Environmental and Human Health Impacts of Nanotechnology

Environmental and Human Health Impacts of Nanotechnology Environmental and Human Health Impacts of Nanotechnology Edite...

Author: Jamie R. Lead | Emma Smith

45 downloads 1199 Views 20MB 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

Environmental and Human Health Impacts of Nanotechnology

Environmental and Human Health Impacts of Nanotechnology Edited by Jamie R. Lead and Emma Smith © 2009 Blackwell Publishing Ltd. ISBN: 978-1-405-17634-7

Environmental and Human Health Impacts of Nanotechnology Edited by JAMIE R. LEAD School of Geography, Earth and Environmental Sciences, University of Birmingham, UK EMMA SMITH Department of Biological and Chemical Sciences, The University of the West Indies, Barbados

A John Wiley and Sond, Ltd., Publication

This edition first published 2009 © 2009 Blackwell Publishing Ltd Registered office John Wiley & Sons, Ltd, The Atrium, Southern Gate, Chichester, West Sussex, PO19 8SQ, United Kingdom. For details of our global editorial offices, for customer services and for information about how to apply for permission to reuse the copyright material in this book please see our website at www.wiley.com. The right of the author to be identified as the author of this work has been asserted in accordance with the Copyright, Designs and Patents Act 1988. All rights reserved. No part of this publication may be reproduced, stored in a retrieval system, or transmitted, in any form or by any means, electronic, mechanical, photocopying, recording or otherwise, except as permitted by the UK Copyright, Designs and Patents Act 1988, without the prior permission of the publisher. Wiley also publishes its books in a variety of electronic formats. Some content that appears in print may not be available in electronic books. Designations used by companies to distinguish their products are often claimed as trademarks. All brand names and product names used in this book are trade names, service marks, trademarks or registered trademarks of their respective owners. The publisher is not associated with any product or vendor mentioned in this book. This publication is designed to provide accurate and authoritative information in regard to the subject matter covered. It is sold on the understanding that the publisher is not engaged in rendering professional services. If professional advice or other expert assistance is required, the services of a competent professional should be sought. The publisher and the author make no representations or warranties with respect to the accuracy or completeness of the contents of this work and specifically disclaim all warranties, including without limitation any implied warranties of fitness for a particular purpose. This work is sold with the understanding that the publisher is not engaged in rendering professional services. The advice and strategies contained herein may not be suitable for every situation. In view of ongoing research, equipment modifications, changes in governmental regulations, and the constant flow of information relating to the use of experimental reagents, equipment, and devices, the reader is urged to review and evaluate the information provided in the package insert or instructions for each chemical, piece of equipment, reagent, or device for, among other things, any changes in the instructions or indication of usage and for added warnings and precautions. The fact that an organization or Website is referred to in this work as a citation and/or a potential source of further information does not mean that the author or the publisher endorses the information the organization or Website may provide or recommendations it may make. Further, readers should be aware that Internet Websites listed in this work may have changed or disappeared between when this work was written and when it is read. No warranty may be created or extended by any promotional statements for this work. Neither the publisher nor the author shall be liable for any damages arising herefrom. Cover photo courtesy of Dr. Ralf Kaegi, Swiss Federal Institude of Aquatic Science and Technology (Eawag) Library of Congress Cataloging-in-Publication Data Environmental and human health impacts of nanotechnology / edited by Jamie R. Lead and Emma Smith. p. cm. Includes bibliographical references and index. ISBN 978-1-4051-7634-7 1. Nanoparticles—Environonmental aspects. 2. Nanoparticles—Toxicology. 3. Nanostructured materials— Environmental aspects. 4. Nanostructured materials—Health aspects. 5. Nanotechnology—Environmental aspects. 6. Nanotechnology—Health aspects. I. Lead, Jamie R. II. Smith, Emma (Emma L.) TD196.N36E58 2009 620′.5—dc22 2009009688 A catalogue record for this book is available from the British Library. Set in 10 on 12 pt Times by SNP Best-set Typesetter Ltd., Hong Kong Printed and bound in Great Britain by CPI Antony Rowe Ltd, Chippenham, Wiltshire

Contents

Preface Biographies Contributors 1. Overview of Nanoscience in the Environment Mohamed Baalousha and Jamie R. Lead 1.1 1.2 1.3 1.4 1.5 1.6 1.7

1.8

1.9

1.10 1.11 1.12 1.13

1.14 1.15 1.16

Introduction History Definitions Investment and International Efforts Development: Four Anticipated Generations Applications of Nanotechnology Potential Benefits of Nanotechnology 1.7.1 Environmental 1.7.2 Human Health Potential Adverse Effects of Nanomaterials 1.8.1 Environmental 1.8.2 Human Health Classification 1.9.1 Chemistry 1.9.2 Origin 1.9.3 Size 1.9.4 State Sources of Nanomaterials in the Environment Properties of Nanomaterials Nanomaterial Structure–Toxicity Relationship Environmental Fate and Behaviour of Nanomaterials 1.13.1 Fate in Air 1.13.2 Fate in Water 1.13.3 Fate in Soil Potential for Human Exposure Detection and Characterization of Nanomaterials Issues to be Addressed 1.16.1 Nomenclature

xiii xv xvii 1 1 2 3 6 6 7 8 8 9 10 10 12 12 12 13 13 14 14 14 15 16 17 17 19 20 21 21 21

vi

Contents

1.16.2 1.16.3 1.16.4 1.16.5

Future Development and Risk Dosimetry Methods of Detection and Characterization Environmental Fate of Nanomaterials and their (Eco)Toxicology 1.17 Conclusion 1.18 References 2. Nanomaterials: Properties, Preparation and Applications Paul Christian 2.1 2.2 2.3

2.4

2.5

2.6

2.7 2.8 2.9

Overview Introduction Nanoparticle Architecture 2.3.1 Nanoparticle Surface 2.3.2 Charge Stabilisation 2.3.3 Steric Stabilisation Particle Properties 2.4.1 Surface Plasmon Resonance 2.4.2 Catalysis 2.4.3 Quantum Confinement 2.4.4 Mechanical Performance 2.4.5 Magnetic Properties 2.4.6 Interfacial Properties 2.4.7 Other Properties Nanoparticle Preparation 2.5.1 The Challenges of Nanoparticle Synthesis: Scale Up 2.5.2 Reactivity 2.5.3 Dispersability 2.5.4 Cost 2.5.5 Methods: Natural Sources 2.5.6 Top Down 2.5.7 Bottom Up 2.5.8 Metal Nanoparticles 2.5.9 Carbon 2.5.10 Graphene 2.5.11 Carbon Black 2.5.12 Inorganic Compounds 2.5.13 Polymers Applications of Nanoparticles and Nanotechnology 2.6.1 The Past 2.6.2 The Present and Near Future Implication for Environmental Issues Conclusions References

23 23 23 23 24 24 31 31 32 35 38 41 42 45 45 46 47 48 49 49 50 53 53 53 53 54 54 55 55 59 60 62 62 63 64 65 65 67 72 73 73

Contents

3. Size/Shape–Property Relationships of Non-Carbonaceous Inorganic Nanoparticles and Their Environmental Implications Deborah M. Aruguete, Juan Liu and Michael F. Hochella, Jr 3.1 3.2 3.3

3.4 3.5

3.6 3.7 3.8

4.4

4.5

4.6

79

Introduction 79 Inorganic Nanoparticle Anatomy 80 Redox Chemistry of Nanoparticles 81 3.3.1 Photoredox Chemistry in Semiconductor Nanoparticles 81 3.3.2 Redox Chemistry in Other Nanoparticle Systems 84 Size Effects in Nanoparticle Sorption Processes 87 Nanoparticle Fate: Dissolution and Solid State Cation Movement 89 3.5.1 Basic Energetic and Kinetic Considerations of Nanoparticle Dissolution 89 3.5.2 Effects of Nanoparticle Morphology 91 3.5.3 Effects of Nanoparticle Coatings and External Substances 92 3.5.4 Case Study: The Dissolution of Lead Sulfide Nanoparticles 94 3.5.5 Solid State Cation Movement in Nanoparticles 96 Effect of Nanoparticle Aggregation on Physical and Chemical Properties 98 Environmental Implications: General Discussion, Recommendations and Outlook 99 References 101

4. Natural Colloids and Nanoparticles in Aquatic and Terrestrial Environments Mohamed Baalousha, Jamie R. Lead, Frank von der Kammer and Thilo Hofmann 4.1 4.2 4.3

vii

Introduction Definition Major Types of Environmental Colloids 4.3.1 Inorganic Colloids 4.3.2 Organic Macromolecules Intrinsic Properties of Environmental Colloidal Particles 4.4.1 Size 4.4.2 Surface Charge 4.4.3 Surface Coating by Natural Organic Matter 4.4.4 Fractal Dimension Interaction Forces Between Colloidal Particles 4.5.1 DLVO Theory 4.5.2 Stability Criteria 4.5.3 Aggregation Kinetics 4.5.4 Non-DLVO Interactions Fate and Behaviour of Colloids in Aquatic Systems 4.6.1 Aggregation

109

109 112 112 114 117 121 121 121 123 124 126 127 129 129 130 136 136

viii

Contents

4.7

4.8 4.9

4.6.2 Disaggregation: Effect of Natural Organic Matter 4.6.3 Sedimentation Behaviour Fate and Behaviour of Colloids and Nanoparticles in Porous Media 4.7.1 Saturated Porous Media 4.7.2 Unsaturated Porous Media Conclusion References

5. Atmospheric Nanoparticles Aurélie Charron and Roy M. Harrison 5.1 5.2

5.3 5.4 5.5

5.6

5.7 5.8

Introduction Sources of Atmospheric Nanoparticles 5.2.1 Sources of Primary Nanoparticles 5.2.2 Secondary Sources Chemical Composition of Atmospheric Nanoparticles Fate and Behaviour of Atmospheric Nanoparticles Atmospheric Concentrations 5.5.1 Spatial Variations 5.5.2 Temporal Variations Measurement Methods for Atmospheric Nanoparticles 5.6.1 Particle Number Concentration 5.6.2 Surface Area 5.6.3 Mass Concentration 5.6.4 Chemical Composition Conclusions References

6. Analysis and Characterization of Manufactured Nanoparticles in Aquatic Environments Martin Hassellöv and Ralf Kaegi 6.1

6.2

Introduction 6.1.1 Nanoparticles in the Aquatic Environment 6.1.2 Concepts and Definitions Relating to Analysis and Characterization Nanoparticle Analysis and Characterization Methods 6.2.1 Important Nanoparticle Characteristics 6.2.2 Sampling, NP Extraction, Sample Preparations 6.2.3 Light Scattering Methods 6.2.4 Other Electromagnetic Scattering Methods 6.2.5 Fractionation and Separation Methods 6.2.6 Microscopic Methods 6.2.7 Spectroscopic Methods 6.2.8 Surface Area Measurements with Nitrogen Gas Adsorption

140 141 141 143 146 147 147 163 163 164 164 175 178 181 182 182 185 187 189 193 194 196 200 201

211 211 212 212 214 214 224 224 229 230 237 249 251

Contents

6.3

6.4 6.5 6.6

6.2.9 Method Validation Analytical Test Strategy in NP Exposure Assessment 6.3.1 Initial Material Characterization 6.3.2 Fate and Behaviour Assessment 6.3.3 Exposure Characterization in Effect Assessment Experiments 6.3.4 Monitoring Nanopollution Conclusions Acknowledgements References

7. Ecotoxicology of Manufactured Nanoparticles Simon C. Apte, Nicola J. Rogers and Graeme E. Batley 7.1 7.2

7.3

7.4 7.5

7.6 7.7

Introduction Physico-Chemical Transformation of Nanoparticles 7.2.1 Particle Dispersion and Aggregation 7.2.2 Nanoparticle Dissolution 7.2.3 Oxidation 7.2.4 Adsorption Reactions Mechanisms of Nanoparticle Toxicity in the Environment 7.3.1 Exposure Routes 7.3.2 Nanoparticle Interactions with Cells: Cellular Uptake 7.3.3 Toxicity Mechanisms 7.3.4 Bioaccumulation Development of Valid/Realistic Toxicity Testing Protocols Review of Ecotoxicity Studies 7.5.1 Overview 7.5.2 Carbon-Based Nanoparticles 7.5.3 Metal Oxides 7.5.4 Silver 7.5.5 Copper 7.5.6 Quantum Dots 7.5.7 Iron General Conclusions and Future Directions References

8. Exposure to Nanoparticles Robert J. Aitken, Karen S. Galea, C. Lang Tran and John W. Cherrie 8.1 8.2

Introduction Physical Characteristics and Properties of Nanoparticles 8.2.1 Terminology and Definitions 8.2.2 Nanoparticle Types

ix

251 252 252 252 253 254 255 256 256 267 267 269 270 272 273 275 275 275 277 280 283 284 285 285 289 293 296 298 298 299 300 301 307 307 309 309 311

x

Contents

8.3

8.4

8.5 8.6

8.2.3 Nanoparticle Production Processes 8.2.4 Nanoparticle Behaviour Nanoparticle Exposure 8.3.1 Exposure Scenarios 8.3.2 Exposure Metrics 8.3.3 Methods of Measuring and Characterising Exposure to Nanoparticles 8.3.4 Studies Investigating Nanoparticle Exposure 8.3.5 Numbers of People Potentially Exposed Control of Exposure 8.4.1 Introduction 8.4.2 Inhalation Exposure 8.4.3 Dermal Exposure 8.4.4 Ingestion Exposure Discussion References

9. Human Toxicology and Effects of Nanoparticles Vicki Stone, Martin J.D. Clift and Helinor Johnston 9.1

9.2

9.3

9.4 9.5 9.6 9.7 9.8

Introduction 9.1.1 Toxicology – What Is It? 9.1.2 Particle Toxicology 9.1.3 Risk Assessment Ultrafine Particle Toxicology 9.2.1 Air Pollution 9.2.2 Testing the Ultrafine Particle Hypothesis 9.2.3 Reactive Oxygen Species and Oxidative Stress 9.2.4 Uptake of Nanoparticles into Cells 9.2.5 Interaction of Nanoparticles with Defence Mechanisms 9.2.6 Nanoparticle Interactions with Other Pollutants and Molecules Engineered Nanoparticles 9.3.1 Fullerenes 9.3.2 Nanotubes and Other Fibre-Like Nanostructures 9.3.3 Metals 9.3.4 Metal Oxides 9.3.5 Quantum Dots Relating Physico-Chemical Properties to Toxicity: Structure-Activity Relationships Suggestions for Future Study Designs Conclusions Abbreviations References

314 316 319 319 327 330 338 344 346 346 347 349 350 350 353

357 357 357 357 358 360 360 361 363 365 366 367 368 369 370 373 374 377 379 381 381 382 382

Contents

10. Risk Assessment of Manufactured Nanomaterials Sophie A. Rocks, Simon J. Pollard, Robert A. Dorey, Paul T. C. Harrison, Len S. Levy, Richard D. Handy, John F. Garrod and Richard Owen 10.1 10.2 10.3 10.4 10.5 10.6 10.7 10.8 Index

Introduction Risk Assessment Process Nanomaterials – Issues for Risk Assessment Assembling Evidence for Safety and Intervention International Case Studies Data Gaps in Risk Assessment of Nanomaterials Summary References

xi

389

389 391 399 403 406 407 410 417 423

Preface Manufactured nanoparticles (NPs) are usually defined as materials purposefully produced by human activity and which have at least one dimension between 1 and 100 nm. It is important to distinguish NPs by source; the main other NPs are incidental: that is produced indirectly by human activities including fossil fuel combustion, and natural: that is produced by processes such as chemical hydrolysis, weathering and microbial action. Other size-based definitions of NPs exist and there are a wide variety of material types which fall within this definition. Nanoscience, which is the science dealing with nanoscale materials, can be seen as simply a subset of traditional colloids science. Nevertheless, a large number of novel processes occur below this size due to effects such as exponential increases in specific surface area and surface energy, quantum effects such as quantum confinement (where wave functions are constrained by the small particle size) and undercoordination of bonds at the particle surface. Processes which occur in this size range are thus different in many ways to traditional colloid chemistry and, in general, the differences become more pronounced at smaller sizes. The current interest in nanotechnology is due to these novel properties and their exploitation in industrial processes and consumer products. Huge and exponentially growing research and development funding from government and private sources has been spent to better develop and exploit these potential uses and NPs are now used widely. Silver NPs are currently used as bacteriocides in cosmetics, fabrics, medical and health-related products and elsewhere. Titanium dioxide NPs are used in sunscreens (along with zinc oxide) and self cleaning surfaces, where they have a photocatalytic effect on organic matter due to the production of reactive oxygen species (ROS) and because of this titania is also used as a bacteriocide. Cerium dioxide is widely used as an additive to diesel to improve fuel efficiency. A wide range of other materials such as carbon nanotubes, fullerenes, gold, iron, iron oxide and more exotic species are being developed and used. The extent of the applications and the possibility of unusual and unknown ‘nano’ effects has led to concern about their environmental and human health effects in the scientific community and equal concern in industry and from regulators and policy makers. A major driver for this in some quarters is undoubtedly the example of genetically modified organisms. The extensive public backlash has made the future of that technology quite uncertain and there has been a different approach in nanotechnology to openness and acknowledgement of the risks and a commitment to reducing these risks. Public response to nanoscience and

xiv

Preface

nanotechnology is currently limited by a lack of knowledge and wider impact but is generally positive with benefits expected in health, energy and the environment to name a few. Nevertheless, it is quite feasible that this attitude will change, particularly in view of developments in next generation nanomaterials, including selforganisation and self-assembly and the increasingly researched interface between ‘bio’ and ‘nano’. There are considerable benefits to be gained from the exploitation of nanoscience but current research tells us that there are indeed potential hazards in this area. It is incumbent on the relevant communities to ensure that NPs and other nanomaterials are used appropriately and designed and tested to be of minimal hazard and that exposure is not widespread; risk needs to be minimised and seen to be minimised to allow the full benefits of nanoscience and nanotechnology to be derived. Understanding the behaviour and impacts of nanotechnology in the environment and in human health is a daunting task and many questions remain to be answered: how do we measure concentrations of NPs in complex biological and environmental media?; what are the concentrations in environmental media and in organisms?; what are the correct metrics of measurement (mass or number concentrations for instance)?; what are the sources to the environment and humans?; what are the environmental transport pathways and ultimate sinks of NPs?; are NPs bioavailable and are they subject to bioaccumulation and biomagnification?; how do NPs distribute in the sub-cellular, organ and body environments?; how are transport, bioavailability and effects related to NP physico-chemical structure? Although a substantial amount of research is being performed, the research spending on the risks of nanotechnology and the health and safety and environmental implications is still tiny in comparison to its development and exploitation. This balance is unlikely to change enormously but there are good arguments to say that this should happen and change should come quickly. The questions above and related questions remain unanswered in the main and the purpose of this volume is to collate and discuss our current knowledge and point to future areas of research which are required. We would like to acknowledge and thank a number of people and institutions which made this book possible. The UK Natural Environment Research Council (NERC) provided funding via a Knowledge Transfer Network entitled Engineered nanoparticles in the natural aquatic environment (Nanonet), which enabled all authors and editors to convene for a two-day workshop to discuss the issues and finalise the chapters. We would like to thank the chapter authors for their efforts and their timely submissions, and the patience and help of the publishing team which was essential to the editors. Jamie Lead Emma Smith March 2009

Biographies

Jamie Lead is Professor of Environmental Nanoscience in the School of Geography, Earth and Environmental Sciences, University of Birmingham, UK. Professor Lead completed his PhD at Lancaster University, UK, in 1994 after investigating lanthanide and actinide speciation in natural waters and soils. At the same institution he later undertook postdoctoral research on the impact of size of natural aquatic colloids on transition metal chemistry. In 1998, he undertook further postdoctoral work at Geneva University, Switzerland, developing and using fluorescence correlation spectroscopy to quantify diffusion coefficients of natural organic macromolecules. In 2000, he became a Lecturer at the University of Birmingham and became full Professor at Birmingham in 2008. Professor Lead is Director of the Facility for Environmental Nanoparticle Analysis and Characterisation (FENAC), which is a national UK centre collaborating with the biological community investigating nanoparticle fate and effects. He has been a visiting researcher at CSIRO, Australia, and is a Fellow of the Royal Society of Chemistry, the International Union of Pure and Applied Chemistry and the Institute of Nanotechnology. Professor Lead’s main research interests, where he has published widely, relate to the relationships between chemistry, transport and bio-uptake of pollutants, especially in relation to the nanoscale in the environment. In particular, he is interested in the structure of natural ‘nanocolloids’ and the role this has in metal and manufactured nanoparticle chemistry, fate and behaviour. He is currently collaborating extensively with the ecotoxicological community by synthesising nanoparticles of silver, cerium, iron oxide and other materials and ensuring their full characterisation. These collaborations are particularly focussed on investigating mechanisms of nanoparticle biological uptake and effects

xvi

Biographies

Dr Emma Smith is currently Lecturer in Environmental Chemistry at the University of the West Indies. She received a degree in Oceanography and Chemistry from the University of Liverpool and a Masters in Marine Resource Development and Protection with distinction from Heriot Watt University. Her PhD thesis, Unresolved Complex Mixtures of Aromatic Hydrocarbons in the Marine Environment: Solubility, Toxicity and Photodegradation Studies, was carried out at Plymouth University in conjunction with Plymouth Marine Laboratory and won the SETAC Young Scientist Award in 2000 at World Congress in Brighton. Dr Smith then worked at Plymouth University on the characterisation of bioaccumulated and unidentified agent(s) causing reduced scope for growth in mussels and the potential ecological effects of chemically dispersed and biodegraded crude oils. She then worked at the University of Toronto within the Environmental NMR Centre, evaluating climatic controls on soil organic carbon composition and potential responses to global warming. Following this Dr Smith worked with Professor Lead at the University of Birmingham implementing the Nanonet project, a Knowledge Transfer (KT) Network in the area of manufactured nanomaterials (MNs) in the natural aquatic environment. In her current position she is responsible for teaching environmental chemistry, oceanography and ecotoxicology at UWI and is working with the Caribbean Ecohealth Programme and an EU Outreach project on assessing the potential environmental and human health effects of pollution.

Contributors

Robert J. Aitken

Institute of Occupational Medicine, Edinburgh, UK

Simon C. Apte Centre for Environmental Contaminants Research, CSIRO Land and Water, Bangor, Australia Deborah M. Aruguete Department of Geosciences, Virginia Polytechnic Institute and State University, Blacksburg, USA Mohamed Baalousha School of Geography, Earth & Environmental Sciences, University of Birmingham, Birmingham, UK Graeme E. Batley Centre for Environmental Contaminants Research, CSIRO Land and Water, Bangor, Australia Aurélie Charron Transport and Environment Laboratory, INRETS – French National Institute for Transport and Safety Research, Bron, France John W. Cherrie Institute of Occupational Medicine, Edinburgh, UK Paul Christian School of Chemistry, University of Manchester, Manchester, UK Martin J. D. Clift Institute for Anatomy, Division of Histology, University of Bern, Bern, Switzerland Robert A. Dorey Microsystems & Nanotechnology Centre, Cranfield University, Cranfield, UK Karen S. Galea


John F. Garrod Department for Environment, Food and Rural Affairs, London, UK

xviii

Contributors

Richard D. Handy School of Biological Sciences, University of Plymouth, Plymouth, UK Paul T. C. Harrison Institute of Environment and Health, Cranfield Health, Cranfield University, Cranfield, UK Roy M. Harrison School of Geography, Earth & Environmental Sciences, University of Birmingham, Birmingham, UK Martin Hassellöv Department of Chemistry, University of Gothenburg, Gothenburg, Sweden Michael F. Hochella, Jr Department of Geosciences, Virginia Polytechnic Institute and State University, Blacksburg, USA Thilo Hofmann Austria

Department of Environmental Geosciences, Vienna University,

Helinor Johnston Applied Research Centre for Health, Environment and Society, Edinburgh Napier University, Edinburgh, UK Ralf Kaegi Swiss Federal Institute of Aquatic Science and Technology (Eawag), Dübendorf, Switzerland Jamie R. Lead School of Geography, Earth & Environmental Sciences, University of Birmingham, Birmingham, UK Leonard S. Levy Institute of Environment and Health, Cranfield Health, Cranfield University, Cranfield, UK Juan Liu Chemical and Material Sciences Division, Pacific Northwest National Laboratory, Rickland, USA Richard Owen

School of Biosciences, University of Westminster, London, UK

Simon J. Pollard Collaborative Centre of Excellence in Understanding and Managing Natural and Environmental Risks, School of Applied Sciences, Cranfield University, Cranfield, UK Sophie A. Rocks Collaborative Centre of Excellence in Understanding and Managing Natural and Environmental Risks, School of Applied Sciences, Cranfield University, Cranfield, UK Nicola J. Rogers Centre for Environmental Contaminants Research, CSIRO Land and Water, Bangor, Australia

Contributors

xix

Vicki Stone Applied Research Centre for Health, Environment and Society, Edinburgh Napier University, Edinburgh, UK C. Lang Tran


Frank von der Kammer Department of Environmental Geosciences, Vienna University, Austria

UV/VIS & metals (scaled to fit) (a.u.)

1.2

Figure 4.2 FFF–ICPMS relative particle and element size distribution of aquifer colloids. The grey area represents the UV/VIS signal at 260 nm (as turbidity) and is a measure for the total colloid concentration. The coloured traces show the distribution of the major elements iron, aluminium and manganese and of the trace element lead. The signals are scaled to fit the graph. (v.d. Kammer, Doubascoux, Lespes, unpublished.)

1.0

UV/VIS Al Fe Mn Pb

0.8 0.6 0.4 0.2 0.0 0

50

100

150

hydrodynamic radius (nm)

Figure 4.6 A tapping mode image of a humic layer that has a 1 × 1 µm2 area machined away in contact mode. Lines a–c that cut across the image are where the cross-sections below the image were taken. (Reprinted with permission from C.T. Gibson, I.J. Turner, C.J. Roberts, J.R. Lead, Quantifying the dimensions of nanoscale organic surface layers in natural waters, Environmental Science & Technology, 41, 1339–44. Copyright 2007, American Chemical Society.) Environmental and Human Health Impacts of Nanotechnology Edited by Jamie R. Lead and Emma Smith © 2009 Blackwell Publishing Ltd. ISBN: 978-1-405-17634-7

200

0 250

Agglomeregation State

Concentration

Shape

Surface Speciation Size

Surface Charge

+ + + + +

Surface Functionality

Porosity / Surface Area

Size Distribution

Composition

Structure / Crystallinity

Figure 6.1 The important properties of manufactured nanoparticles in aqueous media are shown, indicating that the central concept of a homogeneous solid sphere with a clean surface is often an over-simplification. All or several of these properties are needed to understand the fate and behaviour of these nanoparticles in the environment or to characterize a certain ecotoxicology experiment. Therefore, a combination of analytical methods is required to obtain a complete characterization. (Figure partly adopted from Tinke et al., 2006.)

250 nm

250 nm (a)

3.0

3.0

2.0

2.0

1.0

1.0

0

1.0

0

3.0 µm

2.0

nm 200

nm 250

0

0

1.0

2.0

3.0 µm

–250

–200 0

1.0

2.0

3.0

µm

0

1.0

2.0

3.0 µm

(b)

Figure 7.8 (a) TEM images of Escherichia coli cells Left: untreated. Right: treated with halogenated MgO nanoparticles for 60 minutes. (b) Tapping mode AFM images of E. coli cells with the corresponding cross sections below. Left: Untreated (z-height 0–920 nm). Right: treated with halogenated MgO nanoparticles for 20 minutes (z-height 0–450 nm). Note the changes in smoothness and height of the cell indicating damage to the E. coli cell envelope upon nanoparticle treatment. (Reprinted with permission from P. K. Stoimenov, R. L. Klinger, G. L. Marchin and K. J. Klabunde, Metal oxide nanoparticles as bactericidal agents, Langmuir, 18, 6679–86. Copyright 2002 American Chemical Society.)

(a)

(b)

Figure 7.9 Uptake of nanoparticles by aquatic organisms. (a) Left: Silver nanoparticles in the membrane and inside of an Escherichia coli cell; right: EDS elemental mapping showing silver distribution through the sample. (b) Daphnia magna exposed to 5 mg/l of lipid coated single-walled nanotubes showing large numbers of tubes filling the gut track (1 h exposure) and clumps of precipitated tubes around the daphnid (20 h) (bar = 200 µm). ((a) Reprinted with permission from J. R. Morones, J. L. Elechiguerra, A. Camacho et al. (2005) The bactericidal effect of silver nanoparticles, Nanotechnology, 16, 2346–53. Copyright 2005 Institute of Physics. (b) Reproduced with permission from A. P. Roberts, A. S. Mount, B. Seda et al. (2007) In vivo biomodification of lipid-coated carbon nanotubes by Daphnia magna, Environmental Science & Technology, 41, 3025–9. Copyright 2007 American Chemical Society.)

1 Overview of Nanoscience in the Environment Mohamed Baalousha and Jamie R. Lead School of Geography, Earth and Environmental Sciences, University of Birmingham, United Kingdom

1.1

Introduction

Nanotechnology and nanoscience have a long tradition but in their current forms, where it is possible to image, manipulate and quantify materials on a nanoscale, they are relatively new. Understanding the environmental and human health impacts of these nanomaterials is itself very new and also highly multidisciplinary, requiring knowledge of environmental, analytical and physical chemistry, physics, materials science, (eco)toxicology and other disciplines. The breadth of the subject makes it demanding for anyone with a serious interest in studying the subject. Therefore, in this introductory chapter the current knowledge is briefly reviewed as a preamble to the more detailed analysis in subsequent chapters. Specifically, Chapter 2 discusses the architecture (structure and composition) of nanomaterials, the methods of producing stabilized nanoparticles, the properties, preparation methods and applications of nanomaterials and nanotechnology. Chapter 3 discusses the currently available knowledge about nanomaterials (colloidal inorganic material). It focuses on the chemical behaviour of nanomaterials that is likely to determine their environmental fate and toxicity, including size, redox chemistry, sorption processes, cation diffusion kinetics and dissolution. In Chapter 4 the available knowledge of natural aquatic and terrestrial colloids (including nanoparticles) is reviewed, including the major types of natural colloidal

Environmental and Human Health Impacts of Nanotechnology Edited by Jamie R. Lead and Emma Smith © 2009 Blackwell Publishing Ltd. ISBN: 978-1-405-17634-7

2


particles and their properties which are related to environmental processes, followed by a discussion of interaction forces between colloidal particles and their fate and behaviour in aquatic and terrestrial systems. Chapter 5 reviews the available knowledge about natural and adventitious nanoparticles in the atmosphere with a focus on their sources, transformations and concentrations. The analysis and characterization of manufactured nanoparticles in the environment are discussed in Chapter 6. It gives a general overview of the key properties that describe nanomaterials and the methods for sampling, extraction and sample preparation. This is followed by an extensive discussion of analytical tools for the characterization of nanoparticles, such as fractionation, filtration, microscopy and spectroscopic methods. Chapters 7 and 8 discuss the ecotoxicology and toxicology of manufactured nanoparticles, while Chapter 9 reviews the occupational health and exposure of nanoparticles. In Chapter 10 regulation, policy and risk management are discussed. This chapter starts by presenting the risk assessment framework for chemicals and then discusses the risk assessment of nanoparticles. It also discusses the critical issues for risk assessment of nanomaterials and the approach that should be adopted for this purpose.

1.2

History

The basic concept of nanotechnology was outlined by Nobel Prize winning physicist Richard Feynman in 1959 when he said ‘the principles of physics as far as I can see, do not speak against the possibility of manoeuvring things atom by atom. It is not an attempt to violate any laws; it is something, in principle, that can be done; but in practice it has not been done because we are too big’ (Feynman, 1960, 1992). The term ‘nanotechnology’ was first used by the Tokyo Science University professor Norio Taniguchi in 1974 to describe the precision of manufacture of materials at the nanometre scale. This term ‘nanotechnology’ became popular and in use in the public domain in 1980 when Eric Drexler published his book ‘Engines of Creation’. The advent of the scanning tunnelling microscope in 1981 and atomic force microscope in 1986 enabled atom clusters to be seen for the first time (Binnig et al., 1982, 1986). However, the history of nanoparticles goes back much further. ‘Soluble’ (or colloidal) gold appeared around the fifth to fourth century BC in Egypt and China and has been used for both aesthetic and curative purposes. In 1618, the philosopher and medical doctor Francisci Antonnii published a book which is considered as the first book about colloidal (nanoparticulate) gold. In 1676, the German chemist Johann Kunckels published a book in which he spoke about ‘drinkable gold that contains metallic gold in neutral, slightly pink solution that exert curative properties for several diseases’ and concluded that ‘gold must be present in such a degree of communition that it is not visible to the human eye’. In 1818, Jeremias Benjamin Richters noticed the formation of pink or purple solutions of fine gold and yellow solutions when the particles have aggregated (Daniel and Astruc, 2004) and, in 1857, Faraday reported the formation of red solutions of gold by the

Overview of Nanoscience in the Environment

3

reduction of an aqueous solution of AuCl −4 using phosphorus in CS2 (Faraday, 1857). Shortly after that, in 1861, the term ‘colloid’ (of which nanoparticles are the smallest fraction) was coined by Graham (Graham, 1861). This brief discussion shows clearly that nanoparticle usage has a long history. The novelty today is the scale of research and industry and the ability to manipulate and design materials at the nanoscale to create large structures with fundamentally new properties and functions. This will lead to unprecedented understanding and control of the properties of materials to discover novel phenomena, process and tools. Therefore, nanotechnology will enable a wide range of discoveries in all major scientific areas.

1.3

Definitions

Various definitions for nanomaterials have been given, or are actually in debate. Most of these are based on size and imply that there is a size range between that of molecules and bulk materials, where particles have unique properties different than those of molecules or bulk material (Tratnyek and Johnson, 2006). Some of these properties arise only for particles smaller than approximately 10 nm or so, where particle size approaches the length-scale of certain molecular properties (Klabunde et al., 1996). For instance, below 10 nm, particle specific surface area increases exponentially and qualitatively similar trends apply to related properties such as the ratio of surface/bulk atoms. Another example is that of quantum confinement, which arises because the band gap of semi-conducting materials increases as particle size decreases (Klabunde et al., 1996). The decrease in haematite particle size (from 37 to 7.3 nm) greatly promotes the oxidation of aqueous manganese (II) in the presence of molecular oxygen (Madden and Hochella, 2005), quite separate from the surface area related effect. Small magnetite nanoparticles (9 nm) exhibit greater reactivity toward carbon tetrachloride (CCl4) relative to larger nanoparticles (80 nm), both on mass and surface area normalized bases (Vikesland et al., 2007). The decrease in size of ceria nanoparticle alters the oxidation state of the nanoparticles with an increase in the fraction of Ce3+ at sizes less than approximately 15 nm with complete reduction of ceria particles to Ce3+ at sizes less than approximately 3 nm (Wu et al., 2004). Size dependent inhibition of nitrifying bacteria has been observed and the inhibition was correlated to the fraction less than approximately 5 nm in the suspension (Choi and Hu, 2008). These properties of nanomaterials and others are discussed in more details in Chapter 2 and 3. Nanoscience is, however, generally defined more inclusively, as the scientific study of materials on the nanoscale, approximately defined as the length scale between 1 and 100 nm (Borm et al., 2006b). Nanotechnology, as defined by the United States National Nanotechnology Initiative, is ‘the research and technology development at the atomic, molecular or macromolecular levels, in the length scale approximately 1–100 nm; the creation, and use of structures, devices and systems that have novel properties and functions because of their small size; and ability to be controlled or manipulated on the atomic scale’ (NNI, 2004). The Royal Society

4


and the Royal Academy of Engineering define nanotechnology as ‘the design, characterization, production and application of structures, devices and systems by controlling shape and size at the nanometre scale’ (Royal Society and Royal Academy of Engineering, 2004). Nanomaterials are a major component of nanotechnology and can be defined as materials that have one or more dimensions in the range 1–100 nm (Lead and Wilkinson, 2006). Importantly, nanomaterials have novel properties that differ from those of the same material without nanoscale features. A recent attempt to develop a more structured approach has been published by the British Standards Institution (BSI) (BSI, 2007). In its ‘Terminology for nanomaterials’ it defines nanoscale as the ‘size range from approximately 1–100 nm’, an nano-object as a ‘discrete piece of material with one or more external dimensions in the nanoscale’ and a nanoparticle as a ‘nano-object with all three external dimensions in the nanoscale’. A nanomaterial is a ‘material having one or more external dimensions in the nanoscale or which is nanostructured’, with nanostructured being defined as ‘possessing a structure comprising contiguous elements with one or more dimension in the nanoscale’. Definitions are also provided for nanorods: ‘nanoobject with two similar external dimensions in the nanoscale and the third dimension significantly larger than the other two external dimensions’; nanofibres: ‘flexible nanorods’; and nanotubes: ‘hollow nanorods’. The term high aspect ratio nanoparticles can be used to refer to fibres, rods or tubes. These definitions are based largely on particle size and do not account for the issue of particle size distribution satisfactorily or the change in properties as a function of size. A modified definition for a nanomaterial/nanoparticle can be based on the variation of material properties with size. Other definitions are still under discussion and various relevant bodies, for example, the International Organisation for Standardisation (ISO), American Society for Testing and Materials (ASTM) and the Organisation of Economic and Co-operation Development (OECD) are currently working on precise and formal definitions and nomenclature. The nanoscale dimension in comparison to the known dimensional scale of the universe is shown in Figure 1.1 (Hochella, 2002). At the smallest end of the scale (Figure 1.1a) are fundamental particles such as electrons and quarks, which are smaller than 10−18 m, and may approach 10−30 m in size or smaller, but such dimensions are not physically measurable at least at this time. At the larger end of the scale are the size of the Earth (107 m in diameter) and the sun (109 m in diameter). The nanoscale with other related objects is described in Figure 1.1b and is in the range 1–100 nm. A nanometre is a billionth of a metre (i.e. 10−9 m). The size of a single atom is of the order of several angstroms (0.1 nm). The size of a bacterium is about 1 µm (1000 nm), approximately the limit of visibility in light microscopes. In contrast, 100 nm is approximately equal to the size of a virus. Nanoparticles, like viruses, cannot be detected through standard light microscopes, because they are smaller than wavelengths of light (approximately 400–700 nm). They can be observed only with higher resolution microscopes such as scanning electron microscope (SEM, resolution of the order of 10 nm), transmission electron microscope (TEM) and atomic force microscope (AFM), which both have a resolution of the order of 400 °C) and closed and carefully controlled atmospheres. These factors combine to make solar cells expensive, rigid devices with limited production volumes. These restrictions can be traced to the simple requirement of having to prepare thin films of the various components. Other methods exist which allow the preparation of thin films, for example spin coating is routinely used to prepare thin films of polymers. However, the insolubility of common semiconducting materials used in the preparation of solar cells makes this approach unsuitable. The advent of nanotechnology allows the preparation of stable suspensions of nanoparticles that may be processed in a similar manner to a solution of the bulk material. This means that spin coating technology can be used to prepare solar cells directly from nanoparticles in an on-line fashion using flexible materials as the support. The application of this technology has been repeatedly demonstrated for solar cells based both solely on nanoparticles, such as cadmium telluride and cadmium selenide, as well as cells based on semiconducting

Nanomaterials: Properties, Preparation and Applications

53

polymers, such as polythiophenes in conjunction with a nanoparticle such as cadmium selenide. Some of the current challenges lie in being able to prepare such solar cells with high efficiency and using more benign materials.

2.5

Nanoparticle Preparation

There is an ever increasing number of methods and techniques being developed to prepare and manipulate nanoparticles. Whilst some of these methods have been commercialised on a large scale many have yet to be scaled up to mass production. As already seen, nanomaterials are actually complex mixtures. Therefore, having an understanding of how they are prepared can be key to understanding their behaviour. There are several challenges currently facing the community which need to be overcome before mass production of many materials can become a reality and, as will be seen later, the exact quality, properties and amounts of materials required can vary widely from application to application. Therefore, some time is spent at the end of this chapter considering the applications in which nanoparticle have been, are being and may be used. 2.5.1 The Challenges of Nanoparticle Synthesis: Scale Up The so-called bottom-up approach to nanoparticle preparation currently offers the best route to mass produced nanoparticles. This type of preparation method essentially builds nanoparticles up from molecules or atoms. It generally requires low concentrations of nanoparticle to be prepared in order to maintain a narrow size distribution of nanoparticle diameters. Simply increasing the concentration of these reactions generally results in the formation of larger nanoparticles with a wider distribution of sizes. For some applications this is not a significant issue; for others, however, it is critical. 2.5.2

Reactivity

As already discussed, nanoparticles have very large surface areas. This can make them prone to reactions at the surface which modify and degrade the performance of the particles. Many nanomaterials will undergo oxidation or hydrolysis at the surface, resulting in the alteration of the surface chemistry of the nanoparticles. It is well known that the surface chemistry can be key to the properties of the nanoparticle. In fact, in some cases it has been speculated that this oxidation process is the reason why some nanoparticles exhibit the properties which are desired. Perhaps the only exception to this problem is oxide-based nanoparticles, those of noble metals and some polymer nanoparticles. Attempt to circumvent this problem currently focus on developing new capping agents and surfactants and in producing layered nanoparticles with chemically inert coatings. 2.5.3

Dispersability

The ability to form a stable colloidal dispersion of many nanoparticles is key to their processing and incorporation into the final product. In some cases, enhance-

54


ment of performance can be achieved despite some agglomeration of the nanoparticles. However, this might be improved further if a macroscopically homogeneous dispersion could be prepared. The issue of dispersion is closely related to that of reactivity in that the search for a better capping agent to prevent further reaction at the surface once the nanoparticle has been prepared must also meet the need to produce stable dispersion of nanoparticles. In addition to this, it is advantageous if the dispersions remain stable at a range of concentrations, so that very concentrated dispersion of nanoparticles may be prepared and then processed in order to prepare the final material or device. 2.5.4

Cost

The preparation of some nanoparticles requires the use of very expensive reagents and solvents. This, in many cases, increases the cost and complexity of scale up significantly. The need to find low cost, simple routes to materials is key in scaling up the preparation of certain products. 2.5.5

Methods: Natural Sources

Perhaps the simplest way to obtain a nanomaterial is to use a raw mineral with a minimal amount of post processing. A few commercial nanomaterials are prepared by simple processing of minerals. For example, asbestos has six main forms (Table 2.4), of which chrysotile, amosite and amphibole were the main forms. They are employed in applications along with small amounts of termolite. Asbestos fibres are silicate-based minerals and are either magnesium or iron silicates. They have a highly anisotropic wire-like structure with diameters in the nanometre range and lengths of several microns. It is well known now that asbestos and other fibrous materials with similar length scales are very harmful to human heath (Castleman, 2006) and are no longer in widespread use in the many countries. Another form of nanomaterial which has found commercial application is the so called nano-clays. These are a post processed form of layered silicates such as montmorillonite. These minerals have a layered structure with intercalated metal ions which serve both to balance the charge on the silicate plates and to bind the layers together. These layers may be separated by exchanging the metal ions for

Table 2.4 Forms of asbestos, their common names and formulas. (Other names may be used and can be found in Nolan et al., 1999) Name

Mineral group

Asbestos type

Formula

Chrysotile Actinolite Tremolite Grunerite Anthophyllite Riebeckite

serpentine amphibole amphibole amphibole amphibole amphibole

white

Mg3Si2O5(OH)4 Ca2(Mg,Fe)5Si8O22(OH)2 Ca2Mg5Si8O22(OH)2 (Fe(5–7), Mg(2–0)) Si8O22(OH)2 (Fe2+, Mg)7Si8O22(OH)2 Na2Fe32+(Fe 2+ , Mg )3 Si8O22(OH)2

brown grey blue


Interchelation

55

Polymer mixing

C12H25NH3+Br-

Figure 2.14 Steps in the production of a nanocomposite based on exfoliated clays.

quaternary ammonium ions with long aliphatic chains or by melt processing with a suitable polymer such as nylon (Figure 2.14). This then results in a powder that can be further compounded with a polymer to produce a clay-reinforced nanocomposite that contains the platelets, which are well distributed through the composite. 2.5.6 Top Down Whilst some nanomaterials may be simply mined there are a whole range of materials which have to be prepared in a more laborious manner. There are two main methods for preparing nanomaterials, top down and bottom up. The top down approach works on the basis of breaking down a large piece of material into a smaller piece, in this case with dimensions in the nanometer range. This method can be used on most nanomaterials and is usually related to the patterning of a surface by either lithography and etching, or by electron or fast atom bombardment (Mendes et al., 2004). More recent interest has grown in the ability to use scanning near field pattering methods to chemically alter surfaces at resolutions down to 9 nm using conventional light (Leggett, 2006) or to use atomic force microscopy tips to plough soft polymer films on the nanoscale (Kunze, 2002). All of these methods result in essentially flat patterned surfaces; as these surfaces have some depth they are often termed 2½D techniques. If these techniques are used to produce a nanoscale pattern in a hard material they can then be used to prepare moulds and be further used for nano-imprinting and other related methods for more large scale production of devices. 2.5.7

Bottom Up

The bottom up approach relies on using small molecules to prepare the nanoparticles. For example, it is well known that the addition of a solution containing sulfate ions to a solution containing calcium ions will result in the rapid production of calcium sulfate. Calcium sulfate, being very insoluble in water, will form a precipitate, the solution will turn cloudy and the final product will settle out at the bottom

56


of the flask. The bottom up approach would use similar chemistry but seek to halt the growth of the particles of calcium sulfate before they became too large, thereby producing a stable dispersion of calcium sulfate where the particle sizes are less than 100 nm. The exact method used can vary enormously as will be seen here but generally the dimensions of the particles are controlled by careful control of the precipitation process. The preparation of nanoparticles by the bottom up approach relies on the principle of supersaturation (Zaizer and Lamer, 1948). In simple terms, at the start of the reaction there is a homogeneous solution of the reagents (Figure 2.15). As the reaction starts the product will begin to form; at very low concentrations it will have some solubility in the medium. As the concentration of the product increases the point of saturation will be reached. At this point the solution can no longer support the ever-increasing amounts of product being formed. However, as the product is being formed very rapidly and precipitation is limited by diffusion of the molecules in the media, which is comparatively slow, the point of saturation may be exceeded, producing a super saturated solution. At some point the supersaturation of the solution is relieved by the formation of the precipitated particles. These particles will be very small and represent the first nuclei of the final particles. It is now more favourable for the formation of more product to occur at the surface of the particle and therefore these first nuclei essentially act as seed for the growth of the particles. This is an idealised example of the processes involved and the occurrence the formation of second batches of seed nuclei is not unusual.

Supersaturation occurs

Concentration of Product

Saturation Concentration

Rapid relief of super saturation

Particle Growth

Rapid increase in product in solution

Time Figure 2.15

The various stages in the formation of an idealised nanoparticle.


57

There is still great debate in the literature over various other mechanisms for the formation of nanoparticles and there are several other processes that may contribute to the growth stage which are not discussed in detail here. The actual kinetic of the growth process may override any thermodynamic sink for the structure of the final material and this can result in particles which contain more than one crystal phase (Christian and O’Brien, 2005, 2008) This simple method for preparing nanoparticles can take many forms and the medium in which supersaturation occurs can be as diverse as a plasma or a conventional solvent. 2.5.7.1

High Temperature Methods

There are several varieties of high temperature methods for the production of nanoparticles. Generally they are well suited to the formation of either metal oxide or metallic nanoparticles and have been used to prepare a range of commercial materials. One excellent method for the preparation of nanoparticles on a large scale is aerosol flame synthesis or spray pyrolysis. The process is relatively simple. A solution is prepared containing the ions required for the synthesis. The solution is then injected into a high temperature flame of up to 3000 K and the particles essentially form in the flame and are collected as they settle out of the atmosphere. Materials prepared by this method tend to be crystalline and well formed. The method has been used for some time for the preparation of nanoparticles of materials such as carbon, titania, silica and alumina at rates of tens of tonnes per hour (Ulrich, 1984). The result of this type of preparation methods is a nanomaterial with an uncoated surface. It is, therefore, not unusual for the particles to be aggregated and difficult to redisperse. There are some excellent reviews on the application of this method to the preparation of metal and metal oxide nanoparticleS with application in catalysis (Wooldridge, 1998; Pratsinis, 1998). 2.5.7.2 Wet Methods There are several variations on wet methods for the preparation of nanoparticles. The term refers to the use of a solvent in which the reaction is performed. There are two subsets of this method: micelle encapsulation and arrested precipitation. Many of these methods can be used with out supplying energy to the system. However, in cases where energy is required to initiate the reaction there is a lot of work investigating the use of light, cavitation, ionising radiation and microwaves instead of thermal heating. Micelle encapsulation Micelle encapsulation relies on the use of a micelle to control the particle size. As already discussed, it is well known that certain surfactants will form micelles where an oil may be dispersed in water. In fact, it is also possible to form inverse micelles, where water is dispersed in a hydrophobic medium. These water-in-oil microemulsions are used as reactors which control the particle growth, size and, in some cases, shape. Two microemulsions are prepared containing the two reagents (Figure 2.16); for example, if cadmium sulfide was to be prepared then one emulsion might contain cadmium chloride and the other sodium sulfide. Upon mixing the micelles

58

Environmental and Human Health Impacts of Nanotechnology Microemulsion of Na2S

Microemulsion of CdCl2

Mixing of Solutions

Nanoparticles of CdS are formed in micelles

Figure 2.16

Preparation of nanoparticles of cadmium sulfide in a microemulsion.

collide, combine and reform very rapidly, resulting in the mixing of the core solutions whist retaining the micelle structure. The formation of the nanoparticle is then constrained by the micelle itself. This type of approach has been applied to both oil-in-water and water-in-oil emulsions as is discussed later. Arrested precipitation There are some similarities between arrested precipitation and micelle encapsulation, notably that both employ a surfactant to control the particle size. However, whilst in the micelle method the surfactant forms structures which contain the nanoparticles, in arrested precipitation the surfactant simply partitions to the surface of the particle as the particle forms. Micelles are very unstable at elevated temperatures (>50 °C) whereas arrested precipitation methods have been applied at temperatures exceeding 300 °C. During an arrested precipitation reaction (Figure 2.17) generally a solution of one reagent is heated to the required temperature and then treated with a solution of the second reagent. It is not uncommon to conduct the whole reaction in the surfactant or capping agent alone with out any further solvent. The final particles are washed to removes some of the excess capping agent. The overall outcome of this is that flame pyrolysis type preparation methods will result in very pure materials, but with no surface coating to ensure the formation


59

Addition of sulfur source

Solution of cadmium source in capping agent

Nanoparticles coated in capping agent

Figure 2.17 The preparation of a nanoparticle dispersion using the arrested precipitation method.

of stable suspensions, where as the micelle and arrested precipitation methods will results in stable dispersions under various conditions but with the added complexity of at least one further component to consider; the capping agent or surfactant. Furthermore, unless care is exercised in the purification of the materials there is likely to be contamination of the particle by either by products from synthesis or unreacted starting materials. Some of the starting materials for certain nanoparticles are exceptionally toxic. It is important to consider that the formation of nanoparticles by the bottom up approach is a dynamic process and that the surface of the nanoparticle cannot be considered to be unreactive. In fact, various processes are known to affect particle form long after the nanoparticles themselves have been prepared. There are two important factors. One is Oswalds ripening, which results in the sacrificial dissolution of small particles in favour of growth of the larger particles with lower surface energies. The rate of such a process is related to the distribution of particle sizes as well as the specific chemistry of the particles themselves. The other important factor is aggregation of the particles to form larger particles. These may become sintered into a polycrystalline larger particle. These processes have much to do with the fate of nanoparticles and are discussed further in Chapter 3. A brief overview of the types of methods used to prepare a range of nanomaterials is given below. These examples are by no means exhaustive. However, they will give an overview of the general methods, reagent and so on that may be used in preparing some nanomaterials. 2.5.8

Metal Nanoparticles

Metal nanoparticles are perhaps the earliest forms of nanoparticles prepared by man. The general method of preparation has changed little and generally relies on the reduction of a dissolved metal salt in the presence of a suitable capping agent or surfactant. The exact method employed depends on the metal. For example, gold

60


and silver nanoparticles may be prepared by simply heating an aqueous solution of gold chloride or silver nitrate with sodium citrate. This rapidly results in the formation of a red (gold) or yellow (silver) suspension of nanoparticles. This method relies on the binding of the citrate to the surface of the nanoparticle to impart stability and the colloid itself is charge stabilised; the citrate also acts as the reducing agent for the formation of the nanoparticle. The charge stabilisation of the nanoparticles means that these colloids are generally prepared at very low concentrations (800 °C). The thermal process results in particles with diameters of 150–500 nm and the furnace method 13–100 nm. The exact composition of these materials can vary, especially regarding the polyaromatic hydrocarbon (PAH) content. These are often present due to incomplete reduction of the oil starting materials.


2.5.12 2.5.12.1

63

Inorganic Compounds Oxides

Perhaps the most facile and widely used method for the preparation of metal oxides relies on the flame pyrolysis technique. This has been used to prepare a wide range of materials and is used to prepare the Degussa P45 titania commonly used in many ecotox studies. Such materials may also be prepared by the base catalysed hydrolysis of suitable salts. Silica and titania may be prepared by simple hydrolysis of an alkoxide (tetraethylorthosilicate or titanium isopropoxide) in an aqueous medium. Careful tuning of the pH will result in the formation of a stable dispersion of nanoparticles and this is a common method for the production of some silica nanoparticles. The use of various capping agents has also been employed in these types of reaction with some success. One method which has proven useful for a range of nanoparticles is hydrothermal synthesis. In this method the reaction is conducted in a sealed bomb so that the temperature may be raised above the boiling point of the liquid under standard conditions. This increase in temperature and pressure can result in the formation of different nanoparticle phases and sizes. Nanoparticles of ceria (CeO2) and zinc oxide (ZnO) have been prepared by similar methods. In these cases it is important to add a stabiliser or use a micelle to control the particle size. For example, zinc oxide is readily prepared by the reaction of zinc nitrate with a suitable base (Hartlieb et al., 2007). Ceria nanoparticles have been prepared in a similar manner (Liu et al., 2007). 2.5.12.2

Narrow Band Gap Semiconductors

Narrow band gap semiconductors represent a large range of nanoparticles. These are usually compound of metals with p-block elements such as sulfur, selenium, tellurium or phosphorous. Often these materials contain heavy metals such as cadmium, mercury, lead, indium, antimony or bismuth. Clearly the composition of these materials contains elements which, when in their free ionic form, are known to be exceptionally toxic. Whilst their current use is limited their potential is large and therefore it is worth considering their preparation. The sulfides are perhaps the easiest to prepare and are readily prepared by the rapid reaction of a solution of the metal ions with sodium sulfide in the presence of a capping agent or within a micelle. This method has been used to prepare both particle and rod shaped forms of cadmium sulfide (Simmons et al., 2002). However, the most common method for the preparation of the other chalcogenides and phosphides was initially described in 1991 (Murry et al., 1993). More recently there have been developments of similar methods to prepare metal nitrides such as gallium and aluminium nitride (Wells and Janik, 1996). In this general approach a metal precursor is dissolved in a suitable capping agent such as an alkylphoshine oxide and heated. The second precursor, such as selenium, is dissolved in a phosphine, such as trioctylphosphine, and rapidly injected into the reaction. The resulting nanoparticles are precipitated by the addition of methanol. This can result in

64


nanoparticles with excellent optical properties. A range of precursor molecules have been used including dialkyl adducts, such as dimethyl cadmium, as well as simple carboxylates. The capping agents used range from phosphines and phosphine oxides to amines, thiols, pyridines and carboxylic acids. Of particular note is the use of phosphine gas in the preparation of metal phosphides such as indium phosphide. Another approach to this preparation is the use of a single molecule precursor. In this case a molecule is prepared which contains all of the atoms needed to prepare the final particle. This molecule is then dissolved in a phosphine and added to a flask of hot phosphine oxide. Clearly, in many of these cases contamination by by-products is likely and high purity hard to attain or measure. Furthermore, whilst most of these nanoparticles are prepared in a form which will not suspend well in aqueous media, the addition of a polymeric surfactant is often enough to facilitate phase transfer from a hydrophobic environment into the aqueous phase. 2.5.13

Polymers

There are essentially two methods for the production of polymer based nanoparticles. In most cases, if a nanoparticle of a pure polymer is required it may most easily be prepared by using a micro emulsion method. For example, a micro emulsion of styrene in water might be prepared by using sodium dodecyl sulfate as the surfactant. A free radical initiator is added to the aqueous phase, for example hydrogen peroxide or ammonium persulfate, and the reaction heated for several hours. It is well known that this type of polymerisation tends to give excellent conversion of the monomer to the polymer. One of the reasons for the high conversion is the reaction kinetics within the micelles. Whilst transport of a radical from the water phase into the emulsion droplets is relatively slow, the reaction within the droplets is very rapid. This often results in an exponential increase in the viscosity of the monomer/polymer phase and tends to trap unreacted polymer ends within the particles. These unreacted polymer ends will still contain reactive radicals which have been used in the past to reinitiate polymerisations. The final nanoparticles will have a proportion of the surfactant bound tightly to their surface because the surfactants chain may either become entangled in the polymer chains or become grafted onto the polymer via side reactions. A second method for preparing polymer nanoparticles is to prepare a block copolymer (Figure 2.19). If a polymer is prepared with two sections that are soluble in different solvents it is possible to force them to self assemble on a molecular scale. As already discussed, polymers are molecules with lengths on the nanometre scale and, therefore, this molecular self assembly invariable results in a nanoparticle being formed. One way to achieve this is to couple to short polymers together, such as polycaprolactone and polyethylene oxide. Polycaprolactone is a water immiscible biodegradable polyester whereas polyethylene oxide is a biocompatible water miscible polyether. By preparing such a copolymer and dispersing it in water a biodegradable polymer nanoparticle may be prepared. In general, the dimensions of polymer nanoparticles are rarely less than 20 nm due to the large volumes which the molecules themselves occupy.


65

Block copolymer

Hydrophilic section

Hydrophobic section

Nanoparticle

Polymersome

Hydrophobic area

Aqueous core

Hydrophilic area

Figure 2.19 A block copolymer for the preparation of a nano-drug delivery system. Particles (bottom left) and polymersomes (bottom right) may be formed depending on composition and processing methods.

Clearly the preparation of nanoparticles is only part of the challenge. Often the next challenge is analysing them. As complex mixtures there are many different factors to consider. Conventionally, particle size and behaviour in various media are routine. Crystal phase, capping material and purity, however, are key factors which cannot be ignored. Various methods for determining these factors are outlined in detail in Chapter 6. As a minimum it is recommended that the following information is critical, although in some cases more information may be required: • • • • •

Particle size and distribution (including surface area). Hydrodynamic size. Crystal phase of the particle. Nature of the capping agent/surface functionalisation (including surface charge). Purity.

2.6 Applications of Nanoparticles and Nanotechnology 2.6.1 The Past The use of nanomaterials is not new. They were employed by nature long before man even thought about technology. Even man has been employing them for thousands of years. The Egyptians often added gold or cobalt to molten glass to colour it and make costume jewellery. The glass took on a red (gold) colour due to the surface plasmon resonances of the particles it contained. Lamp back has been used as a pigment for inks for thousands of years before carbon black particles saw more

66


widespread use. In the Graeco-Roman period mixtures of calcium hydroxide and lead oxide were used to dye hair, and it has been shown recently that this results in the formation of small (about 5 nm) nanoparticles of lead sulfide in the hair itself (Walter et al., 2006). Similarly, nanoparticles were applied in the glaze layers of pottery in the ancient world and there are examples from the eleventh century of shards of pottery painted with typical Islamic patterns which contain nanoparticles of silver and copper (Darque-Ceretti et al., 2005). Michael Faraday is one of the first people to take a scientific interest in gold sols or nanoparticles. He produced gold nanoparticles by the in situ reduction of gold chloride using white phosphorous and citrate as a capping agent. Carbon black has been produced on a commercial scale for more than 150 years and, similarly, asbestos has seen a rise and fall in use over the past 70 years. More recently, widespread commercial use of carbon black and asbestos probably represents the largest market for nanomaterials. 2.6.1.1 Asbestos Asbestos is a generic term for a range of naturally occurring minerals based on fibrous silicates. Asbestos is probably one of the best known nanomaterials in the general public and its detailed history provides an excellent background to the issues relating to very small particles and especially fibres. There are several reviews and books which deal in detail with the subject (Vallarino, 2001; Williams et al., 2007) and many more research papers and reports detailing its hazards. It is worth noting that as a material it is fairly ill defined, having a range of possible chemistries and an exceptionally broad range of sizes, which often fall at the upper scale of nanotechnology. The mineral crystallises in a fibre-like form which combined with its chemical composition results in its excellent insulation properties. The material itself is ill defined by current standards of engineered nanoparticles, consisting of a wide range of diameters (25–300 nm) and lengths (0.5–>20 µm) (Langer, 1974). The hazards of asbestos are well documented and are related to a combination of the chemistry of the material and its nanoscale, fibrous form. The initial applications of asbestos were in the ship building industry during the second world war where it was employed as a insulation material. By the 1960s the main user of asbestos became the building industry, accounting for more than 74% of the total mined material in the United States (NIOSH, 1972). By the 1980s there were about 3000 industrial or commercial products which had asbestos as one of the components (Anderson et al., 1982). Regulation in the 1970s (OSHA, 1986, 1994; US Environmental Protection Agency, 1988) resulted in a rapid decline in the market and use. The use of asbestos was on a scale of thousands of tonnes per annum and figures for 1979 show that, even after regulation, 130 000 tonnes per annum were being used in floor tiling alone (OSHA, 1986, 1994; US Environmental Protection Agency, 1988). It was even considered as a component of feminine hygiene products (OSHA, 1986, 1994; US Environmental Protection Agency, 1988), although there is no evidence that the concept was ever put into production. By 1997 the total asbestos market in the United States was still as high as 21 000 tonnes per annum distributed mainly in the building industry (Figure 2.20). The huge


67

Other (5%) Friction Products (29%)

Roofing materials (48%)

Gaskets (17%)

Paper (1%)

Figure 2.20 The relative use of asbestos in commercial and industrial applications in the United States in 1997. (Adapted from Spengler et al., 2001.)

market for asbestos and its dark history currently serves as a warning for future manufactures of nanomaterials and probably is currently serving as a break on the rate of commercialisation of nanotechnology in the west. This history of asbestos clearly shows not only the need to understand the toxic effects of nanomaterials, but also how a material with exceptional properties can rapidly find uses in a very wide range of products and applications. 2.6.1.2

Carbon Black

Carbon black containing rubbers have excellent wear properties and the neat material also unrivalled optical properties. In 1995 its world wide production was 7200 tonnes per annum (Gardiner, 1995). Its use as an ink, as already discussed, predates industrial production by thousands of years. However, as a source of black pigment it still has excellent properties. This is because the carbon particles absorb light over a wide range of the optical spectrum with little bleaching of the colour with time. Generally, an ink will actually comprise of a suspension of carbon black in a suitable carrier liquid; in ancient times this might have been an oil or water. To date, the use of carbon black in tyres accounts for the largest use of the material. The incorporation of carbon black into the rubber has a significant impact in increasing the wear properties of the rubber and thereby increasing tyre life. Several studies have been conducted on the effects of carbon black on human and animal health and there are some excellent reviews on the subject. Gardiner (1995) generally found that the carbon black itself has little adverse effect on organisms. However, recently particular concern has been placed on aerial exposures and carbon black particles containing significant levels of PAHs. It is sufficient to say that the effects of carbon black on human health are not considered to be as significant as those relating to asbestos. However, the history of carbon black give a good demonstration of how the complexity of a nanomaterial, in this case a chemically simple one, can cause difficulties in assessing their effects on human health. 2.6.2 The Present and Near Future The use of asbestos and carbon black date back more than five decades. In this section present applications of nanomaterials ( Kads (16 nm) (up to 70-fold increase) Kads much larger for bulk than for NP’s

Organic acids

6 nm, 16 nm

TiO2, anatase (Zhang et al., 1999) TiO2, anatase* (Gao et al., 2004)

Explanation suggested

Size dependence

Adsorbate

Particle sizes

Summary of particle size dependence adsorption studies.

Material

Table 3.1

Yes, greater for 20 nm than 300 nm

N/A

N/A

N/A

No, completely reversible

N/A

Hysteresis?

20 nm, 300 nm variable morphology, aggregated. NOM decreased As adsorption.

Aggregation state and morphology of particles not characterized 88 nm sample had variable morphology and more aggregation

Degree of aggregation not quantitatively assessed

Other comments

Size/Shape–Property Relationships

Γ = Γ max

89

KadsC 1 + KadsC

where C is the activity (effective concentration) of the adsorbate in solution, Kads is the adsorption reaction constant (related to the free energy change for the adsorption reaction), Γ is the number of molecules sorbed per unit area and Γmax is the adsorption capacity (the maximum number of molecules/ions per unit area that can be adsorbed). These parameters are referred to in the studies when applicable. (It should be noted that the Langmuir adsorption equation describes adsorption on a homogeneous surface. Because nanoparticle surfaces are relatively heterogeneous by their very nature, parameters derived using this model should be interpreted with caution.) As with many studies in the emerging field of nano-environmental science, interpretation of the results for size dependent trends is complicated by variation within and/or among the samples. The samples often not only vary in size but also in morphology, aggregation state and even crystal phase. All of these variables can affect particle behaviour. While these added variables can be accounted for via careful characterization, it is not always simple to do, and in some cases may be impossible. As synthetic methods advance further, it should become easier to synthesize or purchase more homogeneous and well characterised nanoparticle samples, and the results from such studies should become easier to interpret and compare.

3.5

Nanoparticle Fate: Dissolution and Solid State Cation Movement

Currently, the fate and degradation pathways of nanoparticles are unknown. One possible fate for nanoparticles is for them to dissolve. As many nanoparticles may contain toxic metals, this is a matter of concern. Here, what is known about nanoparticle dissolution is discussed, especially with respect to size and shape. Solid state cation movement and exchange processes are discussed as well, as these may also alter nanoparticle fate in the environment. 3.5.1

Basic Energetic and Kinetic Considerations of Nanoparticle Dissolution

Classically, the dependence of solubility upon particle size, assuming a spherical particle, can be expressed with a modified form of the Kelvin equation: S  2γ V  = exp   S0  RTr  where S is the solubility of particles with inscribed radius r in m, S0 is the solubility of the bulk material, γ is the surface free energy in mJ/m2, R is the gas constant in – nJ/mol⋅K, T is the temperature in K and V is the molecular volume in m3/mol (Adamson, 1982). According to this relation, as the particle dimensions decrease, the solubility increases exponentially relative to the bulk solubility. An example of

90


Figure 3.4 The deviation of the solubility of small grains of quartz relative to its bulk solubility (S/S0) as a function of the size of the quartz grains being dissolved according to the modified form of the Kelvin equation. The following values were used to produce this curve: T = 298 K, v¯ = 22.68 × 10−6 m3/mol, γ = 350 vmJ/m2. At a particle radius of 100 nm, the solubility is indistinguishable from the bulk value. By the time the particle radius is reduced to 1 nm, the predicted solubility is nearly three orders of magnitude higher. (Reprinted from M.F. Hochella Jr., Nanoscience and technology: the next revolution in the Earth Sciences, Earth and Planetary Science Letters, 203, 593–605, Copyright 2002, with permission from Elsevier.)

this relation is shown in Figure 3.4, which is a plot of

S versus particle radius S0

¯ values for quartz (Hochella, 2002). assuming γ and V Dissolution is generally assumed to be a spontaneous process. As long as particles are in a solution of constant undersaturation, the rate of dissolution should be constant. The relation of the normalized dissolution rate (in mol⋅m−2⋅min−1), R, can be related to the undersaturation, σ, via the relationship: R = kσ n

where k is the rate constant and n is the effective reaction order (Christoffersen et al., 1994; Budz and Nancollas, 1988). From these classical models of dissolution, smaller nanoparticles would be expected to dissolve more quickly than larger particles, and to dissolve to completion. For a number of systems, including nanoparticles of titanium dioxide (Schmidt and Vogelsberger, 2006), silica (Roelofs and Vogelsberger, 2004) and zinc oxide (Yang and Xie, 2006), smaller nanoparticles dissolve more quickly than larger nanoparticles. Despite this, it is not always clear whether classical models apply to


91

all nanoparticulate systems. In some systems, chemical processes not included in the classical models, such as photocatalyzed oxidation, may affect dissolution (Stouwdam et al., 2007; Aldana et al., 2001). Other experimental results indicate that small size does not always result in higher rates of dissolution. In one study of zinc oxides in aqueous systems, the same mass of nanoparticles and bulk solids dissolved at the same rate, even though the increased surface area and smaller size of the nanoparticles would warrant otherwise (Franklin et al., 2007). On the other hand, the nanoparticles in the experiment were highly aggregated, which may have lessened surface or size related effects. In some cases, dissolution at the nanoscale may be slower. Indeed, studies on various calcium phosphate minerals (Tang et al., 2001, 2003, 2004a, 2004b, 2004c, 2005; Tang and Nancollas, 2002) display a phenomenon of self-inhibited dissolution occurring primarily at the nanoscale, in which dissolution rates dwindle over time. To understand one of the means by which inhibited dissolution is possible, it is useful to consider the opposite process of nanoparticle growth from a solution. The energy of particle formation, ∆Gform, can be expressed as: ∆Gform = ∆Gv + ∆Gs where ∆Gv is the negative energy term describing the spontaneous tendency of solute to precipitate as part of a solid particle, and ∆Gs describes the excess free energy to form a new solid-liquid interface. ∆Gv depends upon the degree of saturation of the solvent. Assuming spherical particle morphology, both energy terms are functions of r, the radius of the particle. For a given level of saturation, there is a critical radius, r*, above which the magnitude of ∆Gv will be greater than that of ∆Gs and a particle can form (Tang et al., 2001). An analogous critical radius is believed to exist for dissolution processes. In dissolution, which occurs in an undersaturated solution, there is a favourable energetic driving force for units of the solid particle to become solute. However, the formation of etch pits can increase the area of the solid–liquid interface, which is energetically disfavoured. For such a system, there is a critical radius (of etch pit) at which dissolution is energetically allowed. Below this radius, dissolution is inhibited. Such inhibition phenomena have been observed in a number of systems and particularly well studied for various biologically relevant calcium phosphates (Tang et al., 2003, 2004a, 2004b, 2004c, 2005; Tang and Nancollas, 2002). It is therefore reasonable to expect the possibility of such inhibited dissolution occurring for nanoparticles of sparingly soluble compounds, because the nanoparticle dimensions may be below this critical radius. Currently, there is limited data to confirm or disprove these expectations. 3.5.2

Effects of Nanoparticle Morphology

Nanoparticles released into the environment will not only vary in size but also in their morphology, which may strongly affect dissolution. This particularly applies to cases in which the nanoparticles are crystalline. When nanoparticles of the same

92


crystalline substance assume different shapes, this generally means that different crystal faces comprise their surfaces. For example, consider nanoparticles of a rock salt structured mineral. A cubic nanoparticle displays {100} faces, a truncated cubo-octahedron displays {100}, {111} and {110} faces, and an octahedral nanoparticle will display {111} faces. Different crystal faces will be more or less stable (have different surface energies), depending upon their surface bonding. It is expected that less stable faces would be etched more readily than more stable faces. In our example, assuming that all other conditions are equal (same crystal structure, composition, solution undersaturation, etc.), this would mean that the three differently shaped nanoparticles might dissolve at different rates. While the energetic stability of crystal surfaces affects dissolving crystals of all sizes, it is particularly important for nanoparticles because even minimal dissolution may result in their annihilation. In principle, these concepts are simple, but applying them to quantitatively predict morphology dependent dissolution trends in nanoparticles is difficult. This is because little is known regarding the relative stabilities of nanoscale surfaces. The presence of surface defects, steps or kinks, which may be more evident on nanoparticle surfaces, will also influence the energetics of dissolution. Another complicating factor is the presence of coatings or other external substances, which are discussed in the following section. 3.5.3

Effects of Nanoparticle Coatings and External Substances

As with the surfaces of bulk materials (Zhang and Nancollas, 1990; Casey and Ludwig, 1995; Becker et al., 2005), it has been shown that external substances, particularly those that can coordinate to nanoparticle surfaces, can strongly influence nanoparticle growth and dissolution (Jun et al., 2006; Li et al., 2005, 2006b; Yin and Alivisatos, 2005). Anthropogenic nanoparticles released into the environment are likely to encounter many substances that could interact strongly with or sorb onto their surfaces, and many will already have coatings on their surfaces. One way such coatings or sorbed species may affect dissolution is by stabilizing particular crystal surfaces. Consider the partial dissolution of a truncated cubooctahedral nanoparticle composed of the rock salt structured material introduced in Section 3.5.2. The reaction coordinate for this process is displayed in Figure 3.5. Thermodynamically, the most favoured end product of dissolution for this system is a sphere. (This is not to imply that a sphere is always the most favoured shape for every system.) Imagine now adding a substance to the solution of nanoparticles which binds to and stabilizes the {100} and {110} crystal faces. This will increase the activation energy needed to obtain a spherical nanoparticle. Unless there is enough thermal energy in the system to surmount this kinetic barrier, it is likely that the process with the lower kinetic barrier (lower activation energy) will dominate. In this scenario, the {111} faces are energetically unstable relative to the other faces, so they will etch more readily. This etching results in an octahedrally shaped particle rather than a spherical particle. Coatings or external compounds can affect dissolution in other ways. For example, a coating that forms a micellar structure around a nanoparticle might reduce the


∆G†stab

Energy

{111}

93

{100} {110}

∆G†111

∆G†nostab Truncated cubo-octahedron

All {111} faces

Octahedron

sphere

Reaction coordinate Figure 3.5 Reaction coordinate for the partial dissolution (etching) of a truncated cubooctahedral nanoparticle. For this system, the thermodynamically favoured product is a spherical nanoparticle. In the absence of stabilizing coatings or ligands, formation of a spherical nanoparticle is kinetically favoured as well (the activation energy ∆G†nostab < ∆G†111 ). In the presence of compounds that stabilize the {100} and {110} faces, the activation energy to obtain a spherical nanoparticle increases. In this case, since this activation energy is greater than that necessary to form an octahedron (energy ∆G†stab > ∆G†111 ), formation of a sphere is no longer kinetically favoured. Etching the non-stabilized {111} faces is kinetically favoured. In the absence of enough thermal energy in the system, etching will result in an octahedral end product rather than a spherical one. (Figure adapted from Y.W. Jun, J.S. Choi and J.W. Cheon (2006) Shape control of semiconductor and metal oxide nanocrystals through nonhydrolytical colloidal routes. Angewandte Chemie, 45, 3414–39. Copyright Wiley-VCH Verlag GmbH & Co. KGaA. Reproduced with permission.)

activity of water at the surface. As for external compounds in the solution surrounding the nanoparticle, there are experimental examples of dissolution rates being increased by compounds such as acetic acid (Meulenkamp, 1998) or human serum albumin protein (Yang and Xie, 2006). External compounds might form stable ionic complexes with the constituent metal ions in the nanoparticle, hence energetically favouring dissolution. They also may alter pH, which again can affect nanoparticle stability. It is evident from such considerations that in any study concerning nanoparticle dissolution, as much as possible should be known regarding the composition of the coatings or the compounds in the solution surrounding the nanoparticle. These characteristics can be as significant as the composition of the inorganic part of the nanoparticle.

94

3.5.4


Case Study: The Dissolution of Lead Sulfide Nanoparticles

From the previous discussion of nanoparticle dissolution, it is evident that not only is size important, but morphology, coatings and molecules present in the surrounding solution. Currently, in our laboratory, the first two factors are being studied. The non-oxidative dissolution of ∼15 nm diameter lead(II) sulfide (PbS) nanoparticles in hydrochloric acid (pH 3) is being examined (Liu et al., 2007a). Bright-field transmission electron microscopy (TEM) is used to track changes in particle size and high-resolution TEM is used to measure changes in morphology and structure. The dissolution of nano-sized lead sulfide (galena) may have implications for the behaviour of both synthetic and natural nanomaterials in the environment. Lead sulfide is a low band gap semiconductor used in applications such as infrared detectors. Nanoparticles of lead sulfide are popular in nanoscience research and are commercially available. As for natural systems, it is known that nanoparticulate metal sulfides are present in some environments, and that mineral nanoparticles may be involved in the transport of heavy metals (Hochella et al., 2005b, 2008; Labrenz et al., 2000). Lead sulfide nanoparticles are synthesized under inert atmosphere in organic solution with surfactant via a previously published procedure at high temperature (Joo et al., 2003). This synthetic procedure produces monodispersed, highly crystalline nanoparticles, as confirmed with TEM and X-ray diffraction (XRD). After an initial washing procedure to remove excess free surfactant, nanoparticles are deposited onto a carbon/gold TEM grid substrate. Having the particles on a substrate helps to prevent aggregation, as this would complicate analysis. X-ray photoelectron spectroscopy (XPS) confirms that subsequent washing steps remove the majority of the surfactant (although undetectable trace amounts may remain) and that washing does not significantly affect the presence of any oxidation species on the nanoparticle surfaces. Washed, dried grids are exposed to nitrogen-purged hydrochloric acid solutions (pH 3) under constant stirring for varying periods. Images from samples exposed to the acid for different times are compared with each other using TEM measurements. Two interesting trends are summarized here. Firstly, the morphology of the lead sulfide nanoparticles changes after dissolution. From high resolution TEM measurements (Figure 3.6), the {110} and {111} faces are being etched more quickly than the {100} faces ({111} (data not shown). Such results match what might be expected from our knowledge of bulk crystals. Generally, on a crystal face, the rate at which an atom is removed from that face is inversely proportional to the number of bonds it has (Lasaga and Luttge, 2004). Atoms in the ideal bulk {110} and {111} faces have surface atomic coordination numbers of four and three, respectively, while the {100} faces have an atomic coordination number of five. Therefore, it would be expected that the {100} faces would etch more slowly than the {111} or {110} faces. At least for this system, these results indicate that some of our current knowledge about bulk crystal surfaces can be used to predict how nanoparticles might behave in the environment. Secondly, lead sulfide nanoparticles have been found to dissolve at surface area normalized rates higher than those for bulk lead sulfide by approximately one to two orders of magnitude. This difference in dissolution rate may be attributed to

Size/Shape–Property Relationships {110 }

0} {10

} 10 {1

(a)

95

{100}

(b)

{100}

d100

{1 10 }

d110

{100} { 11 0} d100

d110

(c)

Figure 3.6 (a) High resolution TEM images of a nanoparticle before dissolution (left) and a nanoparticle after 2 hours of dissolution (right). Note how the sizes of the {110} faces have increased. (b) Schematic diagrams of the distances measured on particles to determine whether the change in {110} size is statistically significant. (c) Distance ratios of d100/d110 before and after dissolution, along with mean values and 95% confidence intervals. As the {110} face size increases after 2 hours of dissolution, the value of the ratio increases.

the small size of the nanocrystals. As mentioned earlier, the modified Kelvin equation indicates that dissolution is more thermodynamically favoured for smaller particles. Also, due to their size, nanocrystals have a larger fraction of their atoms at corners and edges than bulk crystals. Such undercoordinated atoms are more active in dissolution than ones from flat surfaces. Nanoparticle morphology may

96


also play a role in faster dissolution. While bulk natural lead sulfide mostly displays {100} faces, the lead sulfide nanocrystals exhibit {111} and {110} faces. As described above, these faces dissolved more quickly than the {100} faces. These initial results have important implications for the dissolution behaviour of nanoparticles in the environment. Larger micro-sized particles for further size comparative rate studies are currently in the process of being synthesized. 3.5.5

Solid State Cation Movement in Nanoparticles

Another phenomenon that may affect nanoparticle degradation and fate is solid state cation movement into or out of nanoparticles. One type of cation movement is cation exchange, in which cations in solution replace cations in a lattice. Even if cation exchange does not occur significantly in the bulk form of a particular material (excluding perhaps on its surfaces), this does not preclude this process from happening fully in the nanoparticulate form or in thin films (nanoscaled films 2 and humins are insoluble at all pHs (Thurman and Malcolm, 1981). They cannot be fully defined as yet by structure or function, which are complex, temporally and spatially variable and highly influenced by solution conditions (Hayes et al., 1989). New technical developments have given more insight to the structures of humic substances, which have been explained in different ways including macromolecules (Swift, 1989), supramolecular association of small molecules (Piccolo, 2001), micelles (Thurman et al., 1982; Wershaw, 1999) and soft, semi-permeable spheres (Duval et al., 2005), although these structures are usually technique dependent (Lead and Wilkinson, 2006c). However, it is generally agreed that they are structurally complex macromolecules, strong adsorbers in the UV–Visible range (rich in chromophores) and are weak polyfunctional acids. Humic acids are generally terrestrial, while aquatic humic substances are dominated by FA, showing that the procedural definition has some use in interpreting geochemical behaviour. As with inorganic colloids, humic substances have different sources in different environments. The principle source of humic substances in terrestrial and freshwater environments is the degradation of higher plants, whereas the main source in marine systems is the degradation of plankton (Aiken et al., 1996; Kristensen, 1990). Terrestrial, freshwater and marine humic substances have significantly different chemical characteristics, with a significant contribution from autochthonous sources in freshwater. Freshwater humic substances, have a high C/N ratio (40 to 50), are rich in aliphatic carbon, have strong absorption in the near UV and are often depicted with a highly condensed, cyclic molecular structure. In contrast, marine humic substances have a low C/N ratio (15 to 20) (C/N ratio of fresh microbial material is between 5 and 10 to 1, for comparison), are rich in aromatic carbon, have weak absorption in the near UV and are often depicted with a more open, linear molecular structure. These differences arise from differences in the organic matter sources and formation processes of these two environments (Hedges et al., 1997). Humic substances have important environmental functions. They have an important role in regulating the chemical reactivity, speciation (Tipping, 2002), bioavailability and toxicity of metal ions in the natural environment (Koukal et al., 2003; McGeer et al., 2002; Van Ginneken et al., 2001). They also regulate the speciation (De Paolis and Kukkonen, 1997; Khan and Schnitzer, 1972), bioavailability and toxicity of organic pollutants (Haitzer et al., 1998). They play an important role in stabilizing inorganic and other colloids (Tipping and Higgins, 1982; Tipping and Ohnstad, 1984) through surface sorption and charge and steric stabilization, and have also been shown to be toxic themselves (Bernacchi et al., 1996; Qi et al., 2008).

120


4.3.2.2

Fibrillar Polysaccharides

In freshwaters, a number of organic compounds, in most cases polysaccharides and proteins, but also nucleic acids, peptidoglycan, lipids, lignins, and so on are produced in the water column by exudation (extracellular polymeric substances, EPS) or degradation of phytoplankton, aquatic bacteria and macrophytes. Multiple environmental factors, operating on a large scale, can impose stress (e.g. nutrient deprivation, high toxicant levels) upon some algal and bacterial species in the environment. The stress can lead to biological response such as secretion of organic macromolecules (biopolymers) to alleviate the stress (Leppard, 1995). EPS represent the most abundant organic compounds in the biosphere and constitute the largest fraction of cells. They are important in processes such as mineral dissolution (Welch et al., 1999), biomineralization (Chan et al., 2004), sediment stabilization (Dade et al., 1990), bacterial adhesion (Marshall et al., 1989), biofilm formation (Vandevivere and Kirchman, 1993) and pollutant distribution (Wolfaardt et al., 1994). This section considers mainly fibrillar polysaccharides. Other biopolymers (e.g. protein and peptidoglycan) are not considered as they represent a minor fraction of EPS or have a short turnover time (protein degrades within hours to days and peptidoglycan degrades with days to weeks of their release into the water column) (Nagatal et al., 2003; Smith et al., 1992). Fibrillar polysaccharides can be released from phytoplankton cells during all stages of growth (Strycek et al., 1992). Large amounts of polysaccharides are released during phytoplankton blooms and may comprise 80–90% of the total extracellular release (Myklestad, 1995). They may represent a significant proportion of NOM in freshwater, varying seasonally from about 5 to 30% in surface waters of lakes (Wilkinson et al., 1997a) and likely account for higher proportions (up to 80%) of NOM in marine systems (Aluwihare et al., 1997; McCarthy et al., 1998; Santschi et al., 1998; Verdugo et al., 2004). Polysaccharides are refractory enough to be found in the deep ocean and have a turnover time of hundreds of years (Guo and Santschi, 1997). Polysaccharides are generally rigid due to the large quantity of strongly bound hydration water (up to 80%), their association into double or triple helices that may be stabilized by hydrogen or calcium bridges or helices aggregation (Morris et al., 1980; Norton et al., 1984; Rees, 1981). Transmission electron microscopy (TEM) and atomic force microscopy (AFM) analysis of freshwater and marine polysaccharides suggest that they are a few nanometres in thickness with a length greater than 1 µm and variable conformation as a function of pH and ionic strength (Leppard et al., 1990; Perret et al., 1991; Santschi et al., 1998). Fibrils are important in a variety of environmental functions, such as floc formation via bridging mechanisms which enhance particle sedimentation (Buffle et al., 1998), formation of the matrix component of biofilms and facilitation of microbial adhesion to surfaces (Leppard, 1997), and binding of metal contaminants (Lamelas et al., 2005, 2006; Plette et al., 1996). Due to the complexity, which is both speciesspecific and a function of environmental conditions (Myklestad, 1995), and difficulties in isolating environmental polysaccharides, the study of their colloidal properties is usually performed on model polysaccharides purchased from chemical companies, or ideally produced in bacterial and algal culture within research laboratories

Natural Colloids and Nanoparticles in Aquatic and Terrestrial Environments

121

(Alasonati et al., 2007; Wilkinson et al., 1999). Far less work on these polysaccharides has been performed when compared to HS, primarily for this reason.

4.4

Intrinsic Properties of Environmental Colloidal Particles

This section focuses on the colloidal properties that are relevant to their environmental behaviour and characterization principles, especially as they might enable us to understand the fate and behaviour of manufactured nanoparticles. 4.4.1

Size

Size is the primary means of defining colloids in natural systems (Section 4.2) and is a useful parameter, as other physical and chemical parameters relevant to colloidal behaviour, for example diffusion coefficient, are influenced by, or correlate with, size. Therefore, potentially their behaviour and role in the biogeochemical cycling of trace pollutants and other processes may be understood in terms of size. Colloids, that is particles smaller than 1 µm, tend to stay in suspension, while larger particles tend to sediment (Buffle and Leppard, 1995) and so can be thought of as fundamentally different from particles, even if they are composed identically in chemical terms. As noted, at sizes 7

Al—O–

(c) Dissolution of ionic solids Agl I– Ag+ Ag+ + – I– I + Ag I– (d) Isomorphous substitution

(b) Ion adsorption

Clay

– e.g. SDS, – CH3(CH2)10CH2OSO3 Na+

Al3+

Si4+

Figure 4.4 The methods of charging a solid surface immersed in electrolyte. (T. Cosgrove, Charge in colloidal systems, in Colloid science: principles, methods and applications, 2005. Reproduced with permission from Blackwell Publishing.)

size and crystal structure, and so can be positively charged in many environmental compartments. In the presence of natural organic matter (i.e. humic substances), colloids generally become negatively charged and the point of zero charge shifts to lower values (Amal et al., 1992; Baalousha et al., 2008; Ramos-Tejada et al., 2003). In natural systems, for example freshwater, estuarine, marine and groundwaters, colloids have been observed to have a narrow range of electrophoretic mobilities consistent with the formation of NOM surface coating on all other types of colloids (Beckett and Le, 1990; Hunter and Liss, 1982). Thus, adsorbed NOM molecules dominate colloids surface charge and will have important consequences on their environmental functions and their fate and behaviour. In a few cases, colloids rich in iron oxides (Kaplan et al., 1995; Loder and Liss, 1985; Newton and Liss, 1987) were reported to have a positive surface charge. In aqueous media, the colloidal system as a whole is electrically neutral; oppositely charged ions surround charged particles which balance their surface charge. The distribution of ions in the vicinity of charged particle surfaces may be described by the electric double layer theory (e.g. Stern–Grahame–Gouy–Chapman), which describes the development of the potential with increasing distance from the surface (Figure 4.5). In this model ions are distributed across two layers, a compact inner layer (Stern layer), where the counterions are immobile and a diffuse outer layer, which extends over a certain distance from the particle surface and decays exponentially with increasing distance into the bulk liquid phase. The distribution of ions in the diffuse layer depends on the concentration of the electrolyte, the charge of the ions and the potential at the boundary between the compact inner layer and the diffuse outer layer. The potential at this interface is called the Stern potential. The potential at the shear plane, that is the transition plane from fixed ions and water molecules to those which can be sheared of by fluid motion, is called the zeta potential (ζ), which can be measured by electrokinetic methods (e.g. elec-

Natural Colloids and Nanoparticles in Aquatic and Terrestrial Environments Stern layer

diffuse layer

123

bulk solution

_ _ _ _ _ _

+

+

+

_

+

+ +

+

_

+

_

+ Ψ Ψs ζ Ψs/e xs

1/κ

x

Figure 4.5 Schematic diagram of the diffuse double layer (DDL) forming from the surface of a colloidal particle into the bulk solution. Abbreviations: zeta potential (ζ), electrostatic potential (Ψ), electrostatic potential at the stern layer (ΨS), Euler’s number (e), Boltzmann constant (k). X is a distance from the surface, Xs is the shear plane, the distance from where ions and molecules are mobile and can be sheared off. (With kind permission from Springer Science+Business Media: Ecotoxicology, 17, 2008, 287–314, The ecotoxicology and chemistry of manufactured nanoparticles, R. D. Handy, F. von der Kammer, J. R. Lead, M. Hassellöv, R. Owen and M. Crane, Figure 2.)

trophoresis). Under conditions of very low ionic strength, the decay of the potential between the Stern layer and the shear plane is negligible and the zeta potential can be seen as an approximate of the Stern potential. For more details about the different models describing the double layer, the reader is referred to the literature (Elimelech et al., 1995a). 4.4.3

Surface Coating by Natural Organic Matter

Natural organic matter (NOM) molecules form surface films of several nanometres on macroscopic surfaces (Lead et al., 2005), manufactured nanoparticles (Baalousha et al., 2008) and natural particles (Baalousha and Lead, 2007; Hunter and Liss, 1982; Loder and Liss, 1985; Wilkinson et al., 1997b), hence the similarity in surface charge of colloidal particles in aquatic environment. Almost all environmental particles, regardless of chemical composition, are negatively charged due to the dissociation

124


of surface functional groups on sorbed NOM (Hunter and Liss, 1982; Loder and Liss, 1985). This adsorbed surface layer is likely to dominate the surface properties of colloids such as charge. Therefore, a useful approximation in terms of surface charge and aggregation may be to treat colloids as a single class of colloidal materials, irrespective of their nature, (Filella and Buffle, 1993; O’Melia, 1980). However, these surface coatings may be patchy (Gibson et al., 2007), depending on the nature of the underlying substrate, the NOM type and the solution conditions, meaning that this assumption must be tested in most circumstances. The presence of NOM surface coating on environmental colloids was first shown using surface charge measurements by electrophoresis. The use of TEM, AFM and field flow fractionation have given further insight into the thickness and nature of such a surface coating. The formation of surface coating on a mica surface from IHSS Suwannee River FA is shown in Figure 4.6 (Gibson et al., 2007). The thickness of film found was of about 0.4–5 nm. It has been shown that humic substances sorbs to iron oxide colloids (Baalousha et al., 2008), resulting in the formation of nanoscale surface coating. The thickness of this surface coating was found to be of the order of 0.8 nm on iron oxide particles in the presence of 25 mg l−1 humic acid, although aggregation was also increased at these concentrations due to bridging and charge neutralisation. Surface coating of colloids by NOM is likely to affect aggregation behaviour resulting in reduced aggregation through charge stabilization (Jekel, 1986) and steric stabilization mechanisms (Tipping and Higgins, 1982) or enhanced aggregation through charge neutralization and bridging mechanisms caused by fibrillar attachment (Buffle et al., 1998).

4.4.4

Fractal Dimension

Aggregation of natural colloids results in the formation of fractal aggregate structures. A fractal object has a self-similar structure at all levels of magnification, that is it can be sub-divided into parts, each of which is a reduced-size copy to the whole structure. There are three types of fractal structures: exact self-similar, quasi selfsimilar and statistical self-similar. The latter is the weakest type of self similarity, in which the fractal has a statistical numerical measure which is preserved across different scales. Natural colloidal aggregates generally fall under this type. The fractal dimension, D, can be defined as a statistical quantity that gives an indication of how a fractal structure appears to fill space. The fractal dimension can be described by a geometric power law scaling each dimensional geometry (volume (v) or mass (m) for three dimensions D3, projected area (A) for two dimensions D2, or perimeter (P) for one dimension D1) and characteristic length scales (L) of the aggregate (Lee and Kramer, 2004). D1 provides information about the morphology of the aggregate related to the irregularity of the aggregate boundary or perimeter. D2 provides information about the projected area of an aggregate and D3 provides information about the mass distribution within the aggregate. m or v ∝ LD3

A ∝ LD2

P ∝ LD1

(4.1)


125

Figure 4.6 A tapping mode image of a humic layer that has a 1 × 1 µm2 area machined away in contact mode. Lines a–c that cut across the image are where the cross-sections below the image were taken. (Reprinted with permission from C.T. Gibson, I.J. Turner, C.J. Roberts, J.R. Lead, Quantifying the dimensions of nanoscale organic surface layers in natural waters, Environmental Science & Technology, 41, 1339–44. Copyright 2007, American Chemical Society.) (See colour plate section for a colour representation)

126


A summary of studies that applied the concept of fractal dimension to environmental colloidal particles and the technique used is given in elsewhere (Filella, 2007). Clearly, only a few studies have applied the concept of fractal dimension to non-fractionated or colloidal environmental samples such as fluvial particulate matter (Lartiges et al., 2001), marine snow, diatom blooms, estuarine and marine suspended particles, or biological aggregates in wastewater treatment plants. The majority of studies have used synthetic particles such as iron oxide, goethite (Hackley and Anderson, 1989), hematite (Amal et al., 1992; Zhang and Buffle, 1996), montmorilonite or fractionated organic compounds (Chakraborti et al., 2003; Österberg and Mortensen, 1992; Rice et al., 1999; Rice and Lin, 1993; Senesi et al., 1996, 1997). Although scattered, the fractal dimension values reflect the aggregation mechanisms (Section 4.5.3), values of D3 of 1.6–1.9 indicate a diffusion limited aggregation while values about 2.1–2.3 indicate a reaction limited aggregation. Aggregate fractal dimension is an important factor controlling their fate and behaviour and their interaction with other environmental components. Fractal aggregates have higher permeability than that of a hard sphere and their permeability increases with the decrease in fractal dimension value. The settling velocity of fractal aggregates is higher than that calculated by Stokes’ law for impermeable spheres of identical size and mass and settling rate is lower for aggregates with lower fractal dimension (Johnson et al., 1996a; Li and Logan, 2001). Particle capture efficiency during sedimentation increases with the decrease in fractal dimension (Li and Logan, 1997). Adsorption/desorption hysteresis of contaminants to fractal aggregates can be explained by the blockage of the pores within the aggregates after sorption takes place, that is variation in their fractal dimension (Cheng et al., 2004). Aggregate structure also influences disaggregation rate (more details are given in Section 4.6.2).

4.5

Interaction Forces Between Colloidal Particles

The interaction forces acting between colloidal particles play an important role in determining their fate and behaviour such as stability, aggregation and sedimentation (Liang et al., 2007). As surfaces or colloidal particles approach each other to a distance smaller than a few hundred nanometres, surface forces take place and control colloidal stability and aggregation phenomena. An enormous effort has been devoted to study these forces in the last 60 years, that is since DLVO theory was elaborated. DLVO theory describes surface or particle interactions in term of two independent surface forces, the van der Waals and the electrostatic double layer forces. Since that time, research on surface forces has progressed continuously, especially in the last few decades with the invention of surface force measurement techniques such as surface force apparatus (Israelachvili and Adams, 1976) and atomic force microscopy (Binnig et al., 1986). These techniques contributed not only to confirmation of the DLVO theory, but revealed the presence of other forces called non-DLVO forces, such as the hydration, hydrophobic, steric and bridging forces. Reviewed briefly below are the current understanding of the interaction


127

forces between colloidal particles and the direct experimental measurements of the force as a function of surface separation carried out for particles immersed in a liquid phase. 4.5.1

DLVO Theory

The abbreviation DLVO refers to the names of Derjaguin, Landau, Verwey and Overbeek. They conducted the first successful attempts to describe colloidal stability interactions in Russia (Derjaguin and Landau, 1941) and Netherlands (Verwey and Overbeek, 1948). The DLVO theory is based on the assumption that forces between surfaces or colloidal particles can be regarded as the sum of two forces. These are the short range, attractive van der Waals and the long range, repulsive electrical double layer forces. The interplay between these two forces has many important consequences on colloid stability and aggregation: VT = VA + VR

(4.2)

where VT is the total interaction energy, VA is the attractive van der Waals energy and VR is the repulsive double layer energy. 4.5.1.1 Van der Waals Forces Van der Waals forces are always short range attractive forces and arise from spontaneous electrical and magnetic polarizations, giving a fluctuating electromagnetic field within the media and in the gap between surfaces or particles (Elimelech et al., 1995a). There are two approaches to calculate the van der Waals forces: microscopic and macroscpic. In the microscopic approach, the interaction force is the pairwise summation of all relevant interatomic interactions (Hamaker, 1937) and can be described in terms of geometrical parameters and a constant A, the “Hamaker constant”. For two spheres of equal radius, R, at a surface to surface separation distance, h, apart along the centre to centre axis, the total interaction energy can be given as (Liang et al., 2007): VA( h) = −

A  2R2 2R2 4R2    + + ln  1 −  2 2   6  h + 4 Rh ( h + 2 R) ( h + 2 R)2  

(4.3)

In the case of the interaction between a sphere and a plane at a distance, h, the interaction energy can given as: VA( h) = −

( )

A R R h  + + ln 6  h h + 2 R h + R 

(4.4)

The Hamaker constant depends on material properties such as density and polarizability. The effective Hamaker constant depends also on the dispersion medium. It is generally of the order of magnitude 10−20–10−21 J (Elimelech et al., 1995a). Aeff ≈ ( Aparticle − Amedium )

2

(4.5)

128


The macroscopic approach overcomes the summation assumption by considering the macroscopic electromagnetic properties of the medium (Lifshitz, 1956) where atomic structure is neglected and large bodies are treated as continuous media and forces are derived in terms of the bulk properties such as dielectric constants and refractive indices. However, the use of the macroscopic approach is limited by the computation required and the lack of appropriate dielectric data. Additional details about these two approaches and calculations of Hamaker constant can be found elsewhere (Bergstrom, 1997; Elimelech et al., 1995a; Israelachvili, 1972). 4.5.1.2

Double Layer Interaction

Colloidal particles often carry an electrical charge and therefore attract or repel each other. When two like-charged particles approach each other, their electrical double layer starts to overlap, resulting in a repulsive force which opposes further approach. For identical particles, sphere–sphere double layer interaction energy can be given by Equation 4.6. There are many expressions available based on various assumptions for sphere–sphere double layer interaction energy and readers are referred to the literature for more details (Bell et al., 1970; Carnie et al., 1994; Genxiang et al., 2001; McCormack et al., 1995; Sader et al., 1995; Stankovich and Carnie, 1996). VR( h) = 32πε R

( )γ kT ze

2

2

exp ( −κ h)

(4.6)

For small values of surface or zeta potential (ζ), this simplifies to: VR( h) = 2π eRζ 2 exp ( −κ h)

(4.7)

where ε is the permittivity of the medium, R is the particle radius, γ is dimensionless functions of the surface potentials, k is the Boltzman constant, T is the absolute temperature (Kelvin), h is the surface-surface separation between particles (m), e is the electron charge and κ is the inverse of Debye–Huckel screening length (m−1). Equation 4.7 is applicable only if κR > 5 and h 2 nm. However, below 2 nm, a repulsive force is probed and was attributed to hydration force (Meagher, 1992). Further, hydration forces were observed between an alumina (Al2O3) tip and a mica surface at different pH values and between a silicon nitride tip and a mica surface at high concentrations of divalent cations (>3 M) (Butt, 1991), between silica surfaces in 1,2-ethanediol and water (Atkins and Ninham, 1997), between gold surfaces in sodium chloride (Biggs et al., 1994), between silica surfaces and silicon on titanium dioxide at high pH (Larson et al., 1993) and between an alumina surface and an aluminum or silicon nitride tip (Karaman et al., 1997). The hydration force was also measured between two mica surfaces in electrolyte solution (Israelachvili and Pashley, 1983; Pashley, 1981; Pashley and Israelachvili, 1984a). They measured a short range repulsive force in addition to van der Waals forces at high salt concentrations, which varied with the type of cation in solution. The more hydrated cations, such as Mg2+ and Ca2+, gave stronger repulsive forces than the less hydrated monovalent ions, such as K+ and Cs+. Other studies have suggested that hydration forces are oscillatory and can be either attractive or repulsive (Israelachvili and Wennerstrom, 1996; Pashley and Israelachvili, 1984b). These hydration forces should be present in aquatic colloidal particles and can be dominant in those with high negative charge densities. Clearly further theoretical and experimental work is needed to explore hydration forces in environmental colloids. 4.5.4.2

Hydrophobic Interactions

A hydrophobic surface is one that has low affinity for water and has no polar or ionic groups or hydrogen bonding sites. The nature of water in contact with such a surface is different from the bulk water. Bulk water is significantly structured via the formation of hydrogen bonds between the water molecules, resulting in the formation of large clusters of hydrogen bonded water molecules. The presence of a hydrophobic surface will most likely restrict such phenomena and water confined between two hydrophobic surfaces will not be able to form clusters larger than a certain size, causing water molecules to tend to migrate to the bulk water where there is unrestricted hydrogen bonding opportunities and a lower free energy (Elimelech et al., 1995a). Hydrophobic forces can be important, giving an extra attraction between surfaces or particles. Attraction between hydrophobized mica sheets, via surface adsorption of hydrocarbon and fluorocarbon surfactants, has been directly measured (Israelachvili and Pashley, 1984) and was found to operate over a long range of about 80 nm and to be much stronger than the van der Waals force (Claesson and Christenson, 1988). The magnitude of hydrophobic forces was found to decrease with the increase in electrolyte concentration (0.01–0.1 M magnesium sulfate) (Christenson et al., 1990). Figure 4.9 shows the long range (about 80 nm) attractive hydrophobic forces measured between a silicon nitride (Si3N4) tip and a hydrophobic mica surface prepared by depositing a monolayer of cetyltrimethylammonium bromide (CTAB) on the surface of freshly cleaved mica surfaces (Teschke and de Souza, 2003). More details can be found elesewhere (Christenson and Claesson,


133

0.0 Dieletric Constant (E)

Force (nN)

0.5

–0.5

–1.0 0

20

80 60 40 20 0

0

10

20 30 40 Separation (nm)

40 60 Separation (nm)

80

50

100

Figure 4.9 Force vs separation curve (•) between a standard silicon nitride (Si3N4) tip and a bare mica surface (䊊) between a neutral tip and hydrophobic mica surface prepared depositing a monolayer of CTAB on the surface of freshly cleaved mica surfaces. (Reprinted from Chemical Physics Letters, 375, 540–6, O. Teschke and E.F. de Souza, Measurements of longrange attractive forces between hydrophobic surfaces and atomic force microscopy tips. Copyright 2003, with permission from Elsevier.)

2001; Eriksson et al., 1989; Ruckenstein and Churaev, 1991). It is possible for the surface of particles dispersed in environmental waters to have some degree of hydrophobicity and the hydrophobic force has been postulated to contribute to the aqueous aggregation of clay particles (Zbik and Horn, 2003). 4.5.4.3

Steric Interactions

Adsorption of natural organic matter to colloidal particles is a well known process in aquatic systems as discussed previously in Section 4.4.3 (Gibson et al., 2007; Tipping and Higgins, 1982; Wilkinson et al., 1997a). Adsorbed NOM molecules can play a very important role in the aggregation and deposition phenomena. In some cases, adsorbed NOM molecules (e.g. polysaccharides) can induce aggregation by bridging mechanisms (Section 4.5.4.4), while in other cases NOM molecules such as fulvic and humic acids can enhance colloidal stability (Wilkinson et al., 1997a) by a mechanism known as steric stabilization. The adsorbed molecule chains extend some distance into the water, giving increased stability to colloidal particles. As particles approach each other, the adsorbed layer of NOM comes into contact, resulting in the interaction between these molecules. As these molecules are hydrated, any interaction will induce hydration repulsive forces as described in the previous section. The steric stabilization effect increases with the load with NOM or the thickness of the adsorbed layer.

134


The stability of colloidal particles in an aquatic environment is often higher than expected on the basis of zeta potential and ionic strength, which is likely related to a steric stabilization effect induced by NOM surface coating (Jekel, 1986) and possibly to the hydration effect explained in Section 4.5.4.1. Direct measurements of forces between colloidal particles are scant, though some have been performed by AFM (Assemi et al., 2004; Mosley et al., 2003; Sander et al., 2004). Assemi et al. (2004) investigated the interaction forces between a goethite coated mica surface (positively charged) and silica colloidal probe (negatively charged). The adsorption of humic substances onto a goethite coated mica surface (imparting negative charge) induces repulsion between the goethite surface and silica colloidal probe. In addition, a high repulsion force was observed at short separation distances (typically < 2–3 nm) and was attributed to steric forces induced by the sorption of humic substances. Mosley et al. (2003) investigated the effect of adsorbed NOM, solution pH and ionic composition on the force–distance curve between natural colloids represented by surface film of iron oxides precipitated onto spherical SiO2 particles. At low ionic strength, the interparticle forces were dominated by electrostatic repulsion from the dissociation of functional groups on the NOM. At small separation distances (100 nm, below 100 nm there is no effect of a change in density from 1 to 3 g/cm3. A reduction of the fluid flow velocity from 10 to 1 m/d (representing typical values for aquifers) reduces the maximum travel distance from 200 to 30 m. Maximum colloid mobility can be expected at a size range close to 1 µm for colloids with a density of 1 g/cm3 and around 0.2 µm for colloids with a density of 3 g/cm3. It is important to notice that the CFT holds for a clean bed filtration approach, where the collision efficiency (i.e. the rate of successful attachment) is close to one (Ryan and Elimelech, 1996). Ripening or blocking, as well as surface heterogeneities of the collector (e.g. roughness or charge heterogeneity), cannot be described with these equations (Elimelech et al., 2000; Johnson et al., 1996b). Neither can filtration under non-favourable, repulsive conditions, marked by an energy barrier (Section 4.5).

Filtration Factor l [1/m]

100

10

1 diffusion interception

sedimentation

0,1 sieving

0,01 0,1

1

10

Colloid Size [µm]

Figure 4.18 Colloid filtration by diffusion, interception, sedimentation and sieving.

146

Environmental and Human Health Impacts of Nanotechnology 1000

1: dcol. = 10 mm, vf = 10 m/d, density = 1 g/cm?

travel distance 99,9 % [m]

2: dcol. = 10 mm, vf = 1 m/d, density = 1 g/cm? 100

3: dcol. = 10 mm, vf = 1 m/d, density = 3 g/cm? 4: dcol. = 1 mm, vf = 1 m/d, density = 3 g/cm?

10

1 2

1

3 4 0,1 0,001

0,01

0,1

1

10

colloid size [mm]

Figure 4.19 Effect of colloid density, flow velocity and collector grain diameter on filtration/ deposition efficiencies calculated by Equations 4.14 to 4.21. dcoll = grain size diameter of the collector, vf = apparent flow velocity of the fluid within the porous media.

Once the colloid is attached to the surface of the collector it can be remobilized by hydrodynamic drag or lift forces. These two forces are balanced by an adhesive force, which can be quantified in terms of free energy of adhesion using an extended DLVO type approach (Section 4.5). The fluid drag and lift forces on a retained colloid can be calculated as summarized by Ryan and Elimelech (1996). For small colloids the lift force, that is the force due to different pressure acting on the top and bottom of the particle, can be neglected. The adhesive force can be calculated using different scaling models, such as the Johnson–Kendall– Roberts model or the Derjaguin–Mullen–Toporov model (Derjaguin et al., 1975; Johnson et al., 1971; Kendall, 2001). Drag forces are typically significant only for particles larger than a few hundred nanometers and when deposition occurs in the primary minimum (according to DLVO theory, Section 4.5). However, for deposition of nanoparticles 1000 tonnes/year?

Y 2-generation reproductive study

Positive result or > 100 tonnes/year?

Long-term toxicity (sediment organisms) Long-term/reproductive toxicity (birds)

N

Y In vivo mutagencity studies

Positive result or > 100 tonnes/year?

Relevant in vivo study

Y

Corrosive, strong acid/ base, flammable, very toxic, or skin irritant; and >10 tonne/year?

Y

> 1 tonne/year?

Substance

Ecotoxicological information

> 100 tonnes/year?

N

N

> 10 tonnes/year?

Aquatic toxicity Short-term toxicity testing (Daphnia) Long-term toxicity testing (Daphnia) Long-term toxicity testing (fish)

STOP

Y

> 1000 tonnes/year?

N

Bioconcentration (fish) Further studies on adsorption/desorption Y Short-term toxicity to earthworms/plants Effects on soil microorganisms Long-term toxicity on earthworms/plants/soil invertebrates

Y Identification of degradation products Further degradation testing

N

Hydrolysis as a function of pH

Ready biodegradability Degradation in surface water Soil simulation testing Sediment simulation testing

Growth inhibition study (algae) Short-term toxicity testing (fish) Activated sludge respiration inhibition testing

< 10 tonnes/years and insoluble or does not cross biological membranes?

Chemical Safety Report

N

> 100 tonnes/year?

Environmental fate/behaviour Adsorption/desorption

Y

N

Methods of detection and analysis

Other available physiochemical, toxicological and ecotoxicological information

Overview of risk assessment by REACH showing information triggers (taken from Rocks et al., 2008).

Acute toxicity Oral/dermal/inhalation route Short-term repeated dose toxicity Sub-chronic toxicity study Reproductive toxicity screening Developmental toxicity study Toxicokinetics

Y

> 10 tonnes/year?

Mutagenicity In vitro gene mutation In vitro cytogenicity In vitro gene mutation

Eye irritation Assessment of: human/animal data acid or alkaline reaction

Skin irritation/corrosion Assessment of: human/animal data acid or alkaline reaction

Toxicological information

N

Is substance a non-isolated intermediate, under customs supervision, radioactive, or a polymer?

Substance state Melting/freezing point Boiling point Relative density Vapour Pressure Surface tension Water solubility Flash-point Flammability Explosive properties Self-ignition temperature Oxidising properties Granulometry Partition coefficient n-octanol/water

Physiochemical properties

STOP

412 Environmental and Human Health Impacts of Nanotechnology

Risk Assessment of Manufactured Nanomaterials

413

Physico-Chemical Properties There are many published sources of physico-chemical data, including the Merck Index and IUPAC Solubility Data Series, that can be considered and used within risk assessments rather than experimental results. However, the data should be considered carefully and the state of the substance and range of values must be evaluated. Included in the physico-chemical properties of a chemical substance is the determination of the particle size distribution, which would account for nanomaterials and particles (Figure 10.A.2). Toxicological Information The toxicological information required under REACH can be split further into eight information groups. While these have been considered separately due to the ethics of animal work, the experiments must be carefully designed with crossovers between groups. The groups are: 1. Skin Irritation/Corrosion. These tests are not necessary if data show the substance is corrosive/irritating. In vitro tests are required for CSA, whilst in vivo tests are required for unclassified substances manufactured/imported in amounts greater than 10 tonnes/year. 2. Eye Irritation. In vitro tests are not necessary if the substance is considered irritant/corrosive to skin (and is classified as irritating to the eye). For substances produced or imported in amounts greater than 10 tonnes/year an in vivo test is required, unless substance has been determined as irritating. 3. Skin Sensitisation. In vitro tests are required, and if there is not enough information to classify the substance as a skin sensitiser in vivo tests necessary. 4. Mutagencity. In vitro and in vivo tests in somatic and germ cells are used to determine whether the substance is genotoxic in somatic and/or germ cells. 5. Acute Toxicity. Physico-chemical data previously collected are used to determine the route of administration of the substance (indicative of the common route of human exposure). Oral in vivo acute toxicity tests are required unless oral exposure is not possible, in which case an inhalation study is necessary. Further in vivo studies for another (different) exposure route are required for substances imported or manufactured in amounts greater than 10 tonnes/year. 6. Reproductive and Developmental Toxicity. If the substance has already been classed as a genotoxic carcinogen or a germ cell mutagen and the appropriate risk management measures are in place then further testing is not necessary. A two-generation reproductive toxicity study is required for substances manufactured or imported in amounts ≥100 tonnes/year, and for those ≥10 tonnes/year if the reproductive and developmental toxicity screening study is positive or if repeated dose toxicity study indicates potential reproductive toxicity. 7. Repeat Dose Toxicity. These tests are only required if there were indications in the acute toxicity testing or if the chemical is over the 10 tonnes/year weight trigger. The tests determine the NOAEL for the substance over 28 days. If the amount of substance is greater than 100 tonnes/year, or data suggest accumulation, then a sub-chronic or chronic repeated dose study (90 days or 12 months) is required.

Methods: Cascade, Laser, Rotating drum and Continuous drop method

Y

N

Water insoluble

Stop testing

Y

Can mass median aerodynamic diameter demonstrate no inhalation risk?

Y

Is inhalation risk study required?

Are particles soluble?

Y

N

N

Inhalation route for toxicity testing

Stop testing

Investigate particles using microscopy, sedimentation, and laser doppler

Water soluble

Determine relative density and water solubility

Particles

Environmental and Human Health Impacts of Nanotechnology

Nanoscience and Nanotechnology: Environmental and Health Impacts

Nanoscience and nanotechnology: environmental and health impacts

Nanoscience and Nanotechnology: Environmental and Health Impacts

Environmental Nanotechnology: Applications and Impacts of Nanomaterials

Desalination Technology: Health and Environmental Impacts

Air Pollution: Health and Environmental Impacts

Environmental Impacts on Reproductive Health and Fertility

Air pollution: Health and environmental impacts

Nanotechnology: Health and Environmental Risks (Perspectives in Nanotechnology)

Foundations of Environmental Physics: Understanding Energy Use and Human Impacts

Environmental Management, Sustainable Development and Human Health

Environmental Impacts of Macroeconomic and Sectoral Policies

Nanotechnology: Environmental Implications and Solutions

Environmental Impacts of Macroeconomic and Sectoral Policies

Environmental Health Impacts of Transport and Mobility (Environmental Science and Technology Library)

Nanotechnology: Consequences for Human Health and the Environment

Polar Regions: Human Impacts

Nanotechnology: Consequences for Human Health & the Environment (Issues in Environmental Science and Technology)

Quantitative Environmental Risk Analysis for Human Health

Environmental Health and Housing

Mine Wastes: Characterization, Treatment and Environmental Impacts

Environmental Impacts of Wind-Energy Projects

Ecosystems and human health: toxicology and environmental hazards

Epigenetics and Human Health: Linking Hereditary, Environmental and Nutritional Aspects

Ecotourism series. No2. Environmental Impacts of Ecotourism

Environmental impacts of pasture-based farming

Floods: Physical Processes and Human Impacts

Environmental Toxicants: Human Exposures and Their Health Effects, 3rd Edition

Human Impacts on Weather and Climate

Environmental Health: From Global to Local (Public Health Environmental Health)

Environmental and Human Health Impacts of Nanotechnology

Nanoscience and Nanotechnology: Environmental and Health Impacts

Nanoscience and nanotechnology: environmental and health impacts

Nanoscience and Nanotechnology: Environmental and Health Impacts

Environmental Nanotechnology: Applications and Impacts of Nanomaterials

Desalination Technology: Health and Environmental Impacts

Air Pollution: Health and Environmental Impacts

Environmental Impacts on Reproductive Health and Fertility

Air pollution: Health and environmental impacts

Nanotechnology: Health and Environmental Risks (Perspectives in Nanotechnology)

Foundations of Environmental Physics: Understanding Energy Use and Human Impacts

Environmental Management, Sustainable Development and Human Health

Environmental Impacts of Macroeconomic and Sectoral Policies

Nanotechnology: Environmental Implications and Solutions

Environmental Impacts of Macroeconomic and Sectoral Policies

Environmental Health Impacts of Transport and Mobility (Environmental Science and Technology Library)

Nanotechnology: Consequences for Human Health and the Environment

Polar Regions: Human Impacts

Nanotechnology: Consequences for Human Health & the Environment (Issues in Environmental Science and Technology)

Quantitative Environmental Risk Analysis for Human Health

Environmental Health and Housing

Mine Wastes: Characterization, Treatment and Environmental Impacts

Environmental Impacts of Wind-Energy Projects

Ecosystems and human health: toxicology and environmental hazards

Epigenetics and Human Health: Linking Hereditary, Environmental and Nutritional Aspects

Ecotourism series. No2. Environmental Impacts of Ecotourism

Environmental impacts of pasture-based farming

Floods: Physical Processes and Human Impacts

Environmental Toxicants: Human Exposures and Their Health Effects, 3rd Edition

Human Impacts on Weather and Climate

Environmental Health: From Global to Local (Public Health Environmental Health)

Recommend Documents