 1 million standard barrels (MMstb). This approach eliminates from the beginning small reservoirs while avoiding using estimates of CO2 storage capacity, which are uncertain, as a screening criterion. While these additional criteria are suitable for Gulf Coast reservoirs Table 2.3 Suggested characteristics of oil reservoirs suitable for miscible CO2–EOR (metric values are given in brackets) Reservoir parameter

Miscible CO2–EOR

Size (ROIP in MMstb; or MtCO2) Depth (ft/m) Temperature (°F/°C) Pressure Porosity (%) Permeability (mD) Oil gravity (API) Oil viscosity (cP/mPa·s) Remaining oil fraction in the reservoir

≥ 1 (whichever condition is met first) > 2000 (> 610) 82–250 (28–121) > MMP and < Pf ≥3 ≥5 27–45 1000 ft, i.e., 305 m) and of a certain thickness (> 1.5 ft, i.e., 0.52 m) (Frailey et al., 2006). At intermediate depths (500–1000 ft, i.e.152–305 m), only coals less than 3.5 ft (1.1 m) and greater than 1.5 ft (0.52 m) in thickness were considered as potential targets for CO2 storage. Another consideration with regard to CO2 storage in coal beds is that the process of adsorption applies to CO2 in gaseous phase, i.e., at depths for which temperature is less than 31 °C and pressure is less than 7.38 MPa (738 m hydrostatic column). Storage of supercritical CO2 is less understood currently and it is not clear if no free phase CO2 will exist in the coal. Another



43

limiting factor in regard to CO2 storage in coal beds is the fact that shallow groundwater resources should be protected, and these extend sometimes at depths of several hundred meters, thus eliminating from consideration shallower coals. Because coal swells in the presence of CO2, with the net effect of reducing permeability by approximately one order of magnitude, coals need to have permeability of at least 1 mD and higher in order to ensure injectivity. In light of the fact that coal permeability decreases with increasing depth as a result of increasing effective stress, the requirement of adequate permeability has the net result of restricting the depth of coals that can be used for CO2 storage at no more than approximately 1000 m. Between the requirements of adequate permeability, protection of groundwater resources and protection of coal as an energy resource, not too much coal is left for CO2 storage. Finally, CO2 storage in coal beds requires an extensive infrastructure comprising a high density of injection wells with the associated CO2 distribution pipeline system, and an equally high density of producing wells and gathering pipelines for the coalbed methane that will be produced as a result of CO2 storage. Since methane is also a greenhouse gas, actually stronger than CO2, it has to be captured and used. The existence and/or costs of such an infrastructure will certainly put CO2 storage in coal beds at a disadvantage in comparison to CO2 storage in deep saline aquifers and hydrocarbon reservoirs.

2.3

Site characterisation

Site characterisation is basically the process by which data, information and knowledge are acquired and processed to provide satisfactory answers to the question: does the site meet the site selection criteria? A more detailed definition is provided by Cook (2006) as: ‘the collection, analysis and interpretation of subsurface, surface and atmospheric data (geoscientific, spatial, engineering, social, economic, environmental) and application of that knowledge to judge, with a degree of confidence, if an identified site will geologically store a specific amount of CO2 for a defined period of time and meet all required health, safety, environmental and regulatory standards’. Site characterisation both predates and follows site selection. Sites should be sufficiently characterised initially to be able to judge them on the basis of site selection criteria and, once selected, further characterisation is needed to demonstrate site performance, including monitoring. Performance predictions are based on numerical simulations of injection and of the various processes that take place in the storage unit. The performance modelling also relies on data used for site selection, but needs additional data and at increased resolution and detail. Another category of site characterisation activities consists of characterisation of risk at the storage site, and this characterisation


44


usually focuses on the overlying strata and their structure and geology, other resources, shallow groundwater, vegetation, animal life and population, marine (if offshore), and atmospheric conditions, etc. Finally, additional site characterisation should be performed for capturing the baseline conditions at the site prior to the start of CO2 injection on which basis a monitoring, measurement and verification (MMV) programme can be designed and implemented. Thus, the aims of site characterisation are: 1. refine storage capacity estimates, therefore confirming capacity requirements; 2. predict the extent of the plume of injected CO2 and the effects of CO2 in free phase or dissolved in aquifer water in reservoir and seal rocks; 3. ascertain that, as far as it can be discerned prior to injection, the site will perform effectively and safely; 4. establish the baseline conditions for the design and implementation of a monitoring programme; 5. assess the risk associated with the storage operation and remediation strategies in case of site non-performance; 6. complete the material necessary for application and permitting of the site. To achieve the aims of site characterisation, the following are recommended: ∑ ∑ ∑ ∑ ∑ ∑ ∑ ∑

geological characterisation; characterisation of reservoir rock properties; characterisation of caprock properties; predictive flow modelling; geochemical assessment; geomechanical assessment; risk assessment; design of monitoring programme.

Geological characterisation The geological characterisation is a prerequisite for predictive fluid flow and geochemical simulations, risk assessment and design of a monitoring programme. The geological characterisation of the storage reservoir and its overburden aims to describe the geometry and lithology or the sedimentary succession comprising the storage unit, the immediately confining caprock and the overburden above. Pressure and temperature distributions are essential in establishing the in situ properties of CO2 such as density and viscosity, which affect storage capacity, injectivity and sweep efficiency. Reservoir top structure, thickness and compartmentalisation by faults and/or depositional



45

and/or diagenetic processes (i.e., lateral extent) are the minimum information that should be produced using geophysical and well data. Properly mapping the aquifer top is critical in the case of open aquifers because it is essential in establishing plume evolution and migration pathway along the top of the aquifer. Properties of the storage unit These properties are derived from well logs, core and cuttings, and geophysical lines. The main rock properties are aquifer porosity, permeability and relative permeability for CO2/brine systems at in situ conditions (Bachu and Bennion, 2008), facies type and distribution (which control, in turn, porosity and permeability), shale fraction and geometry of shale bodies, net-to-gross thickness and mineralogy, which is critical for geochemical evaluations of the effects and fate of the injected CO2. Analysis of core should meet the needs of predictive and geochemical modelling, but most likely will include: ∑ sedimentology (optical and scanning electron microscopy); ∑ mineralogy (X-ray diffraction, particle size analysis); ∑ petrophysical properties (porosity, permeability, relative permeability); ∑ mechanical and thermal properties; ∑ acoustic and elastic properties of rocks, such as Young’s modulus and Poisson’s ratio; ∑ pore water chemistry. Caprock and overburden properties The evaluation of the caprock is key in establishing the long-term safety of the storage site. The presence of additional aquifers and sealing units is of considerable interest as it provides the possibility of early warnings in case of CO2 leakage via seismically-imaged spots, changes in groundwater chemistry or even changes in gravity values. The capillary entry pressure is critical because it is a limiting factor in regard to injection pressures and the underlying CO2 plume that it can sustain. A laboratory programme is necessary to test the capillary entry and breakthrough pressure, mineralogy and geochemical composition of the caprock. Caprock core analysis should include: ∑ ∑ ∑

sedimentology (optical and scanning electron microscopy); mineralogy (X-ray diffraction, particle size analysis, cation exchange capacity – CEC, total organic carbon – TOC); petrophysical properties (capillary entry pressure, porosity, permeability and relative permeability);


46


∑

mechanical and thermal properties (Mohr–Coulomb behaviour, drained bulk modulus, time-dependent creep, dehydration); ∑ acoustic and elastic properties, e.g., Young’s modulus and poisson’s ratio; ∑ pore water chemistry. Geochemical assessment

Once dissolved in aquifer water, CO2 forms a weak carbonic acid which may potentially attack and alter rocks, cements and well casings that come in contact with it. The assessment of geochemical impact is thus essential in assessing the safety of a CO2 storage site. The degree of reactivity between CO2, pore water and rock minerals will influence the long-term storage potential and safety of the reservoir. The geochemical assessment of a prospective storage site should include modelling of short-term geochemical reactions, based on the initial characterisation of the system and constrained by laboratory experiments used to calibrate the predictive geochemical modelling, and long-term predictive modelling to assess the geochemical impact of the injected CO2 over hundreds to thousands of years (Chadwick et al., 2008). Modelling results are crucially dependent on which reactions are taken into account and their underpinning chemical data. The output will depend also on the chosen conceptual model. Geomechanical assessment Examples of such assessment are stress regime (stress magnitude and orientation), fault and fracture characteristics, shear strength, activation of preexisting faults and fractures as a result of pressure buildup, and caprock weakening in hydrocarbon reservoirs as a result of pressure decline due to production. Predictive flow modelling This is a key element in site characterisation because it helps in refining storage capacity estimates and provides a means to evaluate the pressure buildup (e.g., Mathias et al., 2009; Birkholzer and Zhou, 2009) and likely lateral spread of CO2 during injection and in the future (plume footprint), which are essential in identifying possible leakage scenarios, design of a monitoring programme and application and permitting. Key modelling parameters include: ∑

reservoir characteristics: geometry, temperature, pressure, porosity, permeability, relative permeability and capillary pressure;



47

∑

caprock characteristics: thickness, permeability, capillary entry pressures; ∑ fluids: water salinity, CO2 composition (type and amount of impurities), phase behaviour.

Risk assessment This type of characterisation is important for assessing the potential risks to the environment and life posed by the CO2 storage project. In the evaluation of consequences versus environmental criteria, the latter must correspond to amounts or concentrations that are measurable, and acceptable levels and limit values must be determined. To determine site-specific criteria, it is necessary to know the baseline conditions, such as groundwater chemistry and ecosystem composition. It is important to note that authorities are responsible for setting requirements, environmental criteria and limit values, but input from industry, the public and other stakeholders is important in the development and determination of acceptable levels and limits.

2.4

Estimation of carbon dioxide (co2) storage capacity

Static and dynamic methods are used in the oil and gas industry, underground natural gas storage, groundwater and underground disposal of fluids for estimating subsurface volumes. Static methods are volumetric and compressibility-based. The dynamic methods are decline/incline curve analysis, material balance and reservoir simulations, and they can be applied only after the start of active injection; hence only static methods will be summarily described here based on the work of Bachu et al. (2007).

2.4.1 Coal beds The effective storage capacity MCO2e for a given coal bed is given by the relation:

Ê ˆ M CO2e = E ¥ rCO2s ¥ A ¥ h ¥ nC ¥ ÁVL ¥ P ˜ ¥ (1 – fa – fm ) Ë P + PL ¯

[2.1]

where rCO2s = 1.873 kg/m3 is CO2 density at standard (surface) conditions, A and h are the area and effective thickness of the coal zone, respectively, ñC is the bulk coal density (generally ñC ≈1.4 t/m3), P is pressure, PL and VL are Langmuir pressure and volume, respectively, fa and fm are the ash


48


and moisture weight fraction of the coal, respectively, and E is the storage efficiency factor. Monte Carlo simulations performed by USDOE (2007) produced a range for E between 28 and 40 % for a 15–85 % confidence range, with an average of 33 % for 50 % confidence. The coal gas saturation (in units of volume of gas per unit of coal mass) is given by GCS and is the coal gas content (dry, ash free) at saturation, assuming that the coal will be 100 % saturated with CO2. It is generally assumed to follow a pressure-dependent Langmuir isotherm of the form:

GCS = VL ¥

P P + PL

[2.2]

2.4.2 Oil and gas reservoirs Assessments for CO2 storage capacity in oil and gas reservoirs are based on reserves databases that list hydrocarbon reserves and various reservoir characteristics. Solution gas should not be considered in storage capacity calculations because it is implicitly taken into account in oil reservoirs through the shrinkage factor. Since reserves databases indicate the volume of original gas and oil in place (OGIP and OOIP) at surface conditions, the theoretical mass storage capacity for CO2 storage in a reservoir at in situ conditions, MCO2t, is given by:

MCO2t = rCO2r ¥ Rf ¥ (1 – FIG) ¥ OGIP

¥ [(Ps ¥ Zr ¥ Tr)/(Pr ¥ Zs ¥ Ts)]

[2.3]

for gas reservoirs, and by:

MCO2t = rCO2r ¥ [Rf ¥ OOIP ¥ Bf – Viw + Vpw]

[2.4]

for oil reservoirs. An alternate equation for calculating the CO2 storage capacity in oil and gas reservoirs is based on the geometry of the reservoir (areal extent and thickness) as given in reserves databases:

MCO2t = rCO2r ¥ [Rf ¥ A ¥ h ¥ f ¥ (1 – Sw) – Viw + Vpw]

[2.5]

In the above equations rCO2r is CO2 density at reservoir conditions of temperature and pressure, calculated from equations of state. OGIP and OOIP are the initial gas and oil in place, respectively, Rf is the recovery factor, FIG is the fraction of injected gas, P, T and Z denote pressure, temperature and the gas compressibility factor, respectively, Bf is the formation volume factor that brings the oil volume from standard conditions to in situ conditions, Viw and Vpw are the volumes of injected and produced water, respectively (applicable in the case of oil reservoirs), and A, h, f and Sw are reservoir area, thickness, porosity and water saturation, respectively. If gas or miscible



49

solvent is injected in oil reservoirs during tertiary recovery, then the mass balance of these should be added to Equations 2.4 and 2.5. The subscripts ‘r’ and ‘s’ denote reservoir and surface conditions, respectively. The volumes of injected and/or produced water, solvent or gas can be calculated from production records. In the case of reservoirs with strong aquifer support (water drive), the volumes of injected and produced water may be negligible by comparison with the amount of invading water. In the case of reservoirs underlain by aquifers, the reservoir fluid (oil and/ or gas) was originally in hydrodynamic equilibrium with the aquifer water. As hydrocarbons are produced and the pressure in the reservoir declines, a pressure differential is created that drives aquifer water up into the reservoir, invading the reservoir. If CO2 is then injected into the reservoir, the pore space invaded by water may not all be available for CO2 storage, resulting in a net reduction of reservoir capacity. The pore volume invaded by water from underlying aquifers cannot be estimated without detailed monitoring of the oil–water interface and detailed knowledge of reservoir characteristics. As CO2 is injected and pressure increases, some of the invading water may be expelled back into the aquifer; however, the hysteresis caused by relative permeability effects and irreducible saturations will prevent complete withdrawal of invaded water, leading to a permanent loss of storage space. Notwithstanding the effect of an underlying aquifer, three other factors control the effectiveness of the CO2 storage process: CO2 mobility with respect to oil and water; the density contrast between CO2 and reservoir oil and water which leads to gravity segregation; and reservoir heterogeneity. All these processes and reservoir characteristics that reduce the actual volume available for CO2 storage can be expressed by capacity coefficients (C < 1) in the form:

MCO2e = Cm ¥ Cb ¥ Ch ¥ Cw ¥ Ca ¥ MCO2t ∫ E ¥ MCO2t

[2.6]

where MCO2e is the effective reservoir capacity for CO2 storage, the subscripts ‘m’, ‘b’, ‘h’, ‘w’ and ‘a’ stand for mobility, buoyancy, heterogeneity, water saturation and aquifer strength, respectively, and E is the storage efficiency coefficient that incorporates the cumulative effects of all the others. In the cases where good production (and injection) records are available, and particularly when cumulative production is greater than the estimated original oil or gas in place, a production-based method can be used to estimate CO2 storage capacity, in which basically the product Rf ¥ OGIP or Rf ¥ OOIP in Equations 2.3 and 2.4 is replaced by the produced gas or oil, respectively. In some cases, more oil or gas has been produced than originally estimated would be recoverable, resulting in a real Rf > 1. Cumulative production data should be used whenever possible to check and update the real Rf.


50


2.4.3 Deep saline aquifers Storing CO2 in water-saturated structural and stratigraphic traps is similar to storage in depleted oil and gas reservoirs. If the geometric volume (Vtrap) of the structural or stratigraphic trap down to the spill point is known, as well as its porosity (f) and the irreducible water saturation (Swirr), then the theoretical volume available for CO2 storage (VCO2t) can be calculated with the formula:

VCO2t = Vtrap ¥ f ¥ (1 – Swirr) ∫ A ¥ h ¥ f ¥ (1 – Swirr)

[2.7]

where A and h are the trap area and average thickness, respectively. This volume is time-independent, and depends on trap characteristics alone. Relation (2.7) assumes constant porosity and irreducible water saturation, and is applicable when average or characteristic values are used. The effective storage volume (VCO2e) is given by:

VCO2e = E ¥ VCO2t

[2.8]

where E is a storage efficiency coefficient that incorporates the cumulative effects of trap heterogeneity, CO2 buoyancy and sweep efficiency. Calculating the mass of CO2 that corresponds to the effective storage volume is more difficult because CO2 density (rCO2) depends on the pressure in the trap once it is filled with CO2. This pressure is not known a priori but depends on permeability, relative permeability to formation water and CO2, dimensions and volume, and the nature of trap boundaries, and may vary with the injection strategy (injection rate, number and/or inclination of injection wells, etc.). However, this pressure has to be higher than the initial water pressure in the trap (Pi) in order to achieve CO2 injection, but it has to be lower than the maximum bottomhole injection pressure (Pmax) that regulatory agencies usually impose in order to avoid rock fracturing or breaching of the capillary seal. Thus, the mass of CO2 that would be stored in a structural or stratigraphic trap would be between these two limits (Bachu et al., 2007):

minMCO2e = rCO2(Pi, T) ¥ VCO2e ≤ MCO2e ≤ maxMCO2e

= rCO2(Pmax, T) ¥ VCO2e

[2.9]

where T is the average temperature in the trap. The mass capacity of a trap may vary in time if pressure varies because, although the volume of the trap remains constant, CO2 density varies with varying pressure. Relations 2.7 –2.9 can also be applied to the case of a plume of CO2 that is not necessarily contained in a stratigraphic or structural trap, but the area and thickness of the CO2 plume have to be known a priori through numerical simulations. USDOE (2007) proposes to use the entire volume of the aquifer according to the following relation for calculating the volumetric CO2 storage capacity:



MCO2e = E ¥ A ¥ h ¥ f ¥ rCO2

51

[2.10]

where rCO2 is the average CO2 density evaluated at pressure and temperature that represent storage conditions anticipated for a specific deep saline aquifer, and E is a storage efficiency factor that reflects the total pore volume filled with CO2. Application of Relation 2.10 may be difficult for aquifers of large areal extent and great variability, in which case an integral form of Relation 2.10 should be used:

M CO2e = E ¥

ÚÚfr

CO2 d Ad h

[2.11]

The effect of irreducible water saturation is not taken into account explicitly in Relation 2.10 as in Relation 2.7, but is included in the efficiency factor E through the pore-scale displacement efficiency. Monte Carlo simulations performed by members of the project team produced a range for E between 1 and 4 % of the bulk volume of a deep saline aquifer for a 15–85 % confidence range, with an average of 2.4 % for 50 % confidence (USDOE, 2007). More recent work shows that the efficiency factor E varies, depending on aquifer lithology, between 1.4 % for limestone aquifers with a 10 % confidence and 6 % for sandstone aquifers with a 90 % confidence (Gorecki et al., 2009).

2.5

Future trends

Future trends in site selection and characterisation depend on progress in the science and economics of CO2 geological storage. As CO2 capture becomes more widespread and CO2 becomes available in greater volumes and at lower costs, and as infrastructure develops, particularly pipelines, the economic criteria for site selection will probably change, with more sites meeting selection criteria. Similarly, as CO2 capture and storage gains wider public acceptance as part of a portfolio of measures for mitigating climate change, some societal screening criteria will likely change. On the other hand, as more and more projects are implemented, and as the regulatory framework for site selection, permitting, operation and abandonment evolves, the criteria for screening and the requirements for characterisation will also evolve, with the potential to become more stringent or less so, depending on the experience gained to date. With regard to CO2 storage in coal beds, as the technology matures and as uneconomic coal beds will be better defined, the requirements for screening and characterisation will also improve. Similarly for deep saline aquifers, as flow and geochemical processes become better understood, most likely the need for better characterisation will increase (e.g., the emergence of relative permeability as a key factor in both capacity estimates and CO2 plume evolution). Other media, such as organic-rich shales, basalts, ultramaphic


52


rocks such as olivine and serpentine, and hydrates, may prove in the mediumto-long term to be suitable for CO2 storage, in which case a new set of screening criteria and characterisation requirements will emerge. Finally, site characterisation will develop significantly beyond site selection, to meet regulatory requirements for permitting, and also for monitoring the fate and effects of the injected CO2. With regard to performance assessment, as the science of CO2 storage will advance and as predictive models improve, the need for characterisation at increasingly higher resolution/smaller scales will increase. Characterisation will continue past the operational stage of a CO2 storage project into the closure and post-closure phases. To conclude, screening criteria for site selection will evolve as the science of CO2 storage advances, as more field experience is gained, and as regulatory frameworks are implemented. Site characterisation will also expand to cover more aspects at increasing resolution and detail and for longer periods of time, basically covering the entire lifetime of a CO2 storage project, from inception to complete abandonment.

2.6


Generally papers written before 2005 presented cases of site characterisation specific to the CO2 storage projects active or in implementation phase up to that time; these have been reviewed in the IPCC Special Report on CO2 Capture and Storage (Benson et al., 2005). Many cases of and various aspects of site characterisation have been presented at the International Conferences on Greenhouse Gas Technologies, whose papers have been published in conference proceedings. Environmental Geosciences has published two special issues in 2006 (numbers 2 and 3 of volume 13) and Environmental Geology has published a special issue in June 2008 (volume 54, number 8) devoted to the subject of site characterisation. Worth mentioning are the papers describing site-specific cases of site characterisation in North America at the Frio, Mountaineer and Tea Pot Dome sites (Friedmann and Stamp, 2006; Hovorka et al., 2006; Lucier et al., 2006; Chiaramonte et al., 2008; Doughty et al., 2008; Lucier and Zoback, 2008), in Europe (Förster et al., 2006; Meyer et al., 2008) and Australia (Sayers et al., 2006; Gibson-Poole et al., 2008). Several papers of interest have been published in the International Journal of Greenhouse Gas Control. In addition, the American Association of Petroleum Geologists (AAPG) published in 2009 a special volume on site characterisation for CO2 storage (AAPG Studies #59: Carbon Dioxide Sequestration in Geological Media – State of the Art). The American Geophysical Union (AGU) also published in 2009 a special monograph: The Science of CO2 Storage, Geophysical Monograph Series GM148. Currently, every major professional organisation, such as European Association of Geophysicists and Engineers (EAGE), American



53

Geophysical Union (AGU), Society of Petroleum Engineers (SPE), American Chemical Society (ACS), American Association of Petroleum Geologists (AAPG), holds annual conferences and various other meetings where one or more sessions are devoted to CO2 storage in geological media, or even special meetings dedicated only to CO2 capture and storage, like the SPE International Conference on CO2 Capture, Storage and Utilization held in San Diego, California, November 2–4, 2009.

2.7

References

Bachu S (2000) Sequestration of carbon dioxide in geological media: criteria and approach for site selection. Energy Convers. Manage., 41(9), 953–970. Bachu S (2001) From suitability to ultimate capacity: a roadmap for assessing sedimentary basins and selecting sites for CO2 storage in geological media. In: Williams D J, Durie R A, McMullan P, Paulson C A J and Smith A Y (eds), Proceedings of the Fifth International Conference on Greenhouse Gas Control Technologies (GHGT5), Collingwood, VIC, Australia, CSIRO Publishing, 328–333. Bachu S (2002) Sequestration of CO2 in geological media in response to climate change: roadmap for site selection using the transform of the geological space into the CO2–phase space. Energy Convers. Manage., 43, 87–102. Bachu S (2003) Screening and ranking of sedimentary basins for sequestration of CO2 in geological media. Environ. Geol., 44(3), 277–289. Bachu S and Bennion DB (2008) Drainage and imbibition relative permeability relationships for supercritical CO2 and H2S/brine systems in intergranular sandstone, carbonate, shale and anhydrite formations. SPE Res. Eval. Eng., 11(3), 487–496. Bachu S and Gunter WD (1999) Storage capacity of CO2 in geological media in sedimentary basins with application to the Alberta basin. In: Riemer P, Eliasson B and Wokaun A (eds), Proceedings of the Fourth International Conference on Greenhouse Gas Control Technologies (GHGT4), Amsterdam, the Netherlands, Pergamon, 195–200. Bachu S and Shaw JC (2003) Evaluation of the CO2 sequestration capacity in Alberta’s oil and gas reservoirs at depletion and the effect of underlying aquifers. J. Can. Pet. Technol., 42(9), 51–61. Bachu S and Shaw JC (2005) CO2 storage in oil and gas reservoirs in western Canada: effect of aquifers, potential for CO2-flood enhanced oil recovery, and practical capacity. In: Rubin E S, Keith D W and Gilboy C F (eds), Proceedings of the Seventh International Conference on Greenhouse Gas Control Technologies: GHGT7, Cheltenham, UK, IEA GHG, Vol. 1, 361–369. Bachu S, Bonijoly D, Bradshaw J, Burruss R, Holloway S, Christensen NP and Mathiassen OP (2007) CO2 storage capacity estimation: Methodology and gaps. Int. J. Greenhouse Gas Control, 1(4), 430–443. Benson S and Cook P et al. (2005) Transport of CO2. In: Metz B, Davidson O, de Coninck H, Loos M and Meyer L (eds), IPCC Special Report on Carbon Dioxide Capture and Storage, Cambridge, UK, Cambridge University Press, 197–276. Birkholzer JT and Zhou Q (2009) Basin-scale hydrogeologic impacts of CO2 storage: capacity and regulatory implications. Int. J. Greenhouse Gas Control, 3(6), 745–756. Chadwick A, Arts R, Bernstone C, May F, Thibeau S and Zweigel P (2008) Best Practice for the CO2 Storage in Saline Aquifers – Observations and Guidelines from the SACS and CO2STORE Projects. Nottingham, UK, British Geological Survey.


54


Chiaramonte L, Zoback MD, Friedmann J and Stamp V (2008) Seal integrity and feasibility of CO2 sequestration in the Teapot Dome EOR pilot: geomechanical site characterisation. Environ. Geol., 54(8), 1667–1675. Cook PJ (2006) Site characterisation. In: International Symposium on Site Characterisation for CO2 Storage, 20–22 March 2006, Berkeley, CA, Lawrence Berkeley National Laboratory, 3–5. CSLF (Carbon Sequestration Leadership Forum), (2007) Estimation of CO2 storage capacity in geological media, June, available at: http://www.cslforum.org/publications/ documents/PhaseIIReportStorageCapacityMeasurementTaskForce.pdf (accessed January 2010). Doughty C, Freifeld BM and Trautz RC (2008) Site characterisation for CO2 geologic storage and vice-versa: the Frio brine pilot, Texas, USA as a case study. Environ. Geol., 54(8), 1635–1656. Förster A, Norden B, Zinck-Jørgensen K, Frykman P, Kulenkampff J, Spangenberg E, Erzinger J, Cosma C-G and Hurter S (2006) Baseline characterisation of the CO2SINK geological storage site at Ketzin, Germany. Environ. Geosci., 13(3), 145–161. Frailey SM, Rittenhouse S, Knepp R, Anderson A, Leetaru H, Korse C, Rupp R and Finley RJ (2006) Reservoir simulation and GIS modeling as tools for defining geological sequestration capacity: An optimal approach toward more specific assessments, Illinois basin, USA. In: Gale J, Rokke N, Zweigel P and Svenson H (eds), Proceedings of the Eighth International Conference on Greenhouse Gas Control Technologies: GHGT8, Oxford, UK, Elsevier, CD-ROM. Friedmann SJ and Stamp VW (2006) Teapot Dome: characterisation of a CO2-enhanced oil recovery and storage site in Eastern Wyoming. Environ. Geosci., 13(3), 181–199. Ghaderi SM, Keith DW and Leonenko Y (2009) Feasibility of injecting large volumes of CO2 into aquifers. In: Gale J, Herzog H and Braitsch J (eds), Greenhouse Gas Control Technologies 9, Proceedings of the Ninth International Conference on Greenhouse Gas Control Technologies (GHGT9), Energy Procedia, 1, 3113–3120. Gibson-Poole CM, Svendsen L, Underschultz J, Watson MN, Ennis-King J, van Ruth PJ, Nelson EJ, Daniel RF and Cinar Y (2008) Site characterisation of a basin-scale CO2 geological storage system: Gippsland Basin, southeast Australia. Environ. Geol., 54(8), 1583–1606. Gorecki CD, Sorensen JA, bremer JM, Knudsen DJ, Smith SA, Steadman EN and Harju JA (2009) Development of storage coefficients for determining the effective CO2 storage resource in deep saline formations. SPE International Conference on CO2 Capture, Storage and Utilization, San Diego, CA, 2–4 November, SPE Paper 126444. Holtz MH (2009) Geologic CO2 storage in oil fields: Considerations for successful sites. SPE International Conference on CO2 Capture, Storage and Utilization, San Diego, CA, 2–4 November, SPE Paper 126198. Hovorka SD, Benson SM, Doughty C, Freifeld BM, Sakurai S, Daley TM, Khaaraka Y, Holtz MH, Trautz RC, Seay Nance H, Myer LR and Knauss KG (2006) Measuring permanence of CO2 storage in saline formations: the Frio experiment. Environ. Geosci., 13(2), 105–122. IEA (2004) Prospects for CO2 Capture and Storage. IEA/OECD, Paris, France. Kaldi JG and Gibson-Poole CM (eds), (2008) Storage Capacity Estimation, Site Selection and Characterisation for CO2 Storage Projects. Report No: RPT08-1001, Canberra, ACT, AU, CO2CRC. Kopp A, Class H and Helmig R (2009a) Investigations on CO2 storage capacity in saline aquifers – Part 1: Dimensional analysis of flow processes and reservoir characteristics. Int. J. Greenhouse Gas Control, 3(3), 263–276. © Woodhead Publishing Limited, 2010


55

Kopp A, Class H and Helmig R (2009b) Investigations on CO2 storage capacity in saline aquifers – Part 2: Estimation of storage capacity coefficients. Int. J. Greenhouse Gas Control, 3(3), 277–287. Li S, Dong M, Li Z and Huang S (2006) CO2 sequestration in depleted oil and gas reservoirs – caprock characterization and storage capacity. Energy Convers. Manage., 47(11–12), 1372–1382. Lucier A and Zoback M (2008) Assessing the economic feasibility of regional deep saline aquifer CO2 injection and storage: a geomechanics-based workflow applied to the Rose Run sandstone in eastern Ohio, USA. Int. J. Greenhouse Gas Control, 2(2), 248–258. Lucier A, Zoback M, Gupta N and Ramakrishnan TS (2006) Geomechanical aspects of CO2 sequestration in a deep saline reservoir in the Ohio River Valley region. Environ. Geosci., 13(2), 85–103. Mathias SA, Hardisty PE, Trudell MR and Zimmerman RW (2009) Screening and selection of sites for CO2 sequestration based on pressure buildup. Int. J. Greenhouse Gas Control, 3(5), 577–585. Meyer R, May F, Müller C, Geel K and Bernstone C (2008) Regional search, selection and geological characterisation of a large anticlinal structure, as a candidate site for CO2 storage in northern Germany. Environ. Geol., 54(8), 1607–1618. Moritis G (2008) Special report: More US EOR projects start but EOR production continues to decline. Oil & Gas J., 106(15), 41–42, 44–46. Nordbotten JM, Celia MA and Bachu S (2005) Semi-analytical solution for CO2 leakage through an abandoned well. Environ. Sci. Technol., 39(2), 602–611. Nordbotten JM, Kavetski D, Celia MA and Bachu S (2008) A model for CO2 leakage including multiple geological layers and multiple leaky wells. Environ. Sci. Technol., 43(3), 743–749. Núñez-López V, Holtz MH, Wood DJ, Ambrose WA and Hovorka SD (2008) Quicklook assessment to identify optimal CO2 EOR storage sites. Environ. Geol., 54(8), 1695–1706. Oldenburg CM (2008) Screening and ranking framework for geologic CO2 storage site selection on the basis of health, safety, and environmental risk. Environ. Geol., 54(8), 1687–1694. Oruganti YD and Bryant SL (2009) Effect of relative permeability on pressure-induced risk during CO2 injection in aquifers. SPE International Conference on CO2 Capture, Storage and Utilization, San Diego, CA, 2–4 November, SPE Paper 126932. Pooladi-Darvish M, Hong H, Theys S, Stocker R, Bachu S and Dashtgard S (2008) CO2 injection for enhanced gas recovery and geological storage of CO2 in the Long Coulee Glauconite F Pool, Alberta. SPE Annual Technical Conference and Exhibition, Denver, CO, USA, September 21–24, SPE Paper 115789. Ramírez A, Hagedoorn S, Kramers L, Wildenborg T and Hendricks C (2009) Screening storage options in the Netherlands. In: Gale J, Herzog H and Braitsch J (eds), Greenhouse Gas Control Technologies 9, Proceedings of the Ninth International Conference on Greenhouse Gas Control Technologies (GHGT9), Energy Procedia, 1, 2801–2808. Sayers J, Marsh C, Scott A, Cinar Y, Bradshaw J, Henning A, Barclay S and Daniel R (2006) Assessment of a potential storage site for carbon dioxide: A case study, southeast Queensland, Australia. Environ. Geosci., 13(2), 123–142. Shah V, Broseta D, Mouronval G and Montel F (2008) Water/acid gas interfacial tensions and their impact on acid gas geological storage. Int. J. Greenhouse Gas Control, 2(4), 595–604.


56


Shaw JC and Bachu S (2002) Screening, evaluation and ranking of oil reservoirs suitable for CO2-flood EOR and carbon dioxide sequestration. J. Can. Pet. Technol., 41(9), 51–61. USDOE (U.S. Department of Energy, Office of Fossil Energy), 2007. Carbon Sequestration Atlas of United States and Canada. USDOE (U.S. Department of Energy, Office of Fossil Energy), 2008. Carbon Sequestration Atlas of United States and Canada, 2nd edn. van der Meer LGH and Egberts PJP (2008) A general method for calculating subsurface CO2 storage capacity. Offshore Technology Conference, Houston, TX, USA, 5–8 May, OTC Paper 19309. Wilkinson J, Szafranski R, Lee K-S and Kratzing C (2009) Subsurface design considerations for carbon dioxide storage. In: Gale J, Herzog H and Braitsch J (eds), Greenhouse Gas Control Technologies 9, Proceedings of the Ninth International Conference on Greenhouse Gas Control Technologies (GHGT9), Energy Procedia, 1, 3047–3054.


3

Carbon dioxide (CO2) sequestration in deep saline aquifers and formations

R. J. R o s e n b a u e r and B. T h o m a s, US Geological Survey, USA Abstract: Carbon dioxide (CO2) capture and sequestration in geologic media is one among many emerging strategies to reduce atmospheric emissions of anthropogenic CO2. This chapter looks at the potential of deep saline aquifers – based on their capacity and close proximity to large point sources of CO2 – as repositories for the geologic sequestration of CO2. The petrochemical characteristics which impact on the suitability of saline aquifers for CO2 sequestration and the role of coupled geochemical transport models and numerical tools in evaluating site feasibility are also examined. The full-scale commercial CO2 sequestration project at Sleipner is described together with ongoing pilot and demonstration projects. Key words: CO2 storage, CO2 sequestration, saline aquifer, saline formation, geologic storage.

3.1

Inroduction

The carbon cycle is dominated by steady-state reactions that involve carbon dioxide (CO2). Although the climate-stabilizing exchange of carbon among its various reservoirs is geologically rapid, the anthropogenic combustion of fossil fuels has caused a continuous and rapid > 35 % increase of atmospheric CO2 over the past 150 years (Keeling and Whorf, 2005). This increase of CO2, a major greenhouse gas, is expected to affect global climate (Stocker and Schmittner, 1997; Allen et al., 2000; Cox et al., 2000; Falkowski et al., 2000; Albritton and Meira Filho, 2001; Cuffey and Vimeux, 2001; Palmer and Räisänen, 2002; IPCC, 2007). Carbon dioxide capture and storage (CCS) represents a key component of a larger portfolio of advanced energy technologies and climate policies needed to mediate the rise in atmospheric CO2 concentration. Geologic CCS systems are specifically designed to remove CO2 from various point sources and safely deposit the CO2 in secure storage sites deep underground. The disposal of CO2 into deep-saline aquifers is one of several possible geologic storage options (Gupta et al., 1999). Other geologic CO2 repositories under consideration are depleted oil field reservoirs and deep unmineable coal seams (Gale and Freund, 2001; Stevens et al., 2001; Damen et al., 2005). These approaches have a value-added economic benefit because CO2 injection into these reservoirs enhances oil recovery (EOR) and coalbed methane recovery 57 © Woodhead Publishing Limited, 2010

58


(ECBM), but these options may pose a higher environmental risk due to reactions of supercritical CO2 with organic matter (Kolak and Burruss, 2006) and caprock seals (Kharaka et al., 2006b). Organic-rich shale may provide an adsorption substrate for CO2 storage similar to CO2 storage in coal seams (Nuttall et al., 2005; Vermylen et al., 2008). Basalt formations (Matter et al., 2005; Goldberg et al., 2008; Rosenbauer and Bischoff, 2009) and ultramafic rocks (Goff and Lackner, 1998; Maroto-Valer et al., 2005) are also under consideration for sequestering CO2 because of their potential to convert CO2 to a solid mineral form. These latter options provide CO2 sequestration alternatives in regions without subsurface reservoirs such as the Pacific northwest in the USA, Hawaii, and island nations such as Iceland where a pilot study is currently underway to inject CO2 into basalt rocks (Alfredssom et al., 2008; Gislason et al., 2008). Successful implementation of any of these approaches requires knowledge of the fundamental science that governs the physical and chemical behavior of CO2 during and after injection into the deep subsurface.

3.2

Saline aquifers

Broadly defined, deep saline aquifers are underground, water-filled strata. They are distributed widely across much of the world (Fig. 3.1 and 3.2) and are important potential CO2 repositories because many are large, generally contiguous reservoirs that may be capable of storing large volumes of CO2 (Hitchon et al., 1999). These aquifers, beneath both continents and oceans, are often highly permeable, allowing high rates of fluid extraction and injection (Marsily, 1986, p.115). Carbonate (limestone or dolomite) or sandstone formations host saline water in pore spaces between rock grains or within fractures and void spaces. Salinity in such aquifers can vary by orders of magnitude from mg/L in relatively fresh aquifers to more than 10 g/L in evaporitic formations (Kharaka and Hanor, 2007). Within sedimentary basins, salinity tends to increase with depth. Salinity heterogeneity develops from lithologic influences, differing groundwater flow regimes, and potential meteoric recharge. While most sedimentary rocks are formed with connate pore water contents that approximate that of seawater, during sediment diagenesis fluids react with silicates, carbonates, and evaporites. In particular, calcite alteration to dolomite decreases Mg and increases Ca fluid contents and albitization of plagioclase results in a decrease of dissolved Na relative to Ca (Davisson and Criss, 1996). Also, at greater depth, high salinities correlate with low (~300 mg/L) carbonate alkalinity (Kharaka and Hanor, 2007) due to precipitation of carbonates in response to high dissolved Ca concentrations. As a result, increasing salinities are consistent with the evolution of Na–Cl connate brines to Ca–Na–Cl brines with low carbonate alkalinity and lower pH (Kharaka and Hanor, 2007).


Carbon dioxide (CO2) sequestration in deep saline aquifers

59

3.1 Map depicting sedimentary basins within North America containing saline aquifers (image courtesy of the US Department of Energy, National Energy Technology Laboratory).

There are more than 800 sedimentary provinces in the world (St John et al., 1984) distributed on and along continents (Fig. 3.2) that are variously well suited for CO2 storage and sequestration (Bachu, 2003; Holtz, 2003; Bradshaw and Dance, 2005a,b). The highest potential for CO2 storage is in large, continental-sized basins, such as the Illinois, and the northern Gulf of Mexico sedimentary basins in USA. In general, active margin sedimentary basins are less favorable for CO2 storage than passive margin basins because deep-seated faults and high-density fracture zones characteristic of active margins increase the likelihood of hydraulic connection with the surface. Tectonic complexity also decreases the lateral extent of formations and limits the exploitable reservoir volume for a given injection well. Rarely exploited for economic resources, saline formations are the largest, but least defined, of the geologic storage possibilities. Considerable


60


Basins on rigid lithosphere

Perisutural basins

Episutural basins

Atlantic-type basins Foredeeps

Pannonian-type basins

Subduction-related folded belts

Chinese-type

Forearc backarc

Trenches

basins

and California-

Cratonic basins

type basins

3.2 Map depicting major sedimentary basins of the world (image based on St John et al., 1984 in Bachu 2003, AAPG©1984, reprinted by the permission of the AAPG whose permission is required for further use).

uncertainty surrounds reservoir properties, flow regimes and capacities of saline aquifers, but the storage capacity is generally considered to be very large and the reservoir properties are likely favorable (Burruss et al., 2009). The evaluation of a formation for sequestration requires a set of variables that balance the total capacity of the reservoir with the likelihood of retention, the ease of injection, and the risks to environment and human health. Relevant physical and chemical formation variables include depth (temperature and pressure), rock properties such as mineralogy and grainsize, pore properties such as such as porosity and permeability, formation properties such as the structure and thickness of the reservoir, the resident fluid composition, and reservoir confinement. These variables are needed to calculate hydrostatic potentials, to create generalized groundwater and CO2 flow maps, and to estimate a net pore volume for CO2 storage. Below, we describe how these variables relate to the suitability of individual formations for CO2 storage.



61

3.2.1 Saline aquifer characteristics Depth The preferred concept is to inject CO2 into a porous and permeable reservoir covered with a caprock located at sufficient depth and pressure where CO2 can be injected and stored under supercritical conditions (31.1 °C and 73.8 bar) (Fig. 3.3). Minimum and maximum estimates for surface temperatures (15–20 °C), geothermal gradients (25–35°/km), and hydrostatic pressure gradients (105–160 bar/km) (Spycher et al., 2003) yield a range of potential injection conditions from above ambient temperature and pressure (TP) up to 150 °C and 800 bar, shown in fig. 3.4 between the dashed lines. Within this projected TP range there co-exist a dense immiscible CO2-rich fluid (Fig. 3.3) and a H2O-rich liquid phase. At all reservoir conditions, CO2 fluids are buoyant relative to aqueous phases and tend to migrate toward the 100 80 60 50 40 30 20

Pressure (MPa)

Solid

Super critical phase

Liquid

10 8 6 5 4 3

Critical point

2 Gas 1.0 .8 .6 .5 .4

Triple point

.3 .2 .1 –100

–80

–60

–40 –20 0 Temperature (°C)

20

40

60

3.3 Phase diagram of CO2 showing the triple and critical points and regions where CO2 is stable as a gas, liquid, solid, or supercritical fluid. The critical point is 31.1 °C and 73.8 bar (based on Bachu 2000; by permission of Elsevier©2000).


62

Developments and innovation in CCS technology 300

Pressure (bar)

250

Injection conditions

200 Supercritical 150 Liquid

100 50

CP

H+L V

0

0

30

60 90 Temperature (°C)

120

150

3.4 Range of potential injection conditions from above ambient temperature and pressure (TP) up to 150 °C and 300 bar based on minimum and maximum estimates for surface temperatures (15–20 °C), geothermal gradients (25–35°/km), and hydrostatic pressure gradients (105–160 bar/km) (Spycher et al., 2003). Projections are extrapolated to earth surface conditions; actual CO2 injection would take place at conditions above the critical point. CP = critical point; H = hydrate; L = Liquid; V = Vapor.

surface. Depending on the local geothermal gradient, a minimum reservoir depth of 800 m (Hitchon, 1996) is typically necessary for CO2 to remain in its supercritical state. This depth may vary, depending on local geothermal and hydrostatic pressure gradients of the formation. Mineralogy and grain size Target saline reservoirs are typically sandy units. These units may have variable clay and silicate mineral contents and be overlain by shale seals. Quartz sand and clays are minimally reactive with CO2 whereas carbonates, in particular, as well as plagioclase feldspar and mafic minerals are more reactive (Gunter et al., 1993, 1997; Kaszuba et al., 2003; Knauss et al., 2005; Rosenbauer et al., 2005). The presence of carbonate cements in shales and sandstones will influence the reactivity with CO2 as described later in the text. In general, larger grain sizes correlate with increased hydraulic conductivity at the expense of residual trapping efficiency (Holtz and Bryant, 2005). Porosity and permeability Permeability controls the direction and rate of CO2 injection and subsequent movement. Porosity and residual saturation control the size, shape, and



63

dispersion of the CO2 plume (Holtz et al., 2004). Porosity in the range 5–30 % (NRG Associates, 2001; Kuuskraa, 2004; Shafeen et al., 2004; Zweigel et al., 2004; Bertier et al., 2006) and permeability in the range 20–3500 md (McRae and Holtz, 1995; Zweigel et al., 2004; Bertier et al., 2006) provide sufficient reservoir volume and injection rates for the CO2 (Bentham and Kirby, 2005). In low-permeability rocks, pressure gradients dissipate slowly and fluid pressure progressively increases near the injection well. The need to avoid compromising the overlying seal may limit the viable injection rate and may limit the accessible pore volume for a given well. Capacity The area extent of worldwide sedimentary basins, not including those offshore, is approximately 70–80 million km2 (Koide et al., 1992; Hendriks and Blok, 1995; Bruant et al., 2002; Ecofys & TNO-NITG, 2002). The vast majority of these sedimentary basins contain deep brines that are variably suitable for sequestration. Worldwide potential capacity estimates of CO2 storage in saline formations range widely from 350 to > 11 000 Gt of CO2 (Omerod, 1994; Bergman and Winter, 1995; IEA, 1995; Holloway, 1997; Beecy and Kuuskraa, 2001; Davison et al., 2001; White et al., 2003; IPCC, 2005, 2007; Bradshaw et al., 2007). The storage volume for saline aquifers is large compared with other geologic repositories (Table 3.1) and capable of storing up to ~1000 years worth of modern CO2 emissions, based on total annual emissions of 31 Gt (IPCC, 2005; Oelkers and Schott, 2005; EIA, 2006). Estimates differ based on assumptions about volumes of sedimentary basins, aquifer characteristics, CO2 storage density, and technological and economic constraints (Bruant et al., 2002). Deficiencies in the accuracy and reliability of these past estimates are due in large part to a lack of common methodology and guidelines. Recent efforts have focused on developing consistent methodologies and specific standards for storage capacity estimation in deep saline aquifers and for the identification, screening, and prioritization of suitable geological basins

Table 3.1 Estimated storage capacities for major geologic storage reservoirs, (Dooley et al., 2006) Sequestration option

Worldwide capacity (Gt CO2)

Deep saline aquifers Depleted oil reservoir Depleted natural gas reservoir Deep coal seams Deep saline basalt formations Organic shales

100s–10 000s 120 700 140 > 240 unknown


64


for CO2 storage (CSLF, 2005, 2007, 2008; Bachu et al., 2007; Bradshaw et al., 2007; USDOE, 2008; Burruss et al., 2009; Vangkilde-Pedersen et al., 2009). Classification terminology from the petroleum industry is combined with a methodology to facilitate dynamic and risk-adjusted storage estimates (Frailey et al., 2006). The resulting probabilistic ranges of storage volumes are analogous to assessments of oil and gas resources (Burruss, 2004; Brennan and Burruss, 2006; Burruss et al., 2009) and allow differentiation of CO2 resource assessments from CO2 capacity estimates (Bachu, et al., 2007; CSLF, 2005, 2007; USDOE, 2008). Carbon dioxide resource estimates include only the technical screening criteria such as the tectonic setting and geology, the basin geothermal regime, and the hydrodynamic regime of formation waters to estimate the volume of porous and permeable sedimentary rocks available and accessible for CO2 storage. Capacity estimates include economic variables related to access, infrastructure and socio-political conditions (CSLF, 2005; Bachu et al., 2007). As with oil and gas resources, knowledge of storage capacity of saline aquifers is inversely proportional to the spatial scale of the assessment. Certainty increases in a progression from country and region to basin, local, and site-specific scales (Bachu et al., 2007). Carbon dioxide storage capacity at basin and regional scales can be evaluated only in the broadest terms. Capacity estimates are more robust at local and site-specific scales (Bachu et al., 2007). Each storage site is therefore unique and requires a specific technical and operational evaluation.

3.3

Trapping mechanisms

A combination of physical and chemical processes causes CO2 to become trapped and stored in saline aquifers. These processes vary in importance from site to site, depending on the reservoir characteristics, and they are effective over varying timescales (IPCC, 2005; Bachu et al., 2007). CO2 can be physically immobilized by structural, stratigraphic, lithologic, and hydrodynamic traps and can be physically bound by pore capillary forces as an immiscible residue. Chemical processes include dissolution of CO2 fluid into the aqueous phase, geochemical acid-base reactions, and mineral dissolution and precipitation (Chadwick et al., 2003; CSLF, 2005). Initially, the main storage processes are structural or stratigraphic traps (Fig. 3.5). Deformed or fractured reservoir strata, such as folded and faulted sandstones, in combination with overlying low permeability strata, such as mudstones, physically contain CO2 (Fig. 3.5b). Low hydraulic conductivity aquitards or aquicludes may be contiguous or a series of multiple interbedded confining layers. Confining layers are crucial because, following injection, the positively buoyant CO2 will follow a pressure gradient migrating from the injection site to other potentially less desirable reservoirs including



65

(a)

1 km + 100s km (b)

3.5 (a) hydrodynamic stratigraphic trapping where CO2 is trapped by an overlying layer of cap rock coupled with impermeable rock within a narrowing of the formation aquifer; (b) structural trapping, where CO2 is trapped by a fold in the rock formations; (c) trapping by impermeable rock layers shifted along a sealing fault to contain the CO2 (image source: CO2CRC).


66


(c)

3.5 Continued

freshwater aquifers. If unconfined, CO2 plumes can reach surface conditions where they can volatilize and vent to the atmosphere.

3.3.1 Hydrodynamic Hydrodynamic trapping refers to a time-dependent hydrogeological process where injected CO2 is effectively trapped by the existence of very long travel times to the surface. Following injection, the CO2 fluid migrates vertically as a plume along preferentially permeable pathways until it encounters a relatively impermeable caprock. It then travels laterally, driven by buoyancy, to structurally higher levels along the caprock–reservoir boundary. This mechanism is particularly effective in laterally unconfined sedimentary basins with limited structural traps, but with large-scale flow systems and low groundwater and fluid flow rates (fig. 3.5a). In contrast to these open and laterally unconfined systems, some sedimentary basins are bounded laterally by low-permeability zones and impervious seals due to natural heterogeneity or faulting (Fig. 3.5c). In such a closed system where the lateral and vertical migration of the fluids is contained, CO 2 injection rapidly increases pore pressure in the vicinity of injection wells. This is a limiting factor affecting CO2 storage capacity because of the need



67

to avoid geomechanical damage of the overlying seals. Highly structurally compartmentalized reservoirs are likely therefore to be less suited to CO2 storage than large unfaulted or high permeability reservoirs (Bentham and Kirby, 2005).

3.3.2 Residual trapping Residual trapping of injected CO2 occurs by capillary forces in the pores of reservoir rocks. It is the result of brine back filling the trailing edge of a mobile plume of supercritical CO2 that has been injected into the formation and has displaced the resident fluid. As buoyant CO2 moves through the formation, aqueous fluid replaces it, but some CO2 is left behind as disconnected, or residual, droplets in the pore spaces (Benson and Cole, 2008). The magnitude of residual saturation within a rock, and hence the fraction of injected CO2 that can be stored in this form, is a function of the rocks pore network geometry (Holtz and Bryant, 2005). It is likely that changes in the aspect ratio (pore radius/pore throat radius) of the pore network due to changes in inter-granular porosity control residual-phase saturation (Holtz and Bryant, 2005). The amount of residual gas trapping is also a function of both the vertical and horizontal extensions of the CO2 plume (Han et al., 2008). The amount of residual-trapped CO2 increases when the injected CO2 plume sweeps a larger area; thus, estimates differ on the amount of CO2 that can be retained by this process. Based on 3D reservoir modeling, capillary trapping ranges from 15–25 % for typical storage formations (Holtz, 2002), but could exceed 25 %, depending upon the porosity and permeability of the formation. Experimental measurements of supercritical CO2 in a Berea sandstone suggest that the maximum residual gas saturation of supercritical CO2 is about 8–10 % (Kitamura and Xue, 2006). Studies by Hesse et al. (2008) and Ide et al. (2007) suggest that, over time, 100 % of the CO2 in a subsurface plume could be immobilized by capillary trapping.

3.3.3 Carbon dioxide (CO2) dissolution Some injected CO2 will dissolve in the saline formation waters as a result of its solubility. The rate of dissolution is initially rapid (Rosenbauer and Koksalan, 2002) but, as the reservoir becomes density stratified, the dissolution becomes diffusion and convection limited. Carbon dioxide saturated formation water is more dense than pure formation water and therefore induces density instability and convective mixing that accelerates CO2 dissolution (Lindeberg and Wessell-Berg, 1997; Xu et al., 2006). Depending on the relative contact lens area and solubility gradient between the CO2 plume and the formation water, saturation of the formation water by the supercritical plume may take several millenia (Ennis-King and Paterson, 2001). Reservoir engineering


68


entailing forced convection by the extraction of formation fluid could accelerate the rate of dissolution trapping (Leonenko and Keith, 2008). The solubility of CO2 is well known in pure water, especially along the saturation curve, and has been measured up to 350 °C and 3500 bar (e.g. Wiebe and Gaddy, 1939, 1941; Malinin, 1959; Todheide and Franck, 1963; Takenouchi and Kennedy, 1964; Drummond, 1981; Blencoe et al., 2001; Rosenbauer et al., 2001; Rosenbauer and Koksalan, 2002; Chapoy et al., 2004). The solubility of water in CO2 has been less studied but becomes increasingly important at elevated temperature (Takenouchi and Kennedy, 1964; King et al., 1992; Spycher et al., 2003). Carbon dioxide solubility in aqueous NaCl solutions has been measured over a broad range of temperatures, pressures, and ionic strengths, but these data are generally confined to conditions relevant to seawater and hydrates (Stewart and Munjal, 1970; Barton and Chou, 1993; Aya et al., 1997) or to hydrothermal solutions and metamorphic processes (Joyce and Holloway, 1993). The solubility of CO2 in both pure water and NaCl solutions has been modeled by several investigators (Duan et al., 1995, 2006; Duan and Sun, 2003; Spycher et al., 2003), but measurements of the solubility of CO2 in saline fluids typical of P–T conditions within deep aquifers are generally lacking (Rosenbauer et al., 2001; Rosenbauer and Koksalan, 2002; Portier and Rochelle, 2005). Duan et al. (2006) present a critical review of the experimental data and a model of the solubility of CO2 in mixed electrolytes. Carbon dioxide dissolution in water produces carbonic acid (equation 3.1), a fundamentally important reaction because it is generally the aqueous (HCO3–, CO3–2), not the neutral molecular form of CO2, that reacts aquifer minerals (Rosenbauer and Koksalan, 2003).

CO2 + H2O ´ H2CO3

[3.1]

The aqueous solubility of molecular CO2 is temperature-, pressure-, and ionic-strength-dependent, generally lower at elevated temperature and salinity and greater at elevated pressure (Takenouchi and Kennedy, 1964; Rosenbauer and Koksalan, 2002; Duan and Sun, 2003). For example, at 25 °C and 1 bar, the solubility of CO2 in an aquifer fluid equivalent to a 4 M NaCl solution is approximately one-third its solubility in pure water (Rosenbauer and Koksalan, 2002) (Fig. 3.6). At conditions consistent with deep down-hole disposal of 120 °C and 300 bars, the solubility of CO2 in a 4 m NaCl equivalent aquifer fluid is 0.5 m (Rosenbauer and Koksalan, 2002; Duan and Sun, 2003). Isotherms of the solubility of CO2 in NaCl exhibit two distinct patterns of solubility versus pressure. At 21° and 50 °C, the solubility of CO2 increases as a nearly linear function of pressure (Rosenbauer, unpublished data; Fig. 3.6). At and above 100 °C, the solubility of CO2 is markedly low at low pressure and increases as a 2nd to 3rd order polynomial-shaped function of pressure,



69

2 Pure water

1.5

21 °C

CO2 (m)

50 °C 1

100 °C

NaCl 22 wt %

150 °C 0.5

0 0

21 °C 50 °C 100 °C 150 °C

100

200

300 400 Pressure (bar)

500

600

700

3.6 Isotherms of the solubility of CO2 in pure water contrasted the solubility of CO2 in 22 wt% NaCl. Isotherms at 21 and 50 °C are nearly linear functions with pressure. Isotherms at 100 and 150 °C are a fit to a complex polynomial expression. Solubility data are from Rosenbauer and Koksalan, 2002.

due probably in part to changes in the density and dielectric constant of CO2 (Rosenbauer, unpublished data; Fig. 3.6). The dielectric constant of CO2 is relatively independent of temperature but increases with density (Keyes and Kirkwood, 1930). The CO2-solubility isotherms also become more tightly clustered at high salt concentrations (Fig. 3.6). The differences in solubility between isotherms decrease at higher NaCl concentrations implying that, at high salinities, ionic strength increasingly controls solubility relative to temperature and pressure. In the absence of any fluid–rock interaction, this dissolved CO2 results in an acidic solution due to the dissociation of carbonic acid (Equation 3.2). H2CO3 ´ H+ + HCO3–

[3.2]

Liquid water is soluble in supercritical CO2. For example, at 100 °C, 300 bar, the solubility of H2O in supercritical CO2 is ~5 % (Takenouchi and Kennedy, 1964). Brine desiccation by mass transfer of liquid water to the supercritical gas phase has been observed in experiments containing a discrete CO2 phase. This increased ionic activity of the residual aqueous phase may promote porosity-reducing precipitation reactions in aquifers near saturation with mineral phases (Kaszuba et al., 2003; Prevost et al., 2005; Rosenbauer et al., 2005).


70


3.3.4 Ionic trapping Carbon dioxide injection induces disequilibrium in the carbonate buffer system of pore fluids. The subsequent dissolution of CO2 and dissociation of carbonic acid can result in reaction with primary minerals of the host formation. Dissociation of carbonic acid into reactive hydrogen ion and bicarbonate potentially initiates a complex series of reactions with aquifer fluids and formation rocks that serves to fix CO2 in aqueous and mineral phases (Rosenbauer et al., 2005; Rosenbauer, 2006; Oelkers and Cole, 2008). A temperature-dependent dissociation constant K can be defined for Equation 3.2 (Equation 3.2a): aH+ · aHCO–3 g H+ [H + ] · g HCO–3 [HC –3 = [3.2a] aH2 CO3 g H2 CO3 2 3CO ] [H CO ] where a is activity, [X] is concentration, and g is an activity coefficient. Calculations of log K (Equation 3.2a) using the geochemical program SUPCRT92 (Johnson et al., 1992) show that a maximum of dissociation for Reaction 3.2 occurs at about 50 °C, above which log K decreases continuously with increasing temperature such that an initially weak acid becomes increasingly weaker at elevated temperature (Fig. 3.7). Thus, at low temperatures, the increased availability of H+ might be expected to cause log K =

–3 [H+]

[H2CO3] = 1.0 molal

–4

log K or log m

–5

–6 K1 –

–7

–8 0

H2CO3 = H+ + HCO3

50


200

3.7 Log K of the dissociation of carbonic acid versus temperature and associated log m of the concentration of hydrogen ion showing a maximum in the dissociation and hydrogen concentration at ~50 °C, indicating maximum acidity at 50 °C. Data calculated from SUPCRT92 (Johnson et al., 1992).



71

higher rates of wall-rock hydrolysis, in spite of expected temperature-related kinetic effects. Experiments reacting rhyolite with CO2 charged waters indeed showed a much greater degree of rock alteration at 200 °C versus 350 °C (Bischoff and Rosenbauer, 1996). The types of reactions between CO2 saturated fluids and formation rocks are dependent on the mineralogy of the formation and the composition of the pore fluids (Rosenbauer et al., 2005). For example, in a limestone matrix, calcite dissolution in response to the increased acidity produced by the dissociation of carbonic acid (Equation 3.3), CaCO3 + CO2 + H2O Æ Ca+2 + 2HCO3–

[3.3]

causes one additional mole of CO2 to be stored as bicarbonate relative to the simple aqueous solubility of CO2 and is an example of the ionic trapping of CO2 (Gunter et al., 1993, 2000, 2004; Assayag et al., 2009). It is, however, important to note that only one of the two moles of bicarbonate produced by this reaction is from the injected CO2. The second mole is produced from dissolution of a mole of carbonate rock. This reaction occurred rapidly in laboratory experiments that reacted CO2 saturated brine with limestone rocks (Rosenbauer et al., 2005). Similar reactions exist for the dissolution of other Mg- and Fe-bearing carbonates. Most carbonates have a retrograde solubility and are therefore more thermodynamically stable at elevated temperature so Reaction 3.3 is favored at lower temperatures (Fig. 3.8). The permeability –8 –2

Calcite = Ca+2 + CO3

–8.5

log K

–9

–9.5

–10

–10.5

–11

0

50


200

250

3.8 Temperature dependence of the solubility of calcite indicates retrograde solubility; calcite is more stable at elevated temperature and more likely to precipitate from solution. Data calculated from SUPCRT92 (Johnson et al., 1992).


72


of dolomite-limestone is typically lower (range ~0.01–0.1 md, Bear, 1988) than sandstones (range 20–3500 md, e.g. Bertier et al., 2006). So the effect of Reaction 3.3 is to promote mineral dissolution near the injection site, enhancing porosity and permeability and facilitating injection. A porosity change of ~15 % was noted in laboratory experiments reacting CO2 saturated brine with limestone (Rosenbauer et al., 2005).

3.3.5 Mineral trapping Under the right geologic setting, a portion of the injected CO2 may be converted to carbonate minerals. Mineral trapping is advantageous because it increases storage capacities and immobilizes CO2 for very long periods (Gunter et al., 1993, 1997; Perkins and Gunter, 1995; Oelkers and Schott, 2005), although precipitation of minerals within porespace especially near the injection site can dramatically decrease permeability and hamper further injection (e.g. Shiraki and Dunn, 2000). The extent to which mineral trapping is capable of storing CO2 is directly dependent on rock type, sedimentary structure, mineralogy, and diffusion. Dissolution of primary and precipitation of secondary minerals affect porosity and permeability and consequently the flow regime (Piri et al., 2005). The efficiency of the CO2 geological storage in sedimentary basins thus depends on many factors, among the most important being formation water composition, lithological heterogeneity, and mineralogy. Host rock composed of silicates can be dissolved by CO2-rich aqueous solution, and carbonates, other silicates, clays, and silica can be produced. For example, reactions of CO2 saturated aquifer fluids with arkosic sandstone illustrate the mineral trapping of CO2 by silicate dissolution, in particular the anorthitic component of plagioclase (Equation 3.4),

2H+ + CaAl2Si2O8(anorthite) + H2O Æ Ca+2

+ Al2Si2O5(OH)4 (kaolinite)

[3.4]

and the subsequent precipitation of calcite (Equation 3.5).

Ca+2 + HCO3– Æ CaCO3 + H+

[3.5]

The solubility product (log Ksp) of the net reaction (Equation 3.6) decreases with increasing temperature (Johnson et al., 1992),

CaAl2Si2O8(anorthite) + H2CO3 + H2O Æ CaCO3(calcite)

+ Al2Si2O5(OH)4(kaolinite)

[3.6]

resulting in competing effects of favorable thermodynamics at low temperature versus kinetic considerations that favor reaction rates at elevated temperature



73

(Fig. 3.9). The dissolution of Na-rich feldspar (albite) to form dawsonite (Equation 3.7) NaAlSi3O8 + H2CO3 = NaAl(CO3)(OH)2 + 3SiO2

[3.7]

is independent of the Na/H+ as it depends only on the fugacity of CO2 (fCO2). The precipitation of amorphous silica in lieu of quartz (Equation 3.7) may significantly reduce porosity and permeability of the formation due to its 28 % greater molar volume (Cipolli et al., 2004). Dawsonite formation is also thermodynamically favored from high-salinity CO2 charged aqueous solutions (Worden, 2006). Similar reactions can be derived for other illitic (Mg-bearing), or glauconitic (Fe II-bearing) sediments, (e.g. Gunter et al., 1997, 2000; Johnson et al., 2001). In contrast to reactions with carbonate minerals, every mole of bicarbonate produced by the dissolution of silicate mineral is derived from CO2. However, most of these reactions are thought to be comparatively slow because they depend on the dissolution of silicate minerals. The overall impact on the reservoir from mineral trapping by silicate reaction may not be realized for tens to hundreds of years or longer. Laboratory experiments reacting CO 2 saturated brine with arkosic sandstone (120 °C) or shale (200 °C) indeed show that silicate reaction kinetics are slow but occur on measurable timescales (Kaszuba et al., 2003; Rosenbauer et al., 2005). Other workers (Gunter et al., 1997; Sass et al., 2002; Gupta et al., 2000; Bateman et al., 2005; Giammar et al., 2005) noted only limited amount of reaction of CO2 with pure mineral 14

Anorthite + CO2 + H2O = Kaolinite + calcite

12

log K

10

8

6 4

2

0

50


200

250

3.9 Temperature dependence of the reaction of anorthite with dissolved CO2 to produce calcite and kaolinite. This reaction is thermodynamically favored at lower temperature.


74


phases (anorthite, glauconite, or fosterite) or sandstones at 50° and 150 °C and low pressure. Dissolution rates of fosterite determined by flow-through reactor experiments agreed well with similar batch-type experiments and published literature values (Bruant et al., 20003). Recent work on the dissolution of anorthite in the presence of CO2, based on mineral surface measurements demonstrated an initial rapid transitory rate of dissolution followed by a slower steady state (Sorai et al., 2007) resulting in porosity increases of reservoir rock of about 2 %. Oelkers and Schott (2005) suggested that experimental dissolution rates of metal-bearing silicates (Carrol and Knauss, 2005; Golubev et al., 2005) and carbonates (Pokrovsky et al. (2005) can be accurately modeled (Oelkers et al. 1994; Pokrovsky et al., 1999) by simply taking account of the presence of CO2 on solution pH. Ferriciron-bearing saline formations such as redbeds have been proposed as a mineral trap for CO2 (Palandri and Kharaka, 2004, 2005). Sulfur dioxide (SO2), a common component of flue gas, can be used to reduce ferric iron in these deposits to ferrous iron resulting in the precipitation of the iron carbonate, siderite (Equation 3.8):

Fe2O3 + 2CO2(g) + SO2(g) + H2O = 2FeCO3 + H2SO4

[3.8]

This reaction was shown to proceed experimentally at 150 °C, 300 bar (Palandri et al., 2004, 2005). Other experimental results (Rosenbauer et al., 2005) and reactive transport modeling (Knauss et al., 2002, 2005) point out that the presence of SO2 in Ca-rich brines and oxidizing environments significantly alters the quantity of CO2 trapped in carbonate minerals due to the formation of anhydrite and very low pH pore fluids. Reactions involving supercritical CO2 and carbonic acid with aquifer fluids and formation rocks are thus many and varied. In general, CO2 interaction causes the dissolution of carbonate phases in limestones and the dissolution of silicates and precipitation of carbonates in plagiocalse-rich arkosic sandstones, but none of these reactions can be taken in isolation. Multiphase systems are extremely complex; compositional changes can take different pathways and different chemical reactions can occur simultaneously. Reaction path modeling helps describe solution properties and mineral solution equilibria. Experimental studies are useful for providing and constraining input to coupled geochemical and mass transport models aimed at predicting CO2 behavior and the consequences of CO2 injection in different types of subsurface systems.

3.4

Modeling of carbon dioxide (CO2) sequestration

3.4.1 History Mathematical models and numerical simulation tools play an important role in evaluating the feasibility of CO2 storage in subsurface brine reservoirs. © Woodhead Publishing Limited, 2010


75

The basis of geochemical modeling of aqueous electrolyte solutions is the computation of activity coefficients, and hence activities, of relevant solutes and the solvent water through maintenance of mass and charge balances and equilibrium relations. The Debye–Huckle method (Debye and Hückel, 1923a,b) of calculating activity coefficients provided a theoretical understanding and the HKF model (Helgeson et al., 1981) and the software program SUPCRT92 (Johnson et al., 1992) provided a way to calculate the thermodynamic properties of minerals, gases, and aqueous species at elevated temperature and pressure. Early conceptual and numerical models, for example: SOLMINEQ (Kharaka and Barnes, 1973), EQ3/EQ6 (Wolery, 1979, 1992), PHREEQE (Parkhurst et al., 1980) and SOLVEQ (Reed, 1982), utilized these principles to calculate the distribution of aqueous species and mineral equilibria in a variety of natural and experimental solutions. A theoretical model of concentrated electrolytes based on a virial coefficient approach was developed by Pitzer (1973, 1975), implemented by Harvie and Weare (1980) and software created by Kharaka et al. (1988) (SOLMIN 88), Plummer et al. (1988) (PHRQPITZ), and Parkhurst (1995) (PHREEQC) to describe solution properties and mineral solution equilibria at high ionic strength. Later, the Pitzer equations were extended to aqueous solutions of NaCl and CO2 at elevated temperature (Corti et al., 1990). During CO2 injection, geochemical processes are strongly affected by physical processes such as multiphase fluid flow and solute transport. Accurate geochemical simulations therefore require a computational capability that couples multiphase-flow processes with kinetically controlled geochemical processes. State of the art modeling has advanced from separate thermodynamically-derived equilibria-based numerical models for batch process simulation to fully coupled simulation programs for non-isothermal reactive geochemical transport in variable saturated geologic media. Integrated reaction path and solute transport models such as TOUGHREACT (Xu and Pruess, 1998; Xu et al., 2004b) and updated versions of EQ3/6 (Wolery and Daveler, 1992) couple complex geochemical, hydrological, and mechanical interactions following CO2 injection (Marini, 2007, p. 349). Models that include physico-chemical processes, chemical reaction, fluid flow, heat transfer, and mechanical properties vastly improve predictive abilities of reservoir simulations. These and other models have been utilized to investigate CO2 injection, storage, and sequestration processes and analyze the coupled mechanisms that lead to structural, residual, solubility, and mineral trapping. The following are representative examples of reactive transport modeling of different reservoir types.


76


3.4.2 Sedimentary basins Sandstone aquifers The Utsira Sand, Sleipner, northern North Sea. One of the first modeling exercises simulated CO2 injection into the Utsira formation at StatoilHydro’s North-Sea Sleipner facility (Johnson et al., 2001) where the subsurface is well characterized. Johnson et al. (2001) used an integrated approach combining reactive transport simulators NUFT (Nitao, 1998) and GIMRT/OS3D (Steefel and Yabusaki, 1996; Steefel, 2001, 2008), supporting geochemical software SUPCRT92 (Johnson et al., 1992; Shock, 1998), thermodynamic-kinetic databases GEMBOCHS (Johnson and Lundeen, 1994a,b), and a graphic utility Xtool (Daveler, 1998). They created numeric simulations of three distinct sequestration scenarios, differing in the extent of intra-aquifer shale units and lateral permeability. Results of this study showed the relationship of intra-aquifer permeability to solubility and mineral trapping and the importance of the overlying shale caprock to long-term storage. Gaus et al. (2005) also performed PHREEQC (Parkhurst, 1995) 1D diffusive reactive transport simulations of dissolved CO2 in the caprock of the Utsira aquifer (Sleipner project) and concluded that plagioclase feldspar in the shale alters to dawsonite, disordered dolomite, and calcite. Glauconitic sandstone of the Alberta Basin. Two contrasting reaction path models of the potential CO2 sequestration in the glauconitic sandstone aquifer of the Alberta Basin, Western Canada have been proposed (Marini, 2007, p. 396). In fluids super-saturated with CO2, Gunter et al. (1997), utilizing PATHARCH (Perkins and Gunter, 1995a) and anorthite/albite and muscovite proxies for oligoclase and illite, modeled the alteration of primary glauconite (as annite) to siderite. At a fixed under-saturated PCO2, with the same proxies, Gunter et al. (2000), again utilizing PATHARCH, modeled the almost complete consumption of CO2 in the precipitation of siderite, calcite, and dolomite, showing the high reactivity of Fe+2-bearing silicates to CO2. In contrast, Xu et al. (2000, 2004a), utilizing TOUGHREACT and the actual mineral assemblage of this formation, modeled precipitation of the CO2-bearing phases as mainly siderite and ankerite with some dawsonite and dolomite. Later, Strazisar et al. (2006) investigated the periphery of this system at the CO2 reaction front away from the injection site. These investigators point out that precipitation reactions are more likely to occur downstream of the injection site, where the mineral assemblage buffers the pH at higher levels. Utilizing PHREEQC, they found that CO2 was trapped initially in siderite from the pH-dependent dissolution of annite and coupled with kaolinite dissolution and K-feldspar precipitation. As the CO2 front migrated further, K-feldspar dissolved and calcite and dolomite precipitated. They concluded



77

that most of the CO2 is trapped in siderite, consistent with previous studies (Gunter et al., 1997, 2000; Xu et al., 2000, 2004a). Deep sand aquifers of the Powder River Basin of Wyoming. McPherson and Lichtner (2001), utilizing TOUGH2 (Pruess et al., 1999), carried out a numerical model that included multiphase fluid flow of sustained injection of CO2 into the deep sand aquifers of the Powder River Basin of Wyoming. Their calculated CO2 residence time and migration rates show the unintended impact of wide-scale brine displacement out of the target aquifer and potentially spreading into adjacent sealing layers. Tertiary Gulf Coast sediments: The Frio, Vicksberg and Wilcox formations. Apps (1996) first carried out batch geochemical modeling of Gulf Coast sediments as a potential repository basis for the deep injection disposal of hazardous and industrial wastes. Xu et al. (2004a,b), utilizing TOUGHREACT, constructed a reactive transport model of the Frio Formation for potential CO2 sequestration. These investigators incorporated in their model the high organic matter (kerogen) and salt content from diapirs that are characteristic of these sediments, along with a representative mineral assemblage at a CO2 injection pressure of 260 bars. Dawsonite and ankerite were the primary CO2-bearing products; calcite and siderite were modeled to initially precipitate then dissolve. Xu et al. (2005), again utilizing TOUGHREACT, focused on the diffusion of CO2 and acidity into the caprock and found carbonate precipitation extending into the shale. Knauss et al. (2005), utilizing the geochemical software CRUNCH, another computer program for simulating multicomponent multidimensional reactive transport in porous media (Steefel, 2001), investigated the effects of ancillary contaminant gases in the Frio CO2 injection stream. This work concluded that only SO2 might have an impact on reaction processes due to the resulting extremely low pH. Permian White Rims sandstone. White et al. (2005), utilizing ChemTOUGH (White, 1995), simulated CO2 injection into this reservoir rock, situated beneath the Hunter Power plant in central Utah. These investigators found that calcite and dolomite precipitated but predicted that after 1000 years about 17 % of the CO2 had leaked to the ground surface. Carbonate aquifers Tuscan Nappe limestone formation. Cantucci et al. (2008) utilized a modified version of the PRHEEQC (V2.11) software package to investigate the shortand long-term consequences of CO2 storage in an offshore Italian porous carbonate saline aquifer, the Tuscan Nappe limestone formation. Numeric


78


simulations of the fate of CO2 injected into the saline aquifer suggest that solubility trapping prevails within the first 100 years. Nisku carbonate aquifer of the Alberta Basin. Gunter et al. (2000) modeled the interaction of under-saturated CO2 charged fluid with carbonate (calcite, dolomite) rocks of the Nisku and found rapid dissolution of calcite and precipitation of dolomite. Serpentine aquifers Serpentinites of the Gruppo di Voltri (Genova, Italy). The high Mg content of serpentinites has the potential for fixing CO2 as magnesite. Cipolli et al. (2004) carried out reaction path modeling of potential CO2 sequestration in deep aquifers hosted by the serpentinized ultramafic rocks of the southern Piedmont, Italy. These investigators concluded that the capacity for CO2 sequestration is high through dissolution of serpentine and precipitation of magnesite and chalcedony but cautioned about the progressive loss of porosity, especially if amorphous silica precipitates instead of chalcedony. They also point out that reactant armoring may occur during precipitation reactions. Not often noted in the literature, this phenomenon is seen in experimental systems.

3.4.3 Model inter-comparison Pruess et al. (2001a,b, 2002, 2004; Pruess, 2005) initiated and carried out an inter-comparison study of reactive transport models on test sets of representative CO2 sequestration in potential reservoirs. Results were quantitatively similar, indicating broad agreement among the models. Allen et al. (2005) conclude that models are only as reliable as the data and reaction scheme upon which they are based and emphasize the importance of pressure corrections to thermodynamic data. Model inter-comparisons indicate that failure to adjust all equilibrium constants to account for elevated CO 2 pressures results in significant errors in both solubility and mineral formation estimates (Allen et al., 2005). Moreover, Bateman et al. (2005) concluded that model predictions tend to overestimate the degree of reaction compared with experimental results. For example, some mineral phases such as dawsonite that are predicted to form in large quantities by models are not observed in the experimental system. These authors highlight the need for appropriate thermodynamic and kinetic data to address these discrepancies. The most robust analyses of CO2 sequestration in potential reservoirs incorporate a combination of experimental, model, and field evidence that often requires large-scale pilot project support (Fig. 3.10).


3

3 3 1

3 1 3


3

3

1 2 3 3

1

2

Other

3 Unknown

79

3.10 Geologic storage and related projects in operation or proposed around the world. Most are research, development or demonstration projects. Some are commercial operations (image source: CO2CRC).


3

80

3.5


Carbon dioxide (CO2) sequestration pilot sites

3.5.1 The Utsira Sand, Sleipner, northern North Sea The Sleipner project is the prototypical test site for CO2 sequestration into a deep saline aquifer. It is the world’s first commercial CO2 capture and storage project (Gale et al., 2001), motivated by the high tax Norway imposes on offshore oil and gas activity. Started in September 1996, it sequesters about one million metric tons of CO2 each year into the Utsira formation at StatoilHydro’s North-Sea Sleipner facility (Chadwick et al., 2004, 2006). As of 2009, about 10 Mt CO2 had been stored in Utsira at a cost of $17/t (Hermanrud et al., 2009) with a final target of 20 Mt. The injection site lies 1000 m below the seafloor and is composed largely of uncemented finegrained sand, primarily quartz (75 %) with some feldspar (13 %) (Chadwick et al., 2004) (Fig. 3.11). Porosity of the Utsira sand is ~30 %, locally up to 42 % (Chadwick et al., 2004). The caprocks are non-organic mudstones composed of 30 % quartz, 30 % mica, and 14 % kaolinite. As of 2004, the plume area extended to 2.8 km2 and the maximum distance from the injection site was 2560 m (Torp and Gale, 2004). Extensive seismic monitoring was carried out by Saline Aquifer CO2 Storage (SACS) (Fig. 3.12), and these studies indicated only limited interaction with formation rocks (Torp and Gale, 2004). Although the activities of SACS ended in 2002, part of its effort continues in the EU-co-funded project CO2STORE, a research project with 19 participants from industry and research institutes, whose aim is to prepare the ground for widespread underground storage of CO2 in aquifers. Sleipner has been and continues to be the subject of much geophysical study (i.e. Arts et al., 2008; Hermanrud et al., 2009.

3.5.2 The Snøhvit project The Snøhvit plant in the Barents Sea is another StatoilHydro commercial project in which CO2 is captured and stored. In this case, the target formation is below a gas reservoir in the Tubåsen sandstone formation (Kårstad, 2002). This case deals with StatoilHydro’s Snøhvit natural gas field and liquid natural gas (LNG) facility in northern Norway. Gas production began at Snøhvit in 2007 and CO2 storage in April 2008 (Estublier and Lackner, 2009). At full capacity, 700 000 t of CO2 is expected to be stored per year. The field consists of a fully subsea offshore development in the Barents Sea, a 160 km pipeline to shore and a liquification plant for LNG (Kårstad, 2002). The Tubåsen sandstone is 45–75 m thick with an overlying shale cap (Maldal and Tappel, 2004).



Utsira Norway Formation Utsira formation

Sleipner

800 -1000 m depth Scotland CO 2 injection

Gas production

Gas field (3000 m)

81

3.11 Schematic diagram depicting gas production at depth and re-injection of CO 2 into the Utsira sand at StatoilHydro’s NorthSea Sleipner facility. Shown also is a location map for the facility in the North Sea (image courtesy of StatoilHydro, ASA).


Sleipner platform

82

8.4 Mt CO2 (2006)

Top of Utsira sand CO2

Base of Utsira sand 2 km

CO2

200 m


4.3 Mt CO2 (2001)

3.12 Seismic monitoring of Sleipner CO2 injection plume. Time-lapse seismic images showing changes in reflectivity due to injected CO2. Arrows indicate CO2 accumulations (modified from original image created by Ola Eiken, StatoilHydro, ASA).


Pre-injection (1994)


83

3.5.3 The Frio formation The first US sequestration field experiment was completed in 2008 at the Frio site located within the South Liberty oil field, near Houston, Texas (Doughty et al., 2002, 2007; Hovorka et al., 2006, 2009). Injection and updip observation wells perforated the target Frio ‘C’ sandstone at ~ 1540 m, above a zone of oil production at 2900 m (Kharaka et al., 2006b). The Frio ‘C’ unit is a poorly cemented subarkosic sandstone dominantly comprising fine grained quartz with minor amounts of illite/smectite, feldspar, and calcite with a porosity of 32 % and permeability of 2–3 darcys. (Kharaka et al., 2006b). The caprock is the regional Miocene–Oligocene Anahuac Shale (Hovorka et al., 2005). Aquifer fluids are 100 ppt NaCl at a temperature of 65 °C (Kharaka et al., 2006b). A test injection of 1600 t CO2 was carried out while monitoring downhole pressure, temperature, and fluid compositions. Intensive fluid sampling and analyses by methods described in Kharaka and Hanor (2007) showed large pH decreases and increases in concentrations of HCO3–, Ca+2, Fe+2, and Mn+2, likely resulting from the rapid dissolution of carbonate and iron oxyhydroxide minerals in the Frio Formation (Kharaka et al., 2006a,b). The time to CO2 breakthrough and the spatial distribution of the CO2 plume in the subsurface, measured with cross-well tomography and wireline logs, compare well with values and patterns predicted by transport modeling using TOUGH2 (Hovorka et al., 2005) (Fig. 3.13)

3.5.4 Other carbon dioxide (CO2) sequestration projects Other international CO2 sequestration demonstration and pilot projects (Fig. 3.10) either online or planned (IPCC, 2005) include the Minami-Nagaoka in Japan, the Otway and Gorgon in Australia, and the Ketzin in Germany. The Minami-Nagaoka gas field in Nagaoka City, 200 km north of Tokyo is the first pilot CO2 injection site in Japan. The target reservoir is the early Pleistocene sandstone formation that lies 3000 m above the gas reservoir and 1100 m below the surface. About 10 000 t CO2 were injected from 2003 through 2005 (Tanase and Yoshimura, 2008). The CO2CRC Otway Project, located in southwest Victoria is Australia’s first demonstration of deep geological storage of CO2. It will inject CO2 obtained from the Buttress gas field into a nearby depleted gas field in the Waarre Formation at depth of ~1700 m. Although technically not CO2 storage in a deep saline aquifer, the Otway Project is one of the largest research and geologic sequestration demonstration projects with a planned injection of about 100 000 t CO2 by 2010 (Sharma et al., 2009). The Gorgon Project, located in northwest Australia, will extract CO2 from produced gas within the Greater Gorgon gas fields prior to liquefaction into LNG and re-inject it into the Dupuy Formation located 2000 m beneath


84

Developments and innovation in CCS technology (a)

Modeled CO2 plume Obs Nov 30, 2004

75 ft

Inj

sand of C Top

107

ft

Sg: 0 0.05 0.1 0.15 0.2 0.25 0.3 Injection well

(b)

0

Distance (m) 10 20

30

Observation well

Depth (ft below GL)

4950

5000

5050 Sigma

Sigma –1.0 –0.5 0 0.5 Change in velocity (km/s)

3.13 (a) CO2 plume in the sub-surface Frio after CO2 injection, measured with cross-well tomography and wireline logs; (b) CO2 plume in the subsurface Frio predicted by transport modeling using TOUGH2 (Hovorka et al., 2005) (image modified from original created by Christine Doughty and Thomas M. Daley in Hovorka et al. (2006), reprinted with kind permission of Springer Science+Business Media).

Barrow Island. The volume of reservoir CO2 to be re-injected is about 100 Mt (Flett et al., 2008). Thirteen CCS projects involving the capture and/or storage of CO2 in Australia are currently proposed or underway. The CO2



85

storage site near the town of Ketzin in Germany is the first onshore pilot site in Europe. The target formation is the Triassic Stuttgart silt and sandstones at a depth of 700 m. Injection started in June 2008 and will continue for two years. A maximum of 60 000 t CO2 will be injected. Extensive geophysical and geochemical monitoring is being carried out to follow the spread of the CO2 plume (Schilling et al., 2008). Also, each of the seven US Department of Energy (DOE) Regional Carbon Sequestration Partnerships has plans for CO2 capture and storage projects in deep geologic formations. These include tertiary sandstones in the San Joaquin Valley of California, the Wingate (Jurassic–Triassic-aged) sandstone in the southwest USA, the deep carbonate saline formation in the Williston Basin in North Dakota, the Mount Simon sandstone formation in central Illinois, the Tuscaloosa Massive Sandstone in Mississippi and Louisiana, and the Nugget Sandstone formation in southwest Wyoming.

3.5.5 Weyburn The IEA Weyburn CO2 Monitoring and Storage project is neither a deep saline aquifer site nor technically even a CO2 sequestration site. It is a large-scale commercial CO2–enhanced oil recovery (EOR) site that, under the management of the Petroleum Technology Research Center (PTRC), has provided a wealth of geophysical and geochemical research that is applicable to CO2 sequestration in saline aquifers. In the Weyburn EOR Project, located in Prairie Province of Saskatchewan, Canada, CO2 has been used to increase recovery of oil from the carbonate Midale Beds of the Mississippian Charles Formation, where about 3 billion m3 of supercritical CO2 have been injected since 2000 at a rate of 5000 t/day (Riding and Rochelle, 2005; Cantucci et al., 2009). Carbon dioxide is obtained from the Dakota Gasification Company, near Beulah ND and transported 320 km via pipeline to the Weyburn and Midale oil fields where it is injected with water at 800 m depth. A mixture of oil, water, and CO2 is extracted; extracted CO2 is re-injected. The database obtained at this site has provided an opportunity to monitor dynamic reservoir response and study effective trapping mechanisms, seals, hydraulic isolation, hydrogeological regime, and pathways for migration along faults and fractures (Preston et al., 2009). Seismic monitoring has provided detailed CO2 plume distribution and containment within the reservoir (Davis et al., 2003; Preston et al., 2009). Changes in the fluid chemistry and isotopic composition of the produced fluid, pre- and post-injection, provide evidence of dissolution of injected CO2, an increase of total dissolved solids, and both dissolution and precipitation of carbonate minerals (Emberley et al., 2004, 2005; Perez et al., 2006; Mayer et al., 2008). Most of the injected CO2 exists in a supercritical state but has reacted with the reservoir rock sufficiently to mask some of the strontium isotope signature caused by 40 years of water


86


flooding (Riding and Rochelle, 2009). Geochemical modeling suggests that eventually most of the CO2 will be stored by solubility and mineral trapping (via dawsonite precipitation) (Cantucci et al., 2009). Pre- and post-injection soil gas surveys have not found any evidence for leakage of the injected CO2 to surface (Riding and Rochelle, 2009).

3.6

Future trends

The geologic storage of CO2 in saline aquifers has proven technologically feasible (Michael et al., 2009). Moreover, saline aquifers contain the volume necessary to sequester a large fraction of our fossil fuel-derived carbon output (Burruss et al., 2009), but the greatest challenges to the commercial-scale exploitation of saline aquifers still lie in the future. The costs, environmental risks, legal framework, and social acceptance of CO2 sequestration in saline aquifers represent increasingly important challenges as we look toward the future.

3.6.1 Costs of saline aquifer sequestration Saline aquifers only rarely contain exploitable hydrocarbon resources. So the primary economic incentive for industry to sequester CO2 in saline aquifers is determined by the balance between the price of carbon abatement and the cost of operations including assumed risk of leakage and environmental damage. Current estimates of the costs associated with saline aquifer sequestration depend on properties of the rocks and the industrial capacity of the locations, including access to wells, costs of drilling new wells, and the cost of transport to the disposal site. Bock et al. (2003) estimate a base case scenario cost of approximately $3/t CO2, though the estimates vary by an order of magnitude depending on the assumed well density and depth. Another estimate by Hendriks et al. (2004) depicts the depth-dependence of costs associated with both onshore and offshore operations (Fig. 3.14). The difference in offshore operation costs renders offshore operations generally much more expensive, although the environmental and public risk of leakage are greatly reduced. Currently the costs of EOR, ECBM, and disposal in depleted hydrocarbon reservoirs are far lower than the cost of saline aquifer sequestration (Bock et al., 2003).

3.6.2 Environmental safety and concerns The environmental concerns for saline aquifer sequestration are shared with many other sequestration sites. Importantly, the injection into saline formations has the potential to displace brines into freshwater resources. Given the acidic character of these brines, there may also be the potential for



87

12

10

8 Cost (7/t)

Offshore 6 onshore 4

2

0 1000

1500

2000 Depth of injection (m)

2500

3000

3.14 Projected transport and operating costs of saline aquifer injection into onshore and offshore reservoirs from data in Hendriks et al. (2004).

liberation of heavy metals and organic constituents from formation materials (Kharaka et al., 2006b; Kolak and Burruss, 2006). As mentioned previously, high injection pressures can induce hydraulic fracturing that increases the likelihood of brine expulsion into overlying formations. Another important concern is the escape of the CO2 plume or charged aqueous fluid to the surface along faults, fractures, or through permeable strata. In addition to negating the economic benefit of sequestration, the degassing of CO2 at the surface can produce deadly concentrations in topographic depressions such as valleys and household basements.

3.6.3 Policy considerations Unlike hydrocarbon reservoirs where mineral rights are well established, there are few laws related to the ownership of porespaces and porefluids in saline aquifers (IPCC, 2005). This legal uncertainty is problematic when considering the future assumption of risk of leakage and associated environmental damage. Clearly, local and national laws are required to clarify these concerns. In the USA, there are currently proposals under consideration to establish a


88


risk-pooling entity to absorb the costs of monitoring and verification, and the risk of environmental damage (HR 2454, 2009).

3.6.4 Social acceptance While the public increasingly supports action to mitigate climate change, public awareness is still minimal on the subject of CO2 capture and sequestration (Curry et al., 2005). Perhaps the greatest concern will be convincing the public to accept the significant industrial footprint required to implement large-scale sequestration projects. Sequestration into depleted oil and gas reservoirs or coalbed formations has the advantage of an already significant industrial footprint. For saline aquifers, there is unlikely to be a precedent industrial footprint. Thus, public skepticism and the ubiquitous ‘not in my back yard’ conundrum that has plagued other waste disposal projects should not be ignored (Ibitayo, 2002). Given these concerns, it will be important to incentivise landowner cooperation (IPCC, 2005) and to educate the public about the inherent and future risks.

3.7

Conclusions

Deep saline aquifers remain a focus as potential repositories for the geologic storage of CO2 in part due to their widespread occurrence, high capacity, and proximity to point sources. Various trapping mechanisms offer the possibility of long-term storage of the injected CO2. Methods are improving for estimating their geologic and hydrologic properties, but much of our knowledge of the behavior of CO2 in the subsurface is inferred from geochemical laboratory experiments and reactive transport modeling. Demonstration, pilot, and fullscale commercial CO2 sequestration projects such as the Sleipner Project provide invaluable data to the scientific research community and to industry to further our understanding of the geochemical, hydrological, and physical processes involved with supercritical CO2, CO2-charged fluids, and host formations.

3.8

Acknowledgements

We thank Yousif Kharaka, Steve Bouzalakos and Anna Kaminska who provided thorough and thoughtful reviews of earlier versions of this manuscript.

3.9

References

Albritton D L and Meira Filho L G (2001) Climate Change 2001: The Scientific Basis; Contribution of Working Group I to the Third Assessment Report of the Intergovernmental Panel on Climate Change – Technical Summary. IPCC, Geneva, Switzerland.



89

Alfredssom H A, Hardarson B S, Franzson H and Gislason S R (2008) CO2 sequestration in basaltic rock at the Hellisheidi site in SW Iceland: stratigraphy and chemical composition of the rocks at the injection site, Mineralogical Magazine, 72(1), 1–5. Allen M R, Stott P A, Mitchell J F B, Schnur R and Delworth T L (2000) Quantifying the uncertainty in forecasts of anthropogenic climate change, Nature, 407, 617–620. Allen D E, Strazisar B R, Soong Y and Hedges S W (2005) Modeling carbon dioxide sequestration in saline aquifers: Significance of elevated pressures and salinities, Fuel Processing Technology, 86(14–15), 1569–1580. Apps J A (1996) An approach to modeling of the chemistry of waste fluid disposal in deep saline aquifers, in Apps, J.A., Tsang, C.F. (eds), Deep Injection Disposal of Hazardous and Industrial Waste: Scientific and Engineering Aspects. Academic Press, San Diego, Ca, 465–488. Arts R, Chadwick A, Eiken O, Thibeau S and Nooner S (2008) Ten years’ experience of monitoring CO2 injection in the Utsira Sand at Sleipner, offshore Norway, First Break, 26, 65–72. Assayag N, Matter J, Ader M, Goldberg D and Agrinier P (2009) CO2 ionic trapping at meta-sedimentary aquifer, following a CO2 injection push-pull test, in Gale J, Herzog H and Braitsch J (eds), Greenhouse Gas Control Technologies 9, Proceedings of the Ninth International Conference on Greenhouse Gas Control Technologies (GHGT9), Energy Procedia, 1, 2357–2360. Aya I, Yamane K and Nariai H (1997) Solubility of CO2 and density of CO2 hydrate at 30 Mpa, Energy, 22, 263–271. Bachu S (2000) Sequestration of CO2 in geological media: criteria and approach for site selection in response to climate change, Energy Conversion and Management, 41, 953–970. Bachu S (2003) Screening and ranking of sedimentary basins for sequestration of CO2 in geological media in response to climate change, Environmental Geology, 44, 277–289. Bachu S, Gunter W D and Perkins E H (1994) Aquifer disposal of CO2: hydrodynamic and mineral trapping, Energy Conversion and Management, 35, 269–279. Bachu S, Bonijoly D, Bradshaw J, Burruss R, Holloway S, Christensen N P and Mathiassen O M (2007) CO2 storage capacity estimation: methodology and gaps, International Journal of Greenhouse Gas Control, 1(4), 430–443. Barton P B and Chou I (1993) Calculation of the vapor-saturated liquidus for the NaCl– CO2–H2O system, Geochimica et Cosmochimica Acta, 57, 2715–2723. Bateman K, Turner G, Pearce J M, Noy D J, Birchall D and Rochelle C A (2005) Largescale column experiment: Study of CO2, porewater, rock reactions and model test case, Oil & Gas Science and Technology – Rev. IFP, 60(1), 161–175. Bear J (1988) Dynamics of Fluids in Porous Media. Courier Dover Publications, Mineola, NY. Beecy D J and Kuuskraa V A (2001) Status of U.S. geologic carbon sequestration research and technology, Environmental Geosciences, 8, 152–159. Bentham M and Kirby G (2005) CO2 storage in saline aquifers, Oil & Gas Science and Technology – Rev. IFP, 60(3), 559–567. Benson S M and Cole D R (2008) CO2 Sequestration in deep sedimentary formations, Elements, 4, 325–331. Bergman P D and Winter E M (1995) Disposal of carbon dioxide in aquifers in the U.S., Energy Conversion and Management, 36, 523–526. Bertier P, Swennen R, Laenen B, Lagrou D and Dreesen R (2006) Experimental


90


identification of CO2–water–rock interactions caused by sequestration of CO2 in Westphalian and Buntsandstein sandstones of the Campine Basin (NE-Belgium), Journal of Geochemical Exploration, 89, 10–14. Bischoff J L and Rosenbauer R J (1996) The alteration of rhyolite in CO2 charged water at 200 and 350 °C: The unreactivity of CO2 at higher temperature, Geochimica et Cosmochimica Acta, 60(1), 7–16. Blencoe J G, Naney M T and Anovitz L M (2001) The CO2-H2O system: III. A new experimental method for determining liquid-vapor equilibria at high subcritical temperatures, American Mineralogist, 86, 1100–1111. Bock B, Rhudy R, Herzog H, Klett M, Davison J, De la Torre Ugarte D and Simbeck D (2003) Economic evaluation of CO2 storage and Sink options, DOE Research Report DE-FC26-00NT40937. Bradshaw J and Dance T (2005a) Mapping geological storage prospectivity of CO2 for the world’s sedimentary basins and regional source to sink matching, in Rubin E S, Keith D W and Gilboy C F (eds), Proceedings of the Seventh International Conference on Greenhouse Gas Control Technologies: GHGT7, IEA GHG, Cheltenham, UK, 1, 583–592. Bradshaw J and Dance T (2005b) Mapping geological storage prospectivity of CO2 for the world’s sedimentary basins and regional source to sink matching, Geophysical Research Abstracts, 7, 01865. Bradshaw J, Bachu S, Bonijoly D, Burruss R, Holloway S, Christensen N P and Mathiassen O M (2007) CO2 storage capacity estimation: Issues and development of standards, International Journal of Greenhouse Gas Control, 1(1), 62–68. Brennan S T and Burruss R C (2006) Specific storage volumes: A useful tool for CO 2 storage capacity assessment, Natural Resources Research, 15(3), 165–182. Bruant Jr. R G, Celia M A, Guswa A J and Peters C A (2002) Safe storage of CO2 in deep saline aquifiers, Environmental Science and Technology, 36(11), 240A–245A. Bruant Jr. R G, Giammar D E, Myneni S C B and Peters C A (2003) Effect of pressure, temperature, and aqueous carbon dioxide concentration on mineral weathering as applied to geological storage of carbon dioxide, in Gale J and Kaya Y (eds), Proceedings of the Sixth International Conference on Greenhouse Gas Control Technologies: GHGT6, Elsevier (Pergamon), Oxford, UK, 2, 1609–1612. Burruss R C (2004) Geologic Sequestration of carbon dioxide in the next 10 to 50 Years: An energy resource perspective, in Workshop Proceedings, The 10–50 Solution: Technologies and Policies for a Low-Carbon Future, Pew Center on Global Climate Change, March 25–26, Washington, DC. Burruss R C, Brennan S T, Freeman P A, Merrill M D, Ruppert L F, Becker M F, Herkelrath W N, Kharaka Y K, Neuzil C E, Swanson S M, Cook T A, Klett T R, Nelson P H and Schenk C (2009) Development of a Probabilistic Assessment Methodology for evaluation of carbon dioxide storage, U.S. Geological Survey Open-File Report 2009–1035, available only online at: http://pubs.usgs.gov/of/2009/1035/ (accessed January 2010). Cantucci B, Procesi M, Buttinelli M, Montegrossi G, Vaselli O and Quattrocchi F (2008) Mineralogy and geochemical trapping of CO2 in an Italian carbonatic deep saline aquifer: preliminary results, Geophysical Research Abstracts, 10, EGU2008-A-00000, 2008EGU General Assembly. Cantucci B, Montegrossi G, Vaselli O, Tassi F, Quattrocchi F and Perkins E H (2009) Geochemical modeling of CO2 storage in deep reservoirs: The Weyburn Project (Canada) case study, Chemical Geology, 265, 181–197.



91

Carroll S A and Knauss K G (2005) Dependence of labradorite dissolution kinetics on CO2 (aq), Al(aq), and temperature, Chemical Geology, 217, 213–225. Chadwick R A, Zweigel P, Gregersen U, Kirby G A, Holloway S and Johannessen P N (2003) Geological CO2 storage sites: lessons from Sleipner, Northern North Sea, in Gale J and Kaya Y (eds), Proceedings of the Sixth International Conference on Greenhouse Gas Control Technologies: GHGT6, Elsevier (Pergamon), Oxford, UK,1, 321–326. Chadwick R A, Zweigel P, Gregersen U, Kirby G A, Johannessen P N and Holloway S (2004) Characterisation of a CO2 storage site: The Utsira Sand, Sleipner, northern North Sea, Energy, 29, 1371–1381. Chadwick A, Arts R, Eiken O, Williamson P and Williams G (2006) Geophysical monitoring of the CO2 plume at Sleipner, North Sea: an outline review, in Lombardi, S, Altunina, LK, Beaubien, SE (eds) Advances in the Geological Storage of Carbon Dioxide: International Approaches to Reduce Anthropogenic Greenhouse Gas Emissions, NATO Science Series IV: Earth and Environmental Sciences, Springer, Dordrecht, the Netherlands, 303–314. Chapoy A, Mohammadi A H, Chareton A, Tohidi B and Richon D (2004) Measurement and modeling of gas solubility and literature review of the properties for the carbon dioxide–water system, Industrial and Engineering Chemistry Research, 43, 1794– 1802. Cipolli F, Gambardella B, Marini L, Ottonello G and Vetuschi Zuccolini M (2004) Geochemistry of high-pH waters from serpentinites of the Gruppo di Voltri (Genova, Italy) and reaction path modeling of CO2 sequestration in serpentinite aquifers, Applied Geochemistry, 19, 787–802. Corti H R, de Pablo J J and Prausnitz J M (1990) Phase equilibria for aqueous systems containing salts and carbon dioxide. Application of Pitzer’s theory for electrolyte solutions, Journal of Physical Chemistry, 94, 7876–7880. Cox P M, Betts R A, Jones C D, Spall S A and Totterdell I J (2000) Acceleration of global warming due to carbon-cycle feedbacks in a coupled climate model, Nature, 408, 184–187. CSLF (Carbon Sequestration Leadership Forum) (2005) Phase I Final Report from the Task Force for Review and Identification of Standards for CO2 Storage Capacity Measurement, CSLF-T-2005-9, August, available at: http://www.cslforum.org/ publications/documents/PhaseIReportStorageCapacityMeasurementTaskForce.pdf (accessed February 2010). CSLF (Carbon Sequestration Leadership Forum) (2007) Phase II Final Report from the Task Force for Review and Identification of Standards for CO2 Storage Capacity Estimation, CSLF-T-2007-04, June 15, available at: http://www.cslforum.org/ publications/documents/PhaseIIReportStorageCapacityMeasurementTaskForce.pdf (accessed February 2010). CSLF (Carbon Sequestration Leadership Forum) (2008) A CSLF Task Force on CO2 Storage Capacity Estimation, Comparison between Methodologies Recommended for Estimation of CO 2 Storage Capacity in Geological Media by the CSLF Task Force on CO2 Storage Capacity Estimation and the USDOE Capacity and Fairways Subgroup of the Regional Carbon Sequestration Partnerships Program, CSLF-T-2008-4, available at: http://www.cslforum.org/publications/documents/ PhaseIIIReportStorageCapacityEstimationTaskForce0408.pdf (accessed February 2010). Cuffey K M and Vimeux F (2001) Covariation of carbon dioxide and temperature from the Vostok ice core after deuterium-excess correction, Nature, 412, 523–527. © Woodhead Publishing Limited, 2010

92


Curry T, Reiner D, Ansolabehere S and Herzog H (2005) How aware is the public of carbon capture and storage? in Rubin E S, Keith D W and Gilboy C F (eds), Proceedings of the Seventh International Conference on Greenhouse Gas Control Technologies: GHGT7, IEA GHG, Cheltenham, UK, 1, 1001–1010. Damen K, Faaij A, van Bergen F, Gale J and Lysen E (2005) Identification of early opportunities for CO2 sequestration–worldwide screening for CO2-EOR and CO2 ECBM projects, Energy, 30(10), 1931–1952. Daveler S A (1998) XTOOL 7.15 user’s manual, UCRL-MA-130654, Lawrence Livermore National Laboratory, Livermore, CA. Davis T L, Terrell M J, Benson R D, Cardona R, Kendall R R and Winarsky R (2003) Multicomponent seismic characterization and monitoring of the CO2 flood at Weyburn Field, Saskatchewan, The Leading Edge, 22(7), 696–697. Davison J, Freund P and Smith A (2001) Putting Carbon Back in the Ground, IEA Greenhouse Gas R&D Programme, Cheltenham, UK. Davisson M L and Criss R E (1996) Na–Ca–Cl relations in basinal fluids, Geochimica et Cosmochimica Acta, 60, 2743–2752. Debye P and Hückel E (1923a) The theory of electrolytes, I Lowering of freezing point and related phenomena, Physik Zeitschrift, 24, 185–206. Debye P and Hückel E (1923b) The theory of electrolytes. II The limiting law of electrical conductivity, Physik Zeitschrift, 24, 305–325. Dooley J J, Dahowski R T, Davidson C L, Wise M A, Gupta N, Kim S H and Malone E L (2006) Carbon Dioxide Capture and Geologic Storage, A technology report from the second phase of the global energy technology strategy program, Battelle, Joint Global Change Research Institute. Doughty C, Benson S and Pruess K (2002) Capacity investigation of brine-bearing sands for geologic sequestration of CO2, Paper LBNL-51333, Lawrence Berkeley National Laboratory, Berkeley, CA, available at: http://repositories.cdlib.org/lbnl/LBNL-51333 (accessed January 2010). Doughty C, Freifeld B M and Trautz R C (2007) Site characterization for CO2 geologic storage and vice versa: the Frio brine pilot, texas, USA as a case study, Environmental Geology, 54(8), 1635–1656. Drummond S E (1981) Boiling and mixing of hydrothermal fluids: chemical effects on mineral precipitation, PhD Thesis, Pennsylvania State University. Duan Z and Sun R (2003) An improved model calculating CO2 solubility in pure water and aqueous NaCl solutions form 273 to 533 K and from 0 to 2000 bar, Chemical Geology, 193, 257–271. Duan Z H, Moller N and Weare J H (1995) Equation of state for the NaCl–H2O–CO2 system: prediction of phase equilibria and volumetric properties, Geochimica et Cosmochimica Acta, 59, 2869–2882. Duan Z, Sun R, Zhu C and Chou I-Ming (2006) An improved model for the calculation of CO2 solubility in aqueous solutions containing Na+, K+, CA2+, Mg2+, Cl-, and SO42–, Marine Chemistry, 98, 131–139. Ecofys & TNO-NITG (2002) Global carbon dioxide storage potential and costs, Report no EEP-02001, Utrecht, the Netherlands. EIA (2006) International Energy Annual 2006 (June–December 2008), energy Information Administration, Washington DC, available at: www.eia.doe.gov/iea (accessed January 2010). Emberley S, Hutcheon I, Shevalier M, Durocher K, Gunter W D and Perkins E H (2004) Geochemical monitoring of fluid-rock interaction and CO2 storage at the Weyburn



93

CO2-injection enhanced oil recovery site, Saskatchewan, Canada, Energy, 29(9–10), 1393–1401. Emberley S, Hutcheon I, Shevalier M, Durocher K, Mayer B, Gunter W D, and Perkins E (2005) Monitoring of fluid–rock interaction and CO2 storage through produced fluid sampling at the Weyburn CO2-injection enhanced oil recovery site, Saskatchewan, Canada, Applied Geochemistry, 20, 1131–1157. Ennis-King J and Paterson L (2001) Reservoir engineering issues in the geological disposal of carbon dioxide, in Williams D J, Durie R A, McMullan P, Paulson C A J and Smith A Y (eds), Proceedings of the Fifth International Conference on Greenhouse Gas Control Technologies: GHGT5, CSIRO Publishing, Collingwood, VIC, 290–295. Estublier A and Lackner A S (2009) Long-term simulation of the Snøhvit CO2 storage, in Gale J, Herzog H and Braitsch J (eds), Greenhouse Gas Control Technologies 9, Proceedings of the Ninth International Conference on Greenhouse Gas Control Technologies (GHGT9), Energy Procedia, 1, 3221–3228. Falkowski P, Scholes R J, Boyle E, Canadell J, Canfield D, Elser J, Gruber N, Hibbard K, Högberg P, Linder S, Mackenzie F T, Moore B, III; Pedersen T, Rosenthal Y, Seitzinger S, Smetacek V and Steffen W (2000) The global carbon cycle: A test of our knowledge of earth as a system, Science, 290, 291–296. Flett M, Beacher G, Brantjes J, Burt A, Dauth C, Koelmeyer F, Lawrence R, Leigh S, McKenna J, Gurton R, Robinson IV W F and Tankersley T (2008) Gorgon Project: Subsurface Evaluation of Carbon Dioxide Disposal under Barrow Island, SPE Asia Pacific Oil and Gas Conference and Exhibition, 20–22 October, Perth, Australia, 116372-Ms. Frailey S M, Finley R J and Hickman T S (2006) CO2 sequestration: storage capacity guideline needed, Oil & Gas Journal, 104(30), 44–46, 48–49. Gale J and Freund P (2001) Coal-bed methane enhancement with CO2 sequestration worldwide potential, Environmental Geosciences, 8(3), 210–217. Gale J, Christensen N P, Cutler A and Torp T A (2001) Demonstrating the potential for geological storage of CO2: the Sleipner and GESTCO projects, Environmental Geosciences, 8(3), 160–165. Gaus I, Azaroual M and Czernichowski-Lauriol I (2005) Reactive transport modelling of the impact of CO2 injection on the clayey cap rock at Sleipner (North Sea), Chemical Geology, 217, 319–337. Giammar D E, Bruant Jr. R G and Peters C A (2005) Forsterite dissolution and magnesite precipitation at conditions relevant for deep saline aquifer storage and sequestration of carbon dioxide, Chemical Geology, 217, 257–276. Gislason S R, Broecker W, Oelkers E H, Gunnlaugsson E, Stefánsson A, Matter J and Björnsson G (2008) Mineral CO2 sequestration into basalt: the Hellisheidi, Iceland project, International Geological Congress, August 6–14, Oslo, Norway. Goff F and Lackner K S (1998) Carbon dioxide sequestering using ultramafic rocks, Environmental Geoscience, 5(3), 89–101. Goldberg D S, Takahashi T and Slagle A L (2008) Carbon dioxide sequestration in deep-sea basalt, PNAS, 105, 9920–9925. Golubev S V, Pokrovsky O S and Schott J (2005) Experimental determination of the effects of dissolved CO2 on the dissolution kinetics of Mg and Ca silicates at 25 °C, Chemical Geology, 217, 227–238. Gunter W D, Perkins E H and McCann T J (1993) Aquifer disposal of CO2-rich gases reaction design for added capacity, Energy Conversion and Management, 34, 941–948.


94


Gunter W D, Wiwchar B and Perkins E H (1997) Aquifer disposal of CO2 -rich greenhouse gases: extension of the time scale of experiment for CO2-sequestering reactions by geochemical modeling, Mineralogy and Petrology, 59, 121–140. Gunter W D, Perkins E H and Hutcheon I (2000) Aquifer disposal of acid gases: modeling of water–rock reactions for trapping of acid wastes, Applied Geochemistry, 15, 1085–1095. Gunter W D, Bachu S and Benson S (2004) The role of hydrogeological and geochemical trapping in sedimentary basins for secure geological storage of carbon dioxide, Geological Society, London, Special Publications, 233, 129–145. Gupta N, Sass B, Sminchak J, Naymik T and Bergman P (1999) Hydrodynamics of CO2 disposal in a deep saline formation in the mid-western United States, in Riemer P, Eliasson B, Wokaun A (eds), Proceedings of the Fourth International Conference on Greenhouse Gas Control Technologies: GHGT4, Elsevier Science, Oxford, UK, 157–162. Gupta N, Sass B and Ickes (2000) Experimental evaluation of chemical sequestration of carbon dioxide in deep aquifer media – phase ii, Technical Progress Report Prepared for C. Byrer, US DOE National Energy Laboratory, Morgantown, WV. Han W, Lee S, Lu C, Thorne D, Esser R and McPherson B J (2008) Analysis of heterogeneity length-scales on residual gas trapping and buoyancy-driven CO2 migration, AGu, Fall Meeting 2008, 15–19 December, San Francisco, CA, abstract #H12C-04. Harvie C E and Weare J H (1980) The prediction of mineral solubilities in natural waters: the Na–K–Mg–Ca–Cl–SO4–H2O system from zero to high concentration at 25 °C, Geochimica Cosmochimica Acta, 44, 981–997. Helgeson H C, Kirkham D H and Flowers G C (1981) Theoretical prediction of the thermodynamic behavior of aqueous electrolytes at high pressures and temperatures: IV. Calculation of activity coefficients and relative partial molar properties to 600 °C and 5 kb, American Journal of Science, 281, 1249–1516. Hendriks C A and Blok K (1995) Underground storage of carbon dioxide, Energy Conversion and Management, 36, 539–542. Hendriks C, Graus W and van Bergen F (2004) Global carbon dioxide storage potential and costs, Report Ecofys & The Netherland Institute of Applied Geoscience TNO, Ecofys Report EEP02002. Hermanrud C, Andresen T, Eiken O, Hansen H, Janbu A, Lippard J, Bolas H N, Simmenes T H, Teige G M and Ostmo S (2009), Storage of CO2 in saline aquifers–lessons learned from 10 years of injection into the Utsira Formation in the Sleipner area, in Gale J, Herzog H and Braitsch J (eds), Greenhouse Gas Control Technologies 9, Proceedings of the Ninth International Conference on Greenhouse Gas Control Technologies (GHGT9), Energy Procedia, 1, 1997–2004. Hesse M A, Orr Jr F M and Tchelepi H A (2008) Gravity currents with residual trapping, Journal of Fluid Mechanics, 611, 35–60. Hitchon B (1996) Aquifer disposal of carbon dioxide-hydrodynamic and mineral trapping–proof of concept, report of a three-year study by the Alberta Research Council, Alberta, Canada, Geoscience Publishing Ltd, Sherwood Park, AB. Hitchon B, Gunter W D, Gentzis T and Bailey R T (1999) Sedimentary basins and greenhouse gases: a serendipitous association, Energy Conversion and Management, 40, 825–843. Holloway S (1997) Safety of the underground disposal of carbon dioxide, Energy Conversion and Management, 38(Suppl.), S241–S245. Holtz M H (2002) Residual gas saturation to aquifer influx: A calculation method for



95

3–D computer reservoir model construction, SPE Paper 75502, SPE Gas Technologies Symposium, 30 April-2 May, Calgary, AB, Canada. Holtz M H (2003) Pore-scale influences on saline aquifer CO2 sequestration, AAPG Annual Convention, 11–14 May, Salt Lake City, UT. Holtz M and Bryant S (2005) Effects of pore network geometry on permanent storage of sequestered CO2, AAPG’s 2005 Annual Meeting, 19–22 June, Calgary, AB, Canada. Holtz M H, Doughty C, Yeh J and Hovorka S (2004) Modeling of CO2 saline aquifer sequestration and the effects of residual phase saturation, AAPG Annual Convention, 16–21 April, Dallas, TX. Hovorka S D, Doughty C and Holtz M H (2005) Testing efficiency of storage in the subsurface: Frio Brine Pilot Experiment, in Wilson M, Morris T, Gale J and Thambimuthu K (eds), Proceedings of the Seventh International Conference on Greenhouse Gas Control Technologies: GHGT7, IEA GHG, Cheltenham, UK, 2, 1361–1366. Hovorka S D, Benson S M, Warne A G, Doughty C, Freifeld B M, Sakurai S, Daley T M, Kharaka Y K, Holtz M H, Trautz R C, Nance H S, Myer L R and Knauss K G (2006) Measuring permanence of CO2 storage in saline formations: the Frio experiment, Environmental Geosciences, 13(2), 105–121. Hovorka S D, Choi J-W, Meckel T A, Trevino R H, Zeng H, Kordi M, Wang F P and Nicot J-P (2009) Comparing carbon sequestration in an oil reservoir to sequestration in a brine formation—field study, in Gale J, Herzog H and Braitsch J (eds), Greenhouse Gas Control Technologies 9, Proceedings of the Ninth International Conference on Greenhouse Gas Control Technologies (GHGT9), Energy Procedia, 1, 2051–2056. H R 2454 (2009) American Clean Energy and Security Act (ACES), A Bill Before the US House of Representatives. Ibitayo O (2002) Public-private partnerships in the siting of hazardous waste facilities: The importance of trust, Waste Management & Research, 20, 212–222. Ide S T, Jessen K and Orr Jr F M (2007) Storage of CO2 in saline aquifers: effects of gravity, viscous, and capillary forces on amount and timing of trapping, International Journal of Greenhouse Gas Control, 1, 481–491. IEA (1995) Carbon Dioxide Disposal from Power Stations, IEA Greenhouse Gas R&D Programme, Paris, France. IPCC (2005) IPCC Special Report on Carbon Dioxide Capture and Storage, Working Group III of the Intergovernmental Panel on Climate Change, Metz B, Davidson O, de Coninck H C, Loos M and Meyer L A (eds), Cambridge, UK, Cambridge University Press. IPCC (2007) Climate Change 2007: Synthesis Report, Contribution of Working Groups I, II and III to the Fourth Assessment Report of the Intergovernmental Panel on Climate Change, Core Writing Team, Pachauri RK and Reisinger A (eds), IPCC, Geneva, Switzerland. Johnson J W and Lundeen, S R (1994a) Jewel: A graphical-user interface for generating custom GEMBOCHS thermodynamic datafiles for use with geochemical modeling software, LLNL-YMP Milestone report MOL63, Lawrence Livermore National Laboratory, Livermore, CA. Johnson J W and Lundeen S R (1994b) GEMBOCHS thermodynamic datafiles for use with the EQ3/6 software package, LLNL-YMP Milestone report MOL72, Lawrence Livermore National Laboratory, Livermore, CA. Johnson J W, Oelkers E H and Helgeson H C (1992) SUPCRT92: A software package for calculating the standard molal thermodynamic properties of minerals, gases and


96


aqueous species from 0 to 1000 °C and 1 to 5000 bar, Computers and Geosciences, 18, 799–847. Johnson J W, Nitao J J, Steefel C I and Knauss K G (2001) Reactive transport modeling of geologic CO2 sequestration in saline aquifers: the influence of intra-aquifer shales and the relative effectiveness of structural, solubility, and mineral trapping during prograde and retrograde sequestration. First National Conference on Carbon Sequestration, 14–17 May, Washington, DC. Joyce D B and Holloway J R (1993) An experimental determination of the thermodynamic properties of H2O–CO2–NaCl fluids at high pressures and temperatures, Geochimica et Cosmochimica Acta, 57, 733–746. Kårstad O (2002) Geological storage, including costs and risks, in saline aquifers. In: Proceedings of the IPCC workshop on carbon capture and storage, 18–21 November, Regina, SK, Canada. Kaszuba J P, Janecky D R and Snow M G (2003) Carbon dioxide reaction processes in a model brine aquifer at 200 °C and 200 bars: implications for geologic sequestration of carbon, Applied Geochemistry, 18(7), 1065–80. Keeling C D and Whorf T P (2005) Atmospheric CO2 records from sites in the SIO air sampling network, in Trends: A Compendium of Data on Global Change. Carbon Dioxide Information Analysis Center, Oak Ridge National Laboratory, US Department of Energy, Oak Ridge, TN. Keyes F G and Kirkwood J G (1930) The dielectric constant of carbon dioxide as a function of temperature and density, Physical Review, 36, 754–761. Kharaka Y K and Barnes J A (1973) SOLMINEQ: solution-mineral equilibrium computations, Computer Contributions Report, U.S. Geological Survey, Reston, VA, 215–899. Kharaka Y K and Hanor J S (2007) Deep fluid in the continents: I. Sedimentary Basins, in Drever J I (ed.) Treatise on Geochemistry, Vol. 5, Surface and Ground Water, Weathering, and Soils, Elsevier Pergamon, Amsterdam, the Netherlands, 1–48. Kharaka Y K, Gunter D, Aggarwal L K, Perkins E H and DeBraal D (1988) SOLMINEQ.88: a computer program for geochemical modeling of water rock interactions, WaterResources Investigations Report Series, 88-4227, US Geological Survey, Reston, VA. Kharaka Y K, Cole D R, Thordsen J J, Kakouros E and Nance H S (2006a) Gas–water–rock interactions in sedimentary basins: CO2 sequestration in the Frio Formation, Texas, USA, Journal of Geochemical Exploration, 89, 183–186. Kharaka Y K, Cole D R, Hovorka S D, Gunter W D, Knauss K G and Freifeld B M (2006b) Gas-water-rock interactions in Frio Formation following CO2 injection: Implications for the storage of greenhouse gases in sedimentary basins, Geology, 34(7), 577–580. King M B, Mubarak A, Kim J D and Bott T R (1992) The mutual solubilities of water with supercritical and liquid carbon dioxide. Journal of Supercritical Fluids, 5, 296–302. Kitamura K and Xue Z (2006) An experimental study of residual gas saturation of carbon dioxide in water-saturated porous sandstone by using multi-channel seismic wave imaging method, AGU, Fall Meeting 2006, 11–15 December, San Francisco, CA, abstract #H51E-0531. Knauss K G, Steefel C I, Johnson J W and Boram L H (2002) Impact of CO2, contaminant gas, aqueous fluid, and reservoir rock interactions on the geologic sequestration of CO2, Geological Society of America National Meeting, 27–30 October, Denver, CO.



97

Knauss K G, Johnson J W and Steffel C I (2005) Evaluation of the impact of CO2, co-contaminant gas, aqueous fluid and reservoir rock interactions on the geologic sequestration of CO2, Chemical Geology, 217, 339–350. Koide H, Tazaki Y, Noguchi Y, Nakayama S, Iijima M, Ito K and Shindo Y (1992) Subterranean containment and long-term storage of carbon dioxide in unused aquifers and in depleted natural gas reservoirs, in Blok K, Turkenburg W, Hendriks C and Steinberg M (eds), Energy Conversion and Management, 33(5–8), 619–626. Kolak J J and Burruss R C (2006) Geochemical investigation of the potential for mobilizing nonmethane hydrocarbons during carbon dioxide storage in deep coal beds, Energy & Fuels, 20(2), 566–574. Kuuskraa V A (2004) Estimating CO2 storage capacity in saline aquifers, Proceedings, National Energy Technology Laboratory Third Annual Conference on Carbon Capture & Sequestration, 3–6 May, Alexandria, VA, CD-ROM. Leonenko Y and Keith D W (2008) Reservoir engineering to accelerate the dissolution of CO2 stored in aquifers, Environmental Science and Technology, 42, 2742–2747. Lindeberg E and Wessel-Berg D (1997) Vertical convection in an aquifer column under a gas cap of CO2, Energy Conversion and Management, 38 (Suppl.), S229–S234. Maldal T and Tappel I M (2004) CO2 underground storage for Snøhvit gas field development, Energy, 29(9–10), 1403–1411, 6th International Conference on Greenhouse Gas Control Technologies. Malinin S D (1959) The system water – carbon dioxide at high temperatures and pressures, Geokhimiya, 3, 292–306. Marini L (2007) Geological Sequestration of Carbon Dioxide: Thermodynamics, Kinetics, and Reaction Path Modeling, Elsevier, Amsterdam, the Netherlands. Maroto-Valer M M, Fauth D J, Kuchta M E, Zhang Y and Andrésen J M (2005) Activation of magnesium rich minerals as carbonation feedstock materials for CO2 sequestration, Fuel Processing Technology, 86(14–15), 1627–1645. Marsily G de (1986) Quantitative Hydrogeology, Academic Press, San Diego, CA. Matter J M, Goldberg D S and Takahashi T (2005) Potential of basalt aquifers for the permanent sequestration of anthropogenic carbon dioxide, AGU, Fall Meeting, 5–9 December, San Francisco, CA, abstract #B33A–1010. Mayer B, Johnson G, Nightingale M, Maurice S, Raistrick M, Taylor S, Hutcheon I and Perkins E (2008) Isotopic approaches to evaluate the fate of injected CO2 in two geological storage projects in mature oilfields in Canada, AGU, Fall Meeting 2008, 11–15 December, San Francisco, CA, abstract #H13K-04. McPherson B J O L and Lichtner P C (2001) CO2 sequestration in deep aquifers, Proceedings First National Conference on Carbon Sequestration, 14–17 May, Washington, DC. McRae L E and Holtz M H (1995) Reservoir architecture and permeability characteristics of fluvial-deltaic sandstone reservoirs, Frio Formation, Rincon Field, South Texas, AAPG Bulletin, 79, 64. Michael K, Allinson G, Golab A, Sharma S and Shulakova V (2009) CO2 storage in saline aquifers II – experience from existing storage operations, in, Gale J, Herzog H and Braitsch J (eds), Greenhouse Gas Control Technologies 9, Proceedings of the Ninth International Conference on Greenhouse Gas Control Technologies (GHGT9), Energy Procedia, 1, 1973–1980. Nitao J J (1998) Reference manual for the NUFT flow and transport code, version 2.0, UCRL-MA-130651, Lawrence Livermore National Laboratory, Livermore, CA. NRG Associates (2001) The significant oil and gas fields of the United States, NRG Associates, Inc., Colorado Springs, CO [includes data current as of May 23, 2001–


98


database available from NRG Associates, Inc., P.O. Box 1655, Colorado Springs, CO 80901]. Nuttall B C, Drahovzal J A, Eble C and Bustin R M (2005) CO2 sequestration in gas shales of Kentucky, AAPG Annual Convention, 19–22 June, Calgary, AB, Canada. Oelkers E H and Cole D R (2008) Carbon dioxide sequestration: A solution to a global problem, Elements, 4, 305–310. Oelkers E H and Schott J (2005) Geochemical aspects of CO2 sequestering, Chemical Geology, 217(3–4), 183–186. Oelkers E H, Schott J and Devidal J-L (1994) The effect of aluminum, pH, and chemical affinity on the rates of aluminosilicate dissolution reactions, Geochimica et Cosmochimica Acta, 58, 2011–2024. Omerod W (1994) Carbon dioxide disposal from power stations, IEA Greenhouse Gas R&D Programme, Cheltanham, UK. Palandri J K and Kharaka Y K (2004) Sequestration of CO2 in redbeds: Siderite precipitation from CO2 and SO2 waste gas, in Wanty R B and Seal R R, II (eds), Water–Rock Interaction, Proceedings of the Eleventh International Symposium on Water–Rock Interaction, A.A. Balkema, Rotterdam, the Netherlands 1, 567–571. Palandri J L and Kharaka Y K (2005) Ferric iron-bearing sediments as a mineral trap for CO2 sequestration: iron reduction using sulfur-bearing waste gas, Chemical Geology, 217, 351–364. Palandri J L, Rosenbauer R J and Kharaka Y K (2004) Experimental studies of CO2 sequestration in ferric iron-bearing sediments – CO2–SO2 reaction with hematite, Proceedings, National Energy Technology Laboratory Third Annual Conference on Carbon Sequestration, 3–6 May, Alexandria, VA, CD-ROM. Palandri J L, Rosenbauer R J and Kharaka Y K (2005) Ferric iron in sediments as a novel CO2 mineral trap – CO2–SO2 reaction with hematite, Applied Geochemistry, 20, 2038–2048. Palmer T N and Räisänen J (2002) Quantifying the risk of extreme seasonal precipitation events in a changing climate, Nature, 415, 512–514. Parkhurst D L (1995) User’s guide to PHREEQC–A computer program for speciation, reaction-path, advective-transport, and inverse geochemical calculations, WaterResources Investigation Report Series, 95–4227, US Geological Survey, Reston, VA. Parkhurst D L, Thorstenson D C and Plummer L N (1980) PHREEQE–A computer program for geochemical calculations, Water-Resources Investigation Report Series, 80–96, US Geological Survey, Reston, VA. (revised and reprinted August, 1990). Perez R J, Shevalier M, Hutcheon I and Mayer B (2006) A model for partitioning gases among brines and hydrocarbons in oil reservoirs: Examples from the IEA-GHG Weyburn CO2 Monitoring and Storage Project, Saskatchewan, Canada, Journal of Geochemical Exploration, 89, 326–330. Perkins E H and Gunter W D (1995a) A user’s manual for b PATHARCH.94: A reaction path-mass transfer program, ENVTR 95–11, Alberta Research Council, Edmonton, AB. Perkins E H and Gunter W D (1995b) Aquifer disposal of CO2-rich greenhouse gases: Modeling of water–rock interaction paths in a siliciclastic aquifer, in Kaharaka Y K, Chudeav, O V (eds), Water–rock Interactions, Brookfield, Rotterdam, the Netherlands, 895–898. Pitzer K S (1973) Thermodynamics of electrolytes. I Theoretical basis and general equations, Journal of Physical Chemistry, 77, 268–277. © Woodhead Publishing Limited, 2010


99

Pitzer K S (1975) Thermodynamics of electrolytes. V Effects of higher–order electrostatic terms, Journal of Solution Chemistry, 4, 249–265. Piri M, Prevost J H and Fuller R (2005) Carbon dioxide sequestration in saline aquifers: evaporation, precipitation and compressibility effects, Proceedings DOE/National Energy Technology Laboratory Fourth Annual Conference on Carbon Capture and Sequestration, 2–5, May, Alexandra, VA. Plummer L N, Parkhurst D L, Fleming G W and Dunkle S A (1988) A computer program incorporating Pitzer’s equations for calculation of geochemical reactions in brines, Water-Resources Investigations Report Series, 88–4153, US Geological Survey, Reston, VA. Pokrovsky O S, Schott J and Thomas F (1999) Processes at the magnesium-bearing carbonates/solution interface: I. A surface speciation model of magnesite, Geochimica et Cosmochimica Acta, 63, 863–880. Pokrovsky O S, Golubev S V and Schott J (2005) Dissolution kinetics of calcite, dolomite and magnesite at 25 °C and 0 to 50 atm pCO2, Chemical Geology, 217, 239–255. Portier S and Rochelle C (2005) Modeling CO2 solubility in pure water and NaCl-type waters from 0 to 300 °C and from 1 to 300 bar. Application to the Utsira Formation at Sleipner, Chemical Geology, 217, 187–199. Preston C, Whittaker S, Rostron B, Chalaturnyk R, White D, Hawkes C, Johnson J W, Wilkinson A and Sacuta N (2009) IEA GHG Weyburn-Midale CO2 monitoring and storage project – moving forward with the Final Phase, in Gale J, Herzog H and Braitsch J (eds), Greenhouse Gas Control Technologies 9, Proceedings of the Ninth International Conference on Greenhouse Gas Control Technologies (GHGT9), Energy Procedia, 1, 1743–1750. Prevost J H, Fuller R, Altevogt A S, Bruant R and Scherer (2005) Numerical modeling of carbon dioxide injection and transport in deep saline aquifers, in Wilson M, Morris T, Gale J and Thambimuthu K (eds), Proceedings of the Seventh International Conference on Greenhouse Gas Control Technologies: GHGT7, IEA GHG, Cheltenham, UK, Vol. 2, 2189–2193. Pruess K (2005) Numerical studies of fluid leakage from a geologic disposal reservoir for CO2 show self-limiting feedback between fluid flow and heat transfer, Geophysical Research Letters, 32, L14404. Pruess K, Oldenburg C and Muridis G (1999) TOUGH2, User’s Guide, Version 2.0, Report LBNL-43134, Lawrence Berkeley National Laboratory. Berkeley, CA. Pruess K, Tsang C F, Law D H-S and Oldenburg C M (2001a) An inter-comparison study of simulation models for geologic sequestration of CO2, Proceedings, First National Conference on Carbon Sequestration, 14–17 May, Washington, DC. Pruess K, Tsang C F, Law D and Oldenburg C (2001b) Inter-comparison of simulation models for CO2 disposal in underground storage reservoirs, Report LBNL-47353, Lawrence Berkeley National Laboratory, Berkeley, CA. Pruess K, Xu T, Apps J A and Garcia J (2001c) Numerical modeling of aquifer disposal of CO2, SPE paper 66537, SPE Exploration and Production Environmental Conference, 26–28 February, San Antonio, TX. Pruess K, Tsang C, Law D H and Oldenburg C M (2002) An inter-comparison study of simulation models for geologic sequestration of CO2, in Gale J and Kaya Y (eds), Proceedings of the Sixth International Conference on Greenhouse Gas Control Technologies: GHGT6, Elsevier (Pergamon), Oxford, UK, Vol. 1, 507–510. Pruess K, García J, Kovscek T, Oldenburg C, Rutqvist J, Steefel C and Xu T (2004) Code intercomparison builds confidence in numerical simulation models for geologic disposal of CO2, Energy, 29(9–10), 1431–1444. © Woodhead Publishing Limited, 2010

100


Reed M H (1982) Calculation of multicomponent chemical equilibria and reaction processes in systems involving minerals, gases, and an aqueous phase, Geochimica et Cosmochimica Acta, 46, 513–528. Riding J and Rochelle C (2005) The IEA Weyburn CO2 monitoring and storage project, final report of the European research team, British Geological Survey, Keyworth, UK. Riding J and Rochelle C (2009) Subsurface characterization and geological monitoring of the CO2 injection operation at Weyburn, Saskatchewan, Canada, Geological Society, London, Special Publications, 313, 227–256. Rosenbauer R J (2006) Experimental multi-phase CO2–H2O brine–rock interactions at elevated temperature and pressure: Implications for CO2 sequestration in deep-saline aquifers, Eos Trans. AGU, 87(36), West. Pac. Geophys. Meet. Suppl., 24–27 July, Beijing, China, Abstract V34B-01. Rosenbauer R J and Bischoff J L (2009) CO2 Sequestration by basalt: experimental studies and geochemical modeling, Goldschmidt Conference Abstracts 2009, Geochimica at Cosmochimica Acta., 54. Rosenbauer R J and Koksalan T (2002) Experimental determination of the solubility of CO2 in electrolytes: application to CO2 sequestration in deep-saline aquifers, GSA Abstracts with Programs, Annual Meeting, 27–30 October, Denver, CO, 135, Vol. 2, 304. Rosenbauer R J and Koksalan T (2003) Experimental CO2-brine-rock interactions at elevated temperature and pressure: implications for the geologic sequestration of CO2 in deep-saline aquifers, Proceedings, 226th American Chemical Society National Meeting, 7–11 September, New York, NY. Rosenbauer R J, Bischoff J L and Koksalan T (2001) An experimental approach to CO2 sequestration in saline aquifers: Application to Paradox Valley, CO, Eos Trans. AGU, 83(47), Fall Meeting, San Francisco, CA, Suppl., Abstract 2002V32B–0974. Rosenbauer R J, Koksalan T and Palandri J L (2005) Experimental investigation of CO2–brine–rock interactions at elevated temperature and pressure: implications for CO2 sequestration in deep-saline aquifers, Fuel Processing Technology, 86(12–14), 1581–1597. Sass B M, Engelhard M H, Berman P and Byrer C J (2002) Interaction of rock minerals with carbon dioxide and brine: A hydrothermal investigation, Energy & Environmental Research, 2, 23–31. Schilling F R, Borm G, Wuerdeman H, Moeller F, Kuehn M and Liebscher A (2008) Status Report on the First European on-shore CO2 Storage Site at Ketzin (Germany), Eos Trans. AGU, 89(53), Fall Meeting, San Francisco, CA, Suppl., Abstract U44A-04. Shafeen A, Croiset E, Douglas P L and Chatzis I (2004) CO2 sequestration in Ontario, Canada. Part I: storage evaluation of potential reservoirs, Energy Conversion and Management, 45, 2645–2659. Sharma S, Cook P, Berly T and Lees M (2009) The CO2CRC Otway Project: Overcoming challenges from planning to execution of Australia’s first CCS project, in Gale J, Herzog H and Braitsch J (eds), Greenhouse Gas Control Technologies 9, Proceedings of the Ninth International Conference on Greenhouse Gas Control Technologies (GHGT9), Energy Procedia, 1, 1965–1972. Shiraki R and Dunn T L (2000) Experimental study on water–interactions during CO2 flooding in the Tensleep Formation, Wyoming USA, Applied Geochemistry, 15, 265–279. Shock E L (1998) An updated and augmented version (slop07.dat) of the original



101

SUPCRT92 database (sprons92.dat) is available at: http://zonvark.wustl.edu/geopig/ (accessed February 2010). Sorai M, Ohsumi T, Ishikawa M and Tsukamoto K (2007) Feldspar dissolution rates measured using phase-shift interferometry: implications to CO 2 underground sequestration, Applied Geochemistry, 22, 2795–2809. Spycher N, Pruess K and Ennis-King J (2003) CO2–H2O mixtures in the geological sequestration of CO2. I. Assessment and calculation of mutual solubilities from 12 to 100 °C and up to 600 bar, Geochimica et Cosmochimica Acta, 67, 3015–3031. Steefel C I (2001) GIMRT, version 1.2: Software for modeling multicomponent, multidimensional reactive transport. User’s Guide, UCRL-MA-143182, Lawrence Livermore National Laboratory, Livermore, CA. Steefel C I (2008) CRUNCHFLOW: Software for multicomponent reactive transport: USER’S MANUAL, Lawrence Livermore National Laboratory, Livermore, CA. Steefel C I and Yabusaki S B (1996) OS3D/GIMRT, Software for multicomponentmultidimensional reactive transport: User’s Manual and Programmer’s Guide, PNL11166, Pacific Northwest National Laboratory, Richland, WA. Stevens S H, Kuuskraa V A, Gale J and Beecy D (2001) CO2 injection and sequestration in depleted oil and gas fields and deep coal seams: worldwide potential and costs, Environmental Geosciences, 8(3), 200–209. Stewart P B and Munjal P (1970) Solubility of carbon dioxide in pure water, synthetic sea water, and synthetic sea water concentrates at 5° to 25° C and 10 to 45 atm. Pressure. Journal of Chemical and Engineering Data, 15, 67–71. St John B, Bally A W and Klemme H D (1984) Sedimentary provinces of the world hydrocarbon productive and nonproductive, American Association of Petroleum Geologists, Tulsa, OK. Stocker T F and Schmittner A (1997) Influence of CO2 emission rates on the stability of the thermohaline circulation, Nature, 388, 862–865. Strazisar B R, Zhu C and Hedges S W (2006) Preliminary modeling of the long-term fate of CO2 following injection into deep geologic formations, Environmental Geosciences, 13(1), 1–15. Takenouchi S and Kennedy G C (1964) The binary system H2O–CO2 at high temperatures and pressures, American Journal of Science, 262, 1055–1074. Tanase D and Yoshimura T (2008) Nagaoka CO2 injection and monitoring project; a gateway of the intimate understanding of CO2 behavior in the deep reservoir, International Geological Congress, 6–14 August, Oslo, Norway. Todheide K and Franck E U (1963) Das Zweiphasengebiet und die kritische Kurve im System Kohlendioxid-Wasser bis zu Drucken von 3500 bar, Zeitschrift fur Physikalische Chemie, 37, 387–401. Torp T A and Gale J (2004) Demonstrating storage of CO2 in geological reservoirs: The Sleipner and SACS projects, Energy, 29(9–10), 1361–1369, 6th International Conference on Greenhouse Gas Control Technologies. United States Department of Energy, National Energy Technology Laboratory (2008), Carbon Sequestration Atlas of the United States and Canada, March. Vangkilde-Pedersen T, Lyng Anthonsen K, Smith N, Kirk K, Neele F, van der Meer B, Le Gallo Y, Bossie-Codreanud D, Wojcicki A, Le Nindre Y-M, Hendriks C, Dalhoff F and Christensen NP (2009) Assessing European capacity for geological storage of carbon dioxide – the EU GeoCapacity project, in Gale J, Herzog H and Braitsch J (eds), Greenhouse Gas Control Technologies 9, Proceedings of the Ninth International Conference on Greenhouse Gas Control Technologies (GHGT9), Energy Procedia, 1, 2663–2670. © Woodhead Publishing Limited, 2010

102


Vermylen J P, Hagin P N and Zoback M D (2008) Feasibility assessment of CO2 sequestration and enhanced recovery in gas shale reservoir, AGU, Fall Meeting 2008, 15–19 December, San Francisco, CA, abstract #H23D–0990. White S P (1995) Multiphase non-iosthermal transport of systems of reacting chemicals, Water Resources Research, 31(7), 1761–1772. White C M, Strazisar B R, Granite E J, Hoffman J S and Pennline H W (2003) Separation and capture of CO2 from large stationary sources and sequestration in geological formations–coalbeds and deep saline aquifer, Journal of Air and Waste Management Association, 53, 645–715. White S P, Allis R G, Moore J, Chidsey T, Morgan C, Gwynn W, Adams M, (2005) Simulation of reactive transport of injected CO2 on the Colorado Plateau, Utah, USA. Chemical Geology, 217, 387–405. Wiebe R and Gaddy V L (1939) The solubility in water of carbon dioxide at 50, 75 and 100°, at pressures to 700 atmospheres, Journal of the American Chemical Society, 61, 315–318. Wiebe R and Gaddy V L (1941) Vapor phase composition of carbon dioxide-water mixtures at various temperatures and at pressures to 700 atmospheres, Journal of the American Chemical Society, 63, 475–477. Wolery T J (1979) Calculation of chemical equilibrium between aqueous solution and minerals–The EQ3/6 software package, Report UCRL-52658, Lawrence Livermore National Laboratory, Livermore, CA. Wolery T J (1992) EQ3/6, A software package for geochemical modeling of aqueous systems: Package overview and installation guide (Version 7.0), Report UCRL-MA110662 PT 1, Lawrence Livermore National Laboratory, Livermore CA. Wolery T J and Daveler S A (1992) A computer program for reaction path modeling of aqueous geochemical solutions: Theoretical manual, user’s guide, and related documentation (version 7.0), Report UCRL-MA-110662 PT IV, Lawrence Livermore National Laboratory, Livermore, CA. Worden R H (2006) Dawsonite cement in the Triassic Lam Formation, Shabwa Basin, Yemen. A natural analogue for a potential mineral product of subsurface CO 2 storage for greenhouse gas reduction, Marine Petroleum Geology, 23, 67–77. Xu T and Pruess K (1998) Coupled modeling of non-isothermal multiphase flow, solute transport and reactive chemistry in porous and fractured media: 1. Model development and validation, Report LBNL-42050, Lawrence Berkeley National Laboratory, Berkeley, CA. Xu T, Apps J A and Pruess K (2000) Analysis of mineral trapping for CO2 disposal in deep aquifers, Report LBNL-47315, Lawrence Berkeley National Laboratory, Berkeley, CA. Xu T, Pruess K and Apps J A (2004a) Numerical simulation to study mineral trapping for CO2 disposal in deep aquifers, Applied Geochemistry, 19, 917–936. Xu T, Sonnenthal E, Spycher N and Pruess K (2004b) TOUGHREACT user’s guide: a simulation program for non-isothermal reactive geochemical transport in variable saturated geologic media, Report LBNL-55460, Lawrence Berkeley National Laboratory, Berkeley, CA. Xu T, Sonnenthal E, Spycher N and Pruess K (2005) TOUGHREACT – A simulation program for non-isothermal reactive geochemical transport in variable saturated geologic media: Applications to geothermal injectivity and CO2 geologic sequestration, Computers and Geosciences, 32(2), 145–165. Xu T, Chen S and Zhang D (2006) Convective stability analysis of the long-term storage



103

of carbon dioxide in deep saline aquifers, Advances in Water Resources, 29(3), 397–407. Zweigel P, Arts R, Lothe A E and Lindeberg E B G (2004) Reservoir geology of the Utsira Formation at the first industrial-scale underground CO2 storage site (Sleipner area, North sea), in Baines S J and Worden R H (eds), Geological Storage of Carbon Dioxide, GSA Special Publication 233, Geological Society of America, Boulder, CO, 165–180.


4

Carbon dioxide (CO2) sequestration in oil and gas reservoirs and use for enhanced oil recovery (EOR)

B. V e g a and A. R. K o v s c e k, Stanford University, USA Abstract: It is frequently said that oil and gas reservoirs are likely to be the first category of geological formation where carbon dioxide (CO 2) shall be injected for greenhouse gas sequestration on a large scale, if sequestration proves feasible. Carbon dioxide is injected into comparatively few reservoirs at the present time. It is estimated, however, that 80 % of oil reservoirs worldwide might be suitable for CO2 injection to enhance oil recovery. Enhanced oil recovery operations with CO2 have been limited by the availability and cost of CO2, but not necessarily candidate reservoirs. The problem of co-optimizing oil production and CO2 storage differs dramatically from current gas injection practice because of the cost– benefit difference due to the purchase cost of CO2 for enhanced recovery projects. When low-cost CO2 becomes widely available, injection into a wider range of reservoirs is foreseen, with the objective of maximizing the amount of CO2 left in the reservoir at abandonment. In addition to discussion of the conventional oil reservoir setting, we demonstrate, using laboratory experiments, the applicability and potential of low-permeability unconventional hydrocarbon reservoirs to store significant volumes of CO 2. Key words: enhanced oil recovery, carbon dioxide sequestration, storage capacity, unconventional oil production.

4.1

Introduction

This article discusses the storage of greenhouse gases, mainly carbon dioxide (CO2), in oil and gas reservoirs. Carbon dioxide is the most often considered greenhouse gas; hence, the focus on CO2 here. Oil and gas reservoirs, by the very fact that hydrocarbons accumulated, are effective in preventing the upward migration of buoyant fluids over long periods of time. Likewise, gas injection is a widely practised method to enhance the production of oil (Moritis, 2006). Surface and subsurface infrastructure already exists in oil and gas fields that could be adapted to CO2 distribution and storage. Although there has been significant activity in saline aquifer storage (cf. Benson et al. (2005) for a review), there are substantial benefits to injection of anthropogenic CO2 into sedimentary basins containing oil and gas (Kuuskraa and Ferguson, 2008). Notably, CO2 used for enhanced oil recovery (EOR) addresses two major barriers to geological sequestration. In the oil-field EOR setting, mineral (i.e., pore space) rights are well established as are 104 © Woodhead Publishing Limited, 2010

Carbon dioxide (CO2) sequestration in oil and gas reservoirs

105

the questions of permitting and liability. Additionally, coupling of CO2 sequestration and EOR creates a market for CO2 that may help to defray partially the cost of CO2 capture from industrial sources of CO2. Notably, the total so-called ‘stranded’ oil remaining in USA oil reservoirs after conventional recovery efforts is 400 billion barrels. It is estimated that injection of this CO2 will result in the production of an additional 39–48 billion barrels of this oil in the USA alone (Kuuskraa and Ferguson, 2008). The infrastructure created for CO2–EOR is then usable for subsequent CO2 storage in saline aquifers. The remainder of this introduction provides an overview of CO2 sequestration volume estimates in the oil and gas setting as well as a discussion of the potential for co-optimized EOR and CO2 storage. The carbon intensity of both crude oil and natural gas is put into context with respect to CO 2. Subsequent sections explore the mechanisms by which CO2 enhances oil recovery and provide a snapshot of the state of the art of CO2–EOR operations. The differences between standard CO2–EOR and a co-optimized process whereby the volume of oil recovered and the volume of CO2 placed in the ground are maximized are brought out. Perhaps the most intriguing aspect of the potential widespread availability of CO2 is that CO2 injection into unconventional hydrocarbon resources may become more commonplace. The geological setting of fractured and low-permeability reservoirs is attractive as a sink for CO2 where the costs of sequestration are offset by recovery of oil and gas. The applicability of CO2 injection in this setting is explored extensively. A summary and conclusions complete the Chapter.

4.1.1 Oil reservoirs Sequestration of anthropogenic CO2 in oil fields is a reality today. Consider the case of the Rangely Field (Colorado, USA). Roughly 180 miles (295 km) to the northwest of Rangely, a large gas separation facility in LaBarge Wyoming separates natural gas from CO2 and hydrogen sulfide (H2S) (Hunter and Bryan, 1987). The natural gas is shipped to consumers and the H2S is re-injected near the production site. Instead of venting all of the CO2 into the atmosphere, roughly three-quarters of the volume is compressed and shipped via pipeline to the Rangely Field (Bleizeffer, 2008). Because this naturally occurring CO2 would be released to the atmosphere as a result of human activities, it is classified as anthropogenic. At Rangely, the CO2 is injected for EOR operations. The mechanisms of oil recovery enhancement by CO2 are discussed in a subsequent section. This discussion includes miscible and immiscible scenarios. To date, the incremental production that results directly from CO2 injection is more than 120 million bbl of oil, representing an additional displacement of about 7–8 % of the original oil in place (OOIP) within the reservoir (Hervy and


106


Iakovakis, 1991; Masoner and Wackowski, 1995). At Rangely, process economics are driven by the sale of oil and, because the CO2 is purchased, the volume of CO2 injected is minimized to reduce cost. Hence, significant reservoir engineering effort is currently expended to reduce the amount of CO2 needed to recover a barrel of oil. In this sense, current EOR efforts are not optimized to maximize the volume of CO2 stored. Experience with CO2–EOR at Rangely is representative of many aspects of the state-of-the-art for injection of CO2 into oil reservoirs. Green and Willhite (1998) tabulate the results of 22 field- and pilot-scale miscible CO2 injection projects. The average incremental production is 12.6 % OOIP with a low of 7.5 and a high of 22 % OOIP. The field-scale projects reviewed inject, on average, 6.3 MCF (0.33 t) of CO2 to recover a barrel of oil. The process of converting a field to CO2–EOR, including the drilling of new wells, and the injection of substantial quantities of CO2 generally results in increased production rates of oil. The Weyburn (Saskatchewan, Canada) project injects CO2 from a coal gasification plant in North Dakota (USA). Production at the beginning of 2006 was roughly 26 000 bbl/day which is an increase of about 13 000 bbl/day as compared to production volumes prior to the initiation of the Weyburn project (PTRC, 2009). The maximum predicted production rate at Weyburn when CO2 injection is fully implemented is about 31 000 bbl/day. Additional elements of CO2–EOR include: (i) long project lifetimes of 25–40 years; (ii) greater engineering manpower requirements to design, monitor, and remediate problems as compared to conventional oil recovery; (iii) premature breakthrough of CO2 to production wells and the need to manage CO2 mobility within the reservoir to limit CO2 injection; (iv) design of surface facilities to inject and handle the separation of produced oil, water, gas, and CO2 including appropriate materials selection for wells, surface lines, and valves; (v) well surveillance to ensure integrity of well equipment and the cement that bonds the well to the earth. Many of these elements related to the implementation of CO2 injection at East Vacuum (New Mexico, USA) are discussed by Brownlee and Sugg (1987), Harpole and Hallenbeck (1996), Brock and Bryan (1989), Chang and Grigg (1998), and references therein.

4.1.2 Gas reservoirs Carbon dioxide is not injected into gas reservoirs in any significant volumes at present. Nevertheless, studies have shown that the injection of CO2 accelerates natural gas production from a gas reservoir by providing repressurization (Van der Burgt et al., 1992; Oldenburg et al., 2001). Economic analysis suggests that at CO2 purchase prices of $4–10 USD per tonne (t = 1000 kg), enhanced gas recovery is economic (Oldenburg et al., 2004). Process economics are



107

most sensitive to the revenue from natural gas sales and the purchase cost of CO2. The major advantage of injecting CO2 into a gas reservoir is that CO2 is more dense and more viscous than natural gas at virtually all reservoir conditions. Accordingly, gravitationally and viscously stable floods are possible. Average gas recovery factors are quoted as 75 % (Oldenburg et al., 2004). A depletion-drive gas reservoir might have a remaining gas saturation of 10–20 % whereas a water-drive gas reservoir might display 30–40 % remaining gas saturation. In addition to accelerated gas recovery, this remaining gas is a target for enhanced recovery operations. Given the fact that CO2 is not currently injected to enhance gas production from conventional gas reservoirs, we place somewhat less emphasis on conventional gas reservoirs in comparison to oil reservoirs with associated gas.

4.1.3 Capacity estimates In the geological setting of oil and gas reservoirs, a variety of capacity estimates have been made for CO2 storage. For example, the storage capacity for oil and gas reservoirs combined is put at 840–1850 GtCO2 (Parson and Keith, 1998) whereas the IEA has established a volume of 920 GtCO2 that represents a capacity to sequester roughly 25 % of worldwide CO2 emissions over a span of 50 years (IEA, 2008). About one half of the capacity is in natural gas reservoirs. The IPCC consensus values of capacity range from 675–900 GtCO2 (Benson et al., 2005). Hence, CO2-based EOR co-optimized with storage may play a role in reducing the rate of increase of CO2 in the atmosphere. Additionally, the IPCC report (Benson et al., 2005) presents in some detail the methodologies employed for estimating storage capacity and discussing the limited data available for such calculations thereby leading to uncertainty. To put these volumes into another perspective, the carbon emissions reduction wedges concept of Pacala and Socolow (2004) defines a wedge as an activity that reduces atmospheric CO2 emissions by a cumulative of 92 GtCO2 over the span of 50 years. Hence, oil and gas reservoirs represent roughly 10–20 wedges whereas Pacala and Socolow propose that seven wedges solve the carbon and climate problem for a half of a century. On the other hand, a more sobering statistic is that about 87 000 t/day of CO2 are currently injected for EOR worldwide (Moritis, 2006). Over 50 years, this rate accumulates to 1.3 GtCO2 injected. Hence, the volumes of CO2 injected for EOR and stored in a reservoir need to increase roughly 70-fold to grow to represent a wedge.


108


4.1.4 Carbon density Carbon neutrality is mentioned often in relation to carbon sequestration. Many oil industry observers view carbon neutrality of sequestration operations as an ultimate goal (Kuuskraa, 2008). That is, oil or gas is removed from the reservoir, the energy in the hydrocarbon is released, and the resulting CO 2 is put back into the reservoir. Kovscek (2002) illustrates several calculations in this area. He considers the carbon density in oil or gas in comparison to the carbon density in CO2 as a function of pressure and temperature. Generally, the carbon density of liquid hydrocarbons is significantly greater than that of CO2 whereas the carbon density of natural gas is slightly less than that of CO2. For a range of representative reservoir conditions, the carbon density of CO2 is about 20 % greater than that of natural gas. In virtually all cases of practical interest, CO2 is more carbon dense in comparison to natural gas. All of the CO2 from combustion of a volume natural gas can be stored in the original gas reservoir at prevailing reservoir pressure and temperature with a little room to spare. On the other hand, the carbon density of CO2 is only about one-quarter that of crude oil. This implies that carbon neutrality is not feasible in the sense of storing all CO2 produced from combustion of oil in the same volume occupied by that oil in the reservoir. An alternate methodology to view carbon neutrality is to ask whether the total reservoir volume can hold the CO2 created from combusting the oil produced as a result of continuing oil production and EOR. This methodology is relevant because most oil reservoirs worldwide have already produced significant oil volumes to date, implying the creation of significant voidage to be replaced with CO2. The reservoir volume once filled with oil is now filled with water or gas and is also available for CO2 storage. As an illustration, we again consider the Rangely Field. At the onset of CO2 injection, the cumulative recovery of oil was estimated as 37 % of the OOIP (Masoner and Wackowski, 1995). In 1999, the cumulative recovery to date was reported as 43 % of OOIP (Friedmann et al., 1999). For the sake of illustration, let us assume that the total recovery of oil increases to 50 % of the OOIP as a result of EOR and continued operations from the beginning of CO2 injection. The volume of oil produced during the time of CO2 injection is thus 13 % of the OOIP. Again for the sake of illustration, let us assume that the carbon density of crude oil is four times that of CO2 at reservoir conditions. Then, the reservoir volume required to sequester all of the CO2 produced from combusting the oil is 52 % OOIP (= 4 * 13 % OOIP). Thus, the reservoir volume needed to sequester the CO2 emissions of the oil produced, and be carbon neutral, is nearly equal to the volume of the total produced oil.



4.2

109

Carbon dioxide (CO2) enhanced recovery mechanisms

The introductory discussion assumes that an advantage of CO2 injection for EOR is the potential to offset the costs of sequestration through enhanced hydrocarbon recovery. In this section, important aspects related to phase and flow behavior of CO2 injection during enhanced recovery are discussed.

4.2.1 Immiscible versus miscible recovery The displacement efficiency associated with the injection of a fluid into a porous medium is determined by multiphase flow characteristics and the phase behavior among injection gas, crude oil, and reservoir brine. In the case of the flow of immiscible fluids, two important factors to consider are interfacial tension (IFT) among phases and rock wettability. Under immiscible conditions, the presence of two, or more, distinct phases is apparent and evidenced by the existence of a sharp interface. In the case of immiscible displacement, IFT plays a role in the competition among the phases to occupy pore space. In conventional waterflooding and other immiscible displacement fluid systems, large values of IFT are associated with large values of residual oil saturation. Ultimately, oil recovery performance for immiscible CO2 injection is limited by volumetric sweep, displacement efficiency of the injected fluid, and the finite solubility of CO2 in the oil phase that, generally, reduces oil viscosity and oil density. These factors are, in turn, determined by reservoir structural heterogeneities, gravity effects, viscous fingering, rock wettability, crude oil phase behavior, and so on. In most cases, a significant fraction of the oil contacted by the displacing phase is trapped or isolated within unreachable pore spaces because capillary forces immobilize the oil. Thus, the recovery process halts and reduces the oil relative permeability to near zero. The ultimate oil recovery after immiscible fluid injection ranges between 20 and 40 % OOIP, on average (Stalkup, 1984).

4.2.2 Development of miscibility When two miscible phases flow together, they become indistinguishable and the fluids mix in all proportions. Because there is no interface, there is no interfacial tension between the oil and the solvent. Also, the single mixed phase is not affected by relative permeability or wettability of the rock to oil or solvent. In most cases of miscible displacement, microscopic recovery efficiency approaches 100 %, in the absence of water. Hence, the residual oil saturation is quite small, or zero, where the injected phase contacts the oil.


110


Depending on the mixing mechanism of the miscible system composed by injected solvent and oil, the miscible displacement process will be firstcontact miscible (FCM) or multiple-contact miscible (MCM). In an FCM displacement, the injected fluid forms a single phase with the reservoir oil upon first contact, whereas in MCM miscibility is developed in situ through changes in composition to both the injected and the displaced fluids as they advance through the reservoir. Generally, CO2 develops miscibility with oil by the MCM mechanism. A pseudoternary diagram is useful to conceptualize the FCM recovery processes. One vertex represents the heaviest fraction of the oil, C7+, another represents the intermediate components, C2–6, and the third represents gas, C1. On Fig. 4.1, the straight line that spans oil and injected fluid composition points lies outside the two-phase dome that defines multiphase hydrocarbon behavior at reservoir pressure and temperature. Any combination of injectant and oil results in a single-phase fluid. When an FCM solvent is injected for secondary recovery in a batch volume or slug in the reservoir, oil and solvent immediately mix and form a single phase in the contact zone and oil is banked downstream of the contact zone, as shown in Fig. 4.2. The miscible bank features lower viscosity than the pure oil as a consequence

as

er eg ion

100 % C1

Tw o

-ph

Solvent

Oil 100 % C7+

100 % C2–6

4.1 Example of a pseudoternary diagram for a FCM process.

Solvent

(primary slug)

Oil + Solvent

Oil

(miscible zone)

(oil bank)

4.2 Miscible displacement.



111

of the dispersion and molecular diffusion of the solvent in the oil phase. Any trapped oil contacted by the miscible bank is solubilized and therefore displaced. As the concentration profile advances through the reservoir, an increasingly larger volume of the original solvent slug is dispersed in the oil. At the same time, because the viscosity of the injected solvent is usually significantly lower than the oil viscosity, and in the absence of IFT affecting phase relative permeability, the mobility ratio for miscible processes is unfavorable and there is a tendency to viscous fingering. Another potential problem in FCM displacements is asphaltene deposition due to the effect of the solvent on the heaviest fractions of the oil. These may cause pore-plugging, reduce the effective rock permeability, and alter the injectivity and productivity of wells. MCM displacement processes develop miscibility through in situ composition changes prompted by mass transfer and multiple contacts (i.e., mixing) between the solvent and the oil phase. The interplay of solvent composition with the original fluid plays a major role in the type of displacement that results. The ultimate goal of an MCM process is to follow a composition path that brings the mixture to the miscible region of the pseudoternary diagram. This mixing is assumed to take place under equilibrium conditions, up to the miscible region, by condensation of the C2–C6 components from the solvent to the oil. This process is called a condensing-gas drive, and Fig. 4.3 illustrates a typical case. If the injected solvent is mostly composed of methane or natural gas, and the oil composition is high on the intermediate-molecular-weight components, C1

Gas 1

Oil

Liq 1 Liq 2

Lim

itin

Mix 2 Mix 1

Injection gas

g ti e li ne

Gas 2

C7+

C2–6

4.3 Pseudoternary diagram for a case of condensing-gas drive miscibility process.


112


then the gas is subsequently enriched on several equilibrium contacts until its composition is brought to the miscible limiting tie line value. This process is a vaporizing-gas drive as illustrated in Fig. 4.4. It applies mostly to injection fluids such as natural gas, flue gas or nitrogen, provided that miscible conditions, mainly pressure, are possible in the reservoir.

4.2.3 Carbon dioxide (CO2) displacement mechanisms The parameter summarizing the combination of phase behavior and flow is the MMP or minimum miscibility pressure. The MMP is conventionally defined as the pressure needed to recover 90 % of the oil originally in place from a one-dimensional laboratory slim tube with the injection of 1.2 pore volumes of CO2, where pore volumes are computed at test pressure and temperature. Practically, MMP is the pressure necessary to assure the mutual solubility of oil and CO2 and thereby achieve significant recovery. MMP varies with oil composition and density and generally increases as oil becomes more dense. Enhanced recovery mechanisms with CO2 include oil-phase swelling, viscosity reduction, and gas–oil displacement when CO2 is below MMP with the oil. Above MMP, the miscible CO2 injection process is believed to be similar to the vaporizing-gas drive mass transfer process that takes the light fractions out of the crude oil and enriches the gas with them. However, CO2 is capable of extracting higher molecular weight components from the oil than methane, and this makes the required miscibility pressures for C1 Injection gas Gas 1 Gas 2

line

Mix 1

Liq 2

Lim

itin

Liq 1

g ti e

Mix 2

Oil C7+

C2–6

4.4 Pseudoternary diagram for a case of vaporizing-gas drive miscibility process.



113

the CO2 significantly lower than those required for injection gases such as methane, flue gas, natural gas, nitrogen, etc. Also, for a given pressure and temperature, the two-phase region defined by the binodal curve in the pseudoternary diagram is much smaller for the CO2 system than for the methane system at reservoir conditions, Fig. 4.5. These constitute the main advantages of the miscible displacement with CO2 over other gases. In spite of many favorable factors, miscible displacement with CO2 is affected by the unfavorable mobility ratio that is usually present in solvent– oil systems. Reservoir heterogeneities compound the problem, leading to channeling of injectant through zones of large permeability and poor volumetric sweep. Viscous fingering of injectant through oil also occurs due to the high mobility ratio. Because the densities of oil and CO2 are similar at a wide range of reservoir conditions (Stalkup, 1984), gravity segregation between the fluids is relatively unimportant in these systems. The injected solvent volume is typically limited for economic reasons. It is common practice to inject water as a secondary slug to push the injectant deep into the reservoir. Water injection is also useful to achieve mobility control when multiple slugs of CO2 are injected. This is the principle of the so-called water-alternating-gas (WAG) process where volumes of water and CO2 are injected in series. The ratio of injected water and gas volume, the socalled WAG ratio, is a major design parameter. During WAG, the combined mobility of the two phases is less than that of the gas injected alone, and so the mobility ratio of the process is improved and the sweep efficiency of the gas injectant is increased. The injection of water, however, tends to increase the amount of residual oil. Because water reaches some of the oil-filled pore C1 or CO2

CO2 phase boundary C1 phase boundary

Limiting tie line

C7+

C2–6

4.5 Representative comparison of the binodal curves for CO2 and methane.


114


space first, water blocks the advance of CO2 and the subsequent recovery of oil. This effect has been found to be a strong function of the rock wettability and more detrimental in water-wet rocks (Green and Willhite, 1998). At the same time, because CO2 is partially soluble in water, additional watertrapped oil phase can be recovered from occluded pore volumes through mass transfer. Green and Willhite (1998) mention the work from Campbell and Orr, in which they explained the mechanism as CO2 that solubilizes and then diffuses in the water, and then solubilizes in the oil. Subsequently, CO 2 swells the oil and the increase of volume pushes the water phase out of the pore throat, allowing for direct contact with more CO2. A balance of these two counteracting mechanisms can yield a reasonable balance of improved mobility, reduced oil phase trapping, and increased CO2 capture by the oil phase.

4.2.4 Carbon dioxide (CO2) injection in low-permeability fractured systems When there are significant resources still trapped in the matrix block of a fractured reservoir, gas injection is often performed to activate a gravity drainage recovery mechanism. The difference in density between the gas phase in the fracture and the oil phase in the matrix causes oil production until gravitational and capillary forces become equal. When the matrix also presents low permeability and high capillary pressure, injection of a dry gas with the consequent mass transfer between the gas in the fracture and the oil/gas system in the matrix becomes the main recovery mechanism (Kazemi and Jamialahmadi, 2009). Enhanced oil and gas recovery from fractured low-permeability reservoir rock is challenging and, while studied, significant challenges remain in improving recovery factor. Geological and structural properties of the reservoir have a profound impact on both the microscopic and macroscopic diffusion processes that take place on a CO2 miscible displacement. Low permeability compounds the challenge of CO2 accessibility to oil. Additional factors limit process effectiveness, such as asphaltene deposition, water slug occlusion of pores, and unfavorable rock wettability, especially under immiscible conditions (Vega et al., 2008). Near the well, such factors are collectively referred to as ‘formation damage’ or a well ‘skin’ factor. On the other hand, fractures offer relatively low resistance to flow, thereby increasing the effective permeability of the reservoir. Injected CO2 is expected intuitively to flow primarily through the low-flow resistance network of fractures rather than the high-flow resistance matrix. Exchange of CO2 between the fractures and the matrix is thus key to both CO2 storage as well as enhanced recovery. Pressure, chemical composition, and gravitational gradients are driving forces relevant to CO2 exchange.



115

According to the equation for steady-state injectivity defined by Dake (1978),

I=

q 2p k = hDp m(ln(re /rw ) + S )

[4.1]

where q is the volumetric flow rate at the bottom of the well, h is the formation thickness, D p is the pressure drop between the reservoir and the well, k is formation permeability, m is the injected phase viscosity, r the radius, and S is the mechanical skin accounting for formation damage near the well. Subscripts ‘e’ and ‘w’ refer to equivalent drainage radius of the well and the wellbore radius, respectively. Note that I is directly proportional to permeability and inversely proportional to the viscosity of the injected phase. However, an attractive feature of CO2 injection is the relatively low viscosity of the CO2 phase. At 15 MPa and 47 °C, CO2 phase viscosity is only 0.077 mPa-s (Lohrenz et al., 1964) as compared to 0.68 mPa-s for water. Due to its relatively low viscosity, volumetric injection rates of CO2 can be large in both permeable and low-permeability formations (Kovscek, 2002). As suggested by Equation 4.1, even generally low-permeability formations can accept large volumes of CO2 and the storage rate is enhanced by regions of high permeability. A well intersecting many fractures thereby encounters a formation with an effective permeability that is substantially greater than the matrix permeability. Interestingly, heterogeneous, high-permeability paths are generally viewed in a negative fashion for CO2-based EOR. Efficiency of oil recovery is reduced by high-permeability paths and gravity segregation that promotes incomplete reservoir sweep. This is another point where conventional EOR differs from simultaneous sequestration and EOR. CO2 injectivity is an important factor that needs to be considered while designing a sequestration and/or EOR project. It is affected by factors pertinent to the reservoir characteristics and the fluids. Some of these factors include reactions between the CO2 and some of the rock minerals and reservoir brine salts. In some cases, reservoir permeability will increase by dissolving minerals from the walls of flow channels or will decrease permeability by releasing particles that migrate and plug pores and throats in the rock. Also, CO2 might extract some of the medium to heavy components and asphaltene components of the oil could precipitate and reduce CO2 permeability in the reservoir. WAG cycles also reduce the relative permeability of both water and CO2 and injectivity will progressively decrease unless there are other mitigating factors. Diverse results in field trials have been obtained in a range of injection conditions, some with lower than anticipated resulting injectivity for both CO2 and water (Stalkup, 1984). Vega et al. (2008) suggest that primary depletion of low-permeability oilfilled rock creates continuous gas pathways from the fracture to the matrix because, where the reservoir has been pressure depleted and gas released © Woodhead Publishing Limited, 2010

116


from solution, significant remaining gas saturation has been created in the matrix. These gas pathways act as a rough course the CO2 may follow and that aids the ability to deliver injectant to the matrix. They argue that in many cases injected CO2 follows the pre-established gas pathways, the CO2 flow slows, and CO2 mixes with the contacted oil mainly through molecular diffusion, although convective dispersion is also expected to occur.

4.3

Co-optimization of enhanced oil recovery (EOR) and carbon storage

Based on criteria such as reservoir depth and the density of the crude oil, it has been estimated that upwards of 80 % of oil reservoirs would show recovery improvements if the reservoir was subjected to CO2 injection (Taber et al., 1997a,b). While EOR applications using anthropogenic CO2, such as Rangely and Weyburn, are relatively rare at the current time, the geological settings where CO2 injection would benefit oil recovery are numerous. An issue of primary importance to the selection of candidate reservoirs for miscible gas injection is the minimum pressure at which the injected gas develops miscibility with the oil, that is, the MMP. The MMP increases as oil density increases. For this reason, reservoir depth and the gravity of the crude oil are primary screening parameters for miscible CO2 injection. An heuristic based on field experience is that injected CO2 will not develop miscibility unless the oil density is less than 900 kg/m3 (22 °API) and the reservoir depth exceeds 750 m (2500 ft) (Taber et al., 1997a). Miscible injection can be attained at shallower depths for a spectrum of oils. Nevertheless, injection into reservoirs shallower than 2500 ft needs to be considered carefully given the risk of inducing fractures. An important aspect for site selection of simultaneous EOR and CO2 sequestration is the integrity of the reservoir seals including the caprock and reservoir bounding faults. Overpressurization of fluids within reservoir pore space breaches any type of barrier (c.f., Finkbeiner et al., 2001). Leakage through damaged casings and the cement plugs of old and abandoned wells is another area of risk (Nordbotten et al., 2005). Geomechanical site characterization is needed to assess the likelihood of injection-induced stress leading to fault slippage and overpressurization causing hydraulic fracturing of the caprock. Chiaramonte et al. (2008) illustrate a workflow that characterizes seals and traps in the context of CO2 sequestration in ageing oil fields. Although rarely mentioned as a factor to consider during reservoir selection for CO2–EOR, reservoir temperature does play a role with respect to development of miscibility. Generally, the MMP decreases as temperature decreases, all other factors held the same. Carbon dioxide injection in the Permian Basin (USA) has benefitted from reservoir temperatures that



117

are somewhat lower than expected given the earth’s average geothermal temperature gradient (Stalkup, 1984). Figure 4.6 illustrates the correlation between depth and oil density for a selection of active CO2 injection projects (Moritis, 2006). This figure demonstrates that the screening values are merely suggestions. There are a number of rather shallow projects that are likely immiscible. Notable immiscible CO2 injection projects are Bati Raman (Turkey) (Sahin et al., 2008), Lost Hills (Kern Co., CA USA) (Perri et al., 2000), and Lick Creek (Bradley and Union Co., AR USA) (Reid and Robinson, 1981). In addition to oil density, depth, and temperature, other reservoir characteristics of successful CO2 injection projects are oil saturation above 20 % and effective confinement of CO2 within the reservoir. In the scenario of greatly reduced or negative cost for CO2, the criterion of minimum oil saturation is likely to be relaxed as the pore space for storage of CO2 gains value. With respect to CO2 saturated pore space gaining value, traditional reservoir engineering techniques where a large volume of water is injected into the reservoir, usually in a sequential fashion that alternates water and CO2 (WAG), are not conducive to co-optimizing oil recovery and CO2 storage. Pore space becomes filled with water and that pore space could be better used to store CO2. Kovscek and Cakici (2005) show that one pathway to co-optimized oil recovery and carbon storage is to reduce significantly the volume of water injected and to adopt a well control strategy that limits the fraction of gas produced relative to the oil produced. They use a synthetic, 0

Depth (m)

1000

2000

3000

4000

5000 600

700

800 900 Oil density (kg/m3)

1000

4.6 Correlation between depth and oil density at standard conditions for CO2–EOR projects in the USA.


118


three-dimensional reservoir model with a compositional description of the reservoir fluid to investigate various well control schemes via reservoir flow simulation. A scheme that shuts in wells producing large volumes of gas and allows wells to open as reservoir pressure increases is the most successful strategy for co-optimization. Injection wells operate continuously. The well control procedure performs optimally, and results from several equiprobable geostatistical representations of permeability within the reservoir suggest that the well control strategy is general. Simulation also indicates that recovery employing immiscible CO2 injection and the well control scheme is virtually the same as an optimized WAG process. Additionally, more than 50 % of the reservoir pore volume becomes filled with CO2 whereas a WAG process employing the optimal WAG ratio of one volume of gas for every one volume of water fills less than 20 % of the pore volume with CO2. Kovscek and Cakici (2005) investigate in a similar fashion miscible recovery and CO2 storage. The solvent gas is a mixture of CO2, ethane, propane, and butane. Adopting the well control scheme and switching from the solvent gas to pure CO2 injection midway through the recovery process allows recovery of about 80 % of the initial oil in place. Again, CO2 fills more than 50 % of the reservoir volume. Further, the energy required to recompress produced gas for re-injection is not found to be a limiting factor for the well control strategy. The oil recovered is energy dense and the energy for recompression is a few per cent of the produced oil energy. In a related study, Jessen et al. (2005) investigated approaches to increasing CO2 storage in oil recovery. They show that intelligent design of well completions can create injection profiles that reduce the effects of preferential flow of CO2 through high-permeability zones, thereby increasing the reservoir volume contacted by CO2. They also show that reservoir pressurization after the end of the producing life of the field is useful to increase CO2 storage, provided that oil and gas production activities have not damaged the reservoir seal. Additionally, an unconventional suggestion is to inject CO2 into the saline aquifer that generally underlies most oil fields. This injection scheme is less prone to cycling of CO2 from injector to producer, and oil trapped in the vertical capillary transition zone between aquifer and reservoir may be produced.

4.4

Future trends: geologic storage in tight rocks

EOR process applicability is not limited to a particular type of formation, such as carbonate or sandstone. Because CO2 viscosity is low in comparison to oil and water and injectivity is inversely proportional to viscosity, as Equation 4.1 shows, the viscosity of injected CO2 is much less of a limiting factor and injection is relatively easy in all types of formation. In short, formation type and thickness have not been factors that limit oil-recovery



119

performance. This suggests that CO2 may be injected into tight reservoirs for both EOR and CO2 storage. Similar to EOR, recovery of natural gas from tight rocks including coal and shale may be enhanced by CO2 injection. In coal beds and gas shales where a significant volume of natural gas is adsorbed to the solid surface, injection of CO2 and/or mixtures of CO2 and N2 increases ultimate recovery of natural gas while simultaneously leaving CO2 adsorbed to the solid in place of natural gas (e.g., Jessen et al., 2008). Injection of CO2 also helps to establish pressure gradients that drive natural gas toward producers. In this case, the microscopic displacement is miscible and stable. Displacement stability is achieved because CO2 viscosity, while low, is greater than methane. Improving resource recovery from low-permeability resources is strategically important. For instance, it is estimated that about one half of the worldwide petroleum resource is found in fractured siliceous and carbonaceous formations, yet fractured resources make up only roughly 20 % of reserves (Saidi, 1983). As a greater number of reservoirs have matured and are reaching abandonment, recovery options are needed for more difficult to produce hydrocarbon settings. Examples of sizeable low-permeability hydrocarbon resources include the Monterey Shale (CA, USA), West Texas (USA) Carbonates, the North Sea Chalks, and the Asmari Limestone (Iran). To illustrate the potential and difficulties of EOR in low-permeability resources, we present the results of a series of experiments using low permeability (0.02–1.3 mD), medium porosity (30–40 %) siliceous shale reservoir core samples. Cores are initially saturated with either: live oil, depleted to a pressure of 200–300 psi; or dead oil brought to miscible conditions; followed by CO2 injection at pressures proceeding from immiscible to CO2 miscible conditions. In these experiments, two gas injection modes were used: CO2 flow across one face of the core (countercurrent flow) which intends to account for the diffusive transfer mechanism; and CO2 flow along the length of the core (cocurrent injection), where convective displacement was also expected. The experimental set-up was monitored using X-ray computed tomography that helped to visualize phase flow and distribution during the processes. Results reported in this work include imaging techniques that provide images representative of the distribution and connectivity of gas along the central axis of the core sample. The images show CO2 distribution maps obtained by image reconstruction where the light shadowed areas indicate greater concentration or saturation of CO2 for miscible and immiscible tests, respectively. For the immiscible tests, CO2 injection was preceded by pressure depletion that aimed to reproduce the initial conditions of a secondary recovery process. The core was repressurized in stages to simulate an increase in reservoir


120


Dimensionless position, xd

Dimensionless position, xd

pressure. Sample resulting images for such tests are shown in Fig. 4.7. An increasing CO2 saturation becomes apparent as the CO2 injection progresses. The last stage shows an average CO2 saturation of 0.5–0.6 PV. It can also be observed that the CO2 roughly follows the pathways established by the gas produced during the pressure depletion stage. The study also offers images associated with CO2 injection under miscible conditions, like those shown in Fig. 4.8. According to these laboratory experiences, the incremental oil recovery potential for both immiscible and miscible CO2 injection seems significant. In particular, the incremental oil recovery caused by the immiscible CO2 injection ranged from 0–10 % for countercurrent flow mode and from 18–25 % for cocurrent flow mode, totaling 18–35 % of oil recovery. For the miscible injection, countercurrent flow yielded a 54 % oil recovery and the cocurrent flow allowed an additional 39 %, adding up to a total oil recovery of 93 % OOIP. Nevertheless, oil recovery potential by immiscible CO2 injection into siliceous shale rock is challenged by low permeability, rock heterogeneities, distribution of oil within the rock matrix, but it is aided by the presence

1300

1100

Countercurrent

400

900 800 700 600 Pressure [psi] Depletion

600

900 Pressure [psi] CO2 Immiscible injection

400

200

Cocurrent

900

4.7 Gas saturation maps for depletion/immiscible CO2 injection. Numbers below images indicate pressure in psi.



121

CO2 fraction maps for countercurrent miscible injection First cycle

Second cycle 1 0.9 0.8 0.7

Pre-prod

Post-prod

Pre-prod

0.6

Post-prod

0.5 Third cycle

0.4

Fourth cycle

0.3 0.2 0.1

Pre-prod

Post-prod

Pre-prod

Post-prod

0 CO2 fraction

CO2 fraction maps for cocurrent miscible injection First cycle

Pre-prod

Post-prod

Third cycle

Pre-prod

Post-prod

Second cycle

Pre-prod

Post-prod

Fourth cycle

Pre-prod

Post-prod

1 0.9 0.8 0.7 0.6 0.5 0.4 0.3 0.2 0.1 0 CO2 fraction

4.8 Pure CO2 distribution maps for miscible CO2 injection.

of continuous gas pathways that allow CO2 penetration into the matrix. Miscibility seemed to be achieved under multiple contact regime, and it was able to overcome the limitations faced by the immiscible process as demonstrated by the significant increase in oil recovery obtained under both countercurrent and cocurrent injection methods.

4.4.1 Implications on modeling of the carbon dioxide (CO2) injection process In the modeling of a fractured low-permeability system subject to miscible CO2 injection, it is necessary to account for the effects associated with immiscible and miscible displacement that have been previously described, such as the oil


122


swelling, viscosity, and interfacial tension reduction, among others. The use of a compositional simulator allows the definition of pseudoternary diagrams that facilitate the interpretation of the miscible processes, also accounting for the different components in each phase with more rigorous equation of state. Therefore, a good compositional analysis of the oil becomes necessary as an input. Fractures can be modeled by creating a high-permeability/highporosity area on the grid, but there are options to model them using dual porosity/dual permeability techniques.

4.5

Summary and conclusions

In summary, oil and gas reservoirs represent significant volume that could be put into use for long-term storage of CO2. Simple capacity estimates translate to 10–20 climate stabilization wedges where about seven wedges solve the carbon and climate problem (Pacala and Socolow, 2004). Because CO2 is already injected into reservoirs, albeit in a geographically limited area, a degree of acceptance already exists for the practice. Additionally, the opportunity to recover crude oil that was not accessible with standard oil recovery processes such as water injection and/or to accelerate the recovery of natural gas provides economic incentive. As discussed above, enhanced gas recovery employing CO2 is inherently carbon neutral. On the other hand, the available volume in depleted oil reservoirs provides the opportunity to sequester significant CO2. Simulation studies suggest that it should be possible to fill at least one-half of a reservoir’s volume with CO2 during oil production operations. A number of geological factors make oil and gas reservoirs attractive as well. Chief among these is that the reservoir seal held buoyant oil and gas for geological time periods. Provided that oil and gas operations did not damage the reservoir seal, the oil and gas reservoir setting presents a relatively secure and well characterized storage site. The accumulation of significant hydrocarbons in a reservoir also implies that the rock was sufficiently permeable to allow oil migration into the formation and that the void volume (i.e., porosity) was significantly large such that significant fluid volumes were stored in the rock. This chapter also explored the specific and challenging case of placing CO2 into an unconventional shale oil rock. For the immiscible tests, the incremental oil recovery caused by exposing the core to CO2 ranged from 0–10 % for the CO2 countercurrent injection flow mode and from 18–25 % for the cocurrent (forced) injection mode. Countercurrent injection recovery seems to be sensitive to the presence and distribution of the gas phase near the fracture face and within the rock matrix, as revealed by the differences between the immiscible tests (1 and 2). The oil recovery potential of immiscible gas injection remains challenged by factors such as low permeability (as



123

demonstrated by Test 1, performed on a lower perm core sample), rock heterogeneities, oil distribution, and pre-existent gas patterns. Nevertheless, these preliminary tests do indicate that immiscible CO2 may be a technically feasible EOR method based upon the encouraging recovery and relatively rapid oil production obtained in the laboratory. Thus, a new setting for geological carbon sequestration may be possible. At near miscible conditions, the incremental oil recovery caused by the CO2 injection was 25 % and 10 % of the OOIP for the countercurrent and cocurrent modes, respectively. The test reveals slightly greater total recovery than the immiscible tests previously completed. A miscible test may have resulted in even greater recovery. The tests of the EOR potential of siliceous shale resulting from exposure to CO2 continue. Recovery tests under true miscible conditions are underway. Because large pressure gradients are not imposed on the core in either countercurrent or cocurrent conditions and because core permeability is low, tests are relatively slow. One aspect that shall be explored in future tests is the incremental recovery from constant cocurrent injection of CO2 following the countercurrent and cocurrent tests as previously described. Additionally, a complete set of partial molar densities from a compositional description of the oil–CO2 miscible system would be required to validate the image processing technique for miscible and near miscible cases. Such a compositional description may also help to improve image processing.

4.6


CO2–EOR and greenhouse gas sequestration research is being carried out at different instances and scales worldwide and numerous works are being published on a regular basis. At print time, the following are some recommended reads: ∑

On the EOR side, some good sources are: – Green D and Willhite G P (1998) Enhanced Oil Recovery, Textbook Series Vol. 6. Society of Petroleum Engineers, Richardson, TX. – Orr F M (2007) Theory of Gas Injection Processes. 1st edn. Holte, Denmark: Tie-Line Publications. – Stalkup Jr F I (1984) Miscible Displacement, Monograph Series, 2nd ed, Society of Petroleum Engineers, Richardson, TX. ∑ On the CO2 sequestration/storage side, some sources are: – Benson S M and Cook P, Coordinating Lead Authors. J Anderson, S Bachu, HB Nimir, B Basu, J Bradshaw, G Deguchi, J Gale, G von Goerne, W Heidug, S Holloway, R Kamal, D Keith, P Lloyd, P Rocha, B Senior, J Thomson, T Torp, T Wildenborg, M Wilson, F Zarlenga and Zhou D, Lead Authors. M Celia, B Gunter, J Ennis King, E Lindegerg, S Lombardi, C Oldenburg, K Pruess, A Rigg, S © Woodhead Publishing Limited, 2010

124


Stevens, E Wilson and S Whittaker (2005) Underground geological storage, IPCC Special Report on Carbon Dioxide Capture and Storage, Intergovernmental Panel on Climate Change, Cambridge University Press, Cambridge, UK, 195–276. – Jessen K, Kovscek A R and Orr F M Jr (2005) Increasing CO2 storage in oil recovery, Energy Conversion and Management, 46, 293–311. – Kovscek A R (2002) Screening criteria for CO2 storage in oil reservoirs, Petroleum Science and Technology, 20(7&8), 841–866.

4.7

References

Benson S M and Cook P, Coordinating Lead Authors. J Anderson, S Bachu, HB Nimir, B Basu, J Bradshaw, G Deguchi, J Gale, G von Goerne, W Heidug, S Holloway, R Kamal, D Keith, P Lloyd, P Rocha, B Senior, J Thomson, T Torp, T Wildenborg, M Wilson, F Zarlenga and Zhou D, Lead Authors. M Celia, B Gunter, J Ennis King, E Lindegerg, S Lombardi, C Oldenburg, K Pruess, A Rigg, S Stevens, E Wilson and S Whittaker (2005) Underground geological storage, IPCC Special Report on Carbon Dioxide Capture and Storage, Intergovernmental Panel on Climate Change, Cambridge University Press, Cambridge, UK, 195–276. Bleizeffer D (2008) State Weighs ExxonMobil’s CO2 Venting, Caspar Star Tribune, May 14. Brock WR and Bryan LA (1989) Summary results of CO2 EOR field tests, 1972–1987, paper SPE 18977, Proceedings of the SPE Rocky Mountain Regional/Low Permeability Reservoirs Symposium and Exhibition, Denver, CO, 6–8 March. Brownlee MH and Sugg LA (1987) East Vacuum Grayburg-San Andres Unit CO 2 Injection Project: development and results to date, paper SPE 16721, Proceedings of the 62nd Annual Technical Conference and Exhibition of SPE, Dallas, TX, 27–30 September. Chang S-H and Grigg RB (1998) History matching and modeling the CO2-foam pilot test at EVGSAU, paper 39793, Proceedings of the SPE Permian Basin Oil and Gas Recovery Conference, Midland, TX, 23–26 March. Chiaramonte L, Zoback MD, Friedmann J and Stamp V (2008) Seal integrity and feasibility of CO2 sequestration in the teapot dome EOR pilot: geomechanical site characterization, Environmental Geology, 54(8) 1667–1675. Dake LP (1978) Fundamentals of Reservoir Engineering, Elsevier, Amsterdam. Finkbeiner T, Zoback M, Flemings P and Stump B (2001) Stress, pore pressure, and dynamically constrained hydrocarbon columns in the South Eugene Island 330 field, northern Gulf of Mexico, AAPG Bulletin, 85(6), 1007–1031. Friedmann F, Hughes TL, Smith ME, Hild GP, Wilson A and Davies SN (1999) Development and testing of a foam-gel technology to improve conformance of the Rangely CO2 flood, SPE Reservoir Evaluation and Engineering, 2(1), 4–13. Green D and Willhite GP (1998) Enhanced Oil Recovery, Textbook Series Vol. 6. Society of Petroleum Engineers, Richardson, TX. Harpole KJ and Hallenbeck LD (1996) East Vacuum Grayburg San Andres Unit CO2 Flood ten year performance review: evolution of a reservoir management strategy and results of WAG optimization, paper SPE 36710, Proceedings of the SPE Annual Technical Conference and Exhibition, Denver, CO 6–9 October. © Woodhead Publishing Limited, 2010


125

Hervey JR and Iakovakis AC (1991) Performance review of a miscible CO2 tertiary project: Rangely Weber Sand Unit, Colorado, SPE Reservoir Engineering, 6(2), 163–168. Hunter JK and Bryan LA (1987) LaBarge Project: availability of CO2 for tertiary projects, Journal of Petroleum Technology, 39(11), 1407–1410. IEA (2008) Geologic Storage of Carbon Dioxide: Staying Safely Underground, IEA Greenhouse Gas R and D Programme, Cheltenham, UK, available at: http://www. ieagreen.org.uk/glossies/geostoragesafe-web.pdf (accessed January 2010). Jessen K, Kovscek AR and Orr F M Jr (2005) Increasing CO2 storage in oil recovery, Energy Conversion and Management, 46, 293–311. Jessen K, Tang G-Q and Kovscek AR (2008) Laboratory and simulation investigation of enhanced coalbed methane recovery by gas injection, Transport in Porous Media, 73(2) 141–159. Kazemi A and Jamialahmadi M (2009) The effect of oil and gas molecular diffusion in production of fractured reservoir during gravity drainage mechanism by CO2 injection, paper SPE 120894, Proceedings of the SPE EUROPEC/EAGE Annual Conference and Exhibition, Amsterdam, the Netherlands, 8–11 June. Kovscek AR (2002) Screening criteria for CO2 storage in oil reservoirs, Petroleum Science and Technology, 20(7&8), 841–866. Kovscek AR and Cakici MD (2005) Geologic storage of carbon dioxide. II. Cooptimization of storage and recovery, Energy Conversion and Management, 46, 1941–1956. Kuuskraa VA (2008) Maximizing Oil Recovery Efficiency and Sequestration of CO2 With Game-Changer CO2-EOR Technology, Society of Petroleum Engineers, Los Angeles Basin Section, Technology Forum, 8 April. Kuuskraa VA and Ferguson R (2008) Storing CO2 with Enhanced Oil Recovery, Report DOE/NETL-402/1312/02-07-08, Feb. 7, available at: http://www.netl.doe.gov/energyanalyses/pubs/storing%20co2%20w%20eor_final.pdf (accessed February 2010). Lohrenz J, Bray BG and Clark CR (1964) Calculating viscosities of reservoir fluids from their compositions, paper SPE 915, Journal of Petroleum Technology, 1171–1176. Masoner LO and Wackowski RK (1995) Rangely Weber Sand Unit CO2 project update, SPE Reservoir Engineering, 10(3), 203–207. Moritis G (2006) CO2 injection gains momentum, Oil & Gas Journal, 104, 37–41, 45–57. Nordbotten JM, Celia MA, Bachu S and Dahle HK (2005) Semianalytical solution for CO2 leakage through an abandoned well, Environmental Science and Technology, 39(2) 602–611. Oldenburg CM, Pruess K and Benson SM (2001) Process modeling of CO2 injection into natural gas reservoirs for carbon sequestration and enhanced gas recovery, Energy&Fuels, 15, 293–298. Oldenburg CM, Stevens SH and Benson SM (2004) Economic feasibility of carbon sequestration with enhanced gas recovery (CSEGR), Energy, 29(9–10), 1413–1422. Pacala S and Socolow R (2004) Stabilization wedges: solving the climate problem for the next 50 years with current technologies, Science, 305, 968–972. Parson EA and Keith DW (1998) Fossil fuels without CO2 emissions, Science, 282, 1053–1054. Perri PR, Emanuele MA, Fong WS and Morea MF (2000) Lost Hills CO2 pilot: evaluation, design, injectivity test results, and implementation, paper SPE 62526, Proceedings of the SPE/AAPG Western Regional Meeting, Long Beach CA, 19–22 June. PTRC (2009) Oilfield Statistics: Weyburn-Midale CO2 Project, Petroleum Technology Research Centre, Regina, SK, Canada, available at: http://www.ptrc.ca/weyburn_ statistics.php (accessed January 2010). © Woodhead Publishing Limited, 2010

126


Reid TB and Robinson HJ (1981) Lick Creek Meakin Sand Unit Immiscible CO2/ Waterflood Project, Journal of Petroleum Technology, 33, 1723–1729. Sahin S, Kalfa U and Celebioglu D (2008) Bati Raman Field immiscible CO2 application–status quo and future plans, SPE Reservoir Evaluation and Engineering, 11(4), 778–791. Saidi AM (1983) Simulation of naturally fractured reservoirs, paper SPE 12270, SPE Reservoir Simulation Symposium, San Francisco, CA 15–18 November. Stalkup Jr FI (1984) Miscible Displacement, Monograph Series, 2nd ed, Society of Petroleum Engineers, Richardson, TX. Taber JJ, Martin FD and Seright RS (1997a) EOR screening criteria revisited–part 1: introduction to screening criteria and enhanced recovery field projects, SPE Reservoir Engineering, 12(3), 189–198. Taber JJ, Martin FD and Seright RS (1997b) EOR screening criteria revisited–Part 2: applications and impact of oil prices, SPE Reservoir Engineering, 12(3), 199–205. van der Burgt MJ, Cantle J and Boutkan VK (1992) Carbon dioxide disposal from coalbased IGCCS in depleted gas fields, Energy Conversion and Management, 33(5–8), 603–610. Vega B, Tang GQ and Kovscek AR (2008) Experimental Investigation of Oil Recovery from Siliceous Shale by CO2 Injection, paper SPE 115679. Proceedings of the SPE Annual Technical Conference and Exhibition, Denver, CO, 21–24 September.


5

Carbon dioxide (CO2) sequestration in unmineable coal seams and use for enhanced coalbed methane recovery (ECBM)

M. M a z z o t t i and R o n n y P i n i, ETH Zurich, Switzerland, G. S t o r t i, Politecnico di Milano, Italy, and L. B u rl i n i, ETH Zurich, Switzerland Abstract: This chapter discusses the concept of enhanced coalbed methane (ECBM) recovery, a technology where carbon dioxide (CO2) storage in unmineable coal seams is accomplished with concomitant enhancement of coalbed methane production. The chapter first reviews the state-of-theart on the fundamental research aimed at understanding the mechanisms acting during the process. It then describes the practical aspects of the technology and summarizes the results of ECBM field tests. These prove the feasibility of ECBM recovery and highlight substantial opportunities for interdisciplinary research at the interface between earth sciences and chemical engineering. Key words: CO2 storage, coal, ECBM, field tests.

5.1

Introduction

Enhanced coalbed methane (ECBM) recovery is a technique under investigation as a possible approach to the geological storage of carbon dioxide (CO2) in a CO2 capture and storage (CCS) system (White et al., 2005; Mazzotti et al., 2009). This technology allows the recovery of coalbed methane to be enhanced by injecting CO2 in the coal seam at supercritical conditions. Through an in situ sorption/desorption process, the displaced methane is produced and the adsorbed CO2 is permanently stored. In this chapter, the state-of-the-art on the fundamental research aimed at understanding the mechanisms acting during the process is reviewed. The chapter first introduces the concept of CO2 storage in coal seams (Section 5.2) and then presents the main mechanisms acting during the ECBM recovery process (Section 5.3). The issues to be addressed, both at laboratory and field test scales, to assess the potential of a coal seam for ECBM are treated in more detail in the next sections. Pure and competitive sorption data on coal of the gases involved in the process are needed, the former being essential for estimating coal storage capacity and the latter a prerequisite for describing the displacement dynamics (Section 5.4). Studies on the phenomenon of coal swelling and its consequences for 127 © Woodhead Publishing Limited, 2010

128


coal permeability are required, since the latter controls the gas flow through the coal seam affecting, therefore, the overall ECBM operation (Section 5.5). Moreover, ECBM simulation studies aimed at investigating the different injection policies represent an essential tool to be used when assessing the potential of future ECBM operations (Section 5.6). Finally, the results of the ECBM field tests are presented; these are needed to prove the long-term capacity of coal seams to store CO2 and therefore to demonstrate that ECBM technology can be deployed at commercial scale (Section 5.7).

5.2

Storage in unmineable coal seams

Coal seams have been proposed as a possible location for permanent geological storage of CO2 (IPCC, 2005). In particular, the coal seams which can be used for storage purposes are those presenting characteristics precluding economically profitable mining. These so-called unmineable coal seams are either too thin, too deep or too high in sulfur content and mineral matter (White et al., 2005). The estimated storage potential of coal seams is relatively small compared to other geological formations, varying between 3 GtCO2 and 200 GtCO2 (IPCC, 2005). The upper estimate refers to the worldwide distribution of bituminous coal seams whereas the lower estimate refers only to those coal seams where simultaneous CBM production could be carried out. These values, together with the fact that the distribution of potential coal seams does not always match the location of large CO2 sources, suggest that the contribution of coal seams to the underground storage of CO2 will be limited compared to other geological formations. However, compared to the current anthropogenic CO2 emissions of almost 30 GtCO2 per year (IPCC, 2007), these amounts are still significant and need to be taken into account in the effort to find ways to reduce greenhouse gas emissions. Coal seams are fractured porous media, characterized by a large internal surface area. Significant amounts of methane (CH4) are generated and retained in the coal seam during the geological process, leading to the formation of coal seams, the so-called coalification process (Levine, 1993; Gentzis, 2000). The way this coalbed methane is stored in the coal reservoir differs from other geological locations in the fact that, besides filling the available fracture and pore volume, the gas adsorbs on the coal surface and absorbs into the coal structure. The adsorbed phase has a much higher density than gas (Sircar, 2001), allowing for a better exploitation of the reservoir rock as a storage medium for CO2. As explained in Section 5.4, this sorption process is controlled by a thermodynamic equilibrium between the amount of gas adsorbed and absorbed and its corresponding density (or pressure) in the fluid phase. This relation is described by a sorption isotherm and implies that the gas is trapped in this state, as long as the pressure (or density) of the fluid phase in the seam is maintained. As in the case of conventional gas


Carbon dioxide (CO2) sequestration in unmineable coal seams

129

reservoirs, therefore, there needs to be an impermeable rock layer surrounding and sealing the coal reservoir if the gas is to be permanently retained. Such coalbed methane can be recovered from the coal seam and used for energy production. Conventional primary recovery of methane, which is performed by pumping out water and depressurizing the reservoir, allows recovery of 20–60 % of the methane originally present in the reservoir (White et al., 2005). As in the case of enhanced oil recovery (EOR), such primary production could, in principle, be enhanced by injecting CO2 into the coal seam. This process is schematized in Fig. 5.1 and is called ECBM recovery (White et al., 2005). Due to the higher affinity of CO2 to coal with respect to methane, the injected CO2 displaces the methane. Ultimately, most of the methane is recovered and the coal seam contains mainly CO2, which remains there permanently separated from the atmosphere. ECBM is, therefore, attractive from two perspectives. On the one hand, if one is interested in the recovered methane as a fuel or a technical gas, ECBM allows also for a net CO2 sequestration, thanks to the above-mentioned high CO2 sorptivity. On the other hand, if the goal is that of storing captured CO2 the ECBM operation also allows the recovery of methane, thus making CO2 storage economically interesting in this case. Exactly because of this added value, a technique such as ECBM recovery that offers as a byproduct methane is expected to be among the first commercially practised technologies for CO2 storage (White et al., 2005). The expertise gained in the past years for enhanced oil production will play an important role in the implementation of the ECBM technology at a commercial scale. Once injected underground, CO2 is trapped as a dense gas in the coal fractures, the so-called cleats, adsorbed on and absorbed in the coal (see Section 5.4), and solubilized in the formation water. As in the case of other geological formations, optimal storage conditions are attained at a depth in the subsurface where the pressure and temperature conditions allow the CO2 to be stored as a supercritical fluid and therefore at a high density.

5.3

Enhanced coalbed methane recovery

Much laboratory research is taking place to improve understanding of the processes of sorption and displacement. These fundamental studies are supported by the field tests, either completed or planned, whose number has been increasing steadily in the last years (Gunter et al., 2005; Reeves, 2005; Wong et al., 2007; Yamaguchi et al., 2007; Litynski et al., 2009; Vincent et al., 2009; Van Wageningen et al., 2009). By testing different gas injection policies, such as CO2, nitrogen N2, or mixtures of thereof, these field tests have shown the potential of coal seams as storage sites for CO2. However, they have also revealed many factors affecting the success of the ECBM operation which need to be extensively investigated. As an


130

CH4 recovery

© Woodhead Publishing Limited, 2010 5.1 Schematic of an ECBM operation, where CO2 captured from a power plant is injected into the coal seam and CH4 is produced. From an engineering point of view, ECBM recovery is a sorption/desorption process at supercritical conditions in a natural underground coal formation, which is accomplished by injecting CO2 in one or more injection wells and by collecting CH4 from one or more production wells.


CO2 injection


131

example, a common problem encountered during field tests is the reduction of coal permeability upon coal swelling, which forced operators to reduce injection rates in order to avoid a pressure increase above the safety levels at the injection well (Van Bergen et al., 2007; Wong et al., 2007; Yamaguchi et al., 2007). The issues to be addressed both at laboratory and field test scales to assess the potential of a coal seam for ECBM are summarized in the following and will be treated in detail in the next sections: ∑

Pure and competitive sorption data on coal of the gases involved in the process are needed, the former being essential for the coal storage capacity estimates and the latter a prerequisite for the description of the displacement dynamics. ∑ Studies on the coal swelling phenomenon and its consequences on the coal permeability are required, since the latter controls the gas flow through the coal seam, affecting therefore the overall ECBM operation. ∑ ECBM reservoir simulation studies represent an essential tool to be used in the description of the performed field test as well as in the planning of future demonstration projects.

A thorough understanding of the issues listed above will enable a critical assessment of the feasibility of CO2 storage in coal seams coupled with ECBM recovery on a commercial basis.

5.4

Competitive adsorption

The ECBM operation is controlled by a sorption/desorption process and therefore competitive sorption equilibria involving CO2 and CH4, the two main components of any ECBM operation, are required. These measurements should also involve N2, which has been successfully used as a co-injectant with CO2, due to the permeability problems caused by the swelling of the coal (Gunter et al., 2005; Reeves, 2005; Yamaguchi et al., 2007). Moreover, estimates of the coalbed capacity for CO2 storage and of the maximum theoretical amount of coalbed methane (the so-called maximum gas in place, GIPmax) are provided by pure CO2 and CH4 sorption isotherms, respectively. The gravimetric (Bae and Bhatia, 2006), manometric (Ozdemir et al., 2004) and volumetric (Gasem et al., 2002) techniques are the commonly used methods to obtain single-component (pure) sorption isotherms. Prior to the experiments the coal sample is usually ground and sieved to obtain particles of the desired size. On the one hand, the use of a powdered coal sample allows the time for reaching sorption equilibrium to be reduced; on the other hand, during the grinding procedure, care should be taken in order to guarantee that the obtained sample is representative of the different macerals present in the coal (Mukhopadhyay and Hatcher, 1993). In the


132


case of multicomponent sorption measurements, two or more gases compete simultaneously for sorption, leading to a variation in the fluid phase composition (Keller and Staudt, 2005). This requires an additional measurement, such as gas chromatography, to be added to the conventional method used in the pure gas adsorption studies (Yang, 1997; Keller and Staudt, 2005). To be useful for application in the field, the experiments in the laboratory need to reproduce as closely as possible the underground conditions in the coal reservoir. In particular, since the coal seams to be exploited for storage purposes are very deep (more than 750 m), these measurements have to cover a broad range of temperature and pressure conditions, from subcritical to supercritical, and should involve both dry and wet samples, since coal seams are saturated with water. A summary of studies reporting sorption data on dry and wet samples from different coal mines around the world is given in Tables 5.1 and 5.2, for both single and multicompontent sorption experiments. Table 5.1 Studies reporting single-component high-pressure gas sorption measurements on coal at typical coal bed temperature Coal origin

Single component adsorption

Dry/wet Gas

Pmax Method1 Reference

Bowen Basin Dry CO2/CH4 200 G (Australia) Sydney Basin Wet CO2 60 G (Australia) Australia Dry/wet CO2/CH4/N2 200 G Qinshui Basin Wet CO2/CH4 100 V (China) Warndt Colliery Dry/wet CO2 200 V (Germany) Nottinghamshire Dry/wet CO2 200 V Coal Field (Great Britain) Sulcis Coal Dry CO2/CH4/N2 200 G Province (Italy) Ishikari Coal Dry CO2/CH4/N2 60 V Field (Japan) Silesian Basin Dry/wet CO2/CH4 200 V (Poland) Achterhoek Dry/wet CO2/CH4 200 V Area (The Netherlands) San Juan Wet CO2/CH4/N2 150 V Basin (USA) Argonne Dry/wet CO2 150 V/G Premium (USA) 1

Bae and Bhatia, 2006 Saghafi et al., 2007 Sakurovs et al., 2007, Day et al., 2008a,b Yu et al., 2008 Siemons and Busch, 2007 Siemons and Busch, 2007

Ottiger et al., 2006, Ottiger et al., 2008a,b Shimada et al., 2005 Siemons and Busch, 2007, Busch et al., 2004 Krooss et al., 2002

DeGance et al., 1993, Chaback et al., 1996, Fitzgerald et al., 2005 Goodman et al., 2004, Siemons and Busch, 2007, Goodman et al., 2007

V = volumetric; G = gravimetric. Note: maximum pressure, Pmax, is given in bar. © Woodhead Publishing Limited, 2010


133

Table 5.2 Studies reporting competitive high-pressure gas sorption measurements on coal at typical coal bed temperature Coal origin

Competitive adsorption

Dry/wet Gas

Pmax Method1 Reference

Sydney Basin Dry CO2/CH4/N2 52 V–C (Australia) Sulcis Coal Dry CO2/CH4/N2 150 G–C Province (Italy) V–C Ishikari Coal Dry CO2/CH4/N2 60 Field (Japan) Silesian Basin Dry/wet CO2/CH4/N2 230 V–C (Poland) Achterhoek Area Dry/wet CO2/CH4 230 V–C (The Netherlands) San Juan Wet CO2/CH4/N2 150 V–C Basin (USA) Argonne Dry/wet CO2/CH4 180 V–C Premium (USA) Black Warrior Wet CO2/CH4/N2 100 V–C Basin (USA)

Stevenson et al., 1991 Ottiger et al., 2006 Ottiger et al., 2008a,b Shimada et al., 2005 Ceglarska-Stefanska and Zarebska, 2005, Busch et al., 2006 Busch et al., 2006 Arri et al., 1992, DeGance et al., 1993, Fitzgerald et al., 2006 Busch et al., 2003, 2006, 2007 Chaback et al., 1996

1

V = volumetric; G = gravimetric; C = chromatographic. Note: Note: maximum pressure, Pmax, is given in bar.

Comparative studies among laboratories are very useful since they help in defining a standard procedure for measuring sorption isotherm accurately at conditions relevant for ECBM applications (Pini et al., 2010). As an example, the US Department of Energy recently initiated a series of studies on Argonne Premium coal samples, where the most commonly used techniques to measure adsorption isotherms were compared, namely the manometric, volumetric and gravimetric methods (Goodman et al., 2004, 2007). For both dry and moisture-equilibrated samples, the reported CO2 sorption data diverged significantly among the laboratories, suggesting that further comparison studies are needed. Beside experimental artifacts, the main reason for this divergence has been found to lie in the preparation of the coal sample and, in particular, on its moisture content. In the case of coal, the uptake of CO2, CH4 and N2 is a combination of adsorption on its surface and penetration (sorption) into its solid matrix, the latter resulting in coal swelling (Reucroft and Patel, 1986). These two processes act simultaneously, making the coal a challenging material to study, in particular with respect to understanding the fundamental, thermodynamic aspects of adsorption. In the case of non-swelling commercial adsorbents such as zeolites or silica gels, the aforementioned techniques allow the so-called excess adsorbed amount, i.e. the amount adsorbed minus the


134


theoretical quantity held up in the adsorbed phase volume, to be obtained (Sircar, 2001):

mex = ma – rbVa

[5.1]

where mex and ma are the excess and total amount adsorbed on the coal, respectively, rb is the density of the bulk phase and Va the volume of the adsorbed phase. In the case of coal, therefore, the only truly measurable quantity accounts for the effect of both adsorption and absorption, phenomena whose contributions cannot be separated (Milewska-Duda et al., 2000; Ozdemir et al., 2003; Bae and Bhatia, 2006; Romanov et al., 2006; Day et al., 2008b, Ottiger et al., 2008a):

meas = ma + ms – rb(Va + DV s)

[5.2]

where meas is the excess sorption, ms is the amount absorbed in the coal and DVs is the difference between the volume of the mixture of coal and imbibed gas and the initial sample volume. It is worth pointing out that this distinction between the excess adsorbed amount mex and the value which can be truly experimentally determined, the excess sorption meas, is not always acknowledged in the literature. As an example of general validity, Fig. 5.2 shows the single-component sorption isotherms of CO2, CH4 and N2 on a Swiss coal drilled at a depth of 1586 m and measured using a Rubotherm magnetic suspension balance (Ottiger et al., 2006) at two different temperatures, namely 45 °C and 70 °C. The measured data exhibit the usual behavior of excess adsorption isotherms (Fitzgerald et al., 2005; Bae and Bhatia, 2006; Ottiger et al., 2006; Sakurovs et al., 2007). In the case of CO2, the excess sorption increases with the bulk density to reach a maximum, and decreases linearly by further increasing the density. In the case of CH4, the isotherms show only a slight maximum, whereas for N2 it is not visible, the measuring conditions being far above the critical density. Over the whole range of density (that for all the gases corresponds to pressures up to 200 bar), the excess sorption grows with increasing temperature. Moreover, the experimental data show that CO2 adsorbs more than CH4, and CH4 more than N2, in accordance with several other studies (Fitzgerald et al., 2005; Shimada et al., 2005; Bae and Bhatia, 2006; Sakurovs et al., 2007). This property is a prerequisite for a successful ECBM operation. It is worth noting that some early studies hinted at an irregular behavior in the sorption of CO2, particularly close to its critical density (Krooss et al., 2002; Toribio et al., 2005). It is believed that these were artifacts due to the experimental set-ups and procedures, as demonstrated by the fact that when repeating one of the measurements in another setting the irregularities disappeared (Toribio et al., 2005). In this context, the impact of several sources of error on the measured high-pressure sorption isotherms have recently been discussed in detail (Sakurovs et al., 2008).



135

1.2

1 CO2

neas (mmol/g)

0.8 CH4

0.6

0.4

N2

45 °C

0.2

70 °C 0

0

5

10 Density (mol/L)

15

20

5.2 Excess sorption isotherms of CO2, CH4 and N2 on a dry Swiss coal at 45 °C (closed symbols) and 70 °C (open symbols) as a function of the molar bulk density r measured using a Rubotherm magnetic suspension balance as described elsewhere (Ottiger et al., 2006). Solid lines are calculated with a Langmuir equation corrected for the volume occupied by the adsorbed and sorbed phase (Pini et al., 2010).

The theoretical effectiveness of the displacement of CH4 by the injected CO2 during the ECBM operation predicted by the pure sorption isotherms needs to be confirmed by multicomponent sorption measurements, where the different gases compete simultaneously for sorption on coal. In agreement with other studies (Stevenson et al., 1991; Chaback et al., 1996; Shimada et al., 2005; Fitzgerald et al., 2006; Ottiger et al., 2008a), competitive sorption isotherms measured at 45 °C for a ternary gas mixture with feed composition 33.3 % CO2, 33 % CH4, 33.4 % N2 on an Italian coal from the Sulcis Coal Province and shown in Fig. 5.3 confirm this expectation: CO2 is always the most adsorbed component and CH4 is preferentially adsorbed compared to N2 (Ottiger et al., 2008b). There are only two cases in which preferential sorption of CH4 over CO2 on low rank coals has been observed (Busch et al., 2003, 2006). Coal shows a high variability in its chemical and physical properties, being a mixture of many kinds of organic and inorganic materials (Van Krevelen, 1981; Mukhopadhyay and Hatcher, 1993). This heterogeneity is reflected


136


2 T = 45 °C

Total

neas (mmol/g) i

1.5

CO2 1

0.5

CH4 N2

0

0

2

4 6 Density (mol/L)

8

10

5.3 Competitive sorption of a ternary gas mixture with feed composition 33.3 % CO2, 33 % CH4, 33.4 % N2 on an Italian coal at 45 °C. Total and molar excess sorption neas of component i as a function of the bulk density measured using a Rubotherm magnetic suspension balance combined with gas chromatography as described elsewhere (Ottiger et al., 2008a). Symbols are experimental points, whereas solid lines are calculated with a lattice DFT model (Ottiger et al., 2008b).

in differing affinities to adsorb gases: coal samples obtained from different mines worldwide may show different sorption capacities, which affect their suitability for ECBM purposes. As a consequence, correlations between such properties and the amount of gas adsorbed are desirable, since they can be used as a guide to choose the most suitable coal seams for ECBM. The number of studies conducted in this is limited (Siemons and Busch, 2007; Day et al., 2008a; Pini et al., 2010) and often experiments have been carried out at low pressures (Mastalerz et al., 2004; Ozdemir et al., 2004; Saghafi et al., 2007). As anticipated above, the storage capacity of coal seams is mainly given by their ability to adsorb gas. An estimate of the amount of CO2 stored in the coal can, therefore, be obtained by converting the measured excess sorption isotherm into an absolute isotherm, which gives the actual uptake of CO2 by the coal. This can be done by adding to the excess sorption isotherm a term proportional to the bulk density and to the slope of the linear part of the excess isotherm, which corresponds to the sum of the volume of the adsorbed



137

phase and of the swollen part of the coal volume, as given by Equation 5.2 and discussed in detail elsewhere (Ottiger et al., 2006). Alternatively, one can correct the sorption data by using the swelling measurements and by assuming a value for the adsorbed phase density (Murata et al., 2001; Bae and Bhatia, 2006; Day et al., 2008b). The outcome of both procedures is shown in Fig. 5.4 for the case of CO2 sorption measurements at 45 °C on a dry coal from Italy, for which the volumetric behavior at different pressures was measured separately (Ottiger et al., 2008a) (Section 5.5). These results indicate a maximum capacity of dry coal of 0.11 g/g and that swelling has a considerable effect on the sorption isotherm, in particular at large densities. Coal seams are naturally saturated with water and therefore sorption measurements should be carried out on wet coal samples, in order to give reliable estimates for the sorption capacity. Qualitatively all sorption 0.14

0.12 Absolute 0.1

Sorption (g/g)

Excess

0.08

Exce

0.06

– with s wellin

g

ss – w/o swe lling

0.04

0.02

0

CO2

0

0.2

0.4 Density (g/cm3)

0.6

0.8

5.4 CO2 sorption isotherms on a dry Italian coal at 45 °C as a function of the bulk density (Ottiger et al., 2008a). Symbols: () excess sorption without correcting for swelling (truly measurable quantity); () excess sorption corrected for the changes in sample volume caused by absorption using swelling data obtained on the same coal sample (see Fig. 5.5); () total (absolute) CO2 loading capacity obtained by correcting the excess sorption data with the slope of the linear descending part of the sorption isotherm as explained in detail elsewhere (Ottiger et al., 2006). Lines are drawn to guide the eye.


138


measurements show that the uptake of CO2 by wet coal is always less than that by dry coal, because of competitive water sorption. However, the quantitative effect of moisture on CO2 uptake is less certain, because techniques for the measurement of sorption isotherms on wet samples are unfortunately not as well established as those on dry samples. It has been shown, for instance, that a coal containing about 3 % moisture, corresponding to a relative humidity of 50 %, adsorbs 30 % less CO2 than the corresponding dry one (Day et al., 2008c). Typical experiments are conducted on as-received coal samples (Fitzgerald et al., 2005; Siemons and Busch, 2007) or on wet coal samples, that have been prepared by equilibrating them in a sealed chamber with a saturated salt-solution of known water partial pressure (Krooss et al., 2002; Day et al., 2008c). After charging the gas into the measuring chamber, its adsorbed amount is estimated by assuming that the amount of water on coal remains constant as the gas adsorbs, i.e. that the experiment is carried out in such a way as to prevent the gas from drying the sample. Neglecting variation of the moisture content during the experiment may lead to unexpected results precluding a physically sound interpretation of the obtained sorption isotherms (Krooss et al., 2002; Fitzgerald et al., 2003; Siemons and Busch, 2007). The unsatisfactory reproducibility of an inter-laboratory study of CO2 sorption data on wet coals, particularly evident above 80 bar, was attributed to similar causes (Goodman et al., 2004, 2007). Models describing the experimentally obtained sorption isotherms are needed to be used in reservoir simulators which predict the dynamics of ECBM processes. On the one hand, a relatively simple form of these equations is desired, since it allows for a direct implementation in the simulator code. On the other hand, more fundamental approaches are required to understand the complex mechanisms of gas sorption in coal. A variety of semi-empirical isotherm models have been used to describe the experimental sorption data. For single gas sorption these include Langmuir, Toth, Dubinin–Radushkevich (D–R) and Dubinin–Astakhov (D–A) (Clarkson and Bustin, 2000; Bae and Bhatia, 2006; Sakurovs et al., 2007). Since these models describe absolute sorption instead of excess sorption, i.e. the quantity obtained from conventional experimental techniques, a constant value of either the density or the volume of the adsorbed phase is usually assumed (Murata et al., 2001). For the description of multicomponent gas sorption, the ideal adsorbed solution (IAS) theory in conjunction with the aforementioned pure gas isotherm models has been applied (Stevenson et al., 1991; DeGance et al., 1993; Clarkson and Bustin, 2000; Yu et al., 2008). Although the extended Langmuir equation has often been used due to its simplicity (Arri et al., 1992; Chaback et al., 1996), it has been found to be inadequate to describe the behavior of gas mixtures (Fitzgerald et al., 2005). An alternative approach uses an equation of state (EOS), namely a two-dimensional EOS, such as the Eyring and



139

Virial equation of state (DeGance et al., 1993) or the Zhou–Gasem–Robinson two-dimensional equation of state (Fitzgerald et al., 2005). Other methods are based on statistical thermodynamic theory, which uses fluid–fluid and fluid–solid molecular interaction energies and a microscopic description of the pore geometry (Sudibandriyo et al., 2003; Fitzgerald et al., 2006; Ottiger et al., 2008b). When quantitative information about the pore size distribution of the coal is incorporated, insights into the behavior of the adsorbed gas in pores of different sizes, particularly under near- and supercritical conditions, can be obtained (Hocker et al., 2003). The results in the literature show that experimental data can be described well using one or other of these models; however, they may have limited predictive capability when extrapolated. All the models described above treat the adsorbent as a rigid porous solid that does not change its volume as a result of interaction with the gases. As was anticipated previously, gases can dissolve into the coal structure causing it to swell. On the one hand, some modified versions of the original isotherms models, such as modified D–A and D–R, have been proposed to take into account the adsorbent volume change caused by sorption (Ozdemir et al., 2004; Sakurovs et al., 2007). On the other hand, more fundamental approaches, the so-called dual-sorption models, treat the coal as a microporous copolymer containing both elastic polymer-like chains and crystal-like domains and describe the sorption phenomenon as a combination of both penetration of gas molecules into the elastic matrix (absorption) and adsorption of gas into the micropores (Milewska-Duda et al., 2000).

5.5

Swelling and permeability

Swelling and shrinking of coal upon uptake and release of many gases and liquids have been reported (Larsen, 2004). The phenomenon of coal swelling was introduced after the first attempts aimed at determining the coal surface area. CO2 is particularly suitable as a measuring gas since, through dissolution into the coal matrix, it is able to reach both open and closed pores of coal, allowing meaningful estimates of the surface area to be obtained (Mahajan, 1991). Those experiments are conducted at low pressures, and it is only recently that interest has moved towards higher pressures, i.e. at pressures relevant for CO2 storage. Dilatometric, optical or strain measurement methods are used to measure the volume changes of coal samples of given shape upon exposure to a gas at high pressure for several days (Reucroft and Sethuraman, 1987; Harpalani and Chen, 1995; Ottiger et al., 2008a). A summary of the studies reported in the literature together with the corresponding applied experimental conditions is given in Table 5.3. As an illustration and without loss of generality, Fig. 5.5 shows swelling data for two different coal samples from Italy, from the Monte Sinni Coal Mine (Sulcis Coal Province, Sardinia) (Fig. 5.5a) and the Ribolla Coal Mine


140


Table 5.3 Studies reporting swelling measurements on coal Coal origin

Swelling

Method

Gas

Pmax Reference

Australia Optical CO2 150 Canada Strain CO2/CH4/N2 50 France Strain CO2/CH4/He 55 Germany Strain CO2/CH4/He 55 Sulcis Coal Optical CO2/CH4/N2/He 140 Province (Italy) South Island Strain CO2/CH4/N2/He 40 (New Zealand) Poland Dilatometric CO2 40 United Kingdom Strain CO2/CH4/He 55 San Juan Strain CH4/He 104 Basin (USA) Kentucky (USA) Dilatometric CO2 15 USA Dilatometric CO2/N2/He 48

Day et al., 2008b Cui et al., 2007 Durucan et al., 2008 Durucan et al., 2008 Ottiger et al., 2008a St. George and Barakat, 2001 Ceglarska-Stefanska and Czaplinski, 1993 Durucan et al., 2008 Harpalani and Chen, 1995 Reucroft and Sethuraman, 1987 Walker et al., 1988

Note: Maximum pressure, Pmax, is given in bar.

(Grosseto) (Fig. 5.5b) measured in our laboratory using a high-pressure view cell. In both cases, and in common with the other studies reported in Table 5.3, the extent of swelling increases monotonically with pressure up to a few percentage points for adsorbing gases, with CO2 swelling coal more than CH4 and CH4 more than N2. Moreover, it has been observed that for helium, a non-adsorbing gas, volume changes are negligible (St George and Barakat, 2001; Cui et al., 2007; Day et al., 2008b; Ottiger et al., 2008a; Durucan et al., 2008). Therefore, considering an ECBM operation, the displacement of CH4 by CO2 would lead to a net coal swelling, whereas its displacement by N2 would lead to a net shrinking. It is worth pointing out that high swelling coals are not necessarily those coals with high sorption capacity, as confirmed by a recent study (Day et al., 2008b). This observation highlights once more that the mechanisms of sorption and swelling are complex and that many other factors, such as strain response to stress, may come into play. As shown in Fig. 5.5, Langmuir-like equations are suitable for describing the experimentally obtained swelling isotherms (Levine, 1996; Cui et al., 2007). Coal swelling is reversible, with the sample returning to its original size when the pressure is released (Cui et al., 2007; Day et al., 2008b). Similar to many polymer systems, repeated exposure to several organic solvents leads to an isotropic behavior of coal already after the first run (Larsen et al., 1997; Larsen, 2004). When exposed to CO2, however, coal has shown a contradictory behavior which can be traced back to its anisotropic nature.



141

0.05 CO2 0.04

0.03

s (–)

CH4 0.02

0.01

N2

0

–0.01

0

50

Pressure (bar) (a)

100

150

0.03

CO2

s (–)

0.02

CH4

0.01

N2

0

–0.01

0

50

Pressure (bar) (b)

100

150

5.5 Swelling of an unconstrained dry disc of two different Italian coals, Sulcis (a) and Ribolla (b), as a function of the pressure, P, of CO2, CH4, N2 at 45 °C. Measurements have been carried out in a high-pressure view cell equipped with sapphire window by monitoring the change in geometrical dimensions of the disc upon pressurization with pure gases (Ottiger et al., 2008a). Swelling is assumed to be isotropic. The experimental data (symbols) are well described as a function of pressure by Langmuir-like equations (solid lines), i.e. s = smaxbP/(1 + bP), where smax and b are temperature-dependent, gas- and coal-specific constants.


142


In a recent study, repeated CO2 swelling measurements on coal showed that changes are greater in the direction perpendicular to the bedding plane than in that parallel to it (Day et al., 2008b). In another study, the observed differences between the two directions were very limited (Levine, 1996). It is clear that further measurements are needed to clarify whether one or the other conclusion can be drawn. However, we believe that the data presented in Fig. 5.5 are useful and the assumption of isotropic expansion acceptable, the error associated with the experimental technique being similar to the difference arising from assuming either isotropic or anisotropic behavior in the two mentioned works. According to the dual nature of CO2 uptake, i.e. surface adsorption and absorption into the solid matrix, swelling can also be interpreted in two complementary ways. On the one hand, coal expansion may be understood as a consequence of the purely physical adsorption process: adsorption induces a change of the coal-specific surface energy, which can be compensated by the elastic energy change associated with the volume change (Pan and Connell, 2007). On the other hand, as a glassy, strained, cross-linked macromolecular system, coal undergoes structural changes in the presence of high-pressure CO2 that can be explained only by penetration of CO2 into the coal matrix (Karacan, 2003; Larsen, 2004). There are two important consequences of coal swelling which have to be considered, particularly in the context of an ECBM operation. First, in agreement with the suggestion that coal possesses a polymeric nature, CO2 uptake may lead to weakening and plasticization phenomena (Van Krevelen, 1981; Larsen, 2004). A decrease in the coal’s softening temperature upon exposure to CO2 has been reported, suggesting that coal mechanical properties, such as its Young’s elastic modulus, may change over the longterm horizon of CO2 storage (Viete and Ranjith, 2006). Second, volume changes of coal during ECBM operations are of key importance because they affect coal bed permeability which, in turn, controls injection pressure and gas production. This effect can be easily visualized with the help of the so-called matchsticks model, which describes coal as an ensemble of parallel elongated matrix elements separated by fractures, i.e. cleats, that constitute its transport porosity (Seidle et al., 1992; Gentzis, 2000). On the one hand, being deep underground, the external lithostatic pressure (confining pressure) tends to press the matrix elements together, and to reduce porosity. On the other hand, gas absorption swells the coal matrix elements, and therefore consumes the space between them, thus also reducing porosity. Permeability can be dramatically affected by these effects, since the volumetric strains obtained from the experiments (Fig. 5.5) are of the same order of magnitude as coal porosity. For coal seams, a variation in permeability with respect to a reference state 0 is related to a change in porosity e by a cubic law, as given by Equation



143

5.3 (Reiss, 1980; Seidle et al., 1992; Harpalani and Chen, 1997; Palmer and Mansoori, 1998, Cui et al., 2007). Being dependent on the specific stress situation in the coal seam, porosity changes should be calculated by solving a 3D elasticity problem. However, it is common procedure to simplify this situation by describing the coalbed as an isotropic linear poroelastic medium under uniaxial strain and constant overloading (Palmer and Mansoori, 1998; Zhu et al., 2007). Beside the initial stress state, the coal seam porosity is affected by two terms: one depends on the effective pressure (defined as the confining pressure minus the fluid pressure) and the other on the swelling/ shrinkage of the coal upon sorption. Therefore, porosity, e, and permeability, k, in the coal bed can be predicted as (Gilman and Beckie, 2000; Shi and Durucan, 2004, Cui et al., 2007): 3

k = Ê e ˆ = exp [– C (P – P ) – C s ] 1 c 2 k0 ÁË e 0 ˜¯

[5.3]

where P and Pc are fluid and confining pressure, respectively, s is the pressuredependent swelling (see for instance Fig. 5.5), C1 and C2 are coefficients depending on coal properties. In the above equation, the reference values of porosity and permeability apply to an unstressed coal in contact with a non-swelling gas at atmospheric pressure (Pini et al., 2009). It is apparent that the physical description of porosity changes based on the matchsticks model is fully reflected in the functional form of Equation 5.3. Figure 5.6 shows the porosity and permeability behavior as a function of fluid pressure for an Italian coal, for which both the parameters needed in Equation 5.3 and the swelling behavior are known (see Fig. 5.5a and Pini et al., 2009). In agreement with other studies, where a similar relationship has been used (Durucan and Shi, 2009), for non- or weakly swelling gases such as helium or N2, the permeability increases with gas pressure. However, when pure CO2 is injected, this effect can be reversed, leading to the appearance of a minimum at the so-called rebound pressure (Palmer and Mansoori, 1998; Shi and Durucan, 2004). This is due to the fact the CO2 has a much stronger ability to swell the coal compared to N2, as shown in Fig. 5.5. Permeability reduction upon injection of CO 2 is currently the issue of most concern in relation to the feasibility of ECBM operations. As a consequence, experiments aimed at the study of gas flow through the coal have recently attracted increasing attention, in order to provide estimates of the parameters in Equation 5.3. The permeability of a geological structure can be determined in situ through a standard protocol, but experiments with coal are complicated by its interactions with sorbing and swelling gases like CO2, and hence by the possible dynamic variation of permeability in the course of the experiment. Alternatively, 1D dynamic flow experiments can


144


100

He 10–1

k/k0

e/e0

N2

10–2 CO2

10–1

0

20

40 60 Pressure (bar)

80

10–3 100

5.6 Estimated changes in coal porosity (left-hand y-axis) and permeability (right-hand y-axis) as a function of the gas pressure for helium, N2 and CO2 using the permeability relationship given by Equation 5.3 (Pini et al., 2009). e0 = 0.032, k0 = 0.049 mD; C1 = 0.188 MPa–1; C2,CO2 = 28.0; C2,N2 = 107.0.

be carried out in the laboratory using a coal core placed in a hydrostatic or a triaxial cell. In a hydrostatic cell, the coal core is surrounded by a rubber sleeve, through which it is exposed to an external confining pressure, Pc, and it is connected through on–off valves to an upstream and a downstream reservoir containing the gas at the desired pressure levels (Harpalani and Schraufnagel, 1990; Harpalani and Chen, 1997; Mazumder et al., 2006; Mazumder and Wolf, 2008; Durucan and Shi, 2009; Pini et al., 2009). In a triaxial cell, additionally vertical and horizontal stresses on the coal core are imposed independently, so as to better reproduce the in situ conditions (Massarotto et al., 2007; Wang et al., 2007). After saturating the coal sample with a gas at the given pressure and after subjecting it to the desired stresses, typical experiments in both cells are carried out either by imposing a constant gas pressure difference across the sample and by measuring the corresponding gas flow rates (constant pressure difference method), or by imposing an initial pressure difference between the two reservoirs and by monitoring pressure during equilibration (transient step method) (Fisher, 1992). To estimate the parameters in the permeability equation (Equation 5.3, the experimental behavior in terms of pressure or flow rates is compared



145

to the results obtained using a mathematical model of the process (Shi and Durucan, 2003; Wang et al., 2007; Wei et al., 2007a; Pini et al., 2009). In its simplest form, under steady-state conditions, such a model would consist only of Darcy’s law, which describes the flow of a fluid through a porous medium (Hildenbrand et al., 2004). However, in the development of a model describing the unsteady flow of gases through coal, both of the aforementioned phenomena (sorption and swelling) have to be taken into account. In its multicomponent formulation, and with reference to Fig. 5.7, the model consists of the following material balances in the fluid and in the adsorbed phase for each component i, where the rate of diffusion through the coal matrix is simulated by a lumped linear mass transfer driving force (see Section 5.6 for more details):

Confinement

Pressure (bar)

150

100 k/k0 = 0.14 k/k0 = 0.087 PUS

50

PDS

0

0

50

k/k0 = 0.055 k/k0 = 0.035

100

150 Time (min)

200

250

300

(a) 5.7 Results of transient step experiments carried out on Sulcis coal core (diameter 2.54 cm, length 4.50 cm) at 45 °C with (a) helium and (b) CO2 under an external confining pressure of 150 bar measured in a hydrostatic flow cell (Pini et al., 2009). Symbols: experimental () downstream (DS) and () upstream (US) reservoir pressures. Solid lines correspond to simulation results. The estimated values of the relative permeabilities, k/k0, at the end of each transient step are also given, with k0 = 0.162 mD.


146


Confinement 150

Pressure (bar)

100 k/k0 = 0.0033

k/k0 = 0.0020

50 PUS PDS 0

k/k0 = 0.0017 k/k0 = 0.0018

0

1000

2000 Time (min)

3000

4000

(b) 5.7 Continued

∂(e ci ) ∂[(1 – e )qi ] ∂(uci ) + + =0 ∂t ∂t ∂z

[5.4]

∂[(1 – e )qi ] = (1 – e ) kmi (q*i – qi ) ∂t

[5.5]

where ci and qi are, respectively, the gas and adsorbed phase concentration of component i, qi* is its equilibrium concentration in the adsorbed phase, kmi is its mass transfer coefficient; u is the superficial velocity, et the total porosity and t and z are time and space coordinates. The superficial velocity u is given by Darcy’s law, expressing velocity as a function of pressure gradient and permeability:

È Ê ∂c ˆ ˘ u = ne = – k Í∂P – gM m Á z + c˜ ˙ m Î ∂z ¯˚ Ë ∂z

[5.6]

where n is the interstitial velocity and e the cleat porosity, P is the total pore pressure, k is the permeability, m is the dynamic viscosity, g is the gravitational acceleration, Mm the molecular weight of the gas mixture and



147

c its total concentration. The sorption on coal is described by an extended Langmuir equation, giving the amount adsorbed for component I, q*i , as a function of its concentration: qi* =

bi qmi ci j =1

[5.7]

NC

1 + ∑ bjc j

Similar to the pure swelling isotherm equation given in the caption of Fig. 5.5, coal swelling upon exposure to gas mixture can be expressed with an extended Langmuir-like equation: si =

bsi qmi pi

[5.8]

NC

1 + ∑ bsj p j j =1

In Equations 5.6 and 5.7, the saturation capacity and the Langmuir equilibrium constant of component i are given by qmi and bi for the sorption isotherm and by smi and bsi for the swelling isotherm, respectively. The model is completed by the following constitutive equations: an equation of state, e.g. Peng–Robinson EOS, to relate gas density to pressure and temperature; Equation 5.3 expressing permeability and porosity as a function of pressure and swelling. Upstream and downstream boundary conditions are either constant pressure values in the case of the constant pressure difference method, or simple reservoir material balances in the case of the transient step method. Figure 5.7 reports two examples of transient steps carried out by injecting helium and CO2 into a coal core confined in a hydrostatic flow cell (Pini et al., 2009). The two effects presented in Equation 5.3 can now be easily identified: with increasing gas pressure, i.e. decreasing the effective stress, the equilibration time reduces and permeability increases. However, when comparing the same pressure step, the time to reach pressure equilibration increases, going from helium to CO2, i.e. going from the non-swelling gas to the one that swells coal the most. With reference to Equation 5.3 and using a value of the reference cleat porosity that has been determined independently through helium pycnometry, namely 3.2 %, the reference permeability and the coefficient C1 are determined by fitting helium experiments, whereas C2 and the mass transfer coefficient km are estimated from the experiments with CO2. Figure 5.7 shows that the model results and the experimental data are in excellent agreement, and that the permeability obtained with CO2 is clearly the smallest, with variation up to one order of magnitude, as reported too in all other similar studies, including those reporting field data (Mazumder et al., 2006; Shi and Durucan, 2007). Note also that the estimated values of


148


the mass transfer coefficients correspond to time constants of about 3 and 4.5 days for CO2 and N2, i.e. of the same order of magnitude as those used in the literature (Korre et al., 2007; Shi et al., 2008).

5.6

Mass transfer and enhanced coalbed methane (ECBM) modeling

As seen in previous Sections, gas sorption and swelling have complex effects on the variation of coal porosity and permeability and therefore on the performance of an ECBM operation. An accurate description of CO2/CH4 displacement dynamics in the coal seam is essential for the development of reliable ECBM simulators used to history-match field test data obtained during ECBM operations. The amount of CO2 stored and CH4 recovered, and the time needed for the CO2 to break through at the production well, constitute some of the information needed when designing an ECBM process. Input parameters for these models are the laboratory results presented above. Coal reservoirs are fractured systems often saturated with water, consisting of a low-permeability matrix and a high-permeability fracture network. One can distinguish up to four types of pores in coal, namely cleats where gas and water are present, macro- and mesopores where there is only free gas, and micropores where sorption takes place. The complexity of this pore structure also impacts mass transfer mechanisms and how to describe them in ECBM models. The general assumption is that the displacement of CH4 by CO2 results from a multistep process. The gas injected in the coalbed diffuses from the fracture network, through the matrix and macropores and finally to the internal surface of the coal. Here, partial pressure with respect to the adsorbed gas is reduced, causing desorption, and gas exchange takes place. The desorbed gas diffuses through the matrix and micropores, out to the fracture network where it flows to the production well (Gentzis, 2000; Totsis et al., 2004; Seto, 2007). As in the case of Equation 5.5, this mass transfer can be described through a linear driving force model by lumping gas diffusion in the different types of pore using a single mass transfer coefficient or the corresponding time constant (Bromhal et al., 2005; Sams et al., 2005). For the sake of better visualization, Fig. 5.8 shows concentration profiles in a coal seam obtained by solving the model presented in Section 5.5, where this relatively simple description of mass transfer is complemented by the description of porosity and permeability changes in the coal during injection and displacement through Equation 5.3. A typical ECBM scenario has been simulated: the coalbed was originally saturated with CH4 and the model is solved for three different compositions of the injected gas, namely pure CO2, pure N2 and a mixture of the two. Important insights into coalbed displacement dynamics emerge from the figure. In agreement with field observations and more



149

2

CO2

ci (kmol/m3)

1.5

1

CH4

0.5

0

0

50 z (m) (a)

100

N2

ci (kmol/m3)

1

0.5

0

0

1.5

ci (kmol/m3)

CH4

CO2

50 z (m) (b)

100

50 z (m) (c)

100

N2

1

0.5 CH4 0 0

5.8 Density profiles in the coal seam obtained by solving the 1D model described in Section 5.5 for different injection compositions: (a) pure CO2 injection, (b) 50 % CO2/50 % N2 and (c) pure N2 injection. Parameters: initial reservoir pressure P0 = 15 bar, injection pressure Pinj = 40 bar, production pressure Pp = 1 bar, initial permeability k0 = 1 mD. Permeability changes are described using Equation 5.3, for which the constants C1 and C2 are given in the caption of Fig. 5.6.


150


detailed simulations, injection of pure CO2 displaces the CH4 through a sharp front, due to the higher adsorptivity of the former compared to the latter. In contrast, when pure N2 is injected the front is much smoother, resulting in a produced stream of CH4 polluted with N2. As expected, injection of a mixture of CO2 and N2 results in the appearance of both the above-mentioned phenomena. Interestingly, the same trends presented above can be obtained by applying a completely different approach, where the so-called local equilibrium assumption is made, i.e. by assuming that sorption and desorption occur quickly enough that the fluid phase and the coal matrix are always in equilibrium (no mass transfer) (Seto et al., 2006; Seto, 2007; Orr Jr, 2007; Jessen et al., 2008). By neglecting dispersion phenomena and swelling effects, the equations presented in Section 5.5 are further simplified and a powerful mathematical technique, i.e. the method of characteristics, can be used to calculate the multiphase multicomponent flow in a coal bed. Even though it represents a strong simplification of the real coal seam, this model is able to describe the EBCM process in a way that sheds light on the complex injection/ displacement dynamics. In particular, the CO2/CH4 displacement mentioned in the previous section can be described by a sharp front the so-called ‘shock front’, whereas the N2/CH4 is characterized by a much broader front, i.e. the so-called ‘simple wave’. In order to take into consideration the complexity of the pore structure of coal, more detailed approaches have been proposed. These include, for example, the use of a bidisperse pore diffusion model accounting for the diffusivity of both macro-/mesopores and micropores through the corresponding time constants (Shi and Durucan, 2003, 2005a). Such a model has been further improved by extending mass transfer to the overall pore size distribution of the coal, i.e. including convective flow in cleats, convective and diffusive flows in meso- and macropores, sorption and surface diffusion in micropores, whereby diffusion is described using the Maxwell–Stefan equations (Wang et al., 2007). However, a more compact version of such a model has been shown to describe with good agreement a number of displacement experiments carried out on coal cores in different laboratories (Wei et al., 2007a,b). The 1D single-component description of a dry coal core presented above can be extended to a 3D multicomponent multiphase (coal, gas and water) model. Such models have to be solved in a 3D domain that comprises the coalbed and accounts for its geological structure and possibly heterogeneous physical features as well as for the configuration of the injection and production wells. One such reservoir simulator, namely PSU-COALCOMP developed at The Pennsylvania State University and based on a 2D description of the coalbed that assumes vertical homogeneity, has been used to study the effect of the well configuration and design on the amount of stored CO2 as compared to



151

its theoretical amount, given in principle by the sorption isotherm (Bromhal et al., 2005; Sams et al., 2005). It was shown that depending on the sorption time constant used, at the end of the project lifetime a significant portion of the swept region can still be far from equilibrium, resulting in a reduced amount of CO2 stored, i.e. down to 50 % as compared to the thermodynamic limit predicted by the sorption isotherm. By investigating different well configurations, this situation can be improved and useful design criteria can be derived. The METSIM2 simulator has also been used to analyze and optimize ECBM performance (Shi and Durucan, 2007; Durucan and Shi, 2009). With reference to Fig. 5.9, different ECBM schemes, comparing pure CO2 injection with mixed N2/CO2 injection, were investigated for a three-spot pattern of horizontal wells using public-domain coalbed reservoir properties (Durucan and Shi, 2009). For a five-year time period, mixtures rich in N2 significantly improved CH4 production compared to primary recovery, whereas pure CO2 injection led to no enhancement of CH4 production (Fig. 5.9a). This effect was attributed to the higher permeability following N2 injection in the coalbed compared to CO2. For the same reason, injection of a 25 % N2 / 75 % CO2 or 50 % N2/50 % CO2 mixture led to a larger amount of CO2 stored compared to that stored if pure CO2 was injected (Fig. 5.9b). Due to the early breakthrough of N2 at the production well, however, there is a tradeoff between the enhanced CH4 recovery and the purity of the produced gas (Durucan and Shi, 2009).

5.7

Field tests

The long-term capacity of coal seams to store CO2 needs to be proven at the field scale, in order to demonstrate that ECBM technology can be commercially deployed. Table 5.4 summarizes the field tests which have been completed in recent years, together with the field tests whose beginning is planned to occur in the near future. The first ECBM project, the Coal-Seq project at the San Juan Basin in New Mexico (USA), was the one conducted at the largest scale providing the most meaningful results. Two injection policies were tested, namely pure CO2 injection at the Allison Unit and pure N2 injection at the Tiffany Unit, respectively. In both cases, CH4 was successfully produced in a multiwell configuration over a period of more than five years (Reeves, 2005). Right after this first project, a number of other field tests started all around the world. As can be seen in Table 5.4, only a limited amount of CO2 has been injected during these field tests, when compared to the test performed at the San Juan Basin. Besides operational problems, the reason for these smaller-scale applications is that these tests exploited a single well (Gunter et al., 2005; Wong et al., 2007) or a two-well configuration (Van Bergen et al., 2007, Yamaguchi et al., 2007). Moreover,



8

45

6

13 % CO2/87 % N2 (flue gas)

40

25 % CO2/75 % N2

35 30

50 % CO2/50 % N2

25

4 75 % CO2/25 % N2

20 15

2

Primary 10 Pure CO2

0 365

Incremental recovery factor (%)

Incremental cumulative CH4 production (¥ 106 m3)

152

5 0

730 1095 Production time (day) (a)

1460

50 % CO2/50 % N2

Cumulative CO2 injection (¥ 106 m3)

4

3

75 % CO2/25 % N2

Pure CO2

2 25 % CO2/75 % N2

1

0 365

13 % CO2/87 % N2 (flue gas)

730 1095 Production time (day) (b)

1460

5.9 Simulation results of different ECBM schemes (different injection compositions) using the METSIM2 simulator: (a) incremental methane production/recovery and (b) cumulative CO2 injection as a function of time. ECBM production was set to start from the beginning of the second year and was run for three years (from Durucan and Shi, 2009).



ECBM field tests

Location

Project

Year

Gas

# of wells1

Injection

CO2 sw 0.19 kt Alberta CO2 13 % CO2/87 % N2 sw 0.11 kt Fenn Big Valley (Canada) 1999 ECBM 53 % CO2/47 % N2 sw 0.12 kt N2 sw 0.1 kt South Qinshui Basin (China) 2004 CO2 sw 0.19 kt Ishikari Coal Field (Japan) JCOP 2004 CO2 2w 0.15 kt Upper Silesian Basin (Poland) RECOPOL 2004 CO2 2w 0.76 kt Black Warrior Basin (USA) SECARB 2009 CO2 – 1 kt – 1 kt Central Appalachian Basin (USA) 2009 CO2 Illinois Basin (USA) MGSC 2008 CO2 – 0.2 t N2 mw – San Juan Basin (USA) Coal-Seq 1995 CO2 mw 370 kt San Juan Basin (USA) SWP 2008 CO2 – 75 kt 1

Well configuration: sw = single well; 2w = two-well; mw = multi-well.

Reference

Gunter et al., 2005

Wong et al., 2007 Yamaguchi et al., 2007 Van Bergen et al., 2007 Litynski et al., 2009 Litynski et al., 2009 Litynski et al., 2009 Reeves, 2005 Litynski et al., 2009


Table 5.4 Performed and planned ECBM field tests

153

154


their objective was to test the ECBM technology in reservoirs with different geological characteristics as well as to observe CO2 breakthrough within the project lifetime, usually around one year. This is information is very useful, in particular when compared to the result obtained from reservoir modeling studies (Van Bergen et al., 2007). In most cases, it was shown that gas injection can indeed enhance methane recovery. Moreover, CO2 injection yielded a reduction in permeability and, in the cases where N2 was injected, a much more rapid response coupled with an earlier breakthrough of N2 was observed. The former was attributed to the closing of the fracture associated with coal swelling, particularly evident near the well, where the CO2 pressure is high. In the case of N2 coal undergoes a net shrinking, in line with the data shown in Fig. 5.5, but the N2 injection front and the CH4 desorption front overlap so that the injected gas pollutes methane much more than in the case of CO2. These observations are in perfect agreement with the theoretical considerations made in the previous sections. Several attempts have been made to counteract the low injection rates caused by CO2 swelling: at the RECOPOL project, a frac job allowed a substantial increase in the injectivity, at least temporarily (Van Bergen et al., 2007). In the Alberta CO2 ECBM project, shut-in periods were enforced, with the aim of reducing the gas pressure close to the well (Gunter et al., 2005). Moreover, it was shown that during the injection of flue gas, a steady increase of well injectivity was observed (Gunter et al., 2005). This last option is very attractive, since it allows direct injection of the flue gas without the expensive CO2 capture step and, at the same time, keeps the permeability sufficiently high. Key performance parameters of an ECBM operation used to assess the outcome of a field test are the amount of CO2 stored and its rate of injection, as well as the amount and purity of CH4 produced. Knowledge of the CH4 amount within the coalbed before starting its primary and enhanced recovery, i.e. the so-called gas in place (GIP), is important for several reasons. First, when compared to the CH4 sorption isotherm, it defines the reservoir pressure level at which gas starts to be released during primary recovery (White et al., 2005). Second, it provides an indication of the amount of CH4 that has left the coalbed and, therefore, indirect information about the sealing efficiency of the caprock. From a storage point of view, coalbeds with a high GIP content are suitable, since the amount of CH4 which has left the seam is low and therefore the caprock sealing efficiency can be assumed to be high. Finally, the GIP is used as input data for reservoir models to predict the evolution of the ECBM operation, particularly gas injection, CO2 storage and methane production. The data obtained from the field tests constitute the information needed for the calibration of reservoir simulators (history-matching). Once validated, these models represent important tools to design ECBM processes. In fact,



155

only a thorough understanding of the different mechanisms acting during ECBM allows for critical assessment of the success or the failure of the performed field tests and the planning of future demonstration projects. This very important approach has been applied to the data of the South Qinshui Basin single-well micro-pilot test where, after successful historymatching, a larger scale multiwell field test has been designed and planned using the simulator (Shi et al., 2008). Similarly, a reservoir simulator has been used in support of the operations of the Coal-Seq project (Reeves, 2005). Durucan and Shi have consistently and successfully used the Imperial College in-house ECBM simulator METSIM2 to history-match field data from the Coal-Seq project (Allison unit) (Shi and Durucan, 2005), from the Alberta project (flue gas injection) (Shi and Durucan, 2005b) and from the JCOP project for both single- and multiwell tests (Shi et al., 2008). Some of the results referring to the JCOP project are illustrated in Fig. 5.8. In the back-production phase of the single-well test, the gas production rates (Fig. 5.10a) were used to history-match the back-produced gas compositions. An excellent agreement is obtained between field data and model prediction (Fig. 5.10b). Figure 5.10 also shows the results of the multiwell test in terms of injection rates (Fig. 5.10c) and field injection well bottom-hole pressure (Fig. 5.10d). In this case, the agreement between simulated and field measured values was achieved by fine tuning parameters, such as the coal mechanical properties and the swelling coefficient (Shi et al., 2008).

5.8

Future trends

The technical feasibility of injecting CO2 into deep, unmineable coal seams with simultaneous recovery of CH4 has been demonstrated by the performed field tests supported by fundamental studies and reservoir simulations. A unique feature of the geological storage of CO2, which therefore applies to ECBM as well, is that it is an interdisciplinary subject, at the interface between chemical engineering and earth sciences. In order to improve the confidence in the ECBM technology, a thorough knowledge of all the mechanisms taking place during the injection of CO2 as well as of those involved in its permanent containment in the reservoir is needed. Some issues requiring further research were highlighted in the different sections of this chapter and are summarized in the following. Depending on its origin, the injected CO2 may not be pure and therefore the effect of impurities such as SOx, NOx and O2 on the process needs to be studied. Of particular concern is the extrapolation of the laboratory results to the field, because of the problems associated with the heterogeneity of the coal reservoir. Moreover, as highlighted in sections 5.4 and 5.5, experiments in the laboratory need to reproduce the underground conditions as closely



120

1.0

Gas production rate (m3/d)

100

0.8 Gas

80

0.6 60 0.4 Water

40

0.2

20 0 13 Aug

18 Aug

23 Aug

28 Aug 2004 (a)

02 Sep

07 Sep

Water production rate (m3/d)

156

0.0 12 Sep

100

Produced gas composition (%)

CH4 80 Field

60

Model 40 CO2 20

0 13 Aug

18 Aug

23 Aug

28 Aug 2004 (b)

02 Sep

07 Sep

12 Sep

5.10 Results from the field test at the Ishikari coal field CO2 storage pilot project in Japan (Yamaguchi et al., 2007). Single well test: (a) field back produced gas and water rates; (b) field gas composition and model match; multi-well test: (c) CO2 injection and gas production rates; (d) field injection well bottomhole pressure (BPH) and model match (from Shi et al., 2008).


Carbon dioxide (CO2) sequestration in unmineable coal seams 3

400 CO2 injection rate

2.6

300

2.2 200 1.8 100

CO2 injection rate (t/d)

Gas production rate (m3/d)

157

1.4 Gas production rate

0 1 Oct

21 Oct

10 Nov 2004 (c)

30 Nov

1 20 Dec

17

Injection BHP (MPa)

Field 16

Model

15

14

13 9 Nov

12 Nov

15 Nov 18 Nov 2004 (d)

21 Nov

24 Nov

5.10 Continued

as possible, in order to be useful for field application. In this respect, the development of techniques allowing quantification of the effect of water on the storage capacity as well as on the displacement dynamics is very challenging, because of the intrinsic difficulty of the measurements. In addition, innovative methods have to be applied to monitor the ECBM operation properly, in particular, with respect to the long-term stability and fate of the CO2 stored.


158


Finally, the capture of the CO2 is still a cost-intensive process, and therefore direct injection of flue gas in the coal seam seems to be very attractive. As a consequence, designing the overall ECBM process so as to optimize the economics and the effectiveness of the project as a whole will necessarily be an important area of research.

5.9


As highlighted in the chapter, understanding the ECBM technology process requires knowledge both from earth sciences and chemical engineering. Useful background information on the main mechanisms acting during the ECBM process can be found in the following textbooks: ∑ ∑ ∑ ∑ ∑

Atkinson J (2007), An Introduction To the Mechanics of Soils and Foundations, London, UK, Taylor and Francis. Mavko G, Mukerji T and Dvorkin J (2003), The Rock Physics Handbook: Tools for Seismic Analysis of Porous Media, Cambridge, UK, Cambridge University Press. Ruthven D M (1984), Principles of Adsorption and Adsorption Processes, New York, Wiley. Schön J H (2003), Physical Properties of Rocks: Fundamentals and Principles of Petrophysics (Volume 18), Oxford, UK, Elsevier. Yang R T (1997), Gas Separation by Adsorption processes, London, UK, Imperial College Press.

ECBM has been dealt with in detail in the following special issues appearing in scientific journals: ∑

White C M, Smith D H, Jones K L, Goodman A L, Jikich S A, LaCount R B, DuBose S B, Ozdemir E, Morsi B I and Schroeder K T (2005), ‘Sequestration of carbon dioxide in coal with enhanced coalbed methane recovery – A review’, Energy Fuels, 19, 659–724. ∑ Karacan C Ö, Larsen J W and Esterle J S (2009), ‘CO2 sequestration in coals and Enhanced Coalbed Methane Recovery’, Int J Coal Geol, 77, 1–242. More information on ECBM field tests can be found on the web at the following addresses: ∑ ∑

COAL-SEQ field test at the San Juan Basin (USA): http://www.coal-seq. com/ RECOPOL field test in the Upper Silesian Basin (Poland): http://recopol. nitg.tno.nl/



5.10

159

References

Arri L E, Yee D, Morgan W D and Jeansonne M W (1992) ‘Modeling coalbed methane production with binary gas sorption’, SPE Paper 24363, SPE Rocky Mountain Regional Meeting, Casper, Wy, 18–21 May. Bae J S and Bhatia S K (2006) ‘High-pressure adsorption of methane and carbon dioxide on coal’, Energy Fuels, 20, 2599–2607. Bromhal G S, Neal Sams W, Jikich S, Ertekin T and Smith D H (2005) ‘Simulation of CO2 sequestration in coal beds: The effects of sorption isotherms’, Chem Geol, 217, 201–211. Busch A, Gensterblum Y and Krooss B M (2003) ‘Methane and CO2 sorption and desorption measurements on dry Argonne premium coals: pure components and mixtures’, Int J Coal Geol, 55, 205–224. Busch A, Gensterblum Y, Krooss B M and Littke R (2004) ‘Methane and carbon dioxide adsorption/diffusion experiments on coal: an upscaling and modeling approach’. Int J Coal Geol, 60, 151–168. Busch A, Gensterblum Y, Krooss B M and Siemons N (2006) ‘Investigation of highpressure selective adsorption/desorption CO2 and CH4 on coals: An experimental study’, Int J Coal Geol, 66, 53–68. Busch A, Gensterblum Y and Krooss B M (2007) ‘High-pressure sorption of nitrogen, carbon dioxide, and their mixtures on argonne premium coals’, Energy Fuels, 21(3), 1640–1645. Ceglarska-Stefanska G and Czaplinski A (1993) ‘Correlation between sorption and dilatometric processes in hard coals’, Fuel, 72, 413–417. Ceglarska-Stefanska G and Zarebska K (2005) ‘Sorption of carbon dioxide-methane mixtures’, Int J Coal Geol, 62(4), 211–222. Chaback J J, Morgan W D and Yee D (1996) ‘Sorption of nitrogen, methane, carbon dioxide and their mixtures on bituminous coals at in situ conditions’, Fluid Phase Equilib, 117, 289–296. Clarkson C R and Bustin R M (2000) ‘Binary gas adsorption/desorption isotherms: effect of moisture and coal composition upon carbon dioxide selectivity over methane’, Int J Coal Geol, 42, 241–271. Cui X J, Bustin R M and Chikatamarla L (2007) ‘Adsorption-induced coal swelling and stress: Implications for methane production and acid gas sequestration into coal seams’, J Geophys Res [Solid Earth], 112, 1–16. Day S, Duffy G, Sakurovs R and Weir S (2008a) ‘Effect of coal properties on CO 2 sorption capacity under supercritical conditions’, Int J Greenhouse Gas Control, 2, 342–352. Day S, Fry R and Sakurovs R (2008b) ‘Swelling of Australian coals in supercritical CO2’, Int J Coal Geol, 74, 41–52. Day S, Sakurovs R and Weir S (2008c) ‘Supercritical gas sorption on moist coals’, Int J Coal Geol, 74, 203–214. DeGance A E, Morgan W D and Yee D (1993) ‘High-pressure adsorption of methane, nitrogen and carbon-dioxide on coal substrates’, Fluid Phase Equilib, 82, 215– 224. Durucan S and Shi J-Q (2009) ‘Improving the CO2 well injectivity and enhanced coalbed methane production performance in coal seams’, Int J Coal Geol, 77, 214–221. Durucan S, Ahsan M and Shi J Q (2008) ‘Matrix shrinkage and swelling characteristics of European coals’, in Gale J, Herzog H and Braitsch J (eds), Greenhouse Gas Control


160


Technologies 9, Proceedings of the Ninth International Conference on Greenhouse Gas Control Technologies (GHGT9), Energy Procedia, 1, 3055–3062. Fisher G J (1992) ‘The determination of permeability and storage capacity: pore pressure and oscillation method’, in Evans B and Teng-fong W (eds), Fault Mechanics and Transport Properties of Rocks, London, UK, Academic Press, 187–211. Fitzgerald J E, Sudibandriyo M, Pan Z, Robinson Jr. R L and Gasem K A M (2003) ‘Modeling the adsorption of pure gases on coals with the SLD model’, Carbon, 41, 2203–2216. Fitzgerald J E, Pan Z, Sudibandriyo M, Robinson Jr R L, Gasem K A M and Reeves S (2005) ‘Adsorption of methane, nitrogen, carbon dioxide and their mixtures on wet Tiffany coal’, Fuel, 84, 2351–2363. Fitzgerald J E, Robinson Jr. R L and Gasem K A M (2006) ‘Modeling high-pressure adsorption of gas mixtures on activated carbon and coal using a simplified local-density model’, Langmuir, 22, 9610–9618. Gasem K A M, Robinson Jr R L and Reeves S R (2002) Adsorption of Pure Methane, Nitrogen, and Carbon Dioxide and Their Mixtures on San Juan Basin Coal, Topical Report, DOE contract No. DE-FC26–00NT40924. Gentzis T (2000) ‘Subsurface sequestration of carbon dioxide – an overview from an Alberta (Canada) perspective’, Int J Coal Geol, 43, 287–305. Gilman A and Beckie R (2000) ‘Flow of coal-bed methane to a gallery’, Transp Porous Media, 41, 1–16. Goodman A L, Busch A, Duffy G J, Fitzgerald J E, Gasem K A M, Gensterblum Y, Krooss B M, Levy J, Ozdemir E, Pan Z, Robinson Jr. R L, Schroeder K, Sudibandriyo M and White C M (2004) ‘An inter-laboratory comparison of CO2 isotherms measured on Argonne Premium coal samples’, Energy Fuels, 18, 1175–1182. Goodman A L, Busch A, Bustin R M, Chikatamarla L, Day S, Duffy G J, Fitzgerald J E, Gasern K A M, Gensterblum Y, Hartman C, Jing C, Krooss B M, Mohammed S, Pratt T, Robinson R L, Romanov V, Sakurovs R, Schroeder K and White C M (2007) ‘Inter-laboratory comparison II: CO 2 Isotherms measured on moistureequilibrated Argonne premium coals at 55 °C and up to 15 MPa’, Int J Coal Geol, 72, 153–164. Gunter W D, Mavor M J and Robinson J R (2005) ‘CO2 storage and enhanced methane production: field testing at the Fenn-Big Valley, Alberta, Canada, with application’, in Rubin E S, Keith D W and Gilboy C F (eds), Proceedings of the Seventh International Conference on Greenhouse Gas Control Technologies: GHGT7, Cheltenham, UK, IEA GHG, Vol. 1, 413–421. Harpalani S and Chen G (1995) ‘Estimation of changes in fracture porosity of coal with gas emission’, Fuel, 74, 1491–1498. Harpalani S and Chen G (1997) ‘Influence of gas production induced volumetric strain on permeability of coal’, Geotech Geol Eng, 15, 303–325. Harpalani S and Schraufnagel R A (1990) ‘Shrinkage of coal matrix with release of gas and its impact on permeability of coal’, Fuel, 69, 551–556. Hildenbrand A, Schlomer S, Krooss B M and Littke R (2004) ‘Gas breakthrough experiments on pelitic rocks: comparative study with N2, CO2 and CH4’, Geofluids, 4, 61–80. Hocker T, Rajendran A and Mazzotti M (2003) ‘Measuring and modeling supercritical adsorption in porous solids. Carbon dioxide on 13X zeolite and on silica gel’, Langmuir, 19, 1254–1267. IPCC (2005) IPCC special report on carbon dioxide capture and storage, Working Group III of the Intergovernmental Panel on Climate Change, Metz B, Davidson



161

O, de Coninck H C, Loos M and Meyer L A (eds), Cambridge, UK, Cambridge University Press. IPCC (2007) Climate Change 2007: Synthesis Report, Contribution of Working Groups I, II and III to the Fourth Assessment Report of the Intergovernmental Panel on Climate Change, Pachauri R K and Reisinger A (eds), Geneva, Switzerland. Jessen K, Tang G-Q and Kovscek A (2008) ‘Laboratory and simulation investigation of Enhanced Coalbed Methane Recovery by gas injection’, Transp Porous Media, 73, 141–159. Karacan C O (2003) ‘Heterogeneous sorption and swelling in a confined and stressed coal during CO2 injection’, Energy Fuels, 17, 1595–1608. Keller J U and Staudt R (2005) Gas Adsorption Equilibria: Experimmental Methods and Adsorption Isotherms, New York, Springer Science+Business Media, Inc. Korre A, Shi J Q, Imrie C, Grattoni C and Durucan S (2007) ‘Coalbed methane reservoir data and simulator parameter uncertainty modelling for CO2 storage performance assessment’, Int J Greenhouse Gas Control, 1, 492–501. Krooss B M, van Bergen F, Gensterblum Y, Siemons N, Pagnier H J M and David P (2002) ‘High-pressure methane and carbon dioxide adsorption on dry and moistureequilibrated Pennsylvanian coals’, Int J Coal Geol, 51, 69–92. Larsen J W (2004) ‘The effects of dissolved CO2 on coal structure and properties’, Int J Coal Geol, 57, 63–70. Larsen J W, Flowers R A, Hall P J and Carlson G (1997) ‘Structural rearrangement of strained coals’, Energy Fuels, 11, 998–1002. Levine J R (1993) ‘Coalification: The Evolution of Coal as Source Rock and Reservoir Rock for Oil and Gas’, in Law B E and Rice D D (eds), Hydrocarbons from Coal (AAPG Studies in Geology), 38, 39–77. Levine J R (1996) ‘Model study of the influence of matrix shrinkage on absolute permeability of coal bed reservoirs’, in Gayer R and Harris I (eds), Coalbed Methane and Coal Geology, London, Geological Society Special Publication, 197–212. Litynski J, Plasynski S, Spangler L, Finley R, Steadman E, Ball D, Nemeth J K, McPherson B and Myer L (2009) ‘U.S. Department of Energy’s regional carbon sequestration partnership program: overview’, in Gale J, Herzog H and Braitsch J (eds), Greenhouse Gas Control Technologies 9, Proceedings of the Ninth International Conference on Greenhouse Gas Control Technologies (GHGT9), Energy Procedia, 1, 3959–3967. Mahajan O P (1991) ‘CO2 surface-area of coals – the 25-year paradox’, Carbon, 29, 735–742. Massarotto P, Golding S D, Iyer R, Bae J S and Rudolph V (2007) ‘Adsorption, porosity and permeability effects of co2 geosequestration in permian coals’, International Coalbed Methane Symposium, Tuscaloosa, Al, 23–24 May. Mastalerz M, Gluskoter H and Rupp J (2004) ‘Carbon dioxide and methane sorption in high volatile bituminous coals from Indiana, USA’, Int J Coal Geol, 60, 43–55. Mazumder S and Wolf K H (2008) ‘Differential swelling and permeability change of coal in response to CO2 injection for ECBM’, Int J Coal Geol, 74, 123–138. Mazumder S, Karnik A and Wolf K H (2006) ‘Swelling of coal in response to CO2 sequestration for ECBM and its effect on fracture permeability’, SPE Journal, 11, 390–398. Mazzotti M, Pini R and Storti G (2009) ‘Enhanced coal bed methane recovery’, J Supercrit Fluids, 47, 619–617. Milewska-Duda J, Duda J, Nodzenski A and Lakatos J (2000) ‘Absorption and adsorption


162


of methane and carbon dioxide in hard coal and active carbon’, Langmuir, 16, 5458–5466. Mukhopadhyay P K and Hatcher P G (1993) ‘Composition of coal’, in Law B E and Rice D D (eds), Hydrocarbons from Coal (AAPG Studies in Geology), 38, 79–118. Murata K, El-Merraoui M and Kaneko K (2001) ‘A new determination method of absolute adsorption isotherm of supercritical gases under high pressure with a special relevance to density-functional theory study’, J Chem Phys, 114, 4196–4205. Orr Jr F M (2007) Theory of Gas Injection Processes, Copenhagen, Tie-Line Publications. Ottiger S, Pini R, Storti G, Mazzotti M, Bencini R, Quattrocchi F, Sardu G and Deriu G (2006) ‘Adsorption of pure carbon dioxide and methane on dry coal from the Sulcis Coal Province (SW Sardinia, Italy)’, Environ Prog, 25, 355–364. Ottiger S, Pini R, Storti G and Mazzotti M (2008a) ‘Competitive adsorption equilibria of CO2 and CH4 on a dry coal’, Adsorption, 14, 539–556. Ottiger S, Pini R, Storti G and Mazzotti M (2008b) ‘Measuring and modeling the competitive adsorption of CO2, CH4 and N2 on a dry coal’, Langmuir, 24, 9531–9540. Ozdemir E, Morsi B I and Schroeder K (2003) ‘Importance of volume effects to adsorption isotherms of carbon dioxide on coals’, Langmuir, 19, 9764–9773. Ozdemir E, Morsi B I and Schroeder K (2004) ‘CO2 adsorption capacity of Argonne Premium coals’, Fuel, 83, 1085–1094. Palmer I and Mansoori J (1998) ‘How permeability depends on stress and pore pressure in coalbeds: a new model’, SPE Reservoir Eval Eng, 1, 539–544. Pan Z and Connell L D (2007) ‘A theoretical model for gas adsorption-induced coal swelling’, Int J Coal Geol, 69, 243–252. Pini R, Ottiger S, Burlini L, Storti G and Mazzotti M (2009) ‘Role of adsorption and swelling on the dynamics of gas injection in coal’, J Geophys Res [Solid Earth], 114, (B04203). Pini R, Ottiger S, Burlini L, Storti G and Mazzotti M (2010) ‘Sorption of carbon dioxide, methane and nitrogen in dry coals at high pressure and moderate temperature’, Int J Greenhouse Gas Control, 4, 99–101. Reeves S R (2005) ‘The Coal-Seq Project: key results from field, laboratory, and modeling studies’, in Wilson M, Morris T, Gale J and Thambimuthu K (eds), Proceedings of the Seventh International Conference on Greenhouse Gas Control Technologies: GHGT7, Cheltenham, UK, IEA GHG, Vol. 2, 1399–1403. Reiss L H (1980) Reservoir engineering en milieu fissuré, Paris, France, Editions Technip. Reucroft P J and Patel H (1986) ‘Gas-induced swelling in coal’, Fuel, 65, 816–820. Reucroft P J and Sethuraman A R (1987) ‘Effect of pressure on carbon-dioxide induced coal swelling’, Energy Fuels, 1, 72–75. Romanov V N, Goodman A L and Larsen J W (2006), ‘Errors in CO2 adsorption measurements caused by coal swelling’, Energy Fuels, 20, 415–416. Saghafi A, Faiz M and Roberts D (2007), ‘CO2 storage and gas diffusivity properties of coals from Sydney Basin, Australia’, Int J Coal Geol, 70, 240–254. Sakurovs R, Day S, Weir S and Duffy G (2007) ‘Application of a modified DubininRadushkevich equation to adsorption of gases by coals under supercritical conditions’, Energy Fuels, 21, 992–997. Sakurovs R, Day S and Weir S (2008) ‘Causes and consequences of errors in determining sorption capacity of coals for carbon dioxide at high pressure’, Int J Coal Geol, 77, 16–22.



163

Sams W N, Bromhal G, Jikich S, Ertekin T and Smith D H (2005) ‘Field-project designs for carbon dioxide sequestration and enhanced coalbed methane production’, Energy Fuels, 19, 2287–2297. Seidle J P, Jeansonne M W and Erickson D J (1992) ‘Application of matchstick geometry to stress dependent permeability in coals’, SPE Paper 24361, SPE Rocky Mountain Regional Meeting, Casper, WY, 18–21 May. Seto C (2007) Analytical theory for two-phase, multicomponent flow in porous media with adsorption, PhD Dissertation, Stanford University. Seto C J, Jessen K and Orr Jr. F M (2006) ‘A four-component, two-phase flow model for CO2 storage and enhanced coalbed methane recovery’, SPE Paper 102376, SPE Annual Technical Conference and Exhibition, San Antonio, TX, 24–27 September. Shi J Q and Durucan S (2003) ‘A bidisperse pore diffusion model for methane displacement desorption in coal by CO2 injection’, Fuel, 82, 1219–1229. Shi J Q and Durucan S (2004) ‘Drawdown induced changes in permeability of coalbeds: A new interpretation of the reservoir response to primary recovery’, Transp Porous Media, 56, 1–16. Shi J Q and Durucan S (2005) ‘A numerical simulation study of the Allison Unit CO2ECBM pilot: the impact of matrix shrinkage and swelling on ECBM production and CO2 injectivity’, in Rubin E S, Keith D W and Gilboy C F (eds), Proceedings of the Seventh International Conference on Greenhouse Gas Control Technologies: GHGT7, Cheltenham, UK, IEA GHG, Vol. 1, 431–439. Shi J Q and Durucan S (2005a) ‘Gas storage and flow in coalbed reservoirs: Implementation of a bidisperse pore model for gas diffusion in a coal matrix’, SPE Reservoir Eval Eng, 8, 169–175. Shi J Q and Durucan S (2005b) ‘A model for changes in coalbed permeability during primary and enhanced methane recovery’, SPE Reservoir Eval Eng, 8, 291–299. Shi J Q and Durucan S (2007) ‘The assessment of horizontal well option for CO2 storage and ECBM recovery in unmineable thin seams: Pure CO2 vs CO2 enriched flue gas’, in Gale J, Rokke N, Zweigel P and Svenson H (eds), Proceedings of the Eighth International Conference on Greenhouse Gas Control Technologies: GHGT8, Oxford, UK, Elsevier, CD-ROM. Shi J-Q, Durucan S and Fujioka M (2008), ‘A reservoir simulation study of CO2 injection and N2 flooding at the Ishikari coalfield CO2 storage pilot project, Japan’, Int J Greenhouse Gas Control, 2, 47–57. Shimada S, Li H Y, Oshima Y and Adachi K (2005) ‘Displacement behavior of CH4 adsorbed on coals by injecting pure CO2, N2, and CO2-N2 mixture’, Environ Geol, 49, 44–52. Siemons N and Busch A (2007) ‘Measurement and interpretation of supercritical CO2 sorption on various coals’, Int J Coal Geol, 69, 229–242. Sircar S (2001) ‘Measurement of Gibbsian surface excess’, AlChE J, 47, 1169–1176. St. George J D and Barakat M A (2001) ‘The change in effective stress associated with shrinkage from gas desorption in coal’, Int J Coal Geol, 45, 105–113. Stevenson M D, Pinczewski W V, Somers M L and Bagio S E (1991) ‘Adsorption/ desorption of multicomponent gas mixtures at in-seam conditions’, SPE Paper 23026, SPE Asia-Pacific Conference, Perth, WA, 4–7 November. Sudibandriyo M, Fitzgerald J E, Pan Z, Robinson Jr. R L and Gasem K A M (2003) ‘Extension of the Ono-Kondo Lattice Model to high-pressure mixture adsorption’, Proceedings of the AIChE Spring National Meeting, New Orleans, LA, 30 March–3 April.


164


Toribio M, Oshima Y, Shimada S, Pini R, Ottiger S, Storti G and Mazzotti M (2005) ‘Adsorption measurement of supercritical CO2 on coal’, Proceedings of the International Conference on Coal Science and Technology (ICCSandT), Okinawa, Japan, 9–14 October. Totsis T T, Patel H, Najafi B F, Racherla D, Knackstedt M A and Sahimi M (2004) ‘Overview of laboratory and modeling studies of carbon dioxide sequestration in coal beds’, Ind. Eng. Chem. Res., 43, 2887–2901. Van Bergen F, Pagnier H and Krzystolik P (2007) ‘Field experiment of CO2–ECBM in the Upper Silesian Basin of Poland’, in Gale J, Rokke N, Zweigel P and Svenson H (eds), Proceedings of the Eighth International Conference on Greenhouse Gas Control Technologies: GHGT8, Oxford, UK, Elsevier, CD-ROM. Van Krevelen D W (1981) ‘Coal : typology – chemistry – physics – constitution’, in Anderson L L (ed.), Coal Science and Technology, Amsterdam, Elsevier Science Publishers. Van Wageningen W F C, Wentinck H M and Otto C (2009) ‘Report and modeling of the MOVECBM field tests in Poland and Slovenia’, in Gale J, Herzog H and Braitsch J (eds), Greenhouse Gas Control Technologies 9, Proceedings of the Ninth International Conference on Greenhouse Gas Control Technologies (GHGT9), Energy Procedia, 1, 2071–2078. Viete D R and Ranjith P G (2006) ‘The effect of CO2 on the geomechanical and permeability behaviour of brown coal: Implications for coal seam CO2 sequestration’, Int J Coal Geol, 66, 204–216. Vincent C, Dai S, Wenying C, Rongshu Z, Guosheng D, Xu R, Vangkilde-Pedersen T and Dalhoff F (2009) ‘Carbon dioxide storage options for the COACH project in the Bohai Basin, China’, in Gale J, Herzog H and Braitsch J (eds), Greenhouse Gas Control Technologies 9, Proceedings of the Ninth International Conference on Greenhouse Gas Control Technologies (GHGT9), Energy Procedia, 1, 2785–2792. Walker Jr P L, Verma S K, Rivera-Utrilla J and Khan M R (1988) ‘A direct measurement of expansion in coals and macerals induced by carbon dioxide and methanol, Fuel, 67(5), 719–726. Wang F Y, Zhu Z H, Massarotto P and Rudolph V (2007) ‘Mass transfer in coal seams for CO2 sequestration’, AlChE J, 53, 1028–1049. Wei X R, Wang G X, Massarotto P, Golding S D and Rudolph V (2007a) ‘Numerical simulation of multicomponent gas diffusion and flow in coals for CO2 enhanced coalbed methane recovery’, Chem Eng Sci, 62, 4193–4203. Wei X R, Wang G X, Massarotto P, Rudolph V and Golding S D (2007b) ‘Modeling gas displacement kinetics in coal with Maxwell-Stefan diffusion theory’, AlChE J, 53, 3241–3252. White C M, Smith D H, Jones K L, Goodman A L, Jikich S A, LaCount R B, DuBose S B, Ozdemir E, Morsi B I and Schroeder K T (2005) ‘Sequestration of carbon dioxide in coal with enhanced coalbed methane recovery – a review’, Energy Fuels, 19, 659–724. Wong S, Law D, Deng X, Robinson J, Kadatz B, Gunter W D, Ye J, Feng S and Fan Z (2007) ‘Enhanced Coalbed Methane – Micro-pilot Test at South Qinshui, Shanxi, China’, in Gale J, Rokke N, Zweigel P and Svenson H (eds), Proceedings of the Eighth International Conference on Greenhouse Gas Control Technologies: GHGT8, Oxford, UK, Elsevier, CD-ROM. Yamaguchi S, Ohga K, Fujioka M, Nako M and Muto S (2007) ‘Field experiment of Japan CO2 geosequestration in coal seams project (JCOP)’, in Gale J, Rokke N, Zweigel P and



165

Svenson H (eds), Proceedings of the Eighth International Conference on Greenhouse Gas Control Technologies: GHGT8, Oxford, UK, Elsevier, CD-ROM. Yang R T (1997) Gas Separation by Adsorption processes, London, UK, Imperial College Press. Yu H G, Zhou L L, Guo W L, Cheng J and Hu Q T (2008) ‘Predictions of the adsorption equilibrium of methane/carbon dioxide binary gas on coals using Langmuir and Ideal Adsorbed Solution theory under feed gas conditions’, Int J Coal Geol, 73, 115–129. Zhu W C, Liu J, Sheng J C and Elsworth D (2007) ‘Analysis of coupled gas flow and deformation process with desorption and Klinkenberg effects in coal seams’, Int J Rock Mech Min Sci, 44, 971–980.


6

Carbon dioxide (CO2) injection design to maximise underground reservoir storage and enhanced oil recovery (EOR)

R. Q i, T. C. L a F o rc e and M. J. B l u n t, Imperial College London, UK Abstract: We propose designs for CO2 injection to maximise storage in aquifers and to maximise both CO2 storage and enhanced oil recovery (EOR) in oil reservoirs. A review of simulation and experimental studies suggests a carbon storage strategy where CO2 and brine are injected into an aquifer together followed by brine injection alone. Based on simulation studies, this can render 80–95 % of the CO2 immobile in pore-scale droplets within the porous rock. The method does not rely on an impermeable cap rock to contain the CO2; furthermore, the favourable mobility ratio between injected and displaced fluids leads to a more uniform sweep of the aquifer leading to higher storage efficiency than injecting CO2 alone. We then consider CO2 storage in oilfields. We propose to inject more water than the traditional optimum that maximizes only oil recovery. This causes the CO2 to remain in the reservoir, increases the field life and leads to improved storage of CO 2 as a trapped phase. Again, a short period of chase brine injection at the end of the process traps most of the remaining CO2. Key words: carbon dioxide storage, residual trapping, capillary trapping.

6.1

Carbon storage in geological formations

Carbon capture and storage (CCS), the collection of CO2 from industrial sources and its injection underground, could contribute significantly to reductions in atmospheric emissions of this greenhouse gas (IPCC, 2005). Possible sites for injection include coal beds, deep saline aquifers and depleted oil and gas reservoirs. In this work, we focus on CO2 storage in aquifers and oil reservoirs because aquifers have the greatest storage potential and oil reservoirs can provide additional hydrocarbon production. Once CO2 has been injected, the principal public and environmental concern is related to the long-term fate of the stored CO2, i.e can it be guaranteed that the CO2 will remain underground for hundreds to thousands of years? In this chapter, we propose an injection design to ensure that the majority of CO2 injected is trapped rapidly and effectively. Injecting CO2 into depleted oil and gas reservoirs resulting in additional hydrocarbon recovery has the primary advantage of being economically beneficial (Lake, 1989). CO2 flooding is an effective tertiary recovery 169 © Woodhead Publishing Limited, 2010

170


mechanism that uses established injection infrastructure and the experience of the oil industry to extend the profitability of many reservoir systems. While suitable formations are easily located, they are inequitably distributed geographically (Orr, 2004). Compared with oil and gas reservoirs, deep saline aquifers are widely distributed throughout the globe, although they often have poorly characterised geology. These systems could therefore be used for the disposal of anthropogenic CO2 in locations where there are no suitable oil or gas reservoir alternatives. According to the International Energy Agency (IEA) Greenhouse R&D Program, oil and gas reservoirs have an estimated CO2 storage capacity of about 920 Gt while deep saline aquifers could store between 400 and 10 000 Gt (Gale, 2003). This is compared to annual global CO2 emissions of approximately 31 Gt. Although there are significant uncertainties in these estimates, geological formations clearly have a large storage potential. The CO2 will generally be injected underground as a supercritical fluid (the critical pressure of CO2 is 7.38 MPa, corresponding to normally-pressured reservoir depths of around 800 m; the critical temperature is 31 °C). Typical densities in formations of greater than 800 m depth range from 500–900 kg m–3. The injected CO2 will be less dense than the formation fluids and this will cause the sequestered gas to migrate to the top of the rock layer because of buoyancy forces. As we are interested in the long-term trapping of the CO2 for hundreds to thousands of years, it is imperative that the CO2 cannot escape. Over hundreds to thousands of years, the CO2 will dissolve in the formation brine forming a denser phase that will sink; a weakly acidic solution results that may react over thousands to millions of years with the host rock forming solid carbonate. While these are effective storage mechanisms, the timescales for significant dissolution or reaction mean that the CO2 remains mobile in its own phase for many years and needs to be contained under an impermeable caprock (Ennis-King and Paterson, 2002; Xu et al., 2003; Hesse et al., 2007). While in some, well-characterised systems, such as Sleipner (Korbøl and Kaddour, 1995), or in depleted oil and gas fields, there is some reasonable assurance that the caprock will contain buoyant CO2 for many centuries, if CCS is to be implemented at a scale to make a significant impact on atmospheric emissions, storing one or more Gt of carbon per year, it will be necessary, in many sites, to inject CO2 into deep permeable formations where the caprock integrity is highly uncertain. In these cases, another strategy is required to ensure safe long-term storage. Simulation studies of CO2 storage have emphasised the importance of capillary trapping (see, for instance, Ennis-King and Paterson, 2002; Kumar et al., 2005; Obi and Blunt, 2006; Juanes et al., 2006; Ide et al., 2007). When a non-wetting phase is displaced by a wetting phase in a porous


Maximising underground reservoir storage and EOR

171

medium, the non-wetting phase can be trapped in the larger pore spaces; surrounded by the wetting phase, it can no longer move and is effectively trapped. Figure 6.1 shows a two-dimensional cross-section through a threedimensional sandstone sample imaged at a scale of a few microns using micro-CT scanning. The image is obtained after water has displaced oil. This process is well established in the oil industry: water is used to displace oil from reservoirs, but typically only around half the oil is recovered since it remains trapped in the pore space (Lake, 1989). Further water injection simply leads to excessive recycling of water from injection to production wells with little or no further oil recovery. We suggest that this mechanism could be used to trap CO2, as the non-wetting phase, in storage sites. The CO2 would be trapped when it is displaced by water flow in aquifers (Ennis-King and Paterson, 2002; Kumar et al., 2005). This could occur

1350 micrometer

6.1 Pore-scale bubbles of trapped non-wetting phase (shown in grey) surrounded by the wetting phase (white). The rock is shown in black. This is a two-dimensional slice of a three-dimensional image of a Clashach sandstone obtained using micro-CT scanning at a resolution of approximately 5 mm.


172


due to a regional movement of groundwater or when a buoyant CO2 plume migrates upwards: at the trailing edge, water displaces – and potentially traps – the CO2. Juanes et al. (2006) suggested that injecting water into the aquifer would enhance this natural process, while Qi et al. (2009) proposed an injection scheme where CO2 and brine are injected together followed by chase brine. The initial water and CO2 injection – already well established in the oil industry for gas injection projects (Lake, 1989) – ensures that the CO2 has a lower effective mobility contrast with the brine in the aquifer, allowing more of the formation to be contacted by CO2, while the chase brine rapidly and effectively traps the CO2 (Qi et al., 2009). We will now briefly review evidence for the degree of capillary trapping in the literature, and then describe the storage strategy – involving injected brine – in more detail.

6.2

Experiments of capillary trapping

In petroleum engineering applications, the residual saturation – the fraction of the pore space occupied by a trapped phase – is important, since this determines how much oil cannot be recovered during waterflooding to recover oil. In core-flood experiments on rock samples a few cm long, an initial saturation of displaced fluid (typically oil) is first established and then the displacing fluid (water) is injected. The saturation at the end of the displacement is recorded. The same flooding sequence is of relevance in CO2 storage applications: CO2 is injected into an aquifer and will locally reach some saturation – which may be well below the maximum possible, since the CO2 may channel through the formation at low saturation – followed by its displacement by water after the initial injection phase. This could be caused by natural groundwater flow, buoyancy-driven upwards migration of the CO2 plume or deliberate injection of brine (chase brine) to trap the CO2. The residual saturation increases with initial saturation while it usually decreases with increasing rock porosity (Jerauld, 1997; Suzanne et al., 2003). In this discussion we will focus on experiments that have studied water-wet systems, where the trapped oil (or gas, or CO2) is the non-wetting phase. In oilfield applications, long-term contact between the grain surface and the crude may lead to wettability alteration, which may have a significant impact on the trapping of both oil and gas (or CO2). For CO2 storage applications, a more useful concept is the trapping capacity Ctrap which is the product of porosity and residual saturation (Iglauer et al., 2009). Ctrap is the measure of how much CO2 can be stored securely, as a residual phase, without reliance on caprock seals, per unit rock volume. Figure 6.2 shows data on the residual saturation as a function of initial saturation. Since the data refer to different rock types and fluid systems, as well as using different experimental techniques, there is considerable



0.15

Crowell et al., 1966 Land (Berea), 1971 Suzanne et al., 2003

Geffen et al., Jerauld, 1997 Kleppe et al., 1952 1997 Land (alundum), Ma and McKay, 1974 1971 Youngren, 1994 Pentland et al., Al-Mansoori et al., 2008 2009

173

Kralik et al., 2000 Pickell et al., 1966

Ctrap

0.10

0.05

0.00 0.0

0.1

0.2

0.3

0.4

0.5 S(nw)i

0.6

0.7

0.8

0.9

1.0

6.2 Capillary trapping capacity Ctrap (residual saturation times porosity) as a function of initial non-wetting phase saturation S(nw)i for different measurements in the literature, compiled by Iglauer et al. (2009). The plot indicates that within the considerable scatter of the data – caused by taking results from different rock types and fluid systems – there is an increase in trapping capacity with initial saturation. The solid lines are fits to experiments on sandpacks that suggest an approximately linear relationship between trapping capacity with initial saturation, followed by a constant value, as also found by Suzanne et al. (2003).

variation in the results. As expected, the degree of trapping increases with initial saturation. Highlighted are the results on sand packs that show an initially linear trend in trapping capacity with initial saturation followed by an approximately constant, maximum, trapping capacity. This also accords, approximately, with other data in the literature. Figure 6.3 shows a similar compilation of experimental data in the literature, where the maximum trapping capacity (that obtained from a high initial saturation) is plotted as a function of porosity: again, there is considerable scatter with a trend that possibly suggests an optimum porosity for trapping. Again, the scatter is not surprising, since the figure shows data from different rock types and fluid systems with different methods used to determine trapped saturation. Clearly, the lowest porosity rocks will have a low trapping capacity because there is very little available pore space, even though the residual saturation may be high. In contrast, very high porosity (generally unconsolidated) media have a low residual saturation, since the wetting phase is unable to surround the non-wetting phase effectively. The


174


0.15

Ctrap

0.1

0.05

0 0.00

0.10

0.20

Porosity

0.30

0.40

0.50

6.3 Capillary trapping capacity Ctrap (maximum residual saturation times porosity) as a function of porosity for different measurements in the literature, using the data in Fig. 6.2. The plot indicates that within the considerable scatter of the data – caused by taking results from different rock types and fluid systems – there appears to be an optimal porosity of around 0.2–0.3 that provides the maximum trapping capacity.

result is a non-monotonic relationship between trapping capacity and porosity, indicating an ideal range of porosity between approximately 0.2 and 0.3. The conclusion of this section is that at the small scale between 3 and 8 % of the total rock volume can be filled with a residual non-wetting phase. Potentially, this concept could be applied to CO2 storage, where trapped CO2 in aquifers and oil fields could be safely stored, even in the absence of a good caprock. However, to date, most of the experiments have been performed on oil–water or gas–water systems for application to hydrocarbon recovery; further work is needed that focuses on supercritical CO2–brine displacement to confirm the degree of trapping in these cases. Recent research has suggested that supercritical CO2–brine systems are not strongly waterwet which could imply a much lower trapping capacity than presented here (Chiquet et al., 2007; Plug, 2007). The other problem is that while locally CO2 may be trapped, for good overall storage efficiency it is necessary that the CO2 is both placed and trapped throughout the formation. Bearing in mind that CO2 in the subsurface is less dense and much less viscous than the brine it displaces, this is not guaranteed. The next section will use the experimental results in field-scale simulation models to design an efficient and effective storage strategy for CO2 in aquifers and oilfields. © Woodhead Publishing Limited, 2010


6.3

175

Field-scale design of storage in aquifers

We will now discuss the design of carbon storage. The simulation tool used for this work was a streamline-based simulator that used a three-phase (hydrocarbon, water and solid) four-component (oil, CO2, water and salt) formulation. The transport equations were solved along streamlines, giving a computationally efficient solution method that allowed the effects of fine-scale heterogeneity to be captured accurately. Details of the simulation method are given in Qi et al. (2009) while Fig. 6.4 shows a schematic of the phases and components considered. This is a relatively simple model in terms of its geochemistry and phase behavior but is sufficient to study the key processes of interest for this discussion: capillary trapping and dissolution. The relative permeabilities used in this study are those measured on water-wet Berea sandstone by Oak (1990). To accommodate trapping and relative permeability hysteresis we apply the model developed by Spiteri et al. (2008) that is based on pore-scale modeling studies. We will now consider how to design an injection scheme in an aquifer that traps as much CO2 as possible. In the oil industry, it is standard practice to inject gas and water together or, more commonly, in alternating slugs – since the mobility of the combination of the two phases has a lower mobility than gas alone, leading to a more stable displacement and a more efficient sweep of the reservoir (Lake, 1989). We propose the same strategy here: water and CO2 are injected together to provide a more stable displacement, forcing the CO2 into more of the formation. Brine is then injected on its own to trap the CO2. At typical reservoir conditions, CO2 is less dense and much less viscous than water and will tend to rise to the top of the formation, channeling along high-permeability pathways and bypassing most of the storage space. On the other hand, we do wish

Phases (3)

Components (4) Oil

Hydrocarbon Aqueous

CO2

Water

Salt

Solid

6.4 The simulator used to design CO2 storage considers three-phase, four-component flow and reaction. CO2 partitions in both the hydrocarbon and aqueous phases, and can precipitate as carbonate. Oil is only found in the hydrocarbon phase, water can reside in both aqueous and hydrocarbon phases (it can partially evaporate in the presence of CO2), while salt dissolved in water can precipitate if CO2 evaporates sufficient water.


176


locally the saturation of CO2 to be as high as possible, since this allows more to be trapped. Table 6.1 lists the properties used in the simulations presented in this section; we use conditions representative of a saline aquifer under the North Sea. Figure 6.5 shows the corresponding mobility ratio between injected CO2 and brine and resident brine as a function of the CO2 injection fractional flow – that is, the volume of injected fluid that is CO2. In this example, while the displacement is unstable (mobility ratio greater than 1) for pure CO2 injection, an injection fractional flow of 0.85 or lower allows a stable displacement that should lead to an improved sweep of the aquifer and greater storage. Also plotted is the mobility ratio between chase brine in the formation and the injected CO2–brine mixture. This displacement is always stable. The reason for this is that we consider the mobility of chase brine with residual CO2: this residual phase greatly lowers the mobility. We can now consider an injection sequence where both floods are stable – both the injection of CO2 and the use of brine to trap the CO2 are stable displacements that should allow the CO2 to penetrate and be trapped in a large fraction of the aquifer pore volume. Figure 6.6 shows a one-dimensional analytical solution to the governing transport equations compared to a numerical solution: the good agreement helps to validate the solver and enables us to understand the sequence of fluid fronts: the injected CO2 and brine move together, followed by chase brine. The chase brine front moves much faster than the CO2 and will soon catch up, trapping all the CO2. Near the well (distance 0) the chase brine has dissolved the residual CO2. Hence, there is no possibility of CO2 escaping back through the well: there is no CO2 at all nearby, while all the CO2 is trapped further away. Table 6.1 Parameters used in the simulations CO2 viscosity Brine viscosity Temperature Reference pressure CO2 solubility (mole fraction) at 27 MPa Porosity

6 ¥ 10–5 Pa s 5 ¥ 10–4 Pa s 80 °C 27 MPa 0.0228 0.15

1D simulation parameters Darcy velocity

1/15 m/day

3D simulation parameters CO2 injection rate Chase brine injection rate CO2 density Brine density Brine density saturated with CO2

1.065 ¥ 106 kg/day 1500 m3/day 710 kg m–3 1050 kg m–3 1061 kg m–3



177

10 Mobility ratio between CO2/brine mixture and formation brine

Mobility ratio

1 Mobility ratio = 1.0

0.1 Mobility ratio between chase brine and CO2/brine mixture during chase brine injection

0.01 0

0.1

0.2

0.3

0.4

0.5 fgi

0.6

0.7

0.8

0.9

1

6.5 The mobility ratio between an injected CO2–brine mixture and formation brine as a function of the injection fractional flow of CO2, fgi. Also plotted is the mobility ratio between chase brine (including trapped CO2) and the CO2–brine injection mixture. It is possible to choose an injection fractional flow such that both displacements are stable.

We now investigate this design in a heterogeneous three-dimensional geological model of an aquifer. The model is based on SPE10 (Christie and Blunt, 2001), a representation of a Brent-like reservoir in the North Sea that has channels, shale and over four orders of magnitude variation in permeability. CO2 and brine are injected into one corner of the model, while brine is extracted (and re-injected) from the other corner of the model. Figure 6.7 shows some example results where the distribution of CO2 in this three-dimensional displacement is shown. The injected CO2 tends to rise to the top of the system and channel along high-permeability streaks. However, the chase brine injection broadly follows the same flow paths, trapping most of the CO2 and leaving only a small amount of mobile CO2 at the fringes of the plume. This indicates that even when heterogeneity and gravity are taken into account, our injection strategy can lead to a significant fraction of the injected CO2 becoming a trapped phase just a few years after the end of CO2 injection. Figure 6.8 shows a compilation of the results for different injection fractional flows. The storage efficiency is the fraction of the porespace of the aquifer


178

Developments and innovation in CCS technology Chase brine front

Dissolution front

Advancing CO2 front

0.3

0.2

Simulation Trapped CO2

Analytical solution

Sg

Mobile CO2

0.1

0 0

400

800 Distance (m)

1200

170 m

170 m

6.6 Gas (CO2) saturation saturation (Sg) as a function of distance where brine and CO2 injection with a fractional flow of 0.5 for 1000 days is followed by 50 days of brine injection alone: the injection well is at distance zero. Near the injection well all the CO2 has dissolved. Beyond this there is the chase brine front that is moving much faster than the leading CO2. After 89 days the chase water will have trapped all the CO2. Numerical and analytical solutions are shown and are in good agreement.

Z

X 3200 m

Y 2280 m

Trapped CO2 saturation

Z

X 3200 m

Y

2280 m

0.33 0.30 0.27 0.23 0.20 0.17 0.14 0.11 0.07 0.04 0.01

Mobile CO2 saturation

6.7 Saturation distributions near the injection well for a three-dimensional simulation with an injected CO2 fractional flow of 0.85: trapped CO2 (left) and mobile CO2 (right). Twenty years of CO2 and brine injection is followed by two years of chase brine injection.



179

10 1.9 9 Ratio of the mass of brine injected to the mass of CO2 injected

1.7

1.3

7 6

Mass ratio Ratio of the mass of chase brine injected to the mass of CO2

1.1

5

Ratio of the mass of total brine injected to the mass of CO2

0.9

4

Trapping efficiency

0.7

90 % 95 %

3

0.5

2

0.3

1

0.1 0.5

0.55

0.6

0.65

0.7

0.75 fgi

0.8

0.85

0.9

0.95

1

Storage efficiency (%)

8 Storage efficiency

1.5

0

6.8 Graph of storage efficiency and amount of brine injected as a function of injection fractional flow, fgi. The storage efficiency is highest for intermediate values of fgi because of the competing effects of sweep efficiency (highest for mobility ratios less than one–see Fig. 6.5) and local trapped saturation. The total volume of brine needed to trap a given fraction of the CO2 decreases with increasing fgi, but the volume of chase brine is minimized around fgi = 0.85.

filled with CO2, while the trapping efficiency is the fraction of the injected mass that is either dissolved or residual saturation. In this example, it is possible to trap the majority of the CO2 with a relatively modest amount of chase brine injection. The storage efficiency is controlled by two competing effects. The higher the injection fractional flow, the higher the initial CO2 saturation in regions of the field where the CO2 penetrates and hence more can be trapped locally. However, the lower the fractional flow, the more favourable the mobility contrast and the CO2 sweeps more of the aquifer. The optimum is when the injection is first stable – fgi is around 0.85.

6.4

Storage in oilfields

We can perform similar studies for CO2 and brine injection into oilfields. In this case, the field is produced under waterflooding until most of the mobile oil is displaced. Then CO2 and brine are injected, both to store CO2 and to allow additional oil recovery. The details of the work are given elsewhere (Qi et al., 2008): here we simply show some illustrative results. Figure 6.9


180


52 m

Z Y

X

366 m Trapped CO2

52 m Z

Y

Mobile CO2

366 m

X CO2 volumetric fraction 0.000

0.0673 0.135 0.202 0.269 0.336 0.404 0.471

6.9 Saturation distributions for a three-dimensional simulation of CO2 injection into a waterflooded oilfield with an injected CO2 fractional flow of 0.5. The CO2 volumetric fraction is the fraction occupied by CO2 in the hydrocarbon phase. The trapped CO2 (upper) and mobile CO2 (lower) volumetric fractions are shown. 550 days of CO2 injection is followed by 30 days of chase brine injection – in this example the vast majority of the CO2 in the reservoir is trapped very rapidly.

shows the mobile and trapped CO2 saturations for an example where CO2 and brine are injected into an oilfield that has already been flooded by water. This is followed by chase brine injection. The reservoir description is the same as before, but fluids are injected through a central well and produced from wells at the corners of the field. We show that the optimal storage strategy is to inject more water than the traditional optimum for maximal oil recovery (Lake, 1989). Injecting more water helps keeps the CO2 saturation low, reducing its mobility, increasing the fraction of the reservoir contacted by CO2 and facilitating trapping; having less water leads to excessive production of injected CO2 and less is stored in the field.

6.5

Discussion and conclusions

This chapter has emphasised the importance of residual trapping in CO2 storage. In cases where it is not certain that there is a good caprock, or in regions where there may be many wells to act as possible leakage paths, it



181

is necessary to design injection so that the vast majority of the CO2 is safely stored quickly. The injection of CO2 and brine together leads to a more stable displacement, forcing the CO2 into more of the aquifer and leading to a greater storage capacity. If this is followed by a short period of chase brine injection, our simulation studies indicate that the vast majority of the injected CO2 can be trapped within a few years. This ensures storage security with only a relatively small amount of mobile CO2 that could potentially leak. We have indicated how the injection could be designed by ensuring that both the CO2 and brine injections are stable. We extended the work to injection in mature oilfields where we suggest that injecting more water than the traditional optimum for oil recovery leads to greater CO2 storage.

6.6

Future trends

While this discussion suggests a field-scale design strategy, there are two important topics for future study. First, while the compilation of literature data indicates that non-wetting phase can indeed be trapped, there is an enormous range of residual saturation. Furthermore, the experiments have not specifically studied supercritical CO2–brine systems, although now there is some work that deals with these fluids (Suekane et al., 2009). As mentioned before, there are suggestions that such systems are not strongly water-wet which may lead to less trapping than observed in analogue oil–water or gas–water experiments (Chiquet et al., 2007). Careful experimental studies of fluids of interest to carbon storage are required, coupled – as shown here – with detailed field-scale simulation. Second, while it may be persuasive to suggest a safe storage design based on core-scale experiments and numerical models, confidence in this strategy can only be gained through field demonstrations that show that the CO2 can be trapped rapidly and effectively through displacement by water.

6.7


The most comprehensive description of CCS and its potential can be found in the IPCC report: http://www.ipcc.ch/publications_and_data/publications_ and_data_reports_carbon_dioxide.htm (accessed January 2010). Detailed technical papers on trapping mechanisms, simulation studies and field trials can be found in the literature: at present, one of the better, new journals in this field is the International Journal of Greenhouse Gas Control. In addition, there is a wealth of expertise in the oil industry that can be applied to these problems: this literature is often published by the Society of Petroleum Engineers www.spe.org.


182

6.8


Acknowledgments

We would like to thank Shell under the Shell Grand Challenge on Clean Fossil Fuels and Qatar Petroleum, Shell and the Qatar Science and Technology Park under the Qatar Carbonates and Carbon Storage Research Centre for funding this work. We would also like to thank Stefan Iglauer and Christopher Pentland for their help preparing the figures.

6.9

References

Al Mansoori S K, Iglauer S, Pentland C H, Bijeljic B and M J Blunt (2009) ‘Measurements of non-wetting phase trapping applied to carbon dioxide storage,’ in Gale J, Herzog H and Braitsch J (eds), Greenhouse Gas Control Technologies 9, Proceedings of the Ninth International Conference on Greenhouse Gas Control Technologies (GHGT9), Energy Procedia, 1, 3173–3180. Caubit C, Bertin H and Hamon G (2004) ‘Three-phase flow in porous media: wettability effect on residual saturations during gravity drainage and tertiary waterflood,’ SPE 90099, Proceedings of the SPE Annual Technical Conference and Exhibition, Houston, TX, 26–29 September. Chierici G L, Ciucci G M and Long G (1963) ‘Experimental research on gas saturation behind the water front in gas reservoirs subjected to water drive,’ Proceedings of the World Petroleum Conference, Frankfurt am Main, Section II, paper 17, PD6, 483–498. Chiquet P, Broseta D and Thibeau S (2007) ‘Wettability alteration of caprock minerals by carbon dioxide,’ Geofluids 7, 112–122. Christie M A and Blunt M J (2001) ‘Tenth SPE comparative solution project: a comparison of upscaling techniques,’ SPE Reservoir Eval. Eng. 4(4), 308–317. Crowell D C, Dean G W and Loomis A G (1966) Efficiency of gas displacement from a water-drive reservoir, Bureau of Mines, Report of Investigations, 6735, US Bureau of the Interior. Delclaud J (1991) ‘Laboratory measurements of the residual gas saturation,’ Proceedings of the Second European Core Analysis Symposium, London, UK, 20–22 May, 431–451. Ennis-King J and Paterson L (2002) Engineering Aspects of Geological Sequestration of Carbon Dioxide, SPE 77809, Proceedings of the Asia Pacific Oil and Gas Conference and Exhibition, Melbourne, VIC, Australia, 8–10 October. Gale J (2003) ‘Geological storage of CO2: what’s known, where are the gaps, and what more needs to be done?,’ in Rubin E S, Keith D W and Gilboy C F (eds), Proceedings of the Seventh International Conference on Greenhouse Gas Control Technologies: GHGT7, Cheltenham, UK, IEA GHG, Vol. 1, 201–206. Geffen T M, Parrish D R, Haynes G W and Morse R A (1952) ‘Efficiency of gas displacement from porous media by liquid flooding,’ Trans. AIME, 195, 29–38. Hesse M A, Tchelepi H A, Cantwell B J and Orr Jr. F M (2007) ‘Gravity currents in horizontal porous layers: transition from early to late self-similarity,’ J. Fluid Mech., 577, 363–383. Ide T S, Jessen K and Orr Jr F M (2007) ‘Storage of CO2 in saline aquifers: effects of gravity, viscous, and capillary forces on amount and timing of trapping,’ Int. J. Greenhouse Gas Control, 1(4), 481–491.



183

Iglauer S, Wülling W, Pentland C H, Al-Mansoori S K and Blunt M J (2009) ‘Capillary trapping capacity of rocks and sandpacks,’ SPE 120960, proceedings of the SPE EUROPEC/EAGE Annual Conference, Amsterdam, the Netherlands, 8–11 June. IPCC (2005) IPCC Special Report on Carbon Dioxide Capture and Storage, Working Group III of the Intergovernmental Panel on Climate Change, Metz B, Davidson O, de Coninck H C, Loos M and Meyer L A (eds), Cambridge, UK, Cambridge University Press. Jerauld G R (1997) ‘Prudhoe Bay gas/oil relative permeability,’ SPE Reservoir Eng., 12, 66–73. Juanes R, Spiteri E J, Orr Jr F M and Blunt M J (2006) ‘Impact of relative permeability hysteresis on geological CO2 storage,’ Water Resour. Res., 42, W12418. Kantzas A, Dong M and Lee J (2001) ‘Residual gas saturation revisited,’ SPE Reservoir Eval. Eng., 4(6), 467–476. Kleppe J, Delaplace P, Lenormand R, Hamon G and Chaput E (1997) ‘Representation of capillary pressure hysteresis in reservoir simulation,’ SPE 38899, Proceedings of the SPE Annual Meeting, San Antonio, TX, 5–8 October. Korbøl R and Kaddour A (1995) ‘Sleipner vest CO2 disposalinjection of removed CO2 into the Utsira formation,’ Energy Convers. Manage., 36(6–9), 509–512. Kralik J G, Manak L J and Jerauld G R (2000) ‘Effect of trapped gas on relative permeability and residual oil saturation in an oil-wet sandstone,’ SPE 62997, Proceedings of the SPE Annual Meeting, Dallas, TX, 1–4 October. Kumar A, Ozah R, Noh M, Pope G A, Bryant S, Sepehrnoori K and Lake L W (2005) ‘Reservoir simulation of CO2 storage in deep saline aquifers,’ SPE Journal 10(3), 336–348. Lake L W (1989) Enhanced Oil Recovery. Englewood Cliffs, NJ, Prentice-Hall. Land C S (1971) ‘Comparison of calculated with experimental imbibition relative permeability,’ SPE Journal, 11(4), 419–425. Ma T D and Youngren G K (1994) ‘Performance of immiscible water-alternating-gas (IWAG) injection at Kuparuk River Unit, North Slope, Alaska.’ SPE 2860, proceedings of the SPE Annual Meeting, New Orleans, LA, 25–28 September. McKay B A (1974) ‘Laboratory studies of gas displacement from sandstone reservoirs having strong water drive,’ APEA Journal, 14, 189–194. Maloney D and Zornes D (2003) ‘Trapped versus initial gas saturation trends from a single core test,’ SCA 22, International Symposium of the Annual Meeting of the Society of Core Analysts, Pau, France, 21–24 September. Oak M J (1990) ‘Three-phase relative permeability of water-wet Berea,’ SPE 20183, Proceedings of the SPE/DOE Seventh Symposium on Enhanced Oil Recovery, Tulsa, OK, 22–25 April. Obi E I and Blunt M J (2006) ‘Streamline-based simulation of carbon dioxide storage in a North Sea aquifer,’ Water Resour. Res., 42, W03414. Orr, Jr F M (2004) ‘Storage of carbon dioxide in geological formations,’ J. Pet. Technol., 56(3), 90–97. Pentland C H, Al Mansoori S K, Iglauer S, Bijeljic B and Blunt M J (2008) ‘Measurement of non-wetting phase trapping in sand packs,’ SPE 115697, Proceedings of the SPE Annual Technical Conference and Exhibition, Denver, CO, 21–24 September. Pickell J J, Swanson B F and Hickman W B (1966) ‘Application of air-mercury and oil-air capillary pressure data in the study of pore structure and fluid distribution,’ Trans. AIME, 237, 55–61. Plug W J (2007) Measurements of capillary pressure and electric permittivity of gas-water systems in porous media at elevated pressures, PhD thesis, Delft University. © Woodhead Publishing Limited, 2010

184


Qi R, LaForce T C and Blunt M J (2008) ‘Design of carbon dioxide storage in oilfields,’ SPE 115663, Proceedings of the SPE Annual Meeting, Denver, CO, 21–24 September. Qi R, LaForce T C and Blunt M J (2009) ‘Design of carbon dioxide storage in aquifers,’ Int. J. Greenhouse Gas Control, 3, 195–205. Skauge A and Ottesen B (2002) ‘A summary of experimentally derived relative permeability and residual saturation on North Sea reservoir cores,’ SCA2002–12, Proceedings of the International Symposium of the Society of Core Analysts, Monterey, CA, USA, 22–25 September. Spiteri E J, Juanes R, Blunt M J and Orr, Jr F M (2008) ‘A new model of trapping and relative permeability hysteresis for all wettability characteristics,’ SPE Journal 13(3), 277–288. Suekane T, Nguyen H T, Matsumoto T, Matsuda M, Kiyota M and Ousaka A (2009) ‘Direct measurement of trapped gas bubbles by capillarity on the pore scale,’ in Gale J, Herzog H and Braitsch J (eds), Greenhouse Gas Control Technologies 9, Proceedings of the Ninth International Conference on Greenhouse Gas Control Technologies (GHGT9), Energy Procedia, 1, 3189–3196. Suzanne K, Hamon G, Billiotte J and Trocmé V (2003) ‘Experimental relationships between residual gas saturation and initial gas saturation in heterogeneous sandstone reservoirs,’ SPE 84038, Proceedings of the SPE Annual Meeting, Denver, Colorado, 5–8 October. Xu T, Apps J A and Pruess K (2003) ‘Reactive geochemical transport simulation to study mineral trapping for CO2 disposal in deep saline arenaceous aquifers,’ J. Geophys. Res. 108(B2), 2071.


7

Capillary seals for trapping carbon dioxide (CO2) in underground reservoirs

T. A. M e c k e l, The University of Texas at Austin, USA Abstract: Pore-scale capillary processes within geologic reservoirs and seals influence buoyancy-driven fluid migration. This chapter reviews these processes and considers their relevance to CO2 sequestration. The ability of membrane seals in water/brine-wet rocks to retard buoyant fluid migration (including CO2) relates to the capillary pressures and pore throat diameters of the seal rock. An attempt is made here to calculate anticipated ambient capillary pressures in the lowest portions of the seal, using existing laboratory data on the petrophysical properties of the CO2-brine-reservoir system as the basis and calculations carried out using a Monte Carlo approach. The values thus reached can then be used to constrain minimum seal capacities, offering the potential to predict containment capacities. The chapter concludes with a discussion of some aspects of capillary sealing which have been consdiered for hydrocarbon systems but not yet discussed for CO2 systems. Key words: capillary pressure, membrane seal, interfacial tension, contact angle, column height, carbon dioxide, sequestration, CCS.

7.1

Introduction

One of the primary concerns regarding CO2 sequestration in subsurface geologic reservoirs is the ability of overlying confining systems (seals) to effectively inhibit or retard migration of CO2 back to the land surface (containment). Confining systems that have demonstrably retained hydrocarbons are generally considered to be suitable for retaining introduced CO2 as well, although the volumetric retention may differ for different buoyant fluids due to variability in the petrophysical properties between hydrocarbons and CO2. However, the majority of the capacity in sedimentary basins for CO2 sequestration resides in the brine-filled reservoirs, which have traditionally received less attention due to their lack of economic interest (hydrocarbon accumulation). While the geologic characteristics of prospective brine formations (i.e. porosity, permeability, and net sand) may be broadly similar to nearby hydrocarbon reservoirs where reservoir properties are better known, the confining systems that overlie brine reservoirs have generally not been investigated with as much detail. The lack of data regarding seal properties and petrophysical aspects of the confining systems overlying brine reservoirs presents a challenge with regard to assessing the seal capacities of settings for CO2 sequestration with the largest capacity. 185 © Woodhead Publishing Limited, 2010

186


The aims of this chapter are to summarize capillary principles and terminology as traditionally used for hydrocarbon systems and to consider the application of these principles to CO2 sequestration in brine settings. Given the lack of direct observations of sealing characteristics of the confining systems overlying brine reservoirs, an attempt is made to use existing laboratory data on the petrophysical properties of the CO2–brine-reservoir system (i.e. interfacial tension and contact angle) to anticipate the range of expected subsurface capillary pressures and sealing potential of these poorly studied confining systems overlying brine reservoirs. By considering a wide, but laboratory-constrained, range of petrophysical properties and potential seal pore throat diameters representative of most subsurface conditions, the anticipated ambient capillary pressures in the lowest portions of a seal can be calculated. These values can then be used to constrain minimum seal capacities, thus providing a generalized perspective of anticipated containment capabilities. The chapter concludes with a discussion of select aspects of capillary sealing that have been considered for hydrocarbon systems, but have not yet been discussed for CO2 systems.

7.1.1 Capillary principles and terminology for hydrocarbon systems The processes enabling buoyant fluid migration and geologic trapping in subsurface environments are well known and have been described comprehensively for hydrocarbon systems in Thomas et al. (1968), Berg (1975), and Schowalter (1979). These processes involve the local molecular interactions and resulting forces within and between fluids and between fluids and a porous media (Vavra et al., 1992). These forces include cohesion within and between liquids and adhesion between liquids and porous solids. Wetting fluids are dominated by adhesive forces, whereas non-wetting fluids are dominated by cohesive forces. A simple example of these forces is observed when water forms beads on a glass surface. Beads result from stronger cohesive forces within the water than adhesive forces between the water and glass. The degree of adhesion for a wetting fluid is described by the contact angle (q) as measured from the solid interface to the fluid contact through the fluid. When multiple fluids are present, the contact angle is measured through the denser fluid, and the interaction of the fluids is described by the interfacial tension (s), the surface free energy that prevents one fluid from emulsifying into another. Capillary pressure is numerically defined (Thomas et al., 1968; Berg, 1975; Schowalter, 1979; Sneider, 1995; Chiquet et al., 2007; Undershultz, 2007) as the difference between the fluid pressure of the buoyant non-wetting phase (oil or gas in hydrocarbon systems, or CO2) and the ambient brine wetting phase at a given depth, and is normally expressed in terms of the relevant


Capillary seals for trapping CO2 in underground reservoirs

187

petrophysical properties of interfacial tension (s; mN/m), contact angle (q; degrees) and maximum average pore throat radius (R; m):

Pcapillary = Pfluid(non-wetting) – Pfluid(wetting) =

2 ¥ s ¥ cos(q ) R

[7.1]

Capillary pressure thus represents the force available for displacing wetting fluid from a pore or pore throat, allowing the advancement of the nonwetting phase. If the difference in fluid pressure between the non-wetting and wetting phase is greater than the product on the right side of Equation 7.1 for a given pore throat radius, then the non-wetting fluid will displace the wetting fluid and will advance through the pore throat. Any formation that inhibits buoyant fluid migration as a result of capillary forces is referred to hereafter as a membrane seal. The capillary entry pressure (Pce) for a membrane seal is the minimum capillary pressure (Pcapillary) that permits initial migration of a buoyant nonwetting fluid into the seal through a pore throat. The vertical column of nonwetting fluid that corresponds to Pce is given by H (Smith, 1966; Sneider, 1995):

H non-wetting =

Pce (rfluid(wetting) – rfluid(non-w

) ¥ 0.433

[7.2]

respective densities of ambient where rfluid(wetting) and rfluid(non-wetting) are the wetting) brine and buoyant non-wetting phase (hydrocarbon or CO2). The conversion factor 0.433 represents the pressure exerted by a cubic foot of water (hydrostatic gradient; 62.4 lbs/144 in2).

7.1.2 Capillary pressure applications to carbon dioxide (CO2) brine systems The important influence that capillary processes have on various transport and displacement mechanisms for CO2–brine systems have been noted (Chalbaud et al., 2006; Bachu and Bennion, 2008). While the subsurface properties of CO2 (e.g. density, viscosity, chemistry, interfacial tension, and contact angle) differ from hydrocarbons, the fundamental physical processes for migration and trapping remain the same, allowing the transfer of elementary mathematical treatment of hydrocarbon concepts to CO2 storage. For CO2–brine systems, CO2 commonly acts as the non-wetting phase. One difference between hydrocarbon systems and sequestration activities is that the CO2 injected into the deep subsurface will ideally be in the supercritical phase. Supercritical fluids have relatively high densities but low viscosities. The denser nature of supercritical CO2 allows for more storage efficiency of carbon per unit volume of capacity in the subsurface;


188


gas phase CO2 would occupy considerably more space for the same mass due to its extremely low density. Hydrocarbons in subsurface environments are typically not close to their critical points, retaining standard fluid properties. Despite the supercritical nature of CO2 in deep geologic environments, the fundamental physics of capillarity are not adversely affected beyond changes in wettability (s * q). Details on the dependency of these two parameters on pressure and temperature are described below. Some dissolution of CO2 into brine is likely given weak miscibility of CO2 and brine at reservoir pressures and temperatures corresponding to supercritical CO2. This CO2-saturated fluid is denser than ambient brine, and will tend to migrate deeper in the formation and should not contribute to buoyancy forces at the base of the seal, which is considered to be a two-phase brine–CO2 (supercritical) system for the calculations described below.

7.2

Calculations of anticipated capillary pressures and seal capacities

Few data have been published for capillary entry pressures of confining system rocks (low porosity and permeability; seal facies) representative of CO2–brine–seal systems at appropriate subsurface pressures and temperatures (Alberta Geological Survey, 2006). However, the laboratory data that are available can be used to constrain the range in possible ambient capillary pressures in topseals (membrane seals). These capillary pressures can then be used to infer minimum estimates of seal capillary entry pressures and associated column heights by assuming Pce in Equation 7.2 to be equal to Pcapillary in Equation 7.1. Calculations of representative capillary pressures for brine–CO2 systems have been previously difficult because too few CO2–brine–seal system petrophysical data (e.g. contact angles and interfacial tension) were available to provide any constraints on input parameters. The methodology used below incorporates recently available laboratory data to broadly assess performance of topseals for CO2 sequestration. A Monte Carlo methodology is used here to calculate capillary pressures within the lowest portion of a stratigraphic topseal using stochastic combinations of input parameters for Equations 7.1 and 7.2 in order to explore a wide range of realistic subsurface reservoir conditions. Obviously, if specific information were available for a site, more precise calculations could be made. The aim here is to model a broad range of expected conditions in order to generally evaluate probable seal performance. Given that many variables in Equations 7.1 and 7.2 are not known at a particular subsurface site without direct sampling and lab analyses, repeated calculations of capillary pressure (Equations 7.1) and associated column height (Equations 7.2) (assuming Pcapillary = Pce) are made using defined ranges of input parameters (uniform distributions). For Equations 7.2 the component



189

fluid (e.g. CO2) density variation over thickness H is neglected which is not applicable for shallow settings with compressible CO2 phase or very large H (see Chiquet et al., 2007), conditions not considered here. By considering Pcapillary to represent Pce of a seal rock, the method effectively constrains minimum seal capacities for retaining CO2. No assumption is made for dependencies between any of the input parameters in Equations 7.1 and 7.2, allowing any conceivable combination of parameters to be calculated. This provides the most conservative estimates of the ranges for Pcapillary, Pce, and H.

7.2.1 Interfacial tension data

0

0

5

0.5

10

1

15

1.5 2

20 Egermann et al. (2006) 25

Bennion and Bachu (2006a)

2.5

Bachu and Bennion (2008) 30

Chun and Wilkinson (1995)

3

Chiquet et al. (2007) 35

3.5

Hebach et al. (2002) Bennion and Bachu (2006b)

40

Yang et al. (2005) Chalbaud et al. (2006)

45 0

10

20

30 40 50 Interfacial tension (mN/m)

60

70

Depth (km) converted using 10 MPa/km

Pressure (MPa)

Interfacial tension (IFT; s) describes the surface free energy that exists between two fluid phases such as CO2 and brine, preventing one liquid from emulsifying into the other. Figure 7.1 compiles IFT data from nine laboratory measurements in CO2–brine systems (references provided in figure), and illustrates the changes that occur with increased pressure. The pressure scale has been converted to proxy depths using a fixed gradient of 10 MPa/km. It is previously demonstrated that pressure has a much more pronounced

4 4.5 80

7.1 Summary of interfacial tension data from nine laboratory studies (indicated) of brine–CO2 systems. Below depths corresponding to supercritical phase CO2 (~700 m), IFT stabilizes between 20 and 40 mN/m. This is the range used in Monte Carlo calculations of capillary pressure and column height for Fig. 7.3.


190


effect on IFT than temperature does (Chiquet et al., 2007). For depths below approximately 700 m (corresponding to 7 MPa with a pressure gradient of 10 MPa/km (essentially hydrostatic), the approximate depth at which CO 2 becomes supercritical), IFT values stabilize in the range 20–40 (Fig. 6.1). Laboratory experiments indicate that IFT stabilizes at around 30 mN/m for density differences between brine and CO2 of < 600 kg/m3 (Chalbaud et al., 2006). For subsequent calculations, IFT values used to represent a broad range of subsurface conditions are considered to vary uniformly within the range 20–40 mN/m.

7.2.2 Contact angle data The contact angle represents the angle of intersection occurring between two phases, as measured from a solid surface (e.g. clastic particle in a sandstone) through the wetting phase (often termed wettability). There are multiple quantitative techniques for measuring wettability (Amott, 1959; Donaldson et al., 1969; Anderson, 1986). Smaller angles represent water-wet conditions. There are few data constraining the contact angle for brine–CO2–seal systems at varying pressure, temperature, and salinity (density). For CO2–brine systems, Chiquet and Broseta (2005) measured contact angles between 10 and 70° on mica surfaces with varying brine salinities. The brine concentrations used (0.01, 0.1, and 1M NaCl) correspond to brine densities of 1000–1040 kg/ m3 for a temperature of 35 °C and 10 MPa (Chiquet and Broseta, 2005). Using the breakthrough pressure data of Hildebrand et al. (2002), contact angles were determined to be 40–70° at 5–6 MPa (Chiquet and Brosetta, 2005), corresponding to weakly water(brine)-wet conditions. For subsequent calculations, contact angles used are considered to vary uniformly within the range of 20–70°.

7.2.3 Fluid (CO2 and brine) density data Brine densities generally increase with depth as a result of increasing salinity and pressure. Relatively fresh (low-density) groundwater can be found at depth in some stratigraphies, although the likelihood decreases with depth. A conservative uniform range of brine salinities of 1000–1200 kg/m3 was used for the calculations presented here, corresponding to the range in Kharaka and Hanor (2004) as well as Adams and Bachu (2002) for the Alberta Basin, Canada. The lower limit corresponds to fresh water and the upper limit corresponds to brine with a salinity of 300 000 ppm at 80 °C and 30 MPa, although other conditions can result in similar brine density. For reference, seawater salinity is ~ 35 000 ppm. The density of CO2 is a function of pressure and temperature. Ambient fluid pressures vary with depth in the subsurface but are generally between



191

hydrostatic (0.443 psi/ft; 10 MPa/km) and lithostatic (~1 psi/ft; 23 MPa/km). Stratigraphies deeper than a few thousand feet may have distinct pressure compartments with unique zonal gradients, making the prediction of pressure challenging without direct observations. Often, only a range of possible fluid pressures is likely to be known for prospective injection reservoirs. Temperatures also generally increase with depth, but geothermal gradient variability makes specific predictions difficult without prior knowledge. In order to assess the likely range of subsurface CO2 densities, a particularly thorough dataset including the pressure, temperature, and depth of > 12 000 hydrocarbon-producing sands from the offshore Gulf of Mexico (GOM) was used (Minerals Management Service, 2004). These pressure and temperature conditions are considered realistic for similar depths in onshore settings prospective of CO2 capture and sequestration activities. Figure 7.2 is a plot 0 0

20 31 40

60

Temperature (°C) 80 100 120

140

160

180

7.4 10

Depth

200(ft)¥10

4

200 kg/m3 300 kg/m3

20

0.5

400 kg/m3

30 Pressure (MPa)

500 kg/m3

40

1

50 600 kg/m3

1.5

60 70

700 kg/m3

80

800 kg/m3

90 100

2

2.5 1000 kg/m3

900 kg/m3

7.2 Plot of the temperature and pressure conditions for > 12 000 hydrocarbon-producing sands from the offshore Gulf of Mexico. Symbols are shaded according to depth using scale bar at right (depth below seafloor in km). These hydrocarbon reservoir conditions are considered representative of prospective brine sequestration formations at similar elevations both onshore and offshore. Pressure and temperature conditions for pilot CO2 projects and natural accumulations are shown by star and triangle symbols, respectively, and are listed in Table 7.1. The dashed trapezoid contains CO2 densities between 600 and 800 kg/m3, representing 69 % of the calculated CO2 densities for the dataset, and is the range used in the calculations shown in Fig. 7.3.


192


of the MMS data with lines of constant CO2 density in pressure-temperature space. Shades of individual points represent the depth of the sand horizon, and the dashed trapezoid can be seen to constrain the range of CO2 density from a broad range of depths of target sandstone reservoirs to between 600 and 800 kg/m3. Thus, all conceivable targets for sequestration in the offshore GOM (and by inference onshore) result in CO2 densities less than brine, creating buoyancy forces that result in capillary pressures, as is well known. Included in Fig. 7.2 are points representing pressure and temperature conditions for some known and planned sequestration activities as well as natural CO2 accumulations listed in Table 7.1.

7.2.4 Pore throat radius data Both geotechnical and geological literature show that the mean pore throat size of mudrock (seals) declines with increasing burial depth, typically reaching values of < 10 nm at depths of 3–5 km, mainly driven by collapse of large pores (Katsube and Williams, 1994; Dewhurst et al., 1998, 1999). Katsube et al. (1991) reported petrophysical parameters for ten shale samples which show a unimodal pore size distribution mainly concentrated in the 2.5–25 nm range. The geometric means of the pore sizes are on the order of 8.7–16.2 nm. Katsube et al. (1991) suggest that pore sizes greater than 25 nm are likely disconnected from dominant flow paths. The majority Table 7.1 Petrophysical properties used for capillary pressure and column height calculations Location

Depth (m)

P (Mpa)

T (°C)

Sleipner Allison Unit, NM North Hobbs Unit, NM Weyburn Frio Zama Acid Gas Otway Gippsland Basin Flag Formation, Barrow sub-basin SACROC Browse Basin SECARB (Cranfield, MS) Lacq K12-B McElmo Dome Jackson Dome Sheep Mountain Springerville – St John McCallum Dome

880 950 1220 1450 1540 1600 2000 2000 2000 2100 2500 3140 4500 4000 2450 4800 1400 630 1630

8.6 11.37 13 20 15.17 14.5 17.72 20 20.7 18 25.5 32 48 40 24 73 13.7 6.17 16

40 48.89 43 60 57 71 93 95 92 54 123 125 150 128 71 105 58 49 60



193

of the shale samples reported by Katsube and Williams (1994) also show mean pore sizes less than 23 nm. For the purposes of the calculations here, a uniform (and generally conservative) range of pore size between 10 and 25 nm was used to represent sealing horizons associated with the reservoirs investigated, representing the central portion of the range indicated by Yang and Aplin (2007) for 30 natural mudstones (1–100 nm). Lower pore sizes would generally increase Pcapillary and H values (equations 7.1 and 7.2), while higher pore sizes would decrease values.

7.3

Monte Carlo predictions of capillary pressure within a reservoir seal

The various input data and the ranges used in the subsequent calculations are described below and summarized in Table 7.2. Uniform distributions are used to equally sample all ranges of each of the variables. Initially, a total of 50 000 calculations of capillary pressure (1) and column height (2) were made using a Monte Carlo approach, drawing each parameter value independently from uniformly-distributed ranges of properties (Table 7.2, line 1). The distribution of capillary pressures and corresponding column heights for the full range of petrophysical parameters can be seen in Fig. 7.3a,b. Considering the full range of theoretical petrophysical conditions (especially contact angle > 70° and/or IFT < 20 mN/m), the distribution of column heights is skewed to low values (highest frequency < 10 m), indicating generally poor sealing capacity to predicted (calculated) ambient capillary pressures. The distribution of capillary pressures and corresponding column heights for a subset of the input parameters that are more likely to exist in the subsurface (Fig. 7.3c; white box) using the laboratory constrained ranges indicated in Table 7.1 (line 2) can be seen in Fig. 7.3d. It is clear from Fig. 7.3d that the more restricted (anticipated) range of input petrophysical parameters results in a distribution of column heights that center on significantly higher values than the column heights generated from the full range of petrophysical parameters seen in Fig. 7.3a,b. This indicates that the anticipated conditions Table 7.2 Pilot CO2 project and natural accumulation reservoir conditions used in Figure 7.3 Run s (mN/m) q (deg.) rbrine (kg/m3)

rCO2 R (nm) (kg/m3)

Number of calculations

Full range Limited range

500–700 500–700

50 000 61721

0–90 20–40

0–90 20–70

1000–1200 1000–1200

1

10–25 10–25

This number maintains similar data density for both runs, allowing direct comparisons of distribution frequencies to be made.


194


Column height (m)

600 500

800

400

600 400

80

200 0

60 80

40 60

40 20 Contact angle (deg) (a)

0

20 0 IFT (mN/m)

300 200 100

Column height (m)

generally lead to reasonable potential minimum sealing capacity (i.e. > 10 m minimum sustainable column height). It is important to note that the values in Fig. 7.3 are minimum values, assuming that the capillary pressures calculated represent the minimum capillary entry pressure (Pcapillary = Pce). In all likelihood, the entry pressures for the overlying seals can be expected to exceed the minimum values reported here (Alberta Geological Survey, 2006), meaning that topseals are quite likely capable of higher column heights than the values indicated. Indeed, most natural geologic accumulations of

Frequency (% of total)

12 10 8 6 4 2 0 0

50

100 200 Column height (m) (b)

300

7.3 Calculated capillary pressures and minimum column heights for 50 000 independent calculations of Equations 7.1 and 7.2 using uniform distributions with ranges shown in Table 7.2. This number of calculations results in well-defined distributions. (a) Perspective view of the distribution of calculated column heights for the conditions on line 1 of Table 7.2. (b) Histogram showing distribution of calculated column heights in (a). Note highest frequency at lowest column heights (i.e. 10s of m). (c) Map view of data in (a), with white box defining most probable range of IFT and contact angle conditions shown in Table 7.2 line 2. (d) Histogram of calculated column heights for conditions within white box in (c). Note lack of solutions < 10 m, suggesting reasonable seal capacity may be fairly common.



195

90 Contact angle (deg)

80 70 60 50 40 30 20 10

Frequency (% of subset)

0 0

10

20

30

40 50 60 IFT (mN/m) (c)

70

80

90

20 18 16 14 12 10 8 6 4 2 0

0

50

100 200 Column height (m) (d)

300

7.3 Continued

CO2 have column heights on the order of 10s–100s of meters, indicative of good sealing capabilities. Figure 7.4 presents the calculated Monte Carlo ambient capillary pressure values in the context of column heights and capillary pressures for a range of constant fluid density contrasts shown by the solid diagonal lines. Typical values for seal capillary entry pressures are indicated, and are notably higher than the ambient capillary pressures calculated for the range of fluid density contrasts used in the Monte Carlo calculations. These results provide a comfortable level of confidence that subsurface seals can be generally expected to perform adequately in most subsurface environments, suggesting that topseal analyses should be focused as much on structural complications (e.g. (micro)fractures and faults) and geochemical interactions as on general capillary retention.

7.4

Discussion

Some relevant topics regarding petroleum migration and trapping have not been previously discussed in the context of CO2 sequestration and capillary © Woodhead Publishing Limited, 2010

196


106

105

Column height (m)

104

103

102

Calculated ambient capillary pressures Fluid density contrast 10 kg/m3

Typical seals: 1–10s MPa

50 kg/m3

101

100 kg/m3 500 kg/m3

100 100

101

1000 kg/m3

102 103 Capillary pressure (kPa)

104

105

7.4 Calculated minimum capillary pressures (symbols) for 50 000 independent calculations of Equations 7.1 and 7.2 using uniform distributions with ranges shown in Table 7.2. Solid diagonal lines represent the column heights expected for a range of capillary pressures for constant values of fluid density contrast (brine–CO2) indicated in boxes. Note that calculated values of minimum capillary pressure (symbols) are significantly lower than typical seals, suggesting reasonable seal capacity may be fairly common.

sealing. The important elements of those topics are embodied in three related publications (Bjorkum et al., 1998; Clayton et al., 1999; Rodgers et al., 1999), and will be summarized here with relevance to CO2 sequestration. The cited papers focus on the degree to which a continuous water phase exists that connects a water-wet reservoir and a water-wet seal. The presence of such a connection, even at residual saturation, is considered by Bjorkum et al. (1998) to disallow a pressure difference in the water phase between the uppermost reservoir pores and the lowermost seal pores. Thus, they argue that overpressure in a water-wet reservoir should not contribute to, or push, hydrocarbons through a water-wet seal, and overpressured water-wet reservoirs should not be more prone to capillary leakage than normally pressured reservoirs. Given that the petrophysics of capillarity in the hydrocarbon–brine system is mathematically treated in the same manner as in the CO2–brine system, and that injection into brine reservoirs ensures they are water-wet (even if weakly in the presence of CO2), it seems logical that the same arguments can be made for water-wet reservoirs undergoing CO2 sequestration: overpressure in the water phase in a reservoir is not expected to enhance capillary leakage of CO2. Only the pressure within © Woodhead Publishing Limited, 2010


197

free-phase supercritical CO2 at the base of a seal should contribute to CO2 leakage, which should occur at a time when pressure in the CO2 phase exceeds the seal entry pressure to CO2. At long time scales required by low seal permeabilities, brine migration should occur from an overpressured reservoir into the overlying seal without free-phase migration of CO2. This is supported by the conclusions of Underschultz (2007) and Teige et al. (2005). A second element of the discussion topics being considered here has interesting implications for the risk of hydrofracturing. Bjorkum et al. (1998) argue that larger hydrocarbon columns should not present additional hydrofracture risk to top seal and that lower density buoyant phases such as gas should not increase the risk of hydrofracture compared to higher density fluids (i.e. oil). By this logic, CO2 accumulation in brine aquifers should not present additional risk of hydrofracture with increasing free-phase column height relative to the risk of hydrofracture from denser fluids such as oil. This concept underlies a growing consensus that reservoirs that retained oil or gas for geologic periods of time are good candidates for retaining CO 2 at similar timescales. A third element of the cited discussion also has important implications for CO2 migration out of reservoirs through topseals. The discussion of water-phase connectivity suggests brine will move from overpressure in the injection reservoir to lower pressure above the reservoir (i.e. into the seal, assuming normal pressure gradient) while the free-phase CO2 will not. However, CO2 dissolved into the brine phase can potentially migrate with the brine into the seal. Bjorkum et al. (1998) point out that for methane accumulations, this mechanism could allow methane to migrate into and through a seal over geologic timescales and eventually exsolve in gas phase at shallower stratigraphic intervals. They suggest that an explanation for ‘gas chimneys’ observed in seismic data may be a result of exsolution of gas from upward-migrating brine (Heggland, 2005). The degree to which the exsolution of CO2 from brine is possible will be influenced by the magnitude of the pressure drop vertically along the flow path (pressure gradient) as well as other physical gradients affecting solubility such as temperature and salinity (Kuo, 1997). Using solubility information in Duan and Sun (2003) and Duan et al. (2006) and general principles outlined in Kuo (1997) and Spycher et al. (2003), the volume of dissolved CO2 gas entrained with upward-migrating brine can be estimated. Consider a 2M NaCl brine existing in an overpressured sandstone at 3 km (~10,000 ft) with a temperature of 125 °C (260 °F) and a pressure of 35 MPa that migrates vertically due to an imposed pressure gradient related to CO2 injection at an average vertical rate of 1 mm/yr (Garven, 1995). The solubility of CO2 in brine at these conditions is 0.9567 mol CO2/ kg brine. Considering the migration rate per square meter, this is equivalent to 45


198


metric tonnes of CO2 migrating upward with the brine per square kilometer per year, which is trivial compared to a large-scale (i.e. > 1 million tonne per year) injection. This brine will have migrated vertically only 100 m in 100 k.a. If the change in conditions between these zones represents 1 MPa (essentially hydrostatic) and 25 °C (essentially geothermal) and no change in salinity is considered, CO2 solubility would change by 0.04 mol CO2/ kg brine (from 0.956 to 0.914 mol/kg). Thus, similar to the discussion of methane migration in brine presented in Bjorkum et al. (1998), it appears that it would take geologic time of order 105 years for this dissolved transport mechanism to have an appreciable impact on CO2 mass redistribution. Gilfillan et al. (2009) conclude that the solubility of CO2 in formation waters may be a dominant trapping mechanism, and further suggest that due to high potential solubility, long-term anthropogenic CO2 storage models should focus on the mobility of CO2 dissolved in water. While this is certainly reasonable within permeable reservoirs, the calculation above suggests that CO2 migrating vertically through a water-wet seal as a dissolved phase in brine is likely to have minor impact on CO2 mass redistribution over human timescales. Even if average vertical brine migration rates were 10 times larger than above (10 mm/yr; extremely high), the flux would still only be 450 tons CO2 per square kilometer per year that is carried as a dissolved phase in brine. This extremely high flux would have to be occurring over a 100 km2 area to represent annual 5 % loss from a 1 million tonne per year injection. This discussion suggests that discrete rapid flow pathways such as (micro) fractures and faults are more likely to have a significant impact on unwanted brine and/or CO2 migration through a topseal than flow overcoming topseal capillary conditions.

7.5

Conclusions

Existing measured petrophysical property data for the CO2–brine–seal system allow anticipated ambient capillary pressures in the lowermost portion of the topseal to be calculated. If anticipated ambient capillary pressures are (conservatively) assumed to be the minimum capillary entry pressures for seal rocks overlying reservoirs, seal capacities should generally be able to support thicker CO2 columns than typical reservoir thickness. The general conclusion can be made that many subsurface environments with typically low topseal porosities should provide adequate CO2 sealing capacity. A discussion of some controversial topics in hydrocarbon leakage and trapping in the context of CO2 sequestration suggests that (i) overpressure in the water phase of an injection reservoir should not enhance capillary leakage of CO2; (ii) CO2 accumulation in brine aquifers should not present additional risk of hydrofracture with increasing free-phase column height



199

or relative to denser fluids such as oil; and (iii) while CO2 migration as a dissolved phase within brine may represent a concern for discrete high-volume pathways such as faults and fractures, such migration through a topseal is unlikely to represent a significant mechanism for mass loss of CO2 from the injection reservoir.

7.6

Future trends

With increasing interest in the potential for geologic CO2 sequestration, an increasing amount of laboratory data (interfacial tension, contact angle, relative permeability) on the CO2–brine system is expected. Much of this may be relatively site-specific, using known reservoir formations and ambient reservoir conditions for anticipated sequestration projects. An important contribution in coming years would be a summary of confining system properties, including pore throat distributions in fine grained sealing formations overlying prospective brine formations. The geochemical effects of CO2 in seal facies will also be an area of increased interest. Depending on seal facies geochemistry, the potential exists for both seal enhancement and degradation due to presence of CO2. Additionally, the degree of seal facies diagenesis may play an important role in seal capacity, with the mineralogic transition of smectite to illite clays being influential, perhaps mostly from a geomechanical perspective. Finally, the area of fault seal for sequestration is receiving increased attention. This is an area where concepts developed in the hydrocarbon industry will also be applicable for sequestration applications.

7.7


A 2005 volume specific to sequestration edited by Sally Benson entitled Carbon Dioxide Capture for Storage in Deep Geologic Formations – Results from the CO2 Capture Project (Vol. 2) covers storage integrity, storage optimization, monitoring and verification, and risk assessment. General volumes on seals as understood in petroleum applications occur in 1997 Memoir #67 of the American Association of Petroleum (AAPG) Geologists entitled Seals, Traps, and the Petroleum System and 2005 Memoir #85 entitled Faults, Fluid Flow, and Petroleum Traps. The 2005 AAPG Hedberg Series 2 Evaluating Fault and Caprock Seals contains many related articles. There will also be a summary volume from the August 2009 Hedberg Research Conference on carbon dioxide storage, expected in 2010. The title is Carbon Dioxide in Geological Media – State of the Science edited by Grobe, Pashin, and Dodge.


200

7.8


Acknowledgements

Helpful reviews prior to submission were provided by JP Nicot, Jiemin Lu, and Susan Hovorka (TX Bureau of Economic Geology). Additional reviews were provided as part of the submission process. The author acknowledges support from the Gulf Coast Carbon Center and its industrial sponsors, as well as the Bureau of Economic Geology and Jackson School of Geosciences at the University of Texas at Austin.

7.9

References

Alberta Geological Survey (2006) Test Case for Comparative Modeling of CO2 Injection, Migration and Possible Leakage – Wabamun Lake Area, Alberta, Canada, datasets, Edmonton. AB, available at: http://www.ags.gov.ab.ca/co2_h2s/wabamun/Wabamun_ base.html (accessed January 2010). Adams JJ and Bachu S (2002) Equations of state for basin geofluids: algorithm review and intercomparison for brines, Geofluids, 2: 257–271. Anderson WG (1986) Wettability literature survey-Part 2: Wettability measurement, J. Pet. Technol., 37: 1246–1262. Amott E (1959) Observations relating to the wettability of porous rock, Trans. AIME, 216: 156–162. Bachu S and Bennion B (2008) Effects of in-situ conditions on relative permeability characteristics of CO2–brine systems, Environ Geol., 54, 1707–1722. Bennion B and Bachu S (2006a) Dependence on temperature, pressure and salinity of the IFT and relative permeability displacement characteristics of CO2 injected in deep saline aquifers, SPE Annual Technical Conference and Exhibition, 24–27 September, San Antonio, TX, SPE #102138. Bennion B and Bachu S (2006b) The impact of interfacial tension and pore size distribution/ capillary pressure character on CO2 relative permeability at reservoir conditions in CO2–brine systems, SPE Annual Technical Conference and Exhibition, 24–27 September, San Antonio, TX, SPE #99325. Berg RR (1975) Capillary pressures in stratigraphic traps, AAPG Bulletin, 59: 939– 956. Bjorkum PA, Walderhaug O et al. (1998) Physical constraints on hydrocarbon leakage and trapping revisited, Pet. Geosci., 4(3): 237–239. Chalbaud C, Robin M and Egermann P (2006) Interfacial tension data and correlations of brine/CO2 systems under reservoir conditions, SPE Annual Technical Conference and Exhibition, 24–27 September, San Antonio, TX, SPE 102918. Chiquet P and Broseta D (2005) Capillary alteration of shaly caprocks by carbon dioxide, SPE Europe/EAGE Annual Conference, 13–16 June, Madrid, Spain, SPE 94183. Chalbaud C, Robin M, Lombard J-M, Martin F, Egerman P and Bertin H (2009) Interfacial tension measurements and wettability evaluation for geological CO2 storage, Advances in Water Resources, 32: 98–109. Chiquet P, Daridon J-L, Broseta D and Thibeau S (2007) CO2/water interfacial tensions under pressure and temperature conditions of CO2 geological storage, Energy Convers. Manage., 48: 736–744. Chun B-S and Wilkinson GT (1995) Interfacial tension in high-pressure carbon dioxide mixtures, Ind. Eng. Chem. Res., 34: 4371–4377.



201

Clayton CJ, Bjorkum PA et al., (1999) Discussion: ‘Physical constraints on hydrocarbon leakage and trapping revisited’ by P.A. Bjorkum et al., Petroleum Geoscience, 5(1): 99–101. Dewhurst DN, Aplin A and Yang Y (1998) Compaction-driven evolution of porosity and permeability in natural mudstones: an experimental study, J. Geophys. Res., 103(B1), 651–661. Dewhurst DN, Yang Y and Aplin A (1999) Permeability and fluid flow in natural mudstones, in Aplin, A., Fleet, A.J. and Macquaker, J.H.S. (eds), Muds and Mudstones: physical and fluid-flow propertities, Special Publication No. 158. Geological Society, London, UK, 23–43. Donaldson EC, Thomas RD and Lorenz PB (1969) Wettability determination and its effect on recovery efficiency, SPE J., 9: 13–20. Duan ZH and Sun R (2003) An improved model calculating CO2 solubility in pure water and aqueous NaCl solutions from 273 to 533 K and from 0 to 2000 bar, Chem. Geol., 193(3–4): 257–271. Duan ZH, Sun R, Zhu C and Chou I-M (2006) An improved model for the calculation of CO2 solubility in aqueous solutions containing Na+, K+, Ca2+, Mg2+, Cl–, and SO42–, Mar. Geochem., 98: 131–139. Egermann P, Chalbaud C, Duquerroix J–P and Le Gallo Y and (2006) An integrated approach to parameterize reservoir models for CO2 injection in aquifers, SPE Annual Technical Conference and Exhibition, 24–27 September, San Antonio, TX, SPE #102308. Garven G (1995) Continental-scale groundwater flow and geologic processes, Annu. Rev. Earth Planet. Sci., 23: 89–117. Gilfillan SM, Lollar BS, Holland G, Blagburn D, Setvens S, Schoell M, Cassidy M, Ding Z, Zhou Z, Lacrampe-couloume G and Ballentine CJ (2009) Solubility trapping in formation water as dominant CO2 sink in natural gas fields, Nature, 458: 614–618. Hebach A, Oberhof A, Dahmen N, Kogel A, Ederer H and Dinjus E (2002) Interfacial tension at elevated pressures-Measurements and correlations in the water + carbon dioxide system, J. Chem. Eng. Data, 47: 1540–1546. Heggland R (2005) Using gas chimneys in seal integrity analyses: A discussion based on case histories, in Boult P and Kaldi J, (eds), Evaluating Fault and Cap Rock seals, AAPG Hedberg Series, no. 2, American Association of Petroleum Geologists, Tulsa, OK, 237–245. Hildebrand A, Schlomer and Krooss BM (2002) Gas breakthrough experiments on finegrained sedimentary rocks, Geofluids, 2: 3–23. Katsube TJ and Williams MA (1994) Effects of diagenesis on clay nanopore structure and implications for sealing capacity, Clay Minerals, 29: 451–461. Katsube TJ, Mudford BS and Best ME (1991) Petrophysical characteristics of shales from the Scotian Shelf, Geophysics, 56(10): 1681–1689. Kharaka YK and Hanor JS (2004) Deep fluid in the continents: I. Sedimentary basins, in Treatise on Geochemistry, V. 5, Surface and groundwater, weathering, and soils: Elsevier-Pergamon, Oxford, UK, 499–540. Kuo L-C (1997) Gas exsolution during fluid migration and its relation to overpressure and petroleum accumulation, Mar. Pet. Geol., 14(3): 221–229. Minerals Management Service, (2004) Atlas of Gulf of Mexico Gas and Oil Sands, US Department of the Interior, New Orleans, LA. Rodgers S, Bjorkum PA, et al. (1999) Discussion: ‘Physical constraints on hydrocarbon leakage and trapping revisited’ by P. A. Bjorkum et al. – further aspects, Petroleum Geoscience, 5(4): 421–423. © Woodhead Publishing Limited, 2010

202


Schowalter T (1979) Mechanics of secondary hydrocarbon migration and entrapment, AAPG Bull., 63(5): 723–760. Smith DA (1966) Theoretical considerations of sealing and non-sealing faults, AAPG Bull., 50: 363–374. Sneider RM (1995) Evaluation of Seals and flow barriers, RM Sneider Exploration, Houston, TX. Spycher N, Pruess K and Ennis-King J (2003) CO2-H2O mixtures in the geological sequestration of CO2. I. Assessment and calculation of mutual solubilities from 12 to 200 C and up to 600 bar, Geochim. Cosmochim. Acta, 67(16): 3015–3031. Teige GMG, Hermanrud C, Thomas WH, Wilson OB and Bolas HMN (2005) Capillary resistance and trapping of hydrocarbons: a laboratory experiment, Pet. Geosc., 11: 125–129. Thomas LK, Katz DL and Tek MR (1968) Threshold pressure phenomena in porous media. SPE J., 243: 174–84. Underschultz J (2007) Hydrodynamics and membrane seal capacity, Geofluids, 7: 148–158. Vavra CL, Kaldi JG and Sneider RM (1992) Geological applications of capillary pressure: A review, AAPG Bull., 76(6): 840–850. Yang Y and Aplin A (2007) Permeability and petrophysical properties of 30 natural mudstones, J. Geophys. Res., 112, B03206. Yang D, Tontiwachwuthikul P and Gu Y (2005) Interfacial tensions of the crude oil + reservoir brine + CO2 systems at pressures up to 31 MPa and temperatures of 27 °C and 58 °C, J. Chem. Eng. Data, 50: 1242–1249.


8

Measurement and monitoring technologies for verification of carbon dioxide (CO2) storage in underground reservoirs R. A. C h a d w i c k, British Geological Survey, UK Abstract: The chapter reviews some of the current technologies available for storage site monitoring, focusing on a limited range of core monitoring technologies required to provide storage site assurance at the industrial scale. Monitoring strategy has two elements: deep-focused for storage performance testing and verification and the early detection of deviations from predicted behaviour; and shallow-focused for leakage detection, verification of emissions performance and public acceptance. Key deep-focused monitoring technologies include 3D time-lapse seismic and downhole pressure and temperature measurement. For shallow monitoring, key technologies include soil gas, surface flux and atmospheric measurement. Selection of suitable monitoring strategies is highly site-specific, and tool testing and development is ongoing. Key words: CO2 storage, carbon sequestration, monitoring, verification, time-lapse monitoring.

8.1

Introduction

This chapter reviews some of the current technologies available for storage site monitoring and provides examples of their deployment at storage sites worldwide. Previous publications have considered a wide range of potential monitoring tools; the approach taken here is different, focusing on a limited range of key monitoring technologies required to provide storage site assurance at the industrial scale. The first section outlines storage processes and how these relate to generic site monitoring aims. The second section discusses deep-focused monitoring for storage performance verification and early warning of performance deviations. The key technologies, 3D time-lapse seismic and downhole pressure and temperature measurements are described with examples of their deployment at current storage sites. The utility of other deep-focused tools is also discussed. The third section describes shallow monitoring for leakage detection, verification of emissions performance and public acceptance. The key technologies, soil gas, surface flux and atmospheric measurement are described with examples. Other ancillary tools are also outlined. The next section discusses generic monitoring issues and integrated monitoring strategies. The final section concludes with a brief summary and outline of current and future research directions. 203 © Woodhead Publishing Limited, 2010

204

8.2


Background to storage site monitoring

Large-scale storage projects will likely involve very long-term storage of hundreds of millions of tonnes (Mt) of carbon dioxide (CO2). To be widely accepted, storage will have to be safe and demonstrably effective from an emissions reduction standpoint. Storage sites will, therefore, need to be monitored both to establish the current performance of the site and to constrain and calibrate predictions about its future behaviour (Benson et al., 2004; DTI, 2005). The utility of setting site performance thresholds is an issue of current regulatory debate. In addition to the need for maintaining local health and safety (see below), a logical way of establishing satisfactory containment performance would be to estimate how well a nominal storage site should perform in order to fulfil its basic emissions reduction function. Lindeberg (2003) showed how different storage retention times were related to future stabilised atmospheric concentrations – sites retaining CO2 for several thousand years (or longer) can be considered as providing effective mitigation. In a simpler treatment, Hepple & Benson (2003) calculated global site leakage rates, as a percentage of the total amount stored, consistent with a range of IPCC emission scenarios and atmospheric stabilisation targets. They concluded that stabilization at any atmospheric CO2 level less than 550 ppm would require annual leakage rates to be less than 0.01 %. CO2 can be injected into the pore spaces of an underground reservoir rock via one or more wells, permeating the rock, and displacing some of the fluid (commonly saline water) that originally occupied the pore spaces. Given likely injection depths in the range 1000–3000 m, CO2 would form a compressible fluid, with a density of between 300 and 800 kgm–3 (depending on geothermal gradient). Injected CO2 would therefore be buoyant, with a strong tendency to move upwards through the storage reservoir until impeded by a sealing barrier that prevents its further vertical migration. Horizontal or vertical permeability barriers, such as shale layers or faults, will impede movement within the reservoir and favour intra-reservoir trapping. Migration out of the reservoir would be facilitated by transmissive faults, caprock permeability or degraded wellbores (Fig. 8.1). For the purposes of describing the movement of CO2 in and around the primary storage reservoir, it is convenient to define two distinct terms. Migration is here defined as movement of CO2 within the storage reservoir and the surrounding subsurface. Leakage is defined as movement of CO2 from the geosphere either to the atmosphere at the land surface, or to surface water or potable shallow aquifers. In designing a monitoring programme to address migration and potential leakage over both the short term (the injection period) and long term, a detailed site performance assessment is needed to determine a conceptual envelope of possible migration and leakage scenarios. Leaks may not necessarily



Underwater sampling

Soil gas survey

Seismic reflection Gravimetry

Water table

Diffusion

Spontaneous potential

Permanent soil/air gas measurements

Vadose zone CO2 plume

Plume

Plume

Diffusion

Plume Plume

Plume

Casing corrosion

Annular flow

Separate phase CO2

CO2 in low-permeability environment

Caprock CO2

CO2 dissolved in groundwater

CO2

Diffusion CO2

205

8.1 Schematic diagram of storage site migration and leakage scenarios with typical monitoring technologies (from Chadwick et al., 2009a).

Verification of CO2 storage in underground reservoirs

Airborne monitoring for CO2 leaks

206


occur directly above the storage site but will be strongly influenced by the local geological structure. For example, in the case of migration up gently dipping permeable strata, leaks may appear many kilometres from the storage site and the area needing to be covered by a monitoring programme may be much larger than the intended footprint of storage within the primary reservoir itself (Fig. 8.1). Additionally, leaks may not occur for hundreds of years if the leakage path is long, but thereafter could be highly significant. In this respect, realistic long-term simulations of future site behaviour are a prerequisite for satisfactory site operation, monitoring and closure.

8.2.1 Monitoring objectives In light of the above, effective site monitoring needs to address key aspects of predicted site performance (see also Pearce et al., 2007). Monitoring objectives are likely to include the ability to: ∑

demonstrate that the site is currently performing effectively and safely; ∑ identify deviations from expected site performance and guide suitable remediative actions if necessary; ∑ calibrate predictive models of future storage site behaviour to permit satisfactory site closure; ∑ provide warning of any future hazardous surface leakage; ∑ identify and measure surface leakage should it occur.

Given that most well-characterised storage sites are unlikely to show seriously deviant behaviour during operational phase, perhaps the most important function of monitoring is to show that site behaviour is properly understood and that longer-term predictions of site performance are likely to be robust. The latter is likely to form a key element of site closure. Perhaps the ‘holy grail’ of site monitoring is to provide reliable quantitative information on CO2 amounts in the storage system, either in the deeper subsurface or in terms of near-surface fluxes. A monitoring tool can reasonably be considered to provide ‘quantitative’ information if its response has a diagnostic relationship to CO2 amounts or concentrations. A number of geophysical and geochemical tools theoretically fulfil this criterion, but in practice robust quantitative assessment of in situ CO2 is extremely challenging, depending on many factors including lack of full volumetric coverage (sampling deficiency), insufficient resolution and nonunique relationship between CO2 quantity and tool response. Sometimes uncertainty can be reduced by deploying tools in combination.



207

8.2.2 Monitoring tools A comprehensive portfolio of techniques is available for utilisation in storage site monitoring (Fig. 8.1) with over 30 separate tools listed in Chadwick et al. (2009a). Broadly speaking, these can be categorised as deep-focused, for reservoir measurements and tracking of CO2 in the subsurface, and shallow-focused tools for detection and measurement of CO2 leakage, at or close to the surface. The deep-focused tools mainly correspond to mature oil industry technologies, but are relatively untested for CO2 monitoring, whereas the shallow monitoring methodologies are commonly novel and under development. Monitoring tools have been deployed at a number of CO2 injection projects around the world in various combinations (Table 8.1), including industrial-scale sites at Sleipner and Snøhvit (offshore Norway), Weyburn (Canada) and In Salah (Algeria) and also at pilot-scale sites at Nagaoka (Japan), Frio (USA), Ketzin (Germany), K12-B (offshore Netherlands) and Otway (Australia). Tools deployed at the industrial-scale sites are aimed at demonstrating site performance within constraints imposed by cost and logistics. Tool deployment at the pilot-scale sites is more research-oriented, focused on tool testing and calibration, and delineating detailed storage processes within the reservoir. A number of publications (e.g. Benson et al., 2004; Arts and Winthaegen, 2005; Pearce et al., 2005; Chadwick et al., 2009a) describe the characteristics and applicability of a wide range of monitoring tools and suggest ways in which they may be used at storage sites. The International Energy Authority Greenhouse Gas R&D Programme website hosts an interactive tool for the design of CO2 monitoring programmes (IEAGHG 2007), taking into account technical requirements plus a number of other site-specific factors including surface conditions (onshore/offshore, rural, urban, flat, mountainous, etc.) and site geology (reservoir depth, type, etc.). This chapter takes a different approach; the aim here is to show how a limited number of key tools can be used in combination for pragmatic but effective storage site monitoring.

8.3

Detection and measurement of carbon dioxide (CO2) in the subsurface

The current preferred approach to industrial-scale storage site monitoring is to deploy deep-focused monitoring to demonstrate understanding of the storage system in the deep subsurface by confirming predicted storage performance and establishing site integrity. The aim is to detect any deviations from predicted performance as early as possible, such that remediation actions can


208

Subsurface monitoring

Seismic methods

Surface/near-surface monitoring

Non-seismic and borehole methods


2D/3D

Wellbore Micro-

surface

seismic

seismic

Sleipner

Industrial

Weyburn

Industrial

In Salah

Industrial

Snøhvit

Industrial

Cranfield

Industrial

Rangeley

Industrial

K12-B

Industrial

Otway

Research

Ketzin

Research

Nagaoka

Research

Frio

Research

West Pearl Queen

Research

EM/

Gravity Tilt-

seismicity electrical

Pressure/

Downhole Well

meters temperature logs

Seabed

Soil Surface flux/ Eco-

fluids & imaging gas tracers

Remote

atmospheric systems sensing


Table 8.1 Monitoring tools deployed and planned at selected CO2 injection sites


209

be taken before any leakage threat becomes significant. Where practicable, it may also be desirable to quantify in situ amounts of CO2 in the storage reservoir, in order to better understand key stabilisation processes such as capillary trapping, dissolution and, ultimately, the long-term performance of the storage system. Assurance from deep-focused monitoring translates into a number of specific technical objectives: ∑

direct imaging (and, if possible, quantification) of CO2 in the storage reservoir; ∑ measurement of pressure changes in and around the reservoir; ∑ detection of migration of CO2 from the primary reservoir; ∑ detection of migration of CO2 through the overburden to shallower depths. Deep-focused methods split naturally into two: non-invasive (surface-based) and invasive (well-based).

8.3.1 Non-invasive monitoring Non-invasive monitoring systems are generally remotely positioned with respect to the storage reservoir, so their absolute resolution or CO2 detection capability is intrinsically limited. On the other hand, by suitable configuration of survey sensors, these tools have the potential to provide broad coverage and, in some circumstances, continuous 3D coverage of the whole storage reservoir and its overburden (see below). Key tool: surface 3D seismic Under favourable conditions 3D seismic can provide continuous and uniform (in the sense of source–receiver offsets and azimuths) volumetric imaging of the storage system at high resolution (Fig. 8.2). This provides a powerful means of imaging the CO2 plume in the reservoir, and detecting migration of CO2 from the storage reservoir into surrounding strata. As well as testing whether the plume is threatening containment risks, effective plume imaging also provides key data on reservoir performance, tests short-term reservoir flow simulations and helps calibrate longer-term predictive models. Time-lapse seismic does offer potential for useful quantification, but its efficacy is highly site-dependent. In general, storage in shallow, poorlyconsolidated reservoirs is more amenable to seismic quantification than in deeper more lithified rocks (McKenna et al., 2003). The CO2 storage operation at Sleipner in the Norwegian North Sea (Baklid et al., 1996) provides an excellent example of surface seismic monitoring under favourable conditions. Injection commenced at Sleipner in 1996, CO2



100 m

210

500

m

8.2 3D seismic cube showing continuous volumetric imaging of fluid distribution in the subsurface (strong reflections on the front corner).

separated from natural gas being injected into the Utsira Sand, a major saline aquifer of late Cenozoic age (Zweigel et al., 2004). CO2 injection is via a near-horizontal well, at a depth of about 1012 m bsl, some 200 m below the reservoir top, at a rate approaching 1 million tonnes (Mt) per year, with around 12 Mt currently stored. A comprehensive time-lapse surface seismic monitoring programme has been carried out, with 3D surveys in 1994, 1999, 2001, 2002, 2004, 2006 and 2008. Some key findings are outlined here with fuller details in Arts et al. (2004a,b, 2009) and Chadwick et al. (2005a, 2009b). The Sleipner CO2 plume is imaged as a number of bright sub-horizontal reflections within the reservoir, growing with time (Fig. 8.3a). The reflections are interpreted as tuned wavelets arising from thin (mostly < 8 m thick) layers of CO2 trapped beneath thin intra-reservoir mudstones and the reservoir caprock. The plume is roughly 200 m high and elliptical in plan, with a major axis increasing to over 3000 m by 2006, accompanied by development of a prominent northerly extension since 2004 (Fig. 8.3b). A key aim is to track lateral migration of the free CO2 in the reservoir as this determines the likelihood of the plume impacting on containment risks, such as well penetrations. At Sleipner, the topmost layer of the plume (arrowed in Fig. 8.3) is particularly clearly imaged (Chadwick et al., 2009b). Detailed mapping shows it to be spreading quickly northwards, the CO2 front


Verification of CO2 storage in underground reservoirs 1 km

1994

a

north

2001

2004

211 2006

b

1 km

8.3 Time-lapse seismic images of the Sleipner CO2 plume for selected vintages showing its development to 2006. (a) N–S seismic section through the plume. (b) Maps of integrated plume reflectivity.

locally advancing at 1 m/day since 2001, but currently with no short-term threat to existing well infrastructure. As well as its prominent reflectivity, the plume also produces a large velocity pushdown (Fig. 8.4) caused by the seismic waves travelling more slowly through CO2-saturated rock than through the virgin aquifer. These two elements of the seismic signature, when taken together, offer the potential for quantitative analysis: reflectivity, which gives the distribution of acoustic impedance contrasts (the main fluid interfaces), and velocity pushdown, which provides a measure of the total amount of CO2 in a vertical column. A quantitative interpretive approach (combining rock physics with laboratory data on CO2 flow behaviour), allows a CO2 saturation model of the plume to be derived. Chadwick et al. (2004, 2005a), proposed a saturation model for the 1999 dataset which contained around 85 % of the known injected CO2 whilst maintaining a satisfactory match with the seismic data (Fig. 8.5). Given that reservoir flow simulations suggest up to 10 % of the free CO2 would have dissolved into the aqueous phase (thereby becoming seismically invisible to the seismic), this may be considered a remarkably accurate result. Subsequent to this work, questions were raised as to the accuracy of the assumed reservoir temperature at Sleipner which was based on a single downhole measurement. Alternative seismic modelling, with higher reservoir temperatures and modified CO2 properties, showed that by applying different rock physics parameters (in particular by considering the seismic response to patchy mixing of CO2 and water in the plume) it is possible to reproduce


212

0


Pushdown [ms]

1999

50

Base Utsira Sand a

500 m

Base Utsira Sand b

500 m c

8.4 Time-lapse seismic images of the Sleipner CO2 plume showing the development of velocity pushdown. (a) N–S seismic section through reservoir in 1994 showing relatively flat Base Utsira reflection. (b) Same section in 1999 showing pushdown of Base Utsira reflection beneath the plume. (c) Map of velocity pushdown in 1999.


1999

1994

Two-way time

Depth

Two-way time


a

E

W

b

c

8.5 Seismic quantification of the 1999 dataset. (a) E–W seismic section through the 1999 plume. (b) Same section extracted from 3D CO2 saturation model. (c) Synthetic seismogram generated from the CO2 saturations.


E

W

E

W

213

214


the observed seismic response for a range of CO2 distributions and properties (Chadwick et al., 2005b). This highlights a key generic uncertainty in seismic verification; the velocity behaviour of the CO2 – water – rock system, which is heavily dependent on the nature of small-scale mixing processes between the fluid phases. More recently, large-scale water production from the Utsira Sand at the Volve Field, just a few kilometres from Sleipner (O Eiken personal communication), has yielded temperature data which support the original modelling by Chadwick et al. (2005a). A clear lesson here is the need for accurate pressure and temperature data to properly constrain CO2 properties (see below). This is particularly the case for shallow reservoirs such as at Sleipner where conditions are close to the critical point of CO2. Alternative approaches have been used to obtain quantitative information from the data, including pre- and post-stack trace inversion (summarised in Chadwick et al., 2008), with more recent model-based inversion described in Delepine et al. (2009). The inversion approaches all suffer from the difficulty in accounting for tuning effects from very thin layers of CO2. Ongoing work on the Sleipner datasets is using spectral decomposition to further constrain layer thicknesses and velocity, and is also looking in more detail at amplitude – offset changes to extract elastic parameters as well as acoustic information from the plume reflectivity. It appears that the more recent Sleipner datasets are becoming more difficult to model. With time, reflectivity in the deeper plume is fading and velocity pushdown is becoming more difficult to map (Fig. 8.3). These may be seismic imaging effects arising from generally increasing CO2 saturations within the plume envelope, or may signify real and significant changes in CO2 distribution in the deeper part of the plume. A monitoring well through the Sleipner plume would radically reduce uncertainty in all quantitative methods although, since baseline (pre-injection) measurements are no longer possible, the ultimate efficacy of such a well cannot now be realised. In addition to providing detailed information on CO2 within the storage reservoir, the seismic data also indicate that no detectable migration of CO2 into the caprock has so far occurred. The potential detection capability of the Sleipner data can be illustrated by examining differences in time-lapse data between the 1994 baseline survey and the first repeat in 1999 when two small lenses of CO2 had just started to accumulate beneath the caprock seal (Fig. 8.6a). From the reflection amplitudes, the volumes of the two accumulations can be estimated at about 9000 and 11 500 m3, respectively. Other seismic features on the difference map can be attributed to repeatability noise, arising from small intrinsic mismatches between the 1999 and 1994 (baseline) surveys. It is clear that repeatability plays a key role in determining the detectability threshold. Thus, for a patch of CO2 to be identified on the data it must be distinguishable from the largest noise peaks. Preliminary analysis of difference


1000

(a) Repeatability noise peak

(b)


Northern patch (~14 600 m3)

0 50 (c) Noise + CO2

Southern patch (~11 500 m3)

Number

(d)

0 50

Noise CO2

500 m Amplitude

215

0


Noise + CO2

8.6 Detection limits for small amounts of CO2 at Sleipner. (a) Time-slice map of the 1999–94 difference data showing reflection amplitude changes at the top Utsira Sand. High amplitudes (paler greys) correspond to two small CO2 accumulations. Other scattered amplitudes are due to repeatability noise. (b)–(d) Histograms plotting number of seismic traces against reflection amplitude.

216


signal from CO2 compared with repeatability noise (Fig. 8.6b–d) suggests that accumulations larger than about 4000 m3 should fulfil this criterion. This corresponds to about 2400 t CO2 at the top of the reservoir where CO2 has a likely density of about 600 kgm–3, but less than 600 t at 500 m depth, where its density would be considerably lower. Seismic detection depends crucially on the nature of the CO2 accumulation. Small thick accumulations in porous strata would tend to be readily detectable either by their intrinsic reflectivity or via development of a localised velocity pushdown. Conversely, distributed leakage fluxes through low-permeability strata may be difficult to detect with conventional seismic techniques. Similarly, leakage along a fault within low-permeability rocks would be difficult to detect – although it could be argued that such a fault would be unlikely to constitute a flow conduit in any case. To summarise, surface seismic is a key tool for imaging changes across the whole storage system. Favourable reservoir properties and its offshore location render Sleipner ideal for surface seismic monitoring, but the method has also proved successful for plume imaging at Weyburn (Table 8.1), an onshore site where the storage reservoir is deeper, thinner and more complex (Wilson and Monea, 2004). Surface seismic monitoring is a key element of the monitoring programmes at In Salah (Mathieson et al., 2009) and at Snøhvit, both sites utilising thin, deep storage reservoirs. It is stressed that even where reservoir conditions are less than ideal for plume imaging, surface seismic can provide uniform and continuous coverage of the overburden, from the reservoir to shallow depths. Ancillary tools Various non-invasive monitoring tools can be deployed to provide ancillary information to surface seismic. Multicomponent (MC) seismic is a specialised refinement of conventional seismic, whereby both compressional (P-wave) and shear (S-wave) components of ground motion are recorded. The latter are more sensitive than the former to structural fabric, but much less sensitive to fluid changes. By analysing combined P- and S-wave signals, it is possible to obtain a more complete picture of fluid behaviour, including improved discrimination of fluid pressure and saturation changes and better imaging beneath gas accumulations. In particular, geomechanical changes may be observable in low-permeability overburden sequences where the lack of discrete CO2 accumulations may render conventional seismic ineffective. Notable examples of the successful deployment of MC seismic include the CO2–enhanced oil recovery (EOR) operation at the Vacuum Field in Texas (Angerer et al., 2003) and, more recently, at Weyburn (Wilson and Monea, 2004). MC seismic is, however, considerably more expensive than conventional seismic, and the technique would most likely be deployed in



217

certain circumstances such as where there are geomechanical issues. It is likely also that MC seismic would be deployed as a subset of conventional seismic, focusing on identified system vulnerabilities rather than attempting full volumetric coverage of reservoir and overburden. Potential field methods can integrate subsurface information from the whole storage system but are not imaging systems senso strictu and provide radically lower sensitivity and resolution than the seismic methods. Gravimetry measures the gravitational acceleration due to mass distributions within the earth, so in time-lapse mode it can be used to detect and measure changes in fluid distribution. An example of gravimetry offering complementary information to seismic monitoring is illustrated by ongoing studies at Sleipner. A seabed gravity survey was acquired at Sleipner in 2002 (Nooner et al., 2007), with 4.97 Mt of CO2 injected, and a repeat survey in 2005 with 7.75 Mt of CO2 injected (an additional 2.78 Mt). The surveys were based around pre-positioned concrete benchmarks on the seafloor deployed in two perpendicular lines, overlapping the subsurface footprint of the CO 2 plume (Figure 8.7a). Relative gravity and water pressure readings were taken at each benchmark by a customised gravimetry and pressure measurement module mounted on a remotely operated vehicle. Each benchmark was visited at least three times to better constrain instrument drift and other errors, resulting in a single station repeatability of about 4 mGal. For time-lapse measurements an additional uncertainty of 1–2 mGal is associated with the reference null level. The time-lapse gravimetric response due to additional CO2 was obtained by removing the modelled gravimetric time-lapse response from the Sleipner East field (the deeper gas reservoir currently in production) from the measured gravity changes between 2002 and 2005. So far, gravity modelling has focused on constraining the in situ density of CO2, which, given the former uncertainty in reservoir temperature, was thought to be a significant uncertainty in the quantitative seismic analysis. Initial modelling (Nooner et al., 2007) concluded that the average CO2 density in the plume was around 530 kgm–3. More recent modelling, based on an optimisation technique (Fig. 8.7b), and with improved application of the various data corrections (Alnes et al., 2008), has derived a revised CO2 density value of 760 kgm–3. The measurement accuracy of the seabed gravimetric method is not much smaller than the likely time-lapse gravity changes that occurred between 2002 and 2005, so the method is close to its usable limit. However, a second repeat gravity survey was acquired in the summer of 2009, with around 11 Mt of injected CO2. The time-lapse change from 2002 will therefore be more than double that seen in 2005. This should allow much more robust extraction of properties from the modelling. The published studies have assumed, for simplicity, that all of the known injected CO2 is present as a free phase, with


218


Footprint of CO2 plume in 2001 IP

500 m

Gravity signal (microGal)

5

0

–5

–10

–15 Other stations

Long profile stations (a)

–20 0

5

10

15 Benchmark (b)

20

25

30

8.7 (a) Map of the gravity survey layout at Sleipner showing location of seabottom benchmarks. (b) Modelling results – dashed lines indicate modelled gravity with CO2 densities from 500–800 kgm–3. Solid line shows best fit for a CO2 density of 760 kg m–3 (adapted from Alnes et al., 2008).


10


219

none in solution. This is a significant assumption, because when CO2 dissolves in the reservoir water it loses much of its gravimetric signature. Given that reservoir temperatures are now seemingly well-constrained (see above), with a likely CO2 density of around 700 kgm–3, it may be that modelling would be better directed at estimating the amount of CO2 dissolved, a significant uncertainty in predictive flow simulations. It is clear from the work at Sleipner that gravimetry has the potential to provide valuable independent information capable of reducing uncertainty in the seismic analysis. The use of complementary methodologies in this way can be very effective in an integrated monitoring programme.

8.3.2 Invasive methods Invasive methods, essentially synonymous with well-based monitoring, offer detailed measurement capability of key storage reservoir properties, of higher resolution than is possible with the non-invasive methods. A key benefit is the ability to provide detailed information capable of calibrating and reducing uncertainty in surface-based methods. However, with the exception of pressure (see below), coverage is more-or-less 1D, being generally restricted to the volume immediately surrounding the wellbore. Monitoring tools can be deployed in the injection well(s) and also in observation wells, using either pre-existing or purpose-drilled wellbores. Multiple-well monitoring well strategies are, however, likely to be expensive, and introduce additional potential containment risks. Key tool: downhole pressure and temperature measurement Unlike most downhole monitoring tools whose sensitivity is restricted to the vicinity of the wellbore, pressure responds to changes across a wide area, providing a measure of the integrated response of the storage system to injection. Temperature monitoring is very sensitive to fluid flow and can enable detection of CO2 migration around the wellbore. In combination, the two methods constrain in situ CO2 properties, providing key data for calibrating reservoir flow simulations and also quantitative seismic analysis. Pressure monitoring provides two key types of information. Firstly, it can be used to ensure that reservoir pressures are not rising sufficiently to induce geomechanical instability of the reservoir and caprock. Secondly, it can be used to calibrate flow simulations of reservoir performance in terms of pressure evolution. An excellent recent example of downhole pressure monitoring is afforded by the Cranfield CO2 injection operation at the Cranfield oilfield in SW Mississippi, USA (described in detail by Meckel et al., 2008). Oil production was suspended at Cranfield in 1965, and since then reservoir pressures have


220


recovered to near hydrostatic. In 2008, CO2 injection commenced as part of an EOR project. In the first few months, prior to oil production, pressure evolution was comparable to what would be expected from a CO2 aquifer storage operation. Injection is via a number of injection wells (Fig. 8.8) together injecting around 2900 t CO2 per day. Total injection by late 2008 was in the order of 200 000 t, making the operation intermediate; between pilot- and full-scale storage. A number of old wells penetrate the field. One of these, in the midst of the injection wells, was re-entered and instrumented as a monitoring well (Fig. 8.8). Continuous pressure and temperature sensors were placed at casing perforations in the reservoir sand and also some 110 m higher in an overburden sand above the reservoir topseal (full details given in Meckel et al., 2008). The pressure monitoring clearly tracks pressure evolution both within the injection reservoir ‘injection zone’ and in the overlying overburden sand ‘monitor zone’ (Fig. 8.9). From the start of injection in mid-July, a steady increase in reservoir pressure indicates pressure communication in the reservoir between the injection well and the monitoring well. This primary pressure buildup is the key measurement requirement for ensuring the geomechanical integrity of the reservoir and overburden (as a rule of thumb, fracture pressures are typically around 80 % of the overburden pressures). It is also diagnostic of the flow properties and storage capacity of the injection reservoir, and provides valuable calibration data for predictive modelling of reservoir performance. In contrast, pressures in the monitoring zone Monitoring well CO2 injection well Other well

200 m

8.8 Map of well locations at Cranfield (adapted from Meckel et al., 2008).



221

Incremental pressure change 36

Pressure (MPa)

Field-wide increase in injection rates

Injector closest to monitoring well starts injection

Injection zone pressure

34

Shut-in for hurricane Gustav

32

Injection begins

July

Monitor zone pressure

August

September

October

8.9 Cranfield pressure monitoring. Measurements from the reservoir (injection-zone) and from the overburden sand (monitor zone) (adapted from Meckel et al., 2008).

do not show a corresponding increase, indicating that there is no pressure communication through the reservoir topseal and no migration of CO2 out of the reservoir. More transient effects can also be discerned. Injection shut-in for Hurricane Gustav was marked by a pressure decrease, with more subtle changes resulting from a field-wide increase in injection rates in late September and startup of a nearby injection well in October. Subtle effects are more clearly detected by measuring the rate of pressure change (Fig. 8.9 top), which is particularly sensitive to transient changes. It is clear that pressure measurements respond to changes in CO2 input. By implication they would also respond to migration of CO2 out of the reservoir, particularly if this involves transient or time-variant effects. Monitoring in the overburden provides additional valuable assurance on reservoir integrity, perhaps allowing discrimination between along wellbore migration (rapid changes) and migration through the geological seals (slower changes). It should be noted, though, that pressure changes in the overburden would not necessarily indicate CO2 migration out of the reservoir; pressure


222


changes could be transmitted through the topseal via formation water even with the topseal acting as a capillary seal to CO2. Temperature monitoring can also help detect and characterise flow communication in and around the reservoir. A good example of comprehensive injection monitoring is provided by the Ketzin project (Prevedel, et al. 2009), where downhole pressure and temperature are augmented by continuous temperature profiling down the well. This enables very tight control on the thermodynamic properties of the injected CO2, essential for accurate flow simulations. A drawback of pressure monitoring is that it does not provide directional information, so the location of untoward pressure changes is not revealed. Multiple monitoring wells would provide a degree of spatial control on plume extent but would likely be prohibitively expensive. There are also difficulties in detecting ‘steady-state’ migration out of the reservoir; slow leakage, perhaps through the geological seal, would be difficult to distinguish from uncertainty in reservoir flow simulations. Overburden pressure monitoring is, therefore, important, but again would not necessarily be diagnostic of CO2 migration into the overburden as pressure is transmitted through the water phase. Ancillary methods Well-based monitoring tools generally provide high-resolution information, albeit from a volume restricted, to a variable extent, to the vicinity of the wellbore. This type of information is important in two respects: to provide information on wellbore integrity and to provide data for testing and calibrating predictive models of storage performance. Well-based monitoring of oil and gas reservoirs includes a broad array of techniques, using a diverse suite of instruments. During drilling well cuttings, core and fluid samples can be recovered to obtain petrophysical measurements and provide information on fluid properties. These essentially point measurements are complemented by the continuous downhole information provided by wireline logging. The latter provides the highest resolution information but with a coverage range restricted to a radius of a few metres or less around the wellbore. A good summary of recent developments of well logging in CO2 injection monitoring is given by Freifeld et al. (2009). The principal use of well logs is to provide assurance on wellbore integrity and to provide detailed measurement of CO2 saturations around the well: key for calibrating predictive flow simulations. The Frio injection pilot provides a good example of the use of the novel RST (Reservoir Saturation Tool) to test and verify saturations from predictive numerical flow simulations (Fig. 8.10). Wellbore geophysical methods interrogate a volume of the near-borehole



Depth [ft] 1

Day 4

0 1

Day 10

0 1

Day 29

0 1

Day 66

0 1

223 Day 142

0

4995.0 5000

Topseal 5010

Frio ‘C’ sandstone

5020

5030

5040

5050 5055.0

CO2 saturation calculated from RST CO2 saturation calculated by model

Perforations

8.10 Downhole results from Frio. CO2 saturations from RST logging compared with calculated saturations predicted from flow simulations (adapted from Freifeld et al., 2009).

environment and thereby provide spatial information on reservoir processes. Seismic techniques such as vertical seismic profiling (VSP) and cross-hole seismic tomography give direct measurement of velocity and signal attenuation (both key indicators of fluid saturation) and provide finer-scale information complementary to the surface methods. VSPs image specific detail around the wellbore; useful, for example, in the early detection of CO2 migration outside the casing. Cross-hole seismic requires at least two wells through or close to the storage reservoir. Changes in travel-time and signal amplitude between the wells can be used to map velocity and attenuation changes in the cross-section between the wells that can be related to changes in CO2 saturation and/or pressure. Recent practical experience from the Nagaoka CO2 injection experiment (Saito et al., 2006) indicates that amounts of CO2 as small as hundreds of tonnes can be detectable using the cross-hole method. A good example of the use of wellbore geophysics to test predictive reservoir flow simulation in 2D is afforded by the Frio project. Modelled saturation distributions (Fig. 8.11a) can be tested by cross-hole seismic


224

Distance [m] 10 20

30

Observation well

4950

tion erva l l e w

0

d ’ san of ‘C Top 5000

Change due to CO2

sand f ‘C’ Top o

5050

ft 75

Sigma 107

(a)

SCO2

0

0.1

ft

0.2

0.3

Depth (feet)


tion Injec ll e w

Sigma

(b)

Change in velocity (km/s) –1.0

–0.5

0

0.5

8.11 CO2 distributions between the injection and observation wells at Frio: (a) as predicted by reservoir flow simulation; (b) as imaged by crosshole seismic tomography, calibrated by RST well logs (diagrams courtesy of Tom Daley, Christine Doughty and Susan Hovorka).


Obs

Injection well


225

tomography which provides spatial information on seismic velocity changes (Fig. 8.11b). These can, in turn, be compared directly with downhole CO2 saturations measured from RST logging to further reduce uncertainty. A recent development of crosshole seismics deployed at Frio involves a continuous active-source technique (Daley et al., 2007). This measures changes in seismic travel-time between the wellbores and provides realtime monitoring of the evolving CO2 plume with high spatial and temporal resolution. A shortcoming of many wellbore monitoring tools is that they are sensitive mainly to CO2 in the free (gaseous or supercritical) phase, but not to CO2 in solution. Dissolution of CO2 is one of the key medium-to-long-term trapping mechanisms, but is relatively poorly-constrained. At Nagaoka, recent timelapse resistivity logging and downhole fluid sampling have indicated that a layer of dissolved CO2 is progressively forming beneath layers of free CO2 (S. Mito personal communication). This type of detailed information from a pilot-scale test site has the potential to quantify the amount of CO2 going into aqueous solution. Further, it can provide crucial support to the predictive models of site performance that are a regulatory requirement for large-scale storage.

8.3.3 Summary of deep-focused methods Surface seismic and downhole pressure monitoring probably comprise the key deep-focused monitoring technologies. Together, they can provide information on processes from across much of the reservoir and overburden. Ancillary downhole technologies provide radically-improved resolution, but with very limited spatial coverage. On the other hand, they can provide important complementary data capable of reducing uncertainty and supporting predictive modelling.

8.4

Detection and measurement of carbon dioxide (CO2) leakage to surface

Monitoring for CO2 leakage involves the detection or measurement of CO2 in the shallow subsurface, such as in shallow aquifers or in the soil or seabed. In addition, CO2 fluxes may be measured in the air or offshore in the water column. In a well-designed storage operation, leakage, particularly through the geological seals, should not be an issue in the short-term. The main priority of leakage monitoring, therefore, is not so much on actual detection, but on establishing secure baseline conditions against which any future changes can be readily identified. A second priority is to focus on identified containment risks, notably old wells, to test storage integrity.


226


Possible future utilisation of shallow monitoring may be to set quantitative limits on CO2 leakage in order to demonstrate site performance in terms of emissions reduction (see above) and possibly to earn emissions credits as part of a carbon-trading system. If leakage were to occur, shallow or atmospheric monitoring would be deployed to assure identification and remediation of accumulations of CO2. Effective leakage monitoring implies that CO2 leakage over the whole storage footprint can be accurately quantified. An optimal leakage monitoring system, therefore, would have the ability to obtain robust measurements of leakage flux over wide areas whilst retaining high localised detection sensitivity. This requires the deployment of complementary methods in combination. Offshore leakage monitoring is technically and logistically challenging, notably in the acquisition of robust areal monitoring data. This is principally down to difficulties in gaining access to a wide area of seabed with the necessary monitoring tools and in the accurate measurement of CO2 in the water column. The following section will therefore focus on onshore leakage monitoring where a wide range of established techniques has been deployed or is under development. These range from soil and atmospheric measurement tools to airborne remote sensing methods.

8.4.1 Pointwise measurements Pointwise sampling methods, deployed on a grid (regular or otherwise), can provide high detection sensitivity and measurement accuracy, but with accuracy dependent on the sensor being coincident with the leakage (Fig. 8.12). A more likely scenario is that the leak may be partially or wholly displaced from the sensors, meaning that measurements will be incompletely measured or missed altogether. This effect can be to some extent ameliorated by clustering sensors around the main containment risks, but the chance of missing leakage from unexpected migration pathways remains. Key tool: soil gas measurements Because ambient levels of CO2 in soils are much higher than in the air [Welles et al. (2001) quote typical soil gas CO2 concentrations of 2000– 10 000 ppm], soil gas monitoring is considered to provide a robust option for leakage detection and measurement. The equipment needed for soil gas surveying ranges from stationary accumulation chambers to small portable sampling and analysis equipment. In the latter case, sensors are deployed in a grid configuration over the expected leakage ‘footprint’, in or on the soil, with samples analysed periodically to determine soil gas composition and fluxes. A key issue in soil gas surveying is to establish accurate baseline conditions by identifying and removing the effects of naturally occurring



Small leak

227

Leak

Sensor Leak

8.12 Schematic view of a leakage monitoring system deploying a grid of point sensors. Very small leakages can be detected, but only if they are co-incident with the sensor. Conversely, larger leakages may be missed.

CO2 and seasonal variations. A clear requirement, therefore, is to have a robust understanding of climate and seasonal changes in soil use and processes for the site. Issues associated with soil gas monitoring are exemplified by the surveys carried out at Weyburn in Saskatchewan, Canada (Riding and Rochelle, 2009). This is principally an EOR project, but with the secondary aim of ultimately storing 20 Mt of anthropogenic CO2 (Wilson and Monea, 2004). Injection started in late 2000, using CO2 captured from a coal gasification plant in North Dakota. CO2 is injected at rates between 1 and 2 Mt per year into a thin, carbonate reservoir about 1500 m deep. Weyburn differs from Sleipner in having a large number of wells, both active and abandoned, which penetrate the storage reservoir. Baseline surveys were acquired in 2001 to evaluate natural variation in soil gas concentration and to identify sites of higher gas flux that may be indicative of deep gas escape (Strutt et al., 2003). Measurements included gas concentrations in the shallow unsaturated soil horizon (soil gas); mass transfer rates of CO2 across the soil–atmosphere interface (gas flux); and long-term monitoring of radon flow rates as a proxy for CO2 using buried probes. Measurements were acquired on a 360 point grid at 200 m spacing (Fig. 8.13). Repeat surveys in the autumns of 2002 and 2003 traversed anomalies seen on the earlier survey. Selected CO2 and radon anomalies on these profiles were investigated in more detail for signs of natural pathways for deep gas escape using helium, methane and thoron as proxies for potential future


228


1 km July - 2001

September - 2002

(a)

October - 2003

gm–2d–1 440 400 360 320 280 240 200 160 120 80 40 0

%

July - 2001

51 48 45 42 39 36 33 30 27 24 21 18 15 12 9 6 3 0 15.8 14.6 13.4 12.2 11 9.8 8.6 7.4 6.2 5 3.8 2.6 1.4 0.2 –1

September - 2002

(b)

October - 2003

11 10 9 8 7 6 5 4 3 2 1 0 6 5.5 5 4.5 4 3.5 3 2.5 2 1.5 1 0.5 0 2.2 2 1.8 1.6 1.4 1.2 1 0.8 0.6 0.4 0.2 0

8.13 Results from the Weyburn soil gas surveys showing natural and seasonal variations. (a) CO2 flux. (b) CO2 concentrations. Black and grey lines denote major and minor roads, respectively (diagrams courtesy of Dave Jones and Chris Rochelle, British Geological Survey).

CO2 escape. Continuous radon monitoring probes were installed at sites where helium and radon data indicated the potential for deep gas migration, operating virtually continuously from 2001 through to 2004.



229

Marked changes were seen in CO2 concentration and surface flux levels between each of the three datasets (Riding and Rochelle, 2009). Higher values marked the growing season of July 2001, with lower levels in autumn 2002 and further reduction in autumn 2003, when conditions were cooler and the growing season almost over. Some of the CO2 anomalies, based on initial air-photo interpretation, may represent the surface expression of deep faults, but soil gas data indicated that the elevated values in these areas are more likely due to shallow biological reactions in the moist, organic-rich soil. The distribution of radon and thoron anomalies also lacked any clear linear trends that might indicate the presence of a gas-permeable fault or fracture system. Isotopic values of three soil gas samples collected in summer 2001 all indicated that the soil gas CO2 was produced by microbial or root metabolism of organic matter from local plants, illustrating the importance of shallow biological reactions that produce CO2 as a metabolic by-product. Borehole integrity was investigated by measuring soil gas around two decommissioned oil wells, one abandoned and the other suspended due to failed casing. The latter had weakly anomalous CO2 at two sites, but this was not the case for other gases. The abandoned well had normal background CO2 values. Statistical populations of CO2 and radon were generally higher for the suspended well, while those for methane were higher for the abandoned well compared to background values, although all individual values lay well within the range observed across the site. There was one helium anomaly at the abandoned well site, but the lack of correspondence between anomalies of different gases suggests that current leakage from depth in the well is insignificant. Electronic radon sensors were installed up to 2 m deep at six sites on radon and CO2 anomalies selected from the detailed soil gas profiles. Hourly measurements of radon concentration, temperature and atmospheric pressure showed seasonal variations in radon concentration which were modelled against atmospheric parameters, indicating the importance of pressure, rainfall and temperature on gas migration. In addition, CO2 fluxes deeper in the soil were measured. Fluxes at 2 m depth were calculated to be 10–20 times lower than those at surface. This is consistent with declining biogenic CO2 production with depth, and suggests it may be better to monitor flux at this depth where biogenic influences are muted. Other published examples of quantitative surface monitoring of CO2 injection include the operations at Rangeley and West Pearl Queen in the USA. At Rangeley in Colorado, some 25 Mt of CO2 have been stored over more than 20 years as a consequence of EOR. Surface fluxes of deep-sourced CO2 are less than about 170 t year–1, representing an annual leakage rate of < 0.01 %. This fulfils the nominal emissions performance criterion (see above), especially as it is probable that some, perhaps all, of this CO2 is


230


microbially-oxidised methane rather than injected CO2 leaking from the reservoir (Klusman, 2003). At West Pearl Queen in New Mexico, nearly 2100 t CO2 were injected in 2002–03, as a pilot research study (Wells et al., 2007). A soil gas sampling system was installed around the injection well (an old production well). In an effort to enhance detection sensitivity, the system was designed around the detection of PFT (perfluorocarbon) tracers, and comprised sensors mounted in steel tube penetrometers, hammered a few feet into the ground and arranged in a radial pattern around the injection well giving a circular monitoring area of 300 m radius. Baseline conditions were established, before slugs of three different PFT tracers were periodically added to the injected CO2 stream in a known dilution ratio (tracer/CO2). Evidence of leakage associated with the injection well was noted soon after tracer injection. This manifested as tracer hotspots around the injection well with preferred directional migration components. Measurement of the detected tracer amounts was used to quantify CO2 leakage. Integrated measurement for all the tracers indicates a leakage rate of 0.0085 % per year. These calculations are preliminary and include simplifying assumptions such as that of a constant dilution factor; they are thought, therefore, to represent an upper bound for the rate. It is not known how the CO2 and tracers leaked, but the injection wellbore is the likeliest candidate. Injection was accompanied by significant pressure increase (8 MPa), and it may be that this triggered minor migration from the reservoir and up the outside of the well casing (drilling-induced fracturing around the wellbore is ubiquitous in the area). It is clear from the two latter studies that soil gas sampling, particularly with its sensitivity enhanced by tracers, can potentially provide detection capability of the order required to test storage site containment efficacy, provided sampling issues can be resolved.

8.4.2 Areal measurements Areal monitoring tools, though of much lower sensitivity than the pointwise monitoring methods, have a potentially complementary function in providing more-or-less continuous coverage of the storage footprint. In combination, the two approaches have the potential to provide a robust surface monitoring system. Key tool: IR atmospheric monitoring Most atmospheric measurement techniques rely on the preferential absorption by CO2 of infrared (IR) radiation. There are two basic approaches: nondispersive IR gas analysers use a broad-spectrum source in a small closed chamber containing the sample to be analysed; IR diode lasers in ‘open-path’



231

mode, where the free atmosphere can be analysed over short (< 2 m) or extended path lengths, up to hundreds of metres, averaging the concentrations over these distances. Tools fall into two categories: mobile devices capable of measuring fluxes rapidly over large areas and so-called ‘areal integrators’ whereby a limited number of stationary instruments are deployed in such a manner as to detect and measure all CO2 emissions from within a target ‘footprint’. The latter, in particular, depend on the fact that CO2 is denser than air and tends to dissipate laterally rather than vertically upwards. All of these tools are in a developmental stage, particularly with regards to designing optimal deployment configurations. Mobile open-path laser measurements combined with ground-surface portable infrared measurements show great promise for rapid assessment of large areas (Jones et al., 2009). Mobile laser equipment mounted on a quad bike has been tested around the Laacher See – a flooded caldera in the East Eifel volcanic area of Germany, with a number of naturally-occurring surface CO2 emissions around the lakeshore. An area covering some 100 000 m2 was surveyed a number of times in 2007 and 2008 (Jones et al., 2009). There was good agreement between the repeat surveys both in terms of the concentrations of CO2 and the patterns observed (Fig. 8.14). Two main gas anomalies were seen corresponding to known vents, in addition to areas of weaker gas flux at lower concentrations but still above background. The eddy covariance (or eddy correlation) micrometeorological method (Miles et al., 2004) can perform a key areal integration role in leakage monitoring. It essentially consists of a tower-mounted infrared gas analyser alongside a sensitive sonic anemometer to measure wind speed and direction. The aim is to detect CO2 from an area (‘footprint’) upwind. The size and the shape of the footprint is derived mathematically from the wind speed and direction; by measuring over a protracted period, full azimuthal coverage can be obtained. By combining CO2 concentration data with meteorological information, eddy covariance can produce CO2 flux data, expressed as the amount of CO2 released per unit area per unit time and is particularly appropriate in more open terrain. A weakness of the technique is its sensitivity to local topographic conditions and a propensity to detect other anthropogenic sources of CO2 (vehicles, industrial plant, etc.), as well as natural variations (diurnal, seasonal, etc.). These have to be carefully characterised so their effects can be removed. A leak from an underground CO2 storage reservoir could only be detected where it perturbs the total flux significantly beyond background fluxes normally experienced in the area of measurement. This indicates a need to collect and understand baseline measurements prior to CO2 injection. Miles et al. (2004) suggest that in a biologically active area, a detection limit of about 4.4 ¥ 10–7 kg CO2 m–2 s–1 is realistic for hourly measurements using eddy covariance.


232


N

CO2 concentration [ppm] 574.1–650.0 534.1–574.0 513.1–534.0 499.0–513.0 489.0–499.0 480.0–489.0 471.1–480.0 457.1–471.0 < 457

0 50 100

200

300

Meters 400

8.14 Image of mobile laser data from the western shore of the Laacher See overlaid on an air photo, with the two main vents clearly identified as bright spots just to the left of the centre of the figure (diagram courtesy of Dave Jones, British Geological Survey).

Ancillary tools Remote sensing technologies (airborne and satellite) can provide full areal coverage of storage sites, albeit with limited sensitivity, and commonly by measuring changes indirectly related to CO2 leakage. Traditional application of the methods involves detecting changes in floral cover due to the effects of CO2. The use of airborne hyperspectral imaging for mapping floral habitats over the Rangeley CO2–EOR field has suggested that surface seepages of CO2 are minimal (Pickles and Cover, 2004). More recent examples of systematic



233

testing of airborne monitoring capability at naturally-occurring leakage sites are given in Bateson et al. (2008). A more novel application of satellite-based monitoring is the measurement of ground displacements using satellite radar interferometry (InSAR). Unlike traditional remote sensing techniques, InSAR does not provide leakage monitoring, but rather has the potential to give insights into the geomechanical behaviour of the storage system. Recent work on differential satellite interferometry (DInSAR) of the In Salah storage site in Algeria has revealed striking patterns of ground movement associated with CO2 injection (Onuma and Ohkawaba, 2009). There is clear evidence of centimetre-scale relative uplift localised around the injection wells, and relative subsidence above the producing gas field (Fig. 8.15). These movements are related directly to pressure changes in the reservoir and the geomechanical properties of the subsurface – key parameters relating to the overall performance and integrity of the storage system. Inverting these surface measurements to extract these deep-seated parameters is technically very challenging and a focus of current research.

8.4.3 Summary of shallow-focused methods A robust shallow monitoring system will probably require a combination of very sensitive point sampling techniques, such as soil gas and surface flux, and less accurate tools such as mobile IR atmospheric detection that provide more areally uniform coverage. Remote sensing technologies can also provide important areal integration albeit at lower sensitivity or by imaging proxy parameters such as vegetation changes.

8.5

Conclusions and future trends

A wide range of potential monitoring tools have been tested at both industrial and research injection sites worldwide. It is likely that for future industrialscale storage, a limited number of core monitoring technologies will be deployed. Deep-focused monitoring will aim to verify short-term site performance, give early warning of deviations from predicted site behaviour (such that appropriate remediation actions can be taken) and provide the key calibration data required to produce robust predictive models of longer-term site performance. This latter element will be crucial to demonstrating the longerterm safety case for a storage site, and paving the way for site closure and transfer of liability from the operator to the regulatory authority. The purposes of shallow-focused monitoring are for leakage detection, verification of emissions performance and public acceptance; the latter issue is likely to be particularly important for onshore sites.


234


KB-5

KB-503

KB-502

KB-8 KB-CA KB-4

KB-9z KB-CB KB-2 KB-501 KB-6 KB-CC (a)

Vertical displacement (mm)

20 15 10

KB-501 KB-502

5

KB-503 KB-CB

0 –5 CO2 injection started –10

0

200

400

600

800

1000 1200 Days (b)

1400

1600

1800 2000

8.15 DInSAR results from In Salah. (a) Surface displacement 2004–2008 – white discs denote injection wells with darker shading marking uplift. (b) displacements for four specific locations with respect to the start of injection (adapted from Onuma and Ohkawaba, 2009).

Monitoring requirements are highly site-specific both in the types of tools used and the required repeat frequency of monitoring surveys. For example, robust surface monitoring is likely to be practicable only at onshore sites. Offshore, reliable quantification of shallow fluxes over extended areas is unlikely to be a practical proposition in the near future. Monitoring frequency will depend on site-specific requirements. At Sleipner, time-lapse surface seismic has been carried out roughly once every



235

two years, but this reflects the needs of the associated research projects and also the serendipitous acquisition of additional surveys in conjunction with time-lapse surveying of the deeper Sleipner gas field. It is unlikely that such frequently repeated time-lapse coverage would be required at a wellperforming purely industrial storage site. The duration of monitoring, particularly in the post-injection and postclosure operational phases, is the subject of much current debate. Again, this is likely to be very site-specific and a prescriptive approach may well be counter-productive. A basic requirement would be that the monitoring programme should be sufficiently extended to demonstrate that the site is performing appropriately in the short-term and is extremely likely to continue to do so in the future. The role of monitoring as a basis for the allocation of emissions credits, carbon trading and national inventories is still under consideration by regulatory authorities. In Europe, a quantitative approach based on the absolute detection limits and associated uncertainties of the site-specific monitoring system is proposed. How this will work in practice, in terms of monitoring system design and deployment, is currently uncertain.

8.6


In order to continue to improve understanding of the capabilities of monitoring systems, a number of research projects are under way worldwide, testing tools at a wide range of storage sites and naturally-occurring seeps. A selection of these projects is listed below: ∑ ∑ ∑

CO2ReMoVe project: www.co2remove.eu/ CO2GeoNet project: www.co2geonet.eu Frio brine injection project: http://www.beg.utexas.edu/environqlty/ co2seq/fieldexperiment.htm ∑ Nagaoka injection project: http://www.rite.or.jp/English/lab/geological/ demonstration.html ∑ CO2SINK injection project: www.co2sink.de/ ∑ Otway Basin injection project: www.co2crc.com.au/otway/ ∑ K12-B injection project: www.k12-b.nl/

8.7

References

Alnes H, Eiken O and Stenvold T (2008) ‘Monitoring gas production and CO2 injection at the Sleipner field using time-lapse gravimetry,’ Geophysics, 73/6, WA155–WA 161. Angerer E, Crampin S, Li X and Davis T L (2003) ‘Processing, modelling and predicting time-lapse effects of over-pressured fluid-injection in a fractured reservoir,’ Geophysical Journal International, 149, 267–280.


236


Arts R and Winthaegen P (2005) ‘Monitoring options for CO2 storage,’ in Thomas D C and Benson S (eds), Geologic Storage of Carbon Dioxide with Monitoring and Verification, Vol. 2, Elsevier, Oxford, UK, 1001–1013. Arts R, Eiken O, Chadwick R A, Zweigel P, Van Der Meer, L and Kirby G A (2004a) ‘Seismic monitoring at the Sleipner underground CO2 storage site (North Sea),’ in Baines S, Gale J and Worden R J (eds), Geological Storage for CO2 Emissions Reduction. Special Publication 233, The Geological Society, London, UK, 181–191. Arts R, Eiken O, Chadwick R A, Zweigel P, van Der Meer L and Zinszner B (2004b) ‘Monitoring of CO2 injected at Sleipner using time-lapse seismic data,’ Energy, 29, 1383–1393. Arts R J, Trani M, Chadwick R A, Eiken O, Dortland S and van der Meer L G H (2009) ‘Acoustic and elastic modeling of seismic time-lapse data from the Sleipner CO2 storage operation,’ in Grobe M, Pashin J C and Dodge R L (eds), Carbon Dioxide Sequestration in Geological Media – State of the Science, AAPG Studies in Geology 59, American Association of Petroleum Geologists, Tulsa, OK, 391–403. Baklid A, Korbøl R and Owren G (1996) ‘Sleipner Vest CO2 disposal, CO2 injection into a shallow underground aquifer’, SPE Annual Technical Conference and Exhibition, Denver, CO, 6–9 October, SPE paper 36600. Bateson L, Vellico M, Beabien S E, Pearce J M, Annunziatellis A, Ciotoli G, Coren F, Lombardi S and Marsh S (2008) ‘The application of remote sensing techniques to monitor CO2 storage sites for surface leakage: method development and testing at Latera (Italy) where naturally-produced CO2 is leaking to the atmosphere’ International Journal of Greenhouse Gas Control, 2, 388–400. Benson S M, Gasperikova E and Hoversten M (2004) Overview of Monitoring Techniques and Protocols for Geologic Storage Projects, IEA Report Number PH4/29, IEA Greenhouse Gas R&D Programme, Cheltenham, UK. Chadwick R A, Arts R, Eiken O, Kirby G A, Lindeberg E and Zweigel P (2004) ‘4D seismic imaging of an injected CO2 bubble at the Sleipner Field, central North Sea,’ in Davies R J, Cartwright J A, Stewart S A, Lappin M and Underhill J R (eds), 3-D Seismic Technology: Application to the Exploration of Sedimentary Basins, Memoir 29, The Geological Society, London, UK, 305–314. Chadwick R A, Arts R and Eiken O (2005a) ‘4D seismic quantification of a growing CO2 plume at Sleipner, North Sea,’ in Dore AG and Vining B (eds), Petroleum Geology: North West Europe and Global Perspectives – Proceedings of the 6th Petroleum Geology Conference, Petroleum Geology Conferences Ltd, The Geological Society, London, UK, 1385–1399. Chadwick R A, Arts R, Eiken O, Williamson P and Williams G (2005b) ‘Geophysical Monitoring of the CO2 plume at Sleipner, North Sea,’ in Lombardi S, Altunia L K and Beaubien S E (eds), Advances in the Geological Storage of Carbon Dioxide, NATO Science, IV Earth and Environmental Sciences Volume 65, 303–314, Springer, Dordrecht, the Netherlands. Chadwick R A, Arts R, Bernstone C, May F, Thibeau S and Zweigel P (2008) Best Practice for the Storage of CO2 in Saline Aquifers, Occasional Publication No. 14, British Geological Survey, Keyworth, UK. Chadwick R A, Arts R, Bentham M, Eiken O, Holloway S, Kirby G A, Pearce J M, Williamson J P and Zweigel P (2009a) ‘Review of monitoring issues and technologies associated with the long-term underground storage of carbon dioxide,’ in Evans D J and Chadwick R A (eds), Underground Gas Storage: Worldwide Experiences and Future Development in the UK and Europe, Special Publication 313, The Geological Society, London, UK, 255–273. © Woodhead Publishing Limited, 2010


237

Chadwick R A, Noy D J, Arts R J and Eiken O (2009b) ‘Latest time-lapse seismic data from Sleipner yield new insights into CO2 plume development,’ in Gale J, Herzog H and Braitsch J (eds), Greenhouse Gas Control Technologies 9, Proceedings of the Ninth International Conference on Greenhouse Gas Control Technologies (GHGT9), Energy Procedia, 1, 2103–2110. Daley T M, Solbau R D, Ajo-Franklin J B and Benson S M (2007) ‘Continuous active-source monitoring of CO2 injection in a brine aquifer,’ Geophysics, 72/5, A57–A61. Delépine N, Clochard V, Labat K, Ricarte P and Le Bras C (2009) ‘Stratigraphic inversion for CO2 monitoring purposes: a case study for the saline aquifer of Sleipner field,’ EAGE Conference and Exhibition, Amsterdam, the Netherlands, 10–11 June (abstract). DTI (2005) Monitoring Technologies for the Geological Storage of CO2, Technology Status Report DTI/Pub URN 05/1032, Crown copyright, London. Freifeld B M, Daley T M, Hovorka S D, Henninges J, Underschultz and Sharma S (2009) ‘Recent advances in well-based monitoring of CO2 sequestration,’ in Gale J, Herzog H and Braitsch J (eds), Greenhouse Gas Control Technologies 9, Proceedings of the Ninth International Conference on Greenhouse Gas Control Technologies (GHGT9), Energy Procedia, 1, 2277–2284. Hepple R P and Benson S M (2003) ‘Implications of surface seepage on the effectiveness of geological storage of carbon dioxide as a climate change mitigation option’ in Gale J and Kaya Y (eds), Proceedings of the Sixth International Conference on Greenhouse Gas Control Technologies, Elsevier Oxford, UK, Vol. 1, 261–266. IEAGHG (2007) IEA Greenhouse Gas R&D Programme Monitoring Selection Tool, available at: http://www.co2captureandstorage.info/co2tool_v2.2.1/index.php (accessed January 2010). Jones D G, Barlow T, Beaubien S E, Ciotoli G, Lister T R, Lombardi S, May F, Moller I, Pearce J M and Shaw R A (2009) ‘New and established techniques for surface gas monitoring at onshore CO2 storage sites,’ in Gale J, Herzog H and Braitsch J (eds), Greenhouse Gas Control Technologies 9, Proceedings of the Ninth International Conference on Greenhouse Gas Control Technologies (GHGT9), Energy Procedia, 1, 2127–2134. Klusman RW (2003) ‘A geochemical perspective and assessment of leakage potential for a mature carbon dioxide enhanced oil recovery project as a prototype for carbon dioxide sequestration: Rangely field, Colorado,’ Bulletin of the American Association of Petroleum Geologists, 87, 1485–1507. Lindeberg, E (2003) ‘The quality of a CO2 repository: What is the sufficient retention time of CO2 stored underground?,’ in Gale J. and Kaya Y (eds), Proceedings of the Sixth International Conference on Greenhouse Gas Control Technologies, Elsevier, Oxford, UK, Vol. 1, 255–260. Mathieson A, Wright I, Roberts D and Ringrose P (2009) ‘Satellite Imaging to Monitor CO2 Movement at Krechba, Algeria,’ in Gale J, Herzog H and Braitsch J (eds), Greenhouse Gas Control Technologies 9, Proceedings of the Ninth International Conference on Greenhouse Gas Control Technologies (GHGT9), Energy Procedia, 1, 2201–2209. McKenna J, Gurevich B, Urosevic M and Evans B J (2003) ‘Rock physics – application to geological storage of CO2’ Journal of the Australian Petroleum Production and Exploration Association, 43, 567–576. Meckel T A, Hovorka S D and Kalyanaraman N (2008) ‘Continuous pressure monitoring for large volume CO2 injections,’ Presented at Ninth International Conference on Greenhouse Gas Control Technologies, Washington, DC, 16–20 November, GCCC Digital Publication Series # 08–03. © Woodhead Publishing Limited, 2010

238


Miles N L, Davis K J and Wyngaard J C (2004) ‘Detecting leaks from belowground CO2 reservoirs using eddy covariance,’ in Edwards DC and Benson S M (eds) Carbon Dioxide Capture for Storage in Deep Geologic Formations, Vol. 2, Elsevier, Oxford, UK, 1031–1044. Nooner S L, Eiken O, Hermanrud C, Sasagawa G S, Stenvold T and Zumberge M A (2007) ‘Constraints on the in situ density of CO2 within the Utsira formation from time-lapse seafloor gravity measurements,’ International Journal of Greenhouse Gas Control, 1, 198–214. Onuma T and Ohkawaba S (2009) ‘Detection of surface deformation related with CO2 injection by DInSAR at In Salah, Algeria,’ in Gale J, Herzog H and Braitsch J (eds), Greenhouse Gas Control Technologies 9, Proceedings of the Ninth International Conference on Greenhouse Gas Control Technologies (GHGT9), Energy Procedia, 1, 2177–2184. Pearce J M, Kirby G A, Chadwick R A, Bentham M S and Holloway, S (2005) Monitoring Technologies for the Geological Storage of CO2, Cleaner Fossil Fuels Programme, UK Department of Trade and Industry, Technology Status Review TSR025, Department of Trade and Industry. Pearce J M, Chadwick R A, Holloway S and Kirby G A (2007) ‘The objectives and design of generic monitoring protocols for CO2 storage,’ in Gale J, Rokke N, Zweigel P and Svenson H (eds), Proceedings of the Eighth International Conference on Greenhouse Gas Control Technologies: GHGT8, Elsevier, Oxford, UK, CD–ROM. Pickles W L and Cover W A (2004) ‘Hyperspectral geobotanical remote sensing for CO2 storage monitoring’ in Thomas DC and Benson S M (eds), Carbon Dioxide Capture for Storage in Deep Geologic Formations–Results from the CO2 Capture Project, Vol. 2, Elsevier, Oxford, UK, 804–858. Prevedel B, Wohlgemuth L, Legarth B, Henninges J, Schutt H, Schmidt-Hatttenberger C, Norden B, Forster A and Hurter S (2009) ‘The CO2 SINK Boreholes for Geological CO2 storage testing,’ in Gale J, Herzog H and Braitsch J (eds), Greenhouse Gas Control Technologies 9, Proceedings of the Ninth International Conference on Greenhouse Gas Control Technologies (GHGT9), Energy Procedia, 1, 2097–2094. Riding J B and Rochelle C A (2009) ‘Subsurface characterisation and geological monitoring of the CO2 injection operation at Weyburn, Saskatchewan, Canada’ in Evans D J and Chadwick R A (eds) Underground gas storage: worldwide experiences and future development in the UK and Europe, Special Publication 313, The Geological Society, London, UK, 227–256. Saito H, Azuma H, Tanase D and Xue Z (2006) ‘Time-lapse crosswell seismic tomography for monitoring the pilot CO2 injection into an onshore aquifer, Nagaoka, Japan’ Exploration Geophysics, 37/1, 30–36. Strutt M H, Beaubien S E, Baubron J C, Brach M, Cardellini C, Granieri R, Jones D G, Lombardi S, Penner L, Quattrocchi F and Voltatorni N (2003) ‘Soil gas as a monitoring tool of deep geological sequestration of carbon dioxide: preliminary results from the EnCana EOR project in Weyburn, Saskatchewan (Canada),’ in Gale J and Kaya Y (eds) Proceedings of the Sixth International Conference on Greenhouse Gas Control Technologies, Elsevier, Oxford, UK, Vol. 1, 391–396. Welles J M, Demetrides-Shah T H and McDermitt D K (2001) ‘Considerations for measuring ground CO2 effluxes with chambers,’ Chemical Geology, 177(1–2), 3–13. Wells A W, Diehl J R, Bromhal G, Strazisar B R, Wilson T H and White C M (2007) ‘The use of tracers to assess leakage from the sequestration of CO2 in a depleted oil reservoir, New Mexico, USA,’ Applied Geochemistry, 22, 996–1016.



239

Wilson M and Monea M (2004) IEA GHG Weyburn CO2 Monitoring & Storage Operation Summary Report 2000–2004, Petroleum Technology Research Centre, Regina, SK, Canada. Zweigel P, Arts R, Lothe A E and Lindeberg E (2004) ‘Reservoir geology of the Utsira Formation at the first industrial-scale underground CO2 storage site (Sleipner area, North Sea),’ in Baines S, Gale J and Worden R H (eds) Geological Storage for CO2 Emissions Reduction, Special Publication 233, Geological Society, London, UK, 165–180.


9

Mathematical modeling of the long-term safety of carbon dioxide (CO2) storage in underground reservoirs

K. P r u e s s, J. B i r k h o l z e r and Q. Z h o u, Lawrence Berkeley National Laboratory, University of California, USA Abstract: Industrial-scale injection of carbon dioxide (CO2) into geologic media will induce coupled processes of fluid flow, mass and heat transfer, and chemical and mechanical interactions between fluids and rocks. Mathematical models play a key role in all aspects of CO2 geologic storage, including site characterization, injection design, and performance monitoring and confirmation. The main challenges for modeling arise from couplings between different processes and the large range of space and time scales that must be considered. This chapter provides a broad perspective of modeling issues in the context of an evolving regulatory environment. Proceeding from large-scale to small, we review model applications to fluid pressurization and brine displacement, the long-term fate of injected CO2, and issues of potential CO2 leakage along faults, fracture zones, and improperly abandoned wells. Selected references offer readers a more in-depth pursuit of the issues addressed in this chapter. Key words: CO2 storage, coupled processes, multi-scale processes, mathematical models, numerical simulation, brine displacement, CO 2 plumes, CO2 leakage.

9.1

Introduction

Subsurface reservoirs being considered for storing Carbon dioxide (CO2) include saline aquifers, oil and gas reservoirs, and unmineable coal seams (Baines and Worden, 2004; IPCC, 2005). By far the greatest storage capacity is in saline aquifers (Dooley et al., 2005), and our discussion will focus primarily on CO2 storage in saline formations. Most issues for safety and security of CO2 storage arise from the fact that, at typical temperature and pressure conditions encountered in terrestrial crust, CO2 is less dense than aqueous fluids. Accordingly, CO2 will experience an upward buoyancy force in most subsurface environments, and will tend to migrate upwards whenever (sub-)vertical permeable pathways are available, such as fracture zones, faults, or improperly abandoned wells (Bachu, 2008; Pruess, 2008a,b; Tsang et al., 2008). CO2 injection will increase fluid pressures in the target formation, thereby altering effective stress distributions and potentially triggering movement along fractures and faults that could increase their 240 © Woodhead Publishing Limited, 2010

Mathematical modeling of the long-term safety of CO2 storage

241

permeability and reduce the effectiveness of a caprock in containing CO2 (Chiaramonte et al., 2008; Rutqvist et al., 2008). Induced seismicity as a consequence of fluid injection is also a concern (Healy et al., 1968; Raleigh et al., 1976; Majer et al., 2007). Dissolution of CO2 in the aqueous phase generates carbonic acid, which may induce chemical corrosion (dissolution) of minerals with associated increase in formation porosity and permeability, and may also mediate sequestration of CO2 as solid carbonate (Gaus et al., 2008). Chemical dissolution of caprock minerals could promote leakage of CO2 from a storage reservoir (Gherardi et al., 2007). Chemical dissolution and geomechanical effects could reinforce one another in compromising CO2 containment. Additional issues arise from the potential of CO2 to mobilize hazardous chemical species (Kharaka et al., 2006), and from migration of the large amounts of brine that would be mobilized by industrial-scale CO2 injection (Nicot et al., 2009; Birkholzer et al., 2008, 2009). Site characterization and selection and careful design of the CO2 storage operation will have sought to minimize such concerns (Birkholzer and Tsang, 2008; Doughty et al., 2008; Gibson-Poole et al., 2008). However, given natural imperfections of geologic media, and the enormous scale at which CO2 storage would have to be carried out to make a noticeable impact on atmospheric emissions (Orr, 2004; Pacala and Socolow, 2004), it seems likely that some CO2 will escape from the primary storage formation (Pruess, 2008a,b). Concerns with leakage include keeping the CO2 out of the atmosphere, protecting valuable groundwater resources from CO2 intrusion and possible mobilization of hazardous trace elements (Wang and Jaffe, 2004; Apps et al., 2009; Zheng et al., 2009), avoiding negative impacts on other subsurface resources and uses, and avoiding detrimental effects at the land surface. The various processes of concern with respect to storage safety may occur on a large range of space and time scales, from pore level to regional, and from seconds to millennia. The multiscale nature of the problem poses great challenges for modeling (BRN, 2007). Key to successful modeling is a sound understanding of the underlying physical and chemical processes, and their mathematical formulation in a set of ‘governing equations.’ For applications to CO2 injection, storage, and migration scenarios, these equations must be complemented with constitutive equations for physical and chemical properties of the interacting gaseous, liquid, and solid media. For site-specific applications, we require hydrogeologic parameters for the host formations, such as porosity, absolute and relative permeability, and others, which usually will display considerable internal variability (‘heterogeneity’), as well as scale dependence. The presence of heterogeneity on multiple scales poses difficult challenges for site characterization and flow modeling alike. Mathematical models can be constructed in different ways, and for


242


different objectives. Models can be designed to approximate and simplify real-world systems by focusing on a few features that are deemed to be especially important. On the other hand, models can be built to include comprehensive process descriptions and to represent in great detail the (known or hypothesized) variability of hydrogeologic conditions and actual injection operations at specific field sites. Simplified, approximate models have certain advantages. They can emphasize specific features and mechanisms, and may often admit analytical or semi-analytical solutions. This facilitates insight into relevant processes and parameters, and may allow broad and robust conclusions to be reached. Detailed and comprehensive site-specific models on the other hand must be implemented through numerical simulators that are capable of representing real-world heterogeneity, and the interplay of many interacting processes on different scales. This will require partitioning of the flow domain into a large number of subdomains or ‘grid cells,’ on the order of one million or more (Zhang et al., 2007; Birkholzer et al., 2008; Yamamoto et al., 2009). Highly spatially resolved models can provide a much more detailed and quantitative outlook on the behavior of flow systems, but the many interacting processes and parameters, and the complexity of data inputs and outputs, pose severe challenges for computation as well as for comprehension of model results. The choice of modeling approach will depend on the objectives of the modeling study. For example, a feasibility assessment of a certain proposed containment mechanism for CO2 may use rather different conceptualizations and mathematical approaches than a model designed to assess the storage capacity and guide injection operations at a specific site. Another important aspect of mathematical models relates to regulatory requirements for CO2 storage, which are in formative stages in different countries (Collord, 2006; Wilson and Gerard, 2007; Wilkinson, 2008). Regulations must be based on a rational, process-based understanding of CO2 behavior in the subsurface (Tsang et al., 2002, 2008). Mathematical models are a chief source of such understanding, informing regulators about what is practical and feasible; at the same time, it is evolving regulations that will influence or even determine the needs and objectives that must be met by mathematical models. This feedback calls for a flexible approach of ‘learning by doing.’ The next section will summarize the main modeling challenges for demonstrating the safety of CO2 storage. This will be followed by a review of recent modeling applications and challenges on issues of central importance to CO2 storage, including (i) fluid pressurization and brine displacement by CO2 injection, (ii) the long-term fate of stored CO2, and (iii) containment of CO2 beneath a caprock, and leakage through fractures, faults, and open wellbores. Selected references to the published literature are given to enable readers a more in-depth pursuit of the issues addressed in this book chapter.



9.2

243

Coupled processes: a challenge for mathematical models

Large-scale injection of CO2 into subsurface reservoirs will induce a complex interplay of coupled processes of multiphase flow, chemical and mechanical interactions between fluids and rocks, and heat transfer. These processes will determine the short-term injection performance of a CO2 storage system as well as the long-term fate of injected CO2. The individual physical and chemical processes and mechanisms affecting the fate of CO2 in the subsurface are well understood ‘in principle,’ based on extensive geoscientific and engineering experience with oil, gas, and geothermal reservoirs, in which similar processes are taking place. However, in practice it is quite challenging to construct accurate models even for a basic process such as CO2 displacing brine in a well-characterized porous medium in the laboratory, with known initial and boundary conditions. This is largely due to the complicated scale- and history-dependence of relative permeability and capillary pressure behavior of two-phase mixtures of water and CO2 (Juanes et al., 2006; Bachu and Bennion, 2007; Doughty, 2007). Applications to geologic CO2 storage systems face additional hurdles, because subsurface flow systems can only be partially characterized and known, introducing significant uncertainty into mathematical models. Even for idealized, hypothetical systems in which all hydrogeologic parameters are prescribed, modeling of CO2 storage faces difficult challenges due to: (1) processes that occur over a broad range of space and time scales, with constitutive parameters that are scale-dependent; and (2) non-linear feedbacks between different processes. We will proceed here with a general overview of modeling challenges posed by multiscale behavior and non-linear feedbacks. Specific examples of modeling approaches developed to address these issues will be given in Section 9.3. CO2 injected into a saline aquifer will initially be present primarily as a free and mobile non-aqueous and non-wetting phase, which for convenience we will refer to as gas. Over time, CO2 may be transferred to different storage modes: it may become partially trapped by capillary forces, some of it may dissolve in the aqueous phase and, eventually, over longer timescales that may extend to hundreds or thousands of years, some of the dissolved CO2 may react with formation minerals to form solid carbonates. The progression from free gas to trapped gas, dissolution in the aqueous phase, and sequestration in solid minerals is very desirable, as CO2 is becoming less mobile during this process, so that permanence and security of CO2 storage are increased (IPCC, 2005). Injection of CO2 into saline aquifers will cause significant fluid pressure


244


increases over very large areas that typically may be two orders of magnitude larger than the extent of the CO2 plume (Pruess et al., 2003; Birkholzer et al., 2009). Fluid pressurization may induce movement on faults and fractures that could enhance the porosity and permeability of permeable pathways, and may compromise caprock integrity. Such effects could be amplified by dissolution of formation minerals and wettability changes induced by CO 2 (Chiquet et al., 2007a). Non-isothermal effects during CO2 injection and storage are usually minor, except for leakage scenarios where expansion of CO2 migrating to shallower horizons can give rise to very strong Joule–Thomson cooling (Katz and Lee, 1990; Skinner, 2003; Pruess, 2005). Boiling of liquid CO2 into gas may occur when stored CO2 leaks upward into sub-critical pressure and temperature regimes, and would be accompanied by strong cooling from latent heat ffects (Pruess, 2008b). Capabilities exist to model the individual processes of fluid flow, chemical and mechanical interactions between rocks and fluids, and heat transfer. Much current research is devoted to more comprehensive modeling of couplings and feedbacks between different processes that may induce strong non-linearities with both self-enhancing and self-limiting features on different space and time scales.

9.3

Ilustrative modeling applications

In this section, we discuss models that involve a range of spatial scales relevant for CO2 storage. We begin with a regional perspective that addresses environmental impacts under conditions where several CO2 storage projects would be conducted simultaneously in a sedimentary basin. As will be seen, pressurization of saline aquifers from CO2 injection could induce brine flow effects over distances of several 100 km, with potential impacts on freshwater aquifers. We then address the long-term fate of CO2 plumes from individual storage projects, that are expected to reach linear dimensions of order 10 km (Pruess et al., 2003). This is followed by a discussion of CO2 leakage along faults, fracture zones and wellbores, which may involve important flow processes on scales as small as 1 m.

9.3.1 Fluid pressurization and brine displacement The amounts of CO2 that would need to be injected and stored underground to make a noticeable impact on atmospheric emissions are very large. Anthropogenic releases of CO2 into the atmosphere are currently almost 30 Gt (billion metric tonnes) per year. At typical in situ densities of stored CO2, the corresponding fluid volume would be about eight times larger than current world oil production. This means that geologic storage of just 15 % of the CO2 that, due to human activities, is currently released into the atmosphere would require a fluid handling system larger than world oil. A CO2 storage



245

project at a large coal-fired power plant of 1000 MWe generating capacity will, over a typical lifetime of 30 years, generate a subsurface plume with linear dimensions of order 10 km or more (Pruess et al., 2003). Pressurization of displaced brine by more than 1 bar, corresponding to a hydraulic head of 10 m, would occur over a region with dimensions of order 100 km. Figure 9.1 shows schematically the large-scale subsurface impacts that will be experienced during and after industrial-scale injection of CO2. While the CO2 plume at depth may be safely trapped under a low-permeability caprock within an anticlinal structure, the footprint area of the region with elevated pressure in the storage formation is much larger than the footprint area of the plume. The environmental impact of large-scale pressure buildup and related brine displacement depends mainly on the hydraulic connectivity between deep saline formations and the drinking water aquifers overlying them. One concern would be a storage formation that extends updip to form a drinking water resource used for domestic or commercial water supply (Bergman and Winter, 1995; Nicot, 2008). If there is direct hydraulic communication, CO 2 storage at depth could impact the shallow portions of the aquifer, which may experience water table rise, changes in discharge and recharge zones, and changes in water quality. Even if separated from deep storage formations by low-permeability seals, freshwater resources may be hydraulically Footprint area of elevated pressure Footprint area of CO2 plume

Possible brine leakage into shallow units

Abandoned well CO2 injection

Brine leakage into upper strata

CO2 plume

on ati to m dip e r fo up urc o ge d ra ten res o St ex ater ay m eshw fr

Pressure perturbation and brine displacement

9.1 Schematic showing different regions of influence related to CO2 storage (from Birkholzer et al., 2009; not to scale).


246


communicating with deeper layers, and the pressure buildup at depth would then provide a driving force for upward brine migration. Interlayer pressure propagation and brine leakage may occur, for example, if high-permeability conduits such as faults and abandoned boreholes are present (see Section 9.3.3, below). Pressure may also propagate in a slow, diffuse process if the sealing layers have a relatively high permeability (Birkholzer et al., 2009). Concerns about large-scale pressure buildup and brine migration caused by industrial-scale CO2 sequestration have been raised as early as the 1990s (van der Meer, 1992; Bergman and Winter, 1995; Gunter et al., 1996). Since then, less emphasis has been placed on evaluating large-scale pressure changes and understanding the fate of the native brines that are being displaced by the injected CO2. Most research into geologic storage of CO2 has instead focused on evaluating the hydrogeological conditions under which the injected volumes of CO2 can be safely stored, addressing issues such as the long-term efficiency of structural trapping of CO2 under sealing layers. Regional estimates of storage capacity for CO2 sequestration have often been based on simple calculations of the fraction of the total reservoir pore space available for safe trapping of CO2 (Bradshaw et al., 2007; USDOE, 2007), implying an underlying assumption of ‘open’ formations from which the native brine can easily escape laterally and make room for the injected CO2. However, recent modeling studies have suggested that environmental concerns related to large-scale pressure buildup may be the limiting factor in CO2 sequestration capacity (e.g., Nicot, 2008; van der Meer and Yavuz, 2009; Zhou et al., 2008). From the standpoint of fluid dynamics, brine pressurization and migration is a much simpler process than two-phase flow of water–CO2 mixtures. The challenge for mathematical modeling is not in fundamental process issues, but rather in obtaining a sufficiently detailed and realistic characterization of large subsurface volumes, to be able to place meaningful limits on quantities and pathways for brine migration. Nicot (2008) employed a single-phase flow model to simulate the regional-scale brine flow processes in response to hypothetical future CO2 sequestration in Texas Gulf Coast Basin, approximating the injection of CO2 by adding equivalent volumes of saline water. Direct comparison of the single-phase flow model with a more complex simulation suggested that the far-field procceses can be represented reasonably well without accounting for local two-phase and variable density effects (Nicot et al., 2009). Other studies have attempted to evaluate regional-scale impacts together with local CO2–water flow processes in one single model. Unless idealized geologic settings are considered (Birkholzer et al., 2009), the need for large model domains combined with non-linear local flow phenomena requires novel grid designs with appropriate spatial resolution, and efficient highperformance computing techniques. Yamamoto et al. (2009) evaluated the



247

impact of CO2 storage on regional groundwater flow in Tokyo Bay, Japan. Their 3D model covered an area of 60 × 70 km2 and was discretized into approximately 10 million gridblocks, with extensive local grid refinement around the ten CO2 injection wells. The model was run on the massivelyparallel Earth Simulator (http://www.jamstec.go.jp/esc/index.en.html), a supercomputer with 5120 CPUs that is among the fastest available to earth scientists worldwide. Birkholzer et al. (2008) developed a regional-scale 3D model for the Illinois Basin with an area of 240 000 km2, that included local mesh refinement and more than one million gridblocks. To illustrate the methods used and results obtained from such large-scale high-performance models, we shall present the Illinois Basin model in more detail below. The Illinois Basin region has annual CO2 emissions of over 300 Mt (million metric tonnes) from fixed sources, primarily from large coal-fired power plants (USDOE, 2007). The primary target for CO2 storage in the area is the Mount Simon Sandstone, a deep saline formation with proven seals, good permeability and porosity, as well as sufficient thickness (Fig. 9.2). With a large (estimated) storage capacity (USDOE, 2007), the Mount Simon is expected to host multiple sequestration sites, based on the current portfolio of Wisconsin

Wisconsin Arch

1000 800 700 600 500 400 300 200 100 0

900

800

700

600

500

800

900

1000

1100

1200

1300

9.2 Boundary of the model domain (solid black line) and thickness of the Mount Simon (in m). Also shown are the core-injection area that hosts 20 hypothetical injection sites (squares), and the Illinois Basin boundary (in grey). Easting and northing coordinates are given in km.


248


industrial point sources and the projected future developments. In northerly direction, the saline formation extends updip to form a freshwater aquifer in southern Wisconsin. Thus, there is concern about potential degradation of freshwater resources due to pressure buildup and brine displacement in response to future deployment of CO2 sequestration in the area. An important objective in the setup of the large-scale numerical model for the Illinois Basin was to predict simultaneously (i) the basin-scale environmental impacts (i.e., pressure buildup and brine displacement) of projected future CO2 storage scenarios involving multiple injection sites, and (ii) the detailed plume-scale phenomena at individual CO2 injection sites and their interaction. The model domain for the Illinois Basin covers an area of roughly 570 km by 550 km (Fig. 9.2). It includes a core area that is suitable for CO2 storage, as well as a far-field area with important groundwater resources where environmental impacts need to be assessed. The core area was selected based on the favorable geological setting, sufficient thickness and depth, and proximity to gas storage fields as well as to various large anthropogenic CO2 sources. In vertical direction, the model comprises the Mt Simon Sandstone as well as the overlying shale unit and a portion of the underlying granite bedrock. A 3D unstructured mesh was constructed with progressive mesh refinement in the core injection areas to capture details of two-phase flow and spatial variability, using grid refinement down to 20 m in the horizontal and 10 m in the vertical direction. A hypothetical carbon sequestration scenario was modeled that assumes 20 individual sequestration sites (spaced about 30 km apart) within the core area. At each site, the assumed annual CO 2 injection rate is 5 Mt for an injection period of 50 years. Illustrative model results showing the characteristics of individual CO2 plumes after 50 years of continuous injection are presented in Fig. 9.3. The maximum size of CO2 plumes, on the order of 6–10 km, is much smaller than the lateral distance between different injection sites, suggesting that merging of plumes would only occur after very long periods of time (hundreds of years or more), if at all. The close-up view in the vertical cross-section highlights the variability of CO2 saturation and how it relates to the internal layering and permeability differences within the Mt Simon. In addition to the local heterogeneity structure, CO2 plume shapes are affected by the thickness of the Mt Simon and the slope of the structural surfaces, while pressure interference from neighboring injection sites has very little effect on plume shape. The details of CO2 saturation evident in Fig. 9.3 emphasize the importance of local mesh refinement to resolve smaller-scale processes, such as structural trapping from internal layering or enhanced dissolution due to spatial heterogeneity. Figure 9.4 shows the simulated pressure buildup (in bar) at the top of the Mt Simon at 10, 50 (end of injection period), 100, and 200 years after start of injection. While the maximum pressures are not a concern with respect


–1800


Elevation (m)

(a) log10 k Eau Claire

Mt Simon

–2200

S

0.0001 mD

Precambrian

(b) CO2 saturation

–1800 Elevation (m)

N

0.001 mD

–2000

–2400

0.4 0.35 0.3 0.25

Eau Claire

N

–2000 Mt Simon

–2200

–2400 703

Precambrian

704

705

706

707 708 (km)

709

710

711

249

9.3 Graph on left shows contours of CO2 saturation after 50 years of injection at elevation of maximum plume extent. Graphs on right show (a) vertical permeability (in millidarcy), and (b) CO2 saturation in a south–north cross section for a selected injection site.

3 2.5 2 1.5


S


10 years

1

0.

0.1

1 5

1

0.1 1 0.

1

0.1

2

2

10

40 30 20 10 5 2 1 0.1

1

1

0.1

1

1

1

1

0.

5

200 years 0.

1

2

40 30 20 10 5 2 1 0.1

0.

0.

2

1

0.1

1

1

5

2

5

0.1

100 years

0.

1

20

30

0.

0.1

10

10

2

10

1

10

0.

20

1

2 10

40 30 20 10 5 2 1 0.1

1

1

5 1

50 years (end of injection)

2

0.

0.

40 30 20 10 5 2 1 0.1

10

250

2

2

2 1 0.1

0.1

1 0.

1

5

0.1

1

0.

5

1 0.1

9.4 Contours of pressure increase (in bar) at the top of the Mount Simon Sandstone at 10, 50, 100, and 200 years after start of injection, assuming a 50-year injection period.

to geomechanical damage and caprock integrity, the pressure buildup from individual storage locations has obvious implications for neighboring sites. After 10 years, a continuous region with pressure buildup of 10 bar or more has evolved in the core injection area, indicating strong pressure interference between different storage sites located at distances of 30 km or more. Such interference may suggest a hierarchical approach to regulating CO2 injection, that would start with a regional assessment of storage capacity and a general permit for a region, prior to permitting individual sites (Nicot and Duncan, 2008). With respect to the far-field impact of CO2 injection and storage, pressure changes may propagate as much as several hundred kilometers away from the core injection area. Moderate pressure buildup (up to 2.0 bar at the top of the Mt Simon) is observed after 50 years of injection in northern Illinois, where valuable and heavily used freshwater aquifers overlie the Mt



251

Simon and Eau Claire. Such pressure changes may have some impact on groundwater recharge and discharge regimes in these freshwater aquifers. In comparison with the magnitude and extent of pressure buildup, the changes in salinity experienced in the subsurface as a result of brine displacement are very small, posing no direct threat to groundwater quality. However, salinity issues could become a concern if deep saline water from the Mt Simon was pushed upward into overlying aquifers via localized pathways, such as transmissive faults or open boreholes, which are currently not included in the model (see Section 9.3.3, below). After CO2 injection ends, the pressure buildup in the core injection area reduces quickly to moderate values around 5–10 bar, while the far-field pressure response initially continues to increase and expand (compare the 50 and 100 year frames in Fig. 9.4). The system then progresses slowly towards an equilibrated state, with pressures eventually returning to hydrostatic conditions, long after the end of the injection period. The simulation results discussed above are for a hypothetical future injection scenario in which one-third of the total CO2 emissions from large point sources in the Illinois Basin are captured and stored. The total injected mass of CO2 after 50 years is 5000 Mt, which is about one sixth of the lower bound of the estimated storage capacity (USDOE, 2007) for the Mt Simon Sandstone in the Illinois Basin. If the CO2 injection were to continue at the assumed injection rate for 250 more years, to fully utilize the estimated storage capacity, the pressure buildup would be much stronger and extend over a larger area than seen in Fig. 9.4. This suggests that estimates of storage capacity, if solely based on effective pore volume of suitable formations, may have to be revised downward, based on assessments of pressure buildup and environmental impacts. It should be noted that the Illinois Basin study discussed here is preliminary, and considerable uncertainty regarding the large-scale geological model needs to be acknowledged. Further site characterization efforts are underway, and model predictions of environmental impacts may change as more details for future storage scenarios are being developed.

9.3.2 Long-term fate of injected carbon dioxide (CO2) When injected into a saline aquifer, supercritical CO2 forms a separate non-aqueous fluid phase, that for convenience we refer to as ‘gas.’ At typical subsurface conditions for terrestrial CO2 storage projects, the in situ density of the gas phase will be less than the density of the aqueous phase, generating a buoyancy force that will drive CO2 towards the top of the permeable interval. CO2 storage would be made into formations that have a suitable caprock of low permeability to contain the CO2, and injection pressures would be limited so as not to exceed the capillary entry pressure of the caprock (Krooss et al., 2005; Bachu and Bennion, 2007; Chiquet


252


et al., 2007b; Gibson-Poole et al., 2008). However, as CO2 spreads beneath the caprock, containment could be compromised if (sub-)vertical permeable pathways with low or vanishing entry pressures were encountered, such as fracture zones and faults, or improperly abandoned wells. Upward buoyancy force could be avoided if CO2 were injected into deep-sea sediments at an ocean depth of more than 3000 m, as this would place CO2 at temperature and pressure conditions that would make it more dense than aqueous phase, thus removing concerns about upward leakage (House et al., 2006). However, storage of CO2 in deep-sea sediments faces great technical and economic challenges, chief among them the difficulty and cost of (i) operating CO2 injection in more than 3000 m water depth, and (ii) transporting the CO2 from land-based sources to deep-ocean injection sites. There is no evidence that these challenges can be overcome in practice, and in the remainder of this section we will focus on issues relating to terrestrial CO2 storage only. CO2 injected into terrestrial saline aquifers will over time become partially trapped by capillary force, rendering it immobile and reducing concerns about leakage. This is due to the well-established fact that in two-phase (gas–liquid) flow, the relative permeabilities of the phases depend not only on their saturation (i.e., the fraction S of pore volume occupied by a phase), but also on the path by which this saturation was reached. Supercritical CO2 injected into an aquifer acts as a non-wetting (nw) phase, and is mobile (nonzero relative permeability) already for very small saturations Snw. However, when aqueous phase reinvades pore space previously occupied by non-wetting phase, the latter becomes immobile or ‘trapped’ at a finite ‘irreducible’ saturation Snw,ir, which may be of the order of 20 %, and increases with the maximum non-wetting phase saturation that had been reached during prior CO2 invasion. CO2 trapped in this way cannot move in spite of experiencing an upward buoyancy force, and would no longer raise concerns about potential leakage (Kumar et al., 2004; Juanes et al., 2006; Doughty, 2007). Doughty (2007) demonstrated the potential significance of hysteretic trapping effects by numerically simulating CO2 injection into a hypothetical aquifer with no caprock whatsoever. Figure 9.5 shows CO2 plumes at different times resulting from injection of 900 000 t CO2 over a 30-day period into the bottom 100 m of a 2D cylindrically symmetric model with uniform permeability of 100 mD. Results are compared for three different assumptions about non-wetting phase relative permeabilities, (a) non-hysteretic with Snw,ir = 0 (‘slippery’ plume), (b) non-hysteretic with Snw,ir = 25 % (‘sticky’ plume), and (c) hysteretic, with Snw,ir varying dynamically from 0 for CO2 invasion to 25 % for water invasion. It is seen that the ‘slippery’ plume advances rapidly towards the land surface, and almost all the injected CO2 leaks out eventually. In contrast, the ‘sticky’ plume never reaches the land surface and remains trapped indefinitely. The hysteretic model predicts fairly rapid upward advancement of the plume at early times, but plume migration slows



Depth (m)

0

Slippery

1 year Sticky

0 –200

–200

–400

–400

–400

–600

–600

–600

–800

–800

–800

–1000

–1000

–1000

0

200

400

Slippery

0

200

0

400

1000 years Sticky

0 –200

–200

–400

–400

–400

–600

–600

–600

–800

–800

–800

–1000

–1000

–1000

200

400

0

200

0

400

200

400

Hysteretic

0

–200

0

Hysteretic

0

–200

0

Depth (m)

253

200

400

Radial distance [m]

Sg 0

0.05

0.1

0.15

0.2

0.25

0.3

0.35

0.4

9.5 Simulated CO2 plumes after one (top) and 1000 years (bottom). The different panels are for a ‘slippery’ plume (left), a ‘sticky’ plume (middle), and for a hysteretic model (right); see text (from Doughty, 2007).

down when increasing amounts of CO2 become trapped at depth as water reinvades pore space previously occupied by CO2. Small amounts of CO2 eventually reach the land surface, but leakage fluxes are much reduced and delayed relative to the ‘slippery’ plume scenario. CO2 may also dissolve in resident aqueous phase, giving rise to a small increase in density so that aqueous phase with dissolved CO2 actually has © Woodhead Publishing Limited, 2010

254


negative buoyancy (Weir et al., 1995; Lindeberg and Bergmo, 2003). As CO2 dissolves into underlying aqueous phase, it is transported away from the phase boundary by molecular diffusion. This is a slow process, but dissolution can be greatly accelerated by convective currents that may form due to the gravitational instability of denser (CO2-rich) aqueous phase above less dense aqueous phase. Accelerating CO2 dissolution is very desirable from the standpoint of storage security, and the process of dissolution–diffusion– convection (DDC) has been studied by many investigators. Mathematical models have focussed on the onset of convective instability, the growth of convective fingers, and the long-term rate of CO2 dissolution (Ennis-King and Paterson, 2003a,b, 2005; Ennis-King et al., 2005; Hesse et al., 2007; Riaz et al., 2006; Xu et al., 2006; Rapaka et al., 2008). Convection induced by CO2 dissolution has similarities to thermally-buoyant convection, as driven by geothermal temperature gradients, and the extensive literature in that field (e.g. Garg and Kassoy, 1981) has helped understanding of convection induced by CO2 dissolution. There are important differences as well, the principal one being the prevalence of a constant (geothermal) temperature gradient as initial condition for thermal convection. This contrasts with the CO2 dissolution problem, in which there is no significant pre-existing gradient of dissolved aqueous CO2, making the system unconditionally unstable with respect to buoyant convection. In numerical simulations of CO2 storage, the need to discretize continuous space and time variables makes it very challenging to properly account for the DDC process with its multiscale nature in both space and time (Zhang et al., 2007). Depending on formation parameters, onset times for convection may range from a fraction of a year to tens of years or more (Ennis-King and Paterson, 2005). The initial spatial scale of the convective instability is determined by the thickness L = Dt of the diffusive boundary layer, where D is diffusivity and t time. Molecular diffusivity of dissolved aqueous CO2 is of order 10–9 m2/s (Tewes and Boury, 2005), so that the length scale corresponding to estimated convective onset times ranges from a few centimeters to a few meters. Numerical simulations of field-scale CO2 storage typically cannot properly resolve these small spatial scales; lack of spatial resolution results in delaying the onset of DDC and reducing its efficiency, thus underestimating the rates and overestimating the time scales for CO2 dissolution (Lindeberg and Bergmo, 2003; Audigane et al., 2007). Alternative approaches that are not subject to limitations arising from space and time discretization are being pursued for modeling the long-term large-scale behavior of injected CO2. An example is the similarity solution technique used by Nordbotten and collaborators to model the long-term behavior of CO2 injected from a single vertical well into a homogeneous confined aquifer of constant thickness (Nordbotten and Celia, 2006). In addition to idealizing flow geometry and formation properties, these authors invoke



255

approximations such as vertical pressure equilibrium, and assume that CO2 invades as a sharp front across which there is a step change in saturation. The ‘payoff’ from these simplifications is that the partial differential equations for two-phase flow of CO2 and water can be transformed into an ordinary differential equation (ODE), that describes the dependence of vertical CO2 plume thickness h on radial distance R and time t as a function of the ‘similarity variable’ z = R2/t. The ODE is non-linear and must be solved numerically but, due to the similarity property, a single solution of the ODE is sufficient to describe the plume behavior for all times and distances. Using similar assumptions of vertical equilibrium and a sharp interface between CO2 and brine, Hesse et al. (2008) considered the migration of a CO2 plume in linear flow geometry for horizontal as well as sloping aquifers. For horizontal aquifers, similarity solutions can be obtained, while sloping aquifers lead to a Riemann (hyperbolic) problem. Approximations to the equations governing two-phase flow of CO2 and brine that will yield self-similar behavior are very valuable, because similarity solutions are inherently ‘multiscale,’ and are thus ideally suited to describe processes extending over a broad range of time and space scales. Dissolved CO2 may chemically interact with formation minerals, leading to both dissolution and precipitation phenomena (Gunter et al., 1997; Johnson et al., 2001; Xu et al., 2005, 2007; Audigane et al., 2007; Gherardi et al., 2007; André et al., 2007; Gaus et al., 2008). Formation of solid carbonates is the most permanenent and desirable form of CO2 storage, but is a slow process at ambient temperature conditions. There is considerable uncertainty about rock–fluid reaction rates applicable to field-scale systems, but it is likely that hundreds of years or more would be required for significant reaction process (Audigane et al., 2007).

9.3.3 Leakage along faults, fracture zones, and wellbores Storage reservoirs and operations for CO2 will be selected and designed in such a way that intact caprock overlying a CO2 storage reservoir will adequately contain CO2, due to sufficiently large entry pressure for nonwetting phase and/or sufficiently small permeability (Krooss et al., 2005; Bachu and Bennion, 2007; Chiquet et al., 2007a,b). Concerns about longterm storage integrity arise from potential caprock imperfections that may provide preferential pathways for upward gas migration, such as fracture zones and faults, or improperly abandoned wells. Geologic storage of CO2 from large fossil-fueled power plants will generate plumes with linear dimensions of order 10 km or more over the lifetime of a power plant (Pruess et al., 2003). On such a scale it seems likely that in most geologic settings some fracture zones and faults will be encountered


256


by the stored CO2. Assessing the feasibility and long-term safety of CO2 storage requires the evaluation of plausible leakage scenarios, to develop an understanding of controlling mechanisms and parameters. From experience with natural CO2 discharges, it is expected that the most likely manner in which CO2 may migrate upwards is ‘diffuse degassing,’ in which low fluxes of CO2 may be discharged over large areas (Barnes et al., 1978; Sorey et al., 1998; Chiodini et al., 2004). Physical and chemical properties of CO2 suggest a potential for feedback processes that could either self-enhance or self-limit the rates of CO2 migration. CO2 is less viscous and less dense than water, so that large increases in volumetric flux and in fluid pressures at shallow horizons are possible when CO2 replaces water in a leakage pathway. CO2 has large compressibility, so that a small pressure reduction can cause large volume expansion. Exsolution of dissolved CO2 also can cause large volume expansion. Depending on geochemical conditions, CO2 may induce dissolution as well as precipitation of minerals, which may serve to enhance or reduce permeabilities along a flow path (Watson et al., 2004; Johnson et al., 2005; LeNindre and Gaus, 2005; Gherardi et al., 2007). Pressure increases associated with CO2 storage and leakage can induce movement along faults, with a potential for increasing permeability (Streit and Hillis, 2004; Streit and Siggins, 2005; Chiaramonte et al., 2008; Rutqvist et al., 2008). Geomechanical and chemical effects may provide feedbacks that conceivably could enhance CO2 leakage beyond what could occur from either mechanism in isolation; however, mathematical modeling is only beginning to address such couplings. When CO2 escapes from the primary storage reservoir and migrates to shallower depths, pressures may drop below the critical point (Pcrit = 73.82 bar), and phase transitions between liquid and gaseous CO2 may occur that will be accompanied by significant latent heat effects. Non-isothermal effects will also arise from decompression of gaseous CO2, the so-called Joule–Thomson effect (Katz and Lee, 1990). Extremely strong cooling effects have been observed on a few occasions where CO2 used for enhanced oil recovery has broken through and discharged from oil production wells (Skinner, 2003). Numerical simulation studies have shown self-enhancing as well as self-limiting feedbacks between fluid flow and heat transfer, with three-phase flow of aqueous phase and liquid and gaseous CO2 playing an important role in limiting overall fluid mobility, due to the generally small relative permeabilities in three-phase conditions (Stone, 1970; Pruess, 2005, 2008a,b). Figure 9.6 shows a leakage scenario in which CO2 migrates upward from the deep storage aquifer along a blind fault, and accumulates in a secondary or ‘parasitic’ reservoir at shallower depth (Pruess, 2008b). Discharge from the secondary accumulation is initiated when CO2 reaches a spill point at which a second fault provides a permeable pathway towards the land surface.



257

Land surface

‘Parasitic’ CO2 reservoir

High-k fault Spill point

High-k fault Aquifer

CO2 upflow from storage reservoir

9.6 Conceptual leakage scenario that involves CO2 migrating up a fault to accumulate in a secondary ‘parasitic’ reservoir at shallower depth. Discharge of CO2 at the land surface occurs after the accumulation reaches a spill point where it can enter another fault (from Pruess, 2008b).

The particular geometric arrangement shown in Fig. 9.6 was intentionally designed to facilitate strong CO2 leakage, by first enabling a substantial accumulation of CO2, and then allowing for self-enhancement of CO2 flow due to depressurization and large volume expansion of CO2 once a discharge gets underway. Simulated CO2 outflow rates at the land surface show only weak self-enhancing feedbacks, as the maximum in CO2 outflow rates increases somewhat super-linearly with applied CO2 injection rate at depth (Fig. 9.7). A profile of fluid saturations as the time of maximum discharge is approaching shows the upper fault being fed with a three-phase mixture of aqueous phase, and liquid and gaseous CO2 (Fig. 9.8). The strong interference between different fluid phases in three-phase flow reduces effective permeabilities for all phases, and limits discharge rates. Figure 9.8 also shows a temperature profile, indicating attainment of very low temperatures in the region where liquid CO2 boils into gas. In many parts of the world, sedimentary basins have been intensely explored and developed for oil and gas production. As an example, the Alberta Basin, Canada, has over 400 000 wells (Celia et al., 2005). Sedimentary basins also are the prime potential sites for geologic storage of CO2, as they often host many sources of anthropogenic CO2, such as fossil-fueled power plants, and have deep saline aquifers, and oil and gas reservoirs, with large capacity for CO2 storage (Bradshaw and Dance, 2005; Bachu, 2008). Improperly abandoned wells may pose a threat to the integrity of CO2 storage projects in sedimentary basins, and many studies have been performed to address


258


1

2

Time [years] 3

CO2 outflow rate (kg/s)

6 ¥ 10–3

4

4

5

6

Injection rate 20.0 ¥ 10–4 kg/s 16.0 ¥ 10–4 kg/s 12.0 ¥ 10–4 kg/s

2

0

0

50

100 Time (s)

200 ¥ 106

150

9.7 Dependence of CO2 outflow behavior on injection rate (from Pruess, 2008b). Flow rates are for a vertical section of 1 m thickness. 20

1.0 Temperature

0.8

Saturation

0.4 10

Liquid CO2 0.2

Temperature (°C)

15 0.6

CO2 gas 5

0.0

–120

–100

–80 –60 –40 Depth along fault (m)

–20

0

9.8 Profile of temperature and CO2 saturations in upper fault at t = 1.50 yr (from Pruess, 2008b).

this problem. A group at Princeton and Bergen universities has developed models for CO2 leakage through wells penetrating formations that are conceptualized as a layer cake of alternating aquifers and aquitards (Fig. 9.9; Celia et al., 2005, 2009; Nordbotten et al., 2004, 2005). Degradation of well cements due to corrosive action of injected CO2 has also been modeled (Duguid et al., 2005). Fully deterministic models are impractical when dealing with very large numbers of potentially leaky wells, and the



Q13(t)

Q23(t)

Q33(t)

M13(t)

M23(t)

M33(t)

Q12(t)

Q22(t)

Q32(t)

Aquifer

M12(t)

M22(t)

M32(t)

Aquitard

Q11(t)

Q21(t)

Q31(t)

M11(t)

M21(t)

M31(t)

Aquifer

CO2 Injection well

259

Abandoned wells

9.9 Conceptual model for leakage through wellbores in a sedimentary basin with layer cake stratigraphy (from Celia et al., 2009).

Princeton/Bergen group used a stochastic approach to model and aggregate the leakage behavior of individual well segments (Fig. 9.10). One limitation of their approach is that flow in leaky well segments is modeled by Darcy’s law. This will be appropriate for wells that provide relatively ‘small’ flow pathways, as e.g. through small cracks in cement plugs. However, there is concern that leakage from a CO2 storage reservoir could be dominated not by a multitude of slightly leaky wellbores, but by a small number of wells with open-hole sections. Flow behavior in open-hole sections, or along an open annulus, cannot be described by Darcy’s law, but requires modeling of velocity-dependent friction and acceleration effects (Zuber and Findlay, 1965; Brill and Mukherjee, 1999; Lu, 2004; Paterson et al., 2008). In the volcanology literature, the possibility of a ‘pneumatic eruption’ has been suggested, so-called because it supposedly would be driven not by thermal energy, but by the mechanical energy stored in a volume of highly compressed gas (Giggenbach et al., 1991; Fischer et al., 1996; Browne and Lawless, 2001). We are not aware of any attempt, either in the volcanology or CO2 storage communities, to mathematically model pneumatic eruptions, and the possibility of such eruptions remains hypothetical (Pruess, 2008a). A quantitative analysis of this problem would require significant advances in capabilities for modeling two- and three-phase flows of water–brine mixtures at high speeds and including non-isothermal effects.

9.4

Conclusions

CO2 injection into subsurface formations for enhanced oil recovery (EOR) has been practiced for over 30 years, and currently about 50 Mt of CO2 per year is © Woodhead Publishing Limited, 2010

260


300 Soft data Bi-modal 250

Count

200

150

100

50

0 –14

–12

–10 –8 –6 –4 Log10 fractional mass to surface

–2

0

9.10 Histogram of the fraction of CO2 leaked to the surface during injection over a 50 yr period into the Nisku formation, Alberta Basin, Canada. Model results are shown for two different stochastic distributions for the permeability of well segments, and using 1000 realizations for each case (from Celia et al., 2009).

injected underground in the USA (http://fossil.energy.gov/programs/reserves/ npr/CO2_EOR_Fact_Sheet.pdf) for this purpose. Clearly, the technology to safely inject CO2 underground exists, but CO2 storage as a climate change mitigation strategy raises new issues, due to the very large amounts that would need to be injected, and the very large space and time scales that come into play. Excellent opportunities for learning about CO2 behavior underground on relevant scales are provided by naturally occurring CO2 discharges in volcanic or tectonically active regions (IEA, 2006; Annunziatellis et al., 2008). Mathematical modeling of the behavior and fate of CO2 injected underground is a key tool for designing and safely operating large-scale CO2 injection projects. Modeling capabilities developed in the context of oil and gas production and storage, geothermal energy extraction, and geologic disposal of industrial wastes have been adapted to the CO2 problem, but further advances are needed to more accurately represent the complex multiscale processes of fluid flow, chemical and mechanical interactions, and heat transfer, that are induced by geologic storage of CO2.



9.5

261

Acknowledgements

The authors wish to thank H.H. Liu for a careful review of the manuscript and the suggestion of improvements. Thanks are also due to Edward Mehnert, Hannes Leetaru, and other colleagues at the Illinois State Geological Survey for their substantial contributions in developing the Illinois Basin model described in Section 9.3.1. The work described in this chapter was funded by the Assistant Secretary for Fossil Energy, Office of Sequestration, Hydrogen, and Clean Coal Fuels, National Energy Technology Laboratory, of the U.S. Department of Energy, and by Lawrence Berkeley National Laboratory under Contract No. DE-AC02-05CH11231.

9.6

References

André L, Audigane P, Azaroual M and Menjoz A (2007) Numerical modeling of fluid–rock chemical interactions at the supercritical co2–liquid interface during CO2 injection into a carbonate reservoir, the Dogger Aquifer (Paris Basin, France), Energy Convers. Manage., 48, 1782–1797. Annunziatellis A, Beaubien SE, Bigi S, Ciotoli G, Coltella M and Lombardi S (2008) Gas migration along fault systems and through the vadose zone in the Latera caldera (central Italy): Implications for CO2 geological storage, Int. J. Greenhouse Gas Control, 2(3), 353–372. Apps JA, Zhang Y, Zheng L, Xu T and Birkholzer JT (2009) Identification of thermodynamic controls defining the concentrations of hazardous elements in potable groundwaters and the potential impact of increasing carbon dioxide partial pressure, in Gale J, Herzog H and Braitsch J (eds), Greenhouse Gas Control Technologies 9, Proceedings of the Ninth International Conference on Greenhouse Gas Control Technologies (GHGT9), Energy Procedia, 1, 1917–1924. Audigane P, Gaus I, Czernichowski-Lauriol I, Pruess K and Xu T (2007) Two-dimensional reactive transport modeling of CO2 injection in a saline aquifer at the Sleipner site, North Sea, Ame. J. Sci., 307, 974–1008. Bachu S (2008) CO2 storage in geological media: role, means, status and barriers to deployment, Prog. Energy Combust. Sci., 34, 254–273. Bachu S and Bennion B (2007) Effects of in-situ conditions on relative permeability characteristics of CO2–brine systems, Env. Geol., 4, 1707–1722. Baines SJ and Worden RH (2004) Geological storage of carbon dioxide, in: S.J. Baines and R.H. Worden (eds), Geological Storage of Carbon Dioxide, Geological Society, London, Special Publications, Vol. 233, 1–6. Barnes I, Irwin WP and White DE (1978) Global distribution of carbon dioxide discharges, and major zones of seismicity, scale 1:40,000,000, in Water Resour. Invest., Open File Report 78–39, U.S. Geological Survey, Washington, DC, 1978. Bergman M and Winter EM (1995) Disposal of carbon dioxide in aquifers in the US, Energy Convers. Manage., 36, 523–526. Birkholzer J and Tsang CF (2008) Introduction to the special issue on site characterization for geological storage of CO2, Environ. Geol., 54(8), 1579–1581. Birkholzer JT, Zhou Q, Zhang K, Jordan P, Rutqvist J and Tsang C-F (2008) CO2 Geological Storage and Groundwater Resources: Large-Scale Hydrological Evaluation


262


and Modeling of the Impact on Groundwater Systems, Lawrence Berkeley National Laboratory Report. Birkholzer JT, Zhou Q and Tsang C-F (2009) Large-scale impact of CO 2 storage in deep saline aquifers: a sensitivity study on the pressure response in stratified systems, Int. J. Greenhouse Gas Control, 3(2), 181–194. Bradshaw J and Dance T (2005) Mapping geological storage prospectivity of CO2 for the world’s sedimentary basins and regional source to sink matching, in Rubin E S, Keith D W and Gilboy C F (eds), Proceedings of the Seventh International Conference on Greenhouse Gas Control Technologies: GHGT7, IEA GHG, Cheltenham, UK, 1, 583–591. Bradshaw J, Bachu S, Bonijoly D, Burruss R, Holloway S, Christensen NP and Mathiassen OM (2007) CO2 Storage capacity estimation: issues and development of standards, Int. J. Greenhouse Gas Control, 1, 62–68. Brill JP and Mukherjee H (1999) Multiphase Flow in Wells, SPE Monograph, 17, Society of Petroleum Engineers, Richardson, TX. BRN (Basic Research Needs for Geosciences: Facilitating 21st Century Energy Systems) (2007) Report from the Workshop Held February 21–23, 2007, Office of Basic Energy Sciences, U.S. Department of Energy, June. Browne PRL and Lawless JV (2001) Characteristics of hydrothermal eruptions, with examples from New Zealand and elsewhere, Earth-Sci. Rev., 52, 299–331. Celia MA, Bachu S, Nordbotten JM, Gasda SE and Dahle HK (2005) Quantitative estimation of CO2 leakage from geological storage: analytical models, numerical models, and data needs, in Rubin E S, Keith D W and Gilboy C F (eds), Proceedings of the Seventh International Conference on Greenhouse Gas Control Technologies: GHGT7, IEA GHG, Cheltenham, UK, 1, 663–671. Celia MA, Nordbotten JM, Bachu S, Dobossy M and Court B (2009) Risk of leakage versus depth of injection in geological storage, in Gale J, Herzog H and Braitsch J (eds), Greenhouse Gas Control Technologies 9, Proceedings of the Ninth International Conference on Greenhouse Gas Control Technologies (GHGT9), Energy Procedia, 1, 2573–2580. Chiaramonte L, Zoback MD, Friedmann J and Stamp V (2008) Seal integrity and feasibility of CO2 sequestration in the Teapot Dome EOR pilot: geomechanical site characterization, Environ Geol., 54(8), 1635–1656. Chiodini G, Cardellini C, Amato A, Boschi E, Caliro S, Frondini F and Ventura G (2004) Carbon dioxide degassing and seismogenesis in central and southern Italy, Geophys. Res. Lett., 31, L07615. Chiquet P, Broseta D and Thibeau S (2007a) Wettability alteration of caprock minerals by carbon dioxide, Geofluids, 7, 112–122. Chiquet P, Daridon JL Broseta D and Thibeau S (2007b) CO2/water interfacial tensions under pressure and temperature conditions of CO2 geological storage, Energy Convers. Manage., 48, 736–744. Collord G (2006) Implementation of SB 1368 Emission Performance Standard, report CEC-700-2006-011, California Energy Commission, Sacramento, CA. Dooley JJ, Kim SH, Edmonds JA, Friedman SJ and Wise MA (2005) A first-order global geological CO2-storage potential supply curve and its application in a global integrated assessment model, in Rubin E S, Keith D W and Gilboy C F (eds), Proceedings of the Seventh International Conference on Greenhouse Gas Control Technologies: GHGT7, IEA GHG, Cheltenham, UK, 1, 573–581. Doughty C (2007) Modeling geologic storage of carbon dioxide: comparison of non-



263

hysteretic and hysteretic characteristic curves, Energy Convers. Manage., 48, 1768–1781. Doughty C, Freifeld BM and Trautz RC (2008) Site characterization for CO2 geologic storage and vice versa: the Frio Brine Pilot, Texas, USA as a case study, Environ. Geol., 54(8), 1635–1656. Duguid A, Radonjic M, Bruant R, Mandecki T, Scherer G and Celia M (2005) The effect of CO2 sequestration on oil well cements, in Wilson M, Morris T, Gale J and Thambimuthu K (eds), Proceedings of the Seventh International Conference on Greenhouse Gas Control Technologies: GHGT7, IEA GHG, Cheltenham, UK, 2, 1997–2001. Ennis-King J and Paterson L (2003a) Rate of dissolution due to convective mixing in the underground storage of carbon dioxide, in Gale J and Kaya Y (eds), Proceedings of the Sixth International Conference on Greenhouse Gas Control Technologies: GHGT6, Elsevier (Pergamon), Oxford, UK, 507–510. Ennis-King J and Paterson L (2003b) Role of convective mixing in the long-term storage of carbon dioxide in deep saline formations, paper SPE-84344, Society of Petroleum Engineers Annual Fall Technical Conference and Exhibition, Denver, CO, 5–8 October. Ennis-King J and Paterson L (2005) Role of convective mixing in the long-term storage of carbon dioxide in deep saline formations, SPE Journal, 10(3), 349–356. Ennis-King J, Preston I and Paterson L (2005) Onset of convection in anisotropic porous media subject to a rapid change in boundary conditions, Phys. Fluids, 17, 084107. Fischer TP, Arehart GB, Sturchio NC and Williams SN (1996) The relationship between fumarole gas composition and eruptive activity at Galeras Volcano, Colombia, Geology, 24(6), 531–534. Garg SK and Kassoy DR (1981) Convective heat and mass transfer in hydrothermal systems, in: Rybach L and Muffler L.JP (eds), Geothermal Systems: Principles and Case Histories, John Wiley & Sons, Chichester, New York, Brisbane, Toronto, 37–76. Gaus I, Audigane P, André L, Lions J, Jacquemet N, Durst P, Czernichowski-Lauriol I and Azaroual M (2008) Geochemical and solute transport modelling for CO2 storage, what to expect from it?, Int. J. Greenhouse Gas Control, 2, 605–625. Gherardi F, Xu T and Pruess K (2007) Numerical modeling of self-limiting and selfenhancing caprock alteration induced by CO2 storage in a depleted gas reservoir, Chem. Geol., 244, 103–129. Gibson-Poole CM, Svendsen L, Underschultz J, Watson MN, Ennis-King J, van Ruth PJ, Nelson EJ, Daniel RF and Cinar Y (2008) Site characterisation of a basin-scale CO2 geological storage system: Gippsland Basin, Southeast Australia, Environ. Geol., 54(8), 1583–1606. Giggenbach WF, Sano Y and Schmincke HU (1991) CO2-rich gases from Lakes Nyos and Monoun, Cameroon; Lacher See, Germany; Dieng, Indonesia, and Mt. Gambier, Australia – variations on a common theme, J. Volcanol. Geotherm. Res., 45, 311–323. Gunter WD, Bachu S, Law D H-S, Marwaha V, Drysdale DL McDonald, DE and McCann TJ (1996) Technical and economic feasibility of CO2 disposal in aquifers within the Alberta Sedimentary Basin, Canada, Energy Convers. Manage., 37(6–7), 1135–1142. Gunter WD, Wiwchar B and Perkins EH (1997) Aquifer disposal of CO2-rich greenhouse gases: extension of the time scale of experiment for CO2-sequestering reactions by geochemical modeling, Mineral. Petrol., 59, 121–140. Healy JH, Rubey WW, Griggs DT and Raleigh CB (1968) The Dever earthquakes, Science, 161, 1301–1310. © Woodhead Publishing Limited, 2010

264


Hesse MA, Tchelepi HA and Orr Jr. FM (2007) Natural convection during aquifer CO2 storage, in Gale J, Rokke N, Zweigel P and Svenson H (eds), Proceedings of the Eighth International Conference on Greenhouse Gas Control Technologies: GHGT8, Elsevier, Oxford, UK, CD-ROM. Hesse MA, Orr Jr. FM and Tchelepi HA (2008) Gravity currents with residual trapping, J. Fluid Mech., 611, 35–60. House KZ, Schrag DP, Harvey CF and Lackner KS (2006) Permanent carbon dioxide storage in deep-sea sediments, Proc. Natl. Acad. Sci., USA 103(33), 12291–12295. IEA (2006) Natural Releases of CO2, Summary Report, IEA Greenhouse Gas R & D Programme, Cheltenham, UK, available at: http://www.ieagreen.org.uk/glossies/ naturalreleases.pdf (accessed January 2010). IPCC (2005) IPCC Special Report on Carbon Dioxide Capture and Storage, Working Group III of the Intergovernmental Panel on Climate Change, Metz B, Davidson O, de Coninck HC, Loos M and Meyer LA (eds), Cambridge University Press, Cambridge, UK. Johnson JW, Nitao JJ, Steefel CI and Knaus KG (2001) Reactive transport modeling of geologic CO2 sequestration in saline aquifers: the influence of intra-aquifer shales and the relative effectiveness of structural, solubility, and mineral trapping during prograde and retrograde sequestration, Proceedings, First National Conference on Carbon Sequestration, Washington, DC, May 14–17. Johnson JW, Nitao JJ and Morris JP (2005) Modeling the long-term isolation performance of natural and engineered geologic CO2 storage sites, in Wilson M, Morris T, Gale J and Thambimuthu K (eds), Proceedings of the Seventh International Conference on Greenhouse Gas Control Technologies: GHGT7, IEA GHG, Cheltenham, UK, 2, 1315–1321. Juanes R, Spiteri EJ Orr Jr., FM and Blunt MJ (2006) Impact of relative permeability hysteresis on geological CO2 storage, Water Resour. Res., 42, W12418. Katz DL and Lee RL (1990) Natural Gas Engineering, McGraw-Hill Publ. Comp., New York, NY. Kharaka Y, Cole DR, Hovorka SS, Gunter WD, Knauss KG and Freifeld BM (2006) Gas–water–rock interactions in Frio formation following CO 2 injection: implications for the storage of greenhouse gases in sedimentary basins, Geology, 34, 577–580. Krooss BM, Hildenbrand A, Alles S, Littke R and Pearce J (2005) Assessment of the CO2 sealing efficiency of pelitic rocks: two-phase flow and diffusive transport, in Wilson M, Morris T, Gale J and Thambimuthu K (eds), Proceedings of the Seventh International Conference on Greenhouse Gas Control Technologies: GHGT7, IEA GHG, Cheltenham, UK, 2, 2003–2006. Kumar A, Noh M, Pope GA, Sepehrnoori K, Bryant S and Lake LW (2004) Reservoir simulation of CO2 storage in deep saline aquifers, paper SPE-89343, Society of Petroleum Engineers Fourteenth Symposium on Improved Oil Recovery, Tulsa, OK, 17–21 April. LeNindre YM and Gaus I (2005) Characterisation of the Lower Watrous Aquitard as a major seal for CO2 geological sequestration (Weyburn Unit, Canada), in Rubin E S, Keith D W and Gilboy C F (eds), Proceedings of the Seventh International Conference on Greenhouse Gas Control Technologies: GHGT7, IEA GHG, Cheltenham, UK, 1, 761–769. Lindeberg E and Bergmo P (2003) The long-term fate of CO2 injected into an aquifer, in Gale J and Kaya Y (eds), Proceedings of the Sixth International Conference on



265

Greenhouse Gas Control Technologies: GHGT6, Elsevier (Pergamon), Oxford, UK, 1, 489–494. Lu X (2004) An Investigation of Transient Two-Phase Flow in Vertical Pipes with Particular Reference to Geysering, PhD thesis, The University of Auckland, New Zealand. Majer EL, Baria R, Stark M, Oates S, Bommer J, Smith B and Asanuma H (2007) Induced seismicity associated with enhanced geothermal systems, Geothermics, 36(3), 185–222. Nicot J-P (2008) Evaluation of large-scale CO2 storage on fresh-water sections of aquifers: an example from the Texas Gulf Coast Basin, Int. J. Greenhouse Gas Control, 2, 582–593. Nicot J-P and Duncan IJ (2008) Science-based permitting of geological sequestration of CO2 in brine reservoirs in the U.S., Environ. Sci. Policy, 11, 14–24. Nicot J-P, Hovorka SD and Choi J-W (2009) Investigation of water displacement following large CO2 sequestration operations, in Gale J, Herzog H and Braitsch J (eds), Greenhouse Gas Control Technologies 9, Proceedings of the Ninth International Conference on Greenhouse Gas Control Technologies (GHGT9), Energy Procedia, 1, 4411–4418. Nordbotten JM and Celia MA (2006) Similarity solutions for fluid injection into confined aquifers, J. Fluid Mech., 561, 307–327. Nordbotten JM, Celia MA and Bachu S (2004) Analytical solutions for leakage rates through abandoned wells, Water Resour. Res., 40, W04204. Nordbotten JM, Celia MA, Bachu S and Dahle HK (2005) Semianalytical solution for CO2 leakage through an abandoned well, Environ. Sci. Technol., 39(2), 602–611. Orr FM Jr. (2004) Storage of carbon dioxide in geologic formations, J. Pet. Technol., 56(9), 90–97. Pacala S and Socolow R (2004) Stabilization wedges: solving the climate problem for the next 50 years with current technologies, Science, 305, 968–972. Paterson L, Lu M, Connell LD and Ennis-King J (2008) Numerical modeling of pressure and temperature profiles including phase transitions in carbon dioxide wells, paper SPE-115946, Society of Petroleum Engineers 2008 Annual Technical Conference and Exhibition, Denver, CO, 21–24 September. Pruess K (2005) Numerical studies of fluid leakage from a geologic disposal reservoir for CO2 show self-limiting feedback between fluid flow and heat transfer, Geophys. Res. Lett., 32(14), L14404. Pruess K (2008a) On CO2 fluid flow and heat transfer behavior in the subsurface, following leakage from a geologic storage reservoir, Environ. Geol., 54(8), 1677–1686. Pruess K (2008b) Leakage of CO2 from geologic storage: role of secondary accumulation at shallow depth, Int. J. Greenhouse Gas Control, 2(1), 37–46. Pruess K, Xu T, Apps J and García J (2003) Numerical modeling of aquifer disposal of CO2, Paper SPE-83695, SPE Journal, 49–60. Raleigh CB, Healy JH and Bredehoeft JD (1976) An experiment in earthquake control at Rangely, Colorado, Science, 191, 1230–1237. Rapaka S, Chen S, Pawar RJ, Stauffer PH and Zhang D (2008) Non-modal growth of perturbations in density-driven convection in porous media, J. Fluid Mech., 609, 285–303. Riaz A, Hesse M, Tchelepi HA and Orr Jr. FM (2006) Onset of convection in a gravitationally unstable diffusive boundary layer in porous media, J. Fluid Mech., 548, 87–111. Rutqvist J, Birkholzer J and Tsang CF (2008) Coupled reservoir-geomechanical analysis of the potential for tensile and shear failure associated with CO2 injection in multilayered reservoir-caprock systems, Int. J. Rock Mech. Mining Sci., 45(2), 132–143. © Woodhead Publishing Limited, 2010

266


Skinner L (2003) CO2 blowouts: an emerging problem, World Oil, 224(1), 38–42. Sorey ML, Evans WC, Kennedy BM, Farrar CD, Hainsworth LJ and Hausback B (1998) Carbon dioxide and helium emissions from a reservoir of magmatic gas beneath Mammoth Mountain, California, J. Geophys. Res., 103(B7), 15303–15323. Stone HL (1970) Probability model for estimating three-phase relative permeability, Trans. SPE of AIME, 249, 214–218. Streit JE and Hillis RR (2004) Estimating fault stability and sustainable fluid pressures for underground storage of CO2 in porous rock, Energy, 29, 1445–1456. Streit JE and Siggins AF (2005) Predicting, monitoring and controlling geomehcanical effects of CO2 injection, in Rubin E S, Keith D W and Gilboy C F (eds), Proceedings of the Seventh International Conference on Greenhouse Gas Control Technologies: GHGT7, IEA GHG, Cheltenham, UK, 1, 643–651. Tewes F and Boury F (2005) Formation and rheological properties of the supercritical CO2–water pure interface, J. Phys. Chem. B, 109(9), 3990–3997. Tsang CF, Benson SM, Kobelski B and Smith RE (2002) Scientific considerations related to regulation development for CO2 sequestration in brine formations, Environ. Geol., 42, 275–281. Tsang CF, Birkholzer J and Rutqvist J (2008) A comparative review of hydrologic issues involved in geologic storage of CO2 and injection disposal of liquid waste, Environ. Geol., 54(8), 1723–1737. USDOE (U.S. Department of Energy) (2007) Carbon Sequestration Atlas of the United States and Canada, National Energy Technology Laboratory, Pittsburgh, PA. van der Meer LGH (1992) Investigation regarding the storage of carbon dioxide in aquifers in the Netherlands, Energy Convers. Manage., 33(5–8), 611–618. van der Meer LGH and Yavuz F (2009) CO2 storage capacity calculations for the Dutch subsurface, in Gale J, Herzog H and Braitsch J (eds), Greenhouse Gas Control Technologies 9, Proceedings of the Ninth International Conference on Greenhouse Gas Control Technologies (GHGT9), Energy Procedia, 1, 2615–2622. Wang S and Jaffe PR (2004) Dissolution of a mineral phase in potable aquifers due to CO2 releases from deep formations; effect of dissolution kinetics, Energy Convers. Manage., 45, 2833–2848. Watson MN, Zwingmann N and Lemon NM (2004) The Ladbroke Grove–Katnook Carbon Dioxide Natural Laboratory: a recent CO2 accumulation in a lithic sandstone reservoir, Energy, 29, 1457–1466. Weir GJ, White SP and Kissling WM (1995) Reservoir storage and containment of greenhouse gases, Energy Convers. Manage., 36(6–9), 531–534. Wilkinson R (2008) Aussie bill sets up CO2 sequestration framework, Oil and Gas Journal, 13 November, available at: http://www.ogj.com/articles/article_display. cfm?ARTICLE_ID=345336&p=7 (accessed January 2010). Wilson EJ and Gerard D (eds) (2007) Carbon Capture and Sequestration: Integrating Technology, Monitoring and Regulation, Blackwell Publishing, Ames, IA. Xu T, Apps JA and Pruess K (2005) Mineral sequestration of carbon dioxide in a sandstone–shale system, Chem. Geol., 217, 295–318. Xu X, Chen S and Zhang D (2006) Convective stability analysis of the long-term storage of carbon dioxide in deep saline aquifers, Adv. Water Res., 29, 397–407. Xu T, Apps JA, Pruess K and Yamamoto H (2007) Numerical modeling of injection and mineral trapping of CO2 with H2S and SO2 in a sandstone formation, Chem. Geol., 242, 319–346. Yamamoto H, Zhang K, Karasaki K, Marui A Uehara H and Nishikawa N (2009)



267

Large-scale numerial simulation of CO2 geologic storage and its impact on regional groundwater flow, in Gale J, Herzog H and Braitsch J (eds), Greenhouse Gas Control Technologies 9, Proceedings of the Ninth International Conference on Greenhouse Gas Control Technologies (GHGT9), Energy Procedia, 1, 1872–1878. Zhang K, Doughty C, Wu YS and Pruess K (2007) Efficient parallel simulation of CO2 geologic sequestration in saline aquifers, Paper SPE 106026, Proceedings of the 2007 SPE Reservoir Simulation Symposium, Houston, TX, 26–28 February. Zheng L, Apps JA, Zhang Y, Xu T and Birkholzer JT (2009) Reactive transport simulations to study groundwater quality changes in response to CO2 leakage from deep geological storage, in Gale J, Herzog H and Braitsch J (eds), Greenhouse Gas Control Technologies 9, Proceedings of the Ninth International Conference on Greenhouse Gas Control Technologies (GHGT9), Energy Procedia, 1, 1887–1894. Zhou Q, Birkholzer JT, Tsang C-F and Rutqvist J (2008) A method for quick-assessment of CO2 storage capacity in closed and semi-closed porous formations, Int. J. Greenhouse Gas Control, 2, 626–629. Zuber N and Findlay JA (1965) Average volumetric concentration in two-phase flow systems, J. Heat Transfer, Trans. ASME, 87(4), 453–468.


10

Terrestrial sequestration of carbon dioxide (CO2) R. L a l, The Ohio State University, USA Abstract: Carbon sequestration in terrestrial ecosystems, comprising soils and biota, is one of several possible strategies being considered to stabilize atmospheric concentration of CO2 and other greenhouse gases. This chapter looks at basic techniques of carbon sequestration in terrestrial ecosystems by which atmospheric CO2 is transferred into the terrestrial and aquatic C pools via the natural process of photosynthesis. The role of the terrestrial pool in the global carbon cycle, agricultural emissions, sequestration of carbon in soil and the potential benefits and challenges of terrestrial carbon sequestration are described together with the role of soil and terrestrial carbon as indicators of climate change. Key words: soil C pool, global warming, greenhouse effect, global C cycle, no-till farming, secondary carbonates, net primary productivity.

10.1

Introduction

The atmospheric concentration of carbon dioxide (CO2) has increased from 280 ppmv in the pre-industrial era to ~385 ppm in 2008, and is increasing at the rate of ~2 ppmv/y (0.5 %/y) primarily because of two anthropogenic activities: fossil fuel combustion and cement manufacturing, and deforestation along with land use conversion and soil cultivation for crop production (IPCC, 2007a,b; Pielke, 2008). The drastic disturbance in the global carbon (C) cycle is widely considered to be the cause of the observed and projected increase in global temperature, sea level rise, and frequency of extreme events (IPCC, 2007a). Consequently, there is a strong interest in reducing the net emissions of CO2 and other greenhouse gases (GHGs). There are several options for stabilizing the atmospheric concentrations of GHGs (Fig. 10.1). These options can be grouped under two broad categories: (i) reducing emissions and (ii) sequestering emissions. Strategies for reducing emissions include those related to enhancing energy use efficiency and developing C-negative or C-neutral fuel sources (e.g., wind, solar, geothermal energy, and biofuels). Technological options for sequestrating emissions, transferring atmospheric CO2 into other C pools with a long residence time, involve geo-engineering technologies of capture and storage in geologic strata (Broecker, 2008). Included among these abiotic and engineering techniques are the options of chemical transformations of CO2 into stable minerals or carbonation. Biotic techniques of C sequestration are based on the natural process 271 © Woodhead Publishing Limited, 2010

272

Options for stabilizing atmospheric concentrations of GHGs

Sequestering emissions


Biotic

Abiotic Clean energy

Alternate energy (biofuels)

Fuel use efficiency

Terrestrial

Aquatic

Mineralization

Clean fossil energy Food production systems

Capture

Trees Transport systems

Materials

Geological strata

Coal seams (CBM)

Wetlands

Buildings & infrastructure Injection

Oil wells (EOR)

Soils

Saline aquifers

Organic C pool

Ocean (biological pump)

Secondary carbonates

Ocean

Basalt

10.1 Technological options for reducing net emissions of CO2 and other greenhouse gases into the atmosphere.


Reducing emissions


273

of photosynthesis of converting atmospheric CO2 into simple sugars and carbohydrates and complex celluloses and lignins. In other words, atmospheric CO2 is transferred into the terrestrial and aquatic C pools via the natural process of photosynthesis. The objective of this chapter is to collate, synthesize, and describe the state-of-the-knowledge about the processes and practices of sequestration of atmospheric CO2 into the terrestrial C pools. The chapter focuses more on the basic concepts, including potential and challenges of abiotic vis-á-vis biotic strategies, rather than on a comprehensive compilation of all the available literature on the topic, for which the readers are referred to other publications (e.g., Prentice et al., 2001; Houghton, 2003a,b).

10.2

The terrestrial pool and its role in the global carbon cycle

There are four principal global C pools: oceanic, geologic, atmospheric, and terrestrial (Prentice et al., 2001). The global C cycle refers to the exchange of C within and between these four pools. The oceanic pool is the largest of the four pools and contains ~39 000 Pg C. Of this total, the surface ocean contains 700 Pg of dissolved inorganic C (DIC) and dissolved organic C (DOC), and it is increasing at the rate of 0.4 Pg C/y. The intermediate and deep ocean contains 38 000 Pg (36 300 Pg of DIC and 975 Pg of DOC) and is increasing at the rate of about 1.4 Pg C/y. About 0.01 Pg C/y is transferred into the sediment. The surface biota in the ocean is comprised of 3 Pg of live and ~11 Pg of detritus materials. The geologic pool is the second largest pool. It is estimated at 5000–10 000 Pg C, comprising coal, oil, and gas. During 2000s, it has been mined and combusted at the rate of 7.5–8 Pg C/y. The atmospheric pool is about 780 Pg and is increasing at the rate of about ~4 Pg/y. The terrestrial C pool is the third largest pool. It has two distinct but related components: biotic or plant C and the pedologic or soil C (Fig. 10.2). The biotic pool contains 550 ± 100 Pg in above-ground and below-ground components, and an additional 300 Pg of litter and detritus materials. The atmospheric pool interacts very closely with the biotic and pedologic pools. The biotic pool photosynthesizes ~120 Pg C/y, of which 59 Pg is returned back to the atmosphere through respiration by plants and 58 Pg through respiration by organisms in soil (Fig. 10.3). Deforestation and land use change contribute 1.7 Pg C/y into the atmosphere. The pedologic pool consists of soil organic C (SOC) pool estimated at 1500–2000 Pg C in the top 1 m, and as much as 2300 Pg up to 2 m depth (Batjes, 1996; Jobbagy and Jackson, 2000; Amundson, 2001). The SOC pool consists of highly active humus and relatively recalcitrant charcoal C, and is collectively called soil organic matter (SOM). The latter comprises the sum of all organic materials in the soil, including mixture of plant and animal residues at various stages of


274


Pedologic C (2500 Pg)

Biotic C (620 Pg)

Above ground

Below ground

Coarse roots

Detritus material

Fine roots

Inorganic (950 Pg)

Lithogenic

Pedogenic

Organic (1550 Pg)

Active/ labile

Intermediate

Passive/ recalcitrant

10.2 Components of terrestrial carbon pool in different global biomes. Predominant biomes include forest, shrub lands, savanna, grasslands, tundra, and deserts (see Table 10.1).


Terrestrial C in global biomass


275

Atmospheric pool 780 Pg (+4 Pg/y)

60

/y

lan

res

pir

ati on du se cha ng e

syn oto

/y

Pg

on

1.7

n

Pg

tio

59

siti

ra spi l re

po

ph

com

soi

/y

de

/y

Pg

er

Pg

120

litt

the

/y

sis

Pg

2.5

Biotic pool Live: 550 ± 100 Pg Litter: 300 Pg (+0.7 Pg/y)

60 Pg/y (plant-C) 2.5 Pg/y (litter-C)

Pedologic pool SOC: 1500–2300 Pg SIC: 950 Pg (+0.1 Pg/y)

10.3 Interaction between the terrestrial and atmospheric C pools (redrawn from Houghton, 2003a,b; Prentice et al., 2001; Batjes, 1996; Jobbágy and Jackson, 2000; Schulze, 2006).

decomposition, substances synthesized microbiologically and/or chemically from the breakdown products, and the bodies of live microorganisms and small animals and their decomposing products (Schnitzer, 1991). In essence, the SOM pool consists of three components: (i) the coarse material (> 2 mm) which comprises leaf litter, coarse woody debris, plant roots, and charcoal, (ii) humus (2 mm) comprising macro- and microorganisms, and (iii) organo-mineral complexes which form and stabilize micro-aggregates. In this context, the SOC pool is a component of the SOM, including organic materials within and on the soil surface. Baldock and Nelson (1998) stated that ‘SOM is the sum of all natural and thermally altered biologically derived organic materials found in the soil or on the surface irrespective of its source, whether it is living or dead, or stage of decomposition, but excluding the above ground portion of the living plants’. The pedologic pool also contains soil inorganic carbon (SIC), comprising primary or lithogenic carbonates and secondary or pedogenic carbonates. The primary carbonates are derived from the weathering of parent material. The secondary carbonates are formed through the dissolution of atmospheric CO2 with water in the soil, and the reaction of the weak carbonic acid with Ca2+, Mg2+, and other cations.


276


The terrestrial C pool is estimated at about 2844 Pg, comprising 558 Pg of plant-C and 2286 Pg of soil-C, with a soil-C:plant-C ratio of about 4:1 (Table 10.1). The largest reservoir of plant-C is the forest ecosystems (Dixon et al., 1994), comprising 415 Pg or 74 % of the total plant-C pool. The soil-C pool in different biomes consists of 675 Pg (29.5 %) in peatlands, 600 Pg (26.2 %) in the forest, and 490 Pg (21.4 %) in savanna and grasslands. Soils of agricultural ecosystems contain about 147 Pg (6.4 %) of the total C pool. The ratio of soil-C:plant-C in natural ecosystems is 77 in extreme desert, 42 in croplands, 30 in tundra, 16 in wetlands, 15 in temperate grasslands, 2 in boreal forest, and about 1 in tropical rain forest (Table 10.1). A judicious management of the terrestrial biosphere C pool to reduce atmospheric CO 2 abundance has been widely suggested (Dyson, 1977; Read, 2008).

10.3

Emissions from agricultural versus other activities

Since the dawn of settled agriculture about 10 000 years ago, agriculture has been a source of GHGs (Ruddiman, 2003, 2005). Until the 1940s, more C was emitted from land use conversion and soil cultivation than from fossil fuel combustion. Total CO2–C emitted from terrestrial sources from the dawn of settled agriculture until 1750 is estimated at ~320 Pg (Ruddiman, 2003). Table 10.1 Estimates of the terrestrial carbon pool in different biomes (adapted from USDOE, 1999; Amthor et al., 1998; Roy et al., 2001; Prentice et al., 2001; Houghton, 2003a,b) Biome Climate

Plant-C pool (Pg)

Soil-C pool (Pg)

Soil-C: Total (Pg) plant-C

Forest

Tropical Temperate Boreal

212–340 (244) 214–340 (277) 59–139 (99) 100–262 (181) 57–88 (72) 135–150 (142)

1.13 1.83 2.0

521 280 214

Woodland

Temperate Chaparral

14–18 (16) 7–9 (8)

22–26 (24) 28–32 (30)

1.50 3.75

40 38

Savanna/ Temperate Grasslands Tropical

9–23 (16) 66–79 (72)

176–295 (235) 247–264 (255)

14.70 3.5

251 327

Tundra Wetlands

Arctic/alpine Natural Peatlands

2–6 (4) 12–15 (14) 0

115–121 (118) 200–240 (220) 455

29.5 15.7 –

122 234 455

Desert

Scrub/semi desert 8–10 (9) Extreme desert 0.2–0.4 (0.3)

159–199 (179) 22–24 (23)

20.0 76.7

188 23.3

41.9

151

Cropland/ Managed permanent crops

3–4 (3.5)

128–165 (147)

—————

———————

———

Total

558

2286

2844



277

At present, about 13.5 % of the total emissions (~1.6 Pg/y) are attributed to agricultural activities. With as much as 11.5 Pg of total annual emissions from all anthropogenic activities in 2000s (Koonin, 2008), approximately 34 % of the total emissions are attributed to non-energy sources such as waste, agriculture, and land use change (IPCC, 2007a; Holdren, 2008; Koonin, 2008). The data in Table 10.2 show absolute and relative contributions of CO2E (E refers to equivalent) from diverse sources. The largest emission of 25.9 % is from energy, with 19.4 % from industry, 13.1 % from transportation, 7.9 % from residential and commercial buildings, and 2.8 % from waste and wastewater treatment (Table 10.2). From 1750–2002, 292 Pg C has been emitted from fossil fuel combustion, and an additional 200 Pg is projected to be emitted between 2003 and 2030 (Holdren, 2008). In comparison, emissions from land use conversion and agricultural activities between 1850 and 2008 are estimated at about 150 Pg. Thus far (until 2008), total emissions may be as much as ~470 Pg from terrestrial ecosystems compared to ~350 Pg from fossil fuel combustion. Therefore, terrestrial ecosystems in general, but agricultural ecosystems in particular, have been a major source of GHG emissions into the atmosphere.

10.3.1 Emissions from different farm operations Several farm operations are highly C-intensive. Agricultural activities currently emit 1.7 Pg C/y (Smith et al., 2008). Cultivation of drained peat soils is a major source of C emission globally. The data in Table 10.3 show that plow-based tillage methods, heavy applications of nitrogenous fertilizers, Table 10.2 Relative emissions from agriculture and forestry compared with other anthropogenic activities (recalculated from IPCC 2007a; Holdren, 2008; Koonin, 2008) Emission (CO2–C Relative emissions equivalent) Pg C (%)

Source/activity I. Energy sources 1. Energy/power supply 2. Industry 3. Transport 4. Residential and commercial buildings Subtotal

2.98 2.23 1.51 0.91 7.63

25.9 19.4 13.1 7.9 66.3

II. Non-energy sources 1. Forestry 2. Agriculture 3. Waste and wastewater Subtotal

2.00 1.55 0.32 3.87

17.4 13.5 2.8 33.7

Grand total

11.5


100

278

Developments and innovation in CCS technology Table 10.3 Estimates of carbon emissions from farm operations (adapted from Lal, 2004) Farm operations

Equivalent carbon emissions (kg CE/ha)

I.

Seed bed preparation 1. Moldboard plowing 2. Chiseling 3. Discing 4. Subsoiling 5. Cultivation 6. Rotary hoeing

15.2 7.9 8.3 11.3 4.0 2.0

II.

Fertilizers 1. Nitrogen 2. Phosphorus 3. Potassium 4. Lime

1.3 ± 0.3 0.2 ± 0.06 0.15 ± 0.06 0.16 ± 0.11

III.

Pesticides 1. Herbicides 2. Insecticides 3. Fungicides

6.3 ± 2.7 5.1 ± 3.0 3.9 ± 2.2

IV.

Other operations 1. Knife down ammonia 2. Spray herbicide 3. Plant/sow/drill 4. No-till planting 5. Combine harvesting 6. Corn silage 7. Fertilizer spreading 8. Forage harvesting 9. Raking

10.1 1.4 3.2 3.8 10.0 19.6 7.6 13.6 1.7

± ± ± ± ± ±

± ± ± ± ± ± ± ± ±

4.1 2.3 2.5 2.8 1.9 0.9

1.5 1.3 0.8 0.1 1.5 6.4 2.5 4.4 0.7

pest control through reliance on chemicals, combine-harvesting, and silage harvesting operations are a major source of CO2 emissions. In view of high emissions from soil tillage, nitrogenous fertilizers, and pesticides, it is imperative that sustainable systems of soil, crop, and weed management be identified to enhance use efficiency, reduce the hidden C costs, and minimize the C footprint of agricultural activities.

10.3.2 Emission reduction from land use conversion and agricultural activities Land use conversion, involving deforestation and biomass burning, along with drainage and cultivation of peat soils are among the major sources of GHGs. Therefore, identifying strategies of emission avoidance from terrestrial and related ecosystems is extremely important. The goal is to intensify agronomic production from soil and water resources already committed to



279

agricultural land use (Ausubel, 2001), so minimizing the need for conversion of natural ecosystems to agricultural land use. The pressure on finite world soils to produce more per unit area and input will exacerbate because of the adverse effects of projected climate change on crop yields (Schimel, 2006). Worldwide mean cropland area needed to feed one person decreased from 0.47 ha in 1950 to 0.27 ha in 2000 because of agricultural intensification. The per capita land area needed to feed one person is projected to decrease to 0.15 ha by 2050. These projections are based on the assumption of expected increase in average crop yield by 2 %/y (Richards, 1990). Adoption of improved agricultural practices since 1950s has increased crop yields substantially. For example, by tripling yield of wheat (Triticum aestivum) since the mid-1960s, India’s wheat farmers have spared about 50 million ha or about 80 % of the present area of India’s woodland (Waggoner, 1996). Sparing land by agricultural intensification and restoring these surplus lands by afforestation is a win–win strategy. Land restoration through afforestation can also enhance the terrestrial C pool in both plant C and soil C components. In addition, there are numerous other ecosystem services of restored land, especially with regard to improvements in water quality and enhancement in biodiversity.

10.4

Basic principles of carbon sequestration in terrestrial ecosystems

Green plants use solar energy to convert atmospheric CO2 into organic compounds, and use this chemical energy through respiration to support metabolic processes. The difference between gross photosynthesis and respiration is called ‘net primary productivity’ (NPP). Increase in plant C and soil C pools depends on NPP, which is strongly influenced by the biome and, more specifically by soil type and the climate. The global NPP is estimated at ~65 Pg C/y (Table 10.4). Of the global total, NPP is 29.6 Pg C/y (45.9 %) for the forest biome and 20.6 Pg C/y (31.9 %) for the savanna/ grassland ecosystem. A portion of the NPP converted into woody biomass accounts for long-term sequestration of atmospheric CO2 into the plant C pool. Some of the detritus material and leaf litter is converted into the soil C pool. A part of the below-ground biomass, especially the fine roots and roots exudates are also converted into the soil C pool.

10.4.1 Soil C density and pool The data in Table 10.5 show estimates of C pool in world soils. Of the total soil C pool of 2466 Pg, C storage in different soils is estimated at 515 Pg (20.9 %) in Aridisols, 353 Pg (14.3 %) in Entisols, 323 Pg (13.1 %) in Gelisols, 237 Pg (9.6 %) in Mollisols, 224 Pg (9.1 %) in Inceptisols, 201 Pg (8.2 %) in


280


Table 10.4 Net primary productivity of different biomes (recalculated from Prentice et al., 2001; Houghon, 2003a,b; USDOE, 1999; Seiler and Crutzen, 1980; Whitaker and Likens, 1975) Biome Climate Area (109 ha)

Net primary productivity (Pg C/y)

% of total

Forest

Tropical Temperate Boreal

1.75 1.04 1.37

13.7–21.9 (20.1) 5.0–8.1 (6.6) 2.6–3.2 (2.9)

31.1 10.2 4.4

Woodland

Temperate Chaparral

0.2 0.25

1.3–1.5 (1.4) 0.8–1.0 (0.9)

2.1 1.4

Savanna/grassland Temperate Tropical

1.50 2.76

1.4–7.0 (4.2) 14.9–17.8 (16.4)

6.5 24.4

Tundra

Arctic/alpine

0.95

0.5–1.0 (0.75)

1.1

Desert

Scrub/semi-desert 2.77 Extreme desert 0.9

1.4–3.5 (2.5) 0–0.2 (0.1)

3.8 0.15

Wetlands

Natural Peatlands

3.3–4.3 (3.5) –

5.4 –

4.1–6.8 (5.5) ———————

8.5 ——

0.35 ?

Croplands/ Managed permanent crops

1.35 ———

Total

15.2

49.0–76.3 (64.5)

100

Alfisols, l80 Pg (7.3 %) in Histosols, 137 Pg (5.6 %) in Ultisols, and 126 Pg (5.1 %) in Oxisols (Table 10.5). The relatively large C pool in Aridisols is attributed to a high concentration of the SIC component. In contrast to the high SIC pool in Aridisols, the SOC component predominates in soils of the humid regions (e.g., Oxisols, Ultisols, Gelisols, Histosols and Inceptisols). The SOC pool is also expressed in terms of C density as kg/m2 (Table 10.6). The highest SOC density is observed in soils under boreal forest with a range of 9.8–10.2 kg/m2 for 0–0.3 m depth and 23.1–24.0 kg/m2 for 0–1.0 m depth. The lowest SOC density is observed in soils of arid regions ranging from 2.0–2.2 kg/m2 for 0–0.3 m depth and 3.7–4.1 kg/m2 for 0–1.0 m depth. High SOC density is also observed in soils of polar and alpine regions, and in those of temperate climates (Table 10.6). The SOC density of 10 kg/m2 is equivalent to a total C pool of 100 Mg/ha. Thus, SOC density of 23.1–24.0 kg/m2 in soils under boreal forests is equivalent to a C pool of 231–240 Mg/ha. The SOC density depends on a multitude of interacting factors (Jenny, 1941; 1961). Important among these are climate, terrain, soil, vegetation, and management. The SOC density increases with increase in mean annual precipitation and decrease in mean annual temperature (Jenny, 1980b). The SOC density also increases with the increase in clay content. Furthermore, landscape position and terrain characteristics also impact SOC density. Poorly drained soils located at foot-slope positions are generally



281

Table 10.5 Estimates of carbon pool in world soils (adapted from Eswaran et al., 2000) Soil order Alfisols Andisols Aridisols Entisols Gelisols Histosols Inceptisols Mollisols Oxisols Rocky Land Shifting Sand Spodosols Vertisols Total

Area 9

10 ha 1.26 0.09 1.57 2.11 1.13 0.15 1.29 0.90 0.98 1.31 0.53 0.34 1.11 0.32 13.09

Soil-C pool (Pg)

% of total

SOC

SIC

Total

158 20 59 90 316 179 190 121 126 – – 64 137 42

43 0 456 263 7 1 34 116 0 – – 0 0 22

201 20 515 353 323 180 224 237 126 – – 64 137 64

1526

940

2466

9.6 0.7 12.0 16.2 8.6 1.2 9.8 6.9 7.5 10.0 4.1 2.6 8.4 2.4 100

% of total 8.2 0.8 20.9 14.3 13.1 7.3 9.1 9.5 5.1 – – 2.6 5.5 2.6 100

Table 10.6 Soil organic carbon density in different agroecological zones (adapted from Batjes, 1999) Agroecological zone

Mean soil organic carbon density (kg/m2)

0–0.3 m depth

Arid Boreal Polar and alpine (excl. ice) Sub-tropics (summer rains) Sub-tropics (winter rains) Temperate (oceanic) Temperate (continental) Tropics (warm humid) Tropics (warm seasonally dry) Tropics (cool)

2.0–2.2 3.7–4.1 9.8–10.2 23.1–24.0 7.0–7.8 20.6–23.2 4.5–4.7 8.6–9.1 3.6–3.9 7.2–8.0 5.8–6.4 11.7–12.9 5.6–5.9 10.8–11.3 5.2–5.4 10.0–10.4 3.6–3.8 7.0–7.3 4.4–4.7 8.4–8.9

0–1.0 m depth

characterized with higher SOC density than well-drained soils located on summit or shoulder slopes, and north-facing slopes may have higher SOC density than south-facing slopes because of differences in soil temperature and moisture regimes.

10.4.2 Plant C pool The pedologic or soil C pool depends also on the above-ground and belowground biomass C and on the input of biomass C into the ecosystem. Thus,


282


the SOC pool increases with increase in input of the biomass C (Equation 10.2):

gain in soil C: input of biomass C > output of C

[10.1]

loss in Soil C: input of biomass C < output of C

[10.2]

The output of C from an ecosystem, over and above the respiration, is due to decomposition, erosion, and leaching. The input of biomass C in natural ecosystems is from the leaf litter, detritus material, and the root biomass. In contrast to natural ecosystems, the input of C in cropland soils is from the residues returned to the soil. Conversion of plow tillage to notill (NT) farming, in conjunction with the use of growing cover crops in the rotation cycle and adoption of the strategy of integrated nutrient management (INM), is another technique for enhancing the input of biomass C into an ecosystem. Input of root biomass is an important factor determining the SOC density and pool. Total root biomass in all biomes is estimated at about 295 Pg (Table 10.7). While woody roots contribute to the total ecosystem C pool, the relative proportion of the fine root biomass strongly impacts the SOC dynamics. The proportion of fine root biomass varies from 100 % in cropland soils to 97 % in temperate grasslands. High total amount (14 Pg) along with proportion of fine roots (97 %) lead to high SOC density in temperate grasslands (Table 10.6) and the total SOC pool on a global basis (Table 10.1). In addition, the live fine root biomass plays an important role in SOC turnover (Table 10.7). Among natural ecosystems, the relative proportion of live fine root biomass is also high in temperate grasslands (60.7 %) followed by that in savannas Table 10.7 Estimates of total root biomass in major world ecosystems (adapted from Jackson et al., 1997) Ecosystem

Total root Fine root biomass (Pg) biomass (Pg)

Live fine root biomass Pg

% of total

1. 2. 3. 4. 5. 6. 7.

Cultivated lands 2.1 Desert 6.6 Forests ∑ Boreal 35.0 ∑ Temperate deciduous 29.0 ∑ Temperate evergreen 22.0 ∑ Tropical rainforest 83.0 ∑ Tropical seasonal 31.0 Savannas 21.0 Temperate grasslands 14.0 Tundra/alpine 10.0 Woodland and shrubland 41.0

2.1 4.9

1.1 2.3

52.3 34.8

7.2 5.6 4.1 9.7 4.3 14.9 13.6 7.7 4.4

2.8 3.1 2.5 5.7 2.1 7.7 8.5 2.7 2.4

8.0 10.7 11.4 6.9 6.8 36.7 60.7 27.0 5.9

Total

78.5

41.2

14.0

294.7



283

(36.7 %) and tundra (27.0 %). Cultivated lands also have a high proportion of the live fine root biomass. The rate of humification of the root biomass depends on the composition of fine roots, and especially of the C:N, C:P, and C:S ratios. The data in Table 10.8 show that the mean elemental ratios are about 40:1 for C:N, 450:1 for C:P, and 550:1 for C:S. In contrast, these elemental ratios in soil humus are about 12.1 for C:N, 50:1 for C:P, and 70:1 for C:S (Himes, 1998). Thus, conversion of biomass C into humus requires additional nutrients. Total elemental pool in the fine roots is estimated at 38 Pg C, 0.9 Pg N, 0.09 Pg P, and 0.07 Pg S. About 50 % of these elements are contained in living fine roots (Table 10.8). Both soil C and plant C pools are likely to be affected by the projected increase in temperature, and by the CO2-fertilization effect as the rate of photosynthesis increases with the increase in atmospheric concentration. Plants differ in their response to CO2 concentration. For example, corn (Zea mays) reaches a saturation level at lower concentration of CO2 than wheat (Triticum aestivum) (Akita and Moss, 1973). The response to elevated CO2 concentration differs between C3 and C4 plants. Examples of C4 plants which reach saturation at lower CO2 concentration are corn and sugarcane (Saccharum ofjicinarum L). In comparison, examples of C3 plants which reach saturation at higher CO2 concentration are wheat, rice (Oryza sativa), alfalfa (Medicago sativa), cotton (Gossipium hirsutum) and potato (Solanum tuberosum). Kimball (1983) reported that crop yield could increase 33 ± 9 % with doubling of CO2 concentration from 330 ppm to 660 ppm. Experiments using the ‘free-air carbon dioxide enrichment’ (FACE) technique (Hendrey and Kimball, 1994) have shown that yield of cotton grown at 550 ppm CO 2 could increase by 43 % (Mooney et al., 1994). Amthor (2001) observed increase in wheat yield by 31 % at 700 ppm of CO2. However, increase in temperature may also enhance the rate of respiration and decrease NPP. The increase in temperature may accentuate the rate of decomposition of SOM and increase losses through high risks of accelerated soil erosion and leaching. The CO2 fertilization effect may increase NPP provided that plant growth is not limited Table 10.8 Estimates of nutrients in fine roots (adapted from Jackson et al., 1997) Nutrient element

Concentration Global nutrient pool (Pg) (g/kg) Total fine roots Living fine roots

% of total fine roots

Carbon (C) Nitrogen (N) Phosphorus (P) Potassium (K) Calcium (Ca) Magnesium (Mg) Sulfur (S)

488.6 11.7 1.1 3.0 4.1 1.4 0.88

52.2 52.2 51.8 50.0 53.2 49.1 52.2

38.1 0.92 0.085 0.24 0.32 0.11 0.069

19.9 0.48 0.044 0.12 0.17 0.054 0.036


284


by deficiency of plant nutrients (especially N and P) and water. In addition to the total biomass amount, the minimum replacement rate (y) also differs among biomes. In annuals grown on croplands, the replacement time is 1.5 y compared with 20–25 y in forest ecosystems and 6–7 y in deserts and swamps. Short rotation wood perennials, grown for production of biofuel feedstock, have a short time of about 7–8 y.

10.3.3 Sequestration of carbon in soil Processes governing C sequestration in soil are different for SOC and SIC components (Fig. 10.4). The SOC sequestration occurs through humification of biomass C into recalcitrant humic substances and stabilization through several protective mechanisms. Conversion of biomass C returned to the soil (as litter, detritus material, crop residues, fine roots, and any other biomass applied to the soil as manure, mulch, or sludge) into stable humus and humic substances increases SOC concentration and pool. Thus, essential conditions for SOC sequestration are: (i) availability of biomass C, (ii) availability of N, P, S, and other elements, (iii) favorable temperature and moisture regimes conducive to appropriate microbial processes, and (iv) presence of clay (and silt) fractions to stabilize humic substances through formation of organomineral complexes. Despite the presence of optimal conditions, the rate of SOC sequestration may be low (Schlesinger, 1990; 1999). Availability of the principal building blocks of humus (C, N, P, S) is essential. Himes (1998) estimated that sequestration of 10 000 kg C into humus requires 28 000 kg C in plant residues, 62 000 kg of oven dry residues (C:N ratio of 12:1), 200 kg P (C:P ratio of 50:1) and 143 kg S (C:S ratio of 70:1). This will produce 17 241 kg of humus. Assuming that the soil weight is 2240 Mg/ha in the surface 15 cm layer, it will increase SOC concentration by 0.77 % (Himes, 1995). The impact of application of nitrogenous fertilizers on SOC sequestration depends on numerous other soil rotated factors (Khan et al., 2007). The humus thus formed can be stabilized against microbial processes by physical, chemical, and/or biological mechanisms. Physical stabilization processes include formation of stable micro-aggregates. The latter are formed through cementing together of domain and micro-aggregates by humic substances (Jastrow et al., 2007). No-till farming There are several agricultural practices which enhance SOC pool. Important among these is NT farming in conjunction with crop residue mulch and cover crops along with INM, collectively called ‘conservation agriculture.’ The global average rate of SOC sequestration with NT farming is 0.57 Mg/ha/y


Processes and mechanisms governing carbon sequestration in soils

Humification

Nutrient requirements

Aggregation

Microbial processes

Stable microaggregates

Weathering of rocks and minerals

Soil inorganic carbon

Illuviation

Per ascensum model

Stable macroaggregates

Per descensum model

Leaching of HCO3–

In situ model

Silicates

Biogenic model

Carbonates

285

10.4 Processes affecting carbon sequestration in soils.

Formation of secondary carbonates



Soil organic carbon

286


(West and Post, 2002). However, NT farming can sequester C only if it can be adapted to specific soils, crops, and climatic conditions and it creates a positive C budget. While there are numerous benefits of NT farming (e.g., erosion control, water conservation, saving in fuel, reduction in time and seedbed preparation), it may not improve crop yield or create a positive C budget in all soils, climates, and cropping systems (Baker et al., 2007; Blanco-Conqui and Lal, 2008). Retention of crop residue mulch is an essential component of NT farming. The rate of SOC sequestration is also more in clayey (heavy-textured) than sandy (light-textured) soils (Lal, 2008). There are numerous determinants of terrestrial C sequestration in soil and biota (Fig. 10.5). Thus it is difficult to specify either a single agricultural practice (such as NT farming) or a mean rate for a range of highly heterogeneous environments. The rate of C sequestration with NT farming may vary from a negative (~500 kg/ha/y) to a strongly positive (3 Mg/ha/y), depending on climate, soil, plant, terrain and management (Fig. 10.5). Pacala and Socolow (2004) estimated that 3 Pg C/y could be sequestered in pedologic and biotic pools through conversion to NT farming, afforestation in tropical and temperate environments, and establishment of biofuel plantations. Biochar Addition of biochar as soil amendment can also increase the SOC pool (Hayes, 2006; Morris, 2006; Woods et al., 2006). Charcoal is chemically stable (Brocksiepe, 2002), and is a recalcitrant material. This chemical stability is a useful criterion to make it a potential C sink. Seifritz (1993) estimated that 3 Mg of air-dried wood in an ignited charcoal pile can produce 750 kg of charcoal (Equation 10.3)

CH1.44 O0.66 (dry wood) + 0.43 O2 (from air)

Æ 0.6 C (charcoal) + 0.4 CO2 (gas) + 0.72 H2O (vapor)

[10.3]

The reaction in Equation 10.3 shows that 60 % of the C in the wood can be converted to charcoal, and the other 40 % is burnt to produce energy. The biochar thus produced can be used as a soil amendment (Lehmann et al., 2006). Peat soil restoration Drainage and cultivation of peat soils has been a major source of CO2. Thus, restoration of wetlands through flooding may be an important strategy for both avoiding emission and sequestering atmospheric CO2. Drained and cultivated peatlands are decomposing at the rate of 1–2 cm/y (Elder and Lal, 2008). While adoption of NT farming can decrease the rate of decomposition,


Determinants of terrestrial carbon sequestration

Plant

Terrain

Management

Precipitation

Texture

Species

Landscape position

Farming/ cropping systems

Temperature

Structure

Root:shoot ratio

Drainage

Plant species & varieties

Growing season

Mineralogy & cation exchange capacity

Root exudates

Slope gradient

Nutrient management

Light/ radiation

SOC and nutrient pools

NPP

Slope length

water management

Extreme events

Soil pH

Litter & its composition

Slope aspect & shape

Residues management & tillage

Carbon loss from the ecosystem

Soil

Ecosystem carbon, elemental, and water budgets

287

10.5 Ecosystem factors which determine the rate and magnitude of terrestrial carbon sequestration and its residence time.



Carbon input into the ecosystem

Climate

288


the most effective strategy is to reflood and restore the wetlands and make them a net C sink. Afforestation of degraded soils The capacity of natural sinks (e.g., terrestrial) is decreasing (Canadell et al., 2007), probably because of the increase in global soil degradation. Most degraded soils are severely depleted of their SOC and nutrient reserves. Therefore, restoring degraded soils and ecosystems can increase the terrestrial pool in both plant C and soil C components. The NPP of forest ecosystems may also be enhanced through the CO2 fertilization effect, provided that water and nutrients (N, P, S) are not limiting. Establishment of productive monoculture plantations on about 350 Mha of tropical regions can offset CO2 emissions by as much as 1 Pg C/y (Pacala and Socolow, 2004). Afforestation on a large scale, however, may adversely affect water resources and biodiversity. Biofuel/energy plantation Production of second-generation biofuels (e.g., cellulosic ethanol) is a high priority worldwide. Establishment of energy plantations by growing dedicated species (warm season grasses or short rotation woody perennials) can increase C sequestration in detritus material (plant C) and in the soil. Similar to the concerns about afforestation of degraded soils, however, largescale establishment of energy plantations would compete with agricultural production for water and nutrients. Production of 1–1.5 Pg of biomass in the USA per annum (Kennedy, 2007; Somerville, 2006) and 4–5 times as much in the world (Table 10.9) necessitates identification of appropriate land of a high soil quality to sustain the desired biomass productivity. Production of biofuel feedstock must not compete with that of food production for essential resources (e.g., land, water and nutrients). Furthermore, harvesting crop residues as biofuel feedstock is not conducive to soil C sequestration (Jenny, 1980a). Burying biomass Burying biomass C under anaerobic conditions, so that it does not decompose, has been suggested as an option to sequester atmospheric CO2. Two sources of biomass often suggested for burial are: (i) crop residues and (ii) the wood. Metzger and Benford (2001) estimated that 596 Tg of crop residues (corn, soybeans, wheat) produced in the USA contains 238 Tg C. Globally, C contained in the crop residues was estimated to be about 1 Pg. They argued that burial of crop residues in the oceans below the thermocline or in river



289

Table 10.9 Estimates of global biomass production (adapted from Whitaker and Likens, 1975, Erbach and Wilhelm, 2009) Biome Climate I. Terrestrial Forests Desert Wetlands Croplands Others

Biomass Minimum ———————————— replacement Amount (Pg) % of total rate (y)

Rainforest 765.0 Monsoon 262.5 Temperate evergreen 175.0 Temperate deciduous 210.0 Boreal 240.0 Mediterranean open 50.4 Scrub/semi-desert 12.6 Extreme desert, rocks, ice 0.5 Swamps and marsh 30.0 Lakes and streams 0.04 14.0 113.4

40.8 14.0 9.3 11.2 12.8 2.7 0.7 0.03 1.6 – 0.7 6.2

20.5 21.9 26.5 25.0 25.0 24.0 7.8 6.7 7.5 0.08 1.54 –.

Total continental

1873.4

100

16.23

II. Aquatic Marine Total marine

Open ocean Upwelling zones Continental shelf Algal beds and reef Estuaries and mangroves

1.0 0.01 0.27 1.20 1.40 3.87

25.8 0.2 7.0 31.0 36.0 100

0.02 0.04 0.03 0.80 0.67 0.07

Grand total

1877.3

11.02

deltas can reduce atmospheric CO2. Strand and Benford (2009) estimated annual crop residues production of 497 Tg in the USA and 4980 Tg in the world. Strand and Benford argued that ocean sequestration of crop residue C is 92 % efficient compared with 32 % efficiency in converting it into cellulosic ethanol and 15 % into soil humus. However, Karlen et al. (2009) argued strongly against removal of crop residues for ocean burial or other purposes because of numerous ecosystem services provided by its retention on the land (e.g., soil quality, reduction in non-point source pollution, habitat, and food for soil biota). Another suggestion for using biomass for C sequestration is via wood burial. Scholz and Hasse (2008) suggested that wood grown at suitable sites can be permanently buried in open coal pits. They argued that by so doing humans would give back to nature what they have used as fossil fuel, and it may be reused in hundreds or thousands of years time. Zeng (2008) estimated that sustainable long-term C sequestration potential for wood burial is 10 ± 5 Pg C/y, and argued that currently about 65 Pg of detritus material or dead wood on the world’s forest floors are suitable for burial. Of the technical potential of 10 Pg C, 4.2 Pg C/y is in tropical forest, 3.7


290


Pg C/y in temperate and 2.1 Pg C/y in the boreal forests. Zeng estimated the cost of wood burial at about $50/Mg C ($16/Mg CO2). Similar to the concerns about other uses of crop residues, wood is also an important fuel source (Richter et al., 2009). Secondary carbonates In addition to SOC, there is also sequestration of SIC as secondary carbonates. There are four mechanisms of formation of secondary carbonates (Fig. 10.4, Monger, 2002). Marion et al. (1985) suggested that secondary carbonates are formed through vertical translocation of DIC and its reprecipitation in the subsoil (per descensum model). In contrast, Sobecki and Wilding (1983) and Rabenhorst and Wilding (1986) proposed capillary rise of Ca2+ from the shallow water table and its precipitation in the surface layer with dissolved CO2 or carbonic acid (per ascensum model). Monger (2002) proposed a biogenic model of formation of secondary carbonates in which he documented that biotic activity (e.g., termites) enhances formation of secondary carbonates. The rate of formation of secondary carbonates is 5–10 Kg/ha/y (Landi et al., 2003), and is lower than that of SOC sequestration. However, leaching of bicarbonates in soils irrigated with good quality water may be much higher (500–1000 Kg/ha/y) (Lal, 2008).

10.5

Potential of terrestrial sequestration

The theoretical potential of C sequestration in terrestrial ecosystems is 6–10 Pg C/y (Table 10.10). Of this, the highest potential of 1–3 Pg C/y is through afforestation, followed by 0.8–1.3 Pg/C/y through restoration of degraded soils, 1.2 Pg C/y in rangelands, and about 0.8 Pg C/y in cropland soils (Lal, 2004). Transport of C in terrestrial sediments, through deposition and subsequent burial of C in depressional sites and aquatic ecosystems, remains a debatable Table 10.10 Technical potential of carbon sequestration in terrestrial ecosystems (adapted from USDOE, 1999) Ecosystem/biome

Technical potential (pg C/y)

Agricultural lands Biomass in croplands Desert and degraded lands Forests Grasslands Peat lands and wetlands (boreal) Rangelands Terrestrial sediments Total

0.85–0.90 0.5–0.80 0.8–1.30 1.0–3.0 0.5 0.1–0.7 1.2 0.7–1.7 5.65–10.1



291

issue (Lal, 2003; Van Oost et al., 2007). Although these estimates shown in Table 10.10 are old, the technical potential of the terrestrial biosphere reported by Read (2008) is also high. A widely accepted estimate of the technical potential of C sequestration in terrestrial ecosystems is about 3 Pg C/y, comprising 1 Pg C/y each in cropland, rangeland, and degraded soils (through afforestation) (Pacala and Socolow, 2004). However, the economic potential depends on the establishment of an effective market for trading C credits. Presently, soil C is not included in Article 3.3 of the Kyoto Treaty and cannot be considered under the Clean Development Mechanism (CDM) or Joint Implementation (JI). With acceptance of soil C and plant C under CDM and JI and adoption of an effective strategy of commodification of C so that it can be bought and sold as any other farm commodity, the economic potential can eventually be increased to the level of technical potential. Nevertheless, terrestrial sequestration remains an economically viable option compared with the high cost of engineering technologies of capture and storage. Implementation of an appropriate cap and trade system may also be essential to promote trading of C sequestered in terrestrial ecosystems.

10.6

Challenges of terrestrial sequestration

The potential of terrestrial sequestration in general and that of soil sequestration in particular has been debated since the early 1990s (Bouwman, 1990; Lal et al., 1995). Despite several decades of intense scientific research, there are numerous knowledge gaps which need to be addressed objectively and credibly. Important among these are the following:

10.6.1 Carbon sink capacity It is necessary to evaluate the C sink capacity of both plant C and soil C components for their theoretical, technical, economic, and realizable/ attainable sequestration potential. Assessment of C sink capacity must be done in relation to ecosystem/biome characteristics including climate, soil type, terrain, species, etc. (Fig. 10.5).

10.6.2 Technological options Once the sink capacity is known, soil-specific technological options must be evaluated to assess the rate, magnitude, and duration of C sequestration. No single technology can be universally applicable for all diverse soils, climates, crop/tree/plant species, and social/cultural and economic situations. The recent debate on application of the NT system (Baker et al., 2007) and nitrogen fertilization (Khan et al., 2007) is indicative of the lack of a strong database


292


with regard to soil and site-specific technologies for terrestrial sequestration. The goal is to identify those soil and plant management systems which create a positive ecosystem C budget by increasing C-input into the system more than C-output from the system (Equations 10.1 and 10.2). Relevant technological options may include those of soil management (e.g., residue management and tillage methods), water conservation and management (e.g., drainage and irrigation), fertility management (involving a judicious use of fertilizers in conjunction with biological N fixation through leguminous cover crops and liberal application of compost/manure and other biomass), and cropping/farming system management (such as crop rotations and cropping sequences including agroforestry). Soil-specific technologies are also needed for sustainable management of rangeland/grazing land soils and those of forestry/energy plantations. Rather than technique-centric, the strategy is to identify those management options which create positive ecosystem C and soil C budgets. A matrix of such technologies must be developed along with the rate of C sequestration as documented by an example outlined in Table 10.11. The low rates reported over the geologic timescale (Table 10.12) can probably be enhanced through adoption of best management practices (BMPs). In general, BMPs are those which create a positive C budget in soil and the ecosystem. Important among these are conservation agriculture, mulch farming, cover cropping, integrated nutrient management and manuring, use of biosolids such as biochar, and complex farming systems such as agroforestry.

10.6.3 Permanence In addition to assessing the soil-specific rate of C sequestration as outlined in the matrix of Table 10.11, the question of permanence must be addressed. There are several concerns about the residence time of C sequestered in terrestrial ecosystems. Residence time of C in soil and biota must be assessed in relation to management options, soil type, climate, and chemical composition of the residues and biomass returned to the soil. Permanence of terrestrial C must also be related to underlying processes including development of stable micro-aggregates, formation of organo-mineral complexes, biogenesis of recalcitrant compounds, illuviation of humic substances into the subsoil, precipitation of organic and inorganic compounds in specific horizons, transfer of carbonaceous materials into the groundwater, etc.

10.6.4 Measurement and monitoring Credible assessment of the pool and fluxes of C in terrestrial ecosystems to 1–m depth, for a 1–2 y period, over a landscape/watershed/farm in appropriate units (kg/ha/y) is essential to realizing the technical or economic potential


Table 10.11 A matrix of soil and ecosystem specific technological options based on a soil guide to create a positive carbon budget Rate of C sequestration (kg/ha/y)

Biome

Management options

Comments

1. Forest Temperate Alfisols

1. 2. 3. 4. 5. 6.

NT farming Cover cropping Integrated nutrient management Residue harvesting Manuring Biochar

Establishing empirical relations

2. Grasslands Tropics Inceptisols

1. 2. 3. 4.

Tillage methods Cropping systems Water conservation Ley farming

Developing pedotransfer functions

3. Wooded savanna Sub-Tropics Vertisols

1. 2. 3. 4.

Runoff management Tillage methods Soil fertility management Farming systems

4. Other ecosystems

Diverse systems

Climate

Sloping Lands

Soil type

Oxisols

Validating conceptual models

Assessing site-specific rates



293

294

Developments and innovation in CCS technology Table 10.12 Long-term rates of soil organic carbon sequestration since the Holocene (adapted from Schlesinger, 1990) Ecosystem

Accumulation period (103 y)

Sequestration rate (kg/ha/y)

Boreal forests Polar desert Temperate forest Temperate grasslands Tropical forests Tundra

3–5 8–9 2 9 4–8 9

57–117 2 7–25 22 23–25 11–24

of terrestrial sequestration. The techniques must be simple, routine, noninvasive and usable in situ under field conditions, and cost-effective. Such a reliable and credible system is essential to commodification of terrestrial C and trading of C credits.

10.6.5 Ancillary benefits In strong contrast to engineering techniques, terrestrial sequestration has numerous advantages through enhancement of ecosystem services. Important among these ancillary benefits is the increase in agronomic/biomass productivity through improvement in soil quality and use efficiency of inputs. While food demand is increasing and the available land area is finite (Wild, 2003), restoration of soil quality through increase in SOC pool is essential to advancing food security (Lal, 2006) and improvement in water quality. The latter is strongly impacted by decrease in water runoff and reduction in transport of dissolved and suspended loads from agricultural ecosystems.

10.6.6 Reference tables To promote trading of C credits, it is neither essential nor feasible to measure the soil-C pool or flux on every acre of land every year. Yet, it is important to develop reference tables about the rates(s) of soil C sequestration. Such tables, to be developed on the basis of informed opinions by appropriate stakeholders (e.g., scientists, land managers, policy-makers), must be created with consideration of climate, soil type, land use, farming/cropping systems, farm size, agronomic practices, and specific farming system components. Furthermore, these reference tables must be developed on the basis of the following criteria to validate the rate: simple, cost-effective, relevant to farmer’s perspective, based on land-use practice, wholly palatable to a region or community on the basis of soil type, relevant to other ecosystem services, and upgradable over time. An example of such a matrix is outlined in Table 10.11.



295

10.6.7 Soil C pool and climate change Being highly liable and dynamic, the active SOC pool is prone to decomposition with changing temperature and moisture conditions under projected global warming. Furthermore, even the inert/passive C pool may become active under changing environment, thereby converting most soils from a sink at present to a source in the near future (Davidson and Janssens, 2006; De Deyn et al., 2008). In this regard, melting of permafrost may create a large positive feedback.

10.7

Extrapolation

Scaling up the data of SOC sequestration from point scale to watershed, farm or regional scale is essential. Such an extrapolation must be done on the basis of soil type (Pedotransfer functions). There being a good relationship between climate and biome (Woodward et al., 2004), the relation between SOC pool and biome characteristics is also important. The impact of these factors on predominant mechanism/processes of SOC/SIC sequestration needs to be assessed. Extrapolation on a regional or global scale can lead to erroneous results as the underlying processes are not clearly understood (Rustad, 2006). Biological control of the terrestrial C sink must be clearly understood (Schulze, 2006).

10.7.1 Ecosystem carbon budget Credible evaluation of the net ecosystem C budget, in relation to land use and state factors (e.g., soil, climate, terrain vegetation), is essential to assess any changes in soil C pool and fluxes. Changes in the SOC pool are determined by the C balance, in view of the input and output (Equations 10.1 and 10.2), and by soil properties (Fig. 10.5).

10.7.2 Charcoal and fire Biomass burning by fire (natural or wild and managed) has an important impact on ecosystem C pool (Cochrane, 2003; Bond et al., 2005). Firegenerated charcoal is also important in the global C cycle. Thus, there is a need to understand the role of fire and charcoal SOC pool and dynamics.

10.7.3 Fine roots and turnover Roots and root exudates play a major role in SOC dynamics (Jackson et al., 1996; Schenk and Jackson, 2002; Matamala et al., 2003; Personeni and Loiseau, 2004; Bardgett et al., 2005; Dijkstra and Cheng, 2007). Careful


296


evaluation of C-input through roots and root exudates is essential to assessing SOC dynamics in diverse (especially forests) land use systems.

10.7.4 Role of soil biota Soil bacterial and mycorrhizal and fungal populations play an important role in SOC dynamics (Rillig and Mummey, 2006; Six et al., 2006). Thus, rates of SOC sequestration need to be assessed in relation to soil fauna and flora.

10.7.5 Plant characteristics Plant characteristics strongly impact the SOC pool and its dynamics (De Deyn et al., 2008). Recent advances in biotechnology and genetic engineering can be used advantageously to create plant types conducive to SOC sequestration, especially with regard to composition of some recalcitrant compounds.

10.7.6 All greenhouse gases About 40 % of the heat trapped by anthropogenic emissions is due to GHGs other than CO2, especially CH4 (Shine and Sturges, 2007). Agricultural practices emit CH4 (e.g., livestock farming, manure management, rice paddy cultivation) and N2O (use of nitrogenous fertilizers). Therefore, emission of these gases must also be considered in evaluating the net impacts.

10.7.7 The biofuel issue With interest in biofuels, it is important to critically and objectively assess whether biofuels (corn-derived ethanol and soybean- or palm oil-based biodiesel) are really C-neutral or C-negative? It is a debatable issue and some have argued that set-asides can be better climate investments than corn ethanol (Piñeiro et al., 2009).

10.8

Soil and terrestrial carbon as indicators of climate change

The magnitude and risks of projected climate change are important drivers of the enhanced interest in terrestrial sequestration. In addition to promoting C trading for creating another income stream for land managers, realization of numerous ancillary benefits and ecosystem services of increasing the terrestrial C pool can also be achieved by creating awareness about using changes in soil and biotic C pools as indicators of climate change. While melting of ice caps and glaciers in tundra and alpine regions has enhanced awareness of the potential threats of global warming, there are numerous



297

advantages of using the soil C pool and its dynamics as indicators of past and future climate change. There are several favorable attributes of soil C that make it a relevant indicator of climate change (Table 10.13). Important among these are: (i) the soil C pool is quantifiable over time and space and lends itself to repeated measurement for the same site, (ii) it is very sensitive to both land use/management as well as climate change, (iii) it is affected by soil forming or state factors, (iv) it has a memory in terms of natural and anthropogenic perturbations, (v) its pathways over the landscape can be traced through new and emerging technologies, and (vi) the uncertainty in its measurement can be assessed as well as minimized through appropriate sampling and statistical technique.

10.9

Conclusions

Carbon sequestration in terrestrial ecosystems, comprising soils and biota, is one of the strategies to stabilize atmospheric concentration of CO2 and Table 10.13 Reasons for soil carbon pool and fluxes as indicators of climate change Parameter

Attribute

1. 2. 3.

A soil constituent Measurement Quantification over time and space

4. 5.

Temporal changes for a specific management Ancillary benefits

It is a familiar property (Jenny, 1941; Albrecht, 1938). It can be measured directly (Nelson and Sommers, 1996). SOC pool can be measured in four dimensions (e.g., length, width, depth, time) (Nelson and Sommers, 1996; Wielopolski, 2006). SOC pool can be measured over time for the same site (Cremers et al., 2001; Ebinger et al., 2003; Wielopolski, 2006). It is a determinant of several ecosystem services (Feller et al., 2006; Janzen, 2006; Lal, 2008). SOC is an important soil forming factor (Jenny, 1961, 1980b). It is an important determinant of soil quality (Doran and Jones, 1996; Gregorich and Carter, 1997). Land use and soil/crop management strategy impact SOC pool and quality. SOC pool has well defined characteristics (e.g., surface area, charge density) (Brady and Weil, 2007). It can be used in synergism with other indicators (e.g., temperature) (Feller et al., 2006). The degree of uncertainty can be quantified (Beckett and Webster, 1972). Pathways and fate across the landscape can be followed (Lal, 2003). It is an important indicator of paleoclimate (Paul et al., 1997; Trumbore, 2000) SOC pool is sensitive to climate change (Bouwmann, 1990; Davidson and Janssens, 2006).

6. Soil formation 7. Soil quality 8. Memory 9. Characteristics 10. Synergism 11. Uncertainty 12. Pathway 13. Paleoclimate 14. Climate change


298


other greenhouse gases. Both soil C and plant C pools are highly reactive and dynamic, and strongly interact with the atmosphere through natural processes of photosynthesis, respiration, decomposition, humification, aggregation, erosion, and leaching. The technical potential of terrestrial sequestration is estimated at about 3 Pg C/y, one each in agricultural soils, afforestation and biofuel plantations, and restoration of degraded/desertified soils. There are strong ancillary benefits of terrestrial sequestration with regard to numerous ecosystem services such as increase in agronomic productivity, progress in achieving food security, improvement in water quality, and enrichment in biodiversity. Land use and soil/vegetation management practices which create a positive C budget and increase in the ecosystem C pool (e.g., mulch farming, integrated nutrient management, conversion to a perennial land use) need to be identified for soil-specific situations. Furthermore, processes, factors, and causes of C sequestration need to be identified and scaling techniques developed to extrapolate the data to farm, state, region, or national scale. Soil and plant attributes, which enhance performance of C sequestered in terrestrial ecosystems, need to be optimized. A multidisciplinary approach is needed to assess technical, economic, achievable, and current potential for site-specific environments. Terrestrial sequestration is a truly win–win strategy because it offsets anthropogenic emissions while improving the quality of soil and water resources and enhancing the environment.

10.10 References Akita S and Moss ON (1973) ‘Photosynthetic response of CO2 and light by maize and wheat leaves adjusted for constant stomatal apertures’, Crop Sci, 13, 234–237. Albrecht WA (1938) ‘Loss of soil organic matter and its restoration’, Soils and Men, USDA Yearbook of Agriculture #7, 347–360. Amthor JS (2001) ‘Effects of atmospheric CO2 concentration on wheat yield: review of results from experiments using various approaches to control CO2 concentration’, Field Crops Res, 73, 1–34. Amthor JS and Huston MA et al. (1998) Terrestrial Ecosystem Responses to the Global Change: A Research Strategy, ORNL/TM-1998/27, Oakridge, TN, Oakridge National Laboratory. Amundson R (2001) ‘The carbon budget in soils’, Annu Rev Earth Planet Sci, 29, 535–562. Ausubel JH (2001) ‘On sparing farmland and spreading forest’, in Clark T and Staebler R (eds), Forestry at the Great Divide; Proceedings of the Society of American Foresters Convention, Bethesda, MD, 127–138. Baker JM, Ochsner TE, Venterea RT and Griffis TJ (2007) ‘Tillage and soil carbon sequestration: what do we really know?’, Agric Ecosyst Environ, 118, 1–5. Baldock JA and Nelson PN (1998) ‘Soil organic matter’, in Sumer M (ed) Handbook of Soil Science, Boca Raton, FL, CRC Press, 25–84. Bardgett RD, Bowman WD, Kaufmann R and Schmidt SK (2005) ‘A temporal approach to linking aboveground and belowground ecology’, Trends Ecol Evol, 20, 634–641.



299

Batjes NH (1996) ‘TAI carbon and nitrogen in soils of the world’, Eur J Soil Sci, 47, 151–163. Batjes NH (1999) Management options for reducing CO2 concentrations in the atmosphere by increasing carbon sequestration in the soils, Tech. paper #30, International Soil Reference and Information Center (ISRIC), Wageningen, the Netherlands. Beckett PHT and Webster R (1972) ‘Soil variability: a review’, Soils & Fertilizers, 34, 1–15. Blanco-Conqui H and Lal R (2008) ‘No-tillage and soil carbon sequestration: an on-farm assessment’, Soil Sci Soc Am J, 72, 693–701. Bond WJ, Woodward FI and Midgley GF (2005) ‘The global distribution of ecosystems in a world without fire’, New Phytol, 165, 525–538. Bouwman AF (ed.) (1990) Soils and the Greenhouse Effect, Chichester, Wiley. Brady N and Weil R (2007) Nature and Properties of Soils, New York, Macmillan. Brocksiepe HG (2002) ‘Charcoal’, in Ullman’s Encyclopedia of Industrial Chemistry, Weinheim, Germany, Wiley-VCH Verlag GmbH & Co. Broecker WS (2008) ‘CO2 capture and storage: possibilities and perspectives’, Elements, 4, 295–297. Canadell JG, Le Quene C, Raupach MR et al. (2007) ‘Contributions to accelerating atmospheric CO2 growth from economic activity, carbon intensity and efficiency of natural sinks’, PNAS, 104, 18866–18870. Cochrane MA (2003) ‘Fire science for rainforests’, Nature, 421, 913–919. Cremers DA, Ebinger MH, Breshears DD, Unkefer PJ, Kammerdiener SA, Ferris MJ, Katlett KM and Brown JR (2001) ‘Measuring total soil carbon with laser-induced breakdown spectroscopy (LIBS)’, J Environ Qual, 30, 2202–2206. Davidson EA and Janssens IA (2006) ‘Temperature sensitivity of soil carbon decomposition and feedbacks to climate change’, Nature, 440, 165–173. De Deyn GB, Cornelissen JHC and Bardgett RD (2008) ‘Plant functional traits and soil carbon sequestration in contrasting biomes’, Ecol Lett, 11, 516–531. Dijkstra FA and Cheng WX (2007) ‘Interactions between soil and tree roots accelerate long term soil carbon decomposition’, Ecol Lett, 10, 1046–1053. Dixon RK, Brown S, Houghton RA, Solomon AM, Trexler MC and Wisniewski J (1994) ‘Carbon pools and flux of global forest ecosystems’, Science, 263, 185–190. Doran JW and Jones AJ (eds) (1996) Methods for Assessing Soil Quality, SSSA Special Publ #49, Madison, WI. Dyson PJ (1977) ‘Can we control the carbon dioxide in the atmosphere?’, Energy, 2, 287–291. Ebinger MH, Norfleet ML, Breshears DD, Cremers DA, Ferris MJ, Unkefer PJ, Lamb MS, Goddard KL and Meyer CW (2003) ‘Extending the applicability of laser-induced breakdown spectroscopy for total soil carbon measurement’, Soil Sci Soc Am J, 67, 1616–1619. Elder JW and Lal R (2008) ‘Tillage effects on gaseous emissions from an intensivelyfarmed organic soil in north central Ohio’, Soil Till Res, 98, 45–55. Erbach DC and Wilhelm WC (2009) ‘Bioenergy: Energy sources and costs from agriculture’, in Pond WG, Nichols BL and Brown DL (eds), Adequate Food for All: Culture, Science and Technology of Food in the 21st Century, Boca Raton, FL, 263–281. Eswaran H, Reich PF, Kimble JM, Beinroth FH, Padmanabhan E and Moncharoenm P (2000) ‘Global carbon stocks’, in Lal R, Kimble JM, Eswaran H and Stewart BA, Global Climate Change and Pedogenic Carbonates, Boca Raton, FL, Lewis, 15–23. Feller C, Manlay R, Swift MJ and Bernoux M (2006) ‘Functions, services and value of


300


soil organic matter for human societies and the environment: a historical perspective’, in Frossard E, Blum WEH and Warkentin BP (eds), Functions of Soils for Human Society and the Environment, Geo Soc Lond, Special Publications, 266, 9–22. Gregorich EG and Carter MR (eds) (1997) Soil Quality for Crop Production and Ecosystem Health, Elsevier, Amsterdam. Hayes MHB (2006) ‘Biochar and biofuels for a brighter future’, Nature, 443, 144. Hendrey GR and Kimball BA (1994) ‘The FACE program’, Agriculture and Forest Meteorology, 70, 3–14. Himes FL (1998) ‘Nitrogen, sulfur and phosphorus and the sequestering of carbon’, in Lal R, Kimble JM, Follett RF and Stewart BA (eds), Soil Processes and the Carbon Cycle, Boca Raton, FL, CRC Press, 315–320. Holdren JP (2008) Meeting the Climate Change Challenge, National Council for Science and Environment, Washington DC. Houghton RA (2003a) ‘The contemporary carbon cycle’, in Schlesinger WH (ed.), Biogeochemistry, Oxford, UK, Elsevier-Pergamon, 473–513. Houghton RA (2003b) ‘Why are estimates of the terrestrial carbon balance so different?’, Global Change Biol, 9, 500–509. IPCC (2007a) Climate Change 2007: The Physical Science Basis, Cambridge, UK, Cambridge University Press. IPCC (2007b) Climate Change 2007: Mitigation of Climate Change, Cambridge, UK, Cambridge University Press. Jackson RB, Canadell J, Ehleringer JR, Mooney HA, Sala OE and Schulze (1996) ‘A global analysis of root distributions for terrestrial biomes’, Oecologia, 108, 389–411. Jackson RB, Mooney HA and Schulze ED (1997) ‘A global budget for fine root biomass, surface area, and nutrient contents’, PNAS, 94, 7362–7366. Janzen HH (2006) ‘The soil carbon dilemma: shall we hoard it or use it?’, Soil Biol Biochem, 38, 419–424. Jastrow JD, Amonette JE and Bailey VL (2007) ‘Mechanisms controlling soil carbon turnover and their potential application for enhancing carbon sequestration, Climatic Change, 80, 5–23. Jenny H (1941) Factors of Soil Formation: A System of Quantitative Pedology, New York/London: McGraw-Hill. Jenny H (1961) ‘Derivation of state factor equations of soils and ecosystems’, Soil Sci Soc Am Proc, 25, 385–388. Jenny H (1980a) ‘Alcohol or humus?’, Science, 209, 444. Jenny H (1980b) The Soil Resource: Origin and Behavior, New York, Springer-Verlag, 37, 377. Jobbágy EG and Jackson RB (2000) ‘The vertical distribution of soil organic carbon and its relation to climate and vegetation’, Ecol Appl, 10, 423–426. Karlen DL, Lal R, Follett RF, Kimble JM, Hatfield JL, Miranowski JM, Camberdella CA, Manale A, Doran J, Baker JM and Rice CW (2009) ‘Crop residues, the rest of the story’ Env Sci Tech, 43, 8011–8015. Kennedy D (2007) ‘The biofuel conundrum’, Science, 316, 515. Khan SA, Mulvaney RL, Ellsworth TR and Boast CW (2007) ‘The myth of nitrogen fertilization for soil carbon sequestration’, J Environ Qual, 36, 1821–1832. Kimball BA (1983) ‘Carbon dioxide and agricultural yields: an assemblage and analysis of 430 prior observations’, Agron J, 75, 779–788. Koonin SE (2008) ‘The challenge of CO2 stabilization’, Elements, 4, 294–295. Lal R (2003) ‘Soil erosion and the global carbon budget’, Environ Int, 29, 437–450.



301

Lal R (2004) ‘Carbon emission from farm operations’, Environ Int, 30, 981–990. Lal R (2006) ‘Enhancing crop yield in developing countries through restoration of soil organic carbon pool in agricultural lands’, Land Degrad Dev, 17, 197–209. Lal R (2008) ‘Carbon sequestration’, Phil Trans R Soc, 363, 815–830. Lal R, Kimble JM, Levine E and Stewart BA (eds) (1995) Soils and Global Change, Boca Raton, CRC Press. Landi A, Mermut AR and Anderson DW (2003) ‘Origin and rate of pedogenic carbonate accumulation in Saskatchewan soils, Canada’, Geodermia, 29, 437–450. Lehmann J, Gaunt J and Rondon M (2006) ‘Bio-char sequestration in terrestrial ecosystem–a review’, Mitigation and Adaptation Strategies for Global Change, 11, 403–427. Marion GM, Schlesginger WH and Fonteyn PI (1985) ‘CALDEP: A regional model for soil CaCO3 (caliche) deposition in southwestern deserts’, Soil Sci, 139, 468–481. Matamala R, Gonzalez-Meler MA, Jastrow JD, Norby RJ and Schlesinger WH (2003) ‘Impacts of fine root turnover on forest NPP and soil C sequestration potential’, Science, 302, 1385–1387. Metzger RA and Benford G (2001) ‘Sequestering of atmospheric carbon through permanent disposal of crop residue 2001’, Climatic Change, 49, 11–19. Monger JD (2002) ‘Pedogenic carbonates: Link between biotic and abiotic carbonates’, Symph #20, Oral Paper #891, 17th World Cong Soil Sci, 14–20 August, Bangkok, Thailand. Mooney JR, Kimball BA, Pinter PJ Jr, LaMorte RL, Lewin KF, Nagy J and Hendre GR (1994) ‘Growth and yield of cotton in response to a free carbon-dioxide enrichment (FACE) environment’, Agric For Mateoral, 70, 49–67. Morris E (2006) ‘Black is the new green’, Nature, 442, 1878–1888. Nelson DW and Sommers LE (1996) ‘Total carbon organic carbon, and organic matter’, in Sparks DL et al., (eds) Chemical Methods, SSSA Book Series 5, Madison, WI, Chapter 34. Pacala S and Socolow R (2004) ‘Stabilization wedges: solving the climate problem for the next 50 years with current technologies’, Science, 305, 968–972. Paul EA, Follett RF, Leavitt SW, Halvorson A, Peterson GA and Lyon DJ (1997) ‘Radio carbon dating for determination of soil organic matter pool sizes and dynamics’, Soil Sci Soc Am J, 61, 1058–1067. Personeni E and Loiseau P (2004) ‘How does the nature of living and dead roots affect the residence time of carbon in the root litter continuum?’, Plant Soil, 267, 129–141. Pielke R Sr (2008) ‘A broader view of the role of humans in the climate system’, Physics Today, November, 54–55. Piñeiro G, Jobbágy EG, Baker J, Murry BC and Jackson RB (2009) ‘Set aside can be better climate investment than corn ethanol’, Ecol Appl, 19, 277–282. Prentice IC, Farquhar GD, Fasham MJR, Goulden M, Heimann VI, Jarmillo HS, Khesghi HS, Le Quere C, Scholes RJ and Wallace DWR (2001) ‘The carbon cycle and atmospheric carbon dioxide’, in Houghton JT, Ding Y, Griggs DJ, Noguer M, van der Linden PJ, Dai X, Maskell K and Johnson CA (eds) Climate Change 2001: The Scientific Basis. Contribution of Working Group I to the Third Assessment Report of the Intergovernmental Panel on Climate Change, Cambridge, New York, 183–237. Rabenhorst MC and Wilding LP (1986) ‘Pedogenic on the Edwards Plateau, Texas: III A new model for formation of petrocalcic horizons’, Soil Sci Soc Am J, 50, 693–699. Read P (2008) ‘Biosphere carbon stock management: addressing the threat of abrupt climate change in the next few decades, an editorial essay’, Climatic Change, 87, 305–320.


302


Richards JF (1990) ‘Land Transformations’, in Turner BL et al. (eds), The Earth As Transformed by Human Action, New York, Cambridge University Press, 1990, 163–178. Richter D deB Jr, Jenkins DH, Karakash JT, Knight J, McCreery LR and Nemestothy KP (2009) ‘Wood energy in America’, Science, 323, 1432–1433. Rillig MC and Mummey DL (2006) ‘Mycorrhixas and soil structure’, New Phytol, 8, 811–818. Roy I, Saugier B and Mooney HA (eds) (2001) Terrestrial Global Productivity, San Diego, CA, Academic Press. Ruddiman WF (2003) ‘The anthropogenic greenhouse era began thousands of years ago’, Climatic Change, 61, 262–292. Ruddiman WF (2005) ‘How did humans first alter global climate’, Sci Am, 292, 429–436. Rustad LE (2006) ‘From transient to steady-state response of ecosystems to atmospheric CO2-enrichment and global climate change: conceptual challenges and need for an integrated approach’, Plant Ecol, 182, 43–62. Schenk HJ and Jackson RB (2002) ‘The global biogeography of roots’, Ecol Monogr, 72, 311–328. Schimel D (2006) ‘Climate change and crop yields: beyond Cassandra’, Science, 312, 1889–1890. Schlesinger WH (1990) ‘Evidence from chronosequence studies’, Nature, 348, 232– 234. Schlesinger WH (1999) ‘Carbon sequestration in soils’, Science, 284, 2095. Schnitzer M (1991) ‘Soil organic matter–the next 75 years’, Soil Sci, 151, 41–58. Scholz F and Hasse U (2008) ‘Permanent wood sequestration: The solution to the global carbon dioxide problem’, ChemSusChem, 1, 381–384. Schulze ED (2006) ‘Biological control of the terrestrial carbon sink’, Biogeosci, 3, 147–166. Seifritz W (1993) ‘Should we store carbon in charcoal?’, Int J Hydrogen Energy, 18, 405–407. Seiler W and Crutzen PI (1980) ‘Estimates of gross and net fluxes of carbon between the biosphere and the atmosphere from biomass burning’, Climatic Change, 2, 207–247. Shine KP and Sturges WT (2007) ‘CO2 is not the only gas’, Science, 315, 1804–1805. Six J, Frey SD, Thiet RK and Batten KM (2006) ‘Bacterial and fungal contributions to carbon sequestration in agroecosystems’, Soil Sci Soc Am J, 70, 555–569. Smith P et al. (2008) ‘Greenhouse gas mitigation in agriculture’, Phil Trans Royal Soc (B), 363, 789–813. Sobecki TM and Wilding LP (1983) ‘Fountain of calcic and argillic horizons in selected soils of the Texas coast Prairie’, Soil Sci Soc Am J, 47, 707–715. Somerville V (2006) ‘The billion ton biofuel vision’, Science, 312, 1277. Strand SE and Benford G (2009) ‘Ocean sequestration of crop residue carbon: recycling fossil fuel carbon back to deep sediments’, Environ Sci Technol, 43, 1000–1007. Trumbore S (2000) ‘Age of soil organic matter and soil respiration: radiocarbon constraints on below-ground C dynamics’, Ecol Appl, 10, 399–411. USDOE (1999) Carbon Sequestration: Research and Development, Springfield, VA, National Technical Information Service. Van Oost K, Quine TA, Glovers G, De Gayze S, Six J and Harden JW et al. (2007) ‘The impact of agricultural soil erosion on the global carbon cycle’, Science, 318, 620–629. © Woodhead Publishing Limited, 2010


303

Waggoner PE (1996) ‘How much land can ten billion people spare for nature?’, Daedalus, 125, 73–93. West TO and Post WM (2002) ‘Soil organic carbon sequestration rates by tillage and crop rotation: a global data analysis’, Soil Sci Am J, 66, 1930–1946. Whitaker RG and Likens GE (1975) ‘The biosphere and man’, in Lieth H (ed.), Primary Productivity of the Biosphere, New York, Springer, 14, 305–328. Wielopolski L (2006) ‘In-situ non-invasive soil carbon analysis: sample size and geostatistical considerations’, in Lal R, Cerri CC, Bernoux M, Etchevers J, Cerri E (eds), Carbon Sequestration in Soils of Latin America, Binghamton, NY, Haworth Press, 443–455. Wild A (2003) ‘Salt, Land and Food: Managing the Land During the 21st Century, Cambridge, Cambridge University Press, UK. Woods WI, Falcao NPS and Teixeira WG (2006) ‘Biochar aims to enrich soils of smallholders’, Nature, 443, 144. Woodward FI, Lomas MR and Kelly CK (2004) ‘Global climate and the distribution of plant biomes’, Proc R Soc Lon B Soil Sci, 359, 1465–1476. Zeng N (2008) ‘Carbon sequestration via wood burial’, Carbon Balance and Management, 3, 1–12.


11

Ocean sequestration of carbon dioxide (CO2) D. G o l o m b and S. P e n n e l l, University of Massachusetts Lowell, USA Abstract: The deep ocean is a potential storage medium for anthropogenic carbon dioxide (CO2). However, there is strong opposition to ocean storage because of the potential acidification of large volumes of seawater around the discharge point. Furthermore, it is not clear whether international regulations would permit deep ocean discharge of anthropogenic CO2. This chapter reviews various proposals for oceanic CO2 injection designed to alleviate the acidification problem and to discharge the CO2 at sufficient depth so it will not rapidly re-emerge into the surface layer and back into the atmosphere. One proposal is described in greater detail, that of discharging the liquid CO2 in the form of an emulsion in seawater stabilized by pulverized limestone (CaCO3). Key words: ocean storage of carbon dioxide, deep ocean, acidification of seawater, legal constraints, sinking plumes, CO2 emulsion.

11.1

Introduction

This chapter attempts to give an overview of the scientific and technological principles of storing anthropogenic carbon dioxide (CO2) in the deep layers of the ocean. Because there is strong public and political opposition to deep ocean storage of CO2, as well as legal constraints enunciated in the London Convention on Ocean Dumping and the United Nations Convention of the Law of the Sea, deep ocean injection is currently not considered as a viable option for disposing of anthropogenic CO2. In the more distant future, if CO2 concentrations in the atmosphere are still on the rise, other storage and sequestration methods prove to be insufficient, and low- and non-carbon energy sources do not adequately supply the rising energy needs of a burgeoning world population, deep ocean storage may again be considered as an option to reduce atmospheric CO2 concentrations. Let us state at the outset that this chapter concerns storage of CO2 in the deep ocean. We are not addressing here storage proposals in the surface layer of the ocean, such as iron fertilization for enhancing the growth of phytoplankton, or dispersing directly gaseous CO2 or an aqueous solution thereof into the surface layer. We define the deep ocean at a depth of 500 m or deeper. The depth of 500 m corresponds to a hydrostatic pressure of about 5 MPa and a temperature of about 4 °C. At these depths, CO2 remains 304 © Woodhead Publishing Limited, 2010


305

in its liquid state. If liquid CO2 were injected at a lesser depth, it would flash into vapor, and the resulting gas bubbles would immediately surge upward, eventually re-emerging into the atmosphere. Why deep ocean storage? The ocean is vast. It covers approximately 70 % of the earth’s surface, its average depth is about 3800 m, and the deeper layers of the ocean are greatly unsaturated in regard to CO2 and its dissolved forms, carbonic and bicarbonic acid, and their salts. A cubic km of deep ocean water contains about 3 ¥ 1010 g carbon (Wilson, 1992). Thus, the total carbon content of the deep ocean (> 500 m) is approximately 5.1 ¥ 1022 g. At present, anthropogenic emissions of carbon equal about 6.8 ¥ 1015 g y–1. If all the emissions were injected into the deep ocean, they would scarcely make a dent in the deep ocean’s carbon content. This is not to say that local effects around the discharge point would be negligible. Caulfield et al. (1997) estimated that the injection of liquid CO2 captured from several large coal-fired power plants would reduce the pH of seawater below 7 in a volume of tens of cubic km. Israelsson et al. (2009), reviewing the latest literature on biological effects of CO2 in the marine environment, concluded that the direct toxicity of dissolved CO2, expressed as pCO2, may be a more important metric than pH. Here pCO2 = [CO2]/K0 in units of pressure, where the square brackets denote concentration in mole/L and K0 is Henry’s law constant, which is a function of the partial pressure of CO2 at the gas/ water interface and temperature. However, Israelsson et al. concluded that discharge scenarios could be engineered to achieve perturbations in the marine environment that approach the natural variability of both ocean pH and pCO2.

11.2

History of carbon dioxide (CO2) deep ocean storage proposals

Marchetti (1977, 1979) proposed first the idea of storing CO2 in the deep ocean. A 1000 km pipeline would carry liquid CO2 from about 10 large power stations to a disposal site, e.g. at the outflow of the Mediterranean Sea into the Atlantic Ocean at the Straits of Gibraltar. There, the outflow carries over one million tonnes of water per second, gently sinking into the deep layers of the Atlantic. Similar, but smaller, thermohaline currents exist in the Red, Weddell and Norwegian Seas. Marchetti considered CO2 separation from the flue gas by a scrubbing process or burning coal in pure oxygen. Mustacchi et al. (1979) proposed the direct (unseparated) flue gas bubbling at a depth of about 240 m. They estimated that absorption of CO 2 in seawater will be complete before the bubbles ascend to the surface. A second option was the release of separated liquefied CO2 at a depth of 160 m. A third option was the release at a depth of only 10 m of a seawater solution of CO2. Because of the scarce solubility of CO2 in water, this would


306


require pumping very large amounts of seawater, and very large contact towers. Albanese and Steinberg (1980) and Steinberg and Cheng (1985) proposed to inject the separated, liquid CO2 at 300 m depth. Baes et al. (1980) investigated the behavior of a jet of liquid CO2 released in the deep ocean, as well as dropping dry ice blocks from a barge. Saji et al. (1992) suggested releasing the CO2 in the form of a hydrate (clathrate). The hydrate consists of a molecule of CO2 imbedded in a cage of six water molecules, hence the name clathrate (caged). The solid hydrate is denser than seawater (approximately 1130 kg m–3), hence it would sink deeper from the injection point. Saji et al. suggested forming the hydrate on a platform constructed over the deep ocean. Liquid CO2 would be barged to the platform in a ship, and cold seawater would be pumped from great depths. The two liquids would be mixed under high pressure in order to form the hydrates. Hydrates form at a minimum temperature of 10 °C and minimum pressure of 4.32 MPa. The hydrates would be discharged from a pipe dangling from the platform at about 2000 m. Golomb et al. (1992) hypothesized that CO 2 hydrates will form spontaneously when liquid CO2 is released below 500 m, where the temperature is lower than 10 °C and the hydrostatic pressure is about 5 MPa. Golomb et al. estimated that, because of the higher density of the hydrates, the plume would sink all the way to the ocean bottom. Brewer et al. (2003) demonstrated that CO2 hydrates form spontaneously when liquid CO2 was injected from a submersible vehicle at depths greater than 3000 m. Tsouris et al. (2004) produced CO2 hydrates in a liquid CO2–seawater co-injector carried on a submersible. The hydrates were injected between 1100 and 1300 m. Depending on the liquid CO2–seawater flow rates in the co-injector and on the release depth, the formed CO2 hydrates were either positively or negatively buoyant. Ohsumi (1995) suggested that liquid CO2 be released below 3000 m. At those depths, the density of liquid CO2 is about 1050 kg m–3, which is denser than seawater (about 1028 kg m–3 at 3000 m), hence it would sink all the way to the ocean bottom and form a CO2 ‘lake’ there. Adams et al. (1995) proposed to release the liquid CO2 in a confinement vessel at 500 m or deeper. The confinement vessel is open at the top and bottom. Within the confinement vessel, a large fraction of the CO2 would dissolve in seawater, forming a dense, negatively buoyant solution. The vessel acts like an inverted chimney, allowing lighter seawater to be drawn in through the open top of the vessel, and allowing the denser solution to flow out through the open bottom of the vessel. Because of entrainment of ambient seawater, eventually the descending plume would density-equilibrate with the hydrostatically increasing density of seawater. Ozaki et al. (1999) proposed the release of liquid CO2 from a pipe towed by a moving ship between 1000 and 2500 m. This concept would promote the dispersion and dissolution of the CO2 over a wider volume of seawater.



307

Israelsson et al. (2009) propose not only to release the CO2 from a moving ship, but that the injection proceed via a liquid CO2-seawater co-injector as described by Tsouris et al. (2004). In such a manner, a part of the CO2 forms a solid hydrate, further promoting negative buoyancy. It will be a naval engineering challenge to design and build a pressurized, refrigerated tanker with an attachable large diameter 1000–2500 m long pipe and co-injector for the discharge of liquid and hydrated CO2 in the open sea. Furthermore, the cost of such conveyance vessels may be prohibitive. Golomb and Angelopoulos (2001) proposed to release liquid CO2 mixed with a slurry of pulverized limestone (CaCO3) in seawater. The amount of limestone in the slurry would balance stoichiometrically the amount of liquid CO2 in order to achieve complete buffering of the carbonic acid formed after dissolution of the liquid CO2 in seawater. This would require about 2.3 tonnes of pulverized limestone per tonne of CO2, raising the cost of deep ocean storage of CO2 substantially in terms of raw materials, handling and transport. Golomb et al. (2007) proposed the release of a CO2/water emulsion stabilized by very fine limestone (CaCO3) particles. As described below, this method of release will require far less than stoichiometric quantities of CaCO3, and may greatly alleviate the seawater acidification problem.

11.3

Legal constraints of deep ocean storage of carbon dioxide (CO2)

There are two international regulations that may constrain the deep ocean storage of CO2: The London Convention on Ocean Dumping and the United Nations Convention of the Law of the Sea.

11.3.1 London convention on ocean dumping This Convention essentially prohibits all dumping activity in the oceans. A resolution in 1991 formally adopted the Precautionary Principle and outlawed the dumping of all radioactive and industrial waste. The resolution defined industrial waste as ‘generated by manufacturing or processing operations.’ Since then, ongoing discussions have not produced a consensus on whether CO2 should be classified as an industrial waste. The London Convention applies only to ships, aircraft and offshore platforms. Apparently, offshore disposal of CO2 by pipeline would not fall under the purview of the London Convention, but would be governed by national laws. In 1997, the Joint Group of Experts on the Scientific Aspects of Marine Environmental Protection (GESAMP) reported that unless two-thirds majority of Contracting Parties amended the Convention, CO2 dumping from ships, specifically dry ice and liquid CO2, violates the London Convention.


308


11.3.2 United Nations Convention on the Law of the Sea (UNCLOS) Part XII, Article 194 states: ‘Minimize to the fullest extent the release of toxic, harmful or noxious substances, especially those which are persistent, from land-based sources, from or through the atmosphere, or by dumping.’ Article 210 states: ‘Dumping within the territorial sea and the exclusive zone or onto the continental shelf shall not be carried out without the express prior approval by the State, which has the right to permit, regulate and control such dumping after due consideration of the matter with other States which by reason of their geographical situation may be adversely affected thereby.’ Ironically, the UNCLOS tries to minimize the release of toxic substances onto the ocean from or through the atmosphere. It is estimated that about one-third of the anthropogenic emissions of CO2 are inevitably landing on, and absorbed by, the ocean surface waters (Houghton, 1997). It is evident that both the London and the United Nations conventions need to define (i) whether CO2 is an industrial waste and (ii) whether CO2 is toxic, harmful or noxious. Clearly, special expert committees will have to be convened to decide on these issues.

11.4

Sources of anthropogenic carbon dioxide (CO2) for ocean storage

In 2007, global emissions of CO2 amounted to 26.5 Gt y–1. Of this, 17.7 Gt y–1 came from fossil fuel combustion (IPCC, 2007). CO2 capture will most likely be economic only from central, large CO2 emission sources, such as coal-fired power plants. Oil-fired power plants constitute only about 2–3 % of fossil-fueled power plants and, with the escalating price of oil, their percentage of power plants will be even less. Gas-fired power plants amount to 8–10 % of all fossil-fueled power plants globally, but a modern gas-fired combined cycle power plant emits only about 50 % of CO2 per kWh of electricity generated compared to a coal-fueled plant. Therefore, CO2 capture and storage is considered only worthwhile from large coalfueled power plants. At present, about 25–30 % of global CO2 emissions come from coal-fueled power plants. Because overland transport in pipelines of liquid CO2 is about twice as expensive as transport by pipes laid on the sea bed (Golomb, 1993), most likely only coastal, coal-fueled power plants can be considered for deep ocean storage. For example, in the USA less than 10 % of all coal-fueled power plants are located on the seacoasts (US EIA, 2008). If that percentage is similar worldwide, deep ocean storage from coastal power plants may not amount to much more than 2.5–3 % of global CO2 emissions. Of course, insular countries, like Japan and the UK, may have a larger percentage of



309

coastal power plants than inland countries. Also, in the future, new coalfueled power plants with CO2 capture may be specifically located near the coasts in order to avail themselves of deep ocean storage if it were permitted by international conventions.

11.5

Ocean structure

The structure of the ocean and the physical-chemical properties of CO2 are thoroughly intertwined in determining the storage capacity and the storage period of CO2. Of foremost importance are the temperature and density profiles of the ocean. For reasons of transporting large quantities, CO 2 will most likely be injected in its liquid rather than in its gaseous phase. Therefore, we must ensure that liquid CO2 will not immediately flash into positively buoyant gas bubbles. Because liquid CO2 is less dense than seawater, we must ensure that the injected liquid CO2 droplets dissolve in seawater before they ascend to a depth where they would flash into gas. A temperature profile for the North Pacific (45° latitude, 165° longitude) is depicted in Fig. 11.1. The well-mixed surface layer extends to about 100 m depth, where the temperature reaches about 5 °C. Below is the thermocline layer where the temperature gradually decreases to about 2 °C at 2000 m 0 200 400

Depth (m)

600 800 1000 1200 1400 1600 1800 2000 0

2

4


10

12

14

11.1 Sample temperature profile for the North Pacific (45° north latitude, 165° east longitude, 17 Sept 1997) (data from World Ocean Database, 2005).


310


depth. Because of the low compressibility of water, the density does not change very much with depth. At mid-latitude, at a depth of 500 m, the density is about 1026 kg m–3, increasing to 1027 kg m–3 at 1000 m, and 1028 kg m–3 at 3000 m (Millero, 2006). At 3000 m, the density of liquid CO2 is greater than seawater (see Fig. 11.2). This is because of the larger compressibility of liquid CO2 compared to water. Liquid CO2 released below 3000 m is negatively buoyant. Figure 11.2 depicts a sample pH profile in the North Pacific. The profile shows a pH of about 8 in the surface layer, declining to 7.4 at about 200 m, and gradually increasing to 7.6 at 1800 m. From an ocean storage point of view, it is important to note that at 500–1000 m depth the pH is approximately 7.5. Dissolving carbonic acid would lower the pH perhaps to the detriment of marine organisms that live in those depths. Caulfield et al. (1997) estimated the acidification of seawater when liquid CO2 is released at depth. The release of CO2 captured from one 500 MW coal-fired power plant (125 kg/s) would render the pH less than 7 in a volume of a few km3. This is the major cause of concern to marine biologists about deep ocean storage of CO2. However, as described below, some proposed injection methods may alleviate that concern.

0 200 400

Depth (m)

600 800 1000 1200 1400 1600 1800 7.4

7.5

7.6

7.7

pH

7.8

7.9

8

8.1

11.2 Sample pH profile for the North Pacific (48° north latitude, 165° east longitude, 17 Sept 1997) (data from World Ocean Database, 2005).



311

20

Pressure (MPa)

15 CO2: 40 °C CO2: 30 °C 10 CO2: 20 °C CO2: 10 °C

5

CO2: 1 °C

H2O: 4 °C

0 0

100

200

300

400 500 600 Density (kg/m3)

700

800

900

1000

11.3 Density–pressure–temperature nomogram of liquid and supercritical CO2 and H2O (adapted from NIST, 2003).

11.6

Properties of carbon dioxide (CO2)

The pressure–temperature–density diagram of CO2 is depicted in Fig. 11.3. In the pressure–temperature interval of 5–10 MPa, 1–10 °C, corresponding to 500–1000 m depth, the density of liquid CO2 is in the range 930–950 kg m–3, far less than seawater, hence positively buoyant. Figure 11.3 also shows the density variation of water with pressure at 4 °C. At 30 MPa, corresponding to a depth of about 3000 m (somewhat outside the frame of Fig. 11.3), the density of water is about 1028 kg m–3. At that depth, liquid CO2 has a density of 1050 kg m–3, therefore liquid CO2 released at that depth becomes negatively buoyant, and would sink all the way to the ocean bottom, forming a ‘CO2 lake’ imbedded on the ocean bottom. This led to several proposals to release CO2 at depths greater than 3000 m, thereby ensuring long storage periods (e.g. Ohsumi et al., 1992). However, it may be very expensive to lay pipes offshore to 3000 m depth, or from a pipe suspended from a platform anchored to the ocean bottom, let alone the transportation cost of pressurized, refrigerated liquid CO2 in a tanker ship. Liquid CO2 is sparsely soluble in water. The solubility increases with pressure and decreases with temperature. In the interval 500–1000 m (5–10 MPa, 3–4 °C), it is about 5–6 % by weight (Stewart and Munjal, 1970). The low density and limited solubility constrain the deep ocean storage of CO2. Liquid CO2 has to be injected through a diffuser or atomizer in order that the bulk liquid is dispersed into tiny droplets. In such a fashion, the droplets


312


have a chance to dissolve in seawater by mutual diffusion of one liquid into another before the positively buoyant droplets ascend to a depth where they would flash into vapor, which would occur at about 500 m depth (about 5 MPa). Another issue is the likely formation of CO2 hydrates (clathrates). The hydrates are solid, ice-like particles of density of about 1130 kg m –3 and composition CO2·6H2O (Ohmura and Mori, 1998). A phase diagram of CO2–H2O is shown in Fig. 11.4. It is seen that in the 0–10 °C, 5–10 MPa regime, corresponding to 500–1000 m depth, hydrates are stable. Brewer et al. (2003) have shown that releases of liquid CO2 at those depths do form hydrates. However, hydrates seem to form a film at the CO2–seawater interface. The composite density of the hydrate-coated droplets is less than that of seawater, hence they buoy upward from the release point. In fact, the hydrate film may be a hindrance to ocean storage because it may limit the dissolution rate of CO2 droplets.

11.7

Modeling of carbon dioxide (CO2) release

Herzog et al. (1991) discussed the injection of liquid CO2 at depths between 500 m and 2000 m. The injected CO2 breaks up into droplets whose radius depends on the size of the injection nozzle and the flow rate. At depths above 3000 m, liquid CO2 is less dense than ambient seawater, hence the droplets are positively buoyant. As the droplets rise, CO2 dissolves into seawater. The

2 O(I)–

CO2 (I)

100

te–H

10

Hydrate

1 Hydr

–CO 2(I)–

CO 2(g)

I Hydrate–CO2(g) O( 2 (g) –H . O 2 C d O(s)– Hy ate–H 2

O )–C

(g

H2O(I)–CO2(I)

(g) O (I)–CO 2 H 2O(I)–C 2

Hydra

Pressure (MPa)

Hydrate–CO2(I)

)

2

H2O(I)–CO2(g)

H2O(s)–CO2(g) H2O(s)–H2O(I)–CO2(g) 0.1 –20

–10


20

30

11.4 Phase diagram of the water–CO2 system (adapted from NIST, 2003).



313

authors calculated the minimum depth at which a droplet of given radius must be released so that it is completely dissolved by the time it reaches a depth of 500 m, at which point it would flash into vapor. The larger the droplet, the greater the release depth must be. These calculations were carried out using four different assumptions regarding the relation between droplet velocity and mass transfer rate between CO2 drop and seawater. For example, a 1 cm radius droplet when released at 700 m would dissolve completely before it reaches a depth of 500 m. Holder et al. (1995) modified the work of Herzog et al. (1991) to take into account the likely formation of hydrates. They assumed that hydrates form on the surface of a rising CO2 droplet at a rate of 0.04 cm/h. Because solid hydrates are denser than seawater, at some point the droplet becomes negatively buoyant and begins to sink. The authors calculated the minimum depth at which a drop of given radius must be released so it would sink by the time it reaches a depth of 500 m. For example, a 1 cm radius droplet must be released at a depth of about 1400 m, so it will sink before it ascends to 500 m. Wannamaker and Adams (2003) modeled the behavior of negatively buoyant solid hydrate particles. They used a double plume model in which the hydrate particles and entrained seawater form a sinking inner plume surrounded by a rising outer plume of detrained seawater. They calculated maximum plume depth, average intrusion depth and average intrusion changes in dissolved inorganic carbon and pH for given initial particle diameter and CO2 release rates. For example, their model showed that a release at a depth of 800 m with a CO2 release rate of 100 kg/s and particle diameter of 2 cm would result in a maximum plume depth of approximately 2500 m below the release point. Adams et al. (1995) modeled the release of liquid CO2 into a confinement vessel open both at the top and the bottom of the vessel, much like an inverted chimney. Because of the confinement, the dissolving droplets form a dense solution of carbonic acid (H2CO3). The dense acid sinks through the bottom of the confinement vessel. In such a fashion, the confinement vessel could be mounted just below the critical depth of 500 m, where liquid CO 2 would flash into vapor.

11.8

Injection of carbon dioxide, water and pulverized limestone (CO2/H2O/CaCO3) emulsion

In the quest to minimize the acidification of seawater and to ensure that the released CO2 sinks to greater depth from the injection point, the authors developed the concept of emulsifying liquid CO2 in water. As noted before, liquid CO2 is sparingly soluble in water. However, when liquid CO2 is mixed © Woodhead Publishing Limited, 2010

314


with water in the presence of very fine hydrophilic particles, an emulsion is formed consisting of tiny CO2 droplets sheathed with the particles. The sheathed droplets are called globules (Pickering, 1907). The globules are dispersed in water (Golomb et al., 2006). Pulverized limestone (CaCO3) is an inexpensive and environmentally benign material that could be used for stabilizing the Pickering emulsion. In laboratory tests, it has been demonstrated that a stable emulsion can be created with a proportion of 33.3 % by volume liquid CO2, 66.7 % by volume artificial seawater (3.5 % NaCl solution) and 0.5 kg pulverized limestone per kg of liquid CO2. The limestone particles range from submicron to a few mm in diameter, with a mean diameter of 2 mm. The emulsion is created in a Kenics-type static mixer. A photo of the emulsion is shown in Fig. 11.5. The globules’ diameters range from 100–200 mm. The emulsion has a gross density of 1087 kg m–3, compared to a density of seawater at 500 m of 1026 kg m–3.

11.8.1 Emulsion release into open ocean For injection of the emulsion into the open ocean, we visualize a system depicted in Fig. 11.6. A floating platform is tethered to the ocean bottom. Liquid CO2 is barged to the platform and stored in a tank. Pulverized limestone is barged to the platform and slurried with seawater pumped from a depth of

11.5 Emulsion consisting of 33.3 % by volume liquid CO2, 66.7 % by volume artificial seawater (3.5 % NaCl solution) and 0.5 kg pulverized limestone per kg of liquid CO2 (adapted from Golomb et al., 2007).



315

Release in the open ocean Material handling

CaCO3

CO2

Surface seawater

z = –200 m

Ambient seawater Static mixer z = –500 m

Stalled plume

z = –700 m

Rainout and dissolution z = bottom

11.6 Schematic of open ocean CO2 release (from Golomb et al., 2007).

about 200 m (below the photic zone). The liquid CO2 and limestone slurry are piped to a depth of about 500 m into a static mixer. Before entering the mixer, the limestone slurry is diluted by ambient seawater, which is drawn by aspiration. As noted above, the mix ensuing from the static mixer has a gross density of 1087 kg m–3. The only power requirements for creating the mix are for pumping seawater from a depth of 200 m into the slurry mixer and the mechanical mixing of the slurry. No additional power is required for the undersea static mixer; the hydrostatic pressures of the liquid CO2 and the pulverized limestone slurry provide adequate force for mixing the ingredients in the static mixer. Golomb et al. (2007) modeled the behavior of the released emulsion plume. The emulsion sinks because it is denser than ambient seawater. It is assumed that the descending plume entrains ambient seawater at a rate proportional to the plume velocity. Entrainment of seawater decreases the plume density. The plume descends to a level at which it becomes neutrally buoyant with the density stratified seawater. The authors calculated the depth to which the plume sinks before becoming neutrally buoyant, assuming the emulsion is injected at a depth of 500 m from a pipe of diameter 1 m. The plume length increases as the mass flow rate of injected emulsion increases and as the ambient density stratification and entrainment coefficient decrease. Because the injection would occur below 500 m depth, which is below the average picnocline at mid-latitudes, the density stratification is very weak. The density stratification is expressed by the buoyancy frequency N, defined as


316

Developments and innovation in CCS technology N2 =

g d ra –2 s ras dz

where ras is the seawater density at the ocean surface, ra is ambient density, z is depth and g is the gravitational constant. Figure 11.7 shows the modeling results. For example, in a weakly density stratified ocean (N2 = 10–6 s–2) a plume containing the CO2 output of a 1000 MW coal-fired power plant (250 kg/s) released at 500 m would descend 800 m before it density equilibrates with ambient seawater.

11.8.2 Emulsion release along sloping continental shelf When a depth of about 500 m can be reached within 100–200 km from shore, a pipe system laid on the continental slope may be more economic than delivering the ingredients of the emulsion to a floating platform by barges. The system is depicted in Fig. 11.8. Liquid CO2 is stored onshore in a tank. A slurry of pulverized limestone in seawater is prepared onshore. The seawater for the slurry is pumped from about 200 m below the surface. Liquid CO2 is pumped from the tank into a pipe, where it flows to a depth 800 E = 0.05; CO2 flux = 250 kg/s E = 0.05; CO2 flux = 125 kg/s

700

E = 0.05; CO2 flux = 62.5 kg/s E = 0.1; CO2 flux = 250 kg/s

Vertical plume length (m)

600

E = 0.1; CO2 flux = 125 kg/s E = 0.1; CO2 flux = 62.5 kg/s

500 400 300 200 100 0 10–6

10–5 N2 (s–2)

10–4

11.7 Dependence of open ocean globulsion plume length on density stratification. A CO2 flux of 250 kg/s corresponds to an output of a 1000 MWel coal-fired power plant. Note that the squares and diamonds overlap (adapted from Golomb et al., 2007).



317

Material handling Ca CO 3

CO2

Surface seawater

z = –200 m

Ambient seawater Static mixer

z = –500 m

Stalled plume

z = –1000 m

Rainout z = bottom

11.8 Schematic of CO2 release on sloping seabed (from Golomb et al., 2007).

of about 500 m, then into the static mixer. The limestone slurry is diluted with ambient seawater by aspiration before it flows into the static mixer. The mixer lies on the bottom slope. The proportion of the ingredients is the same as in the open ocean release. In locations where the sea floor drops off rapidly, it may be more economic to release CO2 through a pipe from the shore rather than from a ship or offshore platform. Drange et al. (1993) modeled the flow of a plume consisting of CO2enriched seawater along a sloping seabed. The model incorporates changes in plume temperature, salinity, alkalinity and total inorganic carbon content. Effects of ambient density stratification, entrainment of ambient seawater and friction between the plume and seabed are included in the model. Plume behavior is found to be sensitive to both the ambient density stratification and the friction coefficient. Small friction coefficients lead to higher plume velocities and more rapid entrainment of ambient seawater, leading to more rapid dilution of the plume. On the other hand, larger friction coefficients lead to slower and longer plumes, allowing the CO2 to be carried to greater depths. The authors considered typical ambient conditions in the North Atlantic, the North Pacific, and the Norwegian Sea. Plumes are found to sink to greater depths under the conditions found in the Norwegian Sea than in the North Atlantic or North Pacific. Golomb et al. (2007) modeled the flow along a sloping seabed of an injected emulsion as described in the previous section. Chemical and thermodynamic changes were not considered; only the fluid dynamics of the flow were described. Effects of ambient density stratification, entrainment of ambient seawater and friction between the plume and seabed were included


318


in the model. The authors calculated the depth to which the plume sinks before becoming neutrally buoyant, assuming the emulsion is injected at a depth of 500 m from a pipe of 1 m diameter. Plume penetration increases with increasing seabed slope, with increasing friction coefficient (because the plume moves more slowly, entraining less ambient seawater), and with decreasing ambient density stratification. Plume penetration on a sloping seabed is found to be greater than in an open ocean release because the entrainment rate is lower. Less plume surface is exposed to the ambient seawater, and friction with the seabed slows the plume, further decreasing the entrainment rate. Model results are shown in Fig. 11.9. For example, with a buoyancy frequency N2 = 10–6 s–2 and with typical entrainment and friction coefficients, a plume containing the CO2 output of a 1000 MW coal-fired power plant (250 kg/s) released at 500 m along a seabed with 10° slope would descend 2200 m before it density equilibrates with ambient seawater. After the plume comes to rest, the globules, together with excess pulverized limestone, will ‘rain-out’ from the plume on their way to the ocean bottom. Using Stokes’ law for the settling of small particles in a viscous medium, it is estimated that 100 mm radius globules sheathed with a 2 mm thick layer of limestone particles sink at a velocity of approximately 2 ¥ 10–3 m s–1, that is, about 200 m d–1 (Golomb et al., 2007). Eventually, the globules may disintegrate due to wave action and bottom friction. A part of the dissolved carbonic acid may be buffered by the CaCO3 particles.

11.8.3 Economics Raw limestone in chunks can be purchased from several quarries for $5–10/t FOB, and milled to the desired size on-site. The total cost of pulverizing the limestone, mixing it with seawater and co-injecting it with liquid CO2 in a static mixer is estimated at $13/t of limestone. Because only 0.5 t of pulverized limestone is required, the total cost is $6.5 per tonne of liquid CO2 (Golomb et al., 2007). Present reports estimate the cost of capturing and liquefying CO2 at a coal-fired power plant at around $50/t CO2 (IPCC, 2007). Thus, the storage of liquid CO2 in the form of an emulsion would add about 13 % to the capture and liquefaction cost of CO2. The additional cost may be justified on account of increasing the storage period of the released CO2, lowering the transport cost to deeper waters and most importantly, of not acidifying the seawater around the injection point.

11.9

Future trends

Given the public opposition and the legal constraints, it is not expected that in the near future further research and development, let alone actual deep ocean injection of CO2, will be practised.


Ocean sequestration of carbon dioxide (CO2) 800

700


600

319

Seabed slope 1° CD CD CD CD CD CD

= = = = = =

0.1, CO2 flux = 250 kg/s 0.1, CO2 flux = 125 kg/s 0.01, CO2 flux = 250 kg/s 0.1, CO2 flux = 62.5 kg/s 0.01, CO2 flux = 125 kg/s 0.01, CO2 flux = 62.5 kg/s

500

400

300

200

100

0 –6 10

10–5 N2 (s–2)

10–4

2500


2000

CD CD CD CD CD CD

= = = = = =

0.1, CO2 flux = 250 kg/s 0.1, CO2 flux = 125 kg/s 0.1, CO2 flux = 62.5 kg/s 0.01, CO2 flux = 250 kg/s 0.01, CO2 flux = 125 kg/s 0.01, CO2 flux = 62.5 kg/s

1500

1000

500

0 –6 10

10–5 N2 (s–2)

10–4

11.9 Vertical globulsion plume length vs density stratification for seabed with slope 1° (top) and 10° (bottom) (adapted from Golomb et al., 2007).


320


11.10 Conclusions The deep ocean would be a logical storage medium for anthropogenic CO2. The ocean occupies about 70 % of the planet’s area; its average depth is 3800 m; and the deep layers of the ocean are greatly undersaturated in regard to its carbonate absorption capacity. Furthermore, the ocean is not isolated from the atmosphere. It is estimated that roughly one third of the yearly anthropogenic emissions of CO2 are absorbed into the ocean. The pH of the surface layer of the ocean may have already decreased by 0.15 units (IPCC, 2007) due to the anthropogenic CO2 emissions over the last century. However, there is strong opposition by marine biologists and environmental groups to deep ocean storage of anthropogenic CO2. The major concern arises from the possibility of acidification of large volumes of seawater due to the formation of carbonic and bicarbonic acid. Furthermore, there are international regulations that prohibit ocean dumping of industrial waste, which may include anthropogenic CO2. Several proposals have been put forward in order to minimize the acidification problem. They include releasing the CO2 through diffusers or atomizers or from a moving ship. In such a fashion, the released CO2 would rapidly disperse over large volumes of seawater, so no concentrated carbonic acid would be created. Also, it has been proposed to release the CO2 in the form of hydrates that are heavier than seawater; hence they would sink to greater depth from the release point. The authors propose to release a CO2–water emulsion stabilized by pulverized limestone (CaCO3). The emulsion is denser than seawater; therefore, it would sink several hundred meters from the release point before the emulsion comes to a rest with the density stratified seawater. The preparation of the emulsion would incur an additional cost to ocean storage, but that cost may be amply compensated by the fact that the emulsion can be released at a relatively shallow depth of 500 m (minimizing transport costs), and that the formed carbonic acid would be partially neutralized by the pulverized limestone used to form the emulsion.

11.11 Sources of further information and advice Most papers on deep ocean storage of CO2 have been published in the Proceedings of the Greenhouse Gas Control Technologies (GHGT) biannual conferences. In GHGT-1, Amsterdam, Netherlands, 1992, 10 papers out of a total of 80 addressed ocean storage. The latest GHGT-9 conference, Washington, DC, 2008, devoted two out of 640 papers to ocean storage. Monographs on ocean storage are available, including Handa and Ohsumi (1995) and Omerod (1997).



321

11.12 References Adams E E, Golomb D, Zhang X Y and Herzog H J (1995) ‘Confined release of CO2 into shallow seawater’, in Handa N and Ohsumi T (eds), Direct Ocean Disposal of Carbon Dioxide, Tokyo, Japan, Terrapub, 153–164. Albanese A S and Steinberg M (1980) Environmental Control Technology for Atmospheric Carbon Dioxide, Springfield, VA, National Technical Information Service, US Dept of Commerce. Baes Jr E F, Beall S E, Lee D W and Garland G (1980) ‘The collection, disposal, and storage of carbon dioxide’, in Bach W, Pankrath J and Williams J (eds), Interactions of Energy and Climate, Boston, MA, Reidel. Brewer P G, Peltzer E T, Rehder G and Dunk R (2003) ‘Advances in deep-ocean CO2 sequestration experiments’, in Gale J and Kaya Y (eds), Proceedings of the sixth International Conference on Greenhouse Gas Control Technologies: GHGT6, Oxford, UK Elsevier (Pergamon), 2, 1667–1670. Caulfield J A, Auerbach D I, Adams E E and Herzog H (1997) ‘Near field impacts of reduced pH from ocean CO2 disposal’, in Herzog H (ed.), Carbon Dioxide Removal: Proceedings of the Third International Conference on Carbon Dioxide Removal, Oxford, UK, Elsevier, 343–348. Drange H, Alendal G and Haugan P M (1993) ‘A bottom gravity current model for CO2enriched seawater’, Energy Convers Manage, 34, 1065–1072. Golomb D (1993) ‘Ocean disposal of CO2: feasibility, economics and effects’, Energy Convers Manage, 34, 967–976. Golomb D and Angelopoulos A A (2001) ‘Benign form of CO2 Sequestration in the Ocean’, in Williams D J, Durie R A, McMullan P, Paulson C A J and Smith A Y (eds), Proceedings of the Fifth International Conference on Greenhouse Gas Control Technologies: GHGT5, Collingwood, VIC, CSIRO Publishing, 463–467. Golomb D, Zemba S G, Dacey J W H and Michaels A F (1992) ‘The fate of CO2 sequestered in the deep ocean’, Energy Convers Manage, 33, 675–685. Golomb D, Barry E, Ryan D, Swett P and Duan H (2006) ‘Macroemulsions of liquid and supercritical CO2-in-water and water-in-liquid CO2 stabilized by fine particles’, Ind Eng Chem Res, 45, 2728–2733. Golomb D, Pennell S, Ryan D, Barry E and Swett P (2007) ‘Ocean sequestration of carbon dioxide: modeling the deep ocean release of a dense emulsion of liquid CO2-in-water stabilized by pulverized limestone particles’, Environ Sci Technol, 41, 4698–4704. Handa N and Ohsumi T (1995) Direct Ocean Disposal of Carbon Dioxide, Tokyo, Japan, Terrapub. Herzog H, Golomb D and Zemba S (1991) ‘Feasibility, modeling and economics of sequestering power plant CO2 emissions in the deep ocean’, Environ Prog, 10, 64–74. Holder G D, Cugini A V and Warzinski R P (1995) ‘Modeling clathrate hydrate formation during carbon dioxide injection into the ocean’, Environ Sci Technol, 29, 276–278. Houghton J (1997) Global Warming: The Complete Briefing, Cambridge, UK, Cambridge University Press. IPCC (2007) Climate Change 2007: The Physical Science Basis, Contribution of Working Group I to the Fourth Assessment Report of the Intergovernmental Panel on Climate Change, Solomon S, Qin D, Manning M, Chen Z, Marquis M, Averyt KB, Tignor M and Miller HL (eds), Cambridge, UK and New York, Cambridge University Press. Israelsson P H, Chow A C and Adams E E (2009) ‘An updated assessment of the acute


322


impacts of ocean carbon sequestration by direct injection’, in Gale J, Herzog H and Braitsch J (eds), Greenhouse Gas Control Technologies 9, Proceedings of the Ninth International Conference on Greenhouse Gas Control Technologies (GHGT9), Energy Procedia, 1, 4929–4936. Marchetti C (1977) ‘On geoengineering and the CO2 problem’, Climate Change, 1, 59–68. Marchetti C (1979) ‘Constructive solutions to the CO2 problem’, in Bach W, Pankrath J and Kellogg W (eds), Man’s Impact on Climate, New York, Elsevier, 299–311. Millero F J (2006) Chemical Oceanography, 3rd edn, Boca Raton, FL, CRC Press. Mustacchi C, Armenante P and Cena V (1979) ‘Carbon dioxide removal from power plant exhausts’, Environ Int, 2, 453–456. NIST (2003) National Institute for Standards and Technology Standard Reference Database Number 69, Washington DC. Ohmura R and Mori Y H (1998) ‘Critical conditions for CO2 hydrate films to rest on submarine CO2 pond surfaces: A mechanistic study’, Environ Sci Technol, 32, 1120–1127. Ohsumi T (1995) ‘Disposal options in view of geochemical cycle of carbon’, in Handa N and Ohsumi T (eds), Direct Ocean Disposal of Carbon Dioxide, Tokyo, Japan, Terrapub, 83–88. Ohsumi T, Nakashiki N, Shitashima K and Hirama K (1992) ‘Density change of water due to dissolution of carbon dioxide and near-field behavior of CO2 from a source on deep-sea floor’, Energy Convers Manage, 33, 685–690. Omerod B (1997) Ocean Storage of CO2: Practical and Experimental Approaches, Cheltenham, UK, IEA Greenhouse Gas R&D Programme. Ozaki M, Ohsumi T and Masuda A (1999) ‘Dilution of released CO2 in the mid ocean depth by moving ship’, in Eliasson B, Riemer P and Wokaun A (eds), Proceedings of the fourth International Conference on Greenhouse Gas Control Technologies: GHGT4, Oxford, UK, Elsevier, 275–280. Pickering S U (1907) ‘Emulsions’, J Chem Soc, 91, 2001–2020. Saji A, Yoshida H, Sakai M, Tanii T, Kamata T and Kitamura H (1992) ‘Fixation of carbon dioxide by clathrate-hydrate’, Energy Convers Manage, 33, 643–651. Steinberg M and Cheng H C (1985) A systems study for the removal, recovery, and disposal of carbon dioxide from fossil fuel power plants in the US, Report BNL-36428, Brookhaven National Laboratory, Upton, NY. Stewart P B and Munjal P (1970) ‘Solubility of carbon dioxide in pure water, synthetic sea water, and synthetic sea water concentrates at –5° to 25° and 10- to 45-atm pressure’, J Chem Eng Data, 15, 67–71. Tsouris C, Brewer P G, Peltzer E, Walz P, Riestenberg D, Liang L and West O R (2004) ‘Hydrate composite particles for ocean carbon sequestration: field verification’, Environ Sci Technol, 38, 2470–2475. US EIA (2008) Annual Energy Outlook 2008, Washington DC, US Energy Information Agency, available at http://www.eia.doe.gov/oiaf/aeo/pdf/0383(2008).pdf (accessed January 2010]. Wannamaker E J and Adams E E (2003) ‘Modeling descending carbon dioxide injections in the ocean’, in Gale J and Kaya Y (eds), Proceedings of the Sixth International Conference on Greenhouse Gas Control Technologies: GHGT6, Oxford, UK, Elsevier, Vol. 1, 753–758. Wilson T R S (1992) ‘The deep ocean disposal of carbon dioxide’, Energy Convers Manage, 33, 627–635.



323

World Ocean Database (2005) Select and Search (National Oceanographic Data Service, U.S. National Oceanic and Atmospheric Administration) [Online] (Updated 14 Jan 2009), available at: http://www.nodc.noaa.gov/OC5/SELECT/dbsearch/dbsearch.html (accessed January 2010)).


12

Environmental risks and impacts of carbon dioxide (CO2) leakage in terrestrial ecosystems

M. D. S t e v e n, K. L. S m i t h and J. J. C o lls, University of Nottingham, UK Abstract: Carbon dioxide (CO2) capture at the site of production and storage at deep underground sites is currently being investigated as a way of reducing emissions to the atmosphere. There is potential underground storage capacity for between 1300 and 11 000 Gt of CO2, but the risks and impacts on ecosystems of possible leakage from these storage areas and the associated transport systems need to be understood. While the risks in transport, e.g. pipeline failure, are reasonably well known, the experience of deep storage is relatively short and risks are estimated on the basis of scenarios. The main focus of this chapter is on the hazards presented by a leak to the surface. It reviews current knowledge on the impacts of CO2 leakage on terrestrial ecosystems, including studies of natural and manmade analogues of soil CO2 enrichment. The effects of elevated atmospheric CO2 concentrations (denoted here as [CO2]) have been intensively studied: a doubling of atmospheric [CO2] is typically expected to increase the instantaneous rate of photosynthesis by 30–50 %. However, fewer studies have been carried out on elevated [CO2] in soil. It is generally found that concentrations of up to 2 % CO2 in the soil increase growth, but that at higher levels root and shoot growth are depressed and the plants are visibly stressed. The effect of elevated soil CO2 is species dependent but, at concentrations above 20 %, CO2 is phytotoxic and vegetation does not grow. Key words: leakage risk, leakage rate, leakage impacts, plant stress effects, leak analogues, soil fauna, atmospheric enrichment, leak monitoring, tracer gases, remote sensing.

12.1

Introduction

Carbon dioxide (CO2) capture and storage (CCS) is proposed as a means of mitigating global warming by capturing the CO2 from stationary fossil fuel combustion, for example in electricity generation, and transporting it to a storage site where it will be stored away from the atmosphere for many centuries. Global emissions in 2009 were approximately 31.5 Gt (IWR, 2009). A large fraction of the emission is from fixed sources: in the UK, power generation alone accounts for about 41 % (Marland et al., 2005; Defra, 2006). CCS has the potential to reduce CO2 emissions from power generation by 80–90 % (DTI, 2002). There are three main options for geological storage: 324 © Woodhead Publishing Limited, 2010

Carbon dioxide (CO2) leakage in terrestrial ecosystems

325

1. Depleted oil and gas reservoirs. These fields are known to have effective seals which have retained the hydrocarbons for geological timescales. The estimated global capacity for CO2 storage in depleted gas fields is 800 Gt (Freund et al., 2003). Carbon dioxide can also be used in enhanced oil recovery (EOR) schemes where it is injected into an oil reservoir and dissolves into the oil, reducing its viscosity and increasing its volume to enhance reservoir pressure and increase production. Production of oil at Weyburn, Canada has been increased by 130 million barrels using this method and, by 2030, between 15 and 20 Mt of anthropogenic CO2 will have been permanently and safely sequestered (DTI, 2005). 2. Deep saline aquifers. These reservoirs are deep underground and contain saline water unsuitable for extraction as potable water. The estimated global capacity for CO2 storage in saline aquifers is between 400 and 10 000 Gt (Freund et al., 2003). 3. Unmineable coal beds. CO2 can be injected into coal beds where it will be adsorbed onto the coal, locking it up indefinitely. It also preferentially displaces methane from the coal, which can then be used as a fuel. The estimated capacity for CO2 storage in unmineable coal beds is 148 Gt (Freund et al., 2003). These storage reservoirs need to be at a minimum depth of 800 m so that the CO2 is in a supercritical state, allowing a large amount of CO2 to be stored in a small volume. The critical point for CO2, which is the highest pressure and temperature at which CO2 can exist as a vapour and liquid in equilibrium, is 73.8 bar and 31.1 °C.

12.2

Leak scenarios

Carbon dioxide captured from its production source will be transported to the storage site via pipeline, marine tanker or rail and road tankers, all of which have the potential for leakage. There is a risk of a catastrophic leak due to pipeline failure, or of smaller leaks from joints. The impact of these leaks will depend on whether the pipelines are over or under the ground. A leak from a high-pressure pipeline will cool drastically in the process of adiabatic expansion. A leak that is directed downwards from an above-ground pipeline will cause the formation of a patch of frozen CO2 which then creates a secondary hazard as it sublimes, (Mazzoldi et al., 2008). This hazard is of particular concern under stable atmospheric conditions with low wind speeds when the dense gas plume tends to follow the topography. With a typical daytime, unstable atmosphere, the CO2 release is expected to mix rapidly into the surrounding atmosphere. Similar issues might be expected with a buried pipeline where the mass of soil acts to slow the release to the atmosphere. With a smaller leak released below ground, the CO2 will diffuse through the


326


soil and into the atmosphere, and may affect both the vegetation and the soil ecosystem. A similar range of responses may be anticipated with leaks from underground storage. Leakage from an underground storage site may occur due to failure of the sealing cap, allowing for a mass outpouring of CO2, or microseepage through fractures in the caprock and overlying formations, which may be increased by the pressurization of a reservoir necessary for injection (Klusman, 2003a). Studies of natural analogue sites indicate that gas migration is typically channeled along discrete, high-permeability pathways within the faults, with release to the surface occurring from localized gas vents, (Annunziatellis et al., 2008). However, leakage mechanisms at such sites may not be typical as they tend to be heavily faulted and are unlikely to meet the long-term stability criteria for CO2 storage. Over 2500 km of CO2 pipelines already exist in the USA, where they carry 50 Mt yr–1 of compressed CO2 to EOR projects. The corrosion rate of steel in supercritical CO2 is around 0.01 mm yr–1, but increases to 0.7 mm yr–1 if free water is present (IPCC, 2005), so pipelines need to be monitored for corrosion and leakage. The Cortez pipeline in Colorado, which passes through two built-up areas, is buried at least 1 m deep and is patrolled by air every two weeks to monitor for encroachment and evidence of leakage (IPCC, 2005). Failure and accident rates of CO2 pipeline systems should be comparable with those of existing oil and gas pipeline systems. The effect of historical differences in engineering practice is that most leakage from natural gas pipelines in the UK occurs from joints whereas in the USA it mostly arises due to corrosion (Williams, 1994). In the 1980s, estimates of 0.01–0.02 m3 km–1 hr–1 mbar–1 were quoted for leaks of natural gas from UK pipelines (Williams, 1994). IPCC (2005) suggests that US and European failure rates have markedly decreased, with failures falling to 0.0002 km–1 yr–1 in 2002 for small pipelines (less than 100 mm diameter). For pipelines of 500 mm diameter or more the failure rate is 0.00005 km–1 yr–1. Total leakage from Russian natural gas pipelines has been estimated as 1.0–2.5 % (Lelieveld et al., 2005). Between 1990 and 2002, there were 10 incidents involving CO2 pipelines worldwide, corresponding to an incident rate of 0.0003 km–1 yr–1 (Gale and Davison, 2003). The reasons for the leaks were relief valve failure, weld/valve packing failure, corrosion and outside force. If impurities such as H2S or hydrocarbons are present in the transported CO2, then this could affect the potential impacts of a pipeline leak or failure. CH4 will be ubiquitous in hydrocarbon reservoirs and coal beds and relatively common in deep saline aquifers; thus it is also necessary to consider methane and light alkane migration along with the migration of CO2 from sequestration projects (Klusman, 2003a). Evidence from oil and gas fields suggests that hydrocarbons and other gases and fluids, including CO2, can remain trapped for millions of years. However, natural traps can leak, as in the case of the McElmo dome (Gerlach



327

et al., 1998; Stevens et al., 2000), which reinforces the need for careful storage site selection (IPCC, 2005). Documented incidents suggest that out of 600 natural gas storage reservoirs, which are mostly converted gas and oil fields and a few aquifer storage projects, significant leakage occurred in only 10, including five events due to loss of well bore integrity, four due to leaks in the caprocks and one due to reservoir selection (Perry, 2005). No estimates exist of the total release, but in one major incident 3000 t of gas was emitted – less than 0.002 % of the total gas in storage in North America (IPCC, 2005). Klusman (2003b) measured microseepage of CH4 and CO2 at the EOR operation at Rangely, Colorado. Currently 5.1 Mm3 day–1 (9.3 ¥ 103 t day–1) of CO2 is injected into the oil field, of which 80 % returns to the surface with the recovered oil and is reinjected. Estimated microseepage rates across the 78 km2 area of the Rangely field are between 170 and 3800 t yr–1, which may include some conversion of seeping CH4 into CO2 by bacteria. In the absence of precise estimates of leakage rates from geological storage, modelling has been applied to evaluate the effects of various leakage scenarios. Hepple and Benson (2005) found that for seepage rates of 0.01 % and 0.001 % per year, 90 and 99 %, respectively, of the total stored CO2 would remain underground for 1000 years. If the seepage rate were 1 % per year, then most of the carbon would return to the atmosphere after 400 years, demonstrating that geologic storage would not be effective if seepage rates were this high. Hepple and Benson (2005) concluded that to stabilize atmospheric [CO2] at between 350 and 550 ppm required a seepage rate less than 0.01 %. Stone et al. (2009) modelled various climate scenarios and found that even with a leakage rate of 1 %, there would be a reduction in climate change over the next 100 years. However, in modelling the economics, van der Zwaan and Smekens (2009) found that at a leakage rate of 0.1 % per year, CCS would offer useful mitigation of climate change, but a leakage rate of 1 % would add considerably to the marginal costs of CO2 and render CCS an ineffective option. The presence of impurities can also significantly affect the level of risk in the event of leakage. H2S is more toxic than CO2, and SO2 creates a stronger acid in groundwater than CO2, leading to greater mobilization of metals in the groundwater. The gases stored at Weyburn contain approximately 2 % H2S (IPCC, 2005).

12.3

Impacts of terrestrial leakage

At low concentrations, CO2 is not directly hazardous to human health, but at concentrations above 2 % it has a strong effect on respiration and at 7–10 % it can cause unconsciousness and death (Airgas, 2002; Pfanz et al., 2004). Lower concentrations may also detrimentally affect environmental processes. Vertical


328


migration of leaking CO2 will lead to dissolution into shallow groundwaters and production of carbonic acid which would reduce pH. A reduction in pH may lead to mobilization of toxic metals, leaching of biological nutrients and modification of proton gradients across biological membranes (Bruant et al., 2002). Leaked CO2 and any accompanying substances may also affect the flora and fauna with which they come into contact. A well-aerated soil should have a CO2 concentration close to that of the atmosphere but, because of biological activity, [CO2] may be between 0.15 and 2.5 % in the surface layers of the soil (Stolwijk and Thimann, 1957; Russell, 1973) and occasionally figures up to 10 or 12 % have been recorded (Chang and Loomis, 1945; Stolwijk and Thimann, 1957; Russell, 1973; Glinski and Stepniewski, 1985). [CO2] increases with depth and soil moisture and rapidly increases after rain because its diffusion through the gas phase is restricted by water saturation (Yoshioka et al., 1998). Leaked CO2 from underground storage facilities may have several combined effects at the surface: ∑ ∑

it may cause soil [CO2] to rise (possibly almost to 100 %); it may diffuse out of the soil and cause a rise in the atmospheric concentration above the surface; ∑ it may also dissolve in the soil water and potentially be taken up by plants in the transpiration stream. Depending on the soil type, the dissolution of CO2 may either decrease pH or, if HCO3– ions accumulate, cause an increase in pH.

These different situations will have different effects on fauna and flora in the surface ecosystem. Whilst elevated [CO2] in the atmosphere can stimulate plant growth, elevated soil [CO2] will usually be detrimental to plant growth (IPCC, 2005). Few investigations have studied the effect of increased soil [CO2] on a field scale, although laboratory investigations on individual plants have been carried out. Emissions from natural sources, such as volcanic springs, or from landfill are the closest analogues to leakage from a CO2 storage site.

12.3.1 Laboratory studies of carbon dioxide (CO2) enrichment Noyes (1914) first studied the effect of soil CO2 saturation on the growth of tomato and maize. Severe stress symptoms were observed. Starting from the lower parts of the plants, leaves drooped, turned brown and curled up and veins darkened. At the end of two weeks, the plants were fully brown, but the maize recovered after CO2 treatment was halted, while the tomatoes died. Boru et al. (2003) found severe effects on shoot and root growth in soybean, with leaf chlorosis occurring after only two days of treatment.



329

They found that soybean survival was not affected by oxygen deficiency but that 50 % CO2 caused 25 % mortality. However, rice plant growth was unaffected by soil CO2, although differences in root length were detected. They suggest that the enhanced resistance of rice to the effects of CO2 is due to the presence of a gas transport pathway in the plant stems of rice which enables them to withstand flooding (which often causes [CO2] to exceed 50 %). Severe effects on root growth have also been found in sorghum (Matocha and Mostaghimi, 1988), peas, beans, field beans and sunflower (Stolwijk and Thimann, 1957). Conversely, oats and barley showed little response to CO2 (Stolwijk and Thimann, 1957). Similar responses were observed by Geisler (1967) in barley and pea: he found that barley was not affected at O2 levels between 21 % and 7 % or 0–2 % CO2; in peas, root length was already reduced at 14 % O2; CO2 at 8 % was inhibiting at all O2 concentrations. Geisler also found a stimulatory effect on root growth at CO2 levels of up to 2 % when O2 was limiting. In creeping bentgrass, Bunnell et al. (2002) found root damage at high CO2 concentrations, but Rodriguez et al. (2005) in a similar study found stress effects only when high CO2 was combined with high temperature. Leonard and Pinckard (1946) found that concentrations of CO2 between 21 and 60 % were needed to inhibit root growth in cotton. Concentrations of 60 % and higher inhibited all root growth, and top growth was also poor. Chang and Loomis (1945) suggested that soil [CO2] below 10 % was probably not toxic but that concentrations above 15 % may be lethal to some plants. Williamson, (1968) used various mixtures of N2, CO2 and O2 in air to determine whether low O2 or high CO2 was more important in root damage to field bean. All air mixtures containing more than 6 % CO2 effectively stopped cell division, and exposure to lesser concentrations produced lasting damage. The secondary root system was particularly sensitive to damage by CO2. Another effect of elevated soil CO2 observed in some species is enhanced seed germination. The higher soil CO2 achieved after rainfall is thought to be a signal for germination of barnyard grass seeds in a state of dormancy (Yoshioka et al., 1998). However, work in our own CO2 injection facility showed that field bean germination was completely inhibited by soil CO2 above about 20 % (Manal Al-Traboulsi, unpublished PhD thesis). Several mechanisms are involved in the responses of plants to soil CO2. Decreases in nutrient uptake were observed by Chang and Loomis (1945) and Kimball et al., (1986). Matocha and Mostaghimi (1988) found in sorghum that soil CO2 enrichment depressed the response to applied Fe leading to iron deficiency-induced chlorosis, possibly due to an increase in pH, and that phosphorus uptake was reduced by almost 90 %. Bunnell et al. (2002) found that high CO2 levels decreased the cytoplasmic pH of root cells in creeping bentgrass, thereby interfering with nutrient and water uptake and


330


stunting root growth. However, CO2 in solution has been found to have a greater toxic effect on green algae than the same pH generated by different acids (Fox, 1933). Glinka and Reinhold (1962) found that soil CO2 inhibited water uptake in sunflower, by 40 % after three minutes, probably due to an alteration in the permeability of the osmotic barriers. In their review of studies of the application of CO2-enriched water, Enoch and Olesen (1993) pointed out a number of methodological issues and concluded that constant CO2 addition was required to show consistent effects. Nevertheless, taking into account the diversity of experimental approaches, these studies indicate that soil CO2 causes damage to root systems independently of the effects of oxygen deprivation and in excess of its effects on pH, and that some species are more susceptible than others.

12.3.2 Man-made analogues of soil carbon dioxide (CO2) enrichment Elevated soil CO2 has been studied by researchers looking into the effects of landfill gas on vegetation. The main constituents of landfill gas are methane (55–60 %), CO2 (~40 %) and low amounts of oxygen (Gilman et al., 1982; Arthur et al., 1985). Chan et al. (1991) studied landfill gas effects on the root growth of trees in a simulation tank. At the bottom of the tank, where [CO2] was 30–50 %, taproot growth of most species was suppressed by high CO2. In the centre, mixing with ambient air reduced the [CO2] range to 5–30 %, which stimulated the longitudinal growth of adventitious roots which formed a horizontal root system in the middle zone. The limited depth of the root system left the trees susceptible to water stress. Low [O2] in soil had less effect than high [CO2]. Zhang et al. (1995) compared the effects on lead tree (Leucaena leucocephala) of simulated landfill gas to those caused by flooding. Stem elongation and stomatal resistance were affected after two days of flooding whereas the effect of gas fumigation was milder and appeared later (4–6 days). Root nodule activity was inhibited severely by all treatments, with high [CO2] having a more severe, and irreversible, effect than low [O2]. Another analogue is leakage from natural gas pipelines. In addition to the displacement of soil O2 by the natural gas, the soil atmosphere has elevated [CO2] and depleted [O2] due to oxidation of methane by methanotrophic bacteria (Steven et al., 2006). Smith et al. (2005a) studied grass, bean, barley and oilseed rape in plots that were injected with natural gas. Visible stress symptoms, caused by oxygen deficiency and increased soil [CO2] occurred within eight days for oilseed rape, 44 days for grass and four months for wheat and bean.



331

12.3.3 Natural analogues of soil carbon dioxide (CO2) enrichment The main visible characteristic at the surface of long-term elevated CO2 zones is lack of vegetation, and new CO2 releases into vegetated areas cause noticeable die-off. Following a swarm of earthquakes at Mammoth Mountain, California in 1989, geologists discovered large volumes of CO2 seeping to the surface. The area of dead and dying trees now covers 50 ha, with their roots killed by high soil [CO2] interfering with water and nutrient uptake (Gerlach et al., 1998; Sorey et al., 2000). In the area of tree kill, CO2 makes up 20–95 % of the gas content of the soil with efflux rates greater than 500 g m–2 d–1. The total rate of gas emission at Mammoth Mountain is close to 300 t day; this corresponds to over 10 % of the rate of CO2 sequestration at the Sleipnir oilfield and, in the CCS context, would correspond to a catastrophic failure. Where leaks from natural carbon dioxide springs (NCDS) have occurred over long periods of time, any vegetation present will have adapted to the higher levels of CO2. Vodnik et al. (2006) found patchy patterns of CO2 enrichment with limited lateral diffusion. Studies have been carried out at Stavešinci, NE Slovenia (Vodnik et al., 2002; Pfanz et al., 2004; Ma�ek et al., 2005), Northland, New Zealand (Newton et al., 1996) and at over 100 geothermal CO2 springs in Italy (Miglietta et al., 1993). In addition to soil [CO2] up to 100 % in some areas, atmospheric concentrations can reach as high as 0.5 % (5000 ppm) at 50 cm above the soil surface and the gas may also be contaminated with methane, nitrogen and sulphurous compounds (Pfanz et al., 2004), so it is not always certain which factor is the primary agent of stress. Kaligaric (2001) produced a vegetation map around the NCDS at Radenci, Slovenia. In the zone of highest [CO2], there was almost no vegetation; in the intermediate zone, the normal vegetation patterns changed due to reduced competition between different species. At Stavešinci CO2 spring, NE Slovenia, plant height corresponded directly to the soil [CO2], with the smallest plants being closest to the vent. This behaviour was noted in several species of both C3 and C4 plants (Vodnik et al., 2002; Pfanz et al., 2004; Ma�ek et al., 2005). Turk et al. (2001) and Ma�ek et al. (2005) found that shoots of Juncus effusus were taller and broader with increasing distance from a CO2 vent and that closer to the vent, the epidermal cells were thicker. Ma�ek et al. (2005) found that decreases in root respiration occurred only when roots were exposed to very high concentrations (up to 48 %) of CO2.

12.3.4 Elevated carbon dioxide (CO2) effects on soil fauna Elevated concentrations of soil CO2 may also affect organisms living in the soil. Sustr and Simek (1996) found behavioural reactions in soil invertebrates © Woodhead Publishing Limited, 2010

332


ranging from changes of movement to temporary paralysis in all species under study. BD50 (the concentration at which visible behavioural responses were seen in 50 % of animals) ranged from 2–39 % CO2 depending on the species. Total paralysis was found in all species at higher CO2 levels (10–59 %), and CO2 was lethal in some species at levels between 11 and 50 % CO2. Atmobiotic springtails and terrestrial isopods were extremely sensitive, while higher resistance was observed in millipedes, earthworms, centipedes and insects. In contrast, Chan et al. (1997) found that soil fauna in Hong Kong landfill sites (4–11 % CO2) were three to five times denser than in the reference sites and concluded that, although the cover soil possesses a high landfill gas content (including CO2), this may not be a limiting factor for the bioactivity. However, the study by Sustr and Simek (1996) was of the effect of a sudden influx of CO2 and may be more applicable to the case of leakage from CCS than Chan et al. (1991) in which the invertebrates had time to adapt.

12.4

Atmospheric enrichment of carbon dioxide (CO2)

Leakage from underground storage sites will elevate atmospheric as well as soil [CO2]. This topic has been heavily researched because of the continuing global increase in atmospheric [CO2]. Much of the work on the atmospheric effects of CO2 on vegetation has been carried out in open-top chambers (OTCs) or free-air carbon dioxide enrichment (FACE) systems. Ainsworth and Long (2005) reviewed data from 120 peer reviewed FACE studies. Plants grown in elevated CO2 are typically larger than those grown in ambient CO2 (Pritchard et al., 1999). Increases in atmospheric [CO2] have direct and immediate effects on the rate of photosynthetic CO2 assimilation and stomatal conductance. A doubling of atmospheric CO2 is expected to lead to an increase of 30–50 % in the instantaneous rate of photosynthesis and net primary production (Kimball et al., 1993; Van Noordwijk et al., 1998; Ghannoum et al., 2000; Moscatelli et al., 2001). Soussana et al. (1996) found an increase of 14–19 % in above-ground biomass in a perennial ryegrass sward at doubled CO2. These effects are also thought to increase water use efficiency (Nowak et al., 2004). In the Swiss FACE experiment, the longest-running grassland CO2 fumigation project, differences were found between species: there was a 10 % average increase in the yield of perennial ryegrass (Lolium perenne) compared to 25 % for white clover (Trifolium repens). Elevated CO2 also had a greater effect on the yield of legumes than of grasses (Luscher et al., 2004). In OTC experiments with enhanced CO2 on the shortgrass steppe (semiarid, mixed grassland on the western edge of the American Great Plains), mid- and whole-season production was enhanced by 26–47 %. Elevated CO2



333

enhanced above-ground biomass by 35 % at the summer harvest and 41 % for total annual production (Morgan et al., 2004). Moscatelli et al. (2001) also found species-dependent effects of elevated CO2: holm oak (Quercus ilex) produced more shoots, which were shorter (but not thinner), and more smaller leaves but less leaf area in total, whereas mock privet (Phillyrea angustifolia) showed a reduction in the number of shoots; mastic tree (Pistacia lentiscus) had no increase in shoot numbers, but they were longer and the leaves were larger. Plants exchange gases with the atmosphere, including CO2 and water vapour, through their stomata – leaf pores that vary in aperture in response to environmental conditions. Elevated atmospheric CO2 causes a reduction in conductance of the stomata which could reduce water use by vegetation, with the greatest effect in crop plants. A doubling of CO2 reduces mean midday conductance by less than 15 % in some crops compared to over 50 % in others. Simulations indicate that the large reductions in stomatal conductance in crops would translate into reductions of less than 10 % in evapotranspiration, partly because of increases in temperature and decreases in humidity in the air around the leaves (Bunce, 2004). Plant species have different biochemical mechanisms or pathways for photosynthesis, and it is expected that there will be differences in photosynthesic response between plants with the C3 pathway (the majority) and the C4 pathway (most other species and particularly prevalent in the tropics). C3 photosynthesis is limited by present-day atmospheric CO2 concentrations while C4 photosynthesis is nearly CO2 saturated. Early predictions suggested that increasing atmospheric [CO2] would favour C3 over C4 species, thus leading to changes in species composition. Yields of wheat (C3) would increase relatively more than yields of maize (C4) and yields of sugar beet (C3) relatively more than those of sugarcane (C4) (Kimball et al., 1993). Maize and sorghum are two major C4 crops that tend to show little response to elevated atmospheric CO2 under well-watered conditions (Ghannoum et al., 2000). However, although CO2-induced growth responses of C3 species exceed those in C4 species, the differences are not as great as expected (Morgan et al., 2004). Elevated CO2 slows transpiration by inducing the closure of guard cells that form stomata on leaf surfaces; this contributes to the increase in water use efficiency. Morison (1987) showed that a doubling of [CO2] to 660 ppm reduced stomatal conductance to 60 % of that at 330 ppm. He also showed that there is no significant difference between C3 and C4 plants in the response of their stomatal conductivity to increasing CO2. Thus maize should derive as much drought tolerance as wheat, which may be particularly important in water-limited croplands. Rogers et al., 1994) pointed out that the effect on transpiration is far more important for C4 than for C3 plants, noting that water stress in soybean was ameliorated by increased [CO2]. On average, the growth stimulation of C4 plants by a


334


doubling of the current [CO2] is about 22–33 % compared with 40–44 % for C3 plants. The magnitude of the growth stimulation in C4 plants increases with decreasing soil water availability (Ghannoum et al., 2000). The differential responses of plant species suggest that relative competitiveness may be altered and that weeds with the C3 photosynthetic pathway could outcompete C4 crops whereas C4 weeds would be less competitive against C3 crops (Kimball et al., 1993; Rogers et al., 1994; Ghannoum et al., 2000). Derner et al. (2003) found that CO2 enrichment increased leaf area and above-ground biomass in cotton (C3), but sorghum (C4) responses did not differ. In mixed cropping cultures, leaf area of sorghum was reduced with elevated CO2 in both low- and high-density mixtures, but not in monoculture. Rogers et al. (1994) found that root dry weight was increased under elevated atmospheric CO2, regardless of species or study conditions because, when nutrients and water are non-limiting, more carbon is allocated to below-ground structures. Most studies have found that increased atmospheric CO2 resulted in more and/or longer roots which may lead to greater penetration and spread (Fitter et al., 1996). However, some studies indicate that root biomass mainly increases in the surface layers of the soil and that root production is slow at the base of the soil profile, which could increase the risk of water stress (Fitter et al., 1996; Dilustro et al., 2002). In contrast, Moscatelli et al. (2001) found a 43 % reduction in below-ground production with the greatest reduction at 30–45 cm depth. The decrease in production in the surface layers was about 25 %. These results were in contrast to previous results from the same site which produced a 47 % increase in root growth. Kandeler et al. (1998) also found a reduction in root biomass in a terrestrial model ecosystem when subjected to elevated temperature (ambient + 2 ºC) and CO2 (ambient + 200 ppm).

12.5

Leak monitoring techniques

Leak detection is important for several reasons: first, any leakage will negate the original purpose of the storage; second, accurate data on the quantity of carbon that has been sequestered will be needed for trading and accounting purposes; third, the leaking CO2 might damage surface ecosystems, as discussed earlier. Bachelor et al. (2008) pointed out that, as with the natural analogues discussed earlier, potential leakage pathways are not necessarily known a priori, so that leak monitoring may be required over a region of 100 km2, as well as in the immediate vicinity of the injection well. A number of methods were reviewed by Leuning et al. (2008). Flux chambers are suitable for slow leaks but require extensive spatial sampling. Similar considerations apply to direct sampling of soil [CO2] anomalies. Micrometeorological techniques such as eddy covariance or flux gradient techniques are appropriate for large area sources. However, such methods are sensitive to biogenic fluctuations



335

in CO2. A field test by Lewicki et al. (2009) found that a release rate of 0.3 t CO 2 d –1 was detectable by eddy covariance, but a release of 0.1 t CO2 d–1 was indistinguishable from natural background. An alternative approach suggested by Carroll et al. (2009) is to monitor pH and carbonate chemistry in groundwater, but this approach is dependent on CO2 solubility and buoyancy, the rate of leak flux and groundwater flow. To circumvent the confusion with biogenic CO2, tracer gases – either naturally occurring or introduced – may be used. White et al. (2005) point out that chemical tracers need to be introduced into the CO2 stream from the start of sequestration in order to ensure a uniform distribution. There is also an issue with differential rates of migration. Natural tracers, however, would be well mixed; White et al. suggest the use of argon. Our own research has tested the ability to detect anomalies (d13C) of CO2 in soil subjected to artificial injection of industrial CO2. Biogenic CO2 has a d13C value (relative to standard) of about –23 ‰. The injected gas was isotopically lighter (–31 to –34 ‰) and could be readily distinguished from biogenic CO2 down to concentrations of about 3 %. CO2 derived from natural gas is isotopically light, but many fossil fuels such as coal and many oils are isotopically similar to biogenic CO2 and would be difficult to distinguish from background (A. Cheung and M. Steven, unpublished report). Bachelor et al. (2008) suggested introducing 14CO2 as a sensitive tracer, specifically to identify leakage pathways, with the advantage (as with 13CO2) that it is chemically identical to the stored gas. A field test of perfluorocarbon tracers was conducted by Wells et al. (2007) to assess leakage from an enhanced oil recovery scheme over a period of 27 months. The technique was sufficiently sensitive to provide an estimated CO2 leakage rate of 0.0085 % per annum. Remote sensing systems cover large areas and would, in principle, circumvent the spatial sampling issues associated with ground-based monitoring. Carbon dioxide can be measured directly by remote sensing: Spinettia et al. (2008) mapped [CO2] in a volcanic plume, in the range 40–350 ppmv above ambient, using an airborne system that measured CO2 absorption lines between 1900 and 2100 nm, but the sensitivity of this method is limited by variability of the background surface. An alternative approach using the stress responses of vegetation to detect leaks of natural gas from pipelines was developed by Smith el al. (2004). As noted earlier in this chapter, plants subjected to soil gassing exhibit classic symptoms of stress such as chlorosis, either (as in the case of natural gas leakage) due to deprivation of oxygen at the roots, or due to direct physiological effects of CO2. Hyperspectral techniques can identify the signs of such stress at sub-visual levels (Smith et al., 2004; Noomen et al., 2008); our recent work has shown similar responses to enhanced soil CO2 (Fig. 12.1). Although it has not as yet proved possible to distinguish stress induced by root asphyxiation from other causes of stress (Smith et al., 2005b), the spatially limited context of expected leakage from CCS wells


336


2.5 2.0

Zeta

1.5 1.0 Gassed

0.5

Control Remote control

0.0 01/05/2006

01/06/2006

01/07/2006 01/08/2006 Date

01/09/2006

01/10/2006

12.1 Temporal variations of a hyperspectral index, zeta on plots of grass at the ASGARD facility. The gassed plots received a continuous injection of CO2 into the soil and had soil concentrations up to 50 %. There were two sets of control plots, the remote plots being at a greater distance. Zeta is the ratio of two peaks in the first derivative spectrum of canopy reflectance, at 725 and 702 nm and has been found to be a sensitive indicator of plant stress (Smith et al., 2004). The results show small differences in the early part of the season, when it was very dry and all plots were stressed, but larger differences appear from July, after rainfall.

means that remote sensing of vegetation anomalies can be applied as a tool for effective targeting of ground survey techniques.

12.6


A catastrophic leak, particularly from a pipeline, presents an immediate human and environmental hazard. Further research is ongoing to quantify and ameliorate the risks of such events, as discussed in other chapters. For non-catastrophic underground leaks, the effect on overlying ecosystems of CO2 leakage depends on the concentrations of CO2 reached in both the atmosphere and the soil, together with the duration of the event and external environmental factors. The mechanisms of response to CO2 in the near-surface environment are complex, and there are considerable variations in sensitivities between different plants and soil organisms. In the event of long-term leakage, such differential sensitivities will impact on the ecosystem balance. Further research is required into the interactions of CO2 with ecosystems and their dependence on species and environmental conditions. The study of natural analogues, such as CO2 springs, continues to provide valuable insights. The interactions are complicated by CO2 enrichment of the air above the leak, which may enhance plant growth, together with stress effects generated in the soil. Three zones are typically observed: a zone of



337

high CO2 concentration where vegetation is absent; an intermediate zone where the competition between different species is affected; and an outer zone where soil concentrations are not limiting, but enhanced atmospheric [CO2] increases vegetation growth (Kaligaric, 2001; Vodnik et al., 2002; Pfanz et al., 2004; Ma�ek et al., 2005). However, such sites provide only a limited analogue for leakage of CO2 from CCS, because other gases are involved, because local ecosystems have adapted over a long period and because the geology of such areas is untypical of sites likely to be selected for CCS. A few studies have begun to systematically investigate the effects of elevated soil [CO2] in non-adapted sites on a field scale. The Zero Emission Research and Technology Center (ZERT) at Montana State University, USA, simulates the effect of leakage by controlled releases from a buried pipeline (Lewicki et al., 2009). The emphasis is on geochemical and hydrological issues related to underground CO2 storage, their measurement and modelling. A similar facility is under development in Australia. An alternative approach, with a greater focus on biological impacts, is applied in the Artificial Soil Gassing and Response Detection (ASGARD) facility at the University of Nottingham, where CO2 is injected into a series of field plots, allowing replicated experiments to be conducted on a range of crops (Fig. 12.2). The facility is similar to one previously reported for the study of natural gas leakage (Smith et al., 2005b), with added control. Investigations conducted in the ASGARD programme include leak detection techniques (remote sensing

12.2 View of the ASGARD field site, showing the grass plots, CO2 injection pipes and gas monitoring tubes.


338


and isotopic analysis as discussed earlier), responses of plant root systems, differential plant sensitivities and their effects on competition, CO2 fluxes and pathways in soil, effects of SO2 contamination in the CO2 stream, soil and soil-water chemistry, effects on soil microbial communities and ecosystem recovery after gassing. It is particularly important to develop an understanding of the biological processes that control plant and ecosystem responses. The impact of leakage on the flora and fauna in the biosphere above a CO2 reservoir needs to be taken into consideration before selecting CCS storage sites. It would be prohibitive to investigate all possible species and ecosystems; an insight into the processes underlying system responses to CO2 leaks will allow predictions to be made and tested and progress to be made faster.

12.7


IPCC (2005) Special report on carbon dioxide capture and storage. 431 pp, Intergovernmental Panel on Climate Change, Geneva.

12.8

References

Ainsworth E and Long S (2005) What have we learned from 15 years of free-air CO2 enrichment (FACE)? A meta-analytic review of the responses of photosynthesis, canopy properties and plant production to rising CO2. New Phytologist, 165, 351–372. Airgas (2002) Carbon Dioxide-Material Safety Data Sheet. Document 1013. Airgas Inc., Radnor, PA. Annunziatellis A, Beaubien SE, Bigi S, Ciotoli G, Coltella M and Lombardi S (2008) Gas migration along fault systems and through the vadose zone in the Latera caldera (central Italy): Implications for CO2 geological storage. International Journal of Greenhouse Gas Control, 2, 353–372. Arthur J, Leone I and Flower F (1985) The response of tomato plants to simulated landfill gas mixtures. Journal of Environmental Science and Health, A20, 913–925. Bachelor PP, McIntyre JI, Amonette JE, Hayes JC, Milbrath BD and Saripalli P (2008) Potential method for measurement of CO2 leakage from underground sequestration fields using radioactive tracers. Journal of Radioanalytical and Nuclear Chemistry, 277, 85–89. Boru B, Vantoai T, Alves J, Hua D and Knee M (2003) Responses of soybean to oxygen deficiency and elevated root-zone carbon dioxide concentration. Annals of Botany, 91, 447–453. Bruant R, Michael C, Guswa A and Peters C (2002) Safe storage of CO2 in deep saline aquifers. Environmental Science and Technology, 36, 240A–245A. Bunce JA (2004) Carbon dioxide effects on stomatal responses to the environment and water use by crops under field conditions. Oecologia, 140, 1–10. Bunnell T, McCarty LB, Dodd RB, Hill HS and Camberato JJ (2002) Creeping bentgrass growth response to elevated soil carbon dioxide. Hortscience, 37, 367–370. Carroll S, Hao Y and Aines R (2009) Geochemical detection of carbon dioxide in dilute aquifers. Geochemical Transactions, 10, 4.



339

Chan G, Wong M and Whitton B (1991) Effects of landfill gas on subtropical woody plants. Environmental Management, 15, 411–431. Chan G, Chu L and Wong M (1997) Influence of landfill factors on plants and soil fauna – an ecological perspective. Environmental Pollution, 97, 39–44. Chang H and Loomis W (1945) Effect of carbon dioxide on absorption of water and nutrients by roots. Plant Physiology, 20, 220–232. Defra (2006) UK-emissions of greenhouse gases-latest figures. e-Digest of environmental statistics. Department for Environment, Food and Rural Affairs, London, UK. Derner JD, Johnson HB, Kimball BA et al. (2003) Above- and below-ground responses of C-3-C-4 species mixtures to elevated CO2 and soil water availability. Global Change Biology, 9, 452–460. Dilustro JJ, Day FP, Drake BG and Hinkle CR (2002) Abundance, production and mortality of fine roots under elevated atmospheric CO2 in an oak-scrub ecosystem. Environmental and Experimental Botany, 48, 149–159. DTI (2002) The feasibility of carbon dioxide capture and storage in the UK – Project definition and scoping paper. DTI, London, UK. DTI (2005) Geoscientific monitoring of the Weyburn CO2 flood. A multidisciplinary approach to CO2 monitoring. Project summary 281. Cleaner fossil fuels programme, DTI, London, UK. Enoch H and Olesen J (1993) Tansley Review No 54. Plant response to irrigation with water enriched with carbon dioxide. New Phytologist, 125, 249–258. Fitter A, Self G, Wolfenden J et al. (1996) Root production and mortality under elevated atmospheric carbon dioxide. Plant and Soil, 187, 299–306. Fox D (1933) Carbon dioxide narcosis. Journal of Cell and Comparative Physiology, 3, 75–100. Freund P, Gale J, Davison J and Riley N (2003) Capture and geological storage of carbon dioxide – a status report on the technology. IEA Greenhouse Gas R&D Programme. COAL R223, British Geological Survey, London, UK. Gale J and Davison J (2003) Transmission of CO2-safety and economic considerations. In Gale J and Kaya Y (eds), Proceedings of the Sixth International Conference on Greenhouse Gas Control Technologies, Elsevier (Pergamon), Oxford, UK, 1, 517–522. Geisler G (1967) Interactive effects of CO2 and O2 in soil on root and top growth of barley and peas. Plant Physiology, 42, 305–307. Gerlach T, Doukas M, McGee K and Kessler R (1998) Three-year decline of magmatic CO2 emissions from soils of a Mammoth Mountain tree kill: Horseshoe Lake, CA, 1995–1997. Geophysical Research Letters, 25, 1947–1950. Ghannoum O, von Caemmerer S, Ziska L and Conroy J (2000) The growth response of C4 plants to rising atmospheric CO2 partial pressure: a reassessment. Plant, Cell and Environment, 23, 931–942. Gilman E, Leone I and Flower F (1982) Influence of soil gas contamination on tree root growth. Plant and Soil, 65, 3–10. Glinka Z and Reinhold L (1962) Rapid changes in permeability of cell membranes to water brought about by carbon dioxide & oxygen. Plant Physiology, 37, 481–486. Glinski J and Stepniewski W (1985) Soil Aeration and its Role for Plants. CRC Press, Boca Raton, FL. Hepple RP and Benson SM (2005) Geologic storage of carbon dioxide as a climate change mitigation strategy: performance requirements and the implications of surface seepage. Environmental Geology, 47, 576–585.


340


IPCC (2005) IPCC Special Report on Carbon Dioxide Capture and Storage, Working Group III of the Intergovernmental Panel on Climate Change, Metz B, Davidson O, de Coninck HC, Loos M and Meyer LA (eds), Cambridge University Press, Cambridge, UK. IWR (2009) Climate protection just a lot of hot air – Global CO2 emissions at record levels, Press Release, 1 August 2009, Internationales Wirtschaftsforum Regenerative Energien, Muenster, Germany, available at: http://www.iwrpressedienst.de/iwr/Global-CO 2emissions-2008-renewable-energy-investment-plan.pdf (accessed January 2010). Kaligaric M (2001) Vegetation patterns and responses to elevated CO2 from natural CO2 springs at Strmec (Radenci, Slovenia). Acta Biologica Slovenica, 44, 31–38. Kandeler E, Tscherko D, Bardgett RD, Hobbs PJ, Kampichler C and Jones TH (1998) The response of soil microorganisms and roots to elevated CO2 and temperature in a terrestrial model ecosystem. Plant and Soil, 202, 251–262. Kimball BA, Mauney JR, Radin J et al. (1986) Effects of increasing atmospheric CO2 on the growth, water relations and physiology of plants grown under optimal and limiting levels of water and nitrogen. Progress Research Paper 039, United States Water Conservation Laboratory, Phoenix, AZ. Kimball BA, Mauney JR, Nakayama FS and Idso SB (1993) Effects of increasing atmospheric CO2 on vegetation. Vegetatio, 104, 65–75. Klusman RW (2003a) Evaluation of leakage potential from a carbon dioxide EOR/ sequestration project. Energy Conversion and Management, 44, 1921–1940. Klusman RW (2003b) A geochemical perspective and assessment of leakage potential for a mature carbon dioxide-enhanced oil recovery project and as a prototype for carbon dioxide sequestration; Rangely field, Colorado. American Association of Petroleum Geologists Bulletin, 87, 1485–1507. Lelieveld J, Lechtenbohmer S, Assonov S, Brenninkmeijer C, Dienst C, Fischedick M and Hanke T (2005) Low methane leakage from gas pipelines. Nature, 434, 841–842. Leonard O and Pinckard J (1946) Effect of various oxygen and carbon dioxide concentrations on cotton root development. Plant Physiology, 21, 18–36. Leuning R, Etheridge D, Luhar A and Dunse B (2008) Atmospheric monitoring and verification technologies for CO2 geosequestration. International Journal of Greenhouse Gas Control, 2, 401–414. Lewicki JL, Hilley GE, Fischer ML, Pan L, Oldenburg CM, Dobeck L and Spangler L (2009) Eddy covariance observations of surface leakage during shallow subsurface CO2 releases. Journal of Geophysical Research, 114, D12302. Luscher A, Daepp M, Blum H, Hartwig UA and Nosberger J (2004) Fertile temperate grassland under elevated atmospheric CO2 – role of feed-back mechanisms and availability of growth resources. European Journal of Agronomy, 21, 379–398. Macˇ ek I, Pfanz H, Francetic V, Batic F and Vodnik D (2005) Root respiration response to high CO2 concentrations in plants from natural CO2 springs. Environmental and Experimental Botany, 54, 90–99. Marland G, Boden T and Andres R (2005) Global, Regional, and National CO2 Emissions. Trends: A Compendium of Data on Global Change. Carbon Dioxide Information Analysis Center, Oak Ridge National Laboratory, US Department of Energy, Oak Ridge, TN. Matocha J and Mostaghimi S (1988) Effect of carbon dioxide and iron enrichment of a calcareous soil on Fe-chlorosis, root and shoot development of grain sorghum. Journal of Plant Nutrition, 11, 1503–1515. Mazzoldi A, Hill T and Colls JJ (2008) CO2 transportation for carbon capture and



341

storage: sublimation of carbon dioxide from a dry ice bank. International Journal of Greenhouse Gas Control, 2, 210–218. Miglietta F, Raschi A, Bettarini I, Resti R and Selvi F (1993) Natural CO2 springs in Italy – a resource for examining long-term response of vegetation to rising atmospheric CO2 concentrations. Plant Cell and Environment, 16, 873–878. Morgan JA, Mosier AR, Milchunas DG, LeCain DR, Nelson JA and Parton WJ (2004) CO2 enhances productivity, alters species composition, and reduces digestibility of shortgrass steppe vegetation. Ecological Applications, 14, 208–219. Morison J (1987) Intercellular CO2 concentration and stomatal response to CO2. In Zeiger E, Farquhar GD and Cowan I (eds), Stomatal Function, Stanford University Press, Stanford, CA, 229–251. Moscatelli M, Fonck M, De Angelis P, Larbi H, Macuz A, Ramelli A and Grego S (2001) Mediterranean natural forest living at elevated carbon dioxide: soil biological properties and plant biomass growth. Soil Use and Management, 17, 195–202. Newton P, Bell C and Clark H (1996) Carbon dioxide emissions from mineral springs in Northland and the potential of these sites for studying the effects of elevated carbon dioxide on pastures. New Zealand Journal of Agricultural Research, 39, 33–40. Noomen MF, Smith KL, Colls JJ, Steven MD, Skidmore AK and van der Meer FD (2008) Hyperspectral indices for detecting changes in canopy reflectance as a result of underground natural gas leakage. International Journal of Remote Sensing, 29, 5987–6008. van Noordwijk MP, Martikainen P, Bottner P, Cuevas E, Rouland C and Dhillon S (1998) Global change and root function. Global Change Biology, 4, 759–722. Nowak RS, Ellsworth DS and Smith SD (2004) Functional responses of plants to elevated atmospheric CO2 – do photosynthetic and productivity data from FACE experiments support early predictions? New Phytologist, 162, 253–280. Noyes H (1914) The effect on plant growth of saturating a soil with carbon dioxide. Science, 40, 792. Perry K (2005) Natural gas storage industry experience and technology: Potential application to CO2 geological storage. In Thomas DC and Benson SM (eds) Carbon Dioxide Capture for Storage in Deep Geologic Formations – Results from the CO2 Capture Project Vol. 2, Geologic Storage of Carbon dioxide with Monitoring and Verification, Elsevier, Oxford, UK, 815–825. Pfanz H, Vodnik D, Wittman C, Aschan G and Raschi A (2004) Plants and geothermal CO2 exhalations – survival in and adaptation to a high CO2 environment. Progress in Botany, 65, 499–538. Pritchard S, Rogers HH, Prior SA and Peterson C (1999) Elevated CO2 and plant structure: a review. Global Change Biology, 5, 807–837. Rodriguez IR, McCarty LB, Toler JE and Dodd RB (2005) Soil CO2 concentration effects on creeping bentgrass grown under various soil moisture and temperature conditions. Hortscience, 40, 839–841. Rogers HH, Runion GB and Krupa SV (1994) Plant-responses to atmospheric CO2 enrichment with emphasis on roots and the rhizosphere. Environmental Pollution, 83, 155–189. Russell E (1973) Soil Conditions and Plant Growth. Longman Group Ltd, London. Smith KL, Steven MD and Colls JJ (2004) Use of hyperspectral derivative ratios in the red-edge region to identify plant stress responses to gas leaks. Remote Sensing of Environment, 92, 207–217.


342


Smith KL, Colls JJ and Steven MD (2005a) A facility to investigate effects of elevated soil gas concentration on vegetation. Water, Air and Soil Pollution, 161, 75–96. Smith KL, Steven MD and Colls JJ (2005b) Plant spectral responses to gas leaks and other stresses. International Journal of Remote Sensing, 26, 4067–4081. Sorey M, Farrar C, Gerlach T et al. (2000) Invisible CO2 gas killing trees at Mammoth Mountain, California. US Geological Survey – Reducing the Risk from Volcanic Hazards. Fact Sheet 172-96. US Geological Survey, California. Soussana J, Casella E and Loiseau P (1996) Long-term effects of CO2 enrichment and temperature increase on a temperate grass sward. 2: Plant nitrogen budgets and root function. Plant and Soil, 182, 101–114. Spinettia C, Carrère V, Fabrizia Buongiornoa M, Suttonc AJ and Elias T (2008) Carbon dioxide of PuùÒò volcanic plume at Kilauea retrieved by AVIRIS hyperspectral data. Remote Sensing of Environment 112, 3192–3199. Steven MD, Smith KL, Beardsley M and Colls JJ (2006) Oxygen and methane depletion in soil affected by leakage of natural gas. European Journal of Soil Science, 57, 800–807. Stevens S, Fox C and Melzer L (2000) McElmo dome and St. Johns natural CO2 deposits: Analogs for geologic sequestration. In Williams D J, Durie R A, McMullan P, Paulson C A J and Smith A Y (eds), Proceedings of the Fifth International Conference on Greenhouse Gas Control Technologies (GHGT5), CSIRO Publishing, Collingwood, VIC, Australia, 328–333. Stolwijk J and Thimann K (1957) On the uptake of carbon dioxide and bicarbonate by roots, and its influence on growth. Plant Physiology, 32, 513–520. Stone EJ, Lowe JA and Shine KP (2009) The impact of carbon capture and storage on climate. Energy and Environmental Science, 2, 81–91. Sustr V and Simek M (1996) Behavioural responses to and lethal effects of elevated carbon dioxide concentration in soil invertebrates. European Journal of Soil Biology, 32, 149–155. Turk B, Pfanz H, Vodnik D, Bernik R, Wittman C, Sinkovic T and Batic F (2001) The effects of elevated CO2 on bog rush (Juncus effusus L.) growing near a natural CO2 spring – I. Effects on shoot anatomy. Phyton, 42, 13–23. Vodnik D, Pfanz H, Macek I, Kastelec D, Lojen S and Batic F (2002) Photosynthesis of cockspur [Echinochloa crus-galli (L.) Beauv.] at sites of naturally elevated CO2 concentration. Photosynthetica, 40, 575–579. Vodnik D, Kastelec D, Pfanz H, Macek I and Turk B (2006) Small-scale spatial variation in soil CO2 concentration in a natural carbon dioxide spring and some related plant responses. Geoderma, 133, 309–316. Wells AW, Diehl JR, Bromhal G, Strazisar BR, Wilson TH and White CM (2007) The use of tracers to assess leakage from the sequestration of CO2 in a depleted oil reservoir, New Mexico, USA. Applied Geochemistry, 22, 996–1016. White CM, Smith DH, Jones KL, Goodman AJ, Jikich SA, LaCount RB, DuBose SB, Ozdemir E, Morsi BI and Schroeder KT (2005) Sequestration of carbon dioxide in coal with enhanced coalbed methane recovery – a review. Energy & Fuels, 19(3), 659–724. Williams A (1994) Methane Emissions, Report of a working group appointed by the Watt Committee on Energy. Report no 28. 171 pp, London. Williamson R (1968) Influence of gas mixtures on cell division and root elongation of broad bean. Agronomy Journal, 60, 317–321. Yoshioka T, Satoh S and Yamasue Y (1998) Effect of increased concentration of soil



343

CO2 on intermittent flushes of seed germination in Echinochloa crus-galli var. crusgalli. Plant, Cell and Environment, 21, 1301–1306. Zhang J, Liang J and Wong M (1995) The effect of high CO2 and low O2 concentrations in simulated landfill gas on the growth and nodule activity of Leucaena leucocephala. Plant Cell Physiology, 36, 1431–1438. van der Zwaan R and Smekens K (2009) CO2 Capture and storage with leakage in an energy – climate model. Environmental Modelling and Assessment, 14, 135–148.


13

Environmental risks and performance assessment of carbon dioxide (CO2) leakage in marine ecosystems J. B l a c k f o r d, S. W i d d i c o m b e and D. L o w e, Plymouth Marine Laboratory, UK, and B. Chen, Heriot Watt University, UK Abstract: This chapter describes the state of the current understanding of the potential for CO2 leaked from carbon dioxide (CO2) capture and storage (CCS) to impact the marine ecosystem. This is a complex problem as it requires an understanding of physical dispersion, the behaviour of plumes, marine chemistry, organism physiology and ecological relationships. Aside from predicting the likelihood of a leak event, the key issue is to understand the spread, persistence and impact of a hypothetical CCS derived leak and contrast this with, for example, trawling impacts and the global long-term consequences of climate change and the uptake of anthropogenically created atmospheric CO2 (ocean acidification), which CCS seeks to mitigate. Excess CO2 in the marine system is undoubtedly harmful to many organisms. In the vicinity of a leak event, it is likely that significant ecological alteration would occur. Initial research indicates that only persistent leaks of a significant proportion of reservoir capacities would cause widespread and unacceptable impacts. However, much more research is required to determine critical leak magnitudes, within sediment interactions and ecosystem recovery before any comprehensive risk assessment of CCS can be delivered. Key words: carbon dioxide capture and storage, leakage, marine, environment, ecosystem, CO2, pH.

13.1

Introduction

In this chapter we review existing knowledge on both the dynamics and dispersion of carbon dioxide (CO2) emanating from a failure of CCS and on the potential environmental impacts that may result. From the marine perspective, two forms of CCS are pertinent, storage in geological reservoirs under shelf seas and deep sea sequestration. Deep sea CO2 sequestration is a proposal by which CO2 is injected into deep water where it forms semi-stable clathrates, although there is much debate as to the stability and longevity of this form of storage. Broadly speaking, from a marine environmental perspective, leaks from geological storage systems may take two forms, either a breakdown of infrastructure such as a pipeline, that would release CO2 directly into the bottom of the water column, or a geological failure 344 © Woodhead Publishing Limited, 2010

Carbon dioxide (CO2) leakage in marine ecosystems

345

resulting in a percolation of CO2 through various geological strata, the biochemically active sediment layers (benthos) and then into the (pelagic) water column. Many scientific disciplines need to be integrated to form a full understanding of leak impacts, amounting to a significant research challenge. These are: ∑

fluid and gas dynamic processes that determine the micro to mesoscale dispersion of CO2 both in water and within sediments; ∑ carbonate chemistry describing the behaviour and chemical impacts of dissolved CO2 in seawater; ∑ marine hydrodynamics addressing dispersion and mixing on meso to macroscales; ∑ physiological and biochemical responses of marine processes, flora and fauna to changes in carbonate chemistry; ∑ ecological assessments of the impacts on communities, ecological functionality and resources. Whilst some research specific to CO2 discharge from CCS exists, focused either on the dynamics of direct sequestration to deep water or geological sequestration under shelf seas, research into ocean acidification also provides relevant insights especially into impacts. Ocean acidification is the process by which anthropogenic CO2 released to the atmosphere is slowly absorbed into the world’s oceans causing a shift in carbonate chemistry and a lowering of pH. This chemical response is identical to that which would occur from a CCS derived CO2 leak, differing only in scale and magnitude. Prevention of ocean acidification and climate change is the objective of mitigation methods such as CCS. In summary, existing research can provide a firstorder answer to questions about dispersion and impact, but further research is required to reduce the significant uncertainties and achieve cross-disciplinary integration. The marine system faces many challenges over the coming decades. It is vulnerable to climate change, increased fishing pressures and pollution whilst at the same time it provides a vital component in the regulatory earth system and is of huge economic and societal importance. It is vital, therefore, that any risk assessment of CCS includes a risk assessment of the potential impacts on the marine ecosystem. In this chapter, Section 13.2 details the physical and chemical behaviour of CO2 in seawater and Section 13.3 details the potential range of ecological impacts. Section 13.4 discusses the options for monitoring the marine system for leaks, with the mitigation of CCS leaks briefly discussed in Section 13.5. Future trends are summarised in Section 13.6.


346

13.2


The physical and chemical behaviour of carbon dioxide (CO2) in the marine system

13.2.1 Carbonate (CO2) chemistry in the marine system The behaviour of dissolved CO2 in seawater (the carbonate system) is well constrained, equilibrium constants are extensively published and, although there are variations in particular constants emerging from different studies, a consistent, robust approach is generally possible (e.g. Zeebe and WoolfGladrow, 2001 amongst others). Four parameters of the carbonate system can be analytically determined: dissolved inorganic carbon (DIC), total alkalinity (TA), pH and the partial pressure of CO2 in the water (pCO2w). Knowledge of any two of these variables is sufficient, given appropriate meta data, to derive the other two and a number of biogeochemically important parameters such as bicarbonate and carbonate ion concentrations and carbonate saturation states (Zeebe and Woolf-Gladrow, 2001). CO2 added to seawater exists as dissolved gaseous CO2, a proportion of which reacts to form carbonic acid (H2CO3) which then dissociates to form bicarbonate (HCO3–), releasing hydrogen ions to solution (equations 13.1 and 13.2). These hydrogen ions then combine with carbonate ions (CO32–) to form more bicarbonate (equation 13.3). CO2 + H2O ¤ H2CO3

[13.1]

+

[13.2]

H2CO3 ¤

HCO3–

+H

CO32– + H+ ¤ HCO3–

[13.3]

Hence, adding CO2 to seawater results in an increase in bicarbonate ions, a decrease in carbonate ions and an increase in hydrogen ions or acidity (decreasing pH) (fig. 13.1). Such changes are problematic for marine biogeochemical cycles and ecosystems as many biochemical and physiological processes are affected by pH, and bicarbonate and carbonate are substrates for some of the most fundamental marine processes such as photosynthesis and calcification. Calcification, the generation of hard shells by many types of marine flora and fauna (corals, coccolithophores, pteropods, molluscs, echinoderms, etc.), is a process that produces CO2 and is therefore inhibited by excess CO2 in the system (equation 13.4). Additionally the decrease in carbonate ions encourages dissolution of calcium carbonate (equation 13.5).

2HCO3– + Ca2+ ¤ CaCO3 + CO2 + H2O

CaCO3 ¤ Ca2+ + CO32–

[13.4] [13.5]

Inhibition of shell formation obviously has consequences for the organism involved. The carbonate saturation state omega (Ω) represents the balance


Carbon dioxide (CO2) leakage in marine ecosystems Bicarbonate

2000

10 9 8

pH 1500

7 6 5

W

1000 W

500

Ar

Ca ag

Carbon 0 2000

4

lci

te

on

3

ite

pH/saturation state

Bicarbonate/carbonate (mmol/kg)

2500

347

2 1

ate 2200

2400 2600 2800 Dissolved inorganic carbon (mmol/kg)

0 3000

13.1 Changes in key carbonate system parameters for rising concentrations of DIC. The present day status is approximated by the left-hand limit of the figure. The approximate maximum predicted perturbation likely from ocean acidification is marked by the dashed vertical line. CCS leakage could provoke perturbations beyond the righthand limits of the figure but only for restricted volumes and durations. Left axis, bicarbonate and carbonate ion concentration (mmol kg–1); right axis, pH and carbonate saturation state. (The saturation states are the product of the concentrations of the reacting ions divided by the product of those ions at equilibrium, hence saturation states below 1 indicate that mineral carbonate will dissolve into the surrounding seawater.) Calculations made for surface waters at 10 °C, a salinity of 36 psu, with an alkalinity of 2324 mmol kg–1.

between dissolution and mineralisation, with values below 1 indicating net dissolution is thermodynamically favoured (Fig. 13.1).

13.2.2 Predicted changes in carbonate chemistry likely from ocean acidification and carbon capture and storage (CCS) leakage The marine system is highly buffered with respect to changes in CO2, but only via slow millennial scale processes such as erosion. Consequently, the marine system has evolved with a very stable carbonate system for potentially over 20 million years (Pearson and Palmer, 2000), with global mean pH in the range 8.0–8.2. However, the comparatively short-term perturbations such as ocean acidification (the uptake of anthropogenic atmospheric CO2 by the oceans) or leaks from CCS systems alter pH, ion concentrations and omega far beyond the ranges experienced over the evolutionary timescales of contemporary species.


348


Broad-scale predictions of the oceanic uptake of atmospheric CO2 and the consequences for the marine carbonate system are tractable and reasonably robust. It is predicted that global mean ocean pH will fall significantly to pH 7.5 or less within the timeframe of 100–300 years (Caldeira and Wickett, 2003, 2005) dependent on emission scenarios and mitigation. This fall represents an extreme and rapid perturbation from estimated marine pH over at least 20 million years (Fig. 13.2). Many marine organisms depend on synthesising calcium carbonate structures, and their ability to perform this synthesis depends at least partly on the carbonate saturation state (Ω). Because of the dependence of saturation state on temperature and pressure, under-saturation (Ω < 1.0) is a property of deeper waters and the depth of the saturation horizon (where Ω = 1.0 and waters, above remain over-saturated and hence conducive to calcifiers) is an important diagnostic of the marine environment. The temperature dependency creates a latitudinal variation in the saturation horizon depth such that this depth is significantly shallower in polar waters. Global predictions of saturation state predict that polar surface waters will become under-saturated within decades due to ocean acidification (Orr et al., 2005; Steinacher et al., 2008). The impacts of ocean acidification will be experienced globally, although some regions will be more sensitive. Further, the perturbation will persist for thousands of years before the natural processes of weathering, ocean circulation and carbonate buffering return pH and omega to normal values (Archer, 2005). In contrast, the perturbation in carbonate chemistry caused by a leak from CCS has the potential to exceed the perturbation expected from ocean acidification but only in the vicinity of the leak and within any plume of CO2 rich water that is created. The natural mixing of seawater caused by tides, etc. would act to disperse and dilute the CO2-rich plume, restricting the spatial 8.6 8.4

pH

8.2 8

1800 2000 2050 2100

Oceanic pH

7.8 7.6 7.4 –25

–20

–15 –10 –5 Time (million years before present)

0

13.2 Past (white diamonds, data from Pearson and Palmer, 2000) and contemporary variability of marine pH (black diamonds with dates). Future predictions are model-derived values based on IPCC mean scenarios (reprinted from Blackford and Gilbert, 2007).


5


349

scale of impact. If the leak was temporary then the water chemistry would return to normal within a relatively short timeframe (days to months, rather than the millennia required to mitigate ocean acidification). Thus a CCS leak could provoke a pH decrease to ~6.0, but its spread and persistence are likely to be restricted. These issues are explored in the following sections.

13.2.3 Fine-scale dynamics of (CO2) droplets/plumes in marine systems The initial behaviour of CO2 injected into seawater is complex, as an understanding of fluid dynamics and bubble/droplet behaviour in the context of ocean turbulence on fine scales is required. The fine-scale ocean can be defined as the volume where leaked CO2 as a separate phase interacts with seawater, and it is important to take the turbulence generated by leaked CO2 on ocean flow fields into account. The fine-scale dynamics of CO2 leaked from the seabed into the water column depend on the physical-chemical properties of the CO2/seawater system, the leakage depth and the form of leakage. CO2 could potentially leak in two general ways, dilute leakage with a limited mass flux from natural dispersion through sediments and dense leakage with a high mass flux from fractures of geo-formation or failures of injection well and pipelines. Observations of small-field experiments (Brewer et al., 2006) and natural leakage (HallSpencer et al., 2008) demonstrate that leaked CO2 enters the seawater column in the form of either bubbles or droplets. The bubbles/droplets are in a variety of sizes, ranging from millimetres to several centimetres, generated from the porous sediments or broken up from a CO2 column leaked from fractures. CO2 is more compressible than seawater at depths above 3000 m and soluble in seawater. CO2 bubbles or droplets released above this depth could ascend, driven by buoyancy, as well as dissolve into seawater, causing the changes to the seawater carbonate system described above. The interaction of ascending CO2 bubbles/droplets with ocean currents creates a bubble/droplet plume and a CO2-enriched seawater plume in a turbulent ocean. A sample scenario of CO2 droplet and CO2-enriched plumes generated by a numerical model (Alendal and Drange, 2001) is shown in Fig. 13.3, for which the mass leakage (CO2 injection) rate from a depth of 700 m is 1.0 kg/s and the initial droplet size is 12 and 1.0 mm in radius. For the fully developed plume, the spatial scale of dilution may vary from 10 2 to104 m and the associated temporal scale from hours to days. The physical properties of carbon dioxide (CO2)/seawater system Two physical properties, the phase diagram and solubility, are important for estimating the behaviour of CO2 in fine-scale seawater. According to the © Woodhead Publishing Limited, 2010

350

Developments and innovation in CCS technology Droplet concentration (kgCO2 m–3)

pH 6

6.5

7.0 7.5 7.75 7.9

0.01 0.02 0.08 0.2 0.5 1.4

Radius = 1 mm

1 mm

Radius = 12 mm

12 mm

660

Depth (m)

700

740

660

700

740

0

40

80 120 0 40 Downstream distance (m)

80

120

13.3 CO2 droplet plume (right by droplet concentration) and CO2-enriched seawater plume (left by pH) from numerical simulations: elapsed time is 30 min; injection rate is 1.0 kg CO2 s–1; ocean current speed is 5 cm s–1 (Alendal and Drange, 2001).

phase diagram, as shown in Fig. 13.4 (Nikolaus et al., 2008), leaked CO2 can exist as a gas at depths roughly above 400m and as a liquid below that depth. A solid-like state, clathrate hydrate of CO2, may form at the surface of CO2 bubbles/droplets if the temperature of seawater is lower than 9 °C. This hydrate layer affects the dynamics of CO2 and seawater, owing to the changes in features of the boundary layer and solubility. In general, CO2 is more soluble in the deep ocean than in the shallower ocean because the solubility increases as pressure increases and temperature decreases. Another property which plays an important role in the development of CO2-enriched seawater plumes is the increase of density of CO2 solutions which provides a negative buoyancy causing high pCO2 seawater to fall on to the seabed and affect benthic organisms. Carbon dioxide (CO2) bubble/droplet dynamics To date, the research on transportation phenomena of an individual CO2 bubble/droplet in seawater has been mostly focused on those at mid-ocean depths (Caldeira et al., 2005). Some ascending velocity data on droplets with and without hydrates have been obtained from both laboratory (Nikolaus et al., 2008) and small-scale in situ experiments (Brewer et al., 2002), while


Carbon dioxide (CO2) leakage in marine ecosystems Density (kg/m3) 912.4 967.9

856.6

5

CO2 (g ) + SW (s CO2 -H ydrate at) + SW (s

at)

351

1024.5

CO2 (g) + SW (sat) CO2 (l) + SW

TSW

10

at) ) + SW (s CO 2 (l + SW (sat) ydrate CO 2-H

Pressure (MPa)

(sat)

15

20

rCO2

rSW


20

25

5

13.4 CO2 phase diagram in the ocean (Nikolaus et al., 2008).

a few data are available for CO2 bubbles (Johnson et al., 1969). Developed models calibrated against the data (Chen et al., 2009) have been applied to simulate the dynamics of leaked CO2 in seawater at various depths from 10 to 800 m (at temperatures from 5 to 25 °C) and for initial droplet/bubble sizes from 3.0 to 40.0 mm in diameter. A diagram of CO2 terminal distance vs dissolution time obtained from model simulations shows that CO2 droplets ascend at a mean speed of 11 cm/s and a mean shrinking rate of 7.0 ¥ 10–3 mm/s in diameter approximately, if leaked from a deep-ocean (800~1000m) source. This speed and shrinking rate increase to 16 cm/s and 30 ¥ 10–3 mm/s at middle-deep ocean (500~650 m) and finally reach 22 cm/s and 0.2 mm/s at shallow ocean (< 150 m), as shown in Fig. 13.5. Fine-scale dynamics of carbon dioxide (CO2) and CO2-enriched seawater plumes Recent observations of natural volcanic CO2 vents (Hall-Spencer et al., 2008) found that continuous leakage of CO2 at a rate of 1.43 ¥ 106 litre a day in an area of about 3000 m2 lowers the pH of the water column at gradients of normal pH (8.1~8.2) to lowered pH (mean 7.8~7.9, minimum 7.4~7.5). It has been noted, however, that a purposeful field experiment on CO2 plumes in the shallow ocean has not been carried out so far. An in situ experiment (Brewer et al., 2006) at mid-ocean depths successfully


352


Terminal distance (m)

50

40

30 Do = Best

20

10

40 mm 30 mm 20 mm 5 mm 3 mm fitted

0 0

1

2 Terminal time (min)

3

4

13.5 CO2 bubble dissolution in shallow ocean, terminal distance vs time. The gradient of the curve is the mean velocity and the shrinking rate of bubble can be estimated by bubble size and associated terminal time (Chen et al., 2009).

monitored the dispersion of CO2 droplet plumes for more than 25 minutes from 1100 m to 750 m. A two-phase turbulent plume model calibrated by the data from the observation predicts that the pH change in seawater owing to CO2 dissolution is directly proportional to the leakage mass flux (m· , kg/s/ m2) and inversely proportional to initial droplet/bubble size (Do, mm) and ocean current speed (Vc, cm/s), as shown in Fig. 13.6. In the case of the mid-ocean, if m· = 0.1, Do = 8.0, and Vc = 2.5, the maximum pH change is about –1.2 with a water volume less than 60 m3 (case A). This pH change increases to –1.7 pH units when leakage mass flux increases to one (case B) and reduces to –1.0 pH units with water volume of 300 m3 when ocean current increases to 25 (case C), respectively. The smaller the droplet size the larger the pH changes that could be produced (Case D and also the top panel of Fig. 13.3). The mean CO2-enriched seawater volume developed from CO2 leakage positioned in a turbulent ocean is also simulated. The results show that a fully-developed plume volume with pH changes greater than –0.1 could reach up to 1.0 ¥ 106 m3 within 30 minutes of onset of the source.



353

1.0E+06 Case-A Case-B Volume of seawater (m3)

Case-C Case-D 1.0E+04

1.0E+02

1.0E+00 –1.8

–1.5

–1.2

Case

· m

D o

Vc

A: B: C: D:

0.1 1.0 1.0 0.3

8.0 8.0 8.0 5.0

2.5 2.5 25.0 2.5

–0.9 –0.6 pH changes

–0.3

0.0

13.6 Full-developed CO2-enriched plume properties indicated by plume volume vs pH changes (Chen et al., 2007).

13.2.4 Regional scale modelling of carbon capture and storage (CCS) leakage, approaching ecosystem scales. Shelf seas are typically subject to disruption from a variety of events such as fishing, dredging, eutrophication and pollution. A leak from a CCS installation would inevitably cause some disruption to the local ecosystem; however, the key question for risk assessment is to determine the spatial and temporal scales of disruption and assess whether the impact would be significant in ecological, regional or economic scales. This contextual aspect has been partially addressed using coupled hydrodynamic–ecosystem models to quantify the possible spatial range and temporal persistence of perturbations arising from a failure of geological storage in shelf seas (Blackford et al., 2008, 2009). In particular, this work has explored what size of leak event would be needed to produce an environmentally damaging response over a significant area. Figure 13.7 illustrates a short-term leak scenario of ~1.5 ¥ 105 t CO2 over one day, at four different times of the year at two sites in the north Sea. The north site is a deep water column (130 m) in the Forties field, the south site a shallow region (30 m) in the Viking field. The leak rate approximates to 50x the input rate at the Sleipner field, i.e. rather large. This leak rate provokes pH perturbations that exceed 0.5 pH units for about a day at the north site and up to five days at the south site. The duration of disturbance is


354

0 –15 –31 Depth (m)

0

–61 –77 –92 –122 10

20

30

100

110

120

191

201

211

283

293

303

–0.4

(a) 0 –3

–0.6

–6 Depth (m)


–0.2

–107

–9 –13

–0.8

–16 –19 –1

–22 –25 (b)

10

20

30

40

100

110

120

130 191 Julian days

201

211

221

283

293

303

313

13.7 Temporary point–source leak scenario (x50 pipeline capacity): (a) South site, (b) North site. All perturbations below 0.01 pH unit have been masked for clarity. Perturbations below 0.2 pH units indicate little likely impact, those between 0.2 & 0.5 pH units indicate potential damage and those over 0.5 pH units indicate that significant damage is likely. These results illustrate the mean perturbation over an area of 50 km2 (adapted from Blackford et al., 2009).


–46


355

less than 10 days for the north site and as much as 20 days for the south site. The perturbation is the mean effect over an area of ~50 km2. This scenario indicates the approximate scale of leakage required to provoke what may be serious environmental consequences on a regional scale, although there are few observations that quantify the consequences of a short sharp pH perturbation, such as simulated here. The difference between sites is simply driven by the relative depth of the water columns and the assumption of the initial distribution of dissolved CO2. A large-scale long-term leak scenario – perhaps representing some catastrophic failure of sequestration – is illustrated for the same two sites in Figs 13.8 and 13.9. Here, the leak rate is ~ 5x the input rate at Sleipner and is continuous over one year, the total injection of CO2 amounting to ~5.5 ¥ 106 t over one year. The results show, for both sites, an area of high perturbation centred over the release (approximately 50 km2). In the north, this perturbation does not exceed 0.5 pH units; in the south, a perturbation sometimes exceeding 1.0 pH units is recorded. The area of maximum disturbance in both cases remains well constrained, although a plume of acidified water is seen to spread from the release point driven by the regional circulation. This plume can be extensive, although the majority of the plume area is acidified by significantly less than 0.1 pH units. Examination of the chemical signal at the leak epicentre (Fig. 13.9) for both release sites clearly shows the influence of the tidal cycle in determining the instantaneous perturbation strength. Perturbation maxima are associated with neap (weak) tides and minima with springs and can differ by 0.4 or 0.8 pH units depending on site and wind strength. This suggests not only that the timing of leaks may be significant but also that a consideration of the siting of CCS systems in relation to tidal mixing patterns would be prudent. A number of clear conclusions can be drawn from this work. First, it would require a very large or persistent leak to produce a perturbation in carbonate chemistry of sufficient magnitude to have ecological implications on a regional scale. Second, the key dispersion vector in the short to medium term is the strong tidally driven mixing in the region, rather than out-gassing of dissolved CO2 to the atmosphere. Third, the spatial and temporal variability in tidal mixing can have a significant influence on the rate of dispersion and, consequently, the chemical signature at the leak epicentre. Parameterising the likely magnitude and duration of a leak event is clearly difficult, as there is little evidence to base estimates on. Scenarios similar to the above using smaller CO2 injection rates showed proportionally smaller perturbations at a regional scale. However, this model system has a horizontal resolution of ~7 km and does not resolve the fine-scale perturbation as described in the previous section.


01-Jan

56

07-Aug

55

58

54

57

53

56

52

55

51

60

15-Mar

59

55

58

54

57

53

56

52

55

51

60

26-May

55

58

54

57

53

56

52 –2

0

2

4

–4 (a)

–2

0

2

0

15-Mar

4

51

–0.1

19-Oct

–0.2

–0.3

–0.4

56

31-Dec

59

55 –4

0.1

56

19-Oct

07-Aug

26-May

31-Dec –0.5

0

2

4

6

8

(b)

0

2

4

6

8

13.8 Continuous point-source leak evolution: (a) North site, (b) South site, instantaneous snapshot times as stated. All perturbations below 0.01 pH unit have been masked for clarity. Damage potential is as described for Fig. 13.7 (adapted from Blackford et al., 2009).



59

01-Jan

356

60

–15 0.1

–31

0

–61 –77

–0.1

–92

–0.2

–122

–0.3

–138 J

F

M

A

M

J

J (a)

A

S

O

N

D

–0.4 –0.5

–3

–0.6

–6 –9 Depth (m)


–107

–0.7

–13

–0.8

–16 –19

–0.9

–22 –1

–25 –28 J

F

M

A

M

J

J (b)

A

S

O

N

D

357

13.9 Continuous point-source leak as in Fig. 13.4 showing the evolution of pH perturbation at the leak location, indicating the mean state for an area of ~50 km2 : (a) North site, (b) South site. Damage potential is as described for Fig. 13.7 (adapted from Blackford et al., 2009).


Depth (m)

–46

358

13.3


Marine ecosystem impacts of carbon dioxide (CO2) leakage

CO2 sequestration in sub-seabed geological formations at depth, and thereby pressure, will result in the formation of hydrates. By contrast, shallow sea sequestration, such as proposed for the North Sea, will not; the gas would remain buried but in a liquid, or semi liquid state that would be buoyant if released. As such, CCS failure leakage scenarios are site specific and each is potentially different from another. Modelling studies have shown that in the case of leakage from sub-seabed CO2 storage sites, any plume of acidified seawater will dissipate rapidly through continuous mixing by tides and currents (Blackford et al., 2008). So, any exposure likely to be experienced by pelagic organisms may be brief. However, a leak that percolates up through the sediment or occurs near the sea floor could significantly alter the chemistry of the seawater both within and just above the seafloor. Consequently, the impacts of leakage could be severe for benthic organisms that have limited capacity to escape from an affected area. There is now a considerable body of evidence that describes the vulnerability of species and processes to exposure to excessive CO2, which are described below. Whilst much of this experimental work has been performed in the context of ocean acidification impact research, the pH manipulations used are often a close mimic of the more extreme pH/CO2 ranges that would occur with a CCS derived event as this enables scientists to better assess the functional response to decreasing pH. The majority of the impact studies to date have concentrated on pH reductions due to atmospheric deposition of CO2 and with no regard for the interactions between any contaminants that may be present in the water body. The failure of a CCS installation would, in some instances, result in the remobilization of CO2 into the surrounding seas. However, the impact of the released CO2 on the biota will vary depending on factors such as its industrial source, the process of liquefaction and residual gases such as H2S, SOx and NOx. In addition, any residual chemicals associated with the original drilling activities as well as those occurring naturally, such as metals, minerals and previously unrecovered oil, could be released. Liquefied CO2 is highly corrosive requiring that the pipework for the delivery system would need to have a protective coating to prevent corrosion. Many of the products used to prevent corrosion are themselves toxic thereby adding yet another confounding factor to any assessment of likely CCS leakage impact. There is a rapidly growing body of published data indicating that exposure to elevated levels of dissolved CO2 and the associated changes in carbonate chemistry (e.g. pH and carbonate saturation state) can have a considerable deleterious effect on many marine organisms and processes. In the following sections we identify the main impacts on an organism’s basic



359

biological functions, examine the consequences for animal health, survival and reproduction, and explore the implications for the maintenance of marine communities and processes.

13.3.1 Organism physiology A number of recent efforts have been made to review the available evidence from experimental observations and describe the likely impacts of exposure to CO2 acidified seawater (including elevated dissolved CO2, H+ and HCO3– concentrations, plus reduced CO32– concentration) on the physiology, performance and survival of marine organisms (Seibel and Walsh, 2001, 2003; Pörtner et al., 2005; Pörtner, 2008; Pörtner and Farrell, 2008; Widdicombe and Spicer, 2008). These studies conclude that exposure to acidified seawater in the pH range 6.0–7.8 has the potential to disrupt a number of intracellular and extracellular physiological processes, across a range of taxonomic groups: echinoderms (Kurihara and Shiriyama, 2004; Kurihara et al., 2004, 2007; Miles et al., 2007; Wood et al., 2008), molluscs (Michaelidis et al., 2005; Berge et al., 2007; Bibby et al., 2007, 2008; Gazeau et al., 2007; Beesley et al., 2008), crustaceans (Metzger et al., 2007; Pane and Barry, 2007; Spicer et al., 2007) and sipunculids (Langenbuch and Pörtner, 2002, 2004; Langenbuch et al., 2006). The primary effect of exposure to acidified seawater is to decrease the pH of the body fluids (i.e. blood or haemolymph); known as acidosis. For many marine animals, the response to this extracellular acidosis is to increase the amount of bicarbonate ions within the body fluids to create near full or partial pH compensation. This is achieved predominantly through active ion transport processes in the gills (see Widdicombe and Spicer, 2008; and refs therein). However, recent studies have shown that for some invertebrates there is only partial, or no, compensation in hypercapnia (too much CO2 in the blood) induced disturbance of extracellular acid-base balance, e.g. the mussel Mytilus edulis (Zange et al., 1990), the crabs Callinectes sapidus (Wood and Cameron, 1985) and Chionoecetes tanneri (Pane and Barry, 2007) and the sea urchin Psammechinus miliaris (Miles et al., 2007). To ensure the effective maintenance of intracellular pH, extracellular pH needs to be maintained 0.5–0.8 pH units above intracellular pH (Pörtner et al., 2004). Consequently, any lack of regulatory capacity is important because countless cellular functions and regulations depend upon maintaining a specific intracellular pH (Putnam and Roos, 1997). Additionally, the maintenance of extracellular pH is also important for the function of respiratory proteins, with both pCO2 (i.e. a specific CO2 effect; Mangum and Burnett, 1986) and acidity having pronounced effects on oxygen binding by respiratory pigments; hemoglobins (Weber, 1980), hemocyanins (Mangum 1997), but particularly the annelid pigments erythrocruorin and chlorocruorin (Weber, 1980).


360


13.3.2 Organism health and survival There is a growing body of research evidence to indicate that a reduced pH environment has a negative effect on animal growth, performance and health. Impacts on general health status reported have included immunosuppression in Mytilus edulis (Bibby et al., 2008) and narrowing of thermal tolerance in the edible crab Cancer pagurus with increasing CO2 concentrations (Metzger et al., 2007). Wood et al. (2008) observed muscle wastage in the brittle star Amphiura filiformis which was associated with stimulation of calcium deposition for new arm growth. The capacity to upregulate calcium production in reduced pH environments receives support from studies by Gil-Martens et al. (2006) who observed bone remodelling and increased bone volume in salmon smolts maintained in a reduced pH environment; however, and again in support of the studies by Wood et al. above, there was evidence of reduced soft tissue growth. Beesley et al. (2008) observed that within seven days of exposure to reduced pH the lysosomes in mussel blood cells were exhibiting reduced membrane stability which is indicative of a decline in health status. Cardiac output was shown to decrease in yellowtail finfish (Seriola quinqueradiata) at reduced pH (Ishimatsu et al., 2005). Sea urchins (Echinocardium incordata) maintained at pH 6.8 for 30 days experienced disruption of the basement membrane of the gut epithelium and changes in the morphology of the mucous-rich apical cytoplasm (Fig. 13.10, PML unpublished data). There is also an extensive literature on the effects of some of the types of chemical contaminants associated with CCS installation leakage listed above to indicate that they also have an adverse effect on the same cellular processes (Sprague and Logan, 1979; Morse et al., 1982; Cranford and Gordon, 1991,1992; Olsgard and Gray, 1995; Fernley et al., 2000; Barlow and Kingston. 2001; Taban et al., 2004; Armsworthy, 2005). By contrast, there is a distinct lack of information on the consequences of any interactions between these two factors on animal health which may be additive or synergistic or both.

13.3.3 Impact on growth and reproduction A general feature of many studies investigating the effects of a reduced pH environment for a range of organisms is that the greatest impacts are seen in eggs followed by larvae and early life stages with adults being the least affected (Haugan, 2004). In addition, Kikkawa et al. (2008) hypothesized that active species are more sensitive to elevated levels of CO2 than inactive ones. By contrast, Seibel and Walsh (2003) hypothesized that deep sea animals with low metabolic rates are more sensitive to high CO2 conditions. Reduced egg production has also been reported in copepods and decreased hatching success in gastropods (Raven et al., 2005). Also in copepods, Kurihara



361

(a)

(b)

13.10 (a) Section of intestinal wall in sea urchin from reference treatment group at pH 8.0. (b) Section of intestinal wall in sea urchin maintained at pH 6.8 showing disruption of basement membrane on coelomic surface and enlarged and disrupted apical surface on luminal surface.

and colleagues (2004) noted decreased hatching rate and increased nauplius mortality at lowered pH (range 6.40–7.48). Other reported impacts have included a reduction in sperm motility broadcast in spawning invertebrates


362


such as sea urchins (Havenhand et al., 2008) and Pacific oysters resulting in reduced fertilization success. Developmental impairment has been observed in sea urchins (Kurihara and Shirayama, 2004) and oysters (Kurihara et al., 2007); weakening of the shell structure (Green et al., 2004; Shirayama and Thornton, 2005; Haugan et al., 2006), dissolution of the protective armour of coccolithophors (Reibesell et al., 2000), forminiforera and pteropods. Slower growth in corals has been reported which could impact on reproductive success in those species where reproductive maturity is related to size and not age (Sakai, 1998a,b). Growth was also shown to be reduced in mussels following reduced pH exposure (Michaelidis et al., 2005; Berge et al., 2006), sea urchins and gastropods (Shirayama and Thornton, 2005). Abnormal larval development including altered skeletal proportions and symmetry was reported in brittle stars by Dupont et al. (2008) following exposure in reduced pH (7.7 and 7.9) after between five and six days with an associated marked increase in mortality. Increased larval mortality was also observed of bivalves during settling (Green et al., 2004; Raven, et al., 2005) and in sea bream (Kikkawa et al., 2004) following exposure of the eggs to seawater at reduced pH (pH 5.9 and pH 6.2). Whilst some capacity to regulate slight changes in pH (0.5 pH units over 8 days) was demonstrated in the sea urchin Psammechinus miliaris, more severe conditions resulted in high mortality rates (Miles et al., 2007). Egg maturation was shown to be severely disrupted in brittlestars (Ophiura ophiura) maintained at pH 6.5 as compared to reference treatment animals maintained at pH 8.0 (Fig. 13.11, PML unpublished data)

13.3.4 Community structure and diversity The world’s oceans contain enormous biological diversity (May, 1994; Reaka-Kudla, 1997), and most of this biodiversity, 98 % of all marine species, is made up of invertebrates either residing in (infauna) or on (epifauna) sediments (Snelgrove, 1999). This incredible diversity results from complex interactions between the underlying physical and environmental conditions, such as depth, temperature, organic supply and granulometry, and the biological interactions operating between and within benthic organisms, such as predation and competition (May, 1994). To date, there have been very few published studies describing the impact of CCS leakage on these important processes and the consequences for the diversity and structure of intact marine communities. Widdicombe et al. (2009) conducted a mesocosm experiment to quantify the direct effects of short (two weeks) and long (20 weeks) term exposure to acidified seawater on the structure and diversity of macrofaunal and nematode assemblages in two different sediment types; one muddy and one sandy. (A mesocosm is a medium sized enclosed experimental facility that aims to mimic the natural environment by maintaining a community of animals under close to natural conditions.) The experiment showed that



363

(a)

(b)

13.11 (a) Section of reproductive tissue in brittlestars from reference treatment group at pH 8.0 showing late maturing eggs. (b) Section of reproductive tissue from brittlestars maintained at pH 6.5 showing severe disruption of late maturing eggs.


364


communities were made up of species with different tolerances to high CO2. Consequently, exposure to acidified seawater significantly altered community structure and reduced diversity for both macrofaunal and nematode assemblages. However, the impact on nematodes was seen to be less severe than the impact on macrofauna. Two other studies have addressed the impacts of high CO2 on biological interactions. The first, by Bibby et al. (2007), showed that the intertidal gastropod Littorina littorea produce thicker shells in the presence of predation (crab) cues but this response was disrupted at low seawater pH. This result demonstrated that CCS leakage could have indirect biological effects by disrupting the capability of organisms to express induced defences, hence, increasing their vulnerability to predation. In the second, Dashfield et al. (2008) used a mesocosm experiment to determine whether the presence of an important bioturbating species, the burrowing urchin Echinocardium cordatum, might influence the response of a nematode community to seawater acidification. They found that the urchin was vulnerable to elevated levels of CO2, and its removal resulted in a significant change in the nematode community. A further concern is that any change in community structure or diversity could lead to a reduction in a number of key ecosystem functions, in particular the cycling of nitrogen, especially as shelf seas are known to host a disproportionately large fraction of productivity (Field et al., 1998).

13.3.5 Nitrogen cycling Of the very few studies that have been conducted to specifically examine the impact of CCS leakage on nitrogen cycling, all have reported significant effects. Huesemann et al. (2002) demonstrated that rates of ammonium oxidation to nitrite or nitrate (nitrification) were reduced by approximately 50 % at pH 7, by more than 90 % at pH 6.5 and were completely inhibited at pH 6. Whilst not directly measuring nitrification rates, the results of Widdicombe and Needham (2007) and Widdicombe et al. (2009) support the assumption that reduced seawater pH will inhibit the microbial oxidation of ammonium. In sediments the majority of nitrate used to fuel denitrification comes from nitrification rather than from the overlying water, particularly within the burrows of benthic species. Consequently, seawater acidification will inhibit the production of nitrate within the sediment and result in a greater sediment uptake of nitrate from the overlying water (Widdicombe and Needham, 2007; Widdicombe et al., 2009). In addition, many benthic species have been shown to modify the exchange of nitrogen between the sediment and the overlying water (e.g. Widdicombe and Austen, 1998; Olsgard et al., 2008). Deleterious effects on these organisms will have significant effects on nutrient flux rates. From the limited evidence available to date, it could be hypothesized that CCS leakage would significantly alter rates



365

of nutrient cycling, either directly through impacts on bacteria or indirectly through changes in the structure and function of infaunal bioturbators (the animals whose burrowing and feeding behaviour has the effect of mixing the sediments).

13.4

Leak monitoring options

Monitoring the marine environment for CO2 leakage could follow one or a combination of three methods, surveying for physical disruption to the sediment surface caused by geological seepage, detection of the chemical signal in the water column or detection of biological impacts. The stochastic nature of a leak event, the difficulty of predicting when, where and how much CO2 and the shape of the dispersion plume, poses severe problems for an economical and efficient monitoring system. Based on existing research, it is impossible to gauge an appropriate spatial temporal distribution of monitoring that would guarantee a high certainty of detecting an event. Two developments can assist; firstly, a refinement of the range of possible leak scenarios, looking towards both geological and engineering research; and secondly, a refinement of the existing marine models, described here, that would better characterize the behaviour and dispersal of CO2 and the likely chemical signature that would result. Given this information, it would be possible to estimate the optimal density of observations required. Leakage from systems hardware, principally delivery pipelines, would most efficiently be detected via pressure sensors within the pipelines, essentially existent technology. Monitoring within the marine environment requires some knowledge of the base-state. Whilst regions like the North Sea are fairly well characterized, they are extremely heterogenous and dynamic with respect to sediments, implying that physical detection of sediment surface disruption would be problematic. Conversely, the technology to examine sediment topology, sidescan sonar, exists, is affordable, and can cover large areas efficiently. The pelagic marine carbonate system is influenced by both physical and biological processes that impart variability on daily and seasonal scales, again with some spatial heterogeneity (Blackford and Gilbert, 2007). More recently, processes specific to benthic systems have been identified which need taking into account (Thomas et al., 2008). These natural fluctuations are reasonably well characterized. Detection of leak driven perturbation sufficient to perturb the system beyond its natural variability could in theory be achieved by monitoring the carbonate system. Detection of leaks that are not large enough to perturb the system beyond its natural range is more problematic, as a range of physical and biological measurements would be required to determine the cause of the fluctuation. A lot of research has been devoted to developing carbonate system measurements in recent years (e.g.


366


Hardman-Mountford et al., 2008). Measurement of pCO2 is operationally routine and affordable; however, pCO2 has a very high variability and may not be the most sensitive indicator of change. pH is tightly buffered, with a natural variability of around 0.3–0.5 units. However, the accuracy required in pH measurement has prevented as yet a fully operational system being developed, although this is a likely development for the near future. To date, no operational deployable system for measuring DIC exists. Monitoring for biological impact is problematic as we are as yet unable to characterize with any certainty the biological response to CO2, given the high degree of species-specific variability in response and the high heterogeneity of shelf seas. One clear challenge is to determine the optimal density of monitoring systems; too sparse a grid would imply that leak events could be missed, a high-resolution grid would have major cost implications. One solution maybe to employ autonomous underwater vehicles or remotely operated underwater vehicles with operational sonar and carbonate system monitors, coupled with fixed monitoring arrays around specific definable leak risk sites. A crucial component of operational systems is the automation of processing and data transfer; however, such systems are becoming routine in marine applications. (Hardman-Mountford et al., 2008). An added bonus of routine monitoring is that the observations will have significant value for understanding the basic behaviour and variability of the marine system and evaluating models.

13.5

Mitigation of leaks

Significant leakage from pipeline infrastructure can be simply prevented by judicious use of sensors, valves and automated shutdown systems, which is already standard practice. The mitigation of leaks from reservoirs through the caprock presents a considerable problem which could probably only be limited by depressurizing the reservoir. Once a significantly large CO 2 stream reaches the ecosystem, it is difficult to envisage an effective and achievable strategy that would limit impacts. Hence, rather than post-event mitigation, the minimization of risk using comprehensive pre-deployment risk assessment provides the best option for mitigation. The main elements of this risk assessment should include investigations of caprock integrity, an understanding of CO2 movement through geological strata, an understanding of the likely sea floor footprint of a leak and the siting of sequestration in regions of naturally high oceanic mixing, away from ecological or economically sensitive areas.

13.6

Future trends

The general consensus is that excess CO2, whether injected, leaked or absorbed into the marine environment, will cause potentially serious and © Woodhead Publishing Limited, 2010


367

harmful impacts to components of the ecosystem in the vicinity of the incursion. Further, it is likely that, if the CO2 incursion is sufficiently large (in magnitude, duration or spatial scale), ecosystem functionality, regionally or globally important biogeochemical cycles and marine resources could be deleteriously affected. Hence Ocean Acidification (OA) – the global uptake of atmospherically borne anthropogenic CO2 emissions – is considered to be a serious threat to the integrity of the marine system. Regarding CO2 leaking from CCS facilities, and ignoring any consideration of leakage probability, it is reasonable to presume that impacts to marine chemistry at the epicentre of a leak may well be larger than what could be generated by OA and that impacts would be noted. However, the experimental and theoretical evidence suggests that the mixing (and dispersion) inherent in shelf seas coupled with the likely scales of any leakage event means that impacts from CCS leakage would be spatially constrained. Thus such events, should they occur, must be put into context. Clearly, as trawling and dredging and similar activities demonstrate, shelf seas can tolerate serious disruption to restricted areas. Some evidence suggests that only a very large (104 t CO2 per day), longterm leak (Blackford et al., 2008) would be capable of causing damage at a regionally significant scale. One of the key challenges, therefore, is to refine our understanding of the spread and dispersion of injected CO2 in a variety of situations, relating to tidal patterns, currents and weather, which together drive the mixing of shelf waters. Whilst the theoretical elements are in place, but not yet coupled, there is a lack of observational evidence by which models could be verified. There are also a number of remaining areas of limited knowledge. The first of these is the impact on sediment chemistry and biota of CO2 percolating from below; many experimental manipulations have, for practical reasons, assessed benthic impacts by adding CO2 to the overlying water. Given that marine sediments exhibit strong vertical chemical and biological gradients, it is likely that direction of the CO2 source would be an important consideration. A second area of limited knowledge is the possible contamination of leaked CO2 and the integrated effect of CO2 and pollutants. Initial experiments indicate the potential for one to exacerbate the impacts of the other, but much more work remains to be done. A third area requiring consideration relates to the potential short to medium term nature of some leakage scenarios and consequently the ability of individuals and systems to recover from perturbations.

13.7


∑ A readable description of the chemistry and ecological implications of adding CO2 to marine systems, in the context of ocean acidification, can be found in the Royal Society report of 2005 by Prof John Raven et al.:


368


‘Ocean acidification due to increasing atmospheric carbon dioxide’. This can be downloaded from: http://www.co2storage.org.uk/Technical/Ocean/ Ocean_acidificatiion_Jun_05_Roy_Soc.pdf (accessed January 2010), ∑ A more detailed description of the chemistry of CO2 in sea water can be found in Dickson, A.G., Sabine, C.L. and Christian, J.R. (eds) 2007. Guide to Best Practices for Ocean CO2 Measurements. PICES special publication 3, 191pp. This can be sourced from http://cdiac.ornl.gov/ oceans/Handbook_2007.html (accessed January 2010). ∑ The Parliamentary Office of Science and Technology has released two ‘Postnotes’ describing succinctly the issues around CCS. – CARBON CAPTURE AND STORAGE (CCS) http://www.parliament. uk/documents/upload/POSTpn238.pdf (accessed January 2010) – CO2 CAPTURE, TRANSPORT AND STORAGE http://www. parliament.uk/documents/upload/POSTpn335.pdf (accessed January 2010) ∑ The Intergovernmental panel on Climate Change released a report on CCS ‘CCS, IPCC Special Report, Working Group III, September 2005’, which is available from Cambridge University Press (http://www.cambridge. org/ipcc), The Edinburgh Building, Shaftesbury Road, Cambridge, CB2 2RU England. A summary for policy makers can be sourced from: http:// www.ipcc.ch/publications_and_data/publications_and_data_reports. htm. The knowledge base has not yet enabled a comprehensive report on the potential ecosystem impacts of a leak from CCS; please refer to this chapter and the references cited below.

13.8

References

Alendal G and Drange H (2001) Two-phase, near field modelling of purposefully released CO2 in the ocean. J. Geophys. Res. Oceans, 106(C1), 1085-1096. Archer D (2005) Fate of fossil fuel CO2 in geologic time. J. Geophys. Res., 110, C09S05. Armsworthy S L (2005) Chronic effects of synthetic drilling mud on sea scallops (Placopectenmagellanicus). In Armsworthy SL, Cranford P.J., Lee K. (eds) Offshore oil and gas Environmental Effects Monitoring: Approaches and Technologies: Batelle Press, Columbus, OH, 243–265. Barlow M J and Kingston P F (2001) Observations on the effects of barite on the gill tissues of the suspension feeder Cerastoderma edule (Linne) and the deposit feeder Macoma balthica (Linne). Mar. Polluti. Bull. 42, 71–76. Beesley A, Lowe D M, Pascoe C K and Widdicombe S (2008) Effects of CO2-induced seawater acidification on the health of Mytilus edulis. Clim. Change, 37, 215–225. Berge J A, Bjerkeng B, Pettersen O, Schaanning M T and Oxnevad S (2006) Effects of increased seawater concentrations of CO2 on growth of the bivalve Mytilus edulis L. Chemosphere, 62, 681–687.



369

Bibby R, Cleall-Harding P, Rundle S, Widdicombe S and Spicer J (2007) Ocean acidification disrupts induced defences in the intertidal gastropod Littorina littorea. Biol. Lett., 3, 699–701. Bibby R, Widdicombe S, Parry H, Spicer J I and Pipe R (2008) Impact of ocean acidification on the immune response of the blue mussel Mytilus edulis. Aquat. Biol., 2, 67–74. Bigalke N, Rehder G and Gust G (2008) Experimental investigation of the rising behavior of CO2 droplets in seawater under hydrate-forming conditions. Environ. Sci. Technol., 42(14), 5241–5246. Blackford J C and Gilbert F J (2007) pH variability and CO2 induced acidification in the North Sea. J Mar. Syst., 64, 229–241. Blackford J C, Jones N, Proctor R and Holt J (2008) Regional scale impacts of distinct CO2 additions in the North Sea. Mar. Pollut. Bull., 56, 1461–1468. Blackford J, Jones N, Proctor R, Holt J, Widdicombe S, Lowe D and Rees A (2009) An initial assessment of the potential environmental impact of CO2 escape from marine carbon capture and storage systems. J. Power Energy, 233, 269–280. Brewer P E, Peltzer G, Friedrich and G Rehder (2002) Experimental determination of the fate of rising CO2 droplets in seawater. Environ. Sci. Technol., 36, 5441–5446. Brewer P G, Chen B, Warzinki R, Baggeroer A, Peltzer E T, Dunk R M and Walz P (2006) Three-dimensional acoustic monitoring and modeling of a deepsea CO2 droplet cloud. Geophys. Res. Lett., 33, L23607. Caldeira K and Wickett M E (2003) Anthropogenic carbon and ocean pH. Nature, 425, 365. Caldeira K and Wickett M E (2005) Ocean model predictions of chemistry changes from carbon dioxide emissions to the atmosphere and ocean. J. Geophys. Res., 110, C09S04. Caldeira K, Akai M, Brewer P, Chen B, Haugan P, Iwama T, Johnston P, Kheshgi H, Li Q, Ohsumi T, Poertner H, Sabine C, Shirayama Y and Thomson J (2005) Ocean storage. In Metz B, Davidson O, de Coninck H, Loos M and Meyer L (eds), IPCC Special Report on Carbon Dioxide Capture and Storage, Cambridge University Press, Cambridge, UK, 277–318. Chen B, Nishio M, Song Y and Akai M (2007) Fate of CO2 leaked from seabed. Geochimi. Cosmochimi. ACTA, 71(15), A164. Chen B, Nishio M, Song Y and Akai M (2009) The fate of CO2 bubble leaked from seabed. In Gale J, Herzog H and Braitsch J (eds) Greenhouse Gas Control Technologies 9, Proceedings of the Ninth International Conference on Greenhouse Gas Control Technologies (GHGT9), Energy Procedia, 1, 4969–4976. Cranford P J and Gordon D C (1991) Chronic sublethal impact of oil based drilling mud cuttings on adult sea scallop. Mar. Pollut. Bull., 22(7), 339–344. Cranford P J and Gordon D C (1992) The influence of dilute clay suspensions on sea scallop (Plactopectin magellanicus) feeding activity and tissue growth. Neth. J. Sea Res., 30, 107–120. Dashfield S L, Somerfield P J, Widdicombe S, Austen M C and Nimmo M (2008) Impacts of ocean acidification and burrowing urchins on within-sediment pH profiles and subtidal nematode communities. J. Exp. Mar. Biol. Ecol., 365, 46–52. Dupont S, Havenhand J, Thorndyke W, Peck L and Thorndyke M (2008) Near-future level of CO2 ocean acidification radically affects larval survival and development in the brittlestar Ophiothrix fragilis. Mar. Ecol. Prog. Ser., 373, 285–294. Fernley P W, Moore M N, Lowe D M, Donkin P and Evans S (2000) Impact of the Sea Empress oil spill on lysosomal stability in mussel blood cells. Mar. Environ. Res., 50, 451–455. © Woodhead Publishing Limited, 2010

370


Field C B, Behrenfeld M J, Randerson J T and Falkowski P (1998) Primary production of the biosphere: Integrating terrestrial and oceanic components. Science, 281, 237–240. Gazeau F, Quiblier C, Jansen J, Gattuso J P, Middleburg J, Heip C (2007) Impact of elevated CO2 on shellfish calcification. Geophys. Res. Lett., 34, 1–15. Gil-Martens L, Witten P, Fivelstad S, Huysseune A, Sævareid B, Vikeså V and Obach A (2006) Impact of high water carbon dioxide levels on Atlantic salmon smolts (Salmo salar L.): effects on fish performance, vertebrae composition and structure. Aquaculture, 261, 80–88. Green M A, Jones M E, Boudreau C L Moore R L and Westman B A (2004) Dissolution mortality of juvenile bivalves in coastal marine deposits. Limnol. Oceanogr., 49, 727–734. Hall-Spencer J M, Rodolfo-Metalpa R, Martin S, Ransome E, Fine M, Turner S M, Rowley S J, Tedesco D and Buia M-C (2008) Volcanic carbon dioxide vents reveal ecosystem effects of ocean acidification. Nature, 454, 96–99. Hardman-Mountford N J, Moore G, Bakker D, Watson A J, Schuster U, Barciela R, Hines A, Moncoiffé G, Brown J, Dye S, Blackford J, Somerfield P J, Holt J, Hydes D J and Aiken J (2008) An operational monitoring system to provide indicators of CO2-related variables in the ocean. ICES J. mar. Sci., 65, 1498–1503. Haugan P M (2004) Possible Effects of Ocean Acidification. Reports in Meteorology and Oceanography, Report Number 2-2004, University of Bergen. Haugan P M, Turley C M and Pörtner H O (2006) Effects on the Marine Environment of Ocean Acidification resulting from Elevated Levels of CO2 in the Atmosphere. OSPAR Commission Report, London, UK. Havenhand J N, Butler F R, Thorndyke M C and Williamson J E (2008) Near-future levels of ocean acidification reduce fertilization success in a sea urchin. Curr. Biol., 18(15), 651–652. Huesemann M H, Skillman A D and Crecelius E A (2002) The inhibition of marine nitrification by ocean disposal of carbon dioxide. Mar. Poll. Bull., 44, 142–148. Ishimatsu A, Hayashi M, Lee K S, Kikkawa T and Kita J (2005) Physiological effects on fishes in a high-CO2 world. J. Geophys. Res., 110, C09S09. Johnson A I, Besik F and Hamielec A E (1969) Mass transfer from a single rising bubble. Can. J. Chem. Eng., 47, 559–564. Kikkawa T, Kita J and Ishimatsu A (2004) Comparison of the lethal effect of CO2 and acidification on red sea bream (Pagurus major) during the early developmental stages. Mar. Pollut. Bull., 48, 108–110. Kikkawa T, Watanabe Y, Katyayama Y, Kita J and Ishimatsu A (2008) Acute CO2 tolwerance limits of juveniles of three marine species of invertebrates, Sepia lycidas, Sepioteuthis lessoniana and Marsupenaeus japonicus. Plankton Benthos Res., 3(3), 184–187. Kurihara H and Shirayama Y (2004) Effects of increased atmospheric CO2 on sea urchin early development. Mar. Ecol. Prog Ser, 274, 161–169. Kurihara H, Shimode S and Shirayama Y (2004) Effects of raised CO2 concentration on the egg production rate and early development of two species of marine copeopods (Acartia steurei and Acartia erythraea). Mar. Pollut. Bull., 49, 721–727. Kurihara H, Kato Sand Ishimatu A (2007) Effects of increased seawater pCO2 on the early development of the oyster Crassostrea gigas. Aquat. Biol., 1, 91–98. Langenbuch M and Pörtner H O (2002) Changes in metabolic rate and N excretion in the marine invertebrate Sipunculus nudus under conditions of environmental hypercapnia: identifying effective acid-base variables. J. Exp. Biol., 205, 1153–1160. © Woodhead Publishing Limited, 2010


371

Langenbuch M and Pörtner H O (2004) High sensitivity to chronically elevated CO2 levels in a eurybathic marine sipunculid. Aquat. Toxicol., 70, 55–61. Langenbuch M, Bock C, Leibfritz D and Pörtner H O (2006) Effects of environmental hypercapnia on animal physiology: A 13C-NMR study of protein synthesis rates in the marine invertebrate Sipunculus nudus. Comp. Biochem. Physiol., 144A, 479–484. Mangum C P (1997) Invertebrate blood oxygen carriers. In Dantzler W H (ed.), Handbook of Physiology: Section 13: Comparative Physiology, Vol. II, Oxford University Press, Oxford, UK, 1097–1131. Mangum C P and Burnett L E (1986) The CO2 sensitivity of the hemocyanins and its relationship to Cl– sensitivity. Biol. Bull., 171, 248–263. May R M (1994) Biological diversity: differences between land and sea. Phil. Trans. Roy. Soc., 343B, 105–111. Michaelidis B, Ouzounis C, Paleras A and Pörtner H O (2005) Effects of long-term moderate hypercapnia on acid-base balance and growth rate in marine mussels (Mytilus galloprovincialis). Mar. Ecol. Prog. Ser., 293, 109–118. Miles H, Widdicombe S, Spicer J I and Hall-Spencer J (2007) Effects of anthropogenic seawater acidification on acidbase balance in the sea urchin Psammechinus miliaris. Mar. Pollut. Bull., 54, 89–96. Metzger R, Sartoris F J, Langenbuch M and Portner H O (2007) Influence of elevated CO2 concentrations on thermal tolerance of the edible crab Cancer pagurus. J. Therm. Biol., 32, 144–151. Morse M P, Robinson, W E and Wehling W E (1982) Effects of sublethal concentrations of drilling mud components attapulgite and Q-Broxin on the structure and function of the gill of scallop, Placopectin magellanicus (Gmelin). In Vernberg W B, Calabrese A, Thurberg F P and Vernberg F J (eds), Physiological Mechanisms of Pollutant Toxicity, Academic Press, London, UK, 235–259. Olsgard F, and Gray J S (1995) A comprehensive analysis of the effects of offshore oil and gas exploration and production on the benthic communities of the norwegian continental shelf. Mar. Ecol. Prog. Ser., 122, 277–306. Olsgard F, Schaanning M T, Widdicombe S, Kendall M A and Austen M C (2008) Effects of bottom trawling on ecosystem functioning: an experimental assessment. J. Exper. Mar. Biol. Ecol., 366, 123–133. Orr J C, Fabry V J, Aumont O, Bopp L, Doney S C, Feely R A, Gnanadesikan A, Gruber N, Ishida A, Joos F, Key R M, Lindsay K, Maier-Reimer E, Matear R, Monfray P, Mouchet A, Najjar R G, Plattner G-K, Rodgers K B, Sabine C L, Sarmiento L J, Schlitzer R, Slater R D, Totterdell I J, Weirig M-F, Yamanaka Y and Yool A (2005) Anthropogenic ocean acidification over the twenty-first century and its impact on calcifying organisms. Nature, 437, 681–686. Pane E F and Barry J P (2007) Extracellular acid-base regulation during short-term hypercapnia is effective in a shallow-water crab, but ineffective in a deep-sea crab. Mar. Ecol. Prog. Ser., 334, 1–9. Pearson P N and Palmer M R (2000) Atmospheric carbon dioxide concentrations over the past 60 millions years. Nature, 406, 695–699. Pörtner H-O (2008) Ecosystem effects of ocean acidification in times of ocean warming: a physiologist’s view. Mar. Ecol. Prog. Ser., 373, 203–217. Pörtner H-O and Farrell A P (2008) Physiology and climate change. Science, 322, 690–692. Pörtner H O, Langenbuch M and Reipschläger A (2004) Biological impact of elevated ocean CO2 concentrations: lessons from animal physiology and earth history. J. Oceanogr., 60, 705–718. © Woodhead Publishing Limited, 2010

372


Pörtner H O, Langenbuch M and Michaelidis B (2005) Synergistic effects of temperature extremes, hypoxia, and increases in CO2 on marine animals: from earth history to global change. J. Geophys. Res., Oceans, 110, 1–15. Putnam R W and Roos A (1997) Intracellular pH regulation. In Hoffman J F and Jamieson J D (eds), Handbook of Physiology, Cell Physiology, Oxford University Press, New York, 389–440. Raven J, et al. (2005) Ocean Acidification Due to Increasing Atmospheric Carbon Dioxide. The Royal Society, London, UK. Reaka-Kudla M L (1997) The global biodiversity of coral reefs: A comparison with rainforests. In: Reaka-Kudla, M L, Wilson D E and Wilson E O (eds.), Biodiversity II: Understanding and Protecting Our Natural Resources, Joseph Henry/National Academy Press, Washington, DC, 83–108. Reibesell U, Zondervan I, Rost B, Tortell P D, Zeebe R E and Morel F M M (2000) Reduced calcification of marine plankton in response to increased atmospheric CO 2. Nature, 407, 364–367. Sakai K (1998a) Delayed maturation in the colonial coral Goniastrea aspersa (Scleractinia): Whole colony mortality, colony growth and polyp egg production. Res. Popul. Ecol., 40, 287–292. Sakai K (1998b) Effect of colonoy size, polyp size and budding mode on egg production in a colonial coral. Biol. Bull., 195, 319–325. Seibel B A and Walsh P J (2001) Potential impacts of CO2 injection on deep sea biota. Science, 294, 319–320. Seibel B A and Walsh P J (2003) Biological impacts of deep sea carbon dioxide injection inferred from indices of physiological performance. J. Exp. Biol., 206, 641–650. Shirayama Y and Thornton H (2005) Effects of increased atmospheric CO2 on shallow water marine benthos. J. Geophys. Res., 110, C09S08) Snelgrove PVR (1999) Getting to the bottom of marine biodiversity: Sedimentary habitats – Ocean bottoms are the most widespread habitat on Earth and support high biodiversity and key ecosystem services, Bioscience, 49 129–138. Spicer J I, Raffo A and Widdicombe S (2007) Influence of CO2-related seawater acidification on extracellular acid-base balance in the velvet swimming crab Necora puber. Mar. Biol., 151, 1117–1125. Sprague J B and Logan W J (1979) Separate and joint toxicity to rainbow-trout of substances used in drilling-fluids for oil-exploration. Environ. Pollut., 19, 269–281. Steinacher M, Joos F, Frölicher T L, Plattner G-K and Doney S C (2008) Imminent ocean acidification projected with the NCAR global coupled carbon cycle-climate model. Biogeosciences Discuss., 5, 4353–4393. Taban I C, Bechmann R K, Torgrimsen S, Baussant T and Sanni S (2004) Detection of DNA damage in mussels and sea urchins exposed to crude oil using comet assay. Marine Environ. Res., 58, 701–705. Thomas H, Schiettecatte L-S, Suykens K, Koné Y J M, Shadwick E H, Prowe A E F, Bozec Y, Baar H J W and de, Borges A V (2008) Enhanced ocean carbon storage from anaerobic alkalinity generation in coastal sediments. Biogeosciences Discuss., 5, 3575–3591. Weber R E (1980) Functions of invertebrate hemoglobins with special reference to adaptations to environmental hypoxia. Amer. Zool., 20, 79–101. Widdicombe S and Austen M C (1998) Experimental evidence for the role of Brissopsis lyrifera (Forbes, 1841) as a critical species in the maintenance of benthic diversity and the modification of sediment chemistry. J. Exper. Mar. Biol. Ecol., 228(2), 241–255. © Woodhead Publishing Limited, 2010


373

Widdicombe S and Needham H R (2007) Impact of CO2 induced seawater acidification on the burrowing activity of Nereis virens and sediment nutrient flux. Mar. Ecol. Prog. Ser., 341, 111–122. Widdicombe S and Spicer J I (2008) Predicting the impact of ocean acidification on benthic biodiversity: what can physiology tell us? J. Exp. Mar. Biol. Ecol., 366, 187–197. Widdicombe S, Dashfield S L, McNeill C L, Needham H R, Beesley A, McEvoy A, Oexnevad S, Clarke K R and Berge J A (2009) Effects of CO2 induced seawater acidification on infaunal diversity and sediment nutrient fluxes. Mar. Ecol. Prog. Ser., 379, 59–75. Wood C M and Cameron J N (1985) Temperature and physiology of intracellular and extracellular acid-base regulation in the blue crab Callinectes sapidus. J. Exp. Biol., 114, 151–179. Wood H L, Spicer J I and Widdicombe S (2008) Ocean acidification may increase calcification rates, but at a cost. Proc. Biol. Sci., 275, 1767–1773. Zange J, Grieshaber M K and Jans A W (1990) The regulation of intracellular pH estimated by 31P-NMR spectroscopy in the anterior byssus retractor muscle of Mytilus edulis L. J. Exp. Biol., 150, 95–109. Zeebe R E and Wolf-Gladrow D A (2001) CO2 in seawater: equilibrium, kinetics and isotopes. Elsevier Oceanography Series 65, Elsevier, Amsterdam, the Netherlands.


14

Industrial utilization of carbon dioxide (CO2) M. A r e s t a and A. D i b e n e d e t t o, University of Bari, Italy Abstract: The chapter presents the various aspects of the utilization of CO2 (technological, chemical, biotechnological) together with an analysis of the benefits derived from such practice. Conditions for correct use of CO2 are defined, and the potential of each technology is highlighted in terms of reducing emission into the atmosphere and lowering energy and/or material consumption, either directly (recycling of carbon) or indirectly, e.g. when the use of CO2 reduces the emission of products having a much higher climate change power (CCP) than CO2 itself. The potential utilization of CO2 as a tool to store excess or intermittent energies is also discussed, and the production of chemicals or energy products is presented, highlighting existing barriers to a full exploitation. The potential of enhanced fixation into aquatic biomass as a means of recyling CO2 and replacing fossil carbon in the production of chemicals or fuels for the transport sector is discussed. Emphasis is placed on the requirement for research into the potential for CO2 utilization to contribute to the reduction of its accumulation in the atmosphere. Key words: CO2 utilization, industrial-technological, biotechnological, eletrochemical, photoelectrochemical, chemicals, energy products, CO2 sources, the conditions for utilizing CO2.

14.1

Introduction

The utilization of CO2 for the synthesis of chemicals has its roots in the origins of the chemical industry, with applications such as the synthesis of soda Solvay (Na2CO3, 1861),1 salycilic acid (1869)2 and urea (1870),3 a process which is more than 140 years old now. CO2 has also long been used for the synthesis of pigments (Group 2 carbonates with the general formula MCO3) and specialty inorganic carbonates. Interestingly, all such applications are thermal reactions which do not need a catalyst. Almost a century went by between these pioneering applications and the development of the first catalytic industrial application of CO2 in 1972, namely the copolymerization of CO2 and olefins such as propene (1972).4 Interest in the chemistry of CO2 was sustained during the 1970s by the desire to understand how nature converts around 200 GtC/y in the carbon cycle. Several fundamental studies, aimed at the elucidation of the role of metals in the activation and conversion of CO2, were inspired by the discovery of the first CO2–transition metal complex, namely (PCy3)2NiCO2,5 and the demonstration that the co-ordination of CO2 to a metal centre was able 377 © Woodhead Publishing Limited, 2010

378


to promote its reduction to CO under very mild conditions, instead of the harsh conditions required by the free molecule.6 Since then, the interaction of CO2 with several metal systems having different electron densities at the metal centre has been thoroughly investigated7 and several modes of coordination of CO2 have been demonstrated.8 At the same time, the use of CO2–metal complexes as catalysts in the functionalization of organic substrates was attempted and several new reactions were discovered.9 The use of transition metal complexes as electrocatalysts was also investigated10 and some interesting electrosyntheses of chemicals and pharmaceuticals were discovered10, 11a which are still of great interest today.11b Such great academic effort was not, however, followed by a real industrial interest, either because there was no apparent ‘easy’ use of CO2 or because the conditions for pushing towards a change in well-established production processes were not in place. This caused a decrease in academic interest in the chemistry of CO2 which survived in the late 1990s and early 2000s mainly due to the dedication of a few research groups around the world, a minority compared to the large number active during the 1970s and 1980s. One noteworthy development in the years from 1980 to 2000 was the appearance of a new aspect of CO2 utilization which could have a strong role in the future: the new technological applications of CO2, mainly of supercritical (sc) CO2. Sc-CO2 has been used profitably and successfully in a number of cases (dry-washing, extraction, fluid in circuits, solvent and reagent), showing a potential contribution to the reduction of the impact of chemicals on climate change. The recent increase in the price of oil and the new understanding of the need to reducing the climate change impact of the chemical and energy industry in general, and the emission of CO2 in particular, has sparked a renewed interest in issues such as the use of renewable sources of energy and alternative feedstock for the chemical industry, two issues that somehow merge into the enhanced industrial utilization of CO2 and its fixation in aquatic biomass. These are new areas of huge potential for C-recycling with reduction of CO2 emissions. Moreover, the idea of using CO2 conversion as a tool for energy storage is attracting much research attention. Overall, there is now a great deal of interest in assessing the potential for ‘CO2 utilization’, not only in chemical processes but, in general, as a tool for reducing the emissions that have a major impact on climate change.

14.2

The conditions for using carbon dioxide (CO2)

Utilization of CO2 with the aim of reducing its immission into the atmosphere (whether technological, biological or chemical) must meet three essential requirements:



379

1. the new process based on CO2 must reduce the overall CO2 emission compared to the process currently on stream; 2. the new process must be safer and more eco-friendly than the old one; 3. the new process or application must be economically viable. These are absolute ‘musts’: there is no point in using CO2 in ways that increase its emission, are less healthy or more risky than current usage and which impact more negatively on the environment or involve a higher economic cost. In sum, the new process must be economically, energetically, environmentally and socially viable. We will now discuss a few examples in order to highlight some specific aspects of these concepts. The reduction of the overall CO2 emission in a given application of CO2 is not easily quantified. Such an objective means that both the energy and mass balance must be minimized in the new process, and this is not a trivial operation. Reducing energy implies control of several process parameters, such as temperature, pressure, energy use (quality and quantity) in general, and post-reaction operations, namely separation, isolation, purification, etc. Mass control requires more direct (fewer steps), effective (high yield) and selective (product entropy control) processes, with waste (gas, liquid, solid) minimized at source, and with lower loss of atoms and a higher degree of C-atoms utilization. The above considerations are valid whatever use one wishes to make of CO2. Implementing safer and eco-compatible processes is of key importance in any field of application. Avoiding toxic reagents and by-products will reduce the cost of processing while producing a lower environmental impact, in terms of end-of-pipe treatments, storage and disposal cost. CO2 is not a toxic substance, but it is important to remember that it may become an asphyxiating agent at concentrations above 10 %. Therefore, under controlled conditions of utilization, it does not give rise to serious concern as safe conditions are easily created and implemented. Usually there is no need to work at very high pressure, the highest (30–40 MPa) being used when sc-CO2 is used as solvent and/or reagent. The new process must be economically acceptable for exploitation. This aspect is also quite complex as the correct term of reference must be considered. The actual mode of carrying out an operation (chemical synthesis or any other industrial process) generates waste and, in general, an environmental loading that has given rise to a number of regulatory actions aimed at saving natural resources for future utilization. Comparison of methodologies or technologies must be performed taking into account the ‘global cost’ of the good or service, comprising not only the production cost (e.g., the cost of a chemical at the level of the production plant), but also all accessory costs linked to the production and utilization of that good or service. A life-cycle analysis (LCA) comparing all steps involved in the production–use–disposal


380


of the good is necessary. Hence the real evaluation of a good or service is obtained by summing the contribution of the following operations: (extraction of primary materials from natural resources + treatment and disposal of produced waste) + (production of raw materials or intermediates + associated waste treatment and disposal) + (production of the good or service + associated waste treatment and disposal) + (utilization costs + associated waste treatment and disposal) + disposal of the useless good or service. It is most likely that the last two terms in the sum above will be the same whatever the origin of the product and the production process as they are associated with the use of the good under consideration. Conversely, the first three terms will strongly depend on the raw material used and the production route, on which also depends the presence of trace elements that can become important and play a key role in the use–disposal of the good. It is only by comparing the values of all the parameters listed above that a sound comparison of two technologies can be made, and this requires a Life Cycle Assessment (LCA). Therefore, when considering CO2, it is important to state the context in which we are locating the emission and to carry out a cost-analysis with respect to CO2 availability.

14.3

The carbon dioxide (CO2) sources and its value

CO2 can be obtained from several sources at different prices. Up to now, natural deposits (wells) have been exploited. The cost of extracting CO2 from the well is relatively low (15–20 Euros per tonne) and the purity can be very high (> 99 %), with toxic compounds absent. This means that such extracted CO2 is very useful, possibly also as a beverage additive or food preservative. CO2 extracted from natural wells has also been used for purposes which do not require a high purity of the gas: for example, in enhanced oil recovery (EOR). It would seem logical that the extraction of CO2 should be stopped and recovered CO2 (from power generation plants or industrial processes) should be used instead. However, such captured CO2 is characterized by a different degree of purity and may require deep purification operations prior to application in the food industry. CO2 recovered from power plant flue gases may be contaminated with SOx, NOy, non- or partially-combusted toxic chemicals; CO2 from chemical industrial plants may be accompanied by chemicals typical of the process in which it is generated; while CO2 from fermentation reactors may have the purity required for food applications. The purification steps will affect the cost of CO2. In principle, products characterized by different degrees of purity may have different applications. The market price of CO2 depends on the geographic area and can be very variable; as high as $400/t12 if food quality CO2 is requested. The utilization



381

of CO2 recovered from fermentation reactors may be very convenient as its purity is very high and, in principle, contaminants should not be present. The drawback of such a source is the seasonality of raw materials, a barrier that can be circumvented by some means. Therefore, two scenarios for the use of CO2 are foreseeable: 1. CO2 is recovered from industrial and energy sources: this will imply that the energy, materials and economic costs of the recovery must be taken into account in assessing the cost and environmental impact of a new process based on CO2. 2. CO2 is available on the market because its capture from flue gases has been implemented on a large scale or new technologies provide concentrated sources of CO2 (IGCC): recovered CO2 is, thus, a waste product requiring disposal: its utilization produces a benefit both in terms of carbon recycling and added value of the derived products. The difference between the two scenarios above is the economic and energy cost of the recovery of CO2, and this is of key importance in a LCA study. Large-scale utilization of CO2 and large-scale recovery of CO2 can be seen as intrinsically linked. In this chapter, it will be assumed that large amounts of CO2 are available because CO2 separation has been implemented on a large scale. In this scenario, the utilization of CO2 has to be compared with its disposal: any one of the uses described below: (i) will produce an economic benefit compared to disposal which itself entails an economic cost, and (ii) may turn out to be energetically more convenient than disposal, the latter being in all cases an energy-consuming technology (because of the energy required for the liquefaction of CO2, transportation, housing).

14.4

Technological uses of carbon dioxide (CO2)

The use of CO2 as a technological fluid includes all those applications in which CO2 is not converted into other chemicals. A list of such uses is given in Table 14.113 together with the overall market. Although this option does not convert CO2 into storable or disposable materials, and it is usually the case that at the end of the application CO2 is vented (it is only rarely recovered and recycled), nevertheless the use of CO2 as technological fluid may contribute to the reduction in the impact on climate change. The benefit comes from the fact that CO2 is a substitute for other chemicals such as chlorofluorocarbons (CFC), which have much higher climate change power (CCP) (Table 14.2), or chemicals that require energy for their production or which produce waste with a strong environmental impact upon use. The former case refers, for example, to the use of CO2 as a fluid in air-conditioners, the latter to its use as a fumigant or for water treatment, among others.


382

Developments and innovation in CCS technology Table 14.1 Technological uses of CO2: 20 Mt/y Technology

Application

Food industry Fumigant Antifire Mechanical industry Electronics Dry washing Fluid in circuits Water treatment Extraction

Additive to beverages Food packaging, dry-ice Extraction of aromas, caffeine ... Antibacterial agent for cereals Extinguishers Moulding, cutting, soldering Cleaning fluid Cleaning fluid Air conditioning pH control of process waters Enhanced oil recovery (EOR) Extraction of bio-oil from biomass

Table 14.2 Comparison of the climate change power of some CFC with that of CO2 (100 y) Chemical

CCP

Chemical

CCP

Carbon dioxide R134a R22

1 1430 1700

CFC-12 CFC-11 HCF-23

8500 4000 14 800

Let us now consider some practical applications to clarify the concepts above. (i) The production and use of CFC causes the emission into the atmosphere of such chemicals. The estimated amount of CCl2F2 (CFC-12, Table 14.2) released into the atmosphere ranged at the end of the 1970s from 420 (estimated by the Chemical Manufacturers Association–CMA) to 500 kt/y evaluated by direct atmosphere monitoring.14 The substitution of equivalent amounts of CO2 for such chemicals produces a great benefit in terms of CCP reduction, considering that the CCP of CCl2F2 is 8500 times that of CO2 (Table 14.2). (ii) Other typical cases are the substitution of (i) perchloroethene (C2Cl4, PERC, ca. 3 Mt/y) in dry cleaning or (ii) fluorinated cooling gases (e.g., R134) in fixed or mobile air conditioners (A/C). Supercritical CO2 (sc-CO2) is now finding several applications.15 It is largely used in cleaning machines15a as a substitute for PERC. PERC is a highly energy-intensive chemical and its synthesis is highly polluting because of the production of chlorinated waste. If we consider current practice to be a stationary state (with the use of PERC as dry-cleaning agent) and take into consideration only the replacement of PERC lost in running the existing equipment, the replacement of the annual loss of PERC16, 17 is equivalent to 2 Mt/y CO2: this is the minimum estimate for the reduction of CO2 emission as a result of substituting PERC with CO2



383

as cleaning agent. Obviously, should we consider the total energy balance for making PERC and running the existing equipment together with the associated waste, then the balance is much more positive in favour of the substitution of PERC with CO2. Similarly, the use of CO2 as a fluid in automobile or fixed A/C would avoid the effect of 3.3 kt/y of lost R-134 (an average of 0.06 kg/y times 55 Mcars circulating), equivalent to 4.7 MtCO2eq. The IPCC estimate for the impact of emissions from A/C in buildings is 0.6 GtCeq or 2.2 GtCO2eq. Therefore, the use of CO2 as fluid in A/C would significantly contribute to reducing the overall CCP, a result that can contribute to a reduction in CO2 emmission into the atmosphere. The figures above are a demonstration of the direct and indirect benefits associated with the utilization of CO2, benefits that are often hidden and difficult to discover. (iii) The use of CO2 as a fumigant avoids the use either of other pharmaceuticals, which would have a complex molecular structure and would generate a lot of waste for their production, or of chemicals that are highly toxic, such as methylbromide, cyanidric acid, methylisocyanide, formaldehyde, sulphonylfluoride, etc.). The production of chemicals or pharmaceuticals usually has an associated waste production in the range 5–250 t of waste per tonne of product. This waste production is known as the E-factor of a given product (Table 14.3).18 Therefore, assuming that a compound used as an anti-parasite or anti-fungal has an associated organic waste production with an average composition equal to C4 and an E-factor of 70 (we consider a simple molecule), for each tonne of marketed product roughly 12.3 kt of CO2 will be emitted. It is clear that the use of CO2 instead of such a chemical (assuming that the two compounds have the same specific efficacy) is a much better solution and on that also substantially reduces the CO2 emmission into the atmosphere if CO2 is vented after use. (iv) When CO2 is used for basic water treatment, it is usually a substitute for sulphuric acid (H2SO4). The emission factor of H2SO4 is of the order of 5 kg SO2 and 0.3 kg SO3 per tH2SO4, while the entire process looks to be exoergonic (–1.1 GJ/t), assuming that: (i) all the released energy produced in the hydration of SO3 is recovered in the form of steam, and (ii) the use of sulphur obtained from the desulphurization of hydrocarbons is considered Table 14.3 The E-factor of several kinds of chemicals Industrial application

Market (t/y)

E-factor (twaste/tproduct)

Petrochemical Intermediates Fine chemicals Pharmaceuticals

109 > 106 105 104

0.1 0.5–1 5–100 100–250


384


not to imply any energy cost.19 As CO2 has the same neutralization power as H2SO4, for each tonne of CO2 used 2.23 t of H2SO4 will be avoided together with the accompanying environmental impact caused by the sulphate accumulation in water plus the above-mentioned emission of SOx into the atmosphere. In the case that heat is not recovered (which may happen), and we take into account the energy cost of sulphur production, then the energy consumption must be considered, making the balance much more in favour of using CO2. (v) Supercritical CO2 (sc-CO2) exists above 31 °C and 7.38 MPa. Properties like density and viscosity can be modulated over a quite wide range by changing the two parameters, pressure and temperature. The ‘dense phase fluid’ has properties close to those of a non-polar organic solvent, such as pentane or dichloromethane. The benefits derived from its use are well known today,15b so that its utilization is spreading in various industrial sectors. Applications for the replacement of traditional organic solvents with sc-CO2 include: ∑ ∑

the decaffeination of coffee beans; the extraction of fragrances and essences from plants, or proteins or fatty acids and hydrocarbons from algae; ∑ the use as solvent for: reactions, crystallizations, preparation of solid thermal-sensitive pharmaceuticals having controlled size distribution; catalysis (homogeneous and heterogeneous); the synthesis and modification of polymers including perfluoropolymers, or as mobile phase for supercritical fluid chromatography (SFC), dyeing, dry-cleaning, nuclear waste treatment. A specific application that is attracting much interest is the use of sc-CO2 as solvent and reagent.

The most important feature attached to the use of sc-CO2 is that it can be easily recovered at the end of the process (by thermal decompression), recompressed and recycled. It is also worth mentioning that most waste organic solvents are usually burned: the substitution with sc-CO2 when possible will avoid large volumes of emitted CO2. In general, thus, CO2 in its technological applications substitutes species which have a strong impact either on the atmosphere or on water or soil: even if at the end of the application CO2 is vented to the atmosphere, the net result is the avoidance of substantial amounts of CO2-equivalents due to the elimination of chemicals having a high CCP, and this results in an overall mitigation of the impact on climate change.

14.5

Biological enhanced utilization

The enhanced biological utilization of CO2 encompasses the fixation of CO2 into biomass under conditions very different (industrial) from natural ones. Typical examples are: (i) the cultivation of terrestrial biomass in greenhouses



385

under a CO2 concentration in the gas phase of ca. 600 ppm compared to the natural occurrence of 380 ppm; (ii) the culturing of aquatic biomass by sparging CO2 in water or under a gas phase concentration up to 150 times the natural one. It is usually only ornamental plants and vegetables that are grown in greenhouses because of the operating cost: energy crops and plants are cultivated in open areas under the atmospheric pressure of CO2. Therefore, aquatic biomass is more suitable than terrestrial for growing under high CO2 concentrations (industrial production). It should be noted that the need to decouple energy issues from both land use and food production is prompting a move away from the first-generation biofuels (or crop-derived biofuels) towards second-(use of cellulosic materials and lignine) and third-generation biofuels, including aquatic biomass. The exploitation of microalgae, macroalgae, plants or any other vegetal biomass growing in water is a strategy that may substantially contribute to the production of large volumes of biofuels and help to meet the target of 20 % substitution of transport fossil fuels with biofuels by 2020, which represents the target of several industrialized countries.20 Algae are better converters of solar energy (h = 6–8 % under natural conditions, up to 9–10 % in bioreactors) than superior plants (h = 1.5–2.2 %), and they also have a better potential for fuel production diversification. In fact, bio-oil and biodiesel, biogas, bioethanol and biohydrogen, can be produced, depending on the type of aquatic biomass used and its composition. Also, microalgae, macroalgae and plants have different compositions and can thus be used for different purposes. Table 14.4 shows the lipid accumulation capacity of two different kinds of aquatic biomass: microalgae and macroalgae. It is evident that microalgae are, in general, richer in lipids than macroalgae. Another point to consider when comparing terrestrial and aquatic biomass is the productivity per hectare. This could be quite an important factor, influencing the choice of which option to implement. Arable land is required for food production and should not be diverted to energy production. Aquatic biomass can be grown on marginal coastal areas, on desert lands close to salty water or offshore. This greatly increases its potential as source of fuels compared to terrestrial. Table 14.5 compares these two kinds of biomass, showing their different requirements. Table 14.4 Lipid accumulation capacity of some microalgae or macroalgae Species or strain Microalgae

Lipids % dry weight

Species or strain Macroalgae

Botryococcus braunii Nannoclhoropsis sp Schizodetrium sp Nitzschia sp

25–75 31–68 50–77 45–47

Codium harveyi 9–12 duthyae 12–21 fragile 21 Cladophora 12–20


Lipids % dry weight

386


Table 14.5 Comparison of the properties of terrestrial and aquatic biomass Terrestrial biomass

Aquatic biomass

• Light efficiency 1.5–2.2 % • Requires land and water for growing • Productivity depends on soil quality (for a given plant) • Soil additives may be required (environmental and economic costs) • Biomass is generally rich in lignocellulosic components • Cereals and seed plants are mostly used • Open area more than greenhouse cultivation

• Light efficiency 6–8 % (or higher when irradiated bioreactors are used) • Richer in water • May not require land for cultivation (coastal ponds, offshore basins); can be grown in process and municipal waters • Low lignocellulose content. Lipid/ protein/polysaccharide content can be adjusted • Easy to grow in bioreactors (light– temperature adjustment); decoupling from climatic conditions

Table 14.6 Comparison of the land requirements for the production of bio-oil for different biomass Terrestrial crop

Bio-oil production (m3/ha)

Corn Soybean Canola Jatropha Coconut Oil palm

0.175 0.447 1.19 1.89 2.7 6.0

Aquatic biomass Microalgae Macroalgae

50–130 20–30

An issue linked to efficiency of light conversion is the productivity per hectare of biofuels. Table 14.6 compares the extent of land required for the production of a given quantity of biodiesel using different terrestrial or aquatic biomass. Also in this case, the use of aquatic biomass looks to be more profitable than the use of terrestrial seed-plants. Figure 14.1 shows a typical algae farm and the raceway technique of cultivating microalgae. The cultivation technique greatly depends on the selected biomass: microalgae require continuous stirring obtained with a paddle-wheel or by flowing the water medium through a succession of basins. Macroalgae can grow floating, either attached or not attached to a hard substrate. The collection or harvesting technique is quite different as well, as is the treatment of the biomass for oil extraction. Microalgae can also be grown in bioreactors installed in many different ways: flat, vertical, slanting, coiled. Solar light or artificial white light can be employed for irradiation.21



387

14.1 An example of an algae farm.

As shown in Table 14.7, the content of lipids may vary over a large range for the same strain, depending on the culture conditions to which micro- and macroalgae usually adapt quite easily. The cellular composition may make a specific strain more or less suitable for the production of a kind of energy vector. Algae richer in lipids are better suited for the production of biodiesel, while algae richer in starch may be used for alcoholic fermentation to produce ethanol and a high protein content, and starch is ideal for biogas production. The adaptation to the growing conditions means that the organisms can be easily manipulated and their lipid content, as well as the protein or starch content, can be adjusted either by genetic modification or, more simply, by physical stress manipulation, i.e. by regulating the N or Ci content of the cultures. Another point of interest is that algal bio-oil rarely comprises a single type of fatty acid; more frequently, the lipid fraction of algae (both microand macroalgae) contains a large variety of fatty acids, as shown in Table 14.8. Nevertheless, as shown in Table 14.7, such distribution can be driven by controlling the CO2 concentration in the culture. Macroalgae (seaweeds) have so far attracted attention only as an agent for bioremediation and waste water treatment and, partly, as a source of chemicals. Their potential for energy production has been only marginally explored.21 As Table 14.4 shows, macroalgae have on average a lower lipid content than microalgae, the best performance reaching 20–30 % lipids. Nevertheless, the lower growing and harvesting costs make them very interesting compared to microalgae as a source for energy or chemicals. Both micro- and macroalgae are rich in chemicals that can be extracted using a series of different technologies, as shown in Table 14.9. Compounds that can be extracted from micro- and macroalgae are: ∑ ∑

coloring substances and antioxidants; enzymes (superoxidedismutase, restriction enzymes, phosphoglycerate kinase, luciferase and luciferin); © Woodhead Publishing Limited, 2010

388

% CO2 in 14:0 16:0 16:1 18:0 18:1 18:2 20:0 20:4 20:5 the gas-phase

Total FAMEs

Control 0.038 Enhanced 10.0

29.1 ± 4.3 55.5 ± 3.7

5.2 ± 1.7 5.0 ± 1.1

9.4 ± 1.6 0.9 ± 0.2 16.2 ± 2.0 0.8 ± 0.1

0.5 ± 0.2 0.5 ± 0.3

5.9 ± 1.8 5.9 ± 1.2 0.2 ± 0.1 11.0 ± 1.6 19.2 ± 2.5 0.3 ± 0.1

0.5 ± 0.2 1.1 ± 0.2

0.6 ± 0.2 1.6 ± 0.4



Table 14.7 Influence of the CO2 concentration on the distribution of fatty acids in Chaetomorpha l. cultured at ambient conditions and under a high concentration (10 % in the gas phase) of CO2


389

Table 14.8 Distribution of fatty acids in lipids present in some macroalgae Fatty acid

Species and relative percentage of organic compounds

Number of carbon atoms/number of unsaturated bonds

Ulva lactuca

Enteromorpha compressa

Padiva pavonica

Laurencia obtuse

Saturated C12 Æ C20

15.0 %

19.6 %

23.4 %

30.15 %

Monounsaturated C14 Æ C20

18.7 %

12.3 %

25.8 %

9 %

Polyunsaturated C16/2 Æ C16/4 C18/2 Æ C18/4, C20/2

66.3 %

68.1 %

50.8 %

60.9 %

Table 14.9 Use of a cascade of technologies for a full use of biomass Very soft Soft Non-destructive Semi-destructive technologies technologies

Hard Destructive technologies

Extraction of molecules with a complex molecular structure; molecular and polymeric compounds for special applications

Breaking of natural complex structures and production of very simple chemicals (CO–H2) that can be used for making new complex molecular compounds (chemicals and fuels) again

Breaking of complex structures Production of energy products or simple chemicals: this is the case with lignine that can be used for the synthesis of phenolic compounds upon hydrolytic treatment.

∑ ∑

polymers (polysaccharides, starch, poly-beta-hydroxybutiric acid); peptides, toxins, aminoacids, steroids, essential oils such as geraniolgeranyl formate or acetate- cytronellol-nonanol-eucalyptol; ∑ pigments, such as chlorophylls, carotenoids, xantophylls; ∑ amines, inorganic compounds. The existence of significant differences in levels of entropy means that the extraction of a product is not always economically viable. However, the fact that algal organisms are able to produce concentrations of a type of substance when subjected to stress may help to reduce the entropy, increasing the concentration of a desired product in the biomass. It is, in fact, possible to grow selected types of algae for the production of, for example, astaxanthine or carotenoids, that are chemicals with high added value. Several of the substances listed above similarly have a high added value that makes their production using this approach economically viable. The implementation of the cascade of technologies allows the extraction, using the most appropriate technology, of substances with a complex structure in addition to the use of the biomass, for example, for the production of more simple molecules such as H2 and CO (Syngas).


390


Table 14.10 shows the differences between macro- and microalgae. The former may afford a much larger productivity and require only very simple growing technologies. Their harvesting is also very simple and low-cost. All together, such positive aspects may compensate for the lower lipid productivity typical of macro- compared to microalgae. In summary, aquatic biomass represents a much larger variety of raw materials compared to fossil fuels and their potential needs to be fully exploited by defining the most appropriate transformation routes and the most suitable technologies. In order to make the most advantageous use of aquatic biomass it is necessary to integrate the existing expertise in the area of aquatic biomass cultivation with nanotechnologies, process intensification and the production of new nanosized materials in a single process. The biorefinery approach is now considered worldwide to be the most suitable for genuine exploitation of the potential of aquatic biomass. The way to evaluate the real energy or economic potential of aquatic biomass is the application of LCA 22 following Fig. 14.2 on page 391. LCA allows us to calculate the energy and substance-flow in the entire process and to establish the real potential of biomass for chemicals and energy production. All in all, aquatic biomass is an interesting source for chemicals and energy that requires accurate investigation in order to discover its full potential. It must nevertheless be emphasized that the fluctuation of the price of fossil carbon (coal, oil, gas) does not favour the implementation of the production of biodiesel from aquatic biomass. With the oil price below 120 US$/barrel it is not economic to produce biodiesel with such biotechnology. Should the concept of biorefinery enter into operation, it will be possible to develop an economically sustainable industry of fuels and chemicals production from aquatic biomass. Should this happen on a large scale, it can be foreseen that quite significant masses of CO2 will be recycled with a considerable reduction in its emmission into the atmosphere. Meeting the target of producing 20 % of transport diesel from biomass will, in fact, avoid the emission of roughly 1 Gt of CO2 per year. More on algae can be found in Chapter 15.

Table 14.10 Performance of micro- and macroalgae Parameters

Microalgae

Macroalgae

Growing season Productivity (dw) Lipid content Production cost Heat value (GJ t–1) Energy cost

250–280 d 33–50 t ha–1 20–75 % dw 100/5000 US$ t–1 dw 21 GJ t–1 dw 0.56 $ MJ–1

210–240 d 10–70 t ha–1 0.3–32 % dw 100 $ t–1 dw 12.2–20 GJ t–1 dw 0.05–0.6 $ MJ–1


Industrial utilization of carbon dioxide (CO2) MWh Fuel

1 – Power plant

kg CO2

kg CO2 Separation and transport

2 – CO2 separation and transport

MJ

kg CO2

3 – CO2 distribution

MJ

4 – Algae production

MJ

5 – Conversion technology

MJ

6 – Net energy produced

MJ

kg CO2 Type of algae kg algae Nutrients

391

Conversion technology Type of algae kg algae Conversion technology kg CO2 kg algae Nutrients

14.2 Representation of the flow-chart for the LCA of the production of biofuels from biomass.

14.6

Carbon dioxide (CO2) conversion as ‘storage’ of excess electric energy or intermittent energies

One of the major problems with the production of electric energy is that technologies for its easy storage are lacking. Also, intermittent energies are often not used as it is neither practical nor convenient to convert them according to their availability. In both cases a solution could be represented by their conversion into chemical energy and then using such a form of energy when and where necessary. A form of chemical energy would be represented by reduced forms of CO2: methane or liquid hydrocarbons being the most suitable as they are already exploited in transportation media. Therefore, in this section the potential conversion of CO2 into energy-rich products by thermal or electrochemical or electrocatalyzed reactions will be discussed. The electrochemical or electrocatalytic reduction10, 11, 23–28 of CO2 to other C1 or Cn molecules (such as CO, CH3OH, CH4, C2H5OH, C2H4, Cnolefins, Cn-hydrocarbons, Cn-alcohols, etc.) as a way to store energy29 or its fixation into chemicals such as formic acid30 has long been investigated. Recent years have seen a revamping of such reactions because of the need to reduce the emmission of CO2 into the atmosphere: the use of excess electric energy for making energy products from CO2 would be exactly the same as using electric energy for pumping water uphill during the night for making


392


electric energy during the day using the hydro-power of falling water, a practice used in hydro-power stations. The conversion of CO2 into energychemicals can be considered, thus, as a form of ‘storage of excess electric energy’ or else of ‘intermittent energies’, which are not usually exploited because they are not suited to continuous generation of energy as requested by most process uses. An analysis will be made of possible processes with the aim of answering the question: how close are we to the exploitation of such conversion of CO2?

14.6.1 Thermal processes Thermal energy must be generated in ways that minimize CO2 emission. Fossil fuel carbon cannot possibly be used to this end. High-temperature gases recovered from industrial processes such as stainless steel plants or from cement plants could represent a solution. The associated heat could be used to produce steam, used in turn to produce electricity or directly for running reactions. Direct use of concentrated solar energy would surely represent an elegant solution. For example, fused salts (obtained by using solar energy) would represent a valuable source of heat for running chemical reductions of CO2. A field of 1 ha of fused salts (Na/K nitrates) would produce a stable temperature of up to 823 K that is suited for running reactions such as those shown in Equation 14.1

CH4 + CO2 Æ 2H2 + 2CO

[14.1a]

C2H6 + 2CO2 Æ 3H2 + 4CO

[14.1b]

Alternatively, long-chain hydrocarbons, which do not find an application as energy vectors, can be reacted with CO2 to produce Syngas (Equation 14.2) used for the production of fuels:

CnH2n + 2 + nCO2 Æ 2nCO + (n + 1)H2

[14.2]

Equations 14.1b and 14.2 show that the longer the C-chain the lower the ratio H2/CO, and this is not good for the quality of the Syngas. In order to improve that ratio, the dry reforming of HC can be combined with wet reforming (Equation 14.3) or with the partial oxidation of methane (Equation 14.4)

CH4 + H2O Æ CO + 3H2

[14.3]

CH4 + 1/2O2 Æ CO + 2H2

[14.4]

This also improves the heat balance, being exothermic (see below). Such an approach may be useful to recycle HCs with a low or high number of carbons which currently have no use.



393

14.6.2 Electrochemical conversion of CO2 One of the main issues with this approach is that, despite all efforts made so far, there is not yet an established procedure that selectively and efficiently creates a product. Table 14.11 lists the potential for multi-electron reduction of CO2. It is noticeable that the mono-electron reduction of CO2 in non-protic media to produce the radical-anion CO2– has a E° = –2.2 V (VHE), much higher than the multi-electron reduction in protic media. Although, in principle, any reduced form of CO2 would be of interest, nevertheless some species such as formic acid (easily and selectively prepared from CO2 and hydrogen under mild conditions30) and oxalate31 have little application as energy vectors and are of no interest from the energetic point of view. Although methane is abundant in nature and does not seem to be the first priority, nevertheless it is often formed as a product of total reduction of CO2. Therefore, it seems quite sound to confine the discussion to the following molecules: CO, methanol, ethanol (or other alcohols), olefins and HCn. Some general considerations are necessary before we look at each of the products short-listed above and describe the state-of-the-art and identify remaining barriers to their exploitation. It must be said that currently there is no selective and efficient means of forming any of the products above so a genuine breakthrough is highly desirable. The key parameters to be taken into consideration for the electrochemical reduction of CO2 are: ∑ the support solvent and the species reduced; ∑ the electrodes; ∑ the eventual use of electrocatalysts; ∑ the competing processes and the faradic efficiency. The support solvent Clearly, the electroreduction reaction is intended to take place in water and not in an organic solvent. This approach is necessary to reduce the cost of operation. In water, CO2 can exist as such or in the form of hydrogen Table 14.11 Electrode-potential (V vs SHE) for some multi-electron reductions of CO2 Reaction and product

E°

Eeq (pH = 7)

CO2 + 2H+ + 2e– Æ HCOOH CO2 + 2H+ + 2e– Æ CO + H2O CO2 + 4H+ + 4e– Æ CH2O CO2 + 6H+ + 6e– Æ CH3OH + H2O CO2 + 8H+ + 8e– Æ CH4 + 2H2O 2CO2 + 2H+ + 2e– Æ (COOH)2

–0.199 –0.103 –0.028 +0.031 +0.169 –0.49

–0.61 –0.52 –0.44 –0.38 –0.25 –0.90


394


carbonate, HCO3– or carbonate, CO32–. There is experimental evidence which indicates that CO2 is reduced32 but not hydrogen carbonate or carbonate. This raises the problem of the solubility of CO2 in water33 and of equilibria which govern the existence of the various forms as a function of the pH of the solution, as depicted in Equations 14.5–14.7.

CO2 (g) + H2O Æ H2O·CO2 (l) HCO3–

H2O·CO2 (l) + H2O Æ

HCO3– + H2O Æ CO32– + H3O+

[14.5] +

+ H 3O

[14.6] [14.7]

The solubility of CO2 is represented by the fraction of ‘free’ CO2 and not by the other species (HCO3–, CO32–) into which CO2 may be converted by reaction with water. Table 14.12 shows the solubility of CO2 in several solvents. It is clear that water is not the best solvent at 0.1 MPa, due to its polarity. Table 14.12 also shows that the solubility increases with pressure. Therefore, one may conduct the electrolysis under pressure. Such an operation would help to increase the concentration of CO2 in the water solution, but would also increase the operating costs. This is a point that must be taken into consideration. The use of organic solvents would not be the best solution due to their market cost, the need to replace the loss (by oxidation, decomposition, evaporation) and pollution issues. Therefore, the development of efficient technologies is necessary in order to use water as solvent with high and selective conversion of CO2. The electrodes Several massive metals have been used as solid electrodes.34 The behaviour of the various metals has been seen to be in some way dependent on their Table 14.12 Solubility of CO2 in several solvents and the influence of pressure1 Solvent

Solubility, 293 K mL/mL at 0.1 MPa

Solubility, 293 K mL/mL at 2.0 MPa

Water Methanol E DMF Acetone B T

0.89 4.20 2.87 5.10 6.98 2.54 2.42

13 110 104.8

1

71.16 57.91

Solubility of Inorganic and Organic Compounds, H. Stephen and T. Stephen, Part 1 and 2, Pergamon Press, London, UK, 1963.



395

electronic configuration, i.e. whether or not they use d-electrons (transition metals or sp-metals).35 The extended work done during the 1970s and 1980s has been summarized in reviews and papers.36–39 Zn, Au and Ag seem to drive the reaction towards the formation of CO, while Cu has good properties for the formation of HCs, alcohols and ethene. Recently, more sophisticated electrodes were used, such as polycrystalline materials,40, 41 supported metals (on polymeric substrates: 35, 42) or porous electrodes 43 or high-throughput (HT) gas diffusion electrodes.44 In general, the nature of the electrode strongly influences the formation of the reduced species as it drives the adsorption of species which can behave as intermediates. An issue to consider is that often the electrodes are consumed during operation, with an impact on operational cost and efficiency. The eventual use of electrocatalysts In order to improve the selectivity of the reduction process and to reduce the effect of electrode passivation which may increase the electrochemical reduction potential, electrocatalysts such as transition metal systems, either homogeneous or heterogeneous, have been used.45 In these systems, the interaction of the substrate is not with the electrode surface but with the catalyst in solution: therefore, the electrode transfers electrons to the substrate through the catalyst. A partial modification of this strategy is the deposition of the catalyst on the electrode: in this way, it is possible to use as cathode a material (carbon electrode) that alone would not be active in the electroreduction but, because it is covered with an active catalyst such as a metal porphyrin.46 results in a quite effective system. Co and Ni complexes have been effectively used as electrocatalysts. The use of suspended particles of a metal is also a technique used for improving the yield and selectivity: Cu has been particularly investigated for its attitude to act as catalyst when coupled with Zn and Pb electrodes.47 The competing processes and the faradic efficiency As the reduction of CO2 is carried out in water, one must expect that the reduction of the proton to dihydrogen can be a competing process that may reduce the selectivity and the yield of reduced CO2. Hydrogen can also play a role as an electrosorbed species Had that could address the formation of the reduced species of CO2. Had can drive the formation of formic acid, or the interaction of adsorbed species (HCOOad) with a product such as methanol to produce acetic acid (Equation 14.8)

HCOOad + CH3OH Æ CH3COOH + OHad

[14.8]

All the above factors influence the CO2 reduction: there is, in fact, not yet


396


a process that presents a high efficiency and a high selectivity towards a single product. In the follow-up to this section, each of the above short-listed chemicals will be considered in detail. Carbon monoxide CO can be used for the production of energy by combustion with dioxygen even if the associated energy is quite modest. Table 14.13 shows the free energy of formation of several C1 species and their combustion heat. It is clear that burning CO affords only a limited amount of heat if compared to other vectors. Nevertheless, such species must be considered as it is easily formed under several working conditions, either alone (photocathode made of p-InP or p-GaAs48, Pt–Pd–Rh alloy49) or together with H2.50 In several other cases, CO is formed in more complex mixtures containing formic acid33, 41, 44f . Mixtures of CO and H2 are of interest as they can be either burned to produce thermal energy or used for the synthesis of HCn. Methanol, ethanol (or other Cn alcohols) Methanol and, more rarely, other Cn oxygenates are formed at variable concentrations under several working conditions. The products40, 43 must be separated from water, and the mixtures obtained are such that they still demand a large energy input for their fractionation so that their best use is the production of thermal energy. As reported above, the search for active catalytic electrodes for CO2 reduction has led to new discoveries that produce real breakthroughs in this area. For example, RuO2 deposited on conductive diamond (boron-doped diamond)51 is very active in such a reduction process.

Table 14.13 Free energy of formation and combustion heat of C1 molecules1 Compound

DG°f (kJ/mol)

DG°comb, (kJ/mol)

CO2g HCOOHl COg CH2Ol CH3OHl CH4g

–394 –361 –137 –102 –166 –51

–262 –257 –560 –714 –881

1

Handbook of Chemistry and Physics, 41st edn, CRC Press, Boca Raton, FL.



397

Olefins Ethene is formed very frequently when Cu-cathodes are used,42, 52 more often in mixture with other species (such as H2 or methane) than as a single product. Nevertheless, it may reach a quite interesting concentration in the range 3253 to 80 %.42 Superior olefins have also been found in some specific conditions: they are found more frequently in thermal reductions of CO2 (see below) but are not common products of electrochemical reductions. Hydrocarbons, HCn Methane is a frequent product of the reduction of CO2. It can be the only product54a,b or it can be generated in mixture with other hydrocarbons,55 CO,32, 56, 57 ethene40, 51, 52, 58 or formic acid.44f, 56 More recently, evidence for the formation of long-chain HC has been reported, but this approach is still in its infancy.59

14.6.3 Photocatalytic reduction of CO2 The use of light harvesting systems able to generate a ‘hole+ –electron’ separation (Fig. 14.3) has been long investigated60, 61 as a tool for photocatalytic reduction of CO2. The barriers to exploitation were the low efficiency of the semiconductors used (h < 0.1 %) under solar light irradiation, the low selectivity of the catalytic system and the need to use sacrificial organic species for the oxidation. The latter makes no sense as often high-value organics (C3-alcohols, other more complex molecules) are oxidized for producing lower value organic molecules (methane, methanol, CO, etc.). Recently, a new interest in photocatalysis for CO2 reduction has developed, essentially because of the discovery of the properties of TiO2-derived catalysts62, 63 and of new transition metal systems.64, 65, 66, 67 The discovery and development of active supramolecular systems has given a strong impetus to such an

hn

CO2 Æ Reduced forms

h+

H2O Æ 1/2O2 + 2H+ + 2e–

e–

14.3 Use of semiconductors for CO2 reduction in water under solar light irradiation.


398


approach.68, 69 Also, new heterogeneous species have shown an interesting activity in water, splitting into H2 and O2,70 a reaction that could be coupled with CO2 reduction. The barriers to overcome are still those listed above: use of solar light, use of water oxidation coupled to CO2 reduction, avoiding organic compounds oxidation. The new systems seem to have a better performance than old ones, and turnover numbers higher than 200 have been reported.68 Still, several improvements to the photocatalytic systems are required because they may find practical application, mainly in relation to the stability, selectivity, utilization with visible light and dioxygen generation from water as oxidized species. The photoelectrochemistry of CO2 is further discussed in Chapter 17.

14.7

Production of chemicals

The key issue is that the new process must be economically and energetically viable. This means that the energy used or/and the waste production must be lower for the CO2-based process compared to existing processes. It should be emphasized, to frame correctly the issue of CO2 utilization, that what is of real interest is not the amount of CO2 that can be used in chemical applications, but more exactly the contribution that the innovative technologies based on CO2 may make to the reduction of the CO2 emission. It is clear that any product made from CO2, after it is used, will release CO2. This makes the ‘storage’ potential of a number of chemicals made from CO2 is very low and not attractive, except for copolymers such as polycarbonates and polyurethanes which can store CO2 for decades. Therefore, the major contribution to CO2 emission control that the chemical use of CO2 can make comes from the development of innovative synthetic routes based on CO2 which may reduce the production of waste via more direct syntheses71 that use less energy and save carbon.

14.7.1 Synthesis of intermediates and fine chemicals The introduction of the carboxylic moiety ‘–COO’ into an organic substrate in a single step using a catalytic process is one of the most challenging reactions in synthetic chemistry. In principle, one can imagine inserting CO2 into a C–E bond, where E is another element such as H, C, O, N, etc. So far, only one process has been implemented at the industrial level, namely, the carboxylation of epoxides to yield molecular organic carbonates or, better, polymers (Fig. 14.4) that is, in effect, an insertion of CO2 into a C–O bond. Polycarbonates (and polyurethanes) have a market of a few Mt/y and are examples of long-living products derived from CO2. Such materials can thus be considered as examples of an economically viable chemical ‘storage’


Industrial utilization of carbon dioxide (CO2) NH2

O H2N C NH2

ONa/K

HO

H3COH

COONa/K O

O

O

O

O

O

O O

O

A

HCOOH RNH2

n

O

D H2

CO COOH COOH H2C = CH2

O

O

O2

B

O

HOOC

COO R

Br C e –, H +

O O

N H

O

N H

H N 3

OR¢ C

O O

COOH

HOOC

O

RNH O

COOH

COOH

R

RC ∫ CR

RNH2 + R¢X

HCONHR CnH2n+2 CnH2n

O

O RO C OR

ROH

399

O OH

n

14.4 Some uses of CO2 in synthetic chemistry. Reactions framed in A are on stream, those in B are under advanced investigation: both do not require an energy input as the whole CO2 moiety is incorporated into the final product through an exoergonic reaction. C and D reactions demand energy.

of CO2. Polycarbonates can be obtained by reacting the epoxide and CO2 in the presence of the appropriate catalyst. Al–porphyrin complexes, 72 the first to be discovered, have recently been used in processes that are on stream. Several other metal-systems (Zn, Cr, Mo, Ru, etc.) compounds in a liquid73 or supercritical phase74 have been intensively investigated with the aim of developing more active and selective systems that may afford a regular alternate insertion of the two co-monomers, avoiding the preferential polymerization of the epoxide. Among molecular carbonates, dimethylcarbonate (DMC), diethylcarbonate (DEC), diallylcarbonate (DAC) and diphenylcarbonate (DPC) are the most interesting compounds. DMC has a market of ca 0.1 Mt/y and is used in several different applications in the chemical, pharmaceutical and electronic industries. Indeed, its total production is one order of magnitude larger and


400


most of it does not reach the market as it is used for the preparation of polymers. A potential new use of DMC or DEC is as an additive to gasoline for which more than 30 Mt/y would be necessary. However, the actual synthetic methodology, which is based on phosgene, cannot be expanded to meet such a demand. This demand also cannot be met by the new processes based on the oxidative carbonylation of methanol (the ENIChem75 or UBE Process76) due to some drawbacks that limit the upscaling of the plants. The direct carboxylation of methanol (route d, Fig. 14.5) is now under investigation. Recent studies have shown that the reaction suffers thermodynamic limitations and requires the elimination of water to give acceptable conversion of the alcohol.77 As reported above, the availability of large volumes of recovered CO2 may sustain the development of new synthetic technologies based on it, assuming that efficient methodologies are developed that are somehow able to overcome the thermodynamic and kinetic issues related to the free energy of formation of CO2 (DG = –394 kJ/mol). Such a low free energy value could suggest, at a glance, that the conversion of CO2 may require a large energy input. In fact, the carboxylation reactions in which the entire –COO moiety is incorporated into a product (organic or inorganic) with an increase of the C/H ratio, are either exoergonic or are characterized by a very limited energy demand. In fact, CO2 reacts promptly at room or lower temperature with electron-rich species (such as olefins, dienes, amines, hydroxo groups, carbanions, etc.) to create chemicals that are currently produced through complex reaction pathways characterized by a high E-factor. The direct CO2 + 2NH3 e –H2O COCl2 + RR≤NH a

H2NC(O)NH2

RR≤NCOCl

+2ROH –2NH3

a¢ ROH CO2 + R¢R≤NH + RX

i –HX

O R¢R¢¢NC OR h ROH

g

f

b + R¢R≤NH – ROH +ROH –R¢R≤NH c

RO RO

C

O

CO2 d + –H2O 2ROH

R¢=H

R≤NCO

14.5 Synthesis of carbamates, carbonates and isocyanates based on CO2 and their inter-conversion through trans-esterification reactions.



401

synthetic methodology based on CO2 may produce a reduction of the emission, because the new route is characterized by higher selectivity and yield.78 It is true, on the other hand, that processes that utilize CO2 may present high kinetic barriers so that, although they are exoergonic, in order to have an appreciable reaction rate it will be necessary to work at high temperature. An example is given by the conversion of silicates into carbonates (natural weathering of silicates), an exoergonic reaction with a very slow kinetics at room temperature: an appreciable reaction rate is shown close to 1000 K. Figure 14.5 shows an interesting network of reactions based on CO2 aimed at substituting phosgene (COCl2), a toxic species with a world market higher than 8 Mt/y, now banned in several countries. The synthesis of carbamates (route i) is of great interest, as carbamates are intermediates for the production of isocyanates (monomers for polymers) and carbonates. The utilization of urea (route f) is also of great interest for the synthesis of carbonates and urethanes.79, 80 The reaction of CO2 with olefins (Fig. 14.4) is of interest as it may be considered as a direct route to acrylates that are monomers for large market polymers (> 2 Mt/y). Despite recent achievements,81 several issues must still be addressed in order to develop the reaction to an application level. The reaction with dienes has been for long investigated82 and the reaction mechanism at Pd-centres (dimerization of butadiene followed by the insertion of CO2 and the release of the carboxylated product) is well known. Also, such knowledge has prompted some research groups to develop very selective syntheses of six-membered lactones used as fragrances: specific ancillary ligands have been identified that may control the entropy of product distribution.83 Such coupling of dienes with CO2 is interesting also because, by changing the metal centre (from Pd to Rh), the diene can be trimerized before CO2 insertion, addressing, thus, the production of long-chain carboxylic acids used as biodegradable emulsifiers or detergents (market of the order of 10 Mt/y) (Fig. 14.6). All the above reactions occur under mild conditions and are thus exoergonic. Remaining with the issue of producing organic acids from CO2, a process of great interest is the synthesis of formic acid (Fig. 14.4) via the direct interaction of CO2 and dihydrogen.84 Formic acid finds substantial use in the chemical industry and is of interest as hydrogen carrier as it can be easily decomposed back to H2 and CO2 by contact with a metal.84b Such direct synthesis is more eco-friendly than existing technologies.71 The use of CO2 as mild oxidant or dehydrogenating agent85 can be of great interest as it would produce hydrogen and useful olefins by converting long-chain hydrocarbons with a limited use. The direct carboxylation of saturated or unsaturated organic substrates is also a process of interest. Either radiations86 or active catalysts87 in ionic liquids are used. The direct functionalization of hydrocarbons requires the activation of C–H bonds: it


402

Developments and innovation in CCS technology +2

Lx—Pd CO2

+3

Rh—Lx M = Rh

M=Pd (Ru, Ni)

C

Lx—Pd

Rh—Lx

O

O

O

O

O

O

O

O [MLx]

O C

O

CO2

O

O

O

O

C

O

O

O

A. Behr, Chem. Ber., 1984, 272, 29

O

Rh—Lx M. Aresta, New J. Chem, 1994, 18, 133

14.6 Coupling of butadiene with CO2 under the catalytic action of different metal systems. The correct choice of the ancillary ligands may control the entropy of product distribution.

is foreseeable that as knowledge in the area of C–H activation advances, so the synthesis of carboxylic acids based on the use of CO2 will progress, ultimately reaching the level of application.

14.7.2 Synthesis of energy products In principle, CO2 can be used as source of carbon for the synthesis of compounds such as alcohols or hydrocarbons (see Fig. 14.4) which are energy products and have a market much larger than chemicals. In fact, 7–10 % of the total extracted fossil carbon is converted into chemicals, the rest being used for the generation of various forms of energy (electric, thermal, mechanical). In the energy compounds, the C/H ratio decreases upon CO2 incorporation, meaning a parallel incorporation of hydrogen. This is the key issue: the production of energy products from CO2 demands hydrogen. Two questions follow. First, why convert CO2 and not simply use hydrogen; second, where or from what should the hydrogen be produced. The answer to the first question has been given in the previous section: hydrocarbons are a much better energy vector than hydrogen: they have the correct energy density and are easy to stock and transport. The answer to the second question is: the source of hydrogen cannot be fossil carbon (neither hydrocarbons nor coal), because CO2 would be generated in its production, defeating the object of the synthetic process. An external source of hydrogen should be used (water, biomass, residual



403

hydrocarbons, currently not used) that decouples the CO2 recycling in the form of a fuel from fossil carbon extraction. Recently, methanol has attracted much attention (from both industry and the scientific world) for its potential use as fuel, intermediate or hydrogen vector (use in fuel-cells). Catalysts are available characterized by 100 % selectivity and high turnover frequency (TOF). A demonstration plant (50 kg/day) using a Cu–ZnO-based catalyst with 99.9 % selectivity during 8000 h operation at 523 K and 5 MPa was built in the late 1990s in Japan.88 The process was further improved by adding silica and Pd to the catalyst.89 Such technology represents an interesting improvement with respect to Syngas conversion.90 A real breakthrough would be represented by coupling the process with the production of hydrogen from non-fossil sources. This better performance of CO2 with respect to CO, despite the higher consumption of hydrogen (Equations 14.9 and 14.10), is due to a different reaction mechanism91 that implies the direct conversion of CO2 into methanol92 without involving the reverse water gas shift reaction and the preliminary conversion of CO2 into CO. It is interesting to note that in the industrial synthesis of methanol from Syngas, CO2 is used (up to 30 % as C), and this addition reduces both the energy consumption (thermal yield 66.5 compared to 64.3 % for Syngas alone) and the CO2 emission. This reduction is due to the fact that CO2 promotes a better utilization of H293, 94 and methane with improvement of both the overall energetic and product yield (50 compared to 42.3 %).95

CO + 2H2 Æ CH3OH

CO2 + 3H2 Æ CH3OH + H2O

[14.9] [14.10]

The dry reforming of methane (or other hydrocarbons) (Equation 14.11) produces Syngas (H2–CO) used for the synthesis of methanol or gasoline. Such technology might be of practical interest as it would enforce the gasto-fuel (GTF) conversion at the well extraction site, avoiding the methane separation from CO2 and LNG distribution, reducing losses. The CO2 dry reforming (Equation 14.11) is often coupled to the steam reforming (Equation 14.12), and to the partial oxidation of methane (Equation 14.13).

CH4 + CO2 Æ 2CO + 2H2

DH0298 = + 247 kJmol–1

[14.11]

CH4 + H2O Æ CO + 3H2

DH0298 = + 206 kJmol–1

[14.12]

CH4 + 1/2O2 Æ CO + 2H2 DH0298 = – 36 kJmol–1

[14.13]

The combination of these three reactions is known as ‘tri-reforming’.94 It produces a CO/H2 ratio equal to 1.7 that is good for methanol or higher hydrocarbons synthesis.96 A key issue here is coke formation when temperatures below 1000 K are used.97 New catalysts are under development that combine


404


a high efficiency with coke-formation inhibition. The catalyzed cold plasma approach to methane dry reforming98 appears to be a promising and energyefficient technology used also for the direct synthesis of oxyfuels99 and, in fact, a pilot plant for GTF conversion under plasma conditions is being operated in Alberta (Canada).100

14.8


The utilization of CO2 is a promising technology that may contribute to reducing the accumulation of CO2 in the atmosphere. Several ways are open that may have different potentials for CO2 mitigation. Innovations in the field of synthetic industrial chemistry may bring about the discovery of cleaner production processes based on CO2 that may reduce the overall emission of CO2 with respect to processes on stream because of more selective and less energy- and carbon-intensive methodologies. These approaches need more research on catalysts and process development, while using new reaction media. In this area, the use of CO2 as reagent and solvent may provide a significant innovation with the development of new processes with low emission trends. The use of CO2 as a technological fluid is also very attractive. Moving from the study phase to the implementation of CO2 as a fluid in air conditioners and refrigerators and extending its use in dry cleaning and as an extraction fluid may enable a reduction in the current use of a number of chemical products with high CCP thus mitigating the impact on climate change. Searching for new technological applications of dense CO2 may be very useful, assuming that the new application based on CO2 will reducte the use of fluids with a much higher CCP. Enhanced fixation in aquatic biomass is a very interesting application that may result in the production of quasi-zero-emission biofuels that may supplant fossil fuels, especially in the transportation sector, a process very much to be wished for. In this area, algal strains with a good productivity are known; their culture–collection–treatment–extraction processes must be improved so that they can provide large volumes of biofuels. The application of the biorefinery concept can be a winning strategy for reducing the overall CO2 emission with concomitant economic benefit. The thermal or electrochemical conversion of CO2 into fuels is of great interest, especially when residual energies or intermittent perennial energies can be applied to this end. This area has a large potential not yet explored. New electrocatalysts are needed and more sophisticated technologies for the production of efficient semiconductors that may use solar light for fostering the CO2 reduction in water, developing a kind of artificial photosynthesis. In sum, the use of CO2 can help to reduce the impact on climate change either directly (reduced emmission into the atmosphere) or indirectly (less



405

emission of chemicals having a CCP much higher than CO2). More investment in research is needed in order to discover new applications and new technologies that may increase the actual 140 Mt/year utilization to 300–400 Mt/year avoided CO2 that is the estimated potential of the utilization technology in the medium term. Research is also necessary for the correct assessment of the potential of utilization and of the reduction of CCP. LCA studies must be developed that allow the benefits to be certified as, at the end of the day, the use of CO2 is not per se a guarantee of environmental, energetic and economic convenience: this needs to be scientifically demonstrated!

14.9


Several books and reviews have been published on the topic discussed in this chapter. Most significant publications are cited in the text above and listed in the references section. Readers may also refer to IPCC Reports for a more general approach to the CO2 problem. However, several other aspects of CO2 capture and storage are discussed in other chapters of this book. We also wish to recall that the fixation into biomass is a topic discussed in Chapter 15.

14.10 References 1. Solvay E, Belgian Patent 1861. 2. (a) Kolbe H and Lautermann E, Ann., 1869, 113, 125; (b) Schmitt R and Burkard E, Ber, 1877, 20, 2699. 3. (a) Bassarov 1870; (b) Morino S, Sakai M, Kashiki I, Suzuki A and Miki M, Bull. Fac. Fish. Hokkaido Univ., 1978, 29(1), 75–88. 4. De Pasquale RJ, JCS Chem. Commun., 1973, 157–158. 5. Aresta M, Nobile CF, Albano VG, Forni E and Manassero M, JCS Chem. Commun., 1975, 636–637. 6. Aresta M, Quaranta E, Giannocaro P, Tommasi I and Ciccarese A, Gazz. Chim. Ital., 1995, 125, 509–538. 7. Ibers JA, in Carbon Dioxide as a Source of Carbon, M. Aresta (ed.), Reidel, Dordrecht, the Netherlands, 1986, 55–70. 8. Gibson DH, Chem. Rev., 1996, 96, 2063–2095. 9. Aresta M and Dibenedetto A, in Activation of small molecules, W.B. Tolman (ed.), Wiley-VCH Weinheim, Germany, 2006, 1–41. 10. Silvestri G, in Carbon Dioxide as a Source of Carbon, M. Aresta (ed.), Reidel Dordrecht, the Nethertlands, 1986, 339–370 11. (a) Derien S, Clinet J-C, Dunach E and Perichon J, JCS Chem. Commun., 1991, 549–550; (b) Fauvarque J F, De Zelicourt Y, Amatore C and Jutand A, J. Appl. Electrochem., 2005, 20, 338–340. 12. Suresh B, Ishikawa Y and Schlag S, Carbon Dioxide, 2007, SRI Consulting, available at: http://www.sriconsulting.com/CEH/Public/Reports/743.2000/ (accessed February 2010).


406


13. Vansant J, in Carbon Dioxide Recovery and Utilization, M. Aresta (ed.), Kluwer Academic, Dordrecht, the Netherlands, 2003, 3–50. 14. Rowland F S, Tyler S C, Montague D C and Makide Y, Geophys. Res. Lett., 1982, 9(4), 481–484. 15. (a) De Simone J, Proceedings of ICCDU VI, 9–14 September, Breckenridge, 2001, 33; (b) Aresta M, (ed.), Carbon dioxide recovery and utilization, Kluwer Academic, Dordrecht, the Netherlands, 2003, Section III: ‘Supercritical carbon dioxide’, 121–207. 16. EPA-USA, 1995: Office of Compliance Sector Notebook Project. Profile of the Dry Cleaning Industry, EPA310-R-95-001. 17. EPA-USA 744-B-98-001, 1998: Cleaner technologies substitutes assessment for Professional Fabricare Project. 18. Sheldon RA, Green Chem., 2007, 9, 1273. 19. UNEP, Mineral Fertilizer Production and the Environment, A Guide to Reducing the Environmental Impact of Fertilizer Production, Technical report N° 26, United Nations Environment Programme Industry and the Environment, 1996. 20. Fossil fuels central to EU’s long-term energy security vision, 14 November 2008, available at: http://www.euractiv.com/en/energy/fossil-fuels-central-eu-long-termenergy-security-vision/article-177134 (accessed February 2010). 21. Aresta M, in Hand-Book of Combustion, M Lackner, F Winter and M Agarwal (eds), Wiley-VCH, Weinheim, Germany 2009, Chapter 13. 22. Aresta M, Dibenedetto A and Barberio G, Fuel Process. Technol., 2005, 86, 1679–1693. 23. Russel PG, Kovac N, Srinivasan S and Steinberg M, J. Electrochem. Soc., 1977, 124, 1329–1328. 24. Ito K, Ikeda Sh, Yamauchi N, Iida T and Takagi T, Bull. Chem. Soc. Japan, 1985, 58, 3027–3028. 25. Halmann MM, Chemical fixation of carbon dioxide, CRC Press, Boca Raton, FL, 1993, Chapter 7. 26. Gennaro A, Isse AA, Severin M-G, Vianello E, Bhugun I and Savéant J-M, J. Chem. Soc., Faraday Trans., 1996, 92, 3963–3968. 27. Frese Jr KW, Leach SC and Summens DP, United States Patent 4609441, 1986 (methanol production). 28. Sammels AF, United States Patent 4608132, 1986 (several products). 29. Williams R, Phys. Teach., 1979, 17, 246–249. 30. (a) Jessop PJ and Leitner W (eds), Chemical Synthesis using supercritical fluids, Wiley-VCH, Weinheim, Germany, 1999, 351; (b) Agnes A, Opre Z, Laurenczy G and Joo F, J. Mol. Cat. A, 2003, 204–205, 143–148, (c) Urakawa A, Jutz F, Laurenczy G and Baiker A, Chem. Eur. J., 2007, 13(14), 3886–3889. 31. (a) Amatore C and Saveant JM, J. Amer. Chem. Soc., 1981, 103, 5021–5024, (b) Eggins BR, Ennis C, McConnell R and Spence M, J. Appl. Electrochem., 1997, 27(6), 706–712. 32. Kaneco S, Hiei N-h, Xing Y, Katsumata H, Ohnishi H, Suzuki T and Ohta K, Electrochim. Acta, 2002, 48, 51–55. 33. Kaneco S, Iiba K, Ohta K, Mizuno T and Saji A, Electrochim. Acta, 1998, 44, 573–578. 34. Sanchez-Sanchez C, Montiel V, Tryk D, Aldaz A and Fujishima A, Pure Appl. Chem., 2001, 73, 1917–1927. 35. Jitaru M, Lowy DA, Toma M, Toma BC, Oniciu L, J. Appl. Electrochem., 1997, 27, 875–889. © Woodhead Publishing Limited, 2010


407

36. Augustinski J, Sartoretti CJ and Kedzierzawski P, in Carbon Dioxide Recovery and Utilization, M. Aresta (ed.), Kluwer Academic, Dordrecht, the Netherlands, 2003, 280–292. 37. Aurian-Blajeni B, in Electrochemistry in Transition, Murphy OJ, Srivnivasan S and Conway BE (eds), Plenum Press, NY 1992, 381–396. 38. Ito K, Ikeda Sh, Noda H, in Solar World Congress: Proc. Bienn. Congr. Int. Solar Energy Society, Pergamon, Oxford UK, 1992, 884–889. 39. Gattrell M and Gupta N, J. Electroanal. Chem., 2006, 594, 1–19. 40. Dubé P and Brisard JM, J. Electroanal. Chem., 2005, 582, 230–240. 41. Isaacs M, Canales JC, Riquelme A, Lucero M, Aguirre MJ and Costamagna J, J. Coord. Chem., 2003, 56, 1193–1201. 42. Koeleli F, Roepke T, Hamann CH, Electrochim. Acta, 2003, 48, 1595–1601. 43. Schrebler R, Cury P, Herrera F, Gomez H and Cordova R, J. Electroanal. Chem., 2001, 516, 23–30. 44. (a) Li H and Oloman C, J. Appl. Electrochem., 2005, 35, 955–965; (b) Akakori Y, Iwanaga N, Kato Y, Hamamoto O and Isahii M, Electrochemistry, 2004, 72, 266–270; (c) Hara K and Sakata T, J. Electrochem. Soc., 1997, 144, 539–545; (d) Oloman C and Li H, 2007, WO/2007/041872; (e) Hara K, Sonoyama N and Sakata T, Stud. Surf. Sci. Catal., 1998, 114, 577–580; (f) Kaneco S, Iwao R, Iiba K, Ohta K and Mizuno T, Energy, 1998, 23, 1107–1112; (g) Ikeda S, Shiozaki S, Susuki J, Ito K and Noda H, Stud. Surf. Sci. Catal., 1998, 114, 225–230. 45. (a) Lehmann T, Schneider R, Weckbecker C, Dunach E and Olivero S, WO/2002/016671; (b) Rios-Ascudero A, Isaacs M, Villagran M, Zagal J and Costamagna J, J. Argenti. Chem. Soc., 2004, 92, 63–71. 46. Riquelme MA, Isaacs M, Lucero M, Trollund E and Aguirre MJ, J. Chil. Chem. Soc., 2003, 48, 89–92. 47. Kaneco S, Ueno Y, Katsumata H, Suzuki T and Ohta K, Chem. Eng. J., 2006, 119, 107–112. 48. Hirota K, Tryk DA, Hashimoto K, Okawa M, Fujishima A, Stud. Surf. Sci. Catal., 1998, 114, 589–592. 49. Aydin R, Koeleli F, Synth. Met., 2004, 144, 75–80. 50. Schrebler R, Cury P, Suarez C, Munoz E, Gomez H, Cordova R, J. Electroanal. Chem., 2002, 533, 167–175. 51. Spataru N, Tokuhiro K, Terashima C, Rao TN, Fujishima A, J. Appl. Electrochem. 2003, 33, 1205–1210. 52. Lee J and Tak Y, Electrochim. Acta, 2001, 46, 3015–3022. 53. Kaneco S, Iiba K, Hiei N-h, Ohta K, Mizuno T and Suzuki T, Electrochim. Acta, 1999, 44, 4701–4706. 54. (a) Ikeda S, Takagi T and Ito K, Bull. Chem. Soc. Japan, 1987, 60, 2517–2522. (b) Aydin R and Koeleli F, J. Electroanal. Chem., 2002, 535, 107–112. 55. Hori Y, Konishi H, Futamura T, Murata A, Koga O, Sakurai H, Oguma K, Electrochim. Acta, 2005, 50, 5354–5369. 56. Magdesieva TV, Zhukov IV, Kravchuk DN, Semenikhin OA, Tomilova LG and Butin KP, Russ. Chem. Bull., 2002, 51, 805–812. 57. Ohmori T, Nakayama A, Mametsuka H and Suzuki E, J. Electroanal. Chem., 2001, 514, 51–55. 58. Kuniko T, Fudeko T, Masahiro K, Yoshio A and Makoto A, Bull. Fac. Human Environ. Sci., 2005, 36, 13–21. 59. Centi G, Perathoner S, Winè G and Gangeri M, Green Chem., 2007, 9, 671–678.


408


60. Eggins BR, Robertson PKJ, Murphy EP, Woods E and Irvine JTS, J. Photochem. Photobiol., 1998, 118, 31–40. 61. Ziessel R, in Carbon Dioxide as a Source of Carbon, Aresta M and Forti G (eds), Reidel, Dordrecht, the Netherlands 1986, 113–138. 62. Dey GR, Belapurkar AD and Kishore K, J. Photochem. Photobiol. A, 2004, 163, 503–508. 63. Zhao Z-H, Fan J-M and Wang Z-Z, J. Cleaner Prod., 2007, 15, 1894–1897. 64. Hirose T, Maeno Y and Himeda Y, J. Mol. Cat. A: Chem., 2003, 193, 27–32. 65. Tsubaki H, Sugawara A, Takeda H, Gholamkhass B, Koike K and Ishitani O, Res. Chem. Internmed., 2007, 33, 37–48. 66. Kurz P, Probst B, Spingler B and Alberto R, Eur. J. Inorg. Chem., 2006, 15, 2966–2974. 67. Takeda H, Koike K, Inoue H and Ishitani O, J. Am. Chem. Soc., 2008, 130, 2023– 2031. 68. Sato S, Koike K, Inoue H, Ishitani O, Photochem. Photobiol. Sci., 2007, 6, 454– 461. 69. Bian Z-Y, Sumi K, Furue M, Sato S, Koike K and Ishitani O, Inorg. Chem., 2008, 47, 10801–10803. 70. Maeda K, Teramura K, Lu D, Takata T, Saito N, Inoue Y and Domen K, Nature, 2006, 440, 295–297. 71. Aresta M and Dibenedetto A in Carbon dioxide recovery and utilization, M. Aresta (ed.), Kluwer Academic, Dordrecht, the Netherlands, 2003, 211–260. 72. Inoue S, in Carbon Dioxide as a Source of Carbon, Aresta M and Forti G (eds), Reidel, Dordrecht, the Netherlands 1986, 321–338. 73. Darensbourg DJ and Holtcamp MW, Coord. Chem. Rev., 1996, 153, 155–174. 74. Early TR, Holmes AB, Lee JK, Quaranta E and Stamp LM, in Carbon dioxide Recovery and Utilization, M. Aresta (ed.), Kluwer Academic, Dordrecht, the Netherlands, 2003, 149–168. 75. (a) Romano U, Tesei R, Massi Mauri M and Rebora P, Ind. Eng. Chem. Prod. Res. Dev., 1980, 19, 396–403; (b) Romano U, Rivetti S and Di Muzio N, US Patent 4 318 862, 1982, ENIChem. 76. Matsuzaki T, Shimamura T, Fujitsu S and Toriyahara Y, US Patent 5 292 916, 1994, Ube Ltd. 77. (a) Aresta M and Dibenedetto A, J. Mol. Catal. A: Chem 2002, 182–183, 399–409; (b) Aresta M, Dibenedetto A, Dileo C, Tommasi I and Amodio E, J. Supercrit Fluids, 2003, 25/2, 177–182; (c) Aresta M, Dibenedetto A and Pastore C, Inorg. Chem., 2003, 42(10), 3256–3261; (d) M. Aresta, A. Dibenedetto, L. Gianfrate and C. Pastore, J. Mol. Catal. A: Chem., 2003, 204–205, 245–252; (e) Aresta M, Dibenedetto A, Gianfrate L and Pastore C, Appl. Catal. A, 2003, 255(1), 5–11; (f) Aresta M, Dibenedetto A and Pastore C, Stud. Surf. Sci. Catal., 2004, 153, 221–226; (g) Aresta M, Dibenedetto A, Devita C, Bourova OA and Chupakhin ON, Stud. Surf. Sci. Catal., 2004, 153, 213–220; (h) Aresta M, Dibenedetto A, Fracchiolla E, Giannoccaro P, Pastore C, Pápai I and Schubert G, J. Org. Chem., 2005, 70, 6177–6186; (i) Aresta M, Dibenedetto A and Pastore C, Catal. Today, 2006, 115, 88–94; (j) Aresta M, Dibenedetto A, Nocito F, Pastore C, Venezia AM, Chirykalova E, Kononenko VI, Shevchenko VG and Chupova IA, Catal. Today, 2006, 115, 117–123; (k) Aresta M, Dibenedetto A, Pastore C, Pápai I and Schubert G, Top. Catal., 2006, 40(1–4), 71–81; (l) Aresta M, Dibenedetto A, Nocito F and Pastore C, J. Mol. Catal, 2006, 257(1–2), 149–153; (m) Dibenedetto A, Aresta M, Pastore C and Nocito F, Prepri.



409

– Am. Chem. Soc., Div. Pet. Chem., 2007, 52(2), 273; (n) Aresta M, Dibenedetto A, Nocito F and Pastore C, Inorg. Chim. Acta, 2008, 361, 3215–3220; (o) Dibenedetto A, Aresta M, Distaso M, Pastore C, Venezia AM, Liu C-j and Zhang M, Catal. Today, 2008, 137, 44–51; (p) Aresta M, Dibenedetto A and Pastore C, Prepr. Pap. – Am. Chem. Soc. Div. Fuel Chem., 2008, 53(1), 244–245; (q) Aresta M, Dibenedetto A, Pastore C and Aresta B, Prepr. Pap. – Am. Chem. Soc., Div Fuel Chem, 2008, 53(1), 322–323; (r) Aresta M, Dibenedetto A, Pastore C, Cuocci C, Aresta B, Cometa S and Degiglio E, Catal. Today, 2008, 137, 125–131. 78. Only very few LCA studies are available in the literature and they show that the use of CO2 decreases the environmental impact of the process with respect to the existing processes based on phosgene. See: Aresta M, Dibenedetto A and Barberio G, Fuel Process. Technol., 2005, 86, 1679–1693. 79. Li Q, Zhang W, Zhao N, Wei W and Sun Y, in Proceedings of the Eighth International Conference on Greenhouse Gas Control Technologies: GHGT8, Gale J, Rokke N, Zweigel P and Svenson H (eds), Elsevier, Oxford, UK, 2005, CD-ROM. 80. Aresta M, Dubois J–L, Dibenedetto A, Nocito F and Ferragina C, EU Patent 08305653.1–2117. 81. (a) Musco A, J. Chem. Soc., Perkin Trans., 1980, 1, 693–698; (b) A. Behr, Ber., 1984, 272, 29–31. 82. Braunstein P, Matt D and Nobel D, J. Amer. Chem. Soc., 1988, 110(10), 3207– 3212. 83. (a) Behr A, Herdtweck E, Hermann WA, Keim W, Kipshagen W, JCS Chem. Commun., 1986, 16, 1262–1263; (b) Behr A and Becker M, Dalton Trans., 2006, 4607–4612; (c) Gassner F, Haack V, Janssen A, Elsagir A and Dinjus E, 2000, European Patent EP1036791. 84. (a) Jessop PJ and Leitner W, (eds), Chemical Synthesis Using Supercritical Fluids, Wiley-WCH, Weinheim, 1999, 351; (b) Urakawa A, Jutz F, Laurenczy G and Baiker A, Chem. Eur. J., 2007, 13, 3886–3899; (c) Fellay C, Jang N, Dyson PJ and Laurenczy G, Chem. Eur. J., 2009, 15, 3752–3760. 85. Park S-E and Yoo JS, Stud Surf. Sci. Catal., 2004, 153, 304–314. 86. (a) Fujita N, Morita H, Matsuura C and Hiroishi D, Radiat. Phys. Chem., 1994, 44(4), 349–357; (b) Fujita N, Fukuda Y, Matsuura C and Saigo K, Radiat. Phys. Chem., 1996, 47(4), 543–549. 87. (a) Tommasi I, Aresta M and Tkatchenko I, 224th ACS Meeting, Preprints of the Workshop “Ionic liquids as green solvents: Progress and Prospects, 2002, 18–22; (b) Tommasi I and Sorrentino F, Tetrahedron. Lett., 2009, 50, 104–105. 88. Ushikoshi K, Mori K, Watanabe T, Takeuchi M and Saito M, in Advances in Chemical Conversions for Mitigating Carbon Dioxide, T. Inui, M. Anpo, K. Izui, S. Yanagida and T. Yamaguchi (eds)1998, Elsevier Science B.V., Amsterdam, the Netherlands, 357–362. 89. Inui T, Catal. Today, 1996, 29(1–4), 329–337. 90. Saito M, Fujitani T, Takeuchi M and Watanabe T, Appl. Catal. A, 1996, 138(2), 311–318. 91. Kieffer R, Fujiwara M, Udron L and Souma Y, Catal. Today, 1997, 36(1), 15–24. 92. Rozovskii A, Russ. Chem. Rev., 1989, 58(1), 41–56. 93. Maroto–Valer MM, Song C and Soong Y (eds), Environmental Challenges and Greenhouse Gas Control for Fossil Fuel Utilization in the 21st Century, Kluwer Academic/Plenum, New York, 2002. 94. (a) Song C, in CO2 Conversion and Utilization, Song C, Gaffney AM, Fujimoto K


410


(eds), ACS Symp Series 809, Washington, DC, 2002, 2–30; (b) Song C, Gaffney AM and Fujimoto K (eds), CO2 Conversion and Utilization, ACS Symp Series 809, Washington, DC, 2002. 95. Lange JP, Ind. Eng. Chem. Res., 1997, 36, 4282–4290. 96. (a) Fox JM III, Catal. Rev. Sci. Eng., 1993, 35(2), 169–212; (b) Rostrup-Nielsen JR, in Curry-Hyde HE and Howe RF (eds), Elsevier, Amsterdam, the Netherlands, 1994, 25. 97. Ross JRH, van Keulen ANJ, Hegarty MES and Seshan K, Catal. Today, 1996, 30(1–3), 193–199. 98. Eliasson B and Kogelschatz U, IEEE Trans. Plasma Sci., 1991, 19(6), 1063–77. 99. Zhang Y, Li Y, Wang Yu, Liu C and Eliasson B, Fuel Process. Technol., 2003, 83(1), 101–109. 100. Czernichowski A, Czernichowski M, Czernichowski P and Cooley TE, Prepr. Pap. Am. Chem. Soc., Div. Fuel Chem., 2002, 47(1), 280–281.


15

Biofixation of carbon dioxide (CO2) by microorganisms B. W a n g and C. Q. L a n, University of Ottawa, Canada Abstract: Microbial biofixation of carbon dioxide (CO2) employs the capacity of autotrophic microorganisms, including photoautotrophs and chemoautotrophs, to fix CO2 for cell growth to reduce CO2 emission. In general, there are two approaches towards microbial CO2 sequestration: enhancing the biological productivity of autotrophs in their natural habitats (e.g., ocean fertilization) and cultivating autotrophic microorganisms in controlled systems (e.g., microalgal farming). This approach is effective, environment friendly and potentially sustainable. It is predicted that rapid technology development will make microbial CO2 bio-sequestration a practical approach in the near future. Key words: autotrophic microorganism, microalgae, ocean fertilization, microalgal farming, CO2 sequestration, photobioreactor, biofuel, bio-energy

15.1

Introduction

The total CO2 emissions from fossil fuels were 6.79 GtC yr–1 in 2000 and was projected to increase to 8.35 GtC yr–1 in 2010 and 9.97 GtC yr–1 in 2020 (Huntley and Redalje, 2007). Accelerating emission of CO2, the primary greenhouse gas (GHG), will likely lead to dramatic changes in the earth’s climate system (Shi and Shen, 2003; Wang et al., 2008). Development of cost-effective and sustainable CO2 fixation strategies has therefore become a focus of extensive research and the principal goal of international environmental policy. Among the many CO2 mitigation technologies that have been developed in the last decades, biological CO2 fixation has attracted much interest as a promising alternative strategy. It was estimated (Huntley and Redalje, 2007) that biological mitigation options could offset 10–20 % of projected fossil fuel emissions by 2050. The biospecies that contribute to CO2 fixation are those primary producers including land plants and the marine phytoplankton. The marine phytoplankton, which comprises mostly unicellular microalgae, offers approximately half of the current annual biological CO2 fixation (approximately 50 billion tonnes of carbon per annum). Ocean fertilization is expected to significantly enhance the bioproductivity of the oceans and hence CO2 bio-sequestrating by the oceans. Microalgal farming is also expected to play an increasingly important role in CO2 biomitigation. This chapter strives to provide an overview on the potentials of different autotrophic microorganisms, including both 411 © Woodhead Publishing Limited, 2010

412


photoautotrophic microorganisms (photoautotrophs) and chemoautotrophic microorganisms (chemoautotrophs) for CO2 fixation.

15.2

Basic principles and methods

CO2 biofixation using microorganisms involves the cultivation of autotrophic microbes, which capture CO2 from the atmosphere or other sources (e.g., flue gases) for cell growth and extracellular product synthesis, in different cultivation systems such as natural waters, open ponds, and photobioreactors. In this process, autotrophic microorganisms fix CO2 to synthesize cell materials and extracellular products. As a result, CO2 is fixed in the form of organic biomass, which can be converted to different chemicals and biofuels through biorefinery. The CO2 biofixation reaction can be represented by the following equation: nCO 2 + nH 2 O ¨ ææ Æ [CH 2 O]n (biomass) [15.1] Energy Despite the enormous diversity of autotrophic microorganisms, they share a few common CO2 fixation pathways, which include (i) Calvin cycle (Calvin–Benson–Bassham cycle or reductive pentose phosphate cycle); (ii) reductive citric acid cycle; (iii) reductive acetyl–CoA pathway; and (iv) 3-hydroxypropinate cycle.

15.2.1 Calvin cycle The Calvin cycle, which is also called the reductive pentose phosphate cycle, is the most widespread CO2 biofixation pathway among autotrophs. It exists in plants and microalgae, as well as photoautotrophic and chemoautotrophic bacteria. As shown in Fig. 15.1, the key step of the Calvin cycle is catalyzed by the enzyme ribulose bisphosphate carboxylase, which fixes a CO2 molecule onto a molecule of ribulose-1,5-diphosphate (RuBP), resulting in two molecules of glyceric acid-3-phosphate (3PG). These 3PG molecules are then converted into two glyceraldehyde-3-phosphate (G3P, aka phosphoglyceraldehyde, PGAL) molecules by adding a high-energy phosphate group from ATP to each molecule. The two 3PG molecules are then converted to a RuBP molecule, which stays in the cycle for another round of CO2, and an organic carbon unit [C]. Three rounds of Calvin cycle lead to fixation of 3 CO2 molecules and the production of a G3P molecule, which can be further converted to glucose, lipids and other cell materials.

15.2.2 Other carbon dioxide (CO2) biofixation pathways The Calvin cycle exists not only in eukaryotic photosynthetic organisms, including eukaryotic plants and algae, but also in prokaryotic autotrophs © Woodhead Publishing Limited, 2010


413

3 CO2

6 3-phosphoglycerate Rubisco

6 ATP 6 ADP

6 1,3=bisphosphoglycerate 3 ADP

6 NADPH

3 ATP

6 NADP+

5 glyceraldehyde 3-phosphate

6 glyceraldehyde 3-phosphate

Glyceraldehyde 3-phosphate

Glucose

15.1 The Calvin cycle: three molecules of CO2 fixed give a net yield of one molecule of glyceraldehyde 3-phosphate at the net cost of nine molecules of ATP and six molecules of NADPH.

including both photosynthetic and chemoautotrophic bacteria. Unlike in eukaryotes, however, the Calvin cycle is just one of several known autotrophic pathways in prokaryotes (Hulger et al., 2003). Three other CO2 fixation pathways have so far been discovered in autotrophic bacteria. Two of them are bidirectional pathways, which have in common the synthesis of acetyl–coenzyme A (acetyl–CoA). One of these two pathways is based on the reversed (reductive) citric acid cycle with 2-oxoglutarate:ferredoxin oxidoreductase and ATP citratelyase as the key enzymes. This cycle uses oxygen-sensitive enzymes that can operate under microaerobic conditions and are found in both strict anaerobes and microaerobic (microaerophilic) bacteria (e.g., Knallgas bacteria). The other bidirectional pathway is the reductive acetyl–CoA pathway, which is characterized by an enzyme complex referred to as acetyl–CoA synthase. The enzyme also catalyzes a partial reaction, the oxidation of carbon monoxide to CO2. this key enzyme, therefore, is also named CO dehydrogenase. The fourth known pathway is 3-hydroxypropionate cycle. In this unidirectional pathway, acetyl–CoA is reductively transformed via 3-hydroxypropionate to succinyl–CoA. The CO2 fixation reactions, i.e., the carboxylation reactions, are catalyzed by acetyl–CoA carboxylase and propionyl–CoA carboxylase


414


(Herter et al., 2001, 2002). Glyoxylate is believed to be the initial CO2 fixation product. Nevertheless, the pathway of glyoxylate assimilation into cell material is not completely understood at present.

15.3

Carbon dioxide (CO2) fixation microorganisms: chemoautotrophs and photoautotrophs

As shown in Table 15.1, microorganisms that can potentially be employed for CO2 fixation are autotrophic microbes, which utilize CO2 or inorganic carbonate as the sole carbon source for cell growth. According to the energy source of autotrophs, they can be classified into two groups: photoautotrophs and chemoautotrophs (Kwak et al., 2006). Photoautotrophs include higher plants, algae (macroalgae and microalgae) and photosynthetic bacteria (e.g., cyanobacteria, purple bacteria and green bacteria).

Table 15.1 Some autotrophic microorganisms and their energy sources

Microorganism Examples groups

Energy source

Photoautotrophic Photosynthetic microorganisms bacteria Microalgae

Purple sulfur bacteria Light (Chromatiaceae); Green sulfur bacteria (Chlorobiaceae); blue green algae (cyanobacteria) Diatoms (Bacillariophyta); green algae (Chlorophyta); Eustigmatophytes; Prymnesiophytes

Chemoautotrophic Sulfur oxidizing microorganisms bacteria Chemoautotrophic nitrifying bacteria hydrogen oxidizing bacteria iron bacteria

Thiobacillus sp. Thiothrix Thioalkalivibrio versutus Nitrosomonas Nitrobacter Acidianus ambivalens Ignicoccus islandicus Thiobacillus ferrooxidans Thiobacillus thiooxidans


Reduced sulfur (e.g. S2O32–, S, H2S) Reduced nitrogen (e.g. NH4+, N2, NO) Molecular hydrogen (i.e., H2) Fe2+


415

15.3.1 Chemoautotrophic bacteria Chemoautotrophic microorganisms are all prokaryotes. They obtain energy for cell growth via the oxidization of reduced inorganic chemicals such as molecular hydrogen (i.e., H2), reduced sulfur (e.g., S2–, S, S2O32−, S4O62−, SO2− 3 ) (Flores Iii et al., 2007; Sorokin et al., 2008), reduced nitrogen (e.g., NH4+, NO2–), and metal (e.g., Fe2+). The most commonly known chemoautotrophic bacteria are hydrogen oxidizing bacteria, sulfur oxidizing bacteria, nitrifying bacteria, and iron oxidizing bacteria (Haddadin et al., 1993; Shively et al., 1998). Examples of sulfur oxidizing bacteria are the colorless sulphur bacteria, which are distinct from the photosynthetic sulfur bacteria that contain chlorophyll for photosynthesis. Colorless sulfur bacteria can oxidize hydrogen sulfide (H2S) to obtain energy for cell growth. Water and sulphur are produced as products from this reaction. Similarly, iron bacteria oxidize reduced iron in compounds and use the energy gained from the oxidization of iron to fix CO2 and synthesize cell materials. Examples of iron bacteria include Thiobacillus ferrooxidans and Thiobacillus thiooxidans. They are most commonly encountered as the rusty colored slimy layer on the inside of toilet tanks. The nitrifying bacteria oxidize reduced nitrogen such as ammonia (NH3) and nitrite (NO2–) to nitrate (NO3−) to obtain reducing power and energy. Examples of chemoautotrophic nitrifying bacteria include Nitrosomonas and Nitrobacter. These nitrifying bacteria are important in the global nitrogen cycle. Chemoautotrophs are important in the cycle of sulfur, nitrogen and other elements in the biosphere. However, the energy they manage to extract from reduced inorganic chemicals is very small in comparison with that can be captured from sunshine by photosynthetic organisms. They thrive only in niches where extreme conditions (e.g., pH, temperature and/or pressure) protect them against competition from other microorganisms. Although they contribute to CO2 fixation, they are in general not considered as important CO2 biofixation species.

15.3.2 Photosynthetic microorganisms There are two groups of photosynthetic microorganisms: eukaryotic microalgae and prokaryotic photosynthetic bacteria. Prokaryotic photosynthetic organisms include blue green bacteria (cyanobacteria), purple bacteria (Rhodospirillineae) the green bacteria (Chlorobiineae) and the Prochlorophyta. Microalgae and cyanobacteria (green blue algae) are the two most important groups of photoautotrophs for CO2 biofixation. Although green blue algae are bacteria rather than algae, they are sometimes treated together with microalgae for convenience (Li et al., 2008b).


416


15.3.3 Microalgae There are more than 40 000 microalgal species identified to date, which can been classified into several main groups according to their pigment composition, biochemical constituents, ultrastructure and life-cycle. Table 15.2 shows the six groups of microalgae that are of primary importance to the biomitigation and biofuel production: diatoms (Class Bacillariophyceae), green algae (Class Chlorophyceae), golden-brown algae (Class Chrysophyceae), prymnesiophytes (Class Prymnesiophyceae) and the eustigmatophytes (Class Eustigmatophyceae), blue green algae or cyanobacteria (Class Cyanophyceae). Diatoms Diatoms are the most common and widely distributed groups of microalgae on earth. This group dominates the phytoplankton of the oceans and is also commonly found in fresh- and brackish waters. The cells of diatoms are golden-brown because of the presence of a high level of fucoxanthin, a photosynthetic accessory pigment. Several other xanthophylls are present at lower levels, as well as b-carotene, chlorophyll a and chlorophyll c. The main storage compounds of diatoms are triglycerides (TAGs) and chrysolaminarin, a b-1, 3-linked carbohydrate. Diatom cell wall contains substantial quantities of polymerized Si. This unique feature has important implications for media preparation and costs in a commercial production facility, because silicate is a relatively expensive chemical. On the other hand, deficiency of silicate Table 15.2 Some algal divisions and their characteristic storage products Kingdom

Division

Common name

Storage product

Prokaryota Cyanophyta Blue green algae

Cyanophycin granules (arginine and aspartic acid)

Eubacteria Prochlorophyta Chlorophyta Green algae

Starchlike Starch (amylase and amylopectin) (oil in some)

Eukaryota Charophyta Stoneworts Euglenophyta Englenoids Phaeophyta Brown algae Chrysophyta Golden and yellow green algae (including diatoms) Pyrrhophyta Dinoflagellates Cryptophyta Cryptomonads Rhodophyta Red algae

Starch Paramylon, oil Laminaran; mannitol Starch (oil in some)


Starch Floridean starch


417

can promote lipid (TAG) accumulation in diatoms. It can be employed to provide a controllable means to induce lipid synthesis in a two-stage production process. Green algae Green algae (chlorophytes) have chlorophyll a and chlorophyll b as photosynthetic pigments. They use starch as their primary carbon storage materials. However, nitrogen deficiency is known to promote storage lipid accumulation in some green algal species. Green algae are believed to be the evolutionary progenitors of higher plants and have received more attention than other groups of algae. Chlamydomonas reinhardtii and closely related species of this group have been studied extensively. Chlamydomonas reinhardtii was the first alga to be genetically transformed. Another genus of green algae that has been studied extensively is Chlorella. Several green algae, for instance, Neochloris oleoabundans, are known to be able to accumulate large quantities of lipids and efficient in CO2 fixation (Li et al., 2008a,b), making them attractive candidates for combined CO2 fixation and biofuel production. Golden-brown algae Golden-brown algae include the chrysophytes and the synurophytes. They are similar to diatoms with respect to pigment composition. Some chrysophytes have lightly silicified cell walls. They are found primarily in freshwater habitats, and lipids and chrysolaminarin are the major carbon storage form of this group. Prymnesiophytes Prymnesiophytes (haptophytes) are primarily marine organisms and account for a substantial proportion of the primary productivity of tropical oceans. They therefore play an important role in nature’s carbon cycle. Prymnesiophytes are often of a golden-brown color because of the presence of the yellowbrown accessory pigments, diadinoxanthin and fucoxanthin. Lipids and chrysolaminarin are the major carbon storage of this group of algae. Eustigmatophytes Eustigmatophytes represent an important component of the ‘picoplankton’, which is the fraction of plankton that is composed of cells between 0.2 and 2 mm. They could be autotrophic, heterotrophic or mixotrophic and are responsible for the most primary productivity in oligotrophic gyres. The


418


genus Nannochloropsis is one of the few marine species of this group and is commonly found in the oceans. The only chlorophyll pigment contained in the cells of Eustigmatophytes is chlorophyll a. However, several xanthophylls serve as accessory photosynthetic pigments. Blue green algae (Cyanobacteria) As mentioned previously, Cyanobacteria are a group of photosynthetic bacteria. They are treated here together with microalgae for convenience. Cyanobacteria are prokaryotes that contain no membrane enclosed nucleus or chloroplasts and have a different gene structure than all the other microalgae. There are approximately 2000 species of cyanobacteria, which occur in a diversity of different habitats. Some members of this group can assimilate atmospheric nitrogen and therefore eliminate the need to provide fixed nitrogen for cell growth. However, no member of this class produces significant quantities of storage lipids, which have the highest energetic value among the major cell components including lipids, polysaccharides, nucleic acids, proteins and hybrid biomacromolecules. Consequently, cyanobacteria are considered less attractive in biofuel production, especially in biodiesel production, although they could be the species of choice for CO2 mitigation and production of novel bioproducts.

15.4

Carbon dioxide (CO2) fixation by microalgae

Microalgae can fix CO2 from different sources, which can be categorized into (i) CO2 from the atmosphere; (ii) CO2 from industrial exhaust gases (e.g., flue gas and flaring gas); and (iii) fixed CO2 in the form of soluble inorganic carbonates (e.g., NaHCO3 and Na2CO3). CO2 mitigation by microalgae can be achieved either by microalgae in their natural habitats, such as oceans and lakes, or in microalgal farming facilities.

15.4.1 Ocean fertilization Oceans are the largest habitats of microalgae. In fact, the marine phytoplankton, which comprises primarily of unicellular marine microalgae, is responsible for half of the planetary annual carbon fixation (approximately 5.0 ¥ 1011 t carbon per annum). The oceans are in general rich in micronutrients. However, their biological productivity is limited by macronutrients such as iron. Ocean fertilization, which refers to the strategy of enriching ocean waters with limiting nutrients such as irons, is therefore proposed to enhance ocean biological production and CO2 sequestering (Zeebe and Archer, 2005; Aumont and Bopp, 2006; Glibert et al., 2008). It has been well established that ocean fertilization can increase the



419

primary production and hence CO2 fixation over a significant fraction of the oceans (Liss et al., 2005; Zeebe and Archer, 2005; Aumont and Bopp, 2006; Cullen and Boyd, 2008), and many advocates see ocean fertilization as modern society’s last hope to slow global warming. However, this strategy is not free of controversies. The two major concerns are (i) the lack of large-scale experimental data with respect to the efficiency of ocean fertilization (Aumont and Bopp, 2006) and (ii) the uncertainty with regard to the biochemical and ecological side effects of ocean fertilization (Fuhrman and Capone, 1991; Liss et al., 2005; Glibert et al., 2008). It is suggested that ocean fertilization should be regulated to ensure precautions are taken in this practice (Orbach, 2008). Inconsistency of current international laws in this field is also recognized as a challenge (Freestone and Rayfuse, 2008).

15.4.2 Microalgal farming Apart from CO2 fixation by microalgae in their natural habitats, such as oceans, lakes and other waters, microalgal farming, i.e., cultivation of microalgae in controlled environment, is the other important means by which microalgae could contribute to CO2 mitigation. Biomass produced in microalgal farming could be consumed as fertilizers, animal feed, health foods or/and converted to biofuels to have a complete carbon cycle. It can also be converted to agrichar as a form of permanent CO2 bio-sequestrating. Traditionally, microalgae are cultivated in closed systems or open ponds, which are aerated or exposed to air to allow microalgae to capture CO2 from the atmosphere for cell growth. Since the atmosphere contains only 0.03–0.06% CO2, it is expected that mass transfer limitation could slow down the cell growth of microalgae (Chelf et al., 1993). On the other hand, industrial exhaust gases such as flue gas contain up to 15 % CO2, providing a CO2-rich source for microalgal cultivation and a potentially more efficient route for CO2 biofixation. The third route is to fix CO2 by chemical reaction to produce carbonates (e.g., Na2CO3) and use the latter as the carbon source for microalgal cultivation. A number of microalgal species have been shown to be able to utilize carbonates such as Na2CO3 and NaHCO3 for cell growth (Ginzburg, 1993; Merrett et al., 1996; Emma Huertas et al., 2000). Some of these species typically have high extracellular carboanhydrase activities (Emma Huertas et al., 2000), which are responsible for the conversion of carbonate to free CO2 to facilitate CO2 assimilation. In addition, the direct uptake of bicarbonate by an active transport system has been found in several species (Colman and Rotatore, 1995; Merrett et al., 1996). Adoption of carbonate-utilizing strains for CO2 fixation could be advantageous in many aspects: (i) CO2 released at night from industrial facilities could be converted to carbonate salts and stored for conversion in daytime; (ii) since only a limited number


420


of microalgal species thrive in media containing high concentration of carbonate salts, species control (i.e., preventing wild-type microalgal species from contaminating the cultivation system) is relatively simple; (iii) most of these species has high pH optima (in the range 9.0–11.0), further simplifying species control (Ginzburg, 1993). Flue gases from power plant are responsible for more than 7 % of the total world CO2 emissions (Sakai et al., 1995). CO2 in flue gas is available at little or no cost. As estimated by the IPCC criteria, the CO2 concentration of flue gas is up to 15 % (Maeda et al., 1995). Therefore, it would be beneficial if microalgae are tolerant to elevated CO2 level should they be used for CO2 fixation from flue gases (Maeda et al., 1995). An early review on flue gas tolerance by microalgae indicated that high levels of CO2 were tolerated by many microalgal species and that moderate levels of SOx and NOx (up to 150 ppm) were also well-tolerated (Matsumoto et al., 1997). Chlorococcum littorale, a marine alga, showed exceptional tolerance to high CO2 concentration of up to 40 % (Murakami and Ikenouchi, 1997; Iwasaki et al., 1998). Microalgae Scenedesmus obliquus and Chlorella kessleri, separated from the waste treatment ponds of the Presidente Médici coal-fired thermoelectric power plant, also exhibited good tolerance to high CO2 contents (de Morais and Costa, 2007b). Chlorella kessleri showed maximum specific growth rates (mmax) of 0.267 d–1 and biomass productivity of approximately 0.087 g L–1 d–1 when cultivated with 6 % (v/v) and 12 % (v/v) CO2 and a maximum biomass productivity of 0.085 g L–1 d–1 was achieved at 6 % CO2. These two microalgae also grew well when the culture was supplemented with enriched air stream containing up to 18 % CO2, indicating their great potentials for CO2 fixation from CO2-rich streams. It was also reported (de Morais and Costa, 2007a) that Scenedesmus obliquus and Spirulina sp. showed good capacities to fix CO2 when they were cultivated at 30 °C in a temperature-controlled three-stage serial tubular photo-bioreactor. For Spirulina sp., the maximum specific growth rate and maximum productivity were 0.44 d–1 and 0.22 g L–1 d–1, with both 6 % (v/v) CO2 and 12 % (v/v) CO2, respectively, while the maximum cell concentration was 3.50 g dry cell L–1 with both CO2 concentrations. For S. obliquus, the corresponding maximum growth rate and maximum productivity were 0.22 d–1 and 0.14 g L–1 d–1, respectively. Murakami and Ikenouchi (1997) selected more than 10 strains of microalgae with high capability of fixing CO2 by extensive screening. Two green algal strains, Chlorella sp. UK001 and Chlorococcum littorale, showed high CO2 fixation rates exceeding 1 g CO2 L–1 d–1. Botryococcus braunii SI-30, which showed the ability of producing high content of hydrocarbons, was recommended as a promising candidate for combined CO2 mitigation and biofuel production. The tolerance of microalgae to relatively high temperature is very important



421

in reducing cooling costs of the feeding flue gases released from industrial facilities at high temperature. These thermotolerant strains may also simplify species control because the temperature optima of most microalgal species lie in the range of 20–30 °C. A few thermotolerant strains have been selected. For instance, several unicellular green algal strains, identified as species of Chlorella, were isolated from hot springs in Japan (Sakai et al., 1995). These strains grew at temperatures up to 42 °C and in air containing more than 40 % CO2. Their tolerance to both high temperature and high CO2 content makes them potentially the appropriate microbial cellar reactors for bio-CO 2 mitigation from flue gas. Table 15.3 summarizes a few microalgal strains that have been studied for CO2 biomitigation. Some of these strains can tolerate high temperature and high CO2 in the gas stream. CO2 fixed through photosynthesis is converted to different organic cell components including carbohydrates, lipids, proteins and nucleic acids (Spolaore et al., 2006). Although the cell carbon content varies with microalgal strains, media and cultivation conditions, it changes in a relatively small range, and the law of material conservation allows us to calculate CO2 fixation rate from biomass productivity at given cell carbon content. In Table 15.3, such calculations were conducted using a reported biomass molecular formula, CO0.48H1.83N0.11P0.01 (Chisti, 2007), when direct data on CO2 fixation rate were not available, based on the assumption that CO2 fixed in the form of extracellular products was negligible.

15.4.3 Combined carbon dioxide (CO2) fixation and biofuel production using microalgae One promising route of CO2 biomitigation using microalgal farming is to convert the biomass produced in the process of CO2 fixation into biofuels for energy production. An estimate made in 2003 indicates that the costs of biofuel production are in general about 2.3 times more compared to fossil fuels (Kondili and Kaldellis, 2007). The fast technology development and the soaring energy prices since then have improved, and will continue to improve, the situation rapidly. Biofuel production from microalgae is deemed to be the most promising biofuel production strategy from a long-term perspective (Li et al., 2008b). As shown in Fig. 15.2, there are several ways to convert microalgae biomass to biofuels, which can be classified into biochemical conversion, chemical reaction, direct combustion and thermochemical conversion (Demirbas, 2001; McKendry, 2002). More specifically, example processes belonging to biochemical conversion include anaerobic digestion for methane production and fermentation for ethanol production (Spolaore et al., 2006). An example chemical conversion process involves extraction of lipids accumulated in microalgae cells and conversion of the extracted lipid to biodiesel via a


422


Table 15.3 Some microalgal strains studied for CO2 biomitigation (Wang et al., 2008) Microalga CO2 % T °C

P PCO2 Reference g l–1 d–1 g l–1 d–1

Chlorococcum 40 30 N/A 1.0 (Iwasaki littorale et al., 1998, Murakami and Ikenouchi, 1997) Chlorella 18 30 0.087 0.1631 (de Morais and kessleri Costa, 2007b) Chlorella sp. 15 35 N/A >1 (Murakami and UK001 Ikenouchi, 1997) Chlorella 15 N/A 0.624 (Yun et al., 1997) vulgaris Chlorella air 25 0.040 0.0751 (Scragg vulgaris et al., 2002) Chlorella air 25 0.024 0.0451 (Scragg vulgaris et al., 2002) Chlorella sp. 40 42 N/A 1.0 (Sakai et al., 1995) Dunaliella 3 27 0.17 0.3131 (Kishimoto et al., 1994) Haematococcus 16–34 20 0.076 0.143 (Huntley and pluvialis Redalje, 2007) Scenedesmus Air – 0.009 0.016 (Gomez-Villa obliquus et al., 2005) Scenedesmus Air – 0.016 0.031 (Gomez-Villa obliquus et al., 2005) Botryococcus – 25–30 1.1 >1.0 (Murakami and braunii Ikenouchi, 1997) Scenedesmus 18 30 0.14 0.26 (de Morais and obliquus Costa, 2007a) Spirulina sp2. 12 30 0.22 0.4131 (de Morais and Costa, 2007a)

Note

Artificial wastewater Watanabe’s medium Low-N medium

High salinity, ß-carotene Commercial scale, outdoor Wastewater, outdoor, winter Wastewater, outdoor, summer Accumulating hydrocarbon

1

Calculated from the biomass productivity according to equation, CO2 Fixation Rate (Pco2) = 1.88 ¥ Biomass Productivity (P), which is derived from the typical molecular formula of microalgal biomass, CO0.48H1.83N0.11P0.01 (Chisti, 2007). 2 All species except Spirulina sp., which is a prokaryotic cyanobacteria (Cyanophyceae) species, are eukaryotic green algae (Chlorophyta) species (NCBI website). – Not specified or not controlled.

simple transesterification reaction (Belarbi et al., 2000; Chisti, 2007). Some example thermochemical conversion processes include pyrolysis (Chiaramonti et al., 2007), gasification (Hirano et al., 1998) and liquefaction (Minowa and Sawayama, 1999). Energy stored in microalgal biomass could also be utilized via direct combustion or co-firing. Thermochemical conversion (Demirbas, 2004) is one of the most practical biomass conversion strategies.


Biofixation of carbon dioxide (CO2) by microorganisms Biochemical conversion

Fermentation

Ethanol, acetone, butanol

Anaerobic digestion

Microalgal biomass

423

Methane, hydrogen

Gasification

Fuel gas

Pyrolysis

Bio-oil, charcoal

Liquefaction

Bio-oil

Chemical reaction

Transesterification

Biodiesel

Direct combustion

Power generation

Electricity

Thermochemical conversion

15.2 Energy production via microalgal biomass conversion using biochemical, thermochemical, chemical and direct combustion processes (Wang et al., 2008).

15.4.4 Large-scale microalgal farming systems Studies have shown that well-designed cultivation systems may lead to significant increase of CO2 fixation efficiency (Javanmardian and Palsson, 1991; Kadam, 1997; Usui and Ikenouchi, 1997). The design of large-scale culture systems must consider many factors, including light intensity, temperature, biology of the algae, nature of the product, mixing, aeration, source of CO 2 and sterilization (Borowitzka, 1999), etc. Although microalgae are considered to be relatively efficient for capturing solar energy for the production of organic compounds via photosynthetic process, the photosynthetic efficiency of microalgae for the conversion of solar energy is typically below 20 % (Li et al., 2008b). On the other hand, increasing the density of cultures decreases photon availability to individual cells, which reduces specific growth rate of cells. Therefore, the poor penetration of light could be the most significant limiting factor in microalgal cultivation. There are two primary types of cultivation systems, open ponds with moderate surface to volume ratios (3–10 m–1) and photobioreactors with high surface to volume ratios (25–125 m–1) (Weissman et al., 1988). The most common design of open ponds (Vonshak and Richmond, 1988) are raceway ponds, in which algal cultures are mixed in a turbulent flow sustained by a paddle wheel. At least two types of open raceway ponds have been used commercially. The first is raceway ponds lined by concrete and the second is a shallow earthen tunnel lined with PVC or other durable plastic. The size of commercial ponds varies from 0.1 ha to 0.5 ha. While open ponds require low cost to cultivate algal biomass,


424


photobioreactors provide other advantages (Rosello Sastre et al., 2007) such as large surface/volume ratios, ability to prevent contamination, realizing the maximum capability to achieve high density of biomass and high biomass productivity and therefore high CO2 fixation rate (see Table 15.4). The culture conditions, including the light strength, irradiation and CO2 density, can be tightly controlled by photobioreactor design in order to use solar light effectively. There are several types of photobioreactors, including airlift reactor, internal luminous stirrer-type reactor, fountain-type reactor, plain plate-type reactor, liquid film-type reactor and algae immobilizing-type reactor (Michiki, 1995). The tubular photobioreactor, one of the most popular configurations of photobioreactors (Travieso et al., 2001), includes an array of straight transparent tubes, which are used to captured sunlight. Relatively small tube diameter, generally 0.1 m or less, is necessary for ensuring high biomass productivity. In a typical arrangement, the solar tubes are placed parallel to each other horizontally. A variety of other photobioreactors has also been studied. For instance, a flat-plate photobioreactor (Hu et al., 1996) has been tested for cultivation of high-CO2 tolerant unicellular green alga Chlorococcum littorale. Bubble sparged (Berberoglu et al., 2007) and airlift photobioreactors (Kaewpintong et al., 2007) are also common choices for microalga cultivation. Analyses of designed systems showed that CO2 mitigation costs depend closely on the productivity of algae and the intensity of solar radiation (Kadam, 1997). The main problem associated with the use of photobioreactors is the intensive capital cost (Terry and Raymond, 1985). To this end, a twostage strategy was demonstrated to be quite advantageous for algal oil production using Haematococcus pluvialis (Huntley and Redalje, 2007). Closed photobioreactors were used for the first stage and open ponds for the second. In the first stage, rich media containing all the required nutrients were used to allow fast cell growth and the employment of tightly controlled Table 15.4 Comparison between raceway ponds and tubular photobioreactors of microalgae System Raceway ponds

Tubular photobioreactors

Light efficiency Temperature control Gas transfer Produced oxygen accumulation Hydrodynamic stress on algae Species control Sterility Cost to scale-up Volumetric productivity

Excellent Excellent Low–high High Low–high Easy Achievable High High Low

Fairly good None Poor Low Low Difficult None Low Low High



425

closed photobioreactors allowed good species control. In the second stage, nitrogen limiting medium was used in open pond for lipid accumulation. Species control was achieved for two reasons: (i) the cell density of the cultivated microalgal species was high in the second stage; and (ii) nutrient limiting prevented the growth of wild-type algae in the open pond. In this demonstration project, a commercial-scale (2 ha) demonstration microalgal farming facility was operated consecutively for four years to produce H. pluvialis for biodiesel production. Daily production of 1.9 kg dry biomass was achieved with a 25 000 L photobioreactor, corresponding to a biomass productivity of 0.076 g l–1 d–1, at a biomass concentration of 0.3 g/l. An annual averaged rate of microalgal oil production, which was equivalent to 420 GJ ha–1 yr–1, was obtained. While the maximum production rate achieved with H. pluvialis was equivalent to 1014 GJ ha–1 yr–1, it was predicted that a rate of 3200 GJ ha–1 yr–1 is feasible using fast-growing Chlorella species. This is a rate which offers the potential to replace the reliance on current fossil fuel usage equivalent to about 300 EJ yr–1 and eliminate fossil fuel emissions of CO2 of about 6.5 gigatons of carbon (GtC) per year using only 7.3 % of the surplus arable land projected to be available by 2050. It was also expected that other microalgal biodiesel processes such as the one being developed at the University of Utah would be cost-competitive with regular diesel by 2009 (Seefeldt, 2007). There is no doubt that global efforts from both the public and private sectors will be continued and accelerated in order to make biofuels from microalgae a practical replacement of fossil fuels in the near future.

15.4.5 Microalgal biomass harvesting and drying Due to light limitation, the biomass concentrations of microalga suspensions are usually low, in the range 0.5–3.0 g/l. This low biomass concentration, combined with the small cell size of microalgae, makes the biomass harvesting and drying costly and energy consuming. Different technologies, including chemical flocculation (Knuckey et al., 2006), biological flocculation (Divakaran and Pillai, 2002), filtration (Molina Grima et al., 2003), centrifugation (Olaizola, 2003) and ultrasonic aggregation (Bosma et al., 2003) have been investigated for microalgal biomass harvesting. In general, chemical and biological flocculation entail low operating costs. However, they have the disadvantage of requiring long processing time and the risk of bioreactive product decomposition. On the other hand, filtration, centrifuge and ultrasonic flocculation are more efficient but more costly. The selection of appropriate harvesting technology depends on the value of the target products, the biomass concentration and the size of microalgal cells of interest. Biomass drying before further extraction and/or thermochemical processing


426


is another step that needs to be taken into consideration. Sun drying is probably the cheapest drying method that has been employed for the processing of microalgal biomass (Millamena et al., 1990; Prakash et al., 1997). However, this method takes long drying time, requires large drying surface and risks the loss of some bioreactive products. Low-pressure shelf drying is another low-cost drying technology that has been investigated (Prakash et al., 1997). It is, nevertheless ineffective, requires long processing time and risks decomposition of bioproducts. More efficient but more costly drying technologies that have been investigated for drying microalgae include drum drying (Prakash et al., 1997), spray drying (Leach et al., 1998; Desmorieux and Decaen, 2005), fluidized bed drying (Leach et al., 1998), freezing drying (Millamena et al., 1990) and refractance window dehydration technology (Nindo and Tang, 2007). It is important to find the balance between drying efficiency and cost-effectiveness.

15.5

Advantages and limitations

The microorganisms-for-CO2-mitigation strategy using microalgae and phototrophic bacteria offers numerous advantages. In general, photosynthetic microorganisms have high CO2 fixation rate, thanks to their vast surface to volume ratio and simple structure. There is a vast diversity of different autotrophic microorganisms that could grow and therefore fix CO2 in different niche environments. Ocean fertilization, which aims at enhancing the biological productivity of the oceans, is expected to make major contributions to the slowing down of the global warming process despite significant uncertainties and controversies at present. Large-scale commercial microalgal farming is also expected to make substantial contributions to CO2 mitigation. It has been reported that microalgae have the ability to fix CO2 while capturing solar energy with an efficiency 10–50 times greater than that of terrestrial plants (Usui and Ikenouchi, 1997; Li et al., 2008b). Combined CO2 mitigation and biofuel production using microalgae is expected to ease the energy crisis and global warming since it could capture solar energy while completely recycle CO2 (Fig. 15.3). While chemical reaction-based CO2 mitigation approaches are energy consuming and costly processes (Lin et al., 2003; Resnik et al., 2004), and the only economical incentive for CO2 sequestrating using the chemical reaction-based approach is the CO2 credits to be generated under the Kyoto protocol, CO2 biomitigation using microalgae could be made profitable from the production of biofuels and other novel bioproducts. Microbial CO2 biomitigation could be made more cost-effective and environmentally sustainable, especially when it is combined with other processes such as wastewater treatment and biorefinery. For instance, utilization of wastewater for microalga cultivation will bring about remarkable advantages including:



427

(i) microalgae have been shown to be efficient in nitrogen and phosphorus removal (Mallick, 2002) as well as in metal ion depletion, and the combination of microalga cultivation with wastewater treatment will significantly enhance the environmental benefit of this strategy; (ii) it will lead to savings in term of minimizing the use of chemicals such as sodium nitrate and potassium phosphorus as exogenous nutrients; and (iii) it will result in conservation of precious freshwater resources. Figure 15.3 depicts a conceptual flow chart for the complete ‘recycling’ of CO2 for solar energy capturing. Nevertheless, there are some limitations in the microbial CO2 mitigation strategy that need to be addressed. While the disadvantage of CO2 fixation involving chemoautotrophic bacteria is their slow growth rate due to the limited energy that can be extracted from the oxidization of reduced inorganic compounds, the primary concern associated with microalgal farming is the relatively high costs of microalgal cultivation, biomass harvesting, drying and downstream processing. In the case of ocean fertilization, the major concern is the uncertainty with regard to the biochemical and ecological side effects of this approach.

15.6

Future trends

CO2 fixation using fast-growing autotrophic microorganisms, especially fastgrowing microalgal species, in their natural habitats or in artificial cultivation systems provides a very promising alternative for mitigation of CO2, the most prominent greenhouse gas. The primary merit of this strategy lies in

Solar energy

CO2

N/P-Rich wastewater

Microalgal cultivation

Low N/P effluent

Biomass refinery

Value-added bioproducts

Biofuel

Energy

CO2

15.3 A conceptual microalgal farming system for combined biofuel production, CO2 biomitigation and N/P removal from wastewater. Inputs: carbon source, CO2; nitrogen and phosphorus sources, N/P-rich wastewater; energy source, solar energy. Outputs: Low N/P effluent, value-added bioproducts and biofuels. (Wang et al., 2008).


428


the fact that microbial CO2 mitigation is efficient, environmentally friendly and potentially cost-effective. Ocean fertilization, despite the uncertainties surrounding this approach, is likely to provide the most effective strategy for large-scale CO2 sequestration. Biofixation of CO2 microalgal farming will also play an important role in preventing and curing CO2 emission and global warming. Integrated strategies such as the combination of flue gas CO2 mitigation with high-value co-product production, wastewater treatment and/or waste heat utilization is expected to enhance the economic viability of the microalgal farming CO2 biofixation strategy. Extensive studies are expected to generate sufficient technology advances for cost-effective CO 2 fixation using microorganisms.

15.7

References

Aumont O and Bopp L (2006) Globalizing results from ocean in situ iron fertilization studies. Global Biogeochemical Cycles, 20. Belarbi E H, Molina E and Chisti Y (2000) A process for high yield and scaleable recovery of high purity eicosapentaenoic acid esters from microalgae and fish oil. Enzyme and Microbial Technology, 26, 516–529. Berberoglu H, Yin J and Pilon L (2007) Light transfer in bubble sparged photobioreactors for H2 production and CO2 mitigation. International Journal of Hydrogen Energy, 32, 2273–2285. Borowitzka M A (1999) Commercial production of microalgae: ponds, tanks, tubes and fermenters. Journal of Biotechnology, 70, 313–321. Bosma R, Van Spronsen W A, Tramper J and Wijffels R H (2003) Ultrasound, a new separation technique to harvest microalgae. Journal of Applied Phycology, 15, 143–153. Chelf P, Brown L M and Wyman C E (1993) Aquatic biomass resources and carbon dioxide trapping. Biomass and bioenergy, 4, 175–183. Chiaramonti D, Oasmaa A and Solantausta Y (2007) Power generation using fast pyrolysis liquids from biomass. Renewable and Sustainable Energy Reviews, 11, 1056–1086. Chisti Y (2007) Biodiesel from microalgae. Biotechnology Advances, 25, 294–306. Colman B and Rotatore C (1995) Photosynthetic inorganic carbon uptake and accumulation in two marine diatoms. Plant, Cell and Environment, 18, 919–924. Cullen J J and Boyd P W (2008) Predicting and verifying the intended and unintended consequences of large-scale ocean iron fertilization. Marine Ecology Progress Series, 364, 295–301. De Morais M G and Costa J A V (2007a) Biofixation of carbon dioxide by Spirulina sp. and Scenedesmus obliquus cultivated in a three-stage serial tubular photobioreactor. Journal of Biotechnology, 129, 439–445. De Morais M G and Costa J A V (2007b) Isolation and selection of microalgae from coal fired thermoelectric power plant for biofixation of carbon dioxide. Energy Conversion and Management, 48, 2169–2173. Demirbas A (2001) Biomass resource facilities and biomass conversion processing for fuels and chemicals. Energy Conversion and Management, 42, 1357–1378. Demirbas A (2004) Current technologies for the thermo-conversion of biomass into fuels and chemicals. Energy Sources, 26, 715–730.



429

Desmorieux H and Decaen N (2005) Convective drying of spirulina in thin layer. Journal of Food Engineering, 66, 497–503. Divakaran R and Pillai V N S (2002) Flocculation of algae using chitosan. Journal of Applied Phycology, 14, 419–422. Emma Huertas I, Colman B, Espie G S and Lubian L M (2000) Active transport of CO2 by three species of marine microalgae. Journal of Phycology, 36, 314–320. Flores Iii A, Nisola G M, Cho E, Gwon E M, Kim H, Lee C, Park S and Chung W J (2007) Bioaugmented sulfur-oxidizing denitrification system with Alcaligenes defragrans B21 for high nitrate containing wastewater treatment. Bioprocess and Biosystems Engineering, 30, 197–205. Freestone D and Rayfuse R (2008) Ocean iron fertilization and international law. Marine Ecology Progress Series, 364, 227–233. Fuhrman J A and Capone D G (1991) Possible biogeochemical consequences of ocean fertilization. Limnology and Oceanography, 36, 1951–1959. Ginzburg B Z (1993) Liquid fuel (oil) from halophilic algae: a renewable source of nonpolluting energy. Renewable Energy, 3, 249–252. Glibert P M, Azanza R, Burford M, Furuya K, Abal E, Al-Azri A, Al-Yamani F, Andersen P, Anderson D M, Beardall J, Berg G M, Brand L, Bronk D, Brookes J, Burkholder J M, Cembella A, Cochlan W P, Collier J L, Collos Y, Diaz R, Doblin M, Drennen T, Dyhrman S, Fukuyo Y, Furnas M, Galloway J, Granéli E, Ha D V, Hallegraeff G, Harrison J, Harrison P J, Heil C A, Heimann K, Howarth R, Jauzein C, Kana A A, Kana T M, Kim H, Kudela R, Legrand C, Mallin M, Mulholland M, Murray S, O’neil J, Pitcher G, Qi Y, Rabalais N, Raine R, Seitzinger S, Salomon P S, Solomon C, Stoecker D K, Usup G, Wilson J, Yin K, Zhou M and Zhu M (2008) Ocean urea fertilization for carbon credits poses high ecological risks. Marine Pollution Bulletin, 56, 1049–1056. Gomez-Villa H, Voltolina A D, Nieves M and Pina P (2005) Biomass production and nutrient budget in outdoor cultures of Scenedesmus obliquus (Chlorophyceae) in artificial wastewater, under the winter and summer conditions of Mazatlan, Sinaloa, Mexico. Vie et Milieu, 55, 121–126. Haddadin J, Morin D, Ollivier P and Fick M (1993) Effect of different carbon dioxide concentrations on ferrous iron and pyrite oxidation by a mixed culture of iron and/or sulfur-oxidizing bacteria. Enzyme and Microbial Technology, 15, 832–841. Herter S, Farfsing J, Gad’on N, Rieder C, Eisenreich W, Bacher A and Fuchs G (2001) Autotrophic CO2 fixation by Chloroflexus aurantiacus: Study of glyoxylate formation and assimilation via the 3-hydroxypropionate cycle. Journal of Bacteriology, 183, 4305–4316. Herter S, Fuchs G, Bacher A and Eisenreich W (2002) A bicyclic autotrophic CO 2 fixation pathway in Chloroflexus aurantiacus. Journal of Biological Chemistry, 277, 20277–20283. Hirano A, Hon-Nami K, Kunito S, Hada M and Ogushi Y (1998) Temperature effect on continuous gasification of microalgal biomass: theoretical yield of methanol production and its energy balance. Catalysis Today, 45, 399–404. Hu Q, Guterman H and Richmond A (1996) A flat inclined modular photobioreactor for outdoor mass cultivation of photoautotrophs. Biotechnology and Bioengineering, 51, 51–60. Hulger M, Huber H, Stetter K O and Fuchs G (2003) Autotrophic CO2 fixation pathways in archaea (Crenarchaeota). Archives of Microbiology, 179, 160–173. Huntley M E and Redalje D G (2007) CO2 mitigation and renewable oil from photosynthetic


430


microbes: a new appraisal. Mitigation and Adaptation Strategies for Global Change, 12, 573–608. Iwasaki I, Hu Q, Kurano N and Miyachi S (1998) Effect of extremely high-CO2 stress on energy distribution between photosystem I and photosystem II in a ‘high-CO2’ tolerant green alga, Chlorococcum littorale and the intolerant green alga Stichococcus bacillaris. Journal of Photochemistry and Photobiology B: Biology, 44, 184–190. Javanmardian M and Palsson B O (1991) High-density photoautotrophic algal cultures: Design, construction, and operation of a novel photobioreactor system. Biotechnology and Bioengineering, 38, 1182–1189. Kadam K L (1997) Power plant flue gas as a source of CO2 for microalgae cultivation: Economic impact of different process options. Energy Conversion and Management, 38, S505–S510. Kaewpintong K, Shotipruk A, Powtongsook S and Pavasant P (2007) Photoautotrophic high-density cultivation of vegetative cells of Haematococcus pluvialis in airlift bioreactor. Bioresource Technology, 98, 288–295. Kishimoto M, Okabura T, Nagashima H, Minowa T, Yokoyama S Y and Yamaberi K (1994) CO2 fixation and oil production using micro-algae. Journal of Fermentation and Bioengineering, 78, 479–482. Knuckey R M, Brown M R, Robert R and Frampton D M F (2006) Production of microalgal concentrates by flocculation and their assessment as aquaculture feeds. Aquacultural Engineering, 35, 300–313. Kondili E M and Kaldellis J K (2007) Biofuel implementation in East Europe: Current status and future prospects. Renewable and Sustainable Energy Reviews, 11, 2137–2151. Kwak K O, Jung S J, Chung S Y, Kang C M, Huh Y I and Bae S O (2006) Optimization of culture conditions for CO2 fixation by a chemoautotrophic microorganism, strain YN-1 using factorial design. Biochemical Engineering Journal, 31, 1–7. Leach G, Oliveira G and Morais R (1998) Spray-drying of Dunaliella salina to produce a b-carotene rich powder. Journal of Industrial Microbiology and Biotechnology, 20, 82–85. Li Y, Horsman M, Wang B, Wu N and Lan C Q (2008a) Effects of nitrogen sources on cell growth and lipid accumulation of green alga Neochloris oleoabundans. Applied Microbiology and Biotechnology, 81, 629–636. Li Y, Horsman M, Wu N, Lan C Q and Dubois-Calero N (2008b) Biofuels from microalgae. Biotechnology Progress, 24, 815–820. Lin C C, Liu W T and Tan C S (2003) Removal of carbon dioxide by absorption in a rotating packed bed. Industrial and Engineering Chemistry Research, 42, 2381–2386. Liss P, Chuck A, Bakker D and Turner S (2005) Ocean fertilization with iron: effects on climate and air quality. Tellus, Series B: Chemical and Physical Meteorology, 57, 269–271. Maeda K, Owada M, Kimura N, Omata K and Karube I (1995) CO2 fixation from the flue gas on coal-fired thermal power plant by microalgae. Energy Conversion and Management, 36, 717–720. Mallick N (2002) Biotechnological potential of immobilized algae for wastewater N, P and metal removal: a review. BioMetals, 15, 377–390. Matsumoto H, Hamasaki A, Sioji N and Ikuta Y (1997) Influence of CO2, SO2 and no in flue gas on microalgae productivity. Journal of Chemical Engineering of Japan, 30, 620–624. Mckendry P (2002) Energy production from biomass (part 3): gasification technologies. Bioresource Technology, 83, 55–63.



431

Merrett M J, Nimer N A and Dong L F (1996) The utilization of bicarbonate ions by the marine microalga Nannochloropsis oculata (Droop) Hibberd. Plant, Cell and Environment, 19, 478–484. Michiki H (1995) Biological CO2 fixation and utilization project. Energy Conversion and Management, 36, 701–705. Millamena O M, Aujero E J and Borlongan I G (1990) Techniques on algae harvesting and preservation for use in culture and as larval food. Aquacultural Engineering, 9, 295–304. Minowa T and Sawayama S (1999) Novel microalgal system for energy production with nitrogen cycling. Fuel, 78, 1213–1215. Molina Grima E, Belarbi E H, Acien Fernandez F G, Robles Medina A and Chisti Y (2003) Recovery of microalgal biomass and metabolites: Process options and economics. Biotechnology Advances, 20, 491–515. Murakami M and Ikenouchi M (1997) The biological CO2 fixation and utilization project by RITE (2): Screening and breeding of microalgae with high capability in fixing CO2. Energy Conversion and Management, 38, S493–S497. Nindo C I and Tang J (2007) Refractance window dehydration technology: a novel contact drying method. Drying Technology, 25, 37–48. Olaizola M (2003) Commercial development of microalgal biotechnology: from the test tube to the marketplace. Biomolecular Engineering, 20, 459–466. Orbach M K (2008) Cultural context of ocean fertilization. Marine Ecology Progress Series, 364, 235–242. Prakash J, Pushparaj B, Carlozzi P, Torzillo G, Montaini E and Materassi R (1997) Microalgal biomass drying by a simple solar device. International Journal of Solar Energy, 18, 303–311. Resnik K P, Yeh J T and Pennline H W (2004) Aqua ammonia process for simultaneous removal of CO2, SO2 and NOx. International Journal of Environmental Technology and Management, 4, 89–104. Rosello Sastre R, Csogor Z, Perner-Nochta I, Fleck-Schneider P and Posten C (2007) Scale-down of microalgae cultivations in tubular photo-bioreactors–a conceptual approach. Journal of Biotechnology, 132, 127–133. Sakai N, Sakamoto Y, Kishimoto N, Chihara M and Karube I (1995) Chlorella strains from hot springs tolerant to high temperature and high CO2. Energy Conversion and Management, 36, 693–696. Scragg A H, Illman A M, Carden A and Shales S W (2002) Growth of microalgae with increased calorific values in a tubular bioreactor. Biomass and Bioenergy, 23, 67–73. Seefeldt L C (2007) Utah group plans to make biodiesel from algae. Industrial Bioprocessing, 29, 5–6. Shi M and Shen Y M (2003) Recent progresses on the fixation of carbon dioxide. Current Organic Chemistry, 7, 737–745. Shively J M, Van Keulen G and Meijer W G (1998) Something from almost nothing: carbon dioxide fixation in chemoautotrophs. Annual Review of Microbiology, 52, 191–230. Sorokin D Y, Tourova T P, Muyzer G and Kuenen G J (2008) Thiohalospira halophila gen. nov., sp. nov. and Thiohalospira alkaliphila sp. nov, novel obligately chemolithoautotrophic, halophilic, sulfur-oxidizing gammaproteobacteria from hypersaline habitats. International Journal of Systematic and Evolutionary Microbiology, 58, 1685–1692.


432


Spolaore P, Joannis-Cassan C, Duran E and Isambert A (2006) Commercial applications of microalgae. Journal of Bioscience and Bioengineering, 101, 87–96. Terry K L and Raymond L P (1985) System design for the autotrophic production of microalgae. Enzyme and Microbial Technology, 7, 474–487. Travieso L, Hall D O, Rao K K, Benitez F, Sanchez E and Borja R (2001) A helical tubular photobioreactor producing Spirulina in a semicontinuous mode. International Biodeterioration and Biodegradation, 47, 151–155. Usui N and Ikenouchi M (1997) The biological CO2 fixation and utilization project by RITE(1): highly-effective photobioreactor system. Energy Conversion and Management, 38, S487–S492. Vonshak A and Richmond A (1988) Mass Production of the blue-green alga spirulina: an overview. Biomass London, 15, 233–247. Wang B, Li Y, Wu N and Lan C Q (2008) CO2 bio-mitigation using microalgae. Applied Microbiology and Biotechnology, 79, 707–718. Weissman J C, Goebel R P and Benemann J R (1988) Photobioreactor design: mixing, carbon utilization, and oxygen accumulation. Biotechnology and Bioengineering, 31, 336–344. Yun Y S, Lee S B, Park J M, Lee C I and Yang J W (1997) Carbon dioxide fixation by algal cultivation using wastewater nutrients. Journal of Chemical Technology and Biotechnology, 69, 451–455. Zeebe R E and Archer D (2005) Feasibility of ocean fertilization and its impact on future atmospheric CO2 levels. Geophysical Research Letters, 32, 1–5


16

Mineralisation of carbon dioxide (CO2) R . Z e v e n h o v e n and J . F a g e r l u n d, Åbo Akademi University, Finland Abstract: Mineralisation of carbon dioxide (CO2), or mineral carbonation, involves the reaction of CO2 with materials containing alkaline-earth oxides like magnesium oxide (MgO) and calcium oxide (CaO). For large-scale CO2 capture and sequestration (CCS) purposes this makes use of the vast resources of magnesium silicate minerals that are available worldwide, resulting in an environmentally benign magnesium carbonate product that needs no post-storage monitoring. As a spin-off technology related to this, the production of valuable calcium carbonates from industrial by-products and wastes quickly develops into profitable technology. Technologies, raw material resources and recent developments are presented here, with chemical reaction kinetics, recovery/re-use of chemical additives and energy economy being important bottlenecks. Key words: CO2 mineralisation, magnesium silicate mineral, magnesium carbonate, industrial by-products and wastes, precipitated calcium carbonate.

16.1

Introduction

This section briefly presents the fixation of carbon dioxide (CO2) as carbonate minerals, addressing natural processes and the general features and potential of this CO2 capture and sequestration (CCS) method. The mineralisation of CO2, also known as mineral carbonation or CO2 mineral sequestration, for the long-term storage of CO2 is a CCS option (IPCC, 2005) that presents presumably the most important alternative for the more widely advocated method of CO2 capture and geological storage (CCGS) in underground formations. It involves the carbonation of natural silicate minerals that contain alkaline-earth oxides like magnesium oxide (MgO) and calcium oxide (CaO), in principle available as in situ rock material. Basically it aims at translating the very slow geological process known as natural weathering of silicate minerals into economically viable technology. To some extent, it can make use of the significant amounts of MgO or CaO containing mining tailings and other residues and by-products from mineral mining and process industry. Although only the carbonation of Mg-based material offers the capacity needed for significantly reducing anthropogenic CO2 emissions, the production of precipitated calcium carbonate (PCC) and other valuable carbonates from industrial by-products and residues forms an increasingly important CCS spin-off technology with an annual CO2 binding potential of a few 100 Mt. For steel-making slags alone, estimates for the CO2 binding 433 © Woodhead Publishing Limited, 2010

434


Ocean carbon

Annual emissions

Biomass Atmosphere

potential range from 60–170 Mt/a (Zevenhoven et al., 2006, Eloneva et al., 2008a,b; Teir, 2008). As illustrated by Fig. 16.1, Mg-containing (and, to a much lesser extent, Ca-containing) mineral carbonation gives the highest capacity and longest storage time of currently known CCS options (Lackner, 2003). There are more than enough materials, such as serpentine and olivine, available on a global scale to sequester all anthropogenic CO2 that could ever be produced (Lackner, 2003; IPCC, 2005; Krevor et al., 2009). In contrast to CCGS1 methods, the post-storage monitoring of CO2 would not be needed and, moreover, the deployment of CCGS is progressing very slowly (seemingly as a result of policy-making and liability issues) (Bachu, 2008) and might not offer the necessary capacity soon enough, putting pressure on the development of alternatives. The total worldwide capacity of ~7 Mt CO2/a in 2008 is expected to increase to 24 Mt CO2/a in 2012 (Thambimuthu, 2009); it is hard to believe that this will soon accelerate to a sequestration

Mineral carbonates

Underground injection

10 000 EOR 1000

Fossil carbon

Characteristic storage time (yr)

100 000

100 10 1

1

10

100 1000 10 000 100 000 1 000 000 Carbon storage capacity (Gt)

16.1 Estimated storage capacities and storage times for various CCS methods. (from Zevenhoven et al., 2006, after Lackner, 2003).

1 Note that in the recent EU directive on geological storage of CO2 (EC, 2009) this CCS method is confused with CCS in general: ‘Carbon dioxide capture and geological storage (CCS) … consists of the capture of carbon dioxide (CO2) from industrial installations, its transport to a storage site and its injection into a suitable underground geological formation for the purposes of permanent storage’.



435

rate of many Gt CO2/a within a few decades. While the energy sector (often intertwined with the oil and gas industry) focuses primarily on CCGS, CO2 mineralisation is receiving increased attention from the mineral, cement, metal and pulp and paper industries, taking benefit from the combination of CO2 emissions mitigation, waste/by-product utilisation and apparently cheap routes to commercial carbonate materials.

16.2

Basic principles and methods

This section presents the basic principles of CO2 mineralisation, addressing the general features of process chemistry and process energy economics, distinguishing between direct (single-step) and indirect (multistep) process routes. The process of natural weathering of silicate minerals is a quite well understood geological process, and a relatively small (yet steadily increasing) number of countries like the USA, Norway, Finland, Switzerland, Japan, the UK, Italy, the Netherlands and the Baltic states but also ‘newcomers’ like Singapore, Australia, Germany, Portugal and Iceland are looking into technical solutions to perform the carbonation of magnesium from silicate minerals with CO2 on a large scale. Mineralisation of CO2 is attractive, especially at locations where absence of underground storage volume excludes CCGS, the risk of leakage of the CO2 stored underground is considered unacceptable or large resources of material suitable for carbonation are present. One of the countries that can make use of this option is Finland. It was assessed that underground storage sites for CCGS are not available in Finland while vast resources of suitable magnesium silicates exist (Koljonen et al., 2004; Rinne, 2008; Teir, 2008) although a renewed assessment is ongoing (Nieminen et al., 2009). Finland has emission reduction commitments under the 1997 Kyoto protocol to reduce its greenhouse gas emissions to the level of year 1990 during 2008–2012, while current emissions of CO2 are 10–15 Mt/a above that level. For example, the serpentinite rock formations in easterncentral Finland alone could bind 2.5–3.5 Gt CO2. However, after nearly 20 years of R&D efforts no commercial silicate mineral carbonation technology for CCS has yet been developed (problems being, particularly, the slow chemistry and concerns about process energy economy for low-temperature processes), although more and more countries report having access to suitable types and amounts of mineral resources or, despite a much smaller CCS potential, calcium-containing materials (e.g. Korea, Lithuania, India). The ‘highly verifiable and unquestionably permanent’ nature of this CO2 storage method (IPCC, 2005), which may lead to a greater public acceptance, or easy access to suitable minerals, or goals like land reclamation using the solid products of a large-scale CO2 mineralisation process are driving forces for the increasing interest in this CCS option.


436


Naturally occurring silicate minerals such as serpentine (sheet silicates) and olivine contain high concentrations of MgO, while so-called pyroxenes and amphiboles (chain silicates) are a potential source for both CaO and MgO (see e.g. Brownlow, 1996). Carbonation of these traps CO2 as environmentally stable solid carbonates, which for serpentine (Reaction 16.1), olivine (forsterite) (Reaction 16.2) and wollastonite (Reaction 16.3) minerals can be described by the following overall chemistry:

Mg3Si2O5(OH)4 (s) + 3CO2 (g) Æ 3MgCO3 (s)

+ 2SiO2 (s) + 2H2O (l)

[16.1]

MgSiO3 (s) + CO2 (g) Æ MgCO3 (s) + SiO2 (s)

[16.2]

CaSiO3 (s) + CO2 (g) Æ CaCO3 (s) + SiO2 (s)

[16.3]

These chemical reactions are exothermic, and the heat released is dependent on the metallic element-bearing mineral used (for the magnesium- or calciumbased silicate minerals olivine: 89 kJ/mol CO2, serpentine: 64 kJ/mol CO2 and wollastonite: 90 kJ/mol CO2 at 298 K). Besides serpentine, olivine and wollastonite, basalt, with MgO + CaO-content typically of the order of 20 wt% or less but widely available, is also gaining rapidly increasing interest (McGrail et al., 2006; Schaef et al., 2009). Although the overall reactions (Reactions 16.1–16.3) are exothermic – which implies that proper optimisation and process integration should allow for operation at zero or negative net energy input – the natural carbonation of silicate minerals is very slow and the reaction must be accelerated considerably to be an economically viable large-scale CO2 storage technology. Moreover, a carbonation process would in practice involve two or more stages where the production of reactive Mg or Ca is followed by its carbonation in a subsequent step. With different temperatures, pressures and energy requirements for the different stages, there is a risk that the overall process needs an energy input. Two major problems must be solved to make large-scale mineralisation of CO2 more attractive (see also Herzog, 2002): ∑

extracting or activating the reactive component MgO from a silicate mineral; ∑ speeding-up the carbonation chemistry kinetics.

When it comes to technology development, the most important results have been reported from the USA since the mid 1990s. While the research in the USA, followed by several other countries, concentrated increasingly on ‘wet’ methods using aqueous solutions, the research in Finland also addresses ‘dry’ methods. The reason for preferring a gas/solid route is that aqueous processes have temperatures much lower than, say, 500 °C where the rate of the gas/ solid carbonation with MgO or Mg(OH)2 appears to become significant. The energy consumption of the ‘wet’ processes is high (to be further discussed © Woodhead Publishing Limited, 2010


437

below) simply because heat of the exothermic reaction is not taken advantage of; it should be much lower (or negative) for a gas/solid process with heat recovery, with the energy output from the exothermic carbonation reaction step as the obvious starting point. An important feature of CO2 mineralisation will be solids mining and handling and the question of what to do with the carbonate product material. A tonne of CO2 will require 2.5–3 t magnesium silicate mineral (for coal combustion-derived CO2 this implies ~ 8 t mineral per tonne coal). Altogether, this will result in a mining activity similar to typical commercial mining of coal or metal-containing ore. An example of a very large-scale processing of solids is the mining and processing of oil sands in Alberta, Canada, where 1 Mt solids material is moved each day (Kunzig, 2009). Using MgCO3-containing material in, for example, building materials or flame retardants may be considered; it cannot however, be used in paper products, (Zevenhoven et al., 2006). On the other hand, because it is also a product of natural weathering of silicate mineral, the carbonate is benign and stable when disposed of in nature. Tests showed that MgCO3 and CaCO3 are stable in rain water and acid solutions down to pH = 1 (Teir et al., 2006), but a very slow dissolution in water bodies and streams could, in fact, be beneficial. Increasing the concentrations of Mg2+, Ca2+ and HCO3– ions can provide a buffer to the decomposition of solid carbonate material under water, as the equilibrium reactions CaCO3(s) + H2O(l) + CO2(aq) = Ca2+(aq) + CO32–(aq) + H2O(l) + CO2(aq) = Ca2+(aq) + 2HCO3–(aq), with HCO3– being much better soluble than CO32–, will then be pushed to the left side. At the same time, dissolved Mg2+ may support increased growth of biomass, which is part of many renewable energy schemes. It should be noted that other solid products of magnesium silicate carbonation are significant amounts of silicon oxides and iron oxides. For example, with Finnish serpentinite (i.e. rock material containing serpentine) containing ~ 13 wt% iron oxides, a large-scale CO2 sequestration effort using this mineral will give a very significant stream of iron oxide by-product too large to be overlooked by iron- and steel-making as was also recognised by Lackner et al. (2008). It is important to distinguish between in situ and ex situ approaches: in situ mineral carbonation is closely connected to CCGS as it involves the injection of CO2 into underground reservoirs. The difference is that in situ mineral carbonation explicitly aims at reacting the CO2 to form carbonates with alkaline-minerals present in the geological formation. In a recent paper (Oelkers et al., 2008), the advantages of in situ mineral carbonation compared to ex situ processes were pointed out for basaltic rock carbonation as applied at Hellisheidi, Iceland within the CarbFix project, where the mass of rock to be moved in ex situ processes is considered to make it ‘impractical’. It should not be forgotten that in CCGS too the final CO2 fixation stage after very long periods of time will be in situ mineralisation.


438


Technology overviews were produced by Newall et al. (2000), Huijgen and Comans (2003, 2005) and very recently by Sipilä et al. (2008), dealing with Mg-based as well as Ca-based materials.

16.3

Technologies and potential applications

Here, more detail is presented on the various wet/semi-dry/dry direct and indirect processes for CO2 mineralisation, discussing calcium- and magnesium-based mineral and other feedstocks and the use and application of the carbonate product. In the open literature, almost no data can be found on the kinetics of chemical reactions between magnesium silicates or oxides with CO2, especially for the period prior to 2000. Langmuir (1965) refers to experimental work from the early 20th century that reports very slow chemistry for MgO and CO2; the catalytic effect of water on MgO carbonation is also mentioned. Since the mid-1990s, many results from mineral carbonation work by consortia in the USA at Los Alamos National Laboratory (LANL) (Lackner and coworkers) and later from Albany Research Center (ARC) (O’Connor and co-workers) and Arizona State University (McKelvy and co-workers) have been published, often in the form of work-in-progress reporting. Roughly after 2000, other teams in the USA, e.g. PennState (Park and co-workers, Maroto-Valer), and in other countries – Switzerland (Mazzotti), Finland (Zevenhoven), Estonia (Kuusik), Italy (Baciocchi), the Netherlands (Comans and Huijgen), Canada (e.g Stolaroff, Keith), Japan (e.g. Yogo), Norway (Munz) and others – followed.

16.3.1 Early developments (1990–2000) Mineral carbonation was first mentioned not that long ago as a CO2 binding concept by Seifritz (1990). A few years after that, the concept of binding CO2 in calcium and magnesium carbonate minerals was further investigated in the USA by Dunsmore (1992) and this process, also known as enhanced natural weathering, was later investigated in more detail by Lackner and coworkers at LANL (Lackner et al., 1995; Goff and Lackner, 1998). Natural silicate minerals such as olivine, serpentine and wollastonite but also basalt rock were identified as the most suitable raw materials, being abundant and cheap. Since then, research around mineral carbonation has accelerated and divided into several different CO2 binding approaches, using either direct (where the carbonation of the mineral takes place in a single process step) or indirect (where calcium or magnesium is first extracted from the mineral and subsequently carbonated) routes – see Fig. 16.2. (In this figure, chemical additive streams are not shown as these are supposed not to be consumed


Mineralisation of carbon dioxide (CO2) Direct carbonation

439

Indirect carbonation Mineral

Mineral

CO2 Extraction/activation

Carbonation

Mg- and Ca-compounds CO2

Carbonated mineral

Residue

Carbonation Carbonates

16.2 A schematic representation of the principles of direct and indirect carbonation.

in an economically feasible large-scale CCS process.) These primarily aim at ex situ processing in a dedicated processing plant (as opposed to in situ carbonation by injection of CO2 into geological formations). Findings on both direct gas–solid carbonation using pressurised CO2 (a carbonation conversion of 25–30 % was reached at 500 °C, 340 bar after 2 h with 50–100 mm size serpentinite particles), as well as an aqueous process via magnesium chloride salts (after decomposing mineral in hydrochloric acid) were reported by Lackner (2002), Lackner et al. (1995, 1997b) and Butt et al. (1998). The carbonation of Mg(OH)2 was soon found to be significantly faster than carbonation of MgO; using Mg(OH) 2 powder (20 mm) a conversion of 90 % was obtained after ½ h at 565 °C, 52 bar (Butt et al., 1997; Lackner et al., 1997a,b). Butt et al. (1996) analysed in detail the Mg(OH)2 decomposition and carbonation in He and in CO2: the presence of CO2 was found to slow down water release from Mg(OH)2 in the temperature range where MgCO3 is thermodynamically stable. (For example, at a CO2 pressure of 1 bar MgCO3 is stable up to temperatures of around 400 °C and at 35 bar CO2, MgCO3 is stable up to around 550 °C). A maximum conversion rate was reported at 375 °C, after 12 h in pure CO2 giving ~16.7 % conversion of Mg to carbonate (50–100 mm particles). Outside the USA, Kojima and co-workers (1997) studied the aqueous carbonation of wollastonite in a continuously stirred tank reactor exposed to CO2 under ambient conditions and found the conversion to be far too slow for an industrial application. A technology assessment that addressed six processing routes was published by IEA GHG in 2000 (Newall et al., 2000) concluding that all suffered from high costs and excessive energy use. These six routes involved CO2 reacting (i) directly with Mg-silicate; with Mg(OH)2 produced from Mg-silicate rock reacted with (ii) HCl, or (iii) MgCl2; with (iv) Mg-rich brine reacted with water; with (v) MgCO3 + NaCl, producing NaHCO3; and with (vi) Ca-silicate rock reacted with HCl. At this point, it was more or less clear that indirect process routes give a benefit of much


440


faster carbonation chemistry. Routes that involved calcium from sources other than silicate minerals did not yet receive much attention.

16.3.2 The next five years (2000–2004) Attempts to produce Mg(OH)2 from Mg-silicates had resulted in the evaluation of a process using hydrochloric acid (HCl in water) with intermediates MgCl2·nH2O and Mg(OH)Cl, followed by gas–solid carbonation of Mg(OH)2 (see Lackner et al., 1997b; Newall et al., 2000). Due to the complexity and energy consumption of this process route, other US researchers proceeded with work on direct routes based on aqueous solutions. Soon after the end of the 1990s, US researchers reported conversion rates such as 65 % conversion after 1 h and 80 % conversion within ½ h after optimisation of solution chemistry, heat treatment and grinding. Currently, the most successful route for mineral carbonation is still considered to be using an aqueous solution of 0.64M NaHCO3 and 1M NaCl at 185 °C/150 bar for olivine, 155 °C/115 bar for heat-treated serpentine or 100 °C/40 bar for wollastonite. The minimum costs for CO2 sequestration would be 54, 78 and 64 US$/t CO2 sequestered, respectively, while for serpentine, thermal pre-treatment (at 615–630 °C) was found more efficient than mechanical treatment (O’Connor et al., 2005, Gerdemann, et al. 2007). Note that the study included mineral pre-treatment costs, but excluded CO2 separation and transport costs. Herzog (2002) concludes in his assessment that the reported carbonation levels and rates are obtained only at high energy cost: ‘a 20 % energy penalty for a coal-fired power plant’. Cost levels mentioned make this method too expensive at this point – see also (IPCC, 2005). Nonetheless, this route still receives attention and improvements are being reported. It has been noted, however, that the process costs are overestimated partly due to unrealistic calculation of energy efficiency and energy costs: the costs of process heat input are significantly overestimated when charged the same way as power input, giving a false impression of overall process economics (Sipilä et al., 2008; Zevenhoven et al., 2008b) – see also the next section. In 2000, work on mineral carbonation with carbonation at elevated temperatures (gas phase) as the central feature was started in Finland. Instead of aiming for successful chemical processing (that might later appear energetically unfeasible), the starting point was making use of the possibility of covering process energy requirements with the heat from the carbonation reaction (Zevenhoven and Kavaliauskaite, 2004; Zevenhoven et al., 2008a). Stepwise carbonation initially involved testing with Finnish serpentine and (calcined) magnesium hydroxide Mg(OH)2 powder in a pressurised thermogravimetric analyser (PTGA). Since elevated pressures did not give the expected increase in MgO carbonation rate, it was decided to proceed with Mg(OH)2 under test conditions where MgO formation is



441

thermodynamically unfavourable (Zevenhoven and Teir, 2004; Zevenhoven et al., 2007). Although the trend of these test results was in the right direction (as can be seen from Fig. 16.3), more work is required to achieve the reaction kinetics necessary for large-scale technology implementation. Still, the chemical kinetics of gas–solid carbonation are considerably slower than for carbonation using a direct aqueous process as developed at ARC in the USA, mentioned above. Conversion levels of the order of 80–90 % within minutes not hours will require a more suitable reactor type that allows for strong intra-particle mechanical effects, such as a fluidised bed (FB), and still higher CO2 pressures (up to and, if beneficial, above the supercritical pressure for CO2, ~ 74 bar) in order to allow for higher temperatures and hence faster carbonation kinetics. Other US researchers published a detailed study on Mg(OH)2 decomposition and the effect of pressure on Mg(OH) 2 carbonation, studying how dehydroxylation and rehydroxylation interact with carbonation (Béarat et al., 2002; Chizmeshya et al., 2002). Surprisingly, above the minimum CO2 pressure for stable MgCO3 at a given temperature (see above), the rate both of the dehydroxylation of Mg(OH)2 and of the carbonation of MgO decreased. Increasing pressures slow down dehydroxylation of Mg(OH)2, generating fewer reactive MgO sites for carbonation. In Finland, besides the PTGA tests, tests with Mg(OH)2 and MgO in an atmospheric bubbling FB reactor were made, with CO2 as fluidising gas. The product of carbonate material that builds up on the reacting particles (eventually slowing down the conversion) was noticeably removed from the particles as fines by attrition and abrasion, which were entrained from the reactor with the exit gas flow. These fines had a considerably higher MgCO3 content (8.1 wt%) than the material in the bed (4.4 wt%) after ~11 ar

s

1000

ye

MgO

63

da

ys

Mg(OH)2

6

1

s

14

ho 0

0.1

2003

2004

2005

2006 Year

ur 1

ho
1 Mt CO2/year mineralisation process up and running during the period 2015–2020.

16.6


Useful technology overviews are given for example in IPCC (2005, Chapter 7) and by Lackner (2002). Important useful information on state-of-theart and developments can be found in the proceedings of the biennial ACEME



453

(Accelerated Carbonation for Environmental and Materials Engineering, e.g. ACEME-08), GHGT (International Conference on Greenhouse Gas Control Technologies, e.g. GHGT9) and ICCDU (International Conference on Carbon Dioxide Utilization, e.g. ICCDU-X) conference series. Besides reporting in a wide range of scientific peer-reviewed journals, literature review reports are being produced every few years: see, for example, Newall et al. (2000), Huijgen and Comans (2003, 2005), Sipilä et al. (2008).

16.7

References

ACEME-08 (2008) Proc. of the 2nd Int. Conf. on Accelerated Carbonation for Environmental and Materials Engineering (ACEME-08), Baciocchi, R., Costa, G., Polettini, A., Pomi, R. (eds), Rome, Italy, 1–3, October. Andreani M, Luquot L, Gouze P, Godard M Hoise E and Gibert B (2009) ‘Experimental study of carbon sequestration reactions controlled by the percolation of CO2-rich brine through peridotites’, Environ. Sci. Technol., 43(4), 1226–1231. Bachu S (2008) ‘CO2 storage in geological media: role, means, status and barriers to deployment’, Prog. Energy Combust. Sci., 34, 254–273. Baciocchi R, Polettini A, Pomi R, Prigiobbe V, Von Zedwitz V N and Steinfeld A (2006) ‘CO2 sequestration by direct gas-solid carbonation of air pollution control (APC) residues’, Energy Fuels, 20, 1933–1940. Baciocchi R, Costa G, Polettini A, Pomi R and Progiobbe V (2009) ‘Comparison of different reaction routes for carbonation of APC residues’, in Gale J, Herzog H and Braitsch J (eds), Greenhouse Gas Control Technologies 9, Proceedings of the Ninth International Conference on Greenhouse Gas Control Technologies (GHGT9), Energy Procedia, 1, 4851–4858. Béarat H, McKelvy MJ, Chizmeshya AVG, Sharma R and Carpenter RW (2002) ‘Magnesium hydroxide dehydroxylation/carbonation reaction processes: implications for carbon dioxide mineral sequestration’, J. Am. Ceram. Soc. 85(4), 742–748. Béarat H, McKelvy MJ, Chizmeshya AVG, Gormley D, Nunez R, Carpenter RW, Squires K and Wolf GH (2006) ‘Carbon sequestration via aqueous olivine mineral carbonation: role of passivating layer formation’, Environ. Sci. Technol., 40, 4802–4808. Beaudoin G, Hébert R, Constantin M, Duchesne J, Cecchi E, Huot F, Vigneau S and Fiola R (2008) ‘Spontaneous carbonation of serpentine in milling and mining waste, Southern Québec and Italy’, 2nd International Conference on Accelerated Carbonation for Environmental and Materials Engineering (ACEME-08), Rome, Italy, 1–3, October, 73–82. Blencoe JG, Palmer DA, Anovitz LM and Beard JS (2004), Carbonation of metal silicates for long-term CO2 sequestration, Patent WO 2004/094043. Bonenfant D, Kharoune L, Sauvé S, Hausler R, Niquette P and Mimeault M (2008) ‘CO2 sequestration by aqueous red mud carbonation at ambient pressure and temperature’, Ind. Eng. Chem. Res., 47, 7610–7616. Boschi C, Dini A, Dallai L, Gianelli G and Ruggieri G (2008) ‘Mineralogical sequestration of carbon dioxide: new insights from the Malentrata magnesite deposit (Tuscany, Italy)’, 2nd International Conference on Accelerated Carbonation for Environmental and Materials Engineering (ACEME-08), Rome, Italy, 1–3, October, 55–61. Brownlow AH (1996) Geochemistry, 2nd edn, Prentice-Hall, Englewood Cliffs, NJ. Butt DP, Lackner KS, Wendt CH, Conzone SD, Kung H, Lu Y-C and Bremser JK (1996)


454


‘Kinetics of thermal dehydroxylation and carbonation of magnesium hydroxide’, J. Am. Ceram. Soc., 7, 1892–1898. Butt DP, Benjamin A, Holesinger T, Park YS, Lackner KS, Wendt CH, Currier RP, Harradine DM, Rising M and Nomura K (1997) ‘A method for permanent disposal of CO2 in solid form’, 8th Global Warming Conference and Expo, Columbia University, NY, 26–29 May. Butt DP, Lackner KS, Wendt CH, Vaidya R, Piled L, Park Y, Holesinger T, Harradine DM and Nomura K (1998) ‘The kinetics of binding carbon dioxide in magnesium carbonate’, 23rd international technical conference on coal utilization and fuel systems, 9–13 March, 1998, Clearwater, FL. Chizmeshya AVG, McKelvy MJ, Sharma R and Carpenter Bearat H (2002) ‘Density functional theory of the decomposition of Mg(OH)2: a lamellar dehydroxylation model’, Mater. Chem. Phys. 77, 416–425. Chizmeshya AVG, McKelvy MJ, Squires KD, Carpenter RW and Béarat H (2007) A novel approach to mineral carbonation: Enhancing carbonation while avoiding mineral pretreatment process cost, Final Report Project DE-FG26-04NT42124, June 2004–June 2007, Arizona State University, Phoenix, AZ. Dilmore R, Lu P, Allen D, Soong Y, Hedges S and Fu JK (2008) ‘Sequestration of CO2 in mixtures of bauxite residue and saline wastewater’, Energy Fuels, 22, 343–353. Dipple GM, Wilson SA, Power IM, Thom JM, Raudsepp M and Southam G (2008) ‘Passive mineral carbonation in mine tailings’, 2nd International Conference on Accelerated Carbonation for Environmental and Materials Engineering (ACEME-08), Rome Italy, 1–3 October, 119–122. Druckenmiller ML and Maroto-Valer MM (2005) ‘Carbon sequestration using brine of adjusted pH to form mineral carbonates’, Fuel Process Technol., 86, 1599–1614. Dunsmore HE (1992) ‘A geological perspective on global warming and the possibility of carbon dioxide removal as calcium carbonate mineral’, Energy Convers Manage, 33, 565–572. EC (2009) ‘Directive 2009/31/EC of the European Parliament and of the Council of 23 April 2009 on the geological storage of carbon dioxide’, Official Journal of the European Union, L140, June 5, 114–135. Eloneva S, Teir S, Fogelholm C-J and Zevenhoven R (2008a) ‘Fixation of CO 2 by carbonating calcium derived from blast furnace slag’, Energy, 33(9), 1461–1467. Eloneva S, Teir S, Salminen J, Fogelholm C-J and Zevenhoven R (2008b) ‘Steel converter slag as a raw material for precipitation of pure calcium carbonate’, Ind. Eng. Chem. Res., 47(18), 7104–7111. Fagerlund J, Teir S, Nduagu E and Zevenhoven R (2009) ‘Carbonation of magnesium silicate mineral using a pressurized gas-solid process’, in Gale J, Herzog H and Braitsch J (eds), Greenhouse Gas Control Technologies 9, Proceedings of the Ninth International Conference on Greenhouse Gas Control Technologies (GHGT9), Energy Procedia, 1, 4907–4914. Fagerlund J, Nduagu E, Romão I and Zevenhoven R (2010) ‘A stepwise process for carbon dioxide sequestration using magnesium silicates’, Front. Chem. Eng. China: 4(2), 133–141. Ferrini V, De Vito C and Mignardi S (2009) ‘Synthesis of nesquehonite by reaction of gaseous CO2 with Mg chloride solution: its potential role in the sequestration of carbon dioxide’, J. Hazard. Mater., 168(2–3), 832–837. Gerdemann SJ, O’Connor WK, Dahlin DC, Penner LR and Rush H (2007) ‘Ex situ aqueous mineral carbonation’, Environ. Sci. Technol., 41, 2587–2593.



455

GHGT9 (2009) 9th International Conference on Greenhouse Gas Control Technologies, Washington DC, 16–20 November 2008, Gale J, Herzog H and Braitsc J (eds), available as Energia Procedia, 1, 1–4984. GHGT9 (2009) Conference Summary. 9th International Conference on Greenhouse Gas Control Technologies, Washington DC, 16–20 November 2008, U.S. DOE, IEA GHG R&D programme and MIT. Goff F and Lackner KS (1998) ‘Carbon dioxide sequestering using ultramafic rocks’, Environ. Geosci., 5, 89–101. Hänchen M, Prigiobbe V, Storti G, Seward TM and Mazzotti M (2006) ‘Dissolution kinetics of fosteritic olivine at 90–150 °C including effects of the presence of CO2’, Geochim. Cosmochim. Acta., 70, 4403–4416. Hänchen M, Prigiobbe V, Storti G and Mazzotti M (2007a) ‘Mineral carbonation: Kinetic study of olivine dissolution and carbonate precipitation’, in Gale J, Rokke N, Zweigel P and Svenson H (eds), Proceedings of the Eighth International Conference on Greenhouse Gas Control Technologies: GHGT8, Elsevier, Oxford, UK, paper 01_013_04, CD–ROM. Hänchen M, Krevor S, Mazzotti M and Lackner K (2007b) ‘Validation of a population balance model for olivine’, Chem. Eng. Sci., 62, 6412–6422. Hänchen M, Prigiobbe V, Baciocchi R and Mazzotti M (2008) ‘Precipitation in the Mg-carbonate system – effects of temperature and CO2 pressure’, Chem. Eng. Sci., 63, 1012–1028. Herzog H (2002) Carbon sequestration via mineral carbonation: Overview and assessment, MIT Laboratory for Energy and the Environment, MIT, Cambridge, MA, March. Huijgen WJJ (2007) Carbon Dioxide Sequestration by Mineral Carbonation. PhD (Eng) thesis, University of Wageningen, Wageningen, the Netherlands. Huijgen WJJ and Comans RNJ (2003) Carbon dioxide sequestration by mineral carbonation. Report number ECN–C-03-016, Energy research Centre of the Netherlands (ECN), Petten, the Netherlands. Huijgen WJJ and Comans RNJ (2005) Carbon dioxide sequestration by mineral carbonation – Literature Review Update 2003–(2004), Report no. ECN-C-OS-022, Energy research Centre of the Netherlands (ECN), Petten, the Netherlands; Report number 2005/11, IEA Greenhouse Gas R&D Programme, Cheltenham, UK. Huijgen WJJ, Witkamp G and Comans RNJ (2006a) ‘Mechanisms of aqueous wollastonite carbonation as a possible CO2 sequestration process’, Chem. Eng. Sci., 61, 4242–4251. Huijgen WJJ, Ruijg GJ, Comans RNJ and Witkamp GJ (2006b) ‘Energy consumption and net CO2 sequestration of aqueous mineral carbonation’, Ind. Eng. Chem. Res., 45, 9184–9194. Huntzinger DN, Gierke JS, Kawatra SK, Eisele TC and Sutter LL (2009) ‘Carbon dioxide sequestration in cement kiln dust through mineral carbonation’, Environ. Sci. Technol., 43, 1986–1992. ICCDU-X (2009) Abstract book for the 10th International Conference on Carbon Dioxide Utilization, Liu C-J (ed.), 17–21 May, Tianjin, China. IEA (2008) Carbon capture and storage: a key carbon abatement option. IEA, Paris, France. IPCC (2005) IPCC Special Report on Carbon Dioxide Capture and storage, Working Group III of the Intergovernmental Panel on Climate Change, Metz B, Davidson O, de Coninck H C, Loos M and Meyer L A (eds), Cambridge University Press, Cambridge, UK.


456


IPCC (2007) Climate change 2007: Mitigation, Contribution of working group III to the fourth assessment report of the intergovernmental panel on Climate Change, Metz B, Davidson O, Bosch P, Dave R and Meyer L (eds), Cambridge UK. Jarvis K, Carpenter RW, Windman T, Kim Y, Nunez R and Alawneh F (2009) ‘Reaction mechanism for enhancing mineral sequestration of CO2’, Environ. Sci. Technol., 43, 6314–6319. Kakizawa M, Yamasaki A and Yanagisawa Y (2001) ‘A new CO2 disposal process via artificial weathering of calcium silicate accelerated by acetic acid’, Energy, 26, 341–354. Katsuyama Y, Yamasaki A, Iizuka A, Fujii M, Kumagai K and Yanagisawa Y (2005) ‘Development of a process for producing high-purity calcium carbonate (CaCO3) from waste cement using pressurized CO2’, Environ. Prog., 24, 162–170. Kelemen PB and Matter J (2008) ‘In-situ carbonation of peridotite for CO2 storage’, Proc. Nat. Acad. Sci. USA, 105(45), 17295–17300. Kelemen PB, Matter J and Streit L (2008) ‘Field observations and theoretical studies relevant to enhanced in situ carbonation of periodite’, Proceedings 2nd Int. Conf. on Accelerated Carbonation for Environmental and Materials Engineering (ACEME-08), Rome, Italy, 1–3 October, 105–112. Kojima T, Nagamine A, Ueno N and Uemiya S (1997) ‘Absorption and fixation of carbon dioxide by rock weathering’, Energy Convens Manage., 38, 461–466. Koljonen T, Siikavirta H, Zevenhoven R and Savolainen I (2004) ‘CO2 capture, storage and reuse potential in Finland’, Energy, 29, 1521–1527. Krevor SC and Lackner KS (2009) ‘Enhancing process kinetics for mineral carbonation’, in Gale J, Herzog H and Braitsch J (eds), Greenhouse Gas Control Technologies 9, Proceedings of the Ninth International Conference on Greenhouse Gas Control Technologies (GHGT9), Energy Procedia, 1, 4867–4871. Krevor SC, Graves CR, Van Gosen BS and McCafferty AE (2009) Mapping the Mineral Resource Base for Mineral Carbon-Dioxide Sequestration in the Conterminous United States, USGS Digital Data Series 14, United States Geological Survey, Reston, VA, available at: http://pubs.usgs.gov/ds/414/ (accessed January 2010). Kunzig R (2009) ‘The Canadian oil boom: scraping bottom’, National Geographic, 215(3), 38–59, available at: http://ngm.nationalgeographic.com/2009/03/canadian-oil-sands/ kunzig-text (accessed January 2010). Lackner KS (2002) ‘Carbonate chemistry for sequestering fossil carbon’, Annu. Rev. Energy Env. 27, 193–232. Lackner KS (2003) ‘A guide to CO2 sequestration’, Science, 300, 1677–1678. Lackner KS and Ziock H (2000) ‘From low to no emissions’, Modern Power Systems, 20, 31–32. Lackner KS, Wendt CH, Butt DP, Joyce EL and Sharp DH (1995) ‘Carbon dioxide disposal in carbonate minerals’, Energy, 20, 1153–1170. Lackner KS, Butt DP and Wendt CH (1997a) ‘Magnesite disposal of carbon dioxide’, Proceedings 22nd International Technical Conference on Coal Utilization and Fuel Systems, Clearwater, FL, 16–19 March, 419–430. Lackner KS, Butt DP and Wendt CH (1997b) ‘Progress on binding CO2 in mineral substrates’, Energy Convery Manage., 38, 259–264. Lackner KS, Ziock H-J and Grimes P (1999) ‘Carbon dioxide extraction from air: Is it an option?’, Proceedings 24th International Technical Conference on Coal Utilization & Fuel Systems, 8–11 March, Clearwater, FL, 885–896. Lackner KS, Grimes P and Ziock H-J (2001) ‘Capturing carbon dioxide from air’, in: Proc. of the First National Conference on Carbon Sequestration, May 14–17, Washington, DC, USA. © Woodhead Publishing Limited, 2010


457

Lackner KS, Duby PF, Yegulalp T, Krevor S and Graves C (2008) Integrating Steel Production with Mineral Carbon Sequestration, report TRP 9957 for US DOE, part III. Langmuir D (1965) ‘Stability of carbonates in the system MgO–CO2–H2O’, J. Geol., 73, 730–754. Lee JK and Dong WK (2005) ‘Characteristic study of Korean magnesium silicates for CO2 mineral carbonation’, 22nd Annual Int. Pittsburgh Coal Conference, 12–15 September, Pittsburgh, PA. Mani D, Nirmal Charan S and Kumar B (2008) ‘Assessment of carbon dioxide sequestration potential of ultramafic rocks in the greenstone belts of southern India’, Current Science, 94(1), 53–60. Maroto-Valer MM, Fauth DJ, Kuchta ME, Zhang Y and Andrésen JM (2005a) ‘Activation of magnesium rich minerals as carbonation feedstock materials for CO2 sequestration’, Fuel Process. Technol., 86, 1627–1645. Maroto-Valer MM, Zhang Y, Kuchta ME and Andresen J Fauth (2005b) Process for sequestering carbon dioxide and sulfur dioxide, US Patent 2005/002847, Jan. 6. McGrail BP, Schaef HT, Ho AM, Chien Y-J, Dooley JJ and Davidson CL (2006) ‘Potential for carbon dioxide sequestration in flood basalts’, J. Geophys. Res., 111, B12201. McKelvy MJ, Chizmesya AVG, Diefenbacher J, Béarat H and Wolf G (2004) ‘Exploration of the role of heat activation in enhancing serpentine carbon sequestration reactions’, Environ. Sci. Technol., 38, 6897–6903. Munz IA, Korneliussen A, Gorset O, Johansen H, Kihle J, Flaathen T and Sandvik K (2007) ‘Added value of industrial minerals and CO2 storage: New possibilities for Norwegian mineral industry’, in Gale J, Rokke N, Zweigel P and Svenson H (eds), Proceedings of the Eighth International Conference on Greenhouse Gas Control Technologies: GHGT8, Oxford, UK, Elsevier, paper P02_01_12, CD-ROM. Munz IA, Kihle J, Brandvoll Ö, Machenbach I, Carey JW, Haug TA, Johansen H and Eldrup N (2009) ‘A continuous process for manufacture of magnesite from olivine, CO2 and H2O’, in Gale J, Herzog H and Braitsch J (eds), Greenhouse Gas Control Technologies 9, Proceedings of the Ninth International Conference on Greenhouse Gas Control Technologies (GHGT9), Energy Procedia, 1, 4891–4898. Nduagu EI (2008), Mineral carbonation: preparation of magnesium hydroxide [Mg(OH)2] from serpentinite rock. MSc (Eng) thesis, Åbo Akademi University, Heat Engineering Lab., Turku (Finland). Newall PS, Clarke SJ, Haywood HM, Scholes H, Clarke NR, King PA and Barley RW (2000) CO2 storage as carbonate minerals, Report number PH3/17. IEA Greenhouse Gas R&D Programme, Cheltenham, UK. Nieminen M et al. (2009) ‘Application of Carbon Capture and Storage in Finland’, in: ClimBus – Business Opportunities in the Mitigation of Climate Change 2004–2008, Programme Report 4/2009/Final report, Finnish Funding Agency for Technology and Innovation Tekes, pp. 324–335. O’Connor WK, Dahlin DC, Rush SJ, Gerdemann SJ and Penner LR (2004) Energy and economic considerations for ex-situ aqueous mineral carbonation, DOE/ARC-2004028, U.S. Department of Energy, Albany Research Center, Albany, OR. O’Connor WK, Dahlin DC, Rush GE, Gerdemann SJ, Penner LR and Nilsen RP (2005) Aqueous mineral carbonation: Mineral availability, pretreatment, reaction parametrics, and process studies, DOE/ARC-TR-04–002, Albany Research Center, Albany, NY. Oelkers EH, Gislason SR and Matter J (2008) ‘Mineral carbonation of CO2’, Elements, 4, 333–337.


458


Park A-HA (2008) ‘Enabling sustainable fossil fuel energy conversion systems: CO2 and SO2 mineral sequestration and utilization of solid by-products’, 2nd International Conference on Accelerated Carbonation for Environmental and Materials Engineering (ACEME-08), Rome, Italy, 1–3 October. Park AA and Fan L (2004) ‘CO2 mineral sequestration: Physically activated dissolution of serpentine and pH swing process’, Chem. Eng. Sci., 59, 5241–5247. Plasynski S, Lang DA and Chizmeshya A (2008) A new approach to mineral carbonation: enhancing carbonation while avoiding mineral pre-treatment process cost. U.S. DOE Project Facts, Carbon Sequestration, project 314, 4/(2008). Prigiobbe V, Polettini A and Baciocchi R (2008) ‘Gas-solid carbonation kinetics of air pollution control residues for CO2 storage’, Chem. Eng. J., 148(2–3), 270–278. Prigiobbe V, Hänchen M, Costa G, Baciocchi R and Mazzotti M (2009a) ‘Analysis of the effect of temperature, pH, CO2 pressure and salinity on the olivine dissolution kinetics’, in Gale J, Herzog H and Braitsch J (eds), Greenhouse Gas Control Technologies 9, Proceedings of the Ninth International Conference on Greenhouse Gas Control Technologies (GHGT9), Energy Procedia, 1, 4881–4884. Prigiobbe V, Hänchen M, Werner M, Baciocchi R and Mazzotti M (2009b) ‘Mineral carbonation process for CO2 sequestration’, in Gale J, Herzog H and Braitsch J (eds), Greenhouse Gas Control Technologies 9, Proceedings of the Ninth International Conference on Greenhouse Gas Control Technologies (GHGT9), Energy Procedia, 1, 4885–4890. Rau GH, Knauss KG, Langer WH and Caldeira K (2007) ‘Reducing energy-related CO2 emissions using accelerated weathering of limestone’, Energy, 32, 1471–1477. Rinne J (2008) CO2 sequestration potential of industrial by-products and waste materials in Finland. MSc (Eng) thesis, Helsinki University of Technology, Espoo (Finland). Rubin ES, Rao AB and Chen C (2005) ‘Comparative assessments of fossil fuel power plants with CO2 capture and storage’, in Rubin E S, Keith D W and Gilboy C F (eds), Proceedings of the Seventh International Conference on Greenhouse Gas Control Technologies: GHGT7, IEA GHG, Cheltenham, UK, Vol. 1, 285–293. Schaef HT, McGrail BP and Owen AT (2009) ‘Basalt–CO2–H2O interactions and variability in carbonate mineralization’, in Gale J, Herzog H and Braitsch J (eds), Greenhouse Gas Control Technologies 9, Proceedings of the Ninth International Conference on Greenhouse Gas Control Technologies (GHGT9), Energy Procedia, 1, 4899–4906. Seifritz W (1990) ‘CO2 disposal by means of silicates’, Nature, 345, 486. Sipilä J, Teir S and Zevenhoven R (2008) Carbon dioxide sequestration by mineral carbonation – Literature Review Update 2005–2007, Åbo Akademi University, Heat Engineering Lab., report VT-2008-1, available at: http://web.abo.fi/~rzevenho/Miner alCarbonationLiteratureReview05–07.pdf (accessed January 2010). Soong, Y, Fauth DL, Howard BH, Jones JR, Harrison DK, Goodman AL, Gray ML and Frommell EA (2006) ‘CO2 sequestration with brine solution and fly ashes’, Energy Convers. Manage., 47, 1676–1685. Stasiulaitiene I, Denafas G and Sliaupa S (2007) ‘Lithuanian serpentines as potential carbon dioxide binders’, Proceedings 32nd International Technical Conference on Coal Utilization & Fuel Systems, 10–15 June, Clearwater, FL (paper 43). Stolaroff JK, Lowry GV and Keith DW (2005) ‘Using CaO- and MgO-rich industrial waste streams for carbon sequestration’, Energy Convers. Manage., 46, 687–699. Szargut J, Morris D and Steward FR (1988) Exergy analysis of thermal, chemical and metallurgical processes, New York, Hemisphere Publishing Co. Teir S (2008) Fixation of carbon dioxide by producing carbonations from minerals and



459

steelmaking slags. PhD (Eng) thesis, Helsinki University, of Technology, Espoo (Finland), available at: http://lib.tkk.fi/Diss/2008/isbn9789512293537/ (accessed January 2010). Teir S, Eloneva S and Zevenhoven R (2005) ‘Production of precipitated calcium carbonate from calcium silicates and carbon dioxide’, Energy Convers. Manage., 46, 2954–2979. Teir S, Eloneva S, Fogelholm C and Zevenhoven R (2006) ‘Stability of calcium carbonate and magnesium carbonate in rainwater and nitric acid solutions’, Energy Convers. Manage., 47, 3059–3068. Teir S, Revitzer H, Eloneva S, Fogelholm C-J and Zevenhoven R (2007a) ‘Dissolution of natural serpentinite in mineral and organic acids’, Int. J. Miner. Process, 83, 36–46. Teir S, Kuusik R, Fogelholm C-J and Zevenhoven R (2007b) ‘Production of magnesium carbonates from serpentinite for long-term storage of CO2’, Int. J. Miner. Process, 85, 1–15. Teir S, Eloneva S, Fogelholm C-J and Zevenhoven R (2009) ‘Fixation of carbon dioxide by producing hydromagnesite from serpentinite’, Appl. Energy, 86, 214–218. Thambimuthu K (2009) in: Conference Summary for the 9th International Conference on Greenhouse Gas Control Technologies (GHGT9), Washington DC, 16–20 November 2008. Ueda A, Kuroda Y, Yamada Y, Xue Z and Matsuoka T (2008) ‘Experimental study of carbonate mineralization by interaction of plagioclase (rock) – CO2 dissolved water around 100 °C’, 33rd Int. Geological Conference (IGC-33), Oslo (Norway), August 6–14. Uibu M (2008) Abatement of CO2 emissions in Estonian oil shale-based power production, PhD (Eng) thesis, Tallinn University of Technology, Tallinn, Estonia. Yogo K, Teng Y, Yashima T and Yamade K (2005) ‘Development of a new CO2 fixation/ utilization process (1): Recovery of calcium from steelmaking slag and chemical fixation of carbon dioxide by carbonation reaction’, in Wilson M, Morris T, Gale J and Thambimuthu K (eds), Proceedings of the Seventh International Conference on Greenhouse Gas Control Technologies: GHGT7, IEA GHG, Cheltenham, UK, Vol. 2, 2427–2430. Zevenhoven R and Kavaliauskaite I (2004) ‘Mineral carbonation for long-term CO2 storage: an exergy analysis’, Int. J. Thermodyn., 7, 22–31. Zevenhoven R and Tier S (2004) ‘Long term storage of CO2 as magnesium carbonate in Finland’, Proceedings Third Annual Conference on Carbon Capture and Sequestration, May 3–6, Alexandria (VA), USA (paper 217). Zevenhoven R, Eloneva S and Teir S (2006) ‘Chemical fixation of CO2 in carbonates: Routes to valuable products and long-term storage’, Catal. Today, 115, 73–79. Zevenhoven R, Eloneva S and Teir S (2007) ‘A study on MgO-based mineral carbonation kinetics using pressurised thermogravimetric analysis’, in Gale J, Rokke N, Zweigel P and Svenson H (eds), Proceedings of the Eighth International Conference on Greenhouse Gas Control Technologies: GHGT8, Elsevier, Oxford, UK, paper P02_01_09, CD–ROM. Zevenhoven R,Teir S and Eloneva S (2008a) ‘Heat optimisation of a two-stage gas/solid mineral carbonation process for long-term CO2 storage’, Energy, 33(2), 362–370. Zevenhoven R, Sipilä J and Teir S (2008b) ‘Motivations for carbonating magnesium silicates using a gas-solid process route’, Proceedings 2nd Int. Conf. on Accelerated Carbonation for Environmental and Materials Engineering (ACEME-08), Rome, Italy, 1–3 October, 45–54. © Woodhead Publishing Limited, 2010

460


Appendix: Energy efficiency of mineral carbonation processes The efficiency of fixation of CO2 by mineral carbonation (and also other long-term CO2 storage methods) can be assessed by considering the use of power and heat by the process. After all, the power and heat used are presumably produced by fossil energy conversion in a combustion process which will result in the production of CO2. Huijgen suggested a simple model expression for an ‘energetic CO2 sequestration efficiency’: CO 2 avoided ¥ 100 % CO 2 sequestered Epower e power Eheat e heat = 100 [16A.1] · 2 + e · 0– ¥ 100 % CO sequestered where Epower and Eheat are the amounts of power and heat used in the carbonation process (kWh) and epower and eheat give the amounts of CO2 produced when generating the necessary amounts of heat (kg CO2/kWh). Following Rubin et al. (2005), the value for epower was taken to be 0.6 kg CO2/kWh (an average value for a natural gas-fired combined cycle system, 0.36 kg CO2/kWh, and a condensing pulverised coal-fired power plant, 0.8 kg CO2/kWh). For eheat, the value 0.2 kg CO2/kWh (based on methane combustion) was used. Using this, Huijgen et al. (2006b) and Huijgen (2007) found efficiencies of around to 75 % for wollastonite carbonation and around 69 % for steel slag carbonation, respectively, using an aqueous process. A more general assessment can be made when considering process heat in more detail. Besides power consumption Epower (which may in theory be < 0 if excess process heat can be used to generate power), the process heat streams can be separated into ingoing heat streams Qin, outgoing heat streams Qout and heat losses Qlosses, each at a given temperature. Exergy analysis (based on the Second Law of Thermodynamics) allows for calculating the maximum power that can be produced from a certain amount of heat, Q, at temperature, T, at temperature of the surroundings, T0, by calculating its so-called exergy:

h(CO 2 ) =

Ê Tˆ Ex(Q ) = Q · Á1 – 0 ˜ Ë T¯

[16A.2]

where the temperatures must be taken in K. This quantifies how much ‘useful energy’ (i.e. work) can be obtained by converting heat into power, taking into account the quality of the heat as given by its temperature with respect to the surrounding environment (see e.g. Szargut et al., 1988). For power, P, the exergy is equal to the energy: Ex(P) = P. For a carbonation process with ni ingoing heat streams Qin,i at temperatures



461

Tin,i, nj outgoing heat streams Qout,j at temperatures Tout,j, and nk heat loss streams Qloss,k at temperatures Tloss,k, a more general expression for h(CO2) can be given, using exergies only: h(CO 2 ) =

CO 2 avoided ¥ 100 % CO 2 sequestered

Ê ˆ È Ê T ˆ˘ x j ·Qout,j · Á1 – 0 ˜ ˙ ÍEpower – ∑ Á ˜ T j Ë out,j ¯ ˚ Î Á ˜ Á ˜ È Ê ˆ Ê ˆ T T Á ˜ + Í∑ Qin,j · Á1– 0 ˜ – ∑ (1 – x j )Qout,j · Á1 – 0 ˜ Á ˜ Ë Tin,j ¯ j Ë Tout,j ¯ Îj Á ˜ Á ˜ Ê T0 ˆ ˘ + ∑ Qloss,k · Á1 – Á ˜ Tloss,k ˜¯ ˙˚ k Ë · e power ˜ ¥ 100 % = Á1 – CO 2 sequestered Ë ¯

[16A.3]

where the parameter 0 ≤ xj ≤ 1 gives the fraction of an outgoing heat stream Qout,j that is converted into power. In the study by Huijgen, no heat losses were considered (nk = 0), one ingoing heat stream (ni = 1, preheating of mineral to carbonation temperature) and one outgoing heat stream (nj = 1, reaction heat) were considered, and no excess heat was converted into power (all xj = 0). This reduces the above equation to CO 2 avoided h(CO 2 ) = ¥ 100 % CO 2 sequestered T0 ˆ Ê T Ê Epower Qin Á Qout 1 0 ˜ ˆ Ê ¯ Ë T È Ê T ˆ˘ + Í · 1 – in – ·Á – out ˜ ˙ e power ˜ ¥ 100 % = Á1 – ¯˚ Ë Î 2 ˜˜ · Á – ˜ CO sequestered Á ÁË ¯ [16A.4] where Tout of Qout is the temperature of the carbonation reactor TR, and Tin for the heat used in the preheater should be slightly higher than that. With Tout = TR ≈ Tin, the expression simplifies to

h(CO 2 ) =

CO 2 avoided ¥ 100 % CO 2 sequestered

ˆ Ê È T0 ˆ ˘ Ê Epower Í Qin Qout Á ˜¯ ˙ ˜ Á Ë T + Î( – ) · 1 – R ˚ · e power ˜ ¥ 100 % = Á1 – 2 sequestered ˜ Á – ˜¯ ÁË CO [16A.5] © Woodhead Publishing Limited, 2010

462


This is identical to the original expression by Huijgen for net heat input Eheat = Qin–Qout, and eheat = (1 – T0/TR) · epower. For the values eheat = 0.2 kg CO2/ kWh and epower = 0.6 kg CO2/kWh and carbonation reaction temperature TR = 150–200 °C, this gives T0 = 9–42 °C for the temperature of the environment, showing that the estimate of eheat for the given epower is rather good. The power use was related to grinding the material to be carbonated to the appropriate particle size, compression of the CO2 and pumping slurries. Power consumptions related to the reactor and for filtration were neglected. Using the relation eheat = epower · (1–T0/T) in (Equation 16A.5) gives Equation 16A.1 suggested by Huijgen, but it must be noted that for heat streams Eheat of different temperature the factors eheat are different. Applying this concept to a comparison of olivine (requiring significant pre-grinding) and serpentine (requiring heat treatment) gives an economic assessment outcome quite different from what was reported for the ARC study mentioned above.


17

Photocatalytic reduction of carbon dioxide (CO2) J e f f r e y C. S. W u, Department of Chemical Engineering, National Taiwan University, Taiwan Abstract: One of the best means of solving the carbon dioxide (CO2) problem is to photocatalytically convert the CO2 to hydrocarbons using solar energy, a process which simultaneously generates renewable energy. This chapter discusses the photocatalytic reduction of CO2 using either artificial light or sunlight and presents the fundamental mechanism of CO 2 photoreduction. An optical-fiber reactor has been designed that transmits and spreads light uniformly inside the reactor. Methanol is the main product from Cu- or Ag-loaded TiO2, and methane and ethylene are produced from Cu–Fe-loaded TiO2–SiO2 photocatalysts. The photoactivity of dye-sensitized TiO2 is significantly enhanced because the dye can harvest full sunlight. The transformation of solar to chemical energy via CO2 conversion is similar to photosynthesis. Key words: CO2, photocatalytic reduction, titania, optical-fiber reactor, renewable energy.

17.1

Introduction

Greenhouse gases such as carbon dioxide (CO 2), methane (CH 4) and chlorofluorocarbons (CFCs) are the primary cause of global warming. CO2 is released mainly by the burning of fossil fuels and, over the past few decades, the atmospheric concentration of CO2 has been increasing due to an increase in human activity, which has consequently accelerated the greenhouse effect. The consumption of fossil fuels worldwide continues to increase year after year because of people’s great demand for energy. The physical or chemical separation/storage of CO2 can solve this problem only temporarily. In addition, the conventional chemical process for CO2 reduction requires energy input. Therefore, converting CO2 to valuable hydrocarbons is one of the best solutions to both the global warming and the energy shortage problems. Figure 17.1 depicts the energy cycle using carbon as a carrier (Wu, 2009). All life is sustained by the utilization of solar energy via photosynthesis, and fossil fuels also act as storage for solar energy received in the past. In the natural world, CO2 is removed from the atmosphere via photosynthesis. The energy obtained from sunlight is used to convert CO2 into glucose – a sugar molecule that stores solar energy in the form of chemical energy. Except for geothermal or nuclear energy, most energy forms, such as fossil fuels, biomaterials, hydropower, wind, etc., are simply the transformations 463 © Woodhead Publishing Limited, 2010

464

Developments and innovation in CCS technology Energy

Combustion

Fossil fuel Coal, oil, gas

Chemical products

Biospheric circle

CO2 H 2O sunlight

Photo synthesis

Biomass

17.1 Energy cycle using carbon as the energy carrier (adapted from Wu, 2009).

of solar energy in either the past or the present. Thus the sun is the earth’s ultimate energy supply. A promising application for this is the implementation of artificial photosynthesis via the photoreduction of CO 2 to produce hydrocarbons. In other words, solar energy is directly transformed and stored as chemical energy. Consequently, the photoreduction of CO2 with H2O to produce chemicals, such as methane or methanol (CH3OH), is particularly interesting, and there is a great desire to achieve a high efficiency for this reaction. Methanol can be easily transported, stored and used as a gasoline additive for automobiles. Moreover, methanol, methane and ethylene can be easily transformed into other useful chemicals using existing chemical technologies. The energy grade of CO2 is low from a thermodynamic perspective, which is why any transformation to hydrocarbons requires the input of energy. Equation 17.1 gives an example of the complete photoreduction of CO 2 to form methanol. Based on thermodynamics, converting one mole of CO2 into methanol requires 228 kJ (DH) of energy at 298 K. Furthermore, the Gibbs free energy of Equation 17.1 is 698.7 kJ (DG) at 298 K, indicating that the equilibrium is highly unfavorable to the product, methanol and oxygen. That is, the process is energy intensive and extremely difficult. The energy for photocatalytic reduction of CO2 should be provided without producing more CO2. In nature, plants use solar energy to perform photosynthesis, but the efficiency of the energy transformation is low because part of the energy is consumed to support their lives. Gust et al. indicated that the evolution of photosynthesis was not driven by the need for maximally efficient energy storage (Gust et al., 2008). The energy conversion efficiency of natural photosynthesis is relatively low, with a maximum of about 6 %, but is usually observed to be < 0.8 %. Besides energy dissipation through leaves, energy loss is mainly due to the energy needed for the growth and maintenance of



465

the plant. Even under optimal artificial conditions, energy efficiency can only reach about 7 % in macroalga in full sunlight (Laws and Berning, 1991).

hn CO 2 + 2 H 2 O æÆ æ CH 3OH + 3 O 2 2

[17.1]

Photocatalysis uses semiconductor materials to drive reactions in the presence of light radiation. Semiconductors exhibit a bandgap, in which an electron–hole pair (e––h+) is generated under light radiation. The photocatalytic process occurs via the direct absorption of photons with energy higher than or equal to the bandgap to generate electron–hole pairs. The initial excitation and electron transfer can trigger subsequent chemical reactions in the photocatalytic process. Many researchers have shown that CO2 can be reduced in water vapor or a solvent by photocatalysts. Titanium oxide (TiO2) is a well-known photocatalyst and has been applied to the reduction of CO2 with H2O to produce CH4 and CH3OH (Anpo et al., 1995; Yamashita et al., 1998; Tseng et al., 2002; Dey et al., 2004; Pathak et al., 2004; Lo et al., 2007). Recently, Guan et al. reported the photoreduction of CO2 with H2O under concentrated sunlight using K2Ti6O13-based photocatalysts (Guan et al., 2003a,b). The selective photoreduction of CO2 into methanol was studied using ZnO and NiO under UV laser irradiation (Yahaya et al., 2004). NiO/InTaO4, an efficient visible-light photocatalyst, was shown to be able to reduce CO2 into methanol (Pan and Chen, 2007; Chen et al., 2008). Organic sensitizers, capable of generating electrons under illumination, were used to initiate CO2 photoreduction on TiO2 thin and thick film surfaces (Hirose et al., 2003; Ozcan et al., 2007a,b; Nguyen et al., 2008). Significantly improved photoconversion of CO2 was reported using nanoscale TiO2 particles embedded in a silver-coated Nafion® membrane film (Pathak et al., 2005). Ru–TiO2/ SiO2 was prepared by the solid-state dispersion method and used for the photoreduction of CO2 in an aqueous medium under ambient conditions (Sasirekha et al., 2006). The CO2 photoreduction of zinc phthalocyanineloaded TiO2 was markedly increased because of the reduced recombination probability for hole–electron pairs (Zhao et al., 2007). Although the various catalysts exhibited different levels of efficiency for photo-energy conversion, much effort is still required to make this process commercially viable. The goal is to demonstrate that artificial photosynthesis can be implemented via the photoreduction of CO2 to produce fuel, such as methanol or methane; in other words, to transform solar energy and store it as renewable energy.

17.2

Fundamentals of photocatalysis

In general, a heterogeneous catalytic reaction consists of five essential steps: (i) diffusion of the reactant to the catalyst surface; (ii) adsorption of


466


reactant(s); (iii) chemical reaction; (iv) desorption of product(s); and (v) diffusion of product(s) away from the catalyst. The activation energy of a reaction can be significantly reduced during the adsorption process with the help of a catalyst, thus accelerating the reaction rate. In a thermal reaction, the driving force is thermal energy, i.e., temperature, while in a photocatalytic reaction, the driving force is photo-energy, i.e. light.

17.2.1 Photocatalysis mechanism In general, a photocatalyst is a semiconductor material that consists of valence and conduction bands. The energy difference between the two bands is referred to as the bandgap, usually in the range 1.8–3.5 eV. The bandgap position of some semiconductors and redox couples in aqueous solution is shown in Fig. 17.2 (Inoue et al., 1979). The position of the conduction band in a semiconductor such as TiO2 indicates that methanol can be formed from CO2 photoreduction under UV light (365 nm equivalent to ~ 3.2 eV) in an aqueous solution. Figure 17.3 shows the basic steps of a photocatalytic reaction on a TiO2 photocatalyst. An electron–hole pair is generated on the photocatalyst under light irradiation. The photo-energy must overcome the bandgap of the semiconductor in order to excite the electron on the valence band to the conduction band. However, the electrons and holes may recombine to release their energy as heat and only a small amount of them might migrate to the surface to drive chemical reactions. Titanium oxide is one of the best photocatalysts in many respects. It is relatively inexpensive, non-toxic and highly stable chemically. As shown in Equations 17.2–17.8, TiO2 generates electron–hole pairs under UV irradiation Conduction band –2.0 –1.0 Potential 0.0 (V vs NHE) +1.0

SiC 3.0 eV

GaP CdS ZnO TiO 2 WO3SnO

2.3 eV 2.4 eV

3.2 eV

+2.0 +3.0

2

3.0 eV 2.8 eV

Valence band

3.8 eV

HCOOH/H2CO3 HCHO/H2CO3 H2/H2O CH3OH/H2CO3 CH4/CO2 O2/H2O Redox system at pH = 5.0

Semiconductors

17.2 Bandgap of semiconductors and redox couples in aqueous solution (adapted from Inoue et al., 1979).


Photocatalytic reduction of carbon dioxide (CO2) Metal

467

CO2 H+



hn CH3OH TiO2 OH–



O2

17.3 Schematics of a photoreaction on photocatalyst (adapted from Tseng et al., 2002).

(Equation 17.2). The photogenerated holes and electrons are capable of oxidation and reduction, respectively. Surface-adsorbed water molecules are dissociated into H+ and OH– ions at chemical equilibrium (Equation 17.3). The photogenerated electron combines with H+ to form adsorbed hydrogen (Equation 17.4). The hydroxide ion, OH–, combines with a photogenerated hole to form an OH radical (Equation 17.5). Subsequently, two OH radicals may combine to form H2O2 (Equation 17.6). H2O2 may receive a hole to produce an oxygen ion (O2–) and two H+ (Equation 17.7). One oxygen molecule may be released after O2– donates an electron to a hole (Equation 17.8), then leaves an oxygen vacancy on the TiO2 surface (Anpo et al., 1984; Inoue et al., 1979; Howe and Gratzel, 1987; Serpone et al., 1994). hn

TiO 2 æÆ æ e– + h+

[17.2]

H2O Æ H+ + OH–

[17.3]

H H +e ÆH Æ Ω TiO 2

[17.4]

+

–

OH– + h+ Æ OH ∑

2OH ∑ Æ H2O2 +

O2–

H 2O 2 + h Æ

O2– + h+ Æ O2

[17.5] [17.6]

+

+ 2H


[17.7] [17.8]

468


The formation of oxygen vacancies has an important role in photocatalysis. Fujishima et al. noted that under UV irradiation, surface oxygen atoms would be oxidized by the photogenerated holes to give oxygen molecules and then leave oxygen vacancies (Fujishima et al., 2000). H2O molecules could then be adsorbed on the oxygen vacancies and split into OH groups. Thus the surface of TiO2 becomes super-hydrophilic, which is the origin of the self-cleaning mechanism in photocatalysis.

17.2.2 Method for carbon dioxide (CO2) photoreduction to hydrocarbons CO2 can be reduced to hydrocarbons by electron transfer and by hydrogen ions under light irradiation in an aqueous solution. This can be done in four ways: direct photolysis, photochemical reduction, electrochemical reduction and photocatalytic reduction. Direct photolysis of CO2 requires high-energy photons. Getoff et al. showed that formaldehyde, acetaldehyde and formic acid were directly produced in a CO2 aqueous solution under 60Co g-ray irradiation without a catalyst (Getoff et al., 1960). The yields depended on the radiation dose and the pH value of the solution. For photochemical reduction, a sacrificial agent is required in the photochemical conversion of CO 2. Ogata et al. used p-terphenyl as a photosensitizer and a tertiary amine as a sacrificial electron donor to photoreduce CO2 to formic acid with cobalt macrocycles in methanolic acetonitrile solutions (Ogata et al., 1995). The efficiency was further enhanced by 24 % by using phenazine as a photosensitizer (Yanagida et al., 1995). Electrical energy is needed to reduce CO2 in an electrochemical process. Hydrogen is supplied from H2O by electrolysis. Mishimura et al. carried out CO2 electrochemical reduction with an electrode/K2Ti6O13 PTFE/solid polymer electrolyte system without a liquid electrolyte. Current efficiency reached up to 90 % for CO production on an Au electrode (Nishimura et al., 1995). The Faradaic efficiencies for methanol, formic acid and methane from CO2 in the electrochemical reduction system achieved 18 % in a KHCO3 solution using a Cu tube as an electrode (Ohta et al., 1995). The well-known photocatalytic (or photoelectrocatalytic) reduction of CO2 was demonstrated by Inoue et al. (1979). Formic acid, formaldehyde and methanol were produced using a series of photocatalysts, including WO3, TiO2, ZnO, CdS, GaP and SiC, in an aqueous solution under light illumination. In summary: direct photolysis may not be practicable because highenergy photons are not easily available. The sacrificial agent is consumed in a photochemical process. Electricity is required for the electrochemical



469

method. Therefore, the photocatalytic reduction of CO2 is clearly superior to these methods because energy can be supplied by unlimited sunlight.

17.2.3 Mechanism of photoreduction of carbon dioxide (CO2) on titanium dioxide (TiO2) When CO2 is adsorbed on TiO2, UV irradiation can produce bicarbonate and carbonate from the adsorbed CO2 as shown in Fig. 17.4. The photogenerated OH group and oxygen vacancy can further increase the amount of bicarbonate and carbonate from gas-phase CO2. However, once the bicarbonate or carbonate is formed, it becomes very stable on the TiO2 surface and no further conversion is possible. Only the linearly adsorbed CO2 on the TiO2 can be further converted to other intermediates. Figure 17.5 illustrates the mechanism of linearly adsorbed CO2, which is further photoreduced to hydrocarbons under UV irradiation. Starting from the adsorbed CO2, formate (HCOO) is formed by accepting an electron and adding one hydrogen. Dioxymethylene (H2COO) is formed from the formate by adding another hydrogen, and then migrates to the adjacent oxygen vacancy to form formaldehyde (H2CO) by accepting one electron. In this step, one oxygen detaches from the dioxymethylene and stays on the previous TiO2 site. Methoxy (CH3O) is then formed by adding another hydrogen. Subsequently, the methoxy reacts with free surface H2O and is converted to methanol. Finally, methanol is desorbed from the surface and leaves one OH group on the TiO2 surface. The reduction of CO2 to methanol requires the lowering of the oxidation state of carbon from C (4+) to C (2–). Six electrons are required for each methanol molecule formation. Overall, three hydrogen atoms and three

CO2 CO2

H

Ti

CO2

O

O Ti

Ti

(1)

(2)

O

–

O Ti

O C

C

C O

Ti

Ti (3)

O

OH

–

Ti

O Ti

–

O Ti

Ti

17.4 Proposed mechanism of CO2 adsorption on TiO2.


470

Developments and innovation in CCS technology O

O

C O Ti

O

C O

H Ti(O2)

H

Ti

e–

C Ti(O2)

Ti

e

O

H

H

O O

H

O –

H C

H

Ti

Ti

e–

Ti

O Ti

Ti

H H

H

Ti(O2)

C

H

O Ti

H

+H2O

Ti

CH3OH

O Ti

Ti

17.5 Proposed mechanism of photocatalytic CO2 reduction on TiO2.

electrons progressively transfer to one linearly adsorbed CO2, resulting in the formation of one methoxy molecule. In combination with one water molecule, the methoxy molecule is eventually converted to methanol. Six holes acquire the electrons one by one from OH– and H2O2, resulting in the production of 1.5 O2 as shown in Equations 17.3, 17.5–17.8 and 17.1. In summary, two H2O molecules are consumed. The photoreduction can be significantly enhanced by transition metals on TiO2. Transition metals, such as Cu, Ag and Pt, can serve as electron traps to prevent the recombination of the electron and hole, as shown schematically in Fig. 17.3. Schilke et al. suggested that a hydrogen atom could also be supplied from the adsorbed hydrogen on the surface of Cu particles by spillover (Schilke et al., 1999). It is possible that a hydrogen atom may migrate onto Cu via the reduction of H+ with a trapped electron. The H–Cu can also enhance hydrogenation of the intermediate species including formate, dioxymethylene and formaldehyde.

17.3

Renewable energy from photocatalytic reduction of carbon dioxide (CO2)

17.3.1 The photoreduction of carbon dioxide (CO2) in aqueous solution; experimental Preparation of powder photocatalyst Sol–gel, spread pyrolysis, coprecipitation and solid fusion are common methods of photocatalyst preparation, but the sol–gel procedure is one of



471

best methods to use on a laboratory scale. It consists of simple steps that are versatile enough to allow modifications for controlling particle size, surface area, heteroatom doping and metal dispersion. Thus the sol–gel technique was chosen to synthesize the photocatalysts. Photocatalysts were prepared using the sol–gel method, as illustrated in Fig. 17.6 (Tseng et al., 2002). The precursor was titanium (IV) butoxide [Ti(OC4H9)4, 97 % in n-butanol] purchased from Aldrich (USA). The hydrolysis process was performed in a glove box maintained at a relative humidity below 25 % by purging with tank nitrogen. To avoid rapid precipitation during polycondensation and the formation of unstable colloidal sols, the hydrolyzing water was homogeneously released by the esterification of butanol and acetic acid (Wu and Yeh, 2001). The intention was to provide the appropriate stoichiometric quantity of water to hydrolyze titanium butoxide during hydrolysis. A typical batch contained 0.02 mole titanium butoxide, 0.08 mole anhydrous butanol (min. 99.8 %) and 0.08 mole glacial acetic acid (min. 99.7 %). The clear solution was stirred for eight hours, at which point the pH value stabilized. The final pH value of the solution was nearly 3.56. The transparent sol was dried from room temperature to 150 °C in an oven, then transferred to a furnace and calcined at 500 °C to burn off hydrocarbons. The sample was crushed to a powder in a mortar. JRC-2 and P25 titania powders were obtained from Fuji Titan (Japan) and Degussa (Germany),

0.02 mole Ti(OC4H9)4 +0.08 mole n-C4H9OH +0.08 mole CH3COOH Stirring 8 hr Stirring 8 hr

TiO2 (Degussa P25) Incipient wetness Adding CuCl2 (or AgNO3)

Stirring 0.5 hr

Drying 3 K/min to 423 K maintain 2 hr, then 5 K/min to 773 K maintain 0.5 hr

Reduced at 573 K for 3 hr (5 % H2/Ar, 30 ml/min)

TiO2

Cu/TiO2

Cu/P25

17.6 Synthesis procedure of the powder photocatalysts (adapted from Tseng et al., 2002).


472


respectively, for comparison. Cu-loaded titania (Cu/TiO2) was impregnated by adding CuCl2 during the sol–gel process and incipient wetness method, respectively, followed by calcination similar to that of TiO 2. The crystalline phase was identified by an X-ray diffractometer, MAC Science M03XHF, scanning from 20–80°. The specific surface areas of the photocatalysts were measured by N2 adsorption in a Micromeritics ASAP 2000. A diffusive reflective UV-Vis spectrophotometer (Varian Cary 100) was employed to measure the UV-Vis absorption at wavelengths from 200– 800 nm. The particle size distribution in aqueous solution was obtained using a Coulter LS230 particle size analyzer. Transmission electron microscopy (TEM) and scanning electron microscopy (SEM) with energy dispersive X-ray spectrometer (SEM–EDS, Philips XL30, EDAX DX4) were used to observe the morphology of the photocatalysts and estimate the elemental ratio. The chemical status of the photocatalyst surface was analyzed by X-ray photoelectron spectroscopy (XPS). The XPS was conducted on a spectrometer of VG Microtech MT500 with Mg–Ka radiation as the excitation source (hn = 1253.6 eV). All binding energies were referenced to oxygen (1s) at 530.7eV (Bhattacharya et al., 1997) and carbon (1s) at 285.6eV. Photoreduction of CO2 in aqueous solution Figure 17.7 shows a schematic illustration of a batch photoreactor (Tseng et al., 2002). The system was illuminated by an 8 W xenon lamp with a peak light intensity at 254 nm in the center of the quartz reactor. The entire system was shielded by a metal case during the reaction to prevent interference from ambient light. Photocatalyst powder (0.15 ~ 0.6 g) was suspended in 300 ml of 0.2 N NaOH aqueous solutions for typical batches. Supercritical-fluid grade CO2 with certified maximum of hydrocarbons less than 20 ppb was purchased from Air Products (USA) to avoid any hydrocarbon contamination. CO2 was bubbled through the reactor for at least 30 min to purge air and to saturate the solution. The reactor was tightly closed during the reaction, and the CO2 pressure was maintained in the range 101.3~135.6 kPa. A magnetic stirrer agitated the photocatalyst-suspended solution at the bottom to prevent sedimentation of the photocatalyst. The steady-state temperature of the solution rose to almost 50 °C during illumination. The release of O2 during the reaction would provide direct evidence of CO2 photoreduction (Equation 17.1). An oxygen sensor (Mettler Toledo InPro® 6000 series) was placed in the reactor to monitor the concentration of dissolved O2 during the photoreduction. The sensor was calibrated using the dual point mode before measuring O2 concentration. The O2 sensor was inserted into oxygen-free gel until the ‘ready’ sign came on, to fix the zero point. The sensor was then put into the O2-saturated solution, until the ‘ready’ came on and the value was tuned to a verified concentration (8.2 ppm,



473

Sampling syringe

O2 sensor

temperature/pH meter

CO2

Acrynitrile cover

Quartz reactor

UV Lamp

Suspended catalyst insolution

Lumen meter

Magnetic stirrer

17.7 Schematics of photoreactor in aqueous solution (adapted from Tseng et al., 2002).

25 °C). A needle-type probe was inserted in the reactor to draw samples. The liquid sample (< 0.5 ml) was collected in a vial wrapped in aluminum foil to reduce interference from the indoor fluorescent light before analysis. After the catalyst’s sedimentation, 1~10 ml liquid sample was withdrawn and analyzed by a gas chromatography/flame ionization detector equipped with a 3 m long Porapak Q column. Analysis results indicated that methanol was the dominant hydrocarbon. Formic acid, formaldehyde and ethanol were detected as well from some other catalytic reactions in amounts much lower than that of methanol. Blank reactions were conducted to ensure that the hydrocarbon production was purely from the photoreduction of CO2, without any interference from the surroundings. One blank was UV-illuminated without the photocatalyst, and another was performed in the dark with the photocatalyst and CO2 under the same experimental conditions. An additional blank was UV-illuminated with the photocatalyst in the presence of N2 rather than CO2. No hydrocarbon was detected in the three blank tests.

17.3.2 Results and discussion Characteristics of powder photocatalysts The XRD spectra in Fig. 17.8 verified the anatase phase of JRC-2 and a series of Cu/TiO2 (Tseng et al., 2002). A previous investigation had indicated that anatase TiO2 is the most active phase for photocatalytic reaction (Linsebigler


474


A

A: anatase

A

A Intensity

Cu

A

Cu

A

6.7 wt% Cu/TiO2

3.3 wt% Cu/TiO2

1.0 wt% Cu/TiO2

JRC-2

0

20

40

2q

60

80

100

17.8 XRD spectra of JRC-2 (pure TiO2) and Cu/TiO2 (adapted from Tseng et al., 2002).

et al., 1995). Two small Cu diffraction peaks appeared near 2q = 43.3 and 52° on 6.7 wt% Cu/TiO2. No Cu peak was observed on 1.0 or 3.3 wt% Cu/ TiO2, perhaps due to the low Cu loading or the extremely small Cu clusters. The grain sizes of all sol–gel-derived TiO2 were nearly 20 nm, as calculated from the Scherrer equation. The particle sizes were consistent with the TEM observations displayed in Fig. 17.9. As shown in Fig. 17.9(a), the particles of sol–gel-derived TiO2 were uniform and their diameter was between 10 and 25 nm. Figure 17.9(b) reveals that Cu clusters were well dispersed on the surface of TiO2. Cu clusters were expected to be present on the surface of the TiO2 particles because the CuCl2 was added after hydrolysis of titanium butoxide during the preparation (Fig. 17.6). Zhang et al. suggested that pure TiO2, with grain sizes ranging from 11–21 nm in diameter, possessed maximum photocatalytic efficiency (Zhang et al., 1998). When the particle size was less than 5−10 nm, the surface recombination of electron–hole pairs became significant, resulting in low photocatalytic efficiency. Figure 17.10 shows the actual particle-size distributions of the photocatalysts in an aqueous suspension. For TiO2 and 2 wt% Cu/TiO2, the median particle sizes were about 70 nm and 50 nm, respectively. P25 exhibited two modes of particle-size distribution, about 45 nm and 75 nm. Table 17.1 summarizes the properties of the photocatalysts. The bandgap



(a)

475

(b)

17.9 TEM photographs; (a) TiO2, (b) 3.3 wt% Cu/TiO2 (adapted from Tseng et al., 2002). 20 2.0 wt% Cu/TiO2 TiO2 TiO2 (P25)

Number (%)

15

10

5

0 40

60

80 100 Particle size (nm)

120

140

17.10 The particle size distributions of P25, TiO2 and Cu/TiO2 (adapted from Tseng et al., 2002).

can be estimated by extrapolating the rising portion of the UV spectrum to the abscissa at zero absorption (Sanchez and Lopez, 1995). The bandgaps of JRC-2 and P25 are 3.27 and 3.47 eV, respectively, while those of sol–gel TiO2 and Cu/TiO2 are nearly 3.00 eV. Notably, the bandgap is governed by the crystalline structure and the defects in the TiO2 network. A related © Woodhead Publishing Limited, 2010

476


Table 17.1 Characteristics of photocatalysts, methanol yields and energy conversion efficiency (adapted from Tseng et al., 2002) Photocatalyst

Surface Crystalline Bandgap2 area size1 (eV) (m2/g) (nm)

6 hr methanol Energy yield conversion (mmol/g-cat.) efficiency (%)3

TiO2 P254 JRC-2 1.0 wt% 2.0 wt% 3.3 wt% 6.7 wt%

63 504 163 40 28 26 28

4.7 38.2 3.4 37.7 72.9 118.5 20.0

Cu/TiO2 Cu/TiO2 Cu/TiO2 Cu/TiO2

18 21 42 20 20 20 17

2.95 3.47 3.27 2.99 2.99 2.92 3.05

0.09 0.81 0.07 1.54 2.50 1.89 –

1

Calculated from the Scherrer equation according to the peak broadening of XRD spectra (at 2q = 25.28°). 2 Estimated from the UV-VIS spectra extrapolating the absorption edge of UV-Vis spectrum to the abscissa of zero absorption. 3 The transform of photon to chemical energy, calculated by heat of combu ustion ¥ moles of methanol yield energy conversion efficiency (%) = ¥ 100 % radiation energy of UV absorbed by catalyst 4

From Degussa. Table 17.2 The element molar ratio of photocatalysts calculated from XPS and EDS analysis (adapted from Tseng et al., 2002) Photocatalyst

Cu/Ti

Bulk

EDS

XPS

0.012 0.02 0.04 0.09

– 0.035 0.053 0.091

0.063 0.119 0.138 0.165

1.0 2.0 3.3 6.7

wt% wt% wt% wt%

Cu/TiO2 Cu/TiO2 Cu/TiO2 Cu/TiO2

EDS = energy-dispersive X-ray spectroscopy; XPS = X-ray photoelectron spectroscopy.

investigation suggested that small bandgaps were caused by the stoichiometric deficiency of Ti/O from the sol–gel processes (Sanchez and Lopez, 1995). The specific surface area of sol–gel-derived photocatalysts ranged from 25–63 m2/g. The specific surface area of sol–gel TiO2 was larger than that of commercial JRC-2 or P25. Table 17.2 lists the elemental ratio of Cu/Ti estimated from XPS and EDS. The bulk ratio of Cu/Ti was the molar ratio of Cu and Ti in the sol–gel preparation (Fig. 17.6). The discrepancy between the Cu/Ti ratio obtained by EDS and XPS indicated that most Cu was on the surface of the TiO2 support. As the source radiation of these two kinds of probes have different



477

incident energies, i.e. X-ray (XPS) at 50 eV and electron (EDS) at 15 keV, this indicates that the elements were detected at different depths from the surface, i.e. ~10 nm and ~1 mm, respectively (Vickerman, 1997). Therefore, the quantitative results of XPS show the outmost surface of photocatalyst, while those of EDS give the deep structural layers and approximately represent the bulk property. Figure 17.11 shows the Ti (2p) XPS spectra of TiO2, P25, JRC-2 and 2 wt% Cu/TiO2. The binding energies of Ti 2p3/2 and 2p1/2 of 2 wt% Cu/TiO2 are the same as those of pure titania at 459.4 and 465.1 eV, respectively, indicating the integrity of the TiO2 structure, which was not modified by Cu impregnation. Figure 17.12 shows the results of Cu (2p) XPS spectra on Cu/ TiO2 with three different Cu loadings (Wu et al., 2005). From the positions of the binding energies (2p3/2, 932.8 eV; 2p1/2, 952.8 eV) and the shape of the peaks, the Cu on the surface of TiO2 might exist in multiple oxidation states with Cu(I) as the primary species (Tseng et al., 2004). Photoreduction of CO2 in aqueous solution Figure 17.13 shows the dependence of methanol formation on UV illumination time. Various yields of methanol were obtained over a certain period,

459.4 eV

Intensity (a.u.)

465.1 eV

P25

JRC–2

2 wt% Cu/TiO2 TiO2

450

455

460 465 Binding energy (eV)

470

475

17.11 XPS spectra of Ti 2p of photocatalysts (adapted from Tseng et al., 2002).


478

Developments and innovation in CCS technology 2p3/2

2p1/2

Intensity (a.u.)

2.06 wt% Cu/TiO2

1.20 wt% Cu/TiO2

0.52 wt% Cu/TiO2

920

930

940 950 Binding energy (eV)

960

970

17.12 XPS of Cu 2p on Cu/TiO2 photocatalysts (adapted from Wu et al., 2005).

depending on the photocatalysts used. In all cases, methanol was generated after nearly 180 min. The formation of methanol was found much more effective on supported Cu titania photocatalyst (Yamashita et al., 1994). Methane, formic acid and other hydrocarbons might have been generated, but in quantities too low to be detected. The results showed that the sol–gelderived Cu/TiO2 outperformed P25, JRC-2, and 2 wt% Cu/P25. Table 17.1 also lists the methanol yields of various photocatalysts after six hours of irradiation. The presence of Cu plays an important role, while the specific surface area is obviously not a major factor in photocatalytic reactions (Anpo et al., 1987). The methanol yields increase with Cu loading, but then decrease when the Cu loading exceeds 2 wt%. Excess Cu loading can mask the TiO2 surface, resulting in a reduction of the photo-exciting capacity of TiO2. The energy conversion efficiency evaluates the transformation of the photon’s energy into chemical energy. It provides a fair comparison of performance in CO2 photoreduction. The highest energy efficiency achieved 2.5 % on the 2.0 wt% Cu/TiO2 catalyst. Contact between TiO2 and metal generally involves a redistribution of electric charge. In the presence of Cu clusters, electrons are enriched owing



479

6.0 wt% Cu/TiO2 3.3 wt% Cu/TiO2 2.0 wt% Cu/TiO2 1.0 wt% Cu/TiO2 TiO2

100

Methanol yield (µmol/g-cat.)

TiO2 (P25) 2.0 wt% Cu/P25

50

0 0

100

200 300 Irradiation time (min)

400

17.13 Time dependence on the methanol yields of various photocatalysts (UV = 254 nm) (adapted from Tseng et al., 2002).

to the alignment of the Fermi levels of the metal and the semiconductor; in other words, the Schottky barrier (Linsebigler et al., 1995). The Cu then serves as an electron trapper and prohibits the recombination of holes and electrons. In addition, the rapid transfer of excited electrons to the Cu cluster increases the separation of holes and electrons (Hirano et al., 1992), significantly promoting the efficiency of photon conversion. The formation of methanol was more efficient than that of other hydrocarbons in the presence of supported Cu+ on TiO2 from the CO2 and H2O system (Yamashita et al., 1994). Oxygen is expected as a product of CO2 reduction (Equation 17.1). Figure 17.14 shows that the time dependency of oxygen generation approximately matches that of the methanol yield. It has been suggested that free O2 is adsorbed on the surface of TiO2 in the presence of water (Yamashita et al., 1994). Consequently, the oxygen sensor detected only a little dissolved oxygen in these experiments. The O2 concentration also plateaued after 16–20 hours of reaction, revealing possible O2 consumption by methanol re-oxidation. NaOH solution was found to be crucial in the photoreduction of CO2 in aqueous solution (Kaneco et al., 1998). The methanol yield substantially increased with the addition of NaOH in our experiments, due perhaps to


480

Developments and innovation in CCS technology 100 Methanol (run1) Methanol (run2) O2 (run1)

Products amount (mmol)

O2 (run2)

50

0 0

5

10 Irradiation time (h)

15

17.14 The dissolved O2 of two 2.0 wt% Cu/TiO2 photocatalysts during reaction (adapted from Tseng et al., 2002).

the following two reasons. First, the high concentration of OH– ions in aqueous solution could allow them to act as strong hole-scavengers to form OH radicals, thereby reducing the recombination of hole–electron pairs. The longer lifetime of the surface electrons would certainly facilitate the reduction of CO2. Second, caustic NaOH solution dissolves more CO2 than pure water does. The initial pH value of the 0.2 N NaOH solution was approximately 13.3, and decreased to nearly 7.3 after bubbling CO2 in the reactor. Notably, photoreduction may have been accelerated by the high concentration of HCO3–. Summary This study demonstrated the transformation of solar energy to chemical energy via the photoreduction of CO2. Methanol was favorably produced on Cu/ TiO2 photocatalysts under UV irradiation. Cu-loaded titania was an efficient photocatalyst for CO2 reduction since Cu is an effective electron trapper, capable of reducing the recombination of electron–hole pairs. Experimental results indicated that the homogeneous hydrolysis of titanium (IV) butoxide via the improved sol–gel method is a promising technique for preparing



481

photocatalysts with uniform nanoparticles. However, there are still many factors that require improvement.

17.3.3 The photoreduction of carbon dioxide (CO2) in the vapor phase Design of optical-fiber photoreactor Practical applications of using a photocatalyst for the photoreduction of CO2 in a continuous steady-state often require the photocatalyst to be immobilized in a photoreactor. However, a traditional packed-bed reactor with light irradiation from either the side or the center always projects a shadow on the other side of the photocatalyst particle, which makes part of the photocatalyst surface ineffective in a photo-driven reaction. Hofstadler et al. used TiO2-coated fused-silica glass fibers for wastewater treatment (Hofstadler et al., 1994). A single TiO2-coated optical-fiber reactor was designed by Danion et al. (Danion et al., 2004a). This reactor was built and studied further in order to optimize different parameters before the development of a multi-optical-fiber reactor (Danion et al., 2004b, 2006a,b 2007). Peill and Hoffmann demonstrated that TiO2-coated optical-fiber reactors had some inherent advantages over packed-bed reactors in photoreactions (Peill et al., 1997; Peill and Hoffmann, 1995, 1996, 1997, 1998; Xu et al., 2008). In general, a bundle of optical fibers can provide a very high exposed surface area per unit volume in a photoreactor. The transmission and uniform distribution of light energy are important in designing a photoreactor that differs completely from a traditional reactor. Figure 17.15 shows schematically how light is transmitted in an optical fiber (Wu et al., 2008). The light is split into two beams when it hits the internal fiber surface, due to the difference in refractive index between the TiO2 coating and the silica core. Part of the light is reflected and transmitted along the fiber, while the rest penetrates and excites the titania film at the interface. The light gradually spreads to the end of the fiber and diminishes. Upon photo-excitation, the electron–hole (e– and h+) pairs are generated on the TiO2 photocatalyst to carry out the subsequent photoreaction. Therefore, optical fiber can be regarded as a medium for delivering light effectively TiO2 Light

Fiber

17.15 The schematic of light transmission and spread of TiO2-coated optical fiber (adapted from Wu et al., 2005).


482


and uniformly to the photocatalyst surface, hence increasing the efficiency of photo-energy conversion in a photoreactor. Prior to the assembling of the optical-fiber reactor, the beam propagation method (BPM) was used to calculate the transmission loss in a 5 mm long TiO2-coated optical fiber (Chu et al., 2006). The power loss, which is due to the absorption of photo-energy by the TiO2 film, decays exponentially along the optical fiber as shown in Fig. 17.16 (Wu et al., 2005). Within the TiO2 film thickness of 50–350 nm, the decay of UV light (365 nm wavelength) in the fiber varies only slightly. Based on this result, it is estimated that the remaining power will decrease to 1 % in an 11.4 cm long optical fiber. Therefore, the length of optical fiber should be designed to be about 11.4 cm so that most photo-energy is spread and absorbed by the TiO2 film in a photoreactor. Experimental Preparation of TiO2-coated optical fiber The TiO2 solutions to be coated onto optical fibers were prepared by the thermal hydrolysis method. Titanium (IV) butoxide and polyethylene glycol (PEG, molecular weight of 20 000, Merck, Darmstadt, Germany) were added to a 0.1M nitric acid solution. The volume ratio of titanium butoxide to HNO3 was 1:6, and the weight of PEG was 50 % of that of TiO2. The appearance of the pure TiO2 solution was milky and slightly transparent. PEG was added to prevent cracking during the drying and calcination of the

Ratio of remaining power (I/Io)

1.00 Incident light Io 0.95

Fiber Length

0.90

Thickness of TiO2 50 nm

0.85 0.80

150 nm

250 nm

350 nm

0.75 0.70 0.00

1.00

2.00 3.00 Length (mm)

4.00

5.00

17.16 The remaining photo energy in an optical fiber along the transmission length (Z) for light wavelength of 365 nm (adapted from Wu et al., 2008).



483

film. Furthermore, it tends to increase the viscosity of the solution as well as the uniformity of TiO2 particles dispersed in the solution. An appropriate amount of metal precursor, such as CuCl2 or AgNO3, was added to obtain the desired metal loading of TiO2. The mixed solution was stirred and heated to 80 °C for eight hours. Details of the preparation procedure were reported in the literature (Wu et al., 2005). Optical fibers were obtained from the E-Tone Technology Company of Taiwan. The polymeric shield on the optical fiber was burned off in a furnace at 400 °C. The remaining quartz fiber had a diameter of 112 mm. Each quartz fiber was cleaned using a 5M NaOH solution in an ultrasonic cleaner, and then rinsed in de-ionized water and dried before the application of the dip-coating procedure. The bare fiber was immersed into the solution vertically, and then pulled up at various rates by a step motor. The pullingup rates ranged from 5–50 mm/min. The TiO2-coated optical fibers were dried in air at 150 °C at a rate of 1 °C/min from the ambient temperature, and maintained at 150 °C for three hours. Then they were calcined at 500 °C for another five hours. The TiO2 film on the optical fiber was durable and passed the ‘Cross-Cut Tape Test’, according to the methods described in ISO 2409 and ASTM D3359 (Lo and Wu, 2005). The synthesis of TiO2–SiO2 mixed oxide was carried out in two steps (Nguyen et al., 2005). The two solutions were mixed and refluxed at 80 °C for one hour while being vigorously stirred (ca 1500 rpm). Acetyl acetone (acac, Merck) was employed as a chelating agent. The first solution consisted of a corresponding volume of chelating agent in 15 ml of solvent (50 % v/v of ethanol and isopropanol). The second one was tetraisopropoxytitanium (TTIP, Merck) diluted in 35 ml of solvent. The resulting deep-yellow transparent liquid (when the acac was used as a chelating agent) or colorless transparent liquid (without the chelating agent) was cooled to ambient temperature for one hour. The solution is denoted as modified TTIP. The hydrolysant (35 % hydrochloric acid and the amount of de-ionized water used in 10 ml of solvent) was added to tetraethyl orthosilicate (TEOS, Merck) solution via a dropping funnel and vigorously stirred at 50 °C for 45 min, which is referred to as the pre-hydrolysis of TEOS. After that, the modified TTIP described above was added into the solution while being vigorously stirred. After 10 min, the hydrolysis was completed by the addition of the residual amount of de-ionized water diluted in 18 ml of solvent while being vigorously stirred. Finally, 15 ml of solvent was introduced after another 10 min had elapsed. The molar ratio of alkoxides : water : hydrochloric acid : solvents was 1 : 4 : 0.19 : 5 and the content of SiO2 in the TiO2–SiO2 mixed oxides was 5 wt%. The sol–gel solution was aged for 24 hours at room temperature and then dried in the oven at 80 °C for 24 hours. The resulting xerogel was calcined in static air at 500 °C for two hours.


484


The molar ratio of 1 for acac/TTIP was employed to prepare the TiO 2– SiO2–acac sample. The sample prepared without using a chelating agent was denoted as TiO2−SiO2. For the preparation of metal doped TiO2–SiO2, the same synthesis procedure as for the TiO2−SiO2 mixed oxide described above was employed, except that the corresponding metal salts were mixed with the derived TiO2−SiO2 sol–gels after the sol–gel processes. Commercial TiO2 powder (P25, Degussa) was used as a TiO2 source to prepare dye-sensitized photocatalysts. Cu(NO3)2·3H2O and Fe(NO3)3·9H2O (Aldrich) were employed as precursors of metal dopants on a P25 support. RuII(2,2¢−bipyridyl−4,4¢−dicarboxylate) 2−(NCS)2 (also called N3 dye, Solaronix) was used as a dye sensitizer and was dissolved in ethanol (99.5 %) to obtain a 3 mM dye solution. P25 slurry with corresponding metal salts was prepared by adding aqueous PEG solution with the metal salts to P25 powder in a mortar while being vigorously ground with a pestle (Nguyen et al., 2006). The prepared uniform lump-free slurry was coated onto optical fibers by the dip-coating method. The dye-adsorbed photocatalyst was obtained by dipping the corresponding photocatalyst into the 3 mM dye solution for 24 hours. Subsequently, the dye-adsorbed photocatalyst was rinsed with ethanol (99.5 %) to obtain a monolayer of dye on the photocatalyst surface. The resulting dye-adsorbed photocatalyst was finally put in an oven at 80 °C for 30 min to remove the residue solvent on the surface of the photocatalyst. Photoreduction of CO2 in the vapor phase An optical-fiber photoreactor (OFPR) was designed and assembled to transmit light to the fiber-supported TiO2 film from one side of the OFPR module, as shown in Fig. 17.17 (Wu et al., 2005, 2008). The fibers, with a length of about 11 cm, were inserted into the OFPR. The optical fibers were supported on circular plates with a diameter of 5.0 cm. The OFPR was irradiated by artificial light or concentrated natural sunlight through the quartz window at one side. The artificial light (320–500 nm) was supplied by an Exfo Acticure® 4000 with a highest intensity of 365 nm using an appropriate color filter. The light intensity could be tuned and measured with a Lumen meter (Exfo). A solar concentrator (Himawari, Japan) was used to collect natural sunlight. The reflection dish of the solar concentrator is able to track the sun’s trajectory during the day so that the maximum sunlight intensity can be obtained. The concentrated sunlight is transmitted via an optical cable and focused on the window of the photoreactor. The reactor was wrapped in a heating tape connected to a temperature controller with a thermocouple placed at the center of the reactor to maintain the reaction temperature. The reactor was purged by CO2 gas bubbling through distilled water for one hour before the reaction. The space velocity



485

Stainless support Quartz tube

Light source Divided wall

Optical fibers

Out

17.17 Schematic of optical-fiber photoreactor (adapted from Wu et al., 2008).

of the CO2 gas and H2O vapor was maintained at approximately 0.72 h–1. The photoreactions were carried out at steady-state and lasted for six to 24 hours. Some reactions were repeated two or three times, and the deactivation of photocatalysts was found to be negligible. The optical-fiber photoreactor was maintained at 75 °C to ensure that all light hydrocarbons could reach the GC detector. Consequently, no hydrocarbon product in the form of a liquid could be observed by the naked eye in the reactor after the photocatalytic reaction. The outlet gases were analyzed by an on-line sampling loop in a GC, which was described in the previous section.

17.3.4 Result and discussion Figure 17.18 shows the cross-section SEM images of TiO2, Cu/TiO2 and Ag/ TiO2 films on top of the fibers. The TiO2, Cu/TiO2 and Ag/TiO2 films are uniformly coated on the optical fiber with thicknesses of 31, 27 and 33 nm, respectively. The films appear to be transparent, colorless and flat without cracks. It can also be observed from Fig. 17.18 that all films consist of very fine close-packed particles with diameters of 10−20 nm. The XRD spectra in Fig. 17.19 show the diffraction patterns of TiO2, Cu/TiO2 and Ag/TiO2 films. Thermal treatment at 500 °C for five hours results in well-crystallized anatase-type TiO2. No other diffraction peak is observed in the XRD spectra, indicating that metal (Cu or Ag) oxide is finely dispersed on TiO2. Figure 17.20 shows the UV-Vis absorption spectra of TiO2, Cu/TiO2 and Ag/TiO2 films. Clearly, TiO2 film absorbs light with a wavelength below 380 nm. The UV-Vis absorption spectra of Cu/TiO2 and Ag/TiO2 photocatalysts are similar to that of pure titania.


486


Table 17.3 summarizes the characteristics of films including crystalline sizes, bandgaps and specific surface area. The average crystalline size of all films ranged from 14–17 nm, which is consistent with the SEM observations

100 nm

Mag = 100.00 K X EHT = 5.00 kV WD = 5 mm Signal A = InLens (a)

100 nm

Mag = 100.00 K X EHT = 5.00 kV WD = 5 mm Signal A = InLens (b)

17.18 Cross-section SEM images of TiO2 (a), Cu/TiO2 (b) and Ag/TiO2 (c) films on optical fibers (adapted from Wu et al., 2008).



487

Mag = 100.00 K X EHT = 5.00 kV WD = 5 mm Signal A = InLens

100 nm

(c)

17.18 Continued

* * Anatase *

*

Ag/TiO2

*

Intensity (a.u.)

*

Cu/TiO2

TiO2

20

30

40

2q (°)

50

60

70

17.19 XRD of TiO2, Cu/TiO2 and Ag/TiO2 (adapted from Wu et al., 2008).


488

Developments and innovation in CCS technology TiO2 Cu/TiO2

Absorption (a.u.)

Ag/TiO2

200

300

400 500 Wavelength (nm)

600

700

17.20 UV spectra of TiO2, Cu/TiO2 and Ag/TiO2 (adapted from Wu et al., 2008). Table 17.3 Characteristics of TiO2, Cu/TiO2, Ag/TiO2 films (adapted from Wu et al., 2008) Film

Crystal size1 (nm)

Bandgap2 (eV)

Specific area3 (m2/g)

TiO2 1.2 % Cu/TiO2 1.0 % Ag/TiO2

14.4 16.8 12.0

3.6 3.6 3.5

59.8 55.8 71.7

1

Calculated from the Scherrer equation according to the peak broadening of XRD spectra (at 2q = 25.28°). 2 Estimated from the UV–Vis spectra extrapolating the absorption edge of VU-Vis spectrum to the abscissa of zero absorption. 3 Measured by N2 adsorption on powder samples.

(Fig. 17.18). In general, the crystalline size is primarily influenced by the calcination temperature, and metal loading has a negligible effect. The bandgaps ranged from 3.5–3.6 eV, indicating that the metal loadings do not affect the band structure of TiO2. Photoreduction of CO2 using transition metal-loaded TiO2 Figure 17.20 shows methanol yield versus light intensity under partial pressures of CO2 and H2O at 1.19 and 0.03 bar, respectively, at 75 °C. The yield increases with light intensity in the range 2–10 W/cm2. The pure TiO2



489

5 TiO2 Cu/TiO2 Ag/TiO2

CH3OH rate (mmole/g-cat.h)

4

3

2

1

0 1

2

3

4 5 6 7 8 Light intensity (W/cm2)

9

10

11

17.21 Methanol yield vs light intensity on TiO2, Cu/TiO2 and Ag/TiO2 (residence time = 5000 sec., PCO2 = 1.19 bar, PH2O = 0.03 bar, 75 °C) (adapted from Wu et al., 2008).

photocatalyst gives the lowest methanol yield, while metal-loaded TiO2 photocatalysts demonstrate improved yields. The activity of Cu/TiO2 is slightly higher than that of Ag/TiO2 when the light intensity is lower than 8 W/cm2. The maximum methanol yield of 4.12 mmole/g-cat•h was obtained when a 1.0 wt% Ag/TiO2 photocatalyst was used under a light intensity of 10 W/cm2. The light dispersion along the fibers is nearly the same, regardless of a TiO2 layer thickness between 50 and 350 nm, based on the waveguide BPM (Wu et al., 2008). Our TiO2, Ag/TiO2 and Cu/TiO2 layer thicknesses are within this range. The photocatalyst layer is fully illuminated by the light; thus the effect of the TiO2 thickness on photoactivity may be insignificant. The heterogeneous photoreduction of CO2 involves two steps: photoactivation and subsequent catalytic reaction on the photocatalyst. The factors affecting photo-activation include: (i) the conversion efficiency of photons to electron–hole pairs; (ii) the recombination of electrons and holes; and (iii) the number of effective electrons available for the photocatalytic reduction of CO2, since a large proportion of the total electrons may be wasted due to migration loss. The factors affecting catalytic reaction include: (i) adsorption of CO2 on the catalyst surface may be limited; (ii) one of the elementary reaction steps is rate-limiting; and (iii) the reverse reaction of Equation 17.1, i.e., oxidation of methanol, may occur.


490


Table 17.4 Production rate of methane and ethylene over various photocatalysts under UVA (adapted from Nguyen and Wu, 2008) Photocatalyst1

Ethylene production rate [mmol/g-cat.•h]2

Methane production rate [mmol/g-cat.•h]2

TiO2–SiO2–acac Cu(0.5 wt%)–Fe(0.5 wt%)/ TiO2–SiO2–acac

~ 0 0.211

0.185 1.860

1

TiO2–SiO2 was synthesized by sol–gel process with 5 wt% of SiO2; acac stands for the acetyl acetone as a promoter during the preparation process of TiO2–SiO2. 2 Methane and ethylene production rate was determined on the basis of average production rate after the reaction time of 4 hr. The irradiation source was in UVA range (320–500 nm) with intensity 225 mW/cm2.

Metal, such as Cu or Ag, serves as an electron trap that prohibits the recombination of holes and electrons. In addition, the rapid transfer of excited electrons to the metal cluster also favors the separation of holes and electrons (Hirano et al., 1992), and therefore significantly promotes the photoreaction. The influence of mixed oxides TiO2–SiO2 Various products were found when using Cu–Fe/TiO2–SiO2 photocatalysts in the photoreduction of CO2 with H2O. The dominant products included significant amounts of ethylene and methane as well as trace amounts of ethane and methanol. Table 17.4 lists the production rate of methane and ethylene over two kinds of photocatalysts under UVA (Nguyen and Wu, 2008). The evolution of methane was observed for two photocatalysts for which a maximum production rate of 1.860 mmol/g-cat•h was measured on a Cu(0.5 wt%)–Fe(0.5 wt%)/TiO2–SiO2−acac photocatalyst. Meanwhile, ethylene was only produced on photocatalysts loaded with Cu and Fe metals. A solar concentrator was employed to conduct sunlight into the reactor. The production rate of methane over TiO2–SiO2 mixed-oxide photocatalysts under natural sunlight on some specific days is listed in Table 17.5 (Nguyen and Wu, 2008). Methane evolution on the Cu(0.5 wt%)–Fe(0.5 wt%)/TiO2– SiO2−acac photocatalyst (0.279 mmol/g-cat•h) was higher than that on its bare TiO2−SiO2–acac counterpart (0.177 mmol/g-cat•h). Since the intensity of natural sunlight is dependent on the weather of the day when the experiment is carried out, the activities of photocatalysts in hydrocarbon production can only be compared if the intensities of sunlight used are the same. Nevertheless, the average intensity of natural sunlight used to photoreduce CO2 over the TiO2–SiO2−acac photocatalyst was 6.35 mW/cm2 on 11 May 2007 from 9.30 am to 3.00 pm in Taipei, Taiwan. Meanwhile, photoreduction of CO 2 over the Cu(0.5 wt%)–Fe(0.5 wt%)/TiO2−SiO2−acac photocatalyst was carried



491

Table 17.5 Production rate of methane over TiO2–SiO2 mixed oxide-based photocatalysts under natural sunlight1 (adapted from Nguyen and Wu, 2008) Photocatalyst2


TiO2–SiO2–acac Cu(0.5 wt%)–Fe(0.5 wt%)/TiO2–SiO2–acac

0.1774 0.2795

1

Natural sunlight was obtained by using the solar concentrator. TiO2–SiO2 was synthesized by sol–gel process with 5 wt% of SiO2; acac stands for the acetyl acetone as a promoter during the preparation process of TiO2–SiO2. 3 Methane production rate was determined on the basis of average production rate after the reaction time of 4 hr. 4 Experiment was carried out on 11 May 2007 from 9.30 am to 3.00 pm in Taipei, Taiwan under average solar light intensity of 6.35 mW/cm2. 5 Experiment was carried out on 12 May 2007, from 10.30 am to 4.00 pm under average solar light intensity of 2.05 mW/cm2. 2

out on 12 May 2007, from 10.30 am to 4.00 pm under an average sunlight intensity of 2.05 mW/cm2. These results imply that the addition of Cu and Fe metals on the TiO2–SiO2–acac photocatalyst significantly improves its photoactivity for methane production under sunlight. It is interesting to observe from Table 17.4 that TiO 2–SiO 2–acac photocatalysts are not photoactive in ethylene production under UVA. This result could be ascribed to the much lower redox potential of methane as compared to that of ethylene (Gattrell et al., 2006). When TiO2−SiO2 is doped with Cu and Fe metals, the resulting photocatalysts show substantial differences in hydrocarbon production as well as product selectivity. It is well known that the electrochemical reduction of CO2 on single-crystal Cu electrodes may form various kinds of hydrocarbon products depending on its index planes (Hori et al., 1995, 2002). Fe, as a co-dopant on a Cu/TiO2–SiO2 photocatalyst, was observed to give rise to the synergistic performance of CO2 reduction to ethylene. On the other hand, only methane is produced on TiO2–SiO2-based photocatalysts under natural sunlight, as listed in Table 17.5. Although Cu and Fe as dopants are observed to improve the activities of photocatalysts, as indicated by the presence of a significant amount of ethylene in the products under UVA, they produce only methane under natural sunlight. This result could be explained by the negligible UVA intensity in the sunlight that is filtered out by the solar concentrator. According to the manufacturer, UVA is filtered out to protect the plastic optical-fiber in the light-transmission cable against damage. It is shown in Tables 17.4 and 17.5 that a bare TiO2–SiO2 photocatalyst produces comparable amounts of methane only under UVA with an intensity of 225 mW/cm2 or under natural sunlight with an average intensity of 6.35 mW/cm2. This phenomenon could imply that the increase in light intensity might be compensated for by the inherent electron–hole


492


recombination in the TiO2−SiO2–acac photocatalyst. On the other hand, Cu–Fe/TiO2–SiO2–acac in Table 17.5 shows strong photoactivity in methane production as compared with that of bare TiO2–SiO2–acac, which could be the result of the visible-light absorption of this photocatalyst. Dye-sensitized P25 (TiO2) photocatalyst The UV-Vis spectra of dye-sensitized P25 (TiO2) thin films are shown in Fig. 17.22 (Nguyen et al., 2008). Bare P25 inherently absorbs UV light. However, owing to the narrow bandgap of Fe2O3, Cu(0.5 wt%)–Fe(0.5 wt%)/P25 shows only a slight absorption in visible light, as shown in Fig. 17.22(b). When N3 dye is adsorbed on Cu(0.5 wt%)–Fe(0.5 wt%)/P25, the resulting photocatalyst shows strong absorption in the entire visible-light region, as shown in Fig. 17.22(c). To test dye stability, the absorption of N3 dye–Cu(0.5 wt%)–Fe(0.5 wt%)/P25 after six hours of photocatalytic reaction is also measured and illustrated in Fig. 17.22(d). The UV-Vis spectrum of the dye-adsorbed photocatalyst after the reaction is found to shift to the red region. It is generally agreed that some dyes, either in aqueous

Absorption (a.u.)

(a) (b) (c) (d)

200

300

400 500 600 Wavelength (nm)

700

800

17.22 UV-Vis spectroscopy of different P25 (TiO2) thin films: (a) P25, (b) Cu(0.5 wt%)–Fe(0.5 wt%)/P25, (c) N3 dye–Cu(0.5 wt%)–Fe(0.5wt %)/P25–as prepared, (d) N3 dye–Cu(0.5 wt%)–Fe(0.5 wt%)/P25–after six hours of reaction (adapted from Nguyen et al., 2008).



493

solution (Habibi et al., 2006) or as a monolayer on Langmuir–Blodgett films (Tachibana et al., 2000; Kuroda, 2004), could be aggregated under a very low pH value or UV irradiation to form dimer molecules. These dimer molecules, also called J-aggregate dimers, were found to shift the UVVis spectra of their corresponding monomer molecules to the red region. Therefore, the red-shift observed in Fig. 17.22(d) could be ascribed to the above J-aggregate phenomenon. In other words, N3 dye that is adsorbed on the TiO2 photocatalyst is found to be stable under UV irradiation as well as CO2 photoreduction. This result could be ascribed to the efficient charge transfer in the N3 dye–TiO2 system. Otherwise, N3 dye would be observed to have decomposed after the photoreaction. The rates of methane and ethylene production over dye-sensitized Cu–Fe/ P25 photocatalysts under artificial UVA light are shown in Table 17.6 (Nguyen et al., 2008). It can be seen that the production rates of methane and ethylene are similar over both Cu(0.5 wt%)−Fe(0.5 wt%)/P25 and N3 dye−Cu(0.5 wt%)−Fe(0.5 wt%)/P25 photocatalysts. The results imply that N3 dye is not effective for improving the production of methane and ethylene over the photocatalyst because N3 dye mostly absorbs visible light and only covers the surface of the photocatalyst that was coated on the optical fiber. Before the light reaches the adsorbed dye molecules, it may be scattered within the TiO2 matrix; as a result, the capability of the dye may not be fully exhibited. The results of the experiments using concentrated natural sunlight as the irradiation source are shown in Table 17.7 (Nguyen et al., 2008). Interestingly, N3 dye is observed to increase the methane production of the Cu(0.5 wt%)−Fe(0.5 wt%)/P25 photocatalyst by over 100 %. The superior photoactivity of the dye-adsorbed photocatalyst could be ascribed to the strong absorption of N3 dye in the visible-light region as shown in Fig. 17.22(c). Although the dye-adsorbed photocatalyst is observed to produce both methane and ethylene under artificial light, it is only photoactive for methane production under concentrated natural sunlight as shown in Table 17.7. This result could be attributed to the UV cut-off, which might not supply enough driving force to reduce CO2 to ethylene. Table 17.6 Production rate of methane and ethylene over dye-sensitized P25 under artificial light (adapted from Nguyen et al., 2008) Photocatalyst

Ethylene production rate (mmol/g-cat.•h)1

Cu(0.5 wt%)–Fe(0.5 wt%)/P25 0.575 N3 dye–Cu(0.5 wt%)–Fe(0.5 wt%)/P25 0.562 1

Methane production rate (mmol/g-cat.•h)1 0.914 0.847

Methane and ethylene production rate were determined on the basis of average production rate after the reaction time of 4 hr. The artificial light was in the range 320–500 nm with intensity 225 mW/cm2.


494


Table 17.7 Production rate of methane over dye-sensitized P25 photocatalysts under concentrated natural sunlight1 (adapted from Nguyen et al., 2008) Photocatalyst


Cu(0.5 wt%)–Fe(0.5 wt%)/P25 N3 dye–Cu(0.5 wt%)–Fe(0.5 wt%)/P25

0.2813 0.6174

1

Concentrated natural sunlight was obtained by using the solar concentrator. Methane production rate was determined on the basis of average production rate after the reaction time of 4 hr. 3 Experiment was carried out on 4 May 2007 in Taipei, Taiwan from 9.20 am to 2.50 pm under average concentrated sunlight intensity of 60 mW/cm2. 4 Experiment was carried out on 12 April 2007 in Taipei, Taiwan from 9.20 am to 2.50 pm under average concentrated sunlight intensity of 20 mW/cm2. 2

It is also interesting to compare the data in Tables 17.6 and 17.7. Under artificial light irradiation with an intensity of 225 mW/cm2 (Table 17.6), the production rates of methane and ethylene over the Cu(0.5 wt%)–Fe(0.5 wt%)/ P25 photocatalyst are 0.575 and 0.914 mmol/g-cat•h, respectively, which are of the same order of magnitude as those of the N3 dye–Cu(0.5 wt%)−Fe(0.5 wt%)/P25 counterpart. However, under concentrated natural sunlight with an average intensity of ca 60 mW/cm2, the Cu(0.5 wt%)−Fe(0.5 wt%)/P25 photocatalyst only produces methane at a very low rate of 0.281 mmol/g-cat•h. Meanwhile, the N3 dye–Cu(0.5 wt%)−Fe(0.5 wt%)/P25 photocatalyst still has a comparable methane production under concentrated natural sunlight to that under artificial light irradiation (Tables 17.6 and 17.7), despite an average intensity of concentrated natural sunlight of only ca 20 mW/cm2. These results once again could be ascribed to the strong visible-light absorption of the N3 dye–Cu(0.5 wt%)−Fe(0.5 wt%)/P25 photocatalyst as compared to that of its Cu(0.5 wt%)−Fe(0.5 wt%)/P25 counterpart. Note that the wavelength of our artificial light is in the range 320−500 nm, while that of the concentrated sunlight is extended to the range 500–800 nm. The N3 dye–Cu(0.5 wt%)– Fe(0.5 wt%)/P25 photocatalyst can fully absorb the light energy of 500−800 nm but the Cu(0.5 wt%)−Fe(0.5 wt%)/P25 photocatalyst cannot, as shown in Fig. 17.22. Another possible explanation for this phenomenon might be the efficient charge transfer in the N3 dye−TiO2 system, which is supported by the stable UV-Vis spectrum of the dye-adsorbed P25 after six hours of photoreaction, as illustrated in Fig 17.22(d).

17.3.5 Summary An optical-fiber photoreactor was designed and applied to the photoreduction of CO2 with H2O in the vapor phase using TiO2, Cu/TiO2, Ag/TiO2, Cu−Fe/ TiO2−SiO2 and dye-sensitized Cu–Fe/P25 coated optical fibers. The insertion of Fe metal into the TiO2–SiO2 lattice is found to substantially promote



495

the visible-light absorption of the resulting photocatalyst. The superior photoactivity of the N3 dye-adsorbed TiO 2 photocatalyst for methane production under concentrated natural sunlight is ascribed to its extended absorption in the full visible-light region. Compared with a traditional packedbed reactor, the optical fiber provides a medium to transmit light uniformly throughout the reactor. In addition, a higher processing capacity is possible because the photocatalyst can be dispersed on the optical fibers over a large surface area in a given reactor volume. An efficient photoreactor with a highly active photocatalyst is essential for a commercial-scale application to produce renewable fuels. Our results indicate a promising way to harvest sunlight via a photochemical process.

17.4

Advantages and limitations of photocatalytic processes

Photoreduction of CO2 is one of the best ways of obtaining renewable energy in a similar way to photosynthesis. Its advantage is that it benefits from the availability of unlimited solar energy. Although photovoltaics can also harvest solar energy as electricity, it is not easy to store and use as chemical energy, such as fuel. The photoreduction of CO2 has great advantages over green plants as it does not have to support the growth and maintenance of a living system. Ideally, the transformation of photo to chemical energy by a non-living catalyst should be more efficient than that by a life-supporting one. Furthermore, the photoreduction of CO2 does not require fertilizer and a warm/mild environment like plants do. However, the photocatalytic process still has some limitations. Although the solar energy reaching the earth every year corresponds to 10 000 years of fossil fuel use at current rates (SandÈn, 2008), the energy density of sunlight (several mW/cm2) is too low to use it economically. The collection of sunlight requires a large land area, and the construction of such a system can be costly. Furthermore, the intensity of solar energy depends on the geographic location, as its latitude on the earth and the weather conditions have a significant influence on the photoreduction of CO2.

17.5

Future trends

Finding a way to covert solar energy efficiently to fulfill our needs is the most urgent issue for the world. Carbon has been used as an energy carrier in nature since life began on the earth. The concentration of CO 2 in the atmosphere has been kept in a delicate balance between photosynthesis and respiration for millions of years. Recently, however the atmospheric CO2 has rapidly increased due to expanded human activity, particularly with the burning of fossil fuels. This rapid increase in CO2 has been because the rate


496


of photosynthesis cannot keep up with the consumption rate of fossil fuel. The harvesting of sunlight by the photocatalytic process must perform more efficiently and at a faster rate in order to compensate for CO2 generation by humans. The ultimate goal is to use solar energy for the mass production of fuel from CO2. Thus, the problems of global warming and the demand for energy could be solved permanently. However, there are still several hurdles that need to be overcome. (1) The collection of sunlight requires a large land area. Opportunely, a solar-energy system for CO2 photoreduction can be operated in an environmentally harsh area, such as a desert or frozen land, without competing for agricultural land. The ocean may be an alternative area for obtaining solar energy, too. (2) Most photocatalysts have a fixed bandgap so that only a specific wavelength of photon energy can be absorbed for a given photocatalyst. The rest of the photon energy below and above the bandgap is lost as heat. An ideal photocatalyst should be able to utilize the full span of the solar spectrum. (3) The separation/concentration of CO2 from power stations or automobiles is required. However, the energy input and operational cost can be noticeably high. Unless the technology of CO2 photoreduction can be advanced to use the atmosphere’s ppm-level CO2 like plants do, the current method may still not be economically feasible. (4) So far, the energy conversion, or quantum efficiency, of CO2 photoreduction using current photocatalysts remains very low. The efficiency of energy conversion should exceed the 7 % of natural photosynthesis. There still needs to be intensive research work into ways to improve the energy conversion efficiency of photocatalysts. In addition to the development of highly efficient photocatalysts, the process of CO2 photoreduction should be improved simultaneously with other technologies, such as the development of a solar-energy collecting device, optimal photoreactor system, etc. Desai pointed out that a large-scale plant should be able to convert more than 105 t CO2 per year or 5 kg CO2 per m3 of reactor per day (Desai, 2005). This requires the efforts of many scientists and engineers to make the technology competitive with natural photosynthesis. Figure 17.23 displays the concept of a solar-energy harvest system. Mini photoreactors, consisting of optical fibers, mimic the function of green leaves, which can efficiently collect solar energy. Once we have developed a solar-energy harvest process with a higher photon-to-chemical energy conversion rate than that of photosynthesis, the bio-energy from green plants would no longer be needed. Furthermore, it would be possible to make use of non-agricultural and non-forestry land.



497

Array of mini-photoreactors

17.23 Concept of solar-energy harvest system, no fertilizer requirement and no life support consumption under harsh environment.

17.6


∑

More information on the role of catalysis in energy can be found in the review by Jens R. Rostrup−Nielsen (Catalysis Review, 46(3–4), 247–270, 2004). ∑ For artificial photosynthesis, see Artificial Photosynthesis: From Basic Biology to Industrial Application, edited by Anthony F. Collings and Christa Critchley, Wiley-VCH, Weinheim, Germany, 2005.

17.7

References

Anpo M, Aikawa N and Kubokawa Y (1984) Photocatalytic hydrogenation of alkynes and alkenes with water over titanium dioxide. Platinum loading effect on the primary processes. J. Phys. Chem., 88, 3998–4000. Anpo M, Shima T, Kodama S and Kubokawa Y (1987) Photocatalytic hydrogenation of propyne with H2O on small-particle titanium – size quantization effects and reaction intermediates. J. Phys. Chem., 91, 4305–4310. Anpo M, Yamashita H, Ichihashi Y and Ehara S (1995) Photocatalytic reduction of CO2 with H2O on various titanium oxide catalysts. J. Electroanal. Chem., 396, 21–26. Bhattacharya AK, Pyke DR, Reynolds R, Walker GS and Werrett CR (1997) The use of O1s charge referencing for the X-ray photoelectron spectroscopy of Al/Si, Al/Ti and Al/Zr mixed oxides. J. Mater. Sci. Lett., 16, 1–3. Chen H-C, Chou H-C, Wu JCS and Lin H-Y (2008) Sol–gel prepared InTaO4 and its photocatalytic characteristics. J. Mater. Res., 23, 1364–1370.


498


Chu T, Huang H and Tsai D (2006) Optical effect of fibers with TiO2 nano film for photocatalyst reactor. Annual Meeting of Physics, 18 January, Taipei, Taiwan. Danion A, Bordes C, Disdier J, Gauvrit JY, Guillard C, Lanteri P and Jaffrezic-Renault N (2004a) Optimization of a single TiO2-coated optical fiber reactor using experimental design. J. Photochem. Photobiol. A: Chem., 168, 161–167. Danion A, Disdier J, Guillard C, Abdelmalek F and Jaffrezic-Renault N (2004b) Characterization and study of a single TiO2-coated optical fiber reactor. Appl. Catal. B: Environm., 52, 213–223. Danion A, Disdier J, Guillard C, Abdelmalek F and Jaffrezic-Renault N (2006a) Characterization and study of a single TiO2-coated optical fiber reactor. Int. J. Appl. Electrom, 23, 187–201. Danion A, Disdier J, Guillard C, Paisse O and Jaffrezic-Renault N (2006b) Photocatalytic degradation of imidazolinone fungicide in TiO2-coated optical fiber reactor. Appl. Catal. B: Environm., 62, 274–281. Danion A, Disdier J, Guillard C and Jaffrezic-Renault N (2007) Malic acid photocatalytic degradation using a TiO2-coated optical fiber reactor. J. Photochem. Photobiol. A: Chem., 190, 135–140. Desai DK (2005) The artificial photosynthesis system: an engineering approach. In Collings AF and Critchley C (eds), Artifical Photosynthesis: From Basic Biology to Industrial Application. Weinheim, Wiley-VCH. Dey GR, Belapurkar AD and Kishore K (2004) Photo-catalytic reduction of carbon dioxide to methane using TiO2 as suspension in water. J. Photochem. Photobiol. A: Chem., 163, 503–508. Fujishima A, Rao TN and Tryk DA (2000) Titanium dioxide photocatalysis. J. Photochem. Photobiol. C: Rev., 1, 1–21. Gattrell M, Gupta N and Co A (2006) A review of the aqueous electrochemical reduction of CO2 to hydrocarbons at copper. J. Electroanal. Chem., 594, 1–19. Getoff N, Scholes G and Weiss J (1960) Reduction of carbon dioxide in aqueous solutions under the influence of radiation. Tetrahedron Lett., 1, 17–23. Guan GQ, Kida T, Harada T, Isayama M and Yoshida A (2003a) Photoreduction of carbon dioxide with water over K2Ti6O13 photocatalyst combined with Cu/ZnO catalyst under concentrated sunlight. Appl. Catal. A: Gen., 249, 11–18. Guan GQ, Kida T and Yoshida A (2003b) Reduction of carbon dioxide with water under concentrated sunlight using photocatalyst combined with Fe-based catalyst. Appl. Catal. B: Environm., 41, 387–396. Gust D, Kramer D, Moore A, Moore TA and Verrnaas W (2008) Engineered and artificial photosynthesis: human ingenuity enters the game. MRS Bull., 33, 383–387. Habibi MH, Hassanzadeh A and Zeini-Isfahani A (2006) Effect of dye aggregation and azo-hydrazone tautomerism on the photocatalytic degradation of Solophenyl red 3BL azo dye using aqueous TiO2 suspension. Dyes Pigments, 69, 111–117. Hirano K, Inoue K and Yatsu T (1992) Photocatalyzed reduction of CO2 in aqueous TiO2 suspension mixed with copper-powder. J. Photochem. Photobiol. A: Chem., 64, 255–258. Hirose T, Maeno Y and Himeda Y (2003) Photocatalytic carbon dioxide photoreduction by Co(bpy)(3)(2+) sensitized by Ru(bpy)(3)(2+) fixed to cation exchange polymer. J. Mol. Catal. A: Chem., 193, 27–32. Hofstadler K, Bauer R, Novalic S and Heisler G (1994) New reactor design for photocatalytic waste-water treatment with TiO2 immobilized on fused-silica glass-fibers – photomineralization of 4-chlorophenol. Environ. Sci. Technol., 28, 670–674.



499

Hori Y, Wakebe H, Tsukamoto T and Koga O (1995) Adsorption of CO accompanied with simultaneous charge-transfer on copper single-crystal electrodes related with electrochemical reduction of CO2 to hydrocarbons. Surf. Sci., 335, 258–263. Hori Y, Takahashi I, Koga O and Hoshi N (2002) Selective formation of C2 compounds from electrochemical reduction of CO2 at a series of copper single crystal electrodes. J. Phys. Chem. B, 106, 15–17. Howe RF and Gratzel M (1987) EPR study of hydrated anatase under UV irradiation. J. Phys. Chem., 91, 3906–3909. Inoue T, Fujishima A, Konishi S and Honda K (1979) Photoelectrocatalytic reduction of carbon dioxide in aqueous suspensions of semiconductor powders. Nature, 277, 637–638. Kaneco S, Shimizu Y, Ohta K and Mizuno T (1998) Photocatalytic reduction of high pressure carbon dioxide using TiO2 powders with a positive hole scavenger. J. Photochem. Photobiol. A: Chem., 115, 223–226. Kuroda S (2004) J-aggregation and its characterization in Langmuir-Blodgett films of merocyanine dyes. Adv. Colloid Interface Sci., 111, 181–209. Laws EA and Berning JL (1991) Photosynthetic efficiency optimization studies with the macroalga gracilaria-tikvihae – implications for CO2 emission control from powerplants. Bioresour. Technol., 37, 25–33. Linsebigler AL, Lu GQ and Yates JT (1995) Photocatalysis on TiO2 surfaces – principles, mechanisms, and selected results. Chem. Rev., 95, 735–758. Lo CF and Wu JCS (2005) Preparation and characterization of TiO2-coated optical-fiber in a photo reactor. J. Chin. Inst. Chem. Eng., 36, 119–126. Lo CC, Hung CH, Yuan CS and Wu JF (2007) Photoreduction of carbon dioxide with H-2 and H2O over TiO2 and ZrO2 in a circulated photocatalytic reactor. Sol. Energy Mater. Sol. Cells, 91, 1765–1774. Nguyen T-V and Wu JCS (2008) Photoreduction of CO2 to fuels under sunlight using optical-fiber reactor. Sol. Energy Mater. Sol. Cells, 92, 864–872. Nguyen TV, Choi DJ and Yang OB (2005) Effect of chelating agents on the properties of TiO2-SiO2 mixed oxide for photocatalytic water decomposition. Res. Chem. Intermed., 31, 483–491. Nguyen TV, Lee HC and Yang OB (2006) The effect of pre-thermal treatment of TiO2 nano-particles on the performances of dye-sensitized solar cells. Sol. Energy Mater. Sol. Cells, 90, 967–981. Nguyen T-V, Wu JCS and Chiou C-H (2008) Photoreduction of CO2 over Ruthenium dye-sensitized TiO2-based catalysts under concentrated natural sunlight. Catal. Commun., 9, 2073–2076. Nishimura Y, Yoshida D, Mizuhata M, Asaka K, Oguro K and Takenaka H (1995) Solid polymer electrolyte CO2 reduction. Energy Convers. Manage., 36, 629–632. Ogata T, Yanagida S, Brunschwig BS and Fujita E (1995) Mechanistic and kinetic studies of cobalt macrocycles in a photochemical CO2 reduction system. Energy Convers. Manage., 36, 669–672. Ohta K, Hasimoto A and Mizuno T (1995) Electrochemical reduction of carbon dioxide by the use of copper tube electrode. Energy Convers. Manage., 36, 625–628. Ozcan O, Yukruk F, Akkaya EU and Uner D (2007a) Dye sensitized artificial photosynthesis in the gas phase over thin and thick TiO2 films under UV and visible light irradiation. Appl. Catal. B: Environm., 71, 291–297. Ozcan O, Yukruk F, Akkaya EU and Uner D (2007b) Dye sensitized CO2 reduction over pure and platinized TiO2. Top. Catal., 44, 523–528.


500


Pan PW and Chen YW (2007) Photocatalytic reduction of carbon dioxide on NiO/InTaO4 under visible light irradiation. Catal. Commun., 8, 1546–1549. Pathak P, Meziani MJ, Li Y, Cureton LT and Sun YP (2004) Improving photoreduction of CO2 with homogeneously dispersed nanoscale TiO2 catalysts. Chem. Commun., 1234–1235. Pathak P, Meziani MJ, Castillo L and Sun YP (2005) Metal-coated nanoscale TiO2 catalysts for enhanced CO2 photoreduction. Green Chem., 7, 667–670. Peill NJ and Hoffmann MR (1995) Development and optimization of a TiO2 coated fiberoptic cable reactor – Photocatalytic degradation of 4-chlorophenol. Environ. Sci. Technol., 29, 2974–2981. Peill NJ and Hoffmann MR (1996) Chemical and physical characterization of a TiO2coated fiber optic cable reactor. Environ. Sci. Technol., 30, 2806–2812. Peill NJ and Hoffmann MR (1997) Solar-powered photocatalytic fiber-optic cable reactor for waste stream remediation. J. Energ-T Asme, 119, 229–236. Peill NJ and Hoffmann MR (1998) Mathematical model of a photocatalytic fiber-optic cable reactor for heterogeneous photocatalysis. Environ. Sci. Technol., 32, 398–404. Peill NJ, Bourne L and Hoffmann MR (1997) Iron(III)-doped Q-sized TiO2 coatings in a fiber-optic cable photochemical reactor. J. Photochem. Photobiol. A: Chem., 108, 221–228. Sanchez E and Lopez T (1995) Effect of the preparation method on the band gap of titania and platinum-titania sol–gel materials. Mater. Lett., 25, 271–275. SandÈn BA (2008) Solar solution: the next industrial revolution. Mater. Today, 11, 22–24. Sasirekha N, Basha SJS and Shanthi K (2006) Photocatalytic performance of Ru doped anatase mounted on silica for reduction of carbon dioxide. Appl. Catal. B: Environm., 62, 169–180. Schilke TC, Fisher IA and Bell AT (1999) In situ infrared study of methanol synthesis from CO2/H2 on titania and zirconia promoted Cu/SiO2. J. Catal., 184, 144–156. Serpone N, Lawless D, Disdier J and Herrmann J-M (1994) Spectroscopic, photoconductivity, and photocatalytic studies of TiO2 colloids: naked and with the lattice doped with Cr3+, Fe3+, and V5+ cations. Langmuir, 10, 643–652. Tachibana H, Yamanaka Y, Sakai H, Abe M and Matsumoto M (2000) J-aggregate formation of amphiphilic merocyanine in Langmuir-Blodgett films. J. Lumin., 87–9, 800–802. Tseng IH, Chang W-C and Wu JCS (2002) Photoreduction of CO 2 using sol–gel derived titania and titania-supported copper catalysts. Appl. Catal. B: Environm., 37, 37–48. Tseng IH, Wu JCS and Chou H-Y (2004) Effects of sol-gel procedures on the photocatalysis of Cu/TiO2 in CO2 photoreduction. J. Catal., 221, 432–440. Vickerman JC (1997) Surface Analysis: the Principal Techniques, Chichester, UK; New York, John Wiley. Wu JCS (2009) Photocatalytic reduction of greenhouse gas CO2 to fuel. Catal. Surv. Asia, 13, 30–40. Wu JCS and Yeh CY (2001) Sol–gel-derived photosensitive TiO2 and Cu/TiO2 using homogeneous hydrolysis technique. J. Mater. Res., 16, 615–620. Wu JCS, Lin HM and Lai CL (2005) Photo reduction of CO2 to methanol using opticalfiber photoreactor. Appl. Catal. A: Gen., 296, 194–200. Wu J, Wu T-H, Chu T, Huang H and Tsai D (2008) Application of optical-fiber photoreactor for CO2 photocatalytic reduction. Top. Catal., 47, 131–136.



501

Xu JJ, Ao YH, Fu DG, Lin J, Lin YH, Shen XW, Yuan CW and Yin ZD (2008) Photocatalytic activity on TiO2-coated side-glowing optical fiber reactor under solar light. J. Photochem. Photobiol. A: Chem., 199, 165–169. Yahaya AH, Gondal MA and Hameed A (2004) Selective laser enhanced photocatalytic conversion of CO2 into methanol. Chem. Phys. Lett., 400, 206–212. Yamashita H, Nishiguchi H, Kamada N, Anpo M, Teraoka Y, Hatano H, Ehara S, Kikui K, Palmisano L, Sclafani A, Schiavello M and Fox MA (1994) Photocatalytic reduction of CO2 with H2O on TiO2 and Cu/TiO2 catalysts. Res. Chem. Intermed., 20, 815–823. Yamashita H, Fujii Y, Ichihashi Y, Zhang SG, Ikeue K, Park DR, Koyano K, Tatsumi T and Anpo M (1998) Selective formation of CH3OH in the photocatalytic reduction of CO2 with H2O on titanium oxides highly dispersed within zeolites and mesoporous molecular sieves. Catal. Today, 45, 221–227. Yanagida S, Ogata T, Yamamoto Y, Wada Y, Murakoshi K, Kusaba M, Nakashima N, Ishida A and Takamuku S (1995) A novel CO2 photoreduction system consisting of phenazine as a photosensitizer and cobalt cyclam as a CO2 scavenger. Energy Convers. Manage., 36, 601–604. Zhang ZB, Wang CC, Zakaria R and Ying JY (1998) Role of particle size in nanocrystalline TiO2-based photocatalysts. J. Phys. Chem. B, 102, 10871–10878. Zhao ZH, Fan JM and Wang ZZ (2007) Photo-catalytic CO2 reduction using sol–gel derived titania-supported zinc-phthalocyanine. J. Clean Prod., 15, 1894–1897.


Index

acetyl–CoA pathway, 413 acidosis, 359 airlift photobioreactor, 424 amphiboles, 436 areal integrators, 231 areal measurements, 230–3 argon, 335 Artificial Soil Gassing and Response Detection, 337 asbestos, 451 ASGARD see Artificial Soil Gassing and Response Detection ASTM D3359, 483 atmospheric pool, 273 bandgap, 466 Best Practice Manual for Storage of CO2 in Saline Aquifers, 29 biofixation advantages and limitations, 426–7 algal divisions and their characteristic storage products, 416 autotrophic micro-organisms and their energy sources, 414 basic principles and methods, 412–14 Calvin cycle, 412, 413 other CO2 biofixation pathways, 412–14 carbon dioxide, 411–28 chemoautrophs and photoautotrophs, 414–18 chemoautotrophic bacteria, 415 microalgae, 416–18 photosynthetic micro-organisms, 415 CO2 fixation by microalgae, 418–26

CO2 fixation and biofuel production, 421–2 large-scale microalgal farming systems, 423–5 microalgal biomass harvesting and drying, 425–6 microalgal farming, 419–21 ocean fertilisation, 418–19 complete CO2 recycling for solar energy capturing, 427 energy production via microalgal biomass conversion, 423 future trends, 427–8 biomass drying, 425–6 blue green algae, 418 brine density, 190–2 brine displacement, 244–51 brittlestars see Ophiura ophiura bubble sparged photobioreactor, 424 buoyancy frequency, 315–16 burrowing urchin see Echinocardium cordatum calcification, 346 calcium oxide, 433 Callinectes sapidus, 359 Calvin cycle, 412, 413 Cancer pagurus, 360 capillary entry pressure, 187 capillary pressure, 186, 187, 194–5, 196 applications to CO2 brine systems, 187–8 Monte Carlo predictions within a reservoir seal, 193–5 petrophysical properties used, 192 and seal capacities calculations, 188–93

503 © Woodhead Publishing Limited, 2010

504

Index

capillary seals calculations of capillary pressures and seal capacities, 188–93 contact angle data, 190 fluid density data, 190–2 interfacial tension data, 189–90 interfacial tension data summary, 189 petrophysical properties, 192 pore throat radius data, 192–3 temperature and pressure conditions for hydrocarbon-producing sands, 191 carbon dioxide trapping in underground reservoirs, 185–99 capillary pressure applications to CO2 brine systems, 187–8 capillary principles and terminology for hydrocarbon systems, 186–7 Monte Carlo predictions of capillary pressure within a reservoir seal, 193–5 calculated capillary pressures and minimum column heights, 194–5 calculated minimum capillary pressures, 196 pilot CO2 projects and natural accumulation reservoir conditions, 193 capillary trapping, 170 capacity, 173, 174 caprock, 45–6 CarbFix project, 437, 450 carbon capture and sequestration continuous point-source leak evolution, 356 continuous point-source leak showing pH perturbation evolution, 357 leakage regional scale modelling, 353–5 predicted changes in carbonate chemistry, 347–9 temporary point-source leak scenario, 354 carbon capture and storage, 169 carbon density, 108 carbon dioxide biofixation by micro-organisms, 411–28

advantages and limitations, 426–7 basic principles and methods, 412–14 chemoautrophs and photoautotrophs, 414–18 future trends, 427–8 microalgae, 418–26 biological enhanced utilisation, 384–90 algae farm, 387 fatty acids distribution in macroalgae lipids, 389 influence of CO2 concentration on fatty acids distribution in Chaetomorpha I, 388 land requirements for bio-oil production for different biomass, 386 LCA of biofuels production from biomass, 391 lipid accumulation capacity of some microalgae or macroalgae, 385 micro- and macroalgae performance, 390 terrestrial vs aquatic biomass, 386 use of cascade of technologies for full use of biomass, 389 bubble dissolution in shallow ocean, 352 capillary seals for underground reservoirs trapping, 185–99 capture processes and technologies in power plants, 15 chemical production, 398–404 butadiene coupling with CO2, 402 carbamates, carbonates and isocyanates synthesis based on CO2, 400 energy products synthesis, 402–4 intermediate and fine chemicals synthesis, 398–402 compression, transport and injection processes and technologies, 16 conditions for using CO2, 378–80 conversion as storage of excess electric energy or intermittent energies, 391–8 electrochemical conversion, 393–7 photocatalytic reduction, 397–8 thermal processes, 392


Index

droplet plume and CO2-enriched seawater plume from numerical simulations, 350 electrode-potential for some CO2 multielectron reductions, 393 energy cycle using carbon as energy carrier, 464 environmental risks and impacts of leakage in terrestrial ecosystem, 324–38 atmospheric enrichment, 332–4 future trends, 336–8 leak monitoring techniques, 334–6 leak scenarios, 325–7 terrestrial leakage impacts, 327–32 environmental risks and performance assessment of leakage in marine ecosystems, 344–67 future trends, 366–7 leak monitoring options, 365–6 leaks mitigation, 366 marine ecosystem impacts of leakage, 358–65 free energy of formation and combustion heat of C1 molecules, 396 full-developed CO2-enriched plume properties, 353 geological sequestration, 17 geological sequestration screening and selection criteria and characterisation techniques, 27–52 CO2 storage capacity estimation, 47–51 future trends, 51–2 screening for storage suitability and site selection, 28–43 site characterisation, 43–7 site selection, 28–39 geological storage options, 324–5 deep saline aquifers, 325 depleted oil and gas reservoirs, 325 unmineable coal beds, 325 industrial utilisation, 377–405 E-factor of several kinds of chemicals, 383 future trends, 404–5 mineralisation, 433–52

505

advantages and limitations, 449–50 basic principles and methods, 435–8 future trends, 451–2 process energy efficiency, 447 in situ mineral carbonation, 450–1 technologies and potential applications, 438–46 oceanic sequestration, 304–20 anthropogenic CO2 sources, 308–9 CO2 properties, 311–12 future trends, 318 history of deep ocean storage proposals, 305–7 injection of CO2, water and pulverised limestone emulsion, 313–18 legal constraints of CO2 deep ocean storage, 307–8 modelling of CO2 release, 312–13 ocean structure, 309–10 options for stabilising GHGs atmospheric concentrations, 272 phase diagram in the ocean, 351 photocatalytic reduction, 463–97 advantages and limitations, 495 fundamentals of photocatalysis, 465–70 future trends, 495–6 renewable energy, 470–95 physical and chemical behaviour in marine system, 346–55 carbonate chemistry, 346–7 CCS leakage regional scale modelling, 353–5 fine-scale dynamics of droplets/ plumes, 349–52 predicted changes in carbonate chemistry, 347–9 proposed mechanism CO2 adsorption on TiO2, 469 photocatalytic CO2 reduction on TiO2, 470 semiconductors for reduction in water under solar light irradiation, 397 solubility in several solvents and influence of pressure, 394 sources and its value, 380–1 technological uses, 381–4, 382 terrestrial and ocean sequestration, 18


506

Index

terrestrial sequestration, 271–98 basic principles, 279–90 challenges, 291–5 emissions from agricultural vs other activities, 276–9 extrapolation, 295–6 potential, 290–1 soil and terrestrial carbon as indicators of climate change, 296–7 terrestrial pool and it’s role in the global carbon cycle, 273–6 uses in synthetic chemistry, 399 vs CFC climate change power, 382 carbon dioxide brine systems, 187–8 carbon dioxide capture and geological storage, 433 carbon dioxide capture and sequestration, 433 estimated storage capacities and storage times, 434 carbon dioxide capture and storage, 1–20, 57 future trends, 19–20 greenhouse gas emissions and global climate change, 2–8 average annual atmospheric CO2 concentrations, 4–5 carbon dioxide intensity by region and country 1980-2030, 7 EU-15 CO2 emissions 2006, 3 global average air and ocean temperatures, rising global average sea levels and melting of sea-ice, 6 management and stabilisation routes, 8–11 CCS systems schematic diagram, 10 stabilisation wedges concept for reducing carbon emissions, 11 total cost of early commercial projects, 15 and transport technology development and innovation, 11–17 CCS component technologies, 12 CO2 compression, transport and injection processes and technology, 16

economics, regulation and planning, 14–15 global CCS projects, 13 industrial applications, 16–17 processes and technologies in power plants, 15–16 utilisation technology development and innovation, 17–19 advanced concepts, 18–19 geological sequestration, 17 maximising and verifying storage in underground reservoirs, 17–18 terrestrial and ocean sequestration and environmental impacts, 18 carbon dioxide density, 190–2 carbon dioxide dissolution, 67–9 carbon dioxide flooding, 169–70 carbon dioxide injection, 240, 251–5 capillary trapping capacity as a function of initial non-wetting phase saturation, 173 as a function of porosity for different measurements, 174 design to maximise underground reservoir storage and EOR, 169–81 capillary trapping experiments, 172–4 carbon storage in geological formations, 169–72 future trends, 181 storage in oilfields, 179–80 trapped non-wetting phase, 171 field-scale design of storage in aquifers, 175–9 gas saturation as a function of distance, 178 mobility ratio between injected CO2–brine mixture and formation brine, 177 saturation distributions for CO2 injection simulation, 180 saturation distributions near injection wall, 178 simulation parameter, 176 simulator used to design CO2 storage, 175 storage efficiency and amount of brine injected, 179


Index

gas saturation maps, 120 low-permeability fractured systems, 114–16 modelling implications, 121–2 pure CO2 distribution maps, 121 carbon dioxide injectivity, 115 carbon dioxide lake, 311 carbon dioxide leakage detection and measurement, 225–33 areal measurements, 230–3 pointwise measurements, 226–30 shallow-focused methods, 233 carbon dioxide mineral sequestration see mineral carbonation carbon dioxide plumes, 248, 253 carbon dioxide sequestration, 185 CO2 enhanced recovery mechanisms, 109–16 binodal curves for CO2 and methane, 113 CO2 injection in low-permeability fractured systems, 114–16 development of miscibility, 109–12 displacement mechanisms, 112–14 immiscible vs miscible recovery, 109 miscible displacement, 110 CO2 storage capacity estimation, 47–51 coal beds, 47–8 deep saline aquifers, 50–1 oil and gas reservoirs, 48–9 coal seams and use for enhanced coalbed methane recovery, 127–58 ECBM schematic, 130 enhanced coalbed methane recovery, 129–31 future trends, 157–8 storage in unmineable coal seams, 128–9 competitive adsorption, 131–9 CO2 sorption isotherms on dry Italian coal, 137 competitive high pressure gas sorption measurements, 133 competitive sorption of a ternary gas mixture, 136 excess sorption isotherms of CO2, CH4 and N2, 135

507

single-component high pressure gas sorption measurements, 132 deep saline aquifers and formations, 57–88 field tests, 153–6 Ishikari coal field CO2 storage pilot project results, 156–7 performed and planned ECBM tests, 154 future trends, 86–8 environmental safety and concerns, 86–7 policy considerations, 87–8 saline aquifer injection projected transport and operating costs, 87 saline aquifer sequestration costs, 86 social acceptance, 88 geologic storage in tight rocks, 118–22 gas saturation maps for depletion/ imiscible CO2 injection, 120 pure CO2 distribution maps for miscible CO2 injection, 121 mass transfer and enhanced coalbed modelling, 148–53 coal seam density profiles, 149–50 ECBM schemes simulation results, 152 modelling, 74–8 geologic storage and related projects, 79 history, 74–5 model inter-comparison, 78 sedimentary basins, 76–8 oil and gas reservoirs and use for enhanced oil recovery, 104–23 capacity estimates, 107 carbon density, 108 depth and oil density for CO2-EOR projects, 117 EOR and carbon storage cooptimisation, 116–18 gas reservoirs, 106–7 oil reservoirs, 105–6 pilot sites, 80–6 CO2 plume in subsurface Frio, 84 Frio formation, 83 gas production and CO2 re-injection into Utsira sand, 81


508

Index

other projects, 83–5 Sleipner CO2 injection plume seismic monitoring, 82 Snøhvit project, 80 Utsira Sand, Sleipner, northern North Sea, 80 Weyburn, 85–6 pseudoternary diagram condensing-gas drive miscibility, 111 FCM process, 110 vaporising-gas drive miscibility, 112 saline aquifers, 58–64 characteristics, 61–4 CO2 phase diagram, 61 North America sedimentary basins, 59 range of potential injection conditions, 62 storage capacities for major geologic storage reservoirs, 63 world sedimentary basins, 60 screening and selection criteria, and characterisation techniques for CO2, 27–52 future trends, 51–2 screening for storage suitability and site selection, 28–43 depleted hydrocarbon reservoirs and in enhanced oil or gas recovery, 39–42 desirable characteristics of sedimentary basins, 32 eliminatory suitability criteria, 31 oil reservoirs characteristics for CO2-EOR, 41 site selection for CO2 storage, 28–39 uneconomic coal beds CO2 storage, 42–3 site characterisation, 43–7 caprock and overburden properties, 45–6 geochemical assessment, 46 geological characterisation, 44–5 geomechanical assessment, 46 predictive flow modelling, 46–7 risk assessment, 47

storage unit properties, 45 swelling and permeability, 139–48 coal porosity and permeability changes, 144 coal swelling measurements, 140 Italian coals dry disc swelling, 141–2 Sulcis coal core experiments results, 146–7 trapping mechanisms, 64–74 calcite solubility temperature dependence, 71 CO2 dissolution, 67–9 hydrodynamic, 66–7 hydrodynamic stratigraphic trapping, 65–6 ionic trapping, 70–2 isotherms of CO2 solubility in pure water, 69 log K of carbonic acid dissociation vs temperature, 70 mineral trapping, 72–4 residual trapping, 67 temperature dependence of anorthite reaction with dissolved CO2, 73 carbon dioxide storage eliminatory criteria, 34–6 estimation of capacity, 47–51 coal beds, 47–8 deep saline aquifers, 50–1 oil and gas reservoirs, 48–9 field-scale design of storage in aquifers, 175–9 hydrocarbon reservoirs and enhanced oil or gas recovery, 39–42 illustrative modelling applications, 244–59 boundary of model domain and thickness of Mount Simon, 247 CO2 leakage histogram, 260 CO2 outflow behaviour on injection rate, 258 conceptual leakage scenario, 257 conceptual model for leakage through wellbores, 259 contours of CO2 saturation after 50 years of injection, 249 fluid pressurisation and brine displacement, 244–51


Index

injected CO2 long-term fate, 251–5 leakage along faults, fracture zones, and wellbores, 255–9 pressure increase contours, 250 regions of influence, 245 simulated CO2 plumes after one and 1000 years, 253 temperature and CO2 saturations profile, 258 leakage detection and measurement, 225–33 areal measurements, 230–3 leakage monitoring system, 227 mobile laser data image, 232 pointwise measurements, 226–30 shallow-focused methods, 233 surface displacement 2004-2008, 234 Weyburn soil gas surveys results, 228 mathematical modelling of underground reservoirs longterm safety, 240–60 coupled processes, 243–4 measurement and monitoring technologies in underground reservoirs, 203–35 future trends, 233–5 oilfields, 179–80 selection criteria, 36–9 simulator, 175 site selection, 28–39 basin and regional scale screening, 30–3 local and site scale screening, 33–9 Sleipner CO2 plume time-lapse seismic images development to 2006, 211 velocity pushdown development, 212 storage site monitoring background, 204–7 monitoring objectives, 206 monitoring tools, 207 storage site migration and leakage scenarios, 205 tools deployed and planned at CO2 injection sites, 208 subsurface CO2 detection and

509

measurement, 207, 209–25 CO2 detection limits at Sleipner, 215 CO2 distributions between injection and observation wells in Frio, 224 Cranfield pressure monitoring, 221 Cranfield well locations, 220 deep-focused methods, 225 downhole results from Frio, 223 fluid distribution volumetric imaging, 210 invasive methods, 219–25 non-invasive monitoring, 209–19 seismic quantification of the 1999 dataset, 213 Sleipner gravity survey layout, 218 carbon electrode, 395 carbonate changes in key carbonate system parameters, 347 chemistry in marine system, 346–7 predicted changes likely from ocean acidification and CCS leakage, 347–9 carbonate aquifers, 77–8 CCS see carbon dioxide capture and storage chemoautotrophic bacteria, 415 ChemTOUGH, 77 Chionoecetes tanneri, 359 Chlamydomonas reinhardtii, 417 Chlorella, 417, 421, 425 Chlorella kessleri, 420 Chlorella sp. UK001, 420 Chlorococcum littorale, 420, 424 chlorofluorocarbons, 381 chrysolaminarin, 416 chrysotile, 450 clathrate, 306 Clean Development Mechanism, 291 climate change power, 381 CO2-EOR, 106 correlation between depth and oil density at standard conditions, 117 coal, 8 coal beds, 27, 47–8 carbon dioxide storage, 42–3


510

Index

coal ranks, 29 coal seams, 128, 131 carbon dioxide sequestration, 127–58 Coal-Seq project, 153 coal swelling, 139, 142 measurements, 140 Sulcis and Ribolla coals unconstrained dry discs, 141–2 coalification process, 128 CO2CRC Otway Project, 83 Commission Directive 96/61/EC, 20 Commission Directive 85/337/EEC, 20 competitive adsorption, 131–9 condensing-gas drive, 111 conservation agriculture, 284 contact angle, 186, 190 Cortez pipeline, 326 CO2STORE, 80 Coulter LS230 particle size analyser, 472 couple hydrodynamic–ecosystem models, 353 crab see Callinectes sapidus; Chionoecetes tanneri Cranfield oilfield, 219 critical criteria, 31 Cross-Cut Tape Test, 483 cross-hole seismic tomography, 223 CRUNCH, 77 Cyanobacteria, 418 d-electrons, 395 Darcy’s law, 145 Debye-Huckle method, 75 deep-focused monitoring, 233 deep ocean storage history of proposals, 305–7 legal constraints, 307–8 London convention on ocean dumping, 307 United Nations Convention on the Law of the Sea, 308 deep saline aquifers, 27, 50–1, 58, 170 see also saline aquifers carbon dioxide sequestration, 57–88 future trends, 86–8 modelling, 74–8 pilot sites, 80–6 saline aquifers, 58–64 trapping mechanisms, 64–74

dehydrogenating agent, 401 diadinoxanthin, 417 diallylcarbonate, 399 diatoms, 416–17 dichloromethane, 384 diethylcarbonate, 399 differential satellite interferometry, 233 diffuse degassing, 256 dimethylcarbonate, 399 DInSAR see differential satellite interferometry dioxymethylene, 469 diphenylcarbonate, 399 direct photolysis, 468 dissolution-diffusion-convection, 254 double plume model, 313 dual-sorption models, 139 dye-sensitised P25 photocatalyst, 492–4 dynamic storage capacity, 36, 37 E-factor, 383 E-Tone Technology Company, 483 Earth Simulator, 247 ECBM recovery see enhanced coalbed methane recovery Echinocardium cordatum, 364 Echinocardium incordata, 360 eddy covariance, 334 edible crab see Cancer pagurus effective reservoir capacity, 49 effective storage capacity, 47 effective storage volume, 50 efficiency, 30 electrochemical conversion, 393–7 carbon monoxide, 396 competing processes and faradic efficiency, 395–6 electrodes, 394–5 eventual use of electrocatalysts, 395 hydrocarbons, 397 methanol, ethanol, 396 olefins, 397 support solvent, 393–4 electrochemical reduction, 468 eliminatory criteria, 30–1 energetic CO2 sequestration efficiency, 460 enhanced coalbed methane recovery, 17, 127–58


Index field tests, 154 operation schematic, 130 simulation results, 152 enhanced-natural weathering, 438 enhanced oil recovery, 39–42, 104, 116–18, 119, 129, 380 CO2 sequestration in oil and gas reservoirs, 104–23 EQ3/EQ6, 75 essential criteria, 31 Eustigmatophytes, 417–18 excess sorption, 134 Exfo Acticure 4000, 484 faults, 255–9 fine-scale ocean, 349 first-contact miscible, 110 flat-plate photobioreactor, 424 fluid pressurisation, 244–51 flux chamber, 334 flux gradient techniques, 334 formaldehyde, 469 formate, 469 formation damage, 114 forsterite, 436 fossil fuel power plants, 15 fossil fuels, 8 fracture zones, 255–9 free-air carbon dioxide enrichment techniques, 283, 332 Frio, 83 Frio C, 83 Frio project, 223, 225 fucoxanthin, 416, 417 gas displacement recovery, 40 gas in place, 155 gas injection, 104, 116 gas reservoirs, 27, 106–7, 169, 170 CO2 sequestration and use for enhanced oil recovery, 104–23 gas-to-fuel conversion, 403 GCCSI see Global Carbon Capture and Storage Institute GEMBOCHS, 76 geologic pool, 273 geological characterisation, 44–5 geological sequestration, 27–52 carbon dioxide, 17

511

Gibbs free energy, 464 GIMRT/OS3D, 76 global C cycle, 273 Global Carbon Capture and Storage Institute, 14 global climate change, 2–8 global warming potential, 1 globules, 314 glucose, 463 glyoxylate, 414 golden-brown algae, 417 Gorgon Project, 83 green algae, 417 GreenGen, 14 greenhouse effect, 2 greenhouse gas, 2–8, 271, 411, 463 Haematococcus pluvialis, 424–5 Henry’s law constant, 305 Hitura nickel mine, 442 hydrocarbon reservoirs, 39–42 hydrocarbon systems, 186–7 hydrodynamic trapping, 66–7 hydrofracture, 197 3-hydroxypropionate cycle, 413–14 hypercapnia, 359 hyperspectral techniques, 335 ideal adsorbed solution theory, 138 IEA Weyburn CO2 Monitoring and Storage project, 85–6 IGCC see Integrated Gasification Combined Cycle Illinois Basin, 247, 248, 251 industrial waste, 307 infrared radiation atmospheric monitoring, 230–1 InSAR see satellite radar interferometry Integrated Gasification Combined Cycle, 9, 381 integrated nutrient management, 282 interfacial tension, 37, 109, 186 intertidal gastropod see Littorina littorea invasive monitoring, 219–25 ionic trapping, 70–2 IPCC see United Nations Intergovernmental Panel on Climate Change IPCC Special Report on CO2 Capture and Storage, 29


512

Index

iron, 418 ISO 2409, 483 J-aggregate dimers, 493 Joint Implementation, 291 Joule-Thomson effect, 256 JRC-2, 471 XRD spectra, 474 Kenicsy-type static mixer, 314 Knallgas bacteria, 413 Kyoto Protocol, 291, 426, 435 Laacher See, 231 Langmuir equation, 145 Langmuir–Blodgett films, 493 leakage, 204 legal accessibility, 31 life-cycle analysis, 379 Littorina littorea, 364 local-scale site selection criteria, 33 London convention on ocean dumping, 307 low-pressure shelf drying, 426 Lumen meter, 484 MAC Science M03XHF, 472 macroalgae fatty acids distribution in lipids, 389 lipid accumulation capacity, 385 performance, 390 magnesium hydromagnesite, 451 magnesium oxide, 433 magnesium silicates, 435 marine ecosystems carbon dioxide droplets/plumes finescale dynamics, 349–52 bubble/droplet dynamics, 350–1 CO2 and CO2-enriched seawater plumes fine-scale dynamics, 351–2 CO2/seawater system physical properties, 349–50 carbon dioxide leakage environmental risks and performance assessment, 344–67 future trends, 366–7 leak monitoring options, 365–6 leaks mitigation, 366

carbon dioxide physical and chemical behaviour, 346–55 impacts of carbon dioxide leakage, 358–65 brittlestars egg maturation disruption, 363 community structure and diversity, 362, 364 impact on growth and reproduction, 360–2 nitrogen cycling, 364–5 organism health and survival, 360 organism physiology, 359 sea urchin gut basement membrane disruption and morphology changes, 361 marine pH past and contemporary variability, 348 mass storage capacity, 48 McElmo dome, 326–7 measurement, monitoring and verification, 18 membrane seal, 187 mesocosm, 362 METSIM2, 151 Mettler Toledo InPro 6000 series, 472 microalgae, 416–18 biomass harvesting and drying, 425–6 CO2 fixation blue green algae, 418 diatoms, 416–17 Eustigmatophytes, 417–18 golden-brown algae, 417 green algae, 417 Prymnesiophytes, 417 combined CO2 fixation and biofuel production, 421–2 conceptual microalgal farming system, 427 energy production via microalgal biomass conversion, 423 farming, 419–21 large-scale microalgal farming systems, 423–5 lipid accumulation capacity, 385 performance, 390 raceway ponds vs tubular photobioreactors, 424


Index

strains studied for CO2 biomitigation, 422 Micromeritics ASAP 2000, 472 micrometeorological techniques, 334 migration, 204 mild oxidant, 401 Minami-Nagaoka gas field, 83 mineral carbonation, 433–52 basic principles and methods, 435–8 costs related to energy use for mineral pre-treatment, 448 direct and indirect carbonation principles, 439 direct gas–solid route complete carbonation reaction times, 441 energy efficiency, 460–2 future trends, 451–2 MgO/Mg(OH)2 carbonation with supercritical CO2, 444 related issues, 447–51 advantages and limitations, 449–50 process energy efficiency, 447 in situ mineral carbonation, 450–1 results for 0.5 hour fluidised bed experiments, 446 technologies and potential applications, 438–46 early developments (1990-2000), 438–40 next five years (2000-2004), 440–3 recent developments (2005-today), 443–5 mineral trapping, 72–4 miscibility, 109–12 minimum pressure, 41, 112, 116 Monte Carlo predictions, 193–5 Mount Simon Sandstone, 247, 248 multicomponent seismic, 216, 217 multiple-contact miscible, 110, 111 Mytilus edulis, 359, 360 N3 dye, 484, 492 Nafion membrane film, 465 Nannochloropsis, 418 natural carbon dioxide springs, 331 natural weathering, 433 Near Zero Emission Coal, 14 Neochloris oleoabundans, 417 nesquehonite, 451

513

net primary productivity, 279 Nisku carbonate aquifer, 78 nitrification, 364 Nitrobacter, 415 Nitrosomonas, 415 non-invasive monitoring, 209–19 NUFT, 76 NZEC see Near Zero Emission Coal ocean acidification, 345, 367 ocean dumping, 307 ocean fertilisation, 418–19 oceanic pool, 273 oceanic sequestration anthropogenic CO2 sources, 308–9 carbon dioxide, 304–20 properties, 311–12 release modelling, 312–13 water–CO2 system phase diagram, 312 CO2/H2O/CaCO3 emulsion injection, 313–18 CO2 release on sloping seabed, 317 economics, 318 emulsion made from Kenics-type static mixer, 314 open ocean, 314–16 open ocean CO2 release, 315 open ocean globulsion plume length dependence on density stratification, 316 sloping continental shelf, 316–18 vertical globulsion plume length vs density stratification, 319 deep ocean storage history of proposals, 305–7 legal constraints, 307–8 future trends, 318 ocean structure, 309–10 liquid and supercritical CO2 and H2O density–pressure– temperature nomogram, 310 North Pacific sample pH profile, 311 North Pacific sample temperature profile, 309 oil reservoirs, 27, 105–6, 169, 170 CO2 sequestration and use for enhanced oil recovery, 104–23


514

Index

suggested characteristics for miscible CO2-EOR, 41 oilfields, 179–80 olivine, 434, 436, 462 Ophiura ophiura, 362 optical-fibre photoreactor, 484, 494–5 design, 481–2 experimental, 482–5 CO2 photoreduction in vapour phase, 484–5 TiO2-coated optical fibre preparation, 482–4 remaining photo energy in optical fibre, 482 schematic, 485 TiO2-coated optical fibre light transmission and spread, 481 ordinary differential equation, 255 original oil in place, 40 oxygen vacancies, 468 P25, 471 dye-sensitised photocatalyst, 492–4 UV-Vis spectroscopy of different thin films, 492 particle size distribution, 475 PATHARCH, 76 pentane, 384 PERC, 382 perfluorocarbon tracers, 335 permeability, 62 phosgene, 400, 401 photocatalysts characteristics, methanol yields and energy conversion efficiency, 476 dissolved O2 of two Cu/TiO2 during reaction, 480 element molar ratio, 476 time dependence on methanol yields, 479 XPS spectra Cu 2p on Cu/TiO2, 478 Ti 2p, 477 photocatalytic reduction, 397–8, 468 carbon dioxide, 463–97 advantages and limitations, 495 dye-sensitised P25 photocatalyst, 492–4

future trends, 495–6 influence of mixed oxides TiO2SiO2, 490–1 photoreduction on titanium dioxide, 469–70 photoreduction to hydrocarbons, 468–9 transition metal-loaded TiO2, 488–90 energy cycle using carbon as energy carrier, 464 fundamentals, 465–70 mechanism, 466–8 photoreaction on photocatalyst, 467 semiconductors and redox couples bandgap in aqueous solution, 466 JRC-2 and Cu/TiO2 XRD spectra, 474 methane and ethylene production rate over dye-sensitised P25 under artificial light, 493 under concentrated natural sunlight, 494 photoreactor in aqueous solution, 473 renewable energy, 470–95 CO2 in aqueous solution, 470–81 CO2 in vapour phase, 481–5 ethylene and methane production rate over various photocatalysts under UVA, 490 methane production rate over TiO2-SiO2 mixed oxide-based photocatalysts, 491 solar-energy harvest system concept, 497 photochemical reduction, 468 PHREEQC, 76, 77 PHREEQE, 75 phytoplankton, 418 Pickering emulsion, 314 picoplankton, 417 plagioclase rock, 450 pneumatic eruption, 259 polycarbonates, 398 polyethylene glycol, 482 polyurethanes, 398 Porapak Q column, 473 pore throat radius data, 192–3


Index powder photocatalyst characteristics, 473–7 preparation, 470–2 synthesis procedure, 471 Precautionary Principle, 307 precipitated calcium carbonate, 433 predictive flow modelling, 46–7 pressure monitoring, 219, 220, 222 pressurised fluidised bed, 444 pressurised thermogravimetric analyser, 440 Prymnesiophytes, 417 Psammechinus miliaris, 359, 362 pseudoternary diagram, 110 PSU-COALCOMP, 151 pyroxenes, 436 Rangely Field, 105, 108 RECOPOL project, 153 recoverable oil in place, 40 recovery factor, 40 reductive pentose phosphate cycle see Calvin cycle Regional Carbon Sequestration Partnerships, 14 relative permeability, 30 reliability, 30 remote sensing, 335 Reservoir Saturation Tool, 222 reservoir sealed pairs, 29 reservoirs, 17–18 residual saturation, 172 residual trapping, 67 reversed citric acid cycle, 413 ROIP see recoverable oil in place saline aquifers, 58–64, 240 characteristics, 61–4 capacity, 63–4 depth, 61–2 mineralogy and grain size, 62 porosity and permeability, 62–3 sandstone aquifers, 76–7 satellite radar interferometry, 233 Scenedesmus obliquus, 420 Scherrer equation, 474 Schottky barrier, 479 sea urchin see Echinocardium incordata; Psammechinus miliaris

515

Second Law of Thermodynamics, 460 sedimentary basins, 59, 60, 76–8, 257 carbon dioxide storage criteria for assessing and ranking suitability, 28–9 desirable characteristics, 32 eliminatory suitability criteria, 31 carbonate aquifers, 77–8 Nisku carbonate aquifer, 78 Tuscan Nappe limestone formation, 77–8 sandstone aquifers, 76–7 deep sand aquifers, 77 glauconitic sandstone, 76–7 tertiary gulf coast sediments, 77 Utsira Sand, Sleipner, northern North Sea, 76 serpentinites of Gruppo di Voltri, 78 sepiolite, 450 sequestration, 18 Seriola quinqueradiata, 360 serpentine, 434, 436, 462 serpentine aquifers, 78 serpentinite, 78, 435 shallow-focused monitoring, 233 shock front, 151 silicate mineral carbonation technology, 435 simple wave, 151 site characterisation, 43–7 site screening, 28 Sleipner, 209, 210, 211, 214 Sleipner project, 80 Snøhvit project, 80 soil gas measurements, 226–30 soil organic matter, 273 Solaronix, 484 solid-state dispersion method, 465 SOLMINEQ, 75 SOLVEQ, 75 SPE10, 177 Spirulina sp., 420 stabilisation wedges, 9, 11 static storage capacity, 36 steady-state injectivity, 115 Stokes’ law, 318 storage efficiency, 177, 179 sun drying, 426 SUPCRT92, 70, 75, 76


516

Index

supercritical carbon dioxide, 384 supercritical fluid chromatography, 384 superficial velocity, 145 surface 3D seismic, 209–16 Syngas, 389, 392, 400, 403 temperature monitoring, 219, 222 terrestrial ecosystem CO2 leakage environmental risks and impacts, 324–38 ASGARD field site, 337 CO2 atmospheric enrichment, 332–4 future trends, 336–8 hyperspectral index temporal variations, 336 leak monitoring techniques, 334–6 leak scenarios, 325–7 soil CO2 enrichment man-made analogues of, 330 natural analogues, 331 terrestrial leakage impacts, 327–32 CO2 enrichment laboratory studies, 328–30 elevated CO2 effects on soil fauna, 331–2 terrestrial pool, 273 components in different global biomes, 274 estimates in different biomes, 276 interaction with atmospheric C pools, 275 terrestrial sequestration basic principles, 279–90 carbon pool estimates in world soils, 281 different biomes net primary productivity, 280 nutrients in fine roots, 283 plant C pool, 281–4 soil organic carbon density in different agroecological zones, 281 soilC density and pool, 279–81 total root biomass in major world ecosystems, 282 carbon dioxide, 271–98 carbon sequestration technical potential in terrestrial ecosystems, 290 challenges, 291–5

ancillary benefits, 294 carbon sink capacity, 291 measurement and monitoring, 292, 294 permanence, 292 reference tables, 294 soil C pool and climate change, 295 technological options, 291–2 emissions from agricultural vs other activities, 276–9 carbon emissions estimates from farm operations, 278 different farm operations, 277–8 emission reduction from land use conversion and agricultural activities, 278–9 global biomass production estimates, 289 relative emissions from agriculture and forestry, 277 extrapolation, 295–6 all greenhouse gases, 296 biofuel issue, 296 charcoal and fire, 295 ecosystem carbon budget, 295 fine roots and turnover, 295–6 plant characteristics, 296 role of soil biota, 296 potential, 290–1 processes affecting carbon sequestration in soils, 285 soil and ecosystem matrix specific technological options, 293 soil and terrestrial carbon as indicators of climate change, 296–7 soil C sequestration, 284–90 biochar, 286 biofuel/energy plantation, 288 burying biomass, 288–90 degraded soils afforestation, 288 no-till farming, 284, 286 peat soil restoration, 286, 288 secondary carbonates, 290 soil carbon pool and fluxes as indicators of climate change, 297 soil organic carbon sequestration longterm rates, 294 terrestrial C sequestration determinants in soil and biota, 287


Index

terrestrial pool and it’s role in the global carbon cycle, 273–6 Texas Gulf Coast Basin, 246 thermal decompression, 384 Thiobacillus ferrooxidans, 415 Thiobacillus thiooxidans, 415 titanium dioxide CO2 adsorption, 469 particle size distribution, 475 photocatalytic CO2 reduction, 470 TiO2, Cu/TiO2 and Ag/TiO2 cross-section SEM images on optical fibres, 486–7 diffraction patterns, 487 film characteristics, 488 methanol yield vs light intensity, 489, 490 UV spectra, 488 transition metal-loaded TiO2 for CO2 photoreduction, 488–90 transmission electron microscopy photographs, 475 titanium (IV) butoxide, 471, 482 titanium oxide, 465 TOUGH2, 76, 83 TOUGHREACT, 75, 76, 77 transition metals, 395 trapping capacity, 172, 173 trapping efficiency, 179 trapping mechanisms, 64–74 tri-reforming, 403 Triassic Stuttgart, 85 triglycerides, 416 tubular photobioreactor, 424 Tuscan Nappe, 77–8 UBE process, 400 UK Climate Change Act, 7 UNCLOS see United Nations Convention

517

on the Law of the Sea underground reservoirs capillary seals for trapping carbon dioxide, 185–99 carbon dioxide injection design, 169–81 CO2 storage long-term safety mathematical modelling, 240–60 UNFCC see United Nations Framework Convention on Climate Change United Nations Convention on the Law of the Sea, 308 United Nations Framework Convention on Climate Change, 7–8 United Nations Intergovernmental Panel on Climate Change, 3 US Department of Energy Regional Carbon Sequestration Partnerships, 85 Utsira Sand, 76, 80, 210 vaporising-gas drive, 112 vertical seismic profiling, 223 VG Microtech MT500, 472 VSP see vertical seismic profiling WAG ratio, 113 water-alternating-gas, 113 wellbores, 255–9 wettability, 190 wollastonite, 436, 439 Xtool, 76 yellowtail finfish see Seriola quinqueradiata Zero Emission Research Detection, 337


Developments and Innovation in Carbon Dioxide (CO2) Capture and Storage Technology: Volume 2: Carbon Dioxide (CO2) Storage and Utilisation (Woodhead Publishing Series in Energy)

Developments and Innovation in Carbon Dioxide (CO2) Capture and Storage Technology: Volume 1: Carbon Dioxide (CO2) Capture, Transport and Industrial Applications (Woodhead Publishing Series in Energy)

Carbon capture and storage

Carbon Dioxide Populations and Communities

Solubility in Supercritical Carbon Dioxide

Carbon Dioxide and Environmental Stress

Oxy-Fuel Combustion for Power Generation and Carbon Dioxide (CO2) Capture

Carbon Dioxide Sequestration and Related Technologies

Carbon Dioxide and Terrestrial Ecosystems (Physiological Ecology)

Acid Gas Injection and Carbon Dioxide Sequestration

The Phanerozoic Carbon Cycle: CO2 and O2

Carbon Dioxide and Environmental Stress (Physiological Ecology)

Capture and Utilization of Carbon Dioxide with Polyethylene Glycol

Carbon Dioxide as Chemical Feedstock

Carbon Dioxide Angiography: Principles, Techniques, and Practices

Carbon dioxide and climate: a second assessment

Carbon Dioxide as Chemical Feedstock

Supercritical Carbon Dioxide in Polymer Reaction Engineering

Carbon Capture: Sequestration and Storage (Issues in Environmental Science and Technology)

Caching the Carbon: The Politics and Policy of Carbon Capture and Storage

Natural Extracts Using Supercritical Carbon Dioxide

Stand-alone and Hybrid Wind Energy Systems: Technology, Energy Storage and Applications (Woodhead Publishing Series in Energy)

Carbon Dioxide Utilization for Global Sustainability

Natural Extracts Using Supercritical Carbon Dioxide

Acid Gas Injection and Carbon Dioxide Sequestration (Wiley-Scrivener)

Cost and Performance of Carbon Dioxide Capture from Power Generation (IEA Energy Papers)

Transport Energy and CO2 : Moving towards Sustainability

Green Chemistry Using Liquid and Supercritical Carbon Dioxide

Energy Storage

Geological Storage of Carbon Dioxide (Geological Society of London Special Publication No. 233)

Capturing King Coal Deploying Carbon Capture and Storage Systems in the U.S. at Scale

Developments and Innovation in Carbon Dioxide (CO2) Capture and Storage Technology: Volume 2: Carbon Dioxide (CO2) Storage and Utilisation (Woodhead Publishing Series in Energy)

Developments and Innovation in Carbon Dioxide (CO2) Capture and Storage Technology: Volume 1: Carbon Dioxide (CO2) Capture, Transport and Industrial Applications (Woodhead Publishing Series in Energy)

Carbon capture and storage

Carbon Dioxide Populations and Communities

Solubility in Supercritical Carbon Dioxide

Carbon Dioxide and Environmental Stress

Oxy-Fuel Combustion for Power Generation and Carbon Dioxide (CO2) Capture

Carbon Dioxide Sequestration and Related Technologies

Carbon Dioxide and Terrestrial Ecosystems (Physiological Ecology)

Acid Gas Injection and Carbon Dioxide Sequestration

The Phanerozoic Carbon Cycle: CO2 and O2

Carbon Dioxide and Environmental Stress (Physiological Ecology)

Capture and Utilization of Carbon Dioxide with Polyethylene Glycol

Carbon Dioxide as Chemical Feedstock

Carbon Dioxide Angiography: Principles, Techniques, and Practices

Carbon dioxide and climate: a second assessment

Carbon Dioxide as Chemical Feedstock

Supercritical Carbon Dioxide in Polymer Reaction Engineering

Carbon Capture: Sequestration and Storage (Issues in Environmental Science and Technology)

Caching the Carbon: The Politics and Policy of Carbon Capture and Storage

Natural Extracts Using Supercritical Carbon Dioxide

Stand-alone and Hybrid Wind Energy Systems: Technology, Energy Storage and Applications (Woodhead Publishing Series in Energy)

Carbon Dioxide Utilization for Global Sustainability

Natural Extracts Using Supercritical Carbon Dioxide

Acid Gas Injection and Carbon Dioxide Sequestration (Wiley-Scrivener)

Cost and Performance of Carbon Dioxide Capture from Power Generation (IEA Energy Papers)

Transport Energy and CO2 : Moving towards Sustainability

Green Chemistry Using Liquid and Supercritical Carbon Dioxide

Energy Storage

Geological Storage of Carbon Dioxide (Geological Society of London Special Publication No. 233)

Capturing King Coal Deploying Carbon Capture and Storage Systems in the U.S. at Scale

Recommend Documents