Volatile Organic Compounds in the Atmosphere

Volatile Organic Compounds in the Atmosphere Volatile Organic Compounds in the Atmosphere Edited by Ralf Koppmann Uni...

Author: Ralf Koppmann

143 downloads 1325 Views 5MB 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

Volatile Organic Compounds in the Atmosphere

Volatile Organic Compounds in the Atmosphere Edited by Ralf Koppmann University of Wuppertal, Germany

Blackwell Publishing

© 2007 by Blackwell Publishing Ltd Blackwell Publishing editorial offices: Blackwell Publishing Ltd, 9600 Garsington Road, Oxford OX4 2DQ, UK Tel: +44 (0)1865 776868 Blackwell Publishing Professional, 2121 State Avenue, Ames, Iowa 50014-8300, USA Tel: +1 515 292 0140 Blackwell Publishing Asia Pty Ltd, 550 Swanston Street, Carlton, Victoria 3053, Australia Tel: +61 (0)3 8359 1011 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. First published 2007 by Blackwell Publishing Ltd ISBN: 978-1-4051-3115-5 Library of Congress Cataloging-in-Publication Data Volatile organic compounds in the atmosphere/edited by Ralf Koppmann. – 1st ed. p. cm. Includes bibliographical references and index. ISBN-13: 978-1-4051-3115-5 (hardback : alk. paper) 1. Organic compounds–Environmental aspects. 2. Air quality management. I. Koppmann, Ralf. TD885.5.O74V66 2007 511.51 12–dc22 2006034260 A catalogue record for this title is available from the British Library. Set in 10/12 pt Minion by Newgen Imaging Systems (P) Ltd., Chennai, India Printed and bound in Singapore by Markono Print Media Pte Ltd The publisher’s policy is to use permanent paper from mills that operate a sustainable forestry policy, and which has been manufactured from pulp processed using acid-free and elementary chlorine-free practices. Furthermore, the publisher ensures that the text paper and cover board used have met acceptable environmental accreditation standards. For further information on Blackwell Publishing, visit our website: www.blackwellpublishing.com


Contents

Preface

ix

List of Contributors

xi

1

1

2

3

Volatile Organic Compounds in the Atmosphere: An Overview Jonathan Williams and Ralf Koppmann 1.1 Introduction 1.2 Sources 1.3 Sinks 1.4 Atmospheric distribution 1.5 Measurement tools 1.6 Modelling tools 1.7 How organic species affect the atmosphere 1.8 Open questions and future directions References

1 3 5 7 9 10 12 15 19

Anthropogenic VOCs Stefan Reimann and Alastair C. Lewis 2.1 Introduction 2.2 Sources of anthropogenic VOCs 2.3 Atmospheric distribution of VOCs 2.4 Chemical behaviour of VOCs in the atmosphere 2.5 Measurement techniques References

33

Biogenic VOCs Allison H. Steiner and Allen L. Goldstein 3.1 Introduction 3.2 Sources of biogenic VOCs 3.3 Emission inventories of biogenic VOCs 3.4 Global distribution of biogenic VOCs 3.5 Impact on photooxidants and atmospheric chemistry

82

33 33 45 55 60 70

82 83 97 103 107

vi

4

5

6

7

Contents

3.6 Sampling and measurement techniques 3.7 Future directions References

114 116 117

Oxygenated Volatile Organic Compounds Ralf Koppmann and Jürgen Wildt 4.1 Introduction 4.2 Tropospheric mixing ratios and global distribution 4.3 Sources of OVOCs 4.4 Sinks of OVOCs 4.5 Budgets and emission inventories 4.6 Sampling and measurement techniques 4.7 Future directions Acknowledgement References

129

Halogenated Volatile Organic Compounds Simon J. O’Doherty and Lucy J. Carpenter 5.1 Introduction 5.2 Sources of halogenated VOCs 5.3 Atmospheric concentrations: trends and distribution 5.4 Sinks of halogenated VOCs 5.5 Emission inventories 5.6 Sampling techniques 5.7 Measurement techniques References

173

PAN and Related Compounds James M. Roberts 6.1 The chemistry of PANs 6.2 Atmospheric formation 6.3 Measurement and calibration techniques 6.4 Atmospheric measurements 6.5 Modelling and interpretation of ambient measurements 6.6 Conclusions Acknowledgements References

221

Organic Nitrates Paul B. Shepson 7.1 Introduction 7.2 Production mechanism 7.3 Measurement methods 7.4 Atmospheric measurements 7.5 Fate 7.6 Conclusions References

269

129 130 137 149 154 155 160 160 160

173 179 187 192 204 207 210 214

222 229 237 243 249 255 256 256

269 271 274 276 282 285 286

Contents

8

9

vii

High-Molecular-Weight Carbonyls and Carboxylic Acids Paolo Ciccioli and Michela Mannozzi 8.1 Introduction 8.2 Sources 8.3 Atmospheric levels 8.4 Reactivity and impact on the atmosphere 8.5 Sampling and analysis 8.6 Conclusions References

292

Organic Aerosols Thorsten Hoffmann and Jörg Warnke 9.1 Introduction 9.2 Carbonaceous aerosols 9.3 Analysis of organic aerosols Further reading References

342

292 293 309 324 329 333 334

342 345 365 375 375

10 Gas Chromatography-Isotope Ratio Mass Spectrometry Jochen Rudolph 10.1 Introduction 10.2 Fundamentals of stable isotope ratios of VOCs 10.3 Experimental methods 10.4 Kinetic isotope effects 10.5 Stable isotope ratios of atmospheric VOC and their sources 10.6 Conclusions References

388

11 Comprehensive Two-Dimensional Gas Chromatography Jacqueline F. Hamilton and Alastair C. Lewis 11.1 Introduction 11.2 Fundamentals of comprehensive gas chromatography 11.3 Modulators 11.4 Detectors 11.5 Examples of GC × GC use in atmospheric samples 11.6 Conclusions Further reading References

467

Index

489

Color plate appears between pages 268 and 269

388 389 405 420 447 458 460

467 468 471 474 475 482 486 486


Preface

Every day, large quantities of volatile organic compounds (VOCs) are emitted into the atmosphere from both anthropogenic and natural sources. They are the ‘fuel’ that keeps atmospheric photochemistry running. Therefore, their sources, sinks and residence times are the subject of current research. In addition to influencing local, regional and even global photochemistry, several of these compounds have a potential impact on climate, both due to their properties as greenhouse gases and due to their ability to form aerosol particles on oxidation. The formation of gaseous and particulate secondary products caused by the oxidation of VOCs is one of the largest unknowns in the quantitative prediction of the earth’s climate on a regional and global scale, and on the understanding of local air quality. To be able to model and control their impact, it is essential to understand the sources of VOC, their distribution in the atmosphere and the chemical transformations they undergo. Furthermore, organic trace gases can be used as tracer compounds to investigate reactions that are not directly accessible to current measurement techniques or as probes to ‘visualise’ transport processes in the atmosphere or across atmospheric boundaries. In recent years methods and techniques for the analysis of organic compounds in the atmosphere have been developed to increase both the spectrum of detectable compounds as well as the corresponding detection limits. New methods have been introduced to increase the time resolution of those measurements and to resolve more complex mixtures of organic compounds. New technical developments reducing weight and energy requirements made the use of these instruments on various platforms such as balloons or aircraft possible. This book describes the current state of knowledge of the chemistry of VOC as well as the methods and techniques to analyse gaseous and particulate organic compounds in the atmosphere. Chapter 1 is an instructive chapter summarising the variety and the roles of VOC in the atmosphere. Chapters 2 to 9 cover the various compound classes, their distribution in the atmosphere, their chemical transformations and their budgets as well as a survey of currently used measurement techniques. Chapters 10 and 11 describe new methods to measure a large part of the VOC family at a glance and for investigating their stable carbon isotope ratios. In-depth references are provided, enabling each subject to be explored in more detail. The aim is to provide an authoritative review to address the needs of both graduate students and active researchers in the field of atmospheric chemistry research. It may also serve as a desktop resource for experienced scientists in the field of atmospheric research.

x

Preface

Thanks are due to all the chapter authors for their efforts in the completion of this work. I am grateful to many colleagues for numerous discussions, for their patience, their advice and critical reviews of the chapters. Special thanks are due to Sarahjayne Sierra and Dr Paul Sayer of Blackwell Publishing for their patience in answering all my questions and their persistence in the efforts required for completing this book.


Dr Lucy J. Carpenter

Department of Chemistry, University of York, York, UK

Dr Paolo Ciccioli

Istituto di Metodologie Chimiche del C.N.R., Monterotondo Scalo, Italy

Professor Allen L. Goldstein

Department of Environmental Science, Policy, and Management, University of California, Berkeley, CA, USA

Dr Jacqueline F. Hamilton


Professor Thorsten Hoffman

Institut für Anorganische und Analytische Chemie, Johannes Gutenberg-Universität Mainz, Mainz, Germany

Professor Ralf Koppmann

Fachbereich Mathematik und Naturwissenschaften, Atmosphärenphysik, Bergische Universität Wuppertal, Wuppertal, Germany

Professor Alastair C. Lewis


Dr Michela Mannozzi

Istituto Centrale per la Ricerca Scientifica e Tecnologica Applicata al Mare, Roma, Italy

Dr Simon J. O’Doherty

School of Chemistry, University of Bristol, Bristol, UK

Dr Stefan Reimann

Eidgenössische Materialprüfungs- und forschungsanstalt, EMPA, Duebendorf, Switzerland

Dr James M. Roberts

NOAA Earth Systems Research Laboratory, Chemical Science Division, Boulder, CO, USA

Professor Jochen Rudolph

Chemistry Department and Centre for Atmospheric Chemistry, York University, Toronto, Ontario, Canada

Professor Paul B. Shepson

Departments of Chemistry, and Earth and Atmospheric Sciences, Purdue University, West Lafayette, IN, USA

Professor Allison H. Steiner

Department of Atmospheric, Oceanic and Space Science, University of Michigan, Ann Arbor, MI, USA

xii


Dr Jörg Warnke

Institut für Anorganische und Analytische Chemie, Johannes Gutenberg-Universität Mainz, Mainz, Germany

Dr Jürgen Wildt

Institut für Chemie und Dynamik der Geosphäre, Institut 3: Phytosphäre, Forschungszentrum Jülich, Jülich, Germany

Dr Jonathon Williams

Atmospheric Chemistry Department, Max Planck Institute for Chemistry, Mainz, Germany

Volatile Organic Compounds in the Atmosphere Edited by Ralf Koppmann Copyright © 2007 by Blackwell Publishing Ltd

Chapter 1

Volatile Organic Compounds in the Atmosphere: An Overview Jonathan Williams and Ralf Koppmann

1.1

Introduction

The aim of this overview is to highlight the importance of organic trace gases in the atmosphere and to introduce the themes of the chapters to follow. This work is suited to those new to the field and to those seeking to place related activities in a broader context. Tens of thousands of organic compounds have been detected in the air we breathe, and the focus here is on the myriad carbon-containing gases present at mixing ratios of some 10 parts per billion (ppbv, 10−9 or nmol/mol) down to some parts per trillion (ppt, 10−12 or pmol/mol). This excludes the three most abundant, but generally less reactive, organic compounds: carbon dioxide (370 parts per million, ppmv, 10−6 or μmol/mol), methane (1.8 ppmv) and carbon monoxide (0.15 ppmv), which have been discussed in detail elsewhere. Unfortunately various terms have been used in the literature to describe the subset of diverse carbon-containing gases under circa 10 ppbv. One of the first was nonmethane hydrocarbons (NMHC), which was employed originally to distinguish alkanes, such as ethane, propane and butane, from methane. However, strictly speaking the word ‘hydrocarbon’ indicates a molecule containing only carbon and hydrogen atoms, and therefore this term appears to exclude oxygenated species such as alcohols, carbonyls and acids as well as organic compounds containing other heteroatoms, such as nitrogen or sulphur. In an attempt to embrace all the species relevant to atmospheric chemistry, a further term ‘volatile organic compound’ (VOC) was coined, although there is no general quantitative definition of what VOCs are. The Environmental Protection Agency (EPA) in the United States has defined VOC as any compound that participates in atmospheric photochemical reactions; however, there have been subsequent attempts to give a more quantified definition. The result is that VOCs are considered to be those organic compounds having a vapour pressure greater than 10 Pa at 25◦ C, a boiling point of up to 260◦ C at atmospheric pressure, and 15 or less carbon atoms. The remaining compounds are designated as semivolatile organic compounds (SVOCs). This segregation emphasises the volatile gas phase species from those that partition to the aerosol phase which is reasonable since the later undergo different transport and chemistry (see Sections 1.3 and 1.4). Recently, however, the definition has become blurred by the use of OVOCs to specifically identify the oxygenated VOCs. For this chapter we prefer not to draw a divide through the continuum of


Stratospheric chemistry

Upper tropospheric chemistry

NO3 HO O3

Aerosol

N

VOC

Oxidation

siti

on

Advection

Remote region chemistry

Wa sh ou t Wet and dry deposition

olo

Bi

(Br, Cl)

Photochemical products Oxidation

uc

De

po

n

tio lea

C o nv e c ti o n

2

al

gic ke

ta

up

CO2 + H2O

Figure 1.1 Sketch of the various processes which determine the fate of VOC in the atmosphere. The individual processes are discussed in the text and the individual classes of VOC in the following chapters of this book.

compounds and instead use the term ‘trace organic compounds’, referring to the dictionary definitions of organic (designating carbon compounds) and trace (an extremely small amount). Despite being found at extremely low concentrations, trace organic compounds have profound effects in the atmosphere (see Figure 1.1). On the one hand, they are the ‘fuel’ which keeps oxidative atmospheric photochemistry running. Therefore, their sources, sinks and atmospheric residence times are the subject of much current research (see Sections 1.2–1.4). To investigate organic trace gases in the atmosphere it is essential that accurate concentration measurements and careful modelling studies are made (see Section 1.5). In addition to influencing local, regional and even global photochemistry, several such compounds have a potential impact on climate, both due to their properties as greenhouse gases and due to their ability to form aerosol particles on oxidation (see Section 1.6). Organic trace gases can be used as tracer compounds to investigate reactions which are not directly accessible to current measurement techniques or as probes to ‘visualise’ transport processes in the atmosphere or across atmospheric boundaries. Many open questions remain in this field of research, and some of the future challenges in the field are summarised in Section 1.7. The intention here is to provide an up-to-date, referenced overview of the field emphasising the recent progress made in an exciting and rapidly developing area of research. Recent and review-type references have been preferentially cited along with key older articles so that the interested reader may quickly access more detailed information. The authors would

Volatile Organic Compounds in the Atmosphere: An Overview

3

like to point out that the articles cited here represent a tiny fraction of a vast and widespread literature database. We hope that any omissions of particular works by colleagues will be forgiven in the interests of brevity.

1.2

Sources

Almost everything we do in daily life results in the release of organic species to the atmosphere. Driving a car (Fraser et al. 1998), painting the house (Fortmann et al. 1998), cooking (McDonald et al. 2003), making a fire (Andreae and Merlet 2001), cutting the grass (Fall et al. 1999; Kirstine et al. 1998) and even breathing (Barker et al. 2006; Phillips et al. 1999) – all of these processes result in the emission of organic compounds such as carbonyls, alcohols, alkanes, alkenes, esters, aromatics, ethers and amides. In addition to emissions from human activities, the Earth’s vegetation naturally releases huge amounts of organic gases into the air. As plants assimilate carbon dioxide into biomass through photosynthesis, a fraction of this carbon leaks out in to the atmosphere, predominantly in highly reduced forms such as isoprene and terpenes (Fuentes et al. 2000; Guenther 2002; Kesselmeier et al. 2002). Exactly which compounds are emitted from a particular plant, and how much of each, depends on the age and health of the vegetation as well as the ambient temperature, moisture and light levels (Guenther et al. 1995; Kesselmeier and Staudt 1999). Both plants and invertebrates have been shown to use emission of specific organic species into the air for signalling (Greene and Gordon 2003; Krieger and Breer 1999). Examples of elaborate chemical mimicry have been found in have been found in insects (e.g. Cremer et al. 2002) and amongst plants to deter attack by herbivores (Kaori et al. 2002; Kessler and Baldwin 2001). While the natural world uses air as a communication medium, man often uses it as a repository for waste products. The emission rates and their associated uncertainties for VOCs from several source categories are summarised in Table 1.1. The anthropogenic contribution to organic emissions in the atmosphere is dominated by the exploitation of fossil fuels (coal, oil and gas). Approximately 100 TgC/year was estimated to be emitted from ‘technological’ sources and 150 TgC/year from all anthropogenic sources including biomass burning (Müller 1992). Coal production mainly leads to methane emission, but minor emissions of ethane and propane are also present. Liquid fossil fuel production, storage and distribution result in a larger variety of organic gas emissions to the atmosphere. Crude oil production platforms are strong point sources of hydrocarbons such as methane, ethane, propane, butanes, pentanes, hexanes, heptanes, octanes and cycloparaffins (McInnes 1996). The major sources from processing liquid fossil fuels are catalytic cracking (0.25–0.63 kg/m3 of feed), coking (about 0.4 kg/m3 of feed) and asphalt blowing (about 27 kg of VOC/m3 of asphalt) (Friedrich and Obermeier 1999). Furthermore, so-called fugitive emissions can occur from leaks and evaporation from all types of equipment and installations. Evaporative emissions are estimated to be 2.9 kg/t of fuel at service stations (McInnes 1996) and are familiar to anyone who has filled a car with gasoline/petrol. Petrochemical products typically contain a limited number of compound classes (e.g. acyclic alkanes, cyclic alkanes, monoaromatics, diaromatics) each consisting of a very large number (tens of thousands) of individual homologues and isomers (Schoenmakers et al. 2000). Major products of the complete combustion of fossil fuels are carbon dioxide and water. However, in practice, combustion leads to CO and organic gas by-products, mainly due to

4


Table 1.1 Overview of important sources and global annual emission rates of selected groups of VOC per year Emission rate

Uncertainty range

Fossil fuel use Alkanes Alkenes Aromatic compounds

28 12 20

15–60 5–25 10–30

Biomass burning Alkanes Alkenes Aromatic compounds

15 20 5

7–30 10–30 2–10

Terrestrial plants Isoprene Sum of monoterpenes Sum of other VOC

460 140 580

Oceans Alkanes Alkenes Sum of anthropogenic and oceanic emissions Alkanes Alkenes Aromatic compounds

1 6

200–1 800 50–400 150–2 400 0–2 3–12

44 38 25

Terrestrial plants

1 180

Total

1 287

lack of oxygen, imperfect air/fuel mixing and inappropriate combustion temperatures. Tailpipe emissions from gasoline passenger cars with and without three-way catalyst are estimated, respectively, as 0.68 and 18.92 g of HC/kg of fuel, whereas passenger diesel cars (produced after 1996) emit about 1.32 g of HC/kg of fuel. Similarly, diesel heavyduty vehicle emissions are estimated as 5.4 g of HC/kg of fuel. Exhaust gas emissions from motor vehicles strongly depend on parameters such as vehicle speed, motor load and engine temperature. The predominate emissions for gasoline and diesel combustion engines are −C5 for gasoline cars and methane for diesel engines), C2 − −C5 olefins, ethyne, paraffins (C1 − aromatic hydrocarbons (BTEX and C9 aromatics), aldehydes (formaldehyde, acetaldehyde, acrolein, benzaldehyde, tolualdehyde), ketones (acetone) and others (mainly high molecular weight paraffins). A smaller emission contribution of anthropogenic gases comes from the solvents industry, and global inventories of these anthropogenic emissions have been compiled (Friedrich and Obermeier 1999; Olivier et al. 1999). These ‘anthropogenic’ emissions are discussed in detail in Chapter 2 of this book. A further strong source of global emissions is from burning of biomass, and giant smoke plumes can nowadays be seen easily on satellite images, especially in the tropics during the dry season (September–October). These emissions are the most difficult to assess as sources, as they are highly dependent on fuel type, humidity and burn rate amongst other


5

factors (Lobert et al. 1990). Spatial and temporal variability further complicates global budget assessments, and satellite measurements are now being used to monitor the size and location of burning regions (e.g. Duncan et al. 2003). Most burning occurs during human-initiated land clearance but a large component also comes from the domestic use of biomass fuels (Levine 2003). A comprehensive summary of organic gas emissions from biomass burning relative to CO2 has been made recently (Andreae and Merlet 2001). On a global scale, the total amount of reactive biogenic emissions is not well established, although recent estimates indicate that c. 1300 TgC/year are emitted (Guenther 2002). The strongest biogenic emission is thought to be isoprene (C5 H8 ), followed by the less specified so-called other reactive biogenic compounds that are mainly oxygenated compounds and monoterpenes. Biogenic sources in total are considered to be approximately ten times larger than the sum of anthropogenic emissions including fossil fuel emissions and biomass burning (Guenther 2002; Muller 1992; Olivier et al. 1999). In comparison to terrestrial sources, emissions from the ocean are less well constrained although several important species are known to have a predominately marine source dimethyl sulphide (DMS, see Section 1.6) (Groene 1995) and methyl iodide (Lovelock 1975). A relatively small amount of organics is emitted from the ocean in the form of alkanes and alkenes, c. 5 TgC/year (e.g. Broadgate et al. 1997; Ratte et al. 1993). Recently global oceanic isoprene emissions have been estimated from satellite-derived chlorophyll map and laboratory studies 0.1 TgC/year (Palmer and Shaw 2005), that is, much smaller than the terrestrial source (c. 500 TgC/year). However, the surface ocean has been shown recently to be a massive reservoir for oxygenated organic species (Singh et al. 2003; Williams et al. 2004). Furthermore, a recent study of aerosols at a coastal site in Ireland (O’Dowd et al. 2004) showed that the organic fraction contributes significantly (63%) to the sub-micrometer particle mass of aerosols collected over the North Atlantic Ocean during phytoplankton bloom periods. Biogenic emissions in general are discussed in Chapter 3 of this book, while Chapter 4 includes a section concerning the biogenic formation of OVOCs. From a global perspective, geographical location and season determine the relative importance of anthropogenic and biogenic emissions: biogenics are emitted mostly in the tropics whereas most anthropogenic emissions occur in the northern hemisphere between 40◦ N and 50◦ N. All these diverse organic emissions are broken down in the atmosphere into a wider array of partially oxidised species (Atkinson 1994; Atkinson and Arey 2003; Jenkin et al. 1997) and many thousands of gases have been detected in the atmosphere, from the tropics to Antarctica (Ciccioli et al. 1996; Zimmerman et al. 1988).

1.3

Sinks

Since the concentrations of organic trace species do not all simply increase with time there must logically be one or more removal processes (here termed sinks) acting on these compounds. The most important sink for organic trace gases in the atmosphere is chemical oxidation in the gas phase by the hydroxyl radical HO (or to a lesser extent O3 , NO3 and halogen radicals) (Atkinson 1994; Jenkin et al. 2003; Saunders et al. 2003). Certain gas phase organic compounds in the air can absorb sunlight and thereby photolyse to smaller fragments. Some compounds can be efficiently removed physically by dry deposition to surfaces such as vegetation (Doskey et al. 2004; Muller 1992) or aerosol (Cousins and

6


Mackay 2001); or removed by wet deposition in rain (Fornaro and Gutz 2003; Kieber et al. 2002). The gas phase oxidation of organic compounds in air is mostly initiated by the HO radical, with carbon dioxide and water being the final products. In this way atmospheric oxidation is analogous to combustion. Using an everyday example as an analogy, when a cigarette lighter is lit, the hydrocarbon butane burns directly in the flame to form H2 O and CO2 . When the flame is not ignited, then the escaping butane gas is oxidised in the air to the same products, only much more slowly and via many other intermediates. The intermediate oxidation products may have lower vapour pressures, higher polarity or absorb light better than the precursors, making the intermediate products potentially more susceptible to physical removal or photolysis. An alkane must be larger than C20 to be adsorbed onto solid particles (Bidleman 1988), but much smaller multi-functional organic compounds, such as oxalic acid, more readily adsorb and are commonly found on aerosols (Mochida et al. 2003). Further oxidative transformation of these species on the aerosol is also possible (Claeys et al. 2004a; Noziere and Riemer 2003). The overall rate of removal of an organic species from the atmosphere can be derived by summing the reaction rates with radical species, rates of photolysis and the wet and dry deposition rates. From this we may determine the atmospheric lifetime of a species (see spatial distribution section). The rate of reaction of HO with many individual organic compounds under terrestrial conditions is well established from laboratory experiments as a function of temperature and pressure (http://www.iupac-kinetic.ch.cam.ac.uk/ and Mannschreck et al. 2002). Table 1.2 shows the atmospheric lifetime of several commonly measured VOCs with respect to OH, with lifetimes varying from months to minutes. Likewise, global photolysis rates can be calculated for many compounds from laboratory absorption cross section and quantum yield measurements (http://www. iupackinetic.ch.cam.ac.uk/). These rates can be profoundly influenced by clouds and this in turn can affect trace gas concentrations (Tie et al. 2003). The wet and dry deposition rates for organic compounds are highly variable and are generally empirically determined in the field. Generally, organic compounds measured at high and invariable concentrations in the atmosphere are less efficiently removed (Junge 1974). Relationships between the variability of organic gas measurements and their rate of removal by HO have been derived (Jobson et al. 1999; Williams et al. 2000) and exploited to derive HO trends. If a long-lived and hence well-distributed organic compound is known to react predominately with HO, and its emission and HO reaction rate are known, then the global HO concentration can be theoretically estimated. Initial attempts based on methyl chloroform indicated large changes in HO concentrations from 1978 to 2003 (Prinn et al. 2001, 2005) with a maximum in 1989 and a minimum in 1998, although other recent evidence suggests that uncertainty in the temporal and spatial emission pattern of methyl chloroform complicates such trend analysis (Krol et al. 2003). Direct biological uptake can also be an effective atmospheric removal process for some organic species (Kesselmeier 2001; Kuhn et al. 2002). The rate of uptake is dependent on the ambient concentration, being strongest when ambient concentrations are high. Compensation points have been deduced for plants, which mark the crossover point between emission and uptake. A surprising recent discovery is that peroxy acetyl nitrate (PAN), an anthropogenic secondary oxidant like ozone, can also be taken up by plants (Sparks et al. 2003). This is an important development for the atmospheric nitrogen cycle as well as the organic species PAN.


7

Table 1.2 Overview of average tropospheric lifetimes of VOC compound groups and some selected VOCs as examples. Lifetimes are given for an average OH concentration of 6 × 105 per cm3 and an average ozone concentration of 7 × 1011 per cm3 (about 30 ppb)

1.4

Compound

Average lifetime

Alkanes Ethane Propane n-Pentane

Months–days 2.5 months 2.5 weeks 4 days

Alkenes Ethene Propene 1-Butene

Days–hours 1.5 days 11 h 10 h

Cyclic compounds Cyclopentane Methylcyclohexane Cyclohexane

Days–hours 4 days 2 days 3h

Aromatic compounds Benzene Toluene 1,3,5-Trimethylbenzene

Weeks–hours 2 weeks 2 days 7.5 h

Biogenic compounds Isoprene α-Pinene Limonene

Hours–minutes 3h 4h 30 min

Atmospheric distribution

Following emission, volatile organic species are distributed by atmospheric transport processes while undergoing the removal processes described in Section 1.3. The relationship between the atmospheric removal rate (lifetime) of the compound and the average mixing times for the various sections of the atmosphere together determine the extent to which a compound is globally distributed. Atmospheric mixing is physically impeded across the boundary layer temperature inversion (0.5–2 km), the tropopause (10–15 km) and the intertropical conversion zone (ITCZ, 10◦ S–10◦ N), and as a result strong gradients in organic species can develop across these atmospheric interfaces. Typical exchange times are 1–2 days for air to mix vertically out of the boundary layer, 2 weeks to a month for air to be advected zonally around the northern or southern hemisphere, about 1 year for interhemispheric exchange and 4–6 years for an exchange between troposphere and stratosphere. Chemical lifetimes, defined as the time for a chemical concentration to decay to 1/e of its initial value, varies from minutes to hours (terpenes and isoprene), through days to weeks (acetone, methanol, propane), years to decades (methyl chloroform, HCFC 134a), and up to hundreds

8


of years for chlorofluorocarbons (CFC 11 and CFC 12) whose lifetime is determined by the mixing rate into the stratosphere. Short-lived compounds, such as the biogenic species isoprene (CH2 C(CH3 )CHCH2 ), show strong atmospheric gradients within the boundary layer (0–2 km), whereas longer-lived compounds, such as CFC 113 (lifetime c. 8 years), are better mixed and only show strong gradients between the hemispheres (Boissard et al. 1996; Bonsang and Boissard 1999; Rudolph 1998). Some compounds are more or less uniformly distributed in the troposphere (CFC 12, lifetime 79 years, no remaining sources), only showing concentration gradients in the stratosphere. A large archive of aircraft measurements taken at various locations over the globe is available at http://www-gte.larc.nasa.gov/. There are numerous examples in the literature of regional scale advection where organic pollutants found in remote locations have been linked to distant pollution sources by use of back trajectories (e.g. Blake et al. 1996; Traub et al. 2003). Intercontinental pollution events have been reported (Price et al. 2004; Stohl and Trickl 1999) and trajectories have even been used to track southern hemispheric biomass burning, through the ITCZ to the upper troposphere of the northern hemisphere (Andreae et al. 2001). Secondary photooxidants, such as ozone and PAN, that form en route have also similarly been identified in plumes emerging from urban centres (e.g. Rappengluck et al. 2003). Interestingly, there is growing evidence to suggest that migrating birds use chemical gradients as an olfactory aid to navigation (Wallraff 2001, 2003). Where the atmosphere is in contact with the Earth, organic species can interact with the various surfaces, such as snow, soil and water (Ballschmiter 1992). Within these media further production or removal mechanisms may exist such as bacterial uptake, enhanced photolysis (Dominé and Shepson 2002; Klán et al. 2003) or biological production. Such processes will affect the lifetime of these species and hence their global distribution. Some higher molecular weight organics with considerably lower vapour pressures tend to partition predominantly to aerosols following release. When such a compound is unreactive, as with persistent organic pollutants (POPs), which are emitted through incomplete combustion or pesticide use, the lifetime of the transporting aerosol will then determine the distribution of this compound. Examples of such compounds include polyaromatic hydrocarbons (PAHs, e.g. Mastral and Callén 2000), Polychlorinated biphenyls (PCBs) and polychlorinated dibenzo-p-dioxins (PCDDs). Whether in gas form or as particles, these compounds can be transported long distances from source regions (Patton et al. 1991). The distribution of the long-lived semi-volatiles is markedly different to that of the volatiles, and with time through repeated volatilisation and adsorption, such compounds tend to concentrate in polar regions (Burkow and Kallenborn 2000) in a manner that could be likened to a global distillation (from the tropics to the poles). Some of these compounds are toxic (Walker 2001a) and can bioaccumulate through the food web (Tanabe et al. 1984), posing a risk to human health and the environment (UNEP 2001). While the boundary layer (0–2 km) tends to be turbulent, the so-called free troposphere above is less well mixed. In addition to the slow process of diffusion, organic gases may be distributed in the atmosphere by meteorological events such as convection (Collins et al. 1999) and via lifting by frontal systems (Purvis et al. 2003). The overall distribution of the organic species varies with latitude and season as a function of the source and sink strengths, as well as prevailing meteorology (Bonsang and Boissard 1999; Singh and Zimmerman 1992). Certain photochemical products, such as organic nitrates (e.g. PAN or alkyl nitrates), have a hemispheric concentration that is maximum in the spring. This has


9

been explained as the photochemical optimum between the high precursor source and low photochemical sink in wintertime, when PAN precursors are accumulated, and the high photochemical sink in summer (Penkett 1983). In the early years of atmospheric research, it was assumed that after several days to weeks the atmosphere would have effectively removed an organic pollutant, based on the atmospheric lifetime of alkanes. Recently, however, from measurements made between 1 and 13 km over the remote Pacific Ocean, far from source regions, it was shown that volume mixing ratios of oxygenated organic species are some five times higher than those of the NMHC, alkanes and alkenes (Singh et al. 2001, 2004). Similar high mixing ratios of oxygenates and compound diversity have been reported in other airborne studies (Crutzen et al. 2000), in urban centres (Lewis et al. 2000) and in continental outflow from Asia (Jacob et al. 2003; Lelieveld et al. 2001) and Europe (Salisbury et al. 2003). These results concur with earlier theoretical work on the oxidation of organic compounds (Calvert and Madronich 1987; Madronich et al. 1990). Our views about the distribution, sources and role of reactive organic species in the atmosphere are currently being revised rapidly.

1.5

Measurement tools

The human nose is particularly sensitive to several chemical groups (Cain 1979; Firestein 2001). Familiar examples include forests, which emit terpenes (e.g. Geron et al. 2000; Isidorov et al. 2000); oil refineries, which emit aromatic compounds and alkenes (e.g. Doskey et al. 1999); fish markets, which emit amines (Morita et al. 2003) and freshly cut onions, which emit sulphur compounds (e.g. Ferary and Auger 1996). While human subjects are widely used in odour identification studies (Ferreira et al. 2003; Walker 2001b), the nose’s response is inherently subjective (Molhave et al. 1991) and difficult to quantify. Therefore, to investigate the atmosphere quantitatively, researchers have employed a variety of sensitive and specific sensors, including mass spectrometers, flame-ionisation detectors, electroncapture detectors, optical absorption, chemiluminescence and atomic emission detectors (e.g. Apel et al. 1998; Helmig 1999; Kormann et al. 2002; Sigrist 2003). In many cases such detectors are coupled to pre-separation devices, for example, a gas chromatograph so that individual gases may be isolated prior to detection and a single specific compound can be measured. Both animal- and plant-type biological detectors have also been deployed for detection of certain molecules. In some studies the amputated sensory antennae of small insects have been connected into measurement devices (e.g. Murlis et al. 2000) and elsewhere the leaves of plants have been analysed for long-term exposure statistics (Hiatt 1999). Ideally for air studies a detectors should also be capable of measuring the huge range of concentrations in the atmosphere. High mixing ratios of several tens of ppbv (nmol/mol) can be found for alkanes and aromatics in polluted urban areas urban areas (Derwent et al. 2000), while halons must be reliably measured at only 0.045 pptv (pmol/mol) (Fraser et al. 1999). The recently reported compound SF5 CF3 was first detected at 0.005 pptv (Sturges et al. 2000). This means that if only 200 tonnes of such material would be emitted anywhere in the world it would, in time, be detectable by this instrument. Global networks of detectors are in currently place to routinely monitor changes in greenhouse gases (Prinn et al. 2000). Much of our atmospheric knowledge to date has been driven by what can be reliably measured and how fast (Roscoe and Clemitshaw 1997). Although the first atmospheric

10


research on organic trace gases (specifically PAN) was made using infrared spectroscopy (Stephens 1961; Stephens et al. 1956), the following 30–40 years of research on atmospheric organic gases have been dominated by gas chromatography (e.g. Darley et al. 1963; Helmig 1999) coupled to some form of detector. Samples have been either introduced directly into the instrument in the field or collected in pressurised steel canisters, absorbent packed cartridges or filters for later analysis in the laboratory. The alkanes (major components of fossil fuel) were one of the first and most widely investigated subset of the reactive organic species (Blomberg et al. 2002). This is because these fully saturated compounds do not interact strongly with most inlet materials or collection vessel surfaces, and the long established technique of gas chromatographic separation with flame ionisation detection has allowed widely available quantitative analysis (Helmig 1999). Many oxidised gases are more difficult to quantify as they may stick to surfaces, thermally decompose or may even be produced in measurement systems (Bates et al. 2000; Helmig et al. 1996; Kelly and Holdren 1995; Lestremau et al. 2001; Li-Jones et al. 2001; Tanner 2003). These techniques are sensitive and specific, but due to the predetector sample separation, these are limited in sampling frequency. Recently, several important new advances have been made in analytical techniques. These have permitted more organic species to be measured, such as high precision and sensitivity so as to enable δ 13 C isotopic ratios to be determined in organics at mixing ratios below ppbv levels (Rudolph et al. 1997) (see Chapter 10 of this book). Higher frequency measurements have been made possible with chemical ionisation mass spectrometry (CIMS) through use of proton transfer reactions (Hewitt et al. 2003; Lindinger et al. 1998; Williams et al. 2001) (see Chapter 11 of this book), or by other chemical ionisation techniques (Heeb et al. 1999; Leibrock et al. 2003). Several of these high frequency methods have been further developed to measure emission fluxes directly (Bowling et al. 1998; Karl et al. 2004; Warneke et al. 2002). Furthermore, improvements in the field of gas chromatography (e.g. multi-dimensional gas chromatography or comprehensive chromatography, Phillips et al. 1985) have delivered considerable improvements in compound separation, identification and sensitivity (see Chapter 12 of this book). The physical separation of even enantiomeric monoterpenes, or optical isomers is also now possible from ambient air (Yassaa and Williams 2005). With the arrival of this new generation of measurement systems, more species and timescales are accessible and a new golden age of discovery for field measurement has begun. Researchers are now exploiting these latest techniques on planes, ships, balloons and ground sites to establish the global budgets of a wide range of organic species.

1.6

Modelling tools

A variety of numerical models is available today to simulate chemistry and transport in the atmosphere from the level of box models to three-dimensional chemistry and transport models. To simulate atmospheric chemistry in detail, models need to deal with VOCs which play a significant role in all reaction cycles in the atmosphere. In this regard the major input data they require are (a) the emission inventories describing the primary emission of VOC including their specific source compositions and their spatial and temporal variations, (b) the oxidation chemistry of VOC including the kinetics as a function of temperature and pressure and (c) for certain species a consideration of other significant loss processes such as


11

dry and wet deposition. Understandably, due to the huge number of organic compounds in the troposphere, both a complete emission inventory of all possible compounds with the necessary resolution in time and space and a complete coverage of all possible chemical reactions including those of secondary reaction products will never be available. There are only a few comprehensive emission inventories available and these cover the most important compounds relevant for atmospheric chemistry. The global distribution and source strengths of anthropogenic NMHC are usually taken from the Emission Database for Global Atmospheric Research (EDGAR V2.0) database (Olivier et al. 1996, 1999). Detailed information on this database and access to the data is available on http://www.mnp.nl/edgar/. The EDGAR database details sources of fossil-fuel-related activities, biofuel combustion, industrial production and consumption processes (including solvent use) on a per country basis, land-use-related sources, including waste treatment, partially on a grid basis and partially on a per country basis; and natural sources on a grid basis. The database can be used to generate global, regional and national emission data in various formats. For all compounds the reference year is 1990, except for halocarbons, for which 1986 is the reference year. In 2001, version 3.2 was released which comprises emissions by region and source for the period 1990–5. However, organic compounds emissions given in the EDGAR database are known to have large uncertainties in both magnitude and distribution of the emissions. An inventory of biomass burning and natural VOC emissions can be found in the Global Emissions Inventory Activity (GEIA) database. Details of this database can be found on http://www.geiacenter.org/. The biogenic VOC (BVOC) dataset consists of three files that cover isoprene, terpenes and ‘other’ NVOC data. This database provides BVOC emission measurements and modelling parameters and, in addition, an enclosure database that summarises information from literature and identifies the plant species and the BVOC studied, including enclosure and analytical techniques and other parameters. Presently, about 1 800 plant species from which BVOC emissions have been studied are documented. As will be described in the following chapters, for individual VOC compound classes the chemistry of VOC can be quite complex. Two main approaches are used to simulate the complex chemistry of VOC using numerical models. One approach is to significantly reduce the number of organic compounds and hence the complexity of the corresponding reactions. Three methods are used to reduce the number of organic reactions, namely the carbon bond mechanism, the surrogate species method and the lumped species method. In the carbon bond mechanism, individual organic compounds are segregated into one or more bond groups that have a similar chemical reactivity (cf. Gery et al. 1988). With the surrogate species method all VOC of similar reactivity are grouped together. The rate coefficient of each of these compounds is then set equal to that of one particular compound (cf. Atkinson et al. 1982). In the lumped species method, VOC are grouped by their reactivity towards reactions with HO radicals. The rate coefficient is determined by taking the mole fraction weighted average of the reaction coefficient of each compound of the lumped group. Currently the most widely used lumped mechanism is the Regional Acid Deposition Model (RADM) (Stockwell et al. 1990). This mechanism contains 158 chemical reactions and 63 gaseous compounds. Besides primarily emitted inorganic compounds and 16 organic compounds or compound groups, respectively, RADM also includes photochemically produced compounds. This mechanism has meanwhile been updated into a version named

12


Regional Atmospheric Chemistry Model (RACM) (Stockwell et al. 1997) and includes reaction schemes for isoprene and the group of monoterpenes. The RACM mechanism again has been improved (Regional Atmospheric Chemistry Model-Mainz Isoprene Mechanism, RACM-MIM) by implementing an explicit chemistry of isoprene (Geiger et al. 2003), the most important BVOC (see Chapter 3). Today RACM is widely used in a variety of mathematical models, ranging from box models to three-dimensional chemistry and transport models. However, the mechanism is also available as a stand-alone model for simulating tropospheric chemistry at typical ambient trace gas concentration levels. The aforementioned methods of condensing VOC reactions have, of course, some disadvantages. The results of a model investigation of a single hydrocarbon oxidation are likely to be erroneous. Investigation of the RACM simulation showed, for example, that for a simulation of the oxidation of some individual VOC, such as branched alkanes, the error can be considerable. Therefore, other approaches have been developed that try to describe the complex chemistry of organics as explicitly as possible, leading to a vast number of chemicals and reactions which have to be taken into account. For example, the widely used Master Chemical Mechanism (MCM) is a detailed chemical mechanism containing 12 600 reactions and 4 500 chemical species (Jenkin et al. 2003). This mechanism includes the complete tropospheric oxidation of 124 VOC. The VOC which are degraded in this mechanism were selected on the basis of available emission data and provide approximately 90% mass coverage of the emissions of uniquely identifiable chemical species. The majority of the degradation schemes have been constructed using the methodology described by Jenkin et al. (1997). A review and update of the ideas behind the mechanism as well as recent developments can be found on the corresponding web page (http://mcm.leeds.ac.uk/MCM/).

1.7

How organic species affect the atmosphere

In the 1950s Haagen Smit and co-workers showed that the oxidation of organic species in the presence of NOx and sunlight can form ozone (Haagen Smit 1952). Ozone, which is toxic to humans and plants, has become a major air quality problem in cities and larger areas such as the Mediterranean (Lelieveld et al. 2002) and the south-eastern United States (Solomon et al. 2000). Ozone control strategies adopted in the 1970s were initially unsuccessful due to an underestimation of natural organic emissions in the initial models (Trainer et al. 1987). However, more recent emission controls applied to cars (including NOx and organic species reductions) have reduced regionally produced ozone (Derwent et al. 2003). The capacity of the troposphere to oxidise emissions is also dependent on the amount of organic compounds present. Reaction with the main atmospheric oxidant, the HO radical, is the primary loss mechanism of organics from the atmosphere (see Section 1.3). While the initial reaction is a sink, subsequent oxidation steps may be a source of HOx (HO and HO2 ), making the global effect of organic species complicated. In cities where NOx concentrations are high, increasing concentrations of organics increase the ambient HO. However, in most of the free troposphere HO production is not limited by organics but rather by NOx , and increasing organic concentrations generally decreases ambient HO under these conditions (Wang et al. 1998). In the upper troposphere, where water concentrations are low (4.5 yearf

>8 year (Atkinson et al. 1999) >300 days (Reissell et al. 2001) 1.5 year (Corchnoy and Atkinson 1990) 4.1 h (Atkinson et al. 1995) 4.5 h (Atkinson et al. 1995) 6 min (Atkinson et al. 1995) 2.0 year (Atkinson et al. 1999) 7.7 days (Rudich et al. 1996) 9 min (Smith et al. 1996)

Oxygenated VOCs Acetonee Camphor 1,8-Cineole cis-3-hexen-1-ol cis-3-hexenyl acetate Linalool Methanol MBO 6-methyl-5-hepten-2-ol

>235 days (Reissell et al. 2001) >110 days (Atkinson et al. 1990) 6.2 h (Atkinson et al. 1995) 7.3 h (Atkinson et al. 1995) 55 min (Atkinson et al. 1995) >4.5 yearf 1.7 days (Grosjean and Grosjean 1994) 1.0 h (Smith et al. 1996)

Source: Reprinted from Atmospheric Environment, 37: S197–219. Copyright (2003), with permission from Elsevier. a From Calvert et al. (2000) unless noted otherwise. b Assumed OH radical concentration: 2.0 × 106 molecules/cm3 , 12-h daytime average. c Assumed O concentration: 7 × 1011 molecules/cm3 , 24-h average. 3 d Assumed NO radical concentration: 2.5 × 108 molecules/cm3 , 12-h nighttime average. 3 e Photolysis will also occur with a calculated photolysis lifetime of ∼60 days for the lower troposphere, July, 40◦ N Meyrahn et al. (1986). f Estimated.

Biogenic VOCs

105

Similarly, oxVOCs can be important for controlling radical mixing ratios. Lamanna and Goldstein (2002) estimate that MBO accounts for 21% of OH loss over conifers in the Sierra Nevada of California. While biogenic VOCs are normally considered to be sinks for atmospheric radicals, their oxidation in the atmosphere can also provide a new source of radicals. Reactions of ozone with alkenes including terpenes can have high yields of OH radicals (Paulson and Orlando 1996; Paulson et al. 1999). Additionally, photolysis of some oxVOCs such as acetone and acetaldehyde in the upper troposphere can also lead to production of HOx radicals (Wennberg et al. 1998). These impacts are discussed further in Section 3.5.2.

3.4.2

Source distribution

Because of their high reactivity, atmospheric mixing ratios of biogenic VOCs are highly variable and usually concentrated near the emission source. As described in Section 3.3, the source distribution of biogenic VOCs is primarily dependent on the land cover at the Earth’s surface. Emissions are based on the type of vegetation and the amount of biomass present at each location, causing a wide range of temporal and spatial variability in emissions. To give a general idea of the distribution of these source emissions, global isoprene and monoterpene emissions in July (generally considered the global seasonal maximum) are shown in Figures 3.3 and 3.4 as modelled using the G95 algorithm (Guenther et al. 1995). Isoprene is concentrated in regions with large amounts of deciduous forests, such as the southeastern United States, Europe, central Africa, and the tropical regions of Eurasia. Monoterpene emissions, which are considered to be primarily coniferous, are greatest in the western United States, Scandinavia and the forests of Russia. Global distributions of other classes of compounds or VR-BVOCs are still at an early stage of investigation, and thus current global estimates are not available.

3.4.3

Mixing ratio distributions

The mixing ratio distributions in the atmosphere result from the balance between the sources and sinks of biogenic VOC emissions. We have discussed the source distribution of emissions, as well as the reaction with radicals as a major sink for biogenic VOCs. Because of their fast reactivity, transport and deposition processes play minor roles in the global mixing ratio distribution for most biogenic VOCs. Therefore, horizontal surface mixing ratio distributions generally mirror that of the source emissions shown in Figures 3.3 and 3.4. Temporally, different biogenic VOC species have varying diurnal profiles. Isoprene mixing ratios have a strong diurnal cycle with daytime values ranging from 0 to 10 ppb (depending on vegetation sources) peaking at midday, and nighttime values on the order of several to hundreds of ppt. Monoterpenes have a very different diurnal cycle primarily when their emissions are independent of light. Mixing ratios at night can range up to 3 ppb near source regions and are generally higher than daytime because they can accumulate in a shallower nocturnal boundary layer (Biesenthal et al. 1998; Hakola et al. 2000; Lamanna and Goldstein 2002). The diurnal behaviour of oxVOC is highly variable depending on the emission species. Light- and temperature-dependent emissions such as MBO have a similar

0

00

–2

50

0

0

00 –2

00 15

20

0

50

25

–1

–1 12

50

00 10

0–

00

10

0

50 0–

25

50

25

10

0–

0–

10

0


0

106

00 50 0– 10 10 00 00 –1 25 0

25 0– 5

25 0

10 0–

0–

10 0

Figure 3.3 Global distribution of isoprene fluxes in July, in mg C/m2 , based on the GEIA database (Guenther et al. 1995).

Figure 3.4 Global distribution of monoterpene fluxes in July, based on the GEIA database (Guenther et al. 1995).

Biogenic VOCs

107

diurnal cycle to that of isoprene, with peak daytime values reaching 2–5 ppb at midday (Schade and Goldstein 2001). Other biogenic oxVOCs such as acetone and acetaldehyde often have daytime values of approximately several ppb, with mixing ratios that are slightly higher during the daytime than in the nighttime. Methanol typically has mixing ratios about a factor of five higher than other oxVOCs in forested regions (Goldan et al. 1995b; Schade and Goldstein 2001). VR-BVOC mixing ratios have been measured although they are often quite low (on the order of several ppt) due to their high reactivity. Recent work by Holzinger et al. (2004) has found that these emissions may in fact be higher than previously assumed, based on the measurements of their oxidation products and their rapid oxidation in the forest canopy. Vertical variations can also affect the spatial mixing ratio distribution, depending on their transport and fate in the atmosphere. Because of their high reactivity, biogenic VOCs generally do not escape the boundary layer. Based on observations of vertical profiles of biogenic VOCs at various sites, mixing ratios of isoprene and monoterpenes generally decrease rapidly with increasing heights (Andronache et al. 1994; Helmig et al. 1998b; Warneke et al. 2001). Many vertical profiles of oxVOC species have now been performed, yet these compound distributions are more complicated than isoprene and terpenes because of the combination of primary and secondary anthropogenic sources in addition to primary and secondary biogenic sources. Vertical distributions over the Atlantic (Singh et al. 2000) showed decreasing mixing ratios with increasing altitude for acetone and methanol, while in the Pacific (Singh et al. 2001), mixing ratios of methanol (0.9 ppb), acetone (350 ppt) and acetaldehyde (60–100 ppt) were relatively constant with height. Scheeren et al. (2003) measured several oxygenated species in and above the boundary layer over the Eastern Mediterranean, and showed that acetone and methanol (approximately several ppb up to 6 km) decrease with increasing height though much less rapidly than isoprene and monoterpenes, as expected due to their longer lifetimes.

3.5

Impact on photooxidants and atmospheric chemistry

The work of Haagen-Smit in the 1950s determined that photochemical smog was produced by a series of photochemical reactions involving hydrocarbons and nitrogen oxides in the presence of sunlight. In the initial attempts at ground-level ozone control in the 1970s, this led to the control of both nitrogen oxide and anthropogenic hydrocarbon emissions. However, research in the 1980s indicated that in the presence of sufficient NOx , biogenic VOCs can play an important role in the formation of ground-level ozone in both rural and urban regions (Chameides et al. 1988; Trainer et al. 1987), because biogenic VOCs tend to be more reactive than their anthropogenic counterparts (Atkinson and Arey 1998) and their emission rates are relatively high compared to anthropogenic VOCs. As research has progressed, the presence of biogenic VOCs has been shown to strongly influence tropospheric ozone and NOx chemistry at the regional and global scale (e.g. Horowitz et al. 1998; Pierce et al. 1998; Poisson et al. 2000). In this section, we discuss the role of biogenic VOCs in tropospheric-gas- and aerosolphase chemistry. This topic has been reviewed extensively (e.g. Atkinson and Arey 1998, 2003; Monson and Holland 2001); thus, we provide only a brief discussion here. We discuss some of the major impacts of biogenic VOC in the troposphere, including their effects on

108


photooxidants and ground-level ozone, their influence on the NOx distribution, and the ability of their oxidation products to undergo gas-to-particle conversion, leading to the growth of secondary organic aerosols.

3.5.1

Reaction pathways

Depending on the location and time of day, biogenic VOC emissions generally react with OH or NO3 radicals, O3 or the chlorine atom. For most biogenic VOCs, the reaction with oxidants in the atmosphere occurs by either of the following two main mechanisms: 1. the addition of O3 or OH/NO3 radicals to the double carbon bond in the biogenic VOC; 2. the abstraction of an H atom from the hydrogen–carbon bond by OH or NO3 (Finlayson-Pitts and Pitts 2000). Most biogenic VOCs are likely to follow the addition mechanism rather than the abstraction mechanism, with the exception of double carbon bond aldehydes that tend to react via abstraction (Atkinson and Arey 2003). The following discussion describes the cycle of reactions shown in Figure 3.5. Once the biogenic VOC molecule has been oxidised via the addition or abstraction mechanism, Biogenic VOC

OH

R• O2

ROOH

HO2

RO2•

RO2•

NO2 ROONO2 NO

Carbonyl + Alcohol

RONO2

RO•

Decomposition

O2

Isomerisation

Products

Figure 3.5 General biogenic VOC reaction pathway. Reprinted from Atmospheric Environment, 37: S197– 219. Copyright (2003), with permission from Elsevier.

Biogenic VOCs

109

the primary product is an alkyl radical (R• ). The alkyl radical then immediately reacts with O2 to form an alkyl peroxy (RO•2 ) radical. After the alkyl peroxy radical is formed, it can react with NO, NO2 , HO2 or other alkyl peroxys (RO•2 ). Reactions with NO2 and HO2 terminate the reaction sequence by forming peroxy nitrates (ROONO2 ) and peroxides (ROOH), respectively. However, if the reaction sequence continues via reactions with NO and HO2 , they can yield other important reaction products (Seinfeld and Pandis 1998). The NO pathway is important because it can often lead to the formation of NO2 and, hence, ozone (when NO2 is photolysed forming O, then reacting with O2 to form O3 ), while the HO2 pathway is important because it forms relatively stable products and acts as a sink for radicals. Generally, if NO mixing ratios are low (e.g. in a rural environment), the cycle tends towards the HO2 /RO2 reactions (or peroxy-peroxy reactions) that remove radicals from the system and retard ozone production. Reaction of the alkyl peroxy radical with HO2 forms a peroxide (ROOH), which can then be removed via wet/dry deposition, be photolysed (regenerating OH) or react with OH itself. Reactions of RO•2 can form a variety of products including alcohols, aldehydes and ketones. However, if NO is high such as in polluted regions, then peroxy radicals will tend to react with NO or NO2 . Reactions with NO2 form peroxy nitrates (ROONO2 ), which can also act as a reservoir species for NOx as they can be stable at surface-level temperature and pressure conditions. This stability allows them to be transported for long distances to later dissociate. The formation and fate of peroxy nitrates are discussed in greater detail in Chapter 6 of this volume. Reactions of RO•2 and NO can form either an alkyl nitrate (RONO2 ) or an alkoxy radical (RO• ). The alkyl nitrate is a more stable compound and can remove NOx from the cycle. However, if the reaction forms the alkoxy radical (RO• ), it proceeds to isomerise or decompose into a variety of products. This pathway can result in the formation of NO2 , triggering the production of more ozone. The fate of NO reactions (either forming RONO2 or RO) is determined by the ‘branching ratio’, a ratio representing the likelihood of each NO pathway. An example of biogenic VOC oxidation is the reaction of isoprene and the OH radical. There are six different possible hydroxyalkyl radicals that can form from this reaction, and Figure 3.6 shows one possible reaction pathway of isoprene and OH to completion (Atkinson and Arey 2003). In the pathway shown, OH adds to the primary carbon and then reacts with NO to form an alkoxy radical (assuming sufficient NO is present). OH addition on the first or second carbon occurs in approximately 66% of the isoprene–OH reactions and results in the production of methyl vinyl ketone and formaldehyde (Finlayson-Pitts and Pitts 2000). The other third of the reactions involve the OH addition on the third or fourth carbon, producing methacrolein (not shown). The primary products of the isoprene–OH reaction are methyl vinyl ketone, methacrolein and formaldehyde. Another important biogenic VOC oxidation sequence is the α-pinene-ozone reaction, shown in Figure 3.7. This is one of the dominant loss mechanisms for α-pinene and is representative of the type of reactions typically occurring between terpenes and ozone. When ozone adds to the double bond in α-pinene, an ozonide forms that rapidly decomposes to two different Criegee biradicals (Kamens et al. 1999). One of these biradicals (Criegee 1 in Figure 3.7) then reacts to form pinonic acid, norpinonaldehyde and norpinonic acid. The second biradical (Criegee 2) forms pinic acid and several other products. The primary products of this reaction sequence can undergo gas-to-particle conversion and

110


OH

CH3

CH3

+

C H 2C

•

CH2

CH2

C

H2COH

CH

CH O2 CH3 C

H2COH

CH

• OO CH2

NO

CH3 H2COH

C

CH

ONO2 CH2 CH3 H2COH

C

CH

•

O

+

NO2

CH2

Decomposition

O •

CH2O

+

O2 HCHO

+

CH2 CH Methyl vinyl ketone H3C

C

HO2

Figure 3.6 Isoprene and OH reaction pathway. (With permission from Finlayson-Pitts and Pitts 2000.)

form secondary organic aerosols, as well as form other important gas-phase species such as carbon monoxide, formaldehyde, and OH and HO2 radicals.

3.5.2

Effects on photooxidants

The oxidation of biogenic VOCs can lead to production of tropospheric ozone via the formation of NO2 . NO2 is formed from the reaction of NO and the peroxy radical, and is quickly photolysed after formation: NO2 + hν → NO + O(3 P) M

O(3 P) + O2 −→ O3

(3.7a) (3.7b)

where M represents a third body (usually an N2 or O2 molecule) required for the termolecular reaction of 3.7b. Because the cycle with biogenic VOC regenerates radicals, the presence

Biogenic VOCs

+

111

O3 O O O

O O

O O O Criegee 1

CHO

Criegee 2

HO2 + O

COOH

Pinonic acid

O

CHO

CO OH H2

Norpinonaldehyde

HOOC

+ Other products COOH

Pinic acid

oxidant

O COOH Norpinonic acid Figure 3.7 α-Pinene and ozone reaction pathway. Reprinted from Environmental Science and Technology, 33: 1430–8. Copyright (1999), with permission from American Chemical Society.

of these molecules in conjunction with sufficient NOx can increase the production of ozone in the boundary layer. This general reaction sequence represents the primary source of ozone in the troposphere. Modelling studies in the late 1980s and early 1990s indicated that biogenic VOCs, in particular isoprene, could be extremely important in estimating regional ozone in both rural and urban locations (Chameides et al. 1988; Council 1991; Trainer et al. 1987). Isoprene can strongly influence the ozone production efficiency (OPE), and was found to account for a 50% increase in OPE during the summertime in the northeastern United States (Hirsch et al. 1996). Regional studies in the western United States and Europe have also found that isoprene can contribute significantly to ozone production (Dreyfus et al. 2002; Singh et al. 2004; Tsigaridis and Kanakidou 2002). Global modeling studies have also shown that ozone formation is sensitive to the presence of isoprene. Wang et al. (1998) found that the lack of non-methane hydrocarbons (including

112


both anthropogenic and biogenic) in a global photochemistry model reduced ozone mixing ratios by 5–20%, with decreased production near the surface and in the upper atmosphere (above 300 mbar) and increased production in the middle troposphere (300–600 mb), and attribute approximately half of this effect to isoprene. Houweling et al. (1998) found that all non-methane hydrocarbon emissions caused a 17% increase in the global mean tropospheric ozone column and that isoprene from biogenic sources were causing about 70% of this increase. One example of the impacts of oxygenated biogenic VOCs is that of Tie et al. (2003). This study included biogenic methanol emissions in a global tropospheric chemistry model and found that adding these emissions slightly increased O3 (1–2%). Besides the impacts on ozone, the presence of biogenic VOC can also influence other photooxidants such as OH and HO2 . Near source regions, the presence of biogenic VOCs and their fast reactions with OH can lead to a significant sink of OH radicals. The methanol study by Tie et al. (2003) also found that biogenic methanol emissions can slightly increase HO2 mixing ratios (1–2%) and decrease OH (1–3%). This decrease in OH due to the addition of biogenic emissions has also been noted in smaller-scale, canopy-level models (Makar et al. 1999). Biogenic VOCs can also have other impacts on HOx cycling in the troposphere. While isoprene can be a sink for OH, the reaction of ozone with biogenic alkenes can also be a source of OH. This provides a potential OH source at times when mixing ratios would otherwise be negligible (e.g. nighttime) and has been found to be account for 10–15% of the total radical production in the southeastern United States (Paulson and Orlando 1996).

3.5.3

Formation of organic nitrates and sequestering of NOx

Global atmospheric chemistry studies in the late 1980s and 1990s suggested that the export of anthropogenic NOx could influence ground-level ozone formation in other regions (Jacob et al. 1993; Liu et al. 1987; Mauzerall et al. 1996). One of the methods of transporting NOx is via the formation of peroxy nitrates. As shown in Figure 3.5, peroxy nitrates are formed by the reaction of an alkyl peroxy radical (RO2 ) with NO2 and have a generalised structure of ROONO2 . These relatively stable products can then be transported long distances to be dissociated in other regions, leading to the formation of ozone far from the original sources of anthropogenic pollution. Horowitz et al. (1998) found that peroxy nitrates formed from isoprene are the principle peroxynitrates around the globe. The topic of peroxy nitrates is discussed in further detail in Chapter 6 of this volume.

3.5.4

Formation of secondary organic aerosols

Aerosols, which are liquid or solid particles suspended in the atmosphere, play an important role in the chemistry and climate of the atmosphere. They can scatter and absorb incoming radiation, affecting the radiative budget at the surface of the Earth, and provide a site for heterogeneous reactions in the atmosphere. Biogenic secondary organic aerosols (SOA) can be formed via two possible methods. The first is when gas-phase oxidation products condense onto existing particles in the atmosphere. This occurs because some oxidation products have low volatilities, allowing them to condense on pre-existing particles while trying to establish equilibrium between the gas and particle phases. The ability of a biogenic

Biogenic VOCs

113

VOC to form SOA depends on its atmospheric abundance, its chemical reactivity and the volatility of its products (Seinfeld and Pandis 1998). The second formation mechanism is via nucleation, or the formation of new particles, although there is no solid evidence at this time that nucleation is a significant source of aerosol to the atmosphere (Kanakidou et al. 2004). Went (1960) first discussed the biogenic VOC contribution to haze observed over forested regions. Subsequent studies in the 1980s and 1990s found that the reaction of monoterpenes with photooxidants could produce large amounts of aerosols in smog chamber experiments (Kamens et al. 1981; Pandis et al. 1991). Later studies attributed monoterpenes and sesquiterpenes as the primary contributors to the formation of secondary organic aerosols (Griffin et al. 1999a; Hoffmann et al. 1997). These initial smog chamber studies concluded that gas-to-particle conversion was not occurring for the oxidation products of isoprene. Biogenically formed aerosols can make up a large fraction of the particulate mass of the atmosphere. Pandis et al. (1991) estimated that monoterpenes could account for a range of 15–50% of the particulate matter in Los Angeles. Global inventories by Andreae and Crutzen (1997) estimated that SOA formation from biogenic VOCs was 30–270 Tg/year, while a subsequent study by Griffin et al. (1999b) used the equilibrium principle in conjunction with smog chamber experiments to estimate an annual total of 18.5 Tg of aerosol from biogenic hydrocarbons. In contrast to previous studies, recent investigations have concluded that isoprene can be a precursor of secondary organic aerosols via a different mechanism than that of monoterpenes. Compounds such as carbonyls and dienes (e.g. isoprene) can undergo conversion via heterogeneous reactions in the presence of an acid catalyst such as sulphuric acid (Jang and Kamens 2001; Limbeck et al. 2003). Claeys et al. (2004a) found that biogenic aerosols in Amazonia contained large amounts of low-volatility polyols, which are likely derived from the reaction of isoprene and OH, accounting for ∼2 Tg of biogenic SOA/year. Another possible formation pathway is the reaction of isoprene with hydrogen peroxide under acidic conditions in the liquid phase (Claeys 2004b). These newly hypothesised chemical mechanisms indicate a growing understanding and the need for further research into biogenic SOA. For further information, we refer the reader to a recent review by Kanakidou et al. (2004) and to Chapter 8 of this volume.

3.5.5

Impacts of oxVOCs on atmospheric chemistry

Biogenic oxVOC emissions play a significant role in tropospheric chemistry because: (a) they can be carriers of reactive nitrogen and sequester nitrogen in the atmosphere, and (b) they can be photolysed, causing significant production of HOx radicals mainly in the upper atmosphere (Jaegle et al. 2001; Singh et al. 2000; Wennberg et al. 1998). Recent measurements in the remote Pacific indicate that the oxVOC species are about twice as abundant as the C2 –C8 hydrocarbons (Singh et al. 2004), indicating their potential magnitude and importance in understanding atmospheric chemistry. Other measurement campaigns over continental regions have found that oxVOC species can comprise a dominant fraction of the total VOC amount and reactivity, and that a large portion of the total oxVOC is likely from biogenic emissions (Goldan et al. 1995a, 1995b; Lamanna and Goldstein 2002; Millet et al. 2005; Riemer et al. 1998).

114


Besides oxVOCs from primary sources, the oxidation of other biogenic VOCs can create significant amounts of oxygenated products. For example, after biogenic compounds such as isoprene and monoterpenes are oxidised, they can form products such as formaldehyde, acetone and other carbonyl compounds. Globally, the oxidation of isoprene is thought to be the dominant source of formaldehyde. Additionally, secondary formation of acetone by the oxidation of isoprene or MBO can contribute significantly to the global acetone budget.

3.6 3.6.1

Sampling and measurement techniques Mixing ratio measurement methods

Mixing ratios of biogenic VOCs can be measured using a variety of sampling techniques, including sampling air into canisters, collecting air samples on solid adsorbents and using automated in-situ instrumentation. These methods are briefly reviewed here, and a thorough review of recently used methods can be found in Helmig (1999). Samples collected into canisters are generally brought back to a central laboratory facility and analysed following cryocondensation of VOCs and subsequent injection into a gas chromatograph equipped with a capillary column and one of several possible detectors. Typical detectors include the flame ionisation detector (FID), quadrupole mass spectrometer, and photoionisation detector (PID). Due to the poor recovery of some biogenic VOCs from canisters, particularly oxVOCs and many terpenoid compounds, techniques have been developed for collecting air samples directly onto carbon-based solid adsorbent cartridges that are later thermally desorbed for injection into a gas chromatography system. Special considerations are required for measurement of many higher-molecular-weight compounds such as sesquiterpenes, and recent progress has been made in creating methods appropriate to these compounds (Helmig et al. 2003). Automated in-situ instrumentation has been developed and deployed to field sites in order to create more temporally representative observational data sets of VOCs (e.g. Greenberg et al. 1994) and can be coupled to flux measurement systems to measure pairs of simultaneously collected samples from gradient or relaxed eddy accumulation (REA) systems as described below in the flux measurement section (e.g. Goldstein et al. 1995; Lamanna and Goldstein 2002; Schade and Goldstein 2001). Many recent advances have been made in measuring atmospheric VOCs (including biogenics), which are covered in detail in the last chapters of this book. Chapter 8 discusses gas chromatography-isotope ratio mass spectrometry, which is a technique used to measure the isotopic composition of VOC (recently reviewed by Goldstein and Shaw 2003). Chapter 9 of this volume discusses proton transfer reaction mass spectrometry (PTR-MS), which is a fast-response, high-sensitivity measurement with limited selectivity for individual compounds. This method is now widely used for measuring oxVOCs, biogenic VOC fluxes and biogenic VOC oxidation products. Chapter 10 discusses multidimensional gas chromatography, which has been explored but is not yet widely used for biogenic VOC measurements. This technique is able to separate a complex VOC mixture in atmospheric samples, and could be useful for the speciation of biogenic VOCs and their oxidation products.

Biogenic VOCs

3.6.2

115

Flux measurement methods

Several methods are available to measure the emissions of biogenic VOCs from vegetation. Leaf enclosure and branch enclosure are two flux methods described below to determine the emissions from a small, intact sample of plant matter. The next two flux methods, REA and eddy covariance (EC), are micrometeorological methods used to determine fluxes from ecosystems and generally represent fluxes over an area of 105 m2 or more (Cao and Hewitt 1999). The enclosure methods have the advantage of allowing us to understand the contribution of individual species to biogenic emissions and to manipulate the plant environment to determine response to variables such as light and temperature. The micrometeorological methods provide a broader picture of the hydrocarbon fluxes into and out of a whole undisturbed ecosystem, and can be deployed over longer timescales to observe the response of fluxes to changing environmental conditions.

3.6.3

Leaf and branch enclosure measurements

The most direct way to measure biogenic VOC emissions is with leaf and branch chambers. This technique is commonly used to determine emission factors for biogenic VOC emission modelling purposes. Early emissions measurements were taken via branch enclosures (Lamb et al. 1985; Zimmerman 1979), which envelope an entire branch into a Teflon or Tedlar enclosure. Ambient air is then pumped or pulled through the enclosure and an emission rate (ER, mass of emissions per mass of biomass per hour) is determined by the difference in the inlet and outlet air mixing ratios (cin and cout ), the flow rate of the air through the bag, and the biomass within the enclosure: ER =

flow (cout − cin ) mass

(3.8)

ER can also be determined on a per-area basis, by dividing by the leaf area within the enclosure instead of by the mass; however, biogenic VOC emission factors are more typically reported on a per-mass basis (Guenther et al. 1995). Measurements can also be taken at the leaf level, in which a leaf is placed inside a cuvette (chamber). Typically, cuvettes are constructed with the ability to control temperature and light levels to test emissions over a range of conditions and develop emission response curves for modelling purposes. Ambient air is pumped or pulled through the cuvette, and the emission rates are determined in the same manner as Equation (3.8). Past studies have shown that leaf cuvette ER measurements are typically higher than branch enclosure measurements (Guenther et al. 1996), and this has been attributed to light levels and shading within the canopy. When using branch or leaf enclosure techniques, care must be taken not to damage the plant, because that often induces biogenic VOC emissions.

3.6.4

Relaxed eddy accumulation

REA is a micrometeorological method to determine biogenic VOC fluxes in and out of ecosystems or landscapes on a larger scale. This method determines the biosphere–atmosphere flux (F ) of biogenic VOC based on separately collecting and analysing mixing ratios of

116


the target species in updrafts (cup ) and downdrafts (cdown ) of air: F = βσw (cup − cdown )

(3.9)

where β is an empirical coefficient that depends on the atmospheric stability, and σw is the standard deviation of the vertical wind speed. A sonic anemometer is used to determine vertical wind speed and direction, and air samples can be collected into canisters or cartridges for later analysis, or can be analysed by automated in-situ instrumentation. For a complete description of this and an example of its application for biogenic VOC flux measurements, see Bowling et al. (1998). This method is particularly effective for measuring ecosystemscale fluxes of compounds that are not amenable to current fast-response measurement methods. For compounds that can be measured with fast time response (better than 1 Hz), the EC method can be used.

3.6.5

Eddy Covariance (EC)

EC is a more direct micrometeorological method to measure biosphere–atmosphere fluxes of biogenic VOCs on the ecosystem scale. Fluxes (F ) are determined from the covariance of the biogenic VOC mixing ratio and the vertical wind speeds using the following: F = Na w c

(3.10) w

is deviation from the mean vertical where Na is the number density of air molecules, wind velocity, and c is the deviation from the mixing ratio of the biogenic VOC. This method requires that the mixing ratio of biogenic VOC be measured as fast or faster than the eddies carry the flux past the sensor or, in other words, faster than the vertical wind is changing directions. Because measurements for the EC method need to be instantaneous, this technique could not be employed for biogenic VOCs until recently. The first published flux measurements using the EC method biogenic VOCs utilised a fast chemiluminescence sensor to measure isoprene flux (Guenther and Hill 1998), and more recent studies have used PTR-MS for monoterpene (Lee et al. 2005; Rinne et al. 2002) and oxVOC fluxes (Karl et al. 2001a; Warneke et al. 2002).

3.7

Future directions

Biogenic VOC emissions are a dominant source of reactive organic gases in the atmosphere. In the 1980s and 1990s, the primary research focus was to quantify emissions and understand their role in atmospheric chemistry, with particular attention on contributions to formation of tropospheric ozone and to the oxVOC budget. However, as measurements of these compounds have progressed, categories of extremely reactive compounds are being found to be biogenic in origin (e.g. oxVOCs, VR-BVOCs). These highly reactive compounds form oxidation products in the gas phase and contribute to growth of secondary organic aerosols, which can play an important role in the interactions between chemistry and climate. The interaction between emission of very reactive compounds and their role in secondary aerosol production and growth are now a dominant focus in biogenic VOC research, and one that we believe will remain prominent in the decade to come.

Biogenic VOCs

117

References Affek, H.P. and Yakir, D. (2002) Protection by isoprene against singlet oxygen in leaves. Plant Physiology, 129: 269–77. Affek, H.P. and Yakir, D. (2003) Natural abundance carbon isotope composition of isoprene reflects incomplete coupling between isoprene synthesis and photosynthetic carbon flow. Plant Physiology, 131: 1727–36. Allwine, G., et al. (1992) As cited in Fehsenfeld et al. (1992). Andreae, M.O. and Crutzen, P.J. (1997) Atmospheric aerosols: Biogeochemical sources and role in atmospheric chemistry. Science, 276: 1052–8. Andronache, C., Chameides, W.L., Rodgers, M.O., et al. (1994) Vertical distribution of isoprene in the lower boundary layer of rural and urban southern United States. Journal of Geophysical Research, 99 (D8): 16989–99. Arey, J., Winer, A.M., Atkinson, R., et al. (1991) The emission of (z)-3-hexen-1-ol, (z)-3-hexenylacetate and other oxygenated hydrocarbons from agricultural plant species. Atmospheric Environment, 25 (5–6): 1063–75. Atkinson, R. and Arey, J. (1998) Atmospheric chemistry of biogenic organic compounds. Accounts of Chemical Research, 31: 574–83. Atkinson, R. and Arey, J. (2003) Gas-phase tropospheric chemistry of biogenic volatile organic compounds: A review. Atmospheric Environment, 37: S197–219. Atkinson, R., Hasegawa, D. and Aschmann, S.M. (1990) Rate constants for the gas-phase reactions of O3 with a series of terpenes. Journal of Geophysical Research, 97: 6065–73. Atkinson, R., Arey, J., Aschmann, S.M., et al. (1995) Rate constants for the gas-phase reactions of cis3-hexen-1-ol, cis-3-hexenylacetate, trans-2-hexenal, and linalool with OH and NO3 radicals and O3 at 296+/−2K, and OH radical formation yields from the O3 reactions. International Journal of Chemical Kinetics, 27: 941–55. Atkinson, R., Baulch, D.L., Cox, R.A., et al. (1999) Evaluated kinetic and photochemical data for atmospheric chemistry, organic species. Journal of Physical and Chemical Referenece Data, 28: 191–393. Baker, B.M., Guenther, A., Greenberg, J., et al. (1999) Canopy fluxes of 2-methyl-3-buten-2-ol over a ponderosa pine forest by relaxed eddy accumulation: Field data and model comparison. Journal of Geophysical Research, 104 (D21): 26107–44. Baker, B.M., Guenther, A. and Greenberg, J. (2001) Canopy level fluxes of 2-methyl-3-buten-2-ol, acetone, and methanol by a portable relaxed eddy accumulation system. Environmental Science and Technology, 35 (9): 1701–8. Biesenthal, T.A., Bottenheim, J.W., Shepson, P.B., et al. (1998) The chemistry of biogenic hydrocarbons at a rural site in eastern Canada. Journal of Geophysical Research, 103 (D19): 25487–98. Bonsang, B., Poll, C. and Lambert, G. (1992) Evidence of marine production of isoprene. Geophysical Research Letter, 24 (21): 1129–32. Bowling, D.R., Turnipseed, A.A., Delany, A.C., et al. (1998) The use of relaxed eddy accumulation to measure biosphere-atmosphere exchange of isoprene and other biological trace gases. Oecologia, 116: 306–15. Box, E. (1981) Data base of the international biological program and other sources. In: J. Bufalini and R. Arnts (Eds) Atmospheric Biogenic Hydrocarbons. Ann Arbor Science Publishers, Ann Arbor, MI, pp. 121–48. Broadgate, W.J., Malin, G., Kupper, F.C., et al. (2004) Isoprene and other non-methane hydrocarbons from seaweeds: A source of reactive hydrocarbons to the atmosphere. Marine Chemistry, 88 (1–2): 61–73. Calvert, J.G., Atkinson, R., Kerr, J.A., et al. (2000) The Mechanisms of Atmospheric Oxidation of Aromatic Hydrocarbons. New York, NY: Oxford University Press.

118


Cao, X.-L. and Hewitt, C.N. (1999) The sampling and analysis of volatile organic compounds in the atmosphere. In: C.N. Hewitt (Ed.) Reactive Hydrocarbons in the Atmosphere. Academic Press, San Diego, CA, pp. 119–57. Carlslaw, N., Bell, N., Lewis, A.C., et al. (2000) A detailed case study of isoprene chemistry during the EASE96 Mace Head campaign. Atmospheric Environment, 34: 2827–36. Chameides, W.L., Lindsay, R.W., Richardson, J., et al. (1988) The role of biogenic hydrocarbons in urban photochemical smog: Atlanta as a case study. Science, 241: 1473–5. Ciccioli, P., Brancaleoni, E., Frattoni, M., et al. (1999) Emission of reactive terpene compounds from orange orchards and their removal by within-canopy processes. Journal of Geophysical Research, 104 (D7): 8077–94. Claeys, M., Graham, B., Vas, G., et al. (2004a) Formation of secondary organic aerosols through photooxidation of isoprene. Science, 303: 1173–6. Claeys, M., Wang, W., Ion, A.C., et al. (2004b) Formation of secondary organic aerosols from isoprene and its gas-phase oxidation products through reaction with hydrogen peroxide. Atmospheric Environment, 38: 4093–8. Constable, J.V.H., Litvak, M.E., Greenberg, J., et al. (1999) Monoterpene emission from coniferous trees in response to elevated CO2 concentrations and climate warming. Global Change Biology, 5: 255–67. Corchnoy, S.B. and Atkinson, R. (1990) Kinetics of the gas-phase reactions of OH and NO3 radicals with 2-carene, 1,8-cineole, p-cymene and terpinolene. Environmental Science and Technology, 24: 1497–1502. Council, N.R. (1991) Rethinking the Ozone Problem in Urban and Regional Air Pollution. National Academy Press, Washington, DC. DeGouw, J.A., Howard, C.J., Custer, T.G., et al. (2000) Proton-transfer chemical-ionization mass spectrometry allows real-time analysis of volatile organic compounds released from cutting and drying of crops. Environmental Science and Technology, 34 (12): 2640–8. Delwiche, C.F. and Sharkey, T.D. (1993) Rapid appearance of 13 C in biogenic isoprene when 13 C is fed to intact leaves. Plant, Cell and Environmental, 16: 587–91. Dement, W.A., Tyson, B.J. and Mooney, H.A. (1975) Mechanism of monoterpene volatilization in salvia mellifera. Phytochemistry, 14: 1555–7. Di Carlo, P., Brune, W.H., Martinez, M., et al. (2004) Missing OH reactivity in a forest: Evidence for unknown reactive biogenic VOCs. Science, 304: 722–5. Dignon, J. and Logan, J.A. (1990) Biogenic emissions of isoprene: A global inventory. EOS Transaction of the American Geophysical Union, 71: 1260. Donahue, N.M. and Prinn, R.G. (1990) Nonmethane hydrocarbon chemistry in the remote marine boundary layer. Journal of Geophysical Research, 95 (D11): 18387–411. Donahue, N.M. and Prinn, R.G. (1993) In-situ nonmethane hydrocarbon measurements on SAGA-3. Journal of Geophysical Research, 98 (D9): 16915–32. Dreyfus, G.B., Schade, G.W. and Goldstein, A.H. (2002) Observational constraints on the production of isoprene oxidation to ozone production on the western slope of the Sierra Nevada, California. Journal of Geophysical Research, 107 (D19): 4365. Eisenreich, W., Rohdich, F. and Bacher, A. (2001) Deoxyxylulose phosphate pathway to terpenoids. Trends in Plant Science, 6 (2): 78–84. Evans, R., Tingey, D. and Gumpertz, M. (1985) Interspecies variation in terpenoid emissions from Englemann and Sitka Spruce seedings. Forest Science, 31: 132–42. Fall, R. (2003) Abundant oxygenates in the atmosphere: A biochemical perspective. Chemical Reviews, 103: 2941–51. Fall, R. and Monson, R.K. (1992) Isoprene emission rate and intercellular isoprene concentration as influenced by stomatal distribution and conductance. Plant Physiology, 100: 987–992.

Biogenic VOCs

119

Fall, R., Karl, T., Hansel, A., et al. (1999) Volatile organic compounds emitted after leaf wounding: On-line analysis by proton-transfer-reaction mass spectrometry. Journal of Geophysical Research, 104 (D13): 14963–15974. Fehsenfeld, F., Calvert, J., Fall, R., et al. (1992) Emissions of volatile organic compounds from vegetation and the implications for atmospheric chemistry. Global Biogeochemical Cycles, 6 (4): 389–430. Finlayson-Pitts, B.J. and Pitts Jr., J.N. (2000) Chemistry of the Upper and Lower Atmosphere: Theory, Experiments and Applications. Academic Press, San Diego, CA. Fisher, A.J., Baker, B.M., Greenberg, J., et al. (2000) Enzymatic synthesis of methylbutenol from dimethylallyl diphosphate in needles of Pinus sabiniana. Archives of Biochemistry and Biophysics, 383: 128–34. Fuentes, J.D., Wang, D. and Gu, L. (1999) Seasonal variations in isoprene emission from a boreal aspen forest. Journal of Applied Meteorology, 38: 855–69. Fuentes, J.D., Lerdau, M., Atkinson, R., et al. (2000) Biogenic hydrocarbons in the atmospheric boundary layer: A review. Bulletin of the American Meteorological Society, 81 (7): 1537–75. Fukui, Y. and Doskey, P.V. (2000) Identification of nonmethane organic compound emissions from grassland vegetation. Atmospheric Environment, 34 (18): 2947–56. Funk, J.L. (2003) Diurnal variation in the basal emission rate of isoprene. Ecological Application, 13 (1): 269–78. Gabriel, R., et al. (1999) Factors controlling the emissions of volatile organic acids from leaves of Quercus ilex L. (holm oak). Atmospheric Environment, 33: 1347–55. Galbally, I.E. and Kirstine, W. (2002) The production of methanol by flowering plants and the global cycle of methanol. Journal of Atmospheric Chemistry, 43: 195–229. Geron, C., Guenther, A.B. and Pierce, T.E. (1994) An improved model for estimating emissions of volatile organic compounds from forests in the eastern United States. Journal of Geophysical Research, 99 (D6): 12773–91. Geron, C., Guenther, A.B., Sharkey, T.D., et al. (2000a) Temporal variability in the basal isoprene emission factor. Tree Physiology, 20: 799–805. Geron, C., Rasmussen, R., Arnts, R., et al. (2000b) A review and synthesis of monoterpene speciation from forests in the United States. Atmospheric Environment, 34: 1761–81. Goldan, P.D., Kuster, W.C., Fehsenfeld, F., et al. (1993) The observation of a C5 alcohol emission in a North American pine forest. Geophysical Research Letter, 20: 1039–42. Goldan, P.D., Kuster, W.C., Fehsenfeld, F., et al. (1995a) Hydrocarbon measurements in the southeastern United States: The Rural Oxidants in the Southern Environment (ROSE) program 1990. Journal of Geophysical Research, 100 (D12): 25945–63. Goldan, P.D., Trainer, M., Kuster, W.C., et al. (1995b) Measurements of hydrocarbons, oxygenated hydrocarbons, carbon monoxide and nitrogen oxides in an urban basin in Colorado: Implications for emissions inventories. Journal of Geophysical Research, 100 (D11): 22771–83. Goldstein, A.H. and Schade, G.W. (2000) Quanitifying biogenic and anthropogenic contributions to acetone mixing ratios in a rural environment. Atmospheric Environment, 34 (29–30): 4997–5006. Goldstein, A.H. and Shaw, S.L. (2003) Isotopes of volatile organic compounds: An emerging approach for studying atmospheric budgets and chemistry. Chemical Reviews, 103: 5025–48. Goldstein, A.H., Daube, B.C., Munger, J.W., et al. (1995) Automated in-situ monitoring of atmospheric non-methane hydrocarbon concentrations and gradients. Journal of Atmospheric Chemistry, 21: 43–59. Goldstein, A.H., Goulden, M.L., Munger, J.W., et al. (1998) Seasonal course of isoprene emissions from a midlatitude deciduous forest. Journal of Geophysical Research, 103 (D23): 31045–56. Goldstein, A.H., McKay, M., Kurpius, M.R., et al. (2004) Forest thinning experiment confirms ozone deposition to forest canopy is dominated by reaction with biogenic VOCs. Geophysical Research Letter, 31 (22): L22106.

120


Graus, M., Schnitzler, J.-P., Hansel, A., et al. (2004) Transient release of oxygenated volatile organic compounds during light-dark transitions in grey poplar leaves. Plant Physiology, 135: 1967–75. Gray, D.W., Lerdau, M., Goldstein, A.H. (2004) Influences of temperature history, water stress, and needle age on methyl butenol emissions. Ecology, 84 (3): 765–76. Greenberg, J., Lee, B., Helmig, D., et al. (1994) Fully-automated gas chromatograph-flame ionization detector system for the in situ determination of atmospheric non-methane hydrocarbons at low parts per trillion concentration. Journal of Chromatography, 676: 389–98. Griffin, R.J., Cocker, D.R.I., Flagan, R.C., et al. (1999a) Organic aerosol formation from the oxidation of biogenic hydrocarbons. Journal of Geophysical Research, 104 (D3): 3555–67. Griffin, R.J., Cocker, D.R.I., Seinfeld, J.H., et al. (1999b) Estimate of global atmospheric organic aerosol from oxidation of biogenic hydrocarbons. Geophysical Research Letter, 26 (17): 2721–4. Grosjean, E. and Grosjean, D. (1994) Rate constants for the gas-phase reactions of ozone with unsaturated aliphatic alcohols. International Journal of Chemical Kinetics, 26: 1185–91. Guenther, A., Zimmerman, P. and Wildermuth, M. (1994) Natural volatile organic-compound emission rate estimates for United States woodland landscapes. Atmospheric Environment, 28 (6): 1197–210. Guenther, A., Geron, C., Pierce, T.E., et al. (2000) Natural emissions of non-methane volatile organic compounds, carbon monoxide and oxides of nitrogen from North America. Atmospheric Environment, 34 (12–14): 2205–30. Guenther, A.B. and Hill, A.J. (1998) Eddy covariance measurement of isoprene fluxes. Journal of Geophysical Research, 103 (D11): 13145–52. Guenther, A.B., Monson, R.K. and Fall, R. (1991) Isoprene and monoterpene emission rate variability: Observations with eucalyptus and emission rate algorithm development. Journal of Geophysical Research, 96: 10799–808. Guenther, A.B., Zimmerman, P., Harley, P.C., et al. (1993) Isoprene and monoterpene emission rate variability: Model evaluations and sensitivity analyses. Journal of Geophysical Research, 98: 12609–17. Guenther, A.B., Hewitt, C.N., Erickson, D., et al. (1995) A global model of natural volatile organic compound emissions. Journal of Geophysical Research, 100 (D5): 8873–92. Guenther, A.B., Zimmerman, P., Klinger, L., et al. (1996) Estimate of regional natural volatile organic compound fluxes from enclosure and ambient measurement. Journal of Geophysical Research, 101 (D1): 1345–9. Guenther, A.B., Archer, S., Greenberg, J., et al. (1999a) Biogenic hydrocarbon emissions and landcover/climate change in a subtropical savanna. Physics, Chemistry and Earth (B), 24 (6): 659–67. Guenther, A., Baugh, B., Brasseur, G., et al. (1999b) Isoprene emission estimates and uncertainties for the Central African EXPRESSO study domain. Journal of Geophysical Research, 104 (D23): 30625–39. Hakola, H., Laurila, T., Rinne, J., et al. (2000) The ambient concentrations of biogenic hydrocarbons at a northern European boreal site. Atmospheric Environment, 34: 4971–82. Hallquist, M., Wangberg, I., Ljungstrom, E., et al. (1999) Aerosol and product yields from NO3 radical-initiated oxidation of selected monoterpenes. Environmental Science and Technology, 33: 553–9. Hansen, U., Van Eijk, J., Bertin, N., et al. (1997) Biogenic emissions and CO2 gas exchange investigated on four Mediterranean shrubs. Atmospheric Environment, 31 (SI): 157–66. Hanson, D.T., Swanson, S., Graham, L.E., et al. (1999) Evolutionary significance of isoprene from mosses. Americal Journal of Botony, 86: 634–9. Harley, P.C., Guenther, A. and Zimmerman, P. (1996) Effects of light, temperature and canopy position on net photosynthesis and isoprene emission from sweetgum (Liquidambar styraciflua) leaves. Tree Physiology, 16: 25–32.

Biogenic VOCs

121

Harley, P.C., Fridd-Stroud, V., Greenberg, J., et al. (1998) Emission of 2-methyl-3-buten-2-ol by pines: A potentially large natural source of reactive carbon to the atmosphere. Journal of Geophysical Research, 103 (D19): 25479–86. Harley, P.C., Monson, R.K. and Lerdau, M. (1999) Ecological and evolutionary aspects of isoprene emission from plants. Oecologia, 118 (2): 109–23. Harley, P.C., Vasconcellos, P., Vierling, L., et al. (2004) Variation in potential for isoprene emissions among Neotropical forest sites. Global Change Biology, 10: 630–50. He, N.P., Han, X.G., Sun, W., et al. (2004) Biogenic VOCs emission inventory development of temperate grassland vegetation in Xilin River Basin, Inner Mongolia, China. Journal of Environmental Science (China), 16 (6): 1024–32. Heikes, B.G., Chang, W., Pilson, M.E.Q., et al. (2002) Atmospheric methanol budget and ocean implication. Global Biogeochemical Cycles, 16 (4): 1133. Helmig, D. (1999) Review: Gas chromotography air analysis. Journal of Chromatography, 843: 129–46. Helmig, D., Greenberg, J., Guenther, A.B., et al. (1998a) Volatile organic compounds and isoprene oxidation products at a temperate deciduous forest site. Journal of Geophysical Research, 103 (D17): 22397–414. Helmig, D., Balseley, B., Davis, K., et al. (1998b) Vertical profiling and determination of landscape fluxes of biogenic nonmethane hydrocarbons with the planetary boundary layer in the Peruvian Amazon. Journal of Geophysical Research, 103 (D19): 25519–32. Helmig, D., Revermann, T., Pollmann, J., et al. (2003) Calibration system and analytical considerations for quantitative sesquiterpene measurements in air. Journal of Chromatography, 1002: 193–211. Hirsch, A.I., Munger, J.W., Jacob, D.J., et al. (1996) Seasonal variation of the ozone production efficiency per unit NOx at Harvard Forest, Massachusetts. Journal of Geophysical Research, 101 (D7): 12659–66. Hoffmann, T., Odum, J.R., Bowman, F., et al. (1997) Formation of organic aerosols from the oxidation of biogenic hydrocarbons. Journal of Atmospheric Chemistry, 26 (2): 189–222. Holopainen, J.K. (2004) Multiple functions of inducible plant volatiles. Trends in Plant Science, 9 (11): 529–33. Holzinger, R., Sandoval-Soto, L., Rottenberger, S., et al. (2000) Emissions of volatile organic compounds from Quercus ilex L. measured by proton transfer reaction mass spectrometry under different environmental conditions. Journal of Geophysical Research, 105 (D16): 20573–9. Holzinger, R., Lee, A., Paw U, K.T., et al. (2004) Observations of oxidation products above a forest imply biogenic emissions of very reactive compounds. Atmospheric Chemistry and Physics Discussion, 4: 5345–65. Horowitz, L.W., Liang, J., Gardner, G.M., et al. (1998) Export of reactive nitrogen from North American during summertime: Sensitivity to hydrocarbon chemistry. Journal of Geophysical Research, 103 (D11): 13451–76. Houweling, S., Dentener, F.J. and Lelieveld, J. (1998) The impact of nonmethane hydrocarbon compounds on tropospheric chemistry. Journal of Geophysical Research, 103 (D9): 10673–96. Iijima, Y., Davidovich-Rikanati, R., Fridman, E., et al. (2004) The biochemical and molecular basis for the divergent patterns in the biosynthesis of terpenes and phenylpropenes in the peltate glands of three cultivars of basil. Plant Physiology, 136: 3724–36. Jacob, D.J., Logan, J.A., Gardner, G.M., et al. (1993) Factors regulating ozone over the United States and its export to the global atmosphere. Journal of Geophysical Research, 98 (D8): 14817–26. Jacob, D.J., Field, B.D., Jin, E.M., et al.(2002) Atmospheric budget of acetone. Journal of Geophysical Research, 107 (D10): 4100. Jaegle, L., Jacob, D.J., Brune, W.H., et al. (2001) Chemistry of HOx radicals in the upper troposphere. Atmospheric Environment, 35: 469–89.

122


Jang, M. and Kamens, R. (2001) Atmospheric secondary organic aerosol formation by heterogeneous reaction of aldehydes in the presence of a sulfuric acid aerosol catalyst. Environmental Science and Technology, 35: 4758–66. Kamens, R., Jeffries, H.E., Gery, M.W., et al. (1981) The impact of α-pinene on urban smog formation: An outdoor smog chamber study. Atmospheric Environment, 15 (6): 969–81. Kamens, R., Jang, M., Chien, C.-J., et al. (1999) Aerosol formation from the reaction of α-pinene and ozone using a gas-phase kinetics-aerosol partitioning model. Environmental Science and Technology, 33: 1430–8. Kanakidou, M., Seinfeld, J.H., Pandis, S.N., et al. (2004) Organic aerosol and global climate modelling: A review. Atmospheric Chemistry And Physics Discussion, 4: 5855–6024. Karl, T., Guenther, A.B., Jordan, A., et al. (2001a) Eddy covariance measurement of biogenic oxygenated VOC emissions from hay harvesting. Atmospheric Environment, 35 (3): 491–5. Karl, T., Fall, R., Jordan, A., et al. (2001b) On-line analysis of reactive VOCs from urban lawn mowing. Environmental Science and Technology, 35 (14): 2926–31. Karl, T., Curtis, A.J., Rosenstiel, T.N., et al. (2002) Transient releases of acetaldehyde from tree leaves – Products of a pyruvate overflow mechanism. Plant, Cell and Environment, 25: 1121–31. Karlik, J.F. and Winer, A.M. (2001) Measured isoprene emission rates of plants in California landscapes: Comparison to estimates from taxonomic relationships. Atmospheric Environment, 36 (6): 1123–31. Kesselmeier, J., Bode, K., Hofmann, U., et al. (1997) Emission of short chained organic acids, aldehydes and monoterpenes from Quercus ilex L. and Pinus pinea L. in relation to physiological activities, carbon budget and emission algorithms. Atmospheric Environment, 31 (S1): 119–33. Kesselmeier, J. and Staudt, M. (1999) Biogenic volatile organic compounds (VOC): An overview on emission, physiology and ecology. Journal of Atmospheric Chemistry, 33: 23–88. Kirstine, W., Galbally, I.E., Ye, Y., et al. (1998) Emissions of volatile organic compounds (primarily oxygenated species) from pasture. Journal of Geophysical Research, 103 (D9): 10605–19. Klinger, L., Li, Q.-J., Guenther, A., et al. (2002) Assessment of volatile organic compound emissions from ecosystems of China. Journal of Geophysical Research, 107 (D21): 4603. Koenig, G., Brunda, M., Puxbau, H., et al. (1995) Relative contribution of oxygenated hydrocarbons to the total biogenic VOC emissions of selected mid-European agricultural and natural plant species. Atmospheric Environment, 29 (8): 861–74. Kreuzwieser, J., Scheerer, U. and Rennenberg, H. (1999) Metabolic origin of acetaldehyde emitted by poplar (Populus tremula × P. alba) trees. Journal of Experimental Botony, 50 (335): 757–65. Kreuzwieser, J., Graus, M., Wisthaler, A., et al. (2002) Xylem-transported glucose as additional carbon source for leaf formation in Quercus robur. New Phytologist, 156: 171–8. Kuhn, U., Rottenberger, S., Biesenthal, T.A., et al. (2004) Seasonal differences in isoprene and light-dependent monoterpene emission by Amazonian tree species. Global Change Biology, 10: 663–82. Kurpius, M.R. and Goldstein, A.H. (2003) Gas-phase chemistry dominates O3 loss to forest, implying a source of aerosols and hydroxyl radicals to the atmosphere. Geophysical Research Letter, 30: 1371. −C10 volatile organic comLamanna, M.S. and Goldstein, A.H. (2002) In situ measurements of C2 − pounds above a Sierra Nevada ponderosa pine plantation. Journal of Geophysical Research, 104 (D7): 21247–62. Lamb, B., Westberg, H., Allwine, G., et al. (1985) Biogenic hydrocarbon emissions from deciduous and coniferous trees in the United States. Journal of Geophysical Research, 90 (D1): 2380–90. Lamb, B., Guenther, A., Gay, D., et al. (1987) A national inventory of biogenic hydrocarbon emissions. Atmospheric Environment, 21 (8): 1695–1705. Lee, A., Schade, G.W., Holzinger, R., et al. (2005) A comparison of new measurements of total monoterpene flux with improved measurements of speciated monoterpene flux. Atmospheric Chemistry and Physics Discussion, 5: 505–13.

Biogenic VOCs

123

Lehning, A., Zimmer, W., Zimmer, I., et al. (2001) Modeling of annual variations in oak (Quercus robur L.) isoprene synthase activity to predict isoprene emission rates. Journal of Geophysical Research, 106 (D3): 3157–66. Lerdau, M., Guenther, A. and Monson, R.K. (1997) Plant production and emission of volatile organic compounds. Bioscience, 47 (6): 373–83. Lerdau, M. and Throop, H.L. (1999) Isoprene emission and photosynthesis in a tropical forest canopy: Implications for model development. Ecological Application, 9 (4): 1109–17. Lichtenthaler, H.K. (1999) The 1-deoxy-D-xylulose-5-phosphate pathway of isoprenoid biosynthesis in plants. Annual Review of Plant Physiology, 50: 47–65. Lichtenthaler, H.K., Schwender, J., Disch, A., et al. (1997) Biosynthesis of isoprenoids in higher plant chloroplasts proceeds via a mevalonate-independent pathway. FEBS Letter, 400: 271–4. Lieth, H. (1975) Modeling the primary productivity of the world. In: H. Lieth and R. Whittaker (Eds) Primary Productivity of the Biosphere. Springer-Verlag: NY, pp. 237–67. Limbeck, A., Kulmala, M. and Puxbaum, H. (2003) Secondary organic aerosol formation in the atmosphere via heterogeneous reaction of gaseous isoprene on acidic particles. Geophysical Research Letter, 30 (19): 1996. Litvak, M.E. and Monson, R.K. (1998) Patterns of induced and constitutive monoterpene production in conifer needles in relation to insect herbivory. Oecologia, 114: 531–40. Litvak, M.E., Loreto, F., Harley, P.C., et al. (1996) The response of isoprene emission rate and photosynthetic rate to photon flux and nitrogen supply in aspen and white oak trees. Plant, Cell and Environment, 19: 549–59. Litvak, M.E., Madronich, S. and Monson, R.K. (1999) Herbivore-induced monoterpene emissions from coniferous forests: Potential impact on local tropospheric chemistry. Ecological Application, 9 (4): 1147–59. Liu, S., Trainer, M., Fehsenfeld, F., et al. (1987) Ozone production in the rural troposphere and the implications for regional and global ozone distributions. Journal of Geophysical Research, 92 (D4): 4191–207. Logan, B.A. and Monson, R.K. (1999) Thermotolerance of leaf discs from four isoprene-emitting species is not enhanced by exposure to exogenous isoprene. Plant Physiology, 120: 821–5. Logan, B.A., Monson, R.K. and Potosnak, M.J. (2000) Biochemistry and physiology of foliar isoprene production. Trends in Plant Science, 5: 477–81. Loreto, F. and Sharkey, T.D. (1990) A gas exchange study of photosynthesis and isoprene emission in red oak (Quercus rubra L.). Planta, 182: 523–31. Loreto, F., Ciccioli, P., Brancaleoni, E., et al. (1996a) Different sources of reduced carbon contribute to form three classes of terpenoid emitted by Quercus ilex L. leaves. Proceedings of the National Academy of Science USA, 93: 9966–9. Loreto, F., Ciccioli, P., Cecinato, A., et al. (1996b) Influence of environmental factors and air composition on the emission of α-pinene from Quercus ilex leavs. Plant Physiology, 110: 267–75. Loreto, F., Nascetti, P., Graverini, A., et al. (2000) Emission and content of monoterpenes in intact and wounded needles of the Mediterranean pine Pinus pinea. Functional Ecology, 14: 589–95. Loreto, F., Fischbach, R.J., Schnitzler, J.-P., et al. (2001a) Monoterpene emission and monoterpene synthase activities in the Mediterranean evergreen oak Quercus ilex L. grown at elevated CO2 concentrations. Global Change Biology, 7: 709–17. Loreto, F., Mannozzi, M., Maris, C., et al. (2001b) Ozone quenching properties of isoprene and its antioxidant role in leaves. Plant Physiology, 126: 993–1000. Loreto, F., Pinelli, P., Brancaleoni, E., et al. (2004) 13 C labeling reveals chloroplastic and extrachloroplastic pools of dimethylallyl pyrophosphate and their contribution to isoprene formation. Plant Physiology, 135: 1903–7. MacDonald, R. and Fall, R. (1993) Detection of substantial emissions of methanol from plants to the atmosphere. Atmospheric Environment, 27A: 1709–14.

124


Makar, P.A., Fuentes, J.D., Wang, D., et al. (1999) Chemical processing of biogenic hydrocarbons within and above a temperature deciduous forest. Journal of Geophysical Research, 104 (D3): 3581–603. Martin, M.J., Stirling, C.M., Humphries, S.W., et al. (2000) A process-based model to predict the effects of climatic change on leaf isoprene emission rates. Ecological Modelling, 131: 161–74. Mauzerall, D.L., Jacob, D.J., Fan, S.M., et al. (1996) Origin of tropospheric ozone at remote high northern latitudes in summer. Journal of Geophysical Research, 101 (D2): 4175–88. Meyrahn, H., Pauly, J., Schneider, W., et al. (1986) Quantum yields for the photodissociation of acetone in air and an estimate for the life time of acetone in the lower troposphere. Journal of Atmospheric Chemistry, 4: 277–91. Millet, D.B., Donahue, N.M., Pandis, S.N., et al. (2005) Atmospheric VOC measurements during the Pittsburgh Air Quality Study: Results, interpretation and quantification of primary and secondary contributions. Journal of Geophysical Research, 110: D07S07. Monson, R.K. and Fall, R. (1989) Isoprene emission from aspen leaves. Plant Physiology, 90: 267–74. Monson, R.K. and Holland, E.A. (2001) Biospheric trace gas fluxes and their control over tropospheric chemistry. Annual Review of Ecology Systematics, 32: 547–76. Monson, R.K., Jaeger, C.H., Adams, W.W.I., et al. (1992) Relationships among isoprene emission rate, photosynthesis, and isoprene synthase activity as influenced by temperature. Plant Physiology, 98: 1175–80. Monson, R.K., Harley, P.C., Litvak, M.E., et al. (1994) Environmental and developmental controls over the seasonal pattern of isoprene emissions from aspen leaves. Oecologia, 99: 260–70. Mueller, J.-F. (1992) Geographical distribution and seasonal variation of surface emissions and deposition velocities of atmospheric trace gases. Journal of Geophysical Research, 97: 3787–804. Nemecek-Marshall, M., MacDonald, R., Franzen, J.J., et al. (1995) Methanol emissions from leaves (enzymatic detection of gas-phase methanol and relation of methanol fluxes to stomatal conductance and leaf development). Plant Physiology, 108: 1359–68. Niinemets, U. and Reichstein, M. (2002) A model analysis of the effects of nonspecific monoterpenoid storage in leaf tissues on emission kinetics and composition in Mediterranean sclerophyllous Quercus species. Global Biogeochemical Cycles, 16 (4): 1110. Niinemets, U. and Reichstein, M. (2003a) Controls on the emission of plant volatiles through stomata: Differential sensitivity of emission rates to stomatal closure explained. Journal of Geophysical Research, 108 (D7): 4208. Niinemets, U. and Reichstein, M. (2003b) Controls on the emission of plant volatiles through stomata: A sensitivity analysis. Journal of Geophysical Research, 108 (D7): 4211. Niinemets, U., Tenhunen, J.D., Harley, P.C., et al. (1999) A model of isoprene emission based on energetic requirements for isoprene synthesis and leaf photosynthetic properties for Liquidambar and Quercus. Plant, Cell and Environment, 22 (11): 1319–35. Otter, L., Guenther, A., Wiedinmyer, C., et al. (2003) Spatial and temporal variations in biogenic volatile organic compound emissions for Africa south of the equator. Journal of Geophysical Research, 108 (D13): 8505. Palmer, P.I. and Shaw, S.L. (2005) Quantifying global marine isoprene fluxes using MODIS chlorophyll observations. Geophysical Research Letter, 32: L09805, doi:10.1029/2005GL022592. Palmer, P.I., Jacob, D.J., Fiore, A.M., et al. (2003) Mapping isoprene emissions over North American using formaldehyde column observations from space. Journal of Geophysical Research, 108 (D6): 4180. Pandis, S.N., Paulson, S.E., Seinfeld, J.H., et al. (1991) Aerosol formation in the photooxidation of isoprene and β-pinene. Atmospheric Environment, 25A (5/6): 997–1008. Papagni, C., Arey, J. and Atkinson, R. (2001) Rate constants for the gas-phase reactions of OH radicals with a series of unsaturated alcohols. International Journal of Chemical Kinetics, 33: 142–7. Paulson, S.E. and Orlando, J.J. (1996) The reactions of ozone with alkenes: An important source of HOx in the boundary layer. Geophysical Research Letter, 23 (25): 3727–30.

Biogenic VOCs

125

Paulson, S.E., Chung, M.Y. and Hasson, A.S. (1999) OH radical formation from the gas-phase reaction of ozone with terminal alkenes and the relationship between structure and mechanism. Journal of Physical Chemistry A, 103 (41): 8125–38. Pegoraro, E., Rey, A., Bobich, E.G., et al. (2004) Effect of elevated CO2 concentration and vapour pressure deficit on isoprene emission from leaves of Populus deltoides during drought. Functional Plant Biology, 31: 1137–47. Petron, G., Harley, P.C., Greenberg, J., et al. (2001) Seasonal temperature variations influence isoprene emissions. Geophysical Research Letter, 28: 1707–10. Pierce, T.E. and Waldruff, P. (1991) PC-BEIS: A personal computer version of the biogenic emissions inventory system. Journal of Air and Waste Management Association, 41: 937–41. Pierce, T.E., Geron, C., Bender, L., et al. (1998) Influence of increased isoprene emissions on regional ozone modeling. Journal of Geophysical Research, 103 (D19): 25611–29. Plass-Dulmer, C., Koppman, R., Ratte, M., et al. (1995) Light nonmethane hydrocarbons in seawater. Global Biogeochemical Cycles, 9: 79–100. Poisson, N., Kanakidou, M. and Crutzen, P.J. (2000) Impact of non-methane hydrocarbons on tropospheric chemistry and the oxidizing power of the global troposphere: 3-Dimensional modeling results. Journal of Atmospheric Chemistry, 36: 157–230. Potter, C.S., Alexander, S.E., Coughlan, J.C., et al. (2001) Modeling biogenic emissions of isoprene: Exploration of model drivers, climate control algorithms, and use of global satellite observations. Atmospheric Environment, 35: 6151–65. Potter, C., Klooster, S., Bubenheim, D., et al. (2003) Modeling terrestrial biogenic sources of oxygenated organic emissions. Earth Interactions, 7 (7): 1–15. Rapparini, F., Baraldi, R., Miglietta, F., et al. (2004) Isoprenoid emission in trees of Quercus pubescens and Quercus ilex with lifetime exposure to naturally high CO2 environment. Plant, Cell and Environment, 27: 381–91. Rasmussen, R. (1970) Isoprene: Identified as a forest-type emission to the atmosphere. Environmental Science and Technology, 4: 667–71. Rasmussen, R.A. and Khalil, M. (1988) Isoprene over the Amazon basin. Journal of Geophysical Research, 93 (D2): 1417–21. Rasmussen, R.A. (1972) What do hydrocarbons from trees contribute to air pollution? Journal of the Air Pollution Control Association, 22: 537–43. Rasmussen, R.A. and Went, F.W. (1965) Volatile organic material of plant origin in the atmosphere. Proceedings of the National Academy of Science USA, 53: 215–20. Reissell, A., Arey, J. and Atkinson, R. (2001) Atmospheric chemistry of camphor. International Journal of Chemical Kinetics, 33: 56–63. Riemer, D., Pos, W., Milne, P., et al. (1998) Observations of nonmethane hydrocarbons and oxygenated volatile organic compounds at a rural site in the southeastern United States. Journal of Geophysical Research, 103 (D21): 28111–28. Rinne, H.J., Guenther, A., Greenberg, J., et al. (2002) Isoprene and monoterpene fluxes measured above Amazonian rainforest and their dependence on light and temperature. Atmospheric Environment, 36 (14): 2421–6. Robinson, E. and Robbins, R. (1968) Sources, Abundance, and Fate of Gaseous Atmospheric Pollutants. Stanford Research Institute, Menlo Park, CA. Final Report of Project PR-6755. Rosenstiel, T.N., Fisher, A.J., Fall, R., et al. (2002) Differential accumulation of dimethylallyl diphosphate in leaves and needles of isoprene- and methylbutenol-emitting and non-emitting species. Plant Physiology, 129: 1276–84. Rosenstiel, T.N., Potosnak, M.J., Griffin, K.L., et al. (2003) Increased CO2 uncouples growth from isoprene emission in an agriforest ecosystem. Nature, 421: 256–9. Rosenstiel, T.N., Ebbets, A.L., Khatri, W.C., et al. (2004) Induction of a poplar leaf nitrate reductase: A test of extrachloroplastic control of isoprene emission rate. Plant Biology, 6: 12–21.

126


Rudich, Y., Talukdar, R.K., Fox, R.W., et al. (1996) Rate coefficients for reactions of NO3 with a few olefins and oxygenated olefins. Journal of Physical Chemistry, 100: 5374–81. Sanadze, G.A. (1957) Emission of organic matters by leaves of Robinia psudaeacia L. Soobsh. Academy of Nauk Global Sustainable Systems Research, 19: 83–6. Sanadze, G.A. (2004) Biogenic isoprene (a review). Russian Journal of Plant Physiology, 51 (6): 810–24. Sanadze, G.A., Dzhaiani, G.I. and Tevzadze, I.M. (1972) Incorporation into the isoprene molecule of carbon from 13 CO2 assimilated during photosynthesis. Soviet Plant Physiology, 19: 17–20. Schade, G.W. and Goldstein, A.H. (2001) Fluxes of oxygenated volatile organic compounds from a ponderosa pine plantation. Journal of Geophysical Research, 106 (D3): 3111–23. Schade, G.W. and Goldstein, A.H. (2002) Plant physiological influences on the fluxes of oxygenated volatile organic compounds from ponderosa pine trees. Journal of Geophysical Research, 107 (D10): 4082. Schade, G.W. and Goldstein, A.H. (2003) Increase of monoterpene emissions from a pine plantation as a result of mechanical disturbances. Geophysical Research Letter, 30: 1380. Schade, G.W., Goldstein, A.H. and Lamanna, M.S. (1999) Are monoterpene emissions influenced by humidity? Geophysical Research Letter, 26 (14): 2187–90. Schade, G.W., Goldstein, A.H., Gray, D.W., et al. (2000) Canopy and leaf level 2-methyl-3-buten-2-ol fluxes from a ponderosa pine plantation. Atmospheric Environment, 34 (21): 3535–44. Scheeren, H.A., Lelieveld, J., Roelofs, G.J., et al. (2003) The impact of monsoon outflow from India and Southeast Asia in the upper troposphere over the eastern Mediterranean. Atmospheric Chemistry and Physics Discussion, 3: 2285–330. Schnitzler, J.-P., Graus, M., Kreuzwieser, J., et al. (2004) Contribution of different carbon sources for isoprene emitted from poplar leaves. Plant Physiology, 135: 152–60. Scott, K.I. and Benjamin, M.T. (2003) Development of a biogenic volatile organic compound emission inventory for the SCOS97-NARSTO domain. Atmospheric Environment, 37 (2): S39–49. Seinfeld, J.H. and Pandis, S.N. (1998) Atmospheric Chemistry and Physics. John Wiley & Sons, New York. Sharkey, T.D., Loreto, F. and Delwiche, C.F. (1991) The biochemistry of isoprene emission from leaves during photosynthesis. In: T.D. Sharkey, E.A. Holland and H.A. Mooney (Eds) Trace Gas Emissions from Plants. Academic Press, San Diego, CA, pp. 153–84. Sharkey, T.D., Singsaas, E.L., Lerdau, M., et al. (1999) Weather effects on isoprene emission capacity and applications in emissions algorithms. Ecological Application, 9: 1132–7. Sharkey, T.D. and Singsaas, E.L. (1995) Why plants emit isoprene. Nature, 374: 569. Sharkey, T.D. and Yeh, S. (2001) Isoprene emission from plants. Annual Review of Plant Physiology and Plant Molecular Biology, 52: 407–36. Silver, G. and Fall, R. (1991) Enzymatic synthesis of isoprene from dimethylallyl diphosphate in aspen leaf extracts. Plant Physiology, 97: 1588–91. Silver, G.M. and Fall, R. (1995) Characterization of aspen isoprene synthase, an enzyme responsible for leaf isoprene emission to the atmosphere. Journal of Biological Chemistry, 270 (22): 13010–16. Simpson, D., Guenther, A., Hewitt, C.N., et al. (1995) Biogenic emissions in Europe 1. Estimates and uncertainties. Journal of Geophysical Research, 100 (D11): 22875–90. Singh, H.B., O’Hara, D., Herlth, D., et al. (1994) Acetone in the atmosphere: Distribution, sources, and sinks. Journal of Geophysical Research, 99 (D1): 1805–19. Singh, H.B., Chen, Y., Tabazadeh, A., et al. (2000) Distribution and fate of selected oxygenated organic species in the troposphere and lower stratosphere over the Atlantic. Journal of Geophysical Research, 105 (D3): 3795–805.

Biogenic VOCs

127

Singh, H.B., Chen, Y., Staudt, A., et al. (2001) Evidence from the Pacific troposphere for large global sources of oxygenated organic compounds. Nature, 410: 1078–81. Singh, H.B., Tabazadeh, A., Evans, M.J., et al. (2003) Oxygenated volatile organic chemicals in the oceans: Inferences and implications based on atmospheric observations and air-sea exchange models. Geophysical Research Letter, 30 (16): 1862. Singh, H.B., Salas, L.J., Chatfield, R.B., et al. (2004) Analysis of the atmospheric distribution, sources, and sinks of oxygenated volatile organic chemicals based on measurements over the pacific during TRACE-P. Journal of Geophysical Research, 109: D15S07. Singsaas, E.L. and Sharkey, T.D. (2000) The effects of high temperature on isoprene synthesis in oak leaves. Plant, Cell and Environment, 23: 751–7. Singsaas, E.L., Lerdau, M., Winter, K., et al. (1997) Isoprene increases thermotolerance of isopreneemitting species. Plant Physiology, 115: 1413–20. Smith, A.M., Rigler, E., Kwok, E.S.C., et al. (1996) Kinetics and products of the gas-phase reactions of 6-methyl-5-hepten-2-one and trans-cinnamaldehyde with OH and NO3 radicals and O3 at 296+/−2K. Environmental Science and Technology, 30: 1781–5. Solmon, F., Sarrat, C., Serca, D., et al. (2004) Isoprene and monoterpenes biogenic emissions in France: Modeling and impact during a regional pollution episode. Atmospheric Environment, 38 (23): 3853–65. Staudt, M. and Seufert, G. (1995) Light-dependent emissions of monoterpenes by Holm oak (Quercus ilex L.). Naturwissenschaften, 82: 89–92. Staudt, M., Joffre, R., Rambal, S., et al. (2001) Effect of elevated CO2 on monoterpene emission of young Quercus ilex trees and its relation to structural and ecophysiological parameters. Tree Physiology, 21: 437–45. Staudt, M., Mir, C., Joffre, R., et al. (2004) Isoprenoid emissions of Quercus spp. (Q. suber and Q. ilex) in mixed stands contrasting in interspecific genetic introgression. New Phytologist, 163: 573–84. Steinbrecher, R. (1989) Gehalt und Emission von Monoterpenen in oberirdischen Organen von Picea abes (L.). Universitaet Muenchen, Munich, Germany. Steiner, A.L., Luo, C., Huang, Y., et al. (2002) Past and present-day biogenic volatile organic compound emissions in East Asia. Atmospheric Environment, 36: 4895–905. Stewart, H.E., Hewitt, C.N., Bunce, R.G.H., et al. (2003) A highly spatially and temporally resolved inventory for biogenic isoprene and monoterpene emissions: Model description and application to Great Britain. Journal of Geophysical Research, 108 (D20): 4644. Taylor, J., Zimmerman, P. and Erickson, D. (1990) A 3-D modelling study of the sources and sinks of atmospheric carbon monoxide. Australian National University, Canberra. Tie, X., Guenther, A. and Holland, E.A. (2003) Biogenic methanol and its impacts on tropospheric oxidants. Geophysical Research Letter, 30 (17): 1881. Tingey, D., Evans, R. and Gumpertz, M. (1981) Effects of environmental conditions on isoprene emission from live oak. Planta, 152: 565–70. Tingey, D.T., Turner, D.P. and Weber, J.A. (1991) Factors controlling the emissions of monoterpenes and other volatile organic compounds. In: T.D. Sharkey, E.A. Holland and H.A. Mooney (Eds) Trace Gas Emissions by Plants. Academic Press Inc., San Diego, CA, pp. 93–119. Trainer, M., Williams, E.J., Parrish, D.D., et al. (1987) Models and observations of the impact of natural hydrocarbons on rural ozone. Nature, 329: 705–7. Tsigaridis, K. and Kanakidou, M. (2002) Importance of volatile organic compounds photochemistry over a forested area in central Greece. Atmospheric Environment, 36: 3137–46. Turner, D., Baglio, J., Pross, D., et al. (1991) Climate change and isoprene emissions from vegetation. Chemosphere, 23: 37–56. Wang, Y., Jacob, D.J. and Logan, J.A. (1998) Global simulation of tropospheric O3 -NOx -hydrocarbon chemistry, 3. Origin of tropospheric ozone and effects of nonmethane hydrocarbons. Journal of Geophysical Research, 103 (D9): 10757–67.

128


Warneke, C., Holzinger, R., Hansel, A., et al. (2001) Isoprene and its oxidation products methyl vinyl ketone, methacrolein, and isoprene related peroxides measured online over the tropical rain forest of Surinam in March 1998. Journal of Atmospheric Chemistry, 38: 167–85. Warneke, C., Luxembourg, S.L., DeGouw, J.A., et al. (2002) Disjunct eddy covariance measurements of oxygenated volatile organic compounds fluxes from an alfalfa field before and after cutting. Journal of Geophysical Research, 107 (D8): 4067. Warneke, C., DeGouw, J.A., Goldan, P., et al. (2004) Comparison of daytime and nighttime oxidation of biogenic and anthropogenic VOCs along the New England coast in summer during New England Air Quality Study 2002. Journal of Geophysical Research, 109: D10309. Wennberg, P.O., Hanisco, T.F., Jaegle, L., et al. (1998) Hydrogen radicals, nitrogen radicals, and the production of O3 in the upper troposphere. Science, 279: 49–53. Went, F.W. (1960) Blue hazes in the atmosphere. Nature, 187: 641–3. Wiedinmyer, C., Guenther, A., Harley, P.C., et al. (2004) Global organic emissions from vegetation. In: C. Granier, P. Artaxo and C. Reeves (Eds) Emissions of Atmospheric Trace Compounds. Kluwer Academic Publishers, Dordrecht, The Netherlands, pp. 115–70. Williams, J., Holzinger, R., Gros, V., et al. (2004) Measurements of organic species in air and seawater from the tropical Atlantic. Geophysical Research Letter, 31: L23S06. Winer, A.M. et al. (1983) Investigation of the role of natural hydrocarbons on photochemical smog formation in California. California Air Resources Board, Statewide Air Pollution Research Center, Riverside, CA. Yokouchi, Y. and Ambe, Y. (1984) Factors affecting the emission of monoterpenes from red pine (Pinus densiflora)-long-term effects of light, temperature and humidity. Plant Physiology, 74: 1009–12. Zavarin, E. (1968) Chemotaxonomy of the genus Abies – II. Within tree variation of the terpenes in cortical oleoresin. Phytochemistry, 7: 92–107. Zeidler, J.G., Lichtenthaler, H.K., May, H.U., et al. (1997) Is isoprene emitted by plants synthesized via the novel isopentenyl pyrophosphate pathway? Zeitschrift fuer Naturforschung C, 52: 15–23. Zimmer, W., Brueggeman, N., Emeis, S., et al. (2000) Process-based modelling of isoprene emission by oak leaves. Plant, Cell and Environment, 23: 585–595. Zimmerman, P. (1979) Testing of hydrocarbon emissions from vegetation, leaf litter and aquatic surfaces, and development of a methodology for compiling biogenic emission inventories. U.S. Environmental Protection Agency, Research Triangle Park, NC, EPA-450/4-79-004: 1–113.


Chapter 4

Oxygenated Volatile Organic Compounds Ralf Koppmann and Jürgen Wildt

4.1

Introduction

Oxygenated volatile organic compounds (OVOCs) belong to the large family of organic compounds present in the global atmosphere. These compounds include carbonyls, alcohols, ketones, esters, ethers, organic peroxides and organic hydroperoxides. At least the lighter-molecular-weight OVOCs are ubiquitous at relatively high concentrations in the troposphere, and it is widely accepted that they play an important role in atmospheric photochemistry. Despite that recognition, their study has been somewhat neglected, mainly because they are difficult to measure. OVOCs have complex primary and secondary sources. On the one hand they are emitted directly into the atmosphere from a variety of anthropogenic and natural sources. On the other hand, they are products, often the first stable ones, in the gas-phase oxidation pathways of organic compounds in the atmosphere. The removal of OVOCs from the atmosphere occurs mainly by the reaction with OH radicals, but also by photolysis and by wet and dry deposition. However, the removal processes are well understood only for a few compounds. As a consequence, the budgets of OVOCs are either poorly understood or not known at all. While the role of formaldehyde (HCHO) as an oxidation product of methane and other VOCs has been studied for more than two decades, the last years have witnessed a growing interest in the role of other OVOCs owing to improved analytic technologies and a deeper understanding of photochemical processes. As an example, acetone was found to be a key atmospheric compound influencing tropospheric chemistry. It has a significant impact on atmospheric chemistry, especially in the upper troposphere, where photolysis leads to the formation of peroxy acetyl nitrate (PAN). In this way, acetone can influence ozone chemistry by sequestering nitrogen oxides (NOx ) in the form of PAN and by providing free radicals (HOx ) especially in the free troposphere (Jaeglé et al. 2001; McKeen et al. 1997; Singh et al. 1994, 1995; Wennberg et al. 1998). OVOCs may also contribute to organic carbon in aerosols (Jang et al. 2002; Kalberer et al. 2004; Li et al. 2001; Tabazadeh et al. 2004). In addition to their photochemical impact, many of these compounds are harmful to human health especially in urban environments. They affect the respiratory tract and irritate the eyes and some of them are believed to be mutagenic and cancerogenic (Kean et al. 2001; Sin et al. 2001).

130


The increasing awareness that these compounds are important players in tropospheric chemistry has lead to significant ongoing research although a quantitative determination of oxygenated hydrocarbons is still a challenge.

4.2

Tropospheric mixing ratios and global distribution

The abundance of OVOCs in the atmosphere has been studied on local, regional, and global scales. The investigated compounds cover a variety of aldehydes, ketones, alcohols, and carboxylic acids. Understandably, the largest set of compounds can be found in urban air, in the immediate vicinity of the sources. In these cases the measurements benefit from high mixing ratios and short transport times from the source to the measurement location. In the context of air quality studies, a large number of investigations have been carried out concerning OVOC concentrations on a local scale. These studies dealt with a large set of aldehydes, ketones, and sometimes alcohols in the urban atmosphere. There are a fewer number of investigations at semi-rural or rural sites, and some studies at remote sites or on a global scale. The investigations on global scales are typically limited to formaldehyde, acetone, methanol, and ethanol. This can be attributed to their ubiquitous production or long lifetimes in the atmosphere leading to larger scale distribution and accessible mixing ratios in remote areas and in the free troposphere, respectively. Table 4.1 gives an overview of these studies and lists the concentrations resp. mixing ratios of the three most abundant OVOCs in the urban atmosphere, formaldehyde, acetaldehyde and acetone. In the following we will give an overview of OVOC measurements sorted by compound groups. Table 4.2 summarises the mixing ratios also for highermolecular-weight OVOCs obtained from field measurements in rural areas from selected studies. Aldehydes. There are numerous studies on atmospheric mixing ratios of formaldehyde and acetaldehyde. A large part of these investigations has been done in cities and populated areas around the world. Formaldehyde and acetaldehyde are the main OVOCs found in urban environments, together contributing to more than half of the OVOCs by mass (Kean et al. 2001; Sin et al. 2001). Typical concentrations of formaldehyde and acetaldehyde observed in urban air range between 1 and 45 μg/m3 (corresponding to 0.8 and 36 ppb) and 0.7 and 35 μg/m3 (corresponding to 0.5 and 19 ppb), respectively. In rural environments mixing ratios are in the lower ppb range, sometimes less than 1 ppb. At remote locations and in the free troposphere mixing ratios seldom exceed 1 ppb. A few studies report observations of higher-molecular-weight aldehydes in urban air. Higher-molecular-weight aldehydes observed in the rural atmosphere often consist of those formed during lipoxygenase (LOX) activity (C6 aldehydes, see section ‘Biogenic emissions’) −C10 aldehydes such as hexanal, heptanal, octanal, nonanal, and decanal. and saturated C6 − These compounds have been found in mixing ratios from 0.65 >2.3 >8

year year year

2 2.6 >2.3 >0.9 >0.9 5.7

day h year year year h

Reissell et al. (2001) Alvarado et al. (1998) Calogirou et al. (1999) and Atkinson and Aschmann (1993) Calogirou et al. (1999) Atkinson and Aschmann (1993) Alvarado et al. (1998) Alvarado et al. (1998) Atkinson and Aschmann (1993) Baker et al. (2004)

1

326


Table 8.9 Lifetimes due to the photolysis of higher-molecular-weight carbonyls estimated in Hyytiälä (Finland) and Valencia (Spain) and their comparison with those of the reaction OH radicals Lifetime Site

τphot Hyytiälä (Hellén et al. (2004))

τphot Valencia (Moortgat (2001))

Season Solar radiation information Carbonyl compound

Spring 61.30◦ N, Zenith angle 60◦ Time

Unit

Summer 40◦ N, Zenith angle 16.9◦ Time

Formaldehyde Acetaldehyde MAC Propanal Butanal 2-Methylbutanal Pentanal Hexanal Nonanal Pinonaldehyde MEK MVK Nopinone Limonaketone

8 8.1 12 2.4 1.6

h day day day day

1.5 1.3 1.6

day day day

2.9–29

day

>8 >8

day day

τOH Assuming [OH] = 2 × 106 molecule/cm

Unit

Time

Unit

5 4 >6 1.2 1.2 7.3 17 17 1 1

h day day day day h h h day day

>6 >4 >4

day day day

16 9.2 4 7 5.7 6 5.3 5.1 4.8 3.3 4.8 7 10 1.1

h h h h h h h h h h day h h h

radicals (HO2 and RCO2 ) by reaction with O2 , able to convert NO to NO2 . As mentioned earlier, alkylperoxy radicals can further react (or decompose), giving rise to HO2 radicals and smaller molecular weight carbonyls, as products. In the Norrish II type of reaction two stable products usually an olefin and an alkanal are formed (Finalyson-Pitts and Pitts −C7 n-alkanals, ethenol and an olefin were the com2000). However, in the photolysis of C5 − pounds formed (Tadic et al. 2002), both of which can be quite efficient in the production of ozone. The effective quantum yields of high-molecular-weight carbonyls are much smaller than −C9 those of formaldehyde, acetaldehyde and acetone (Moortgat 2001). In the case of C4 − n-alkanals, values are in the range of 0.25–0.3. For MAC and MVK, they are even smaller (0.004). Only in the case of α-branched alkanals values ranging between 0.2 and 0.7 were reported (Moortgat 2001). These data suggest that photodissociation of high-molecularweight carbonyls is not as efficient as the reaction with OH radicals in the ozone production, even at noontime of summer days and at low latitudes. This conclusion is summarised in Table 8.9, in which the lifetimes due to photolysis calculated at two sites are compared with those of the reaction with OH radicals. Results indicate that photodissociation of highmolecular-weight carbonyls can take some part in the ozone production only for a limited number of species and under high solar radiation intensities.

High-Molecular-Weight Carbonyls and Carboxylic Acids

8.4.2

327

Ozone and photochemical oxidant production

A reasonable assessment of the contribution of higher-molecular-weight carbonyl to ozone and total alkyl nitrates (TAN) has been made in two recent publications (Cleary et al. 2005; Hellén et al. 2004). In both cases a relevant contribution was found. By scaling the OH-reactivity of different VOCs with respect to formaldehyde, it was estimated that about 52% of the OH radicals present in Hyytiälä were removed by carbonyl compounds (Hellén et al. 2004). About half of this portion was assigned to high-molecularweight carbonyls, although their mass was less than 24% of the total carbonyl fraction. −C11 carbonyls accounted for about 26% of the ozone proCalculations showed that C4 − duced by all VOCs. The compound contributing most to ozone formation was 6-MHO, present at trace levels in the VOC mixture (30%) to ozone and TAN formation were MAC followed by nonanal and MVK. The contribution of higher-molecular-weight carbonyls to PANs has been accurately measured only for products originated from isobutanal (PiBN, peroxyisobutyric nitric anhydride), and MAC (MPAN, peroxymethacrylic nitric anhydride) (Roberts et al. 1998). It was found to be less than 8% (on a molar basis) of the total fraction in polluted plumes in Texas and Tennessee. It is also possible that other compounds, particularly nonanal and decanal, contribute to the total burden of PANs (Bowman et al. 2003), but no data exist on the atmospheric levels of these peroxyacylnitrates. It should be noted that PANs formed from high-molecular-weight compounds, in which addition of OH radicals prevails over H-abstraction, usually contain one or more hydroxyl groups in the molecule, and they will be easily dissolved in the water film covering fine aerosols. The same considerations hold for nitrates originated from first generation products of unsaturated, cyclic and aromatic carbonyls.

8.4.3

Secondary organic aerosol formation

It is now recognised that about 78% of the mass production of secondary organic aerosols (SOA) in the troposphere comes from monoterpene oxidation products, and that SOA is

328


100

Percent

80 HMW Carbonyls

60

LMW Carbonyls Other VOC

40

20

0 Composition

O3 production

AN production

Figure 8.4 Contribution of lower- (LMW) and higher-molecular-weight carbonyls (HMW) to the VOC composition and to the ozone and alkyl nitrate (AN) production in the urban plume of Sacramento, California. Data are from Cleary et al. (2005).

an important fraction of all particulate matter (Chung and Seinfeld 2002). However, the contribution of higher-molecular-weight carbonyls to SOA is still controversial. According to Bonn et al. (2004), biogenic hydroperoxides are the compounds with the highest contribution (63%) to both SOA formation and SOA mass production. PANs and nitrates contribute for about 11–12%. Carbonyls formed from monoterpene oxidation should not contribute at all to SOA formation and SOA mass production because their vapour pressure is too high and water solubility too low to efficiently transfer them into the water film covering the aerosol phase. Based on these considerations, these carbonyls should be almost exclusively found in the gas-phase. High concentrations should be, thus, measured in forest areas, where the emission of parent monoterpenes and the oxidation capacity of the atmosphere are high. These conclusions are in contrast to the existing observations indicating low levels of these compounds in forest areas. Even the most volatile product (nopinone) is detected at a few sites and always at small concentrations. For this reason, some authors (Jang et al. 2002; Tolocka et al. 2004) believe that carbonyls formed from monoterpene oxidation are heavily involved in both SOA formation and SOA mass production. These conclusions are supported by laboratory experiments indicating strong oxidation of carbonyls in the water film covering the aerosol surface (Jang et al. 2002; Tolocka et al. 2004). Efficient transfer in the liquid phase occurs because carbonyls are converted into highly watersoluble oligomers (Jang et al. 2002). Oxidation preferentially occurs in acid solutions and is catalysed by some metal ions (like iron) very abundant in atmospheric particles (Jang et al. 2002). Precursors of water soluble oligomers are hydrophilic compounds formed by the keto–enolic equilibrium of carbonyls in acid solutions (Jang et al. 2002). Substantial formation of numerous oligomeric compounds has been reported by Tolocka et al. (2004) in the ozonolysis reaction of α-pinene carried out in the presence of seed particles. To identify such large and highly polar molecules, electron-spray Fourier-transform ion-cyclotronmass-spectrometry (ESI-FTCIP) was used. Oligomeric compounds attributed to isoprene


329

oxidation have recently been found in the water extracts of particles collected in the Amazon region (Claeys et al. 2004). While these oxidation mechanisms explain well how carbonyls contribute to SOA mass production, it is not clear how they can promote SOA formation. It is not yet proven that changes in the hygroscopic properties of SOA caused by the accumulation of oligomers in the water film make the aerosols to act as cloud condensation nuclei. Probably, the oxidation mechanism of carbonyls is more complex than what is presently believed and involves the formation of additional compounds affecting the hygroscopic features of SOA. This conclusion is supported by recent experiments indicating the active involvement of OH radicals in SOA formation. Lower yields of SOA were observed in the ozonolysis of monoterpenes when OH radicals were removed from the reaction chamber (Iinuma et al. 2005). How these radicals affect the oxidation processes of carbonyls in the liquid phase and the hygroscopic feature of SOA is still a matter of investigation. High-molecular-weight alkanoic acids have been found in particles emitted from many anthropogenic sources, especially food cooking (Schauer et al. 1999b, 2002b) and biomass burning (Schauer et al. 2001). Low contents were detected, instead, in the extracts of particulate matter released from vehicular emission (Kawamura et al. 2000; Schauer et al. 1999a, 2002a). No specific studies have been made to quantify the contribution of alkanoic acids to SOA. They should contribute to SOA mass production because they are much more soluble than carbonyl compounds in water solutions. However, their contribution should be much smaller than that of acidic compounds containing more than one polar group in the molecule (such as dicarboxylic acids or the ones containing a carbonyl and a carboxylic group in the molecule), as they have much lower vapour pressure and lower water solubility. They are ubiquitous constituents of atmospheric particulate matter but they are present at very low concentrations.

8.5 8.5.1

Sampling and analysis Carbonyl compounds

Basically, only limited advances in the collection and analysis of higher-molecular-weight carbonyl compounds have occurred in the past 10 years. The most widely used methods remain those based on the preconcentration of carbonyls on traps filled with solid sorbents, followed by their analysis by capillary GC or HPLC. Liquid extraction or thermal desorption can be used to recover VOCs from the adsorption media. Thermal desorption is usually performed on uncoated cartridges filled with graphitic carbon adsorbents (Ciccioli et al. 1992, 1993a, 1993b, 1994, 2002) and porous polymers of the Tenax family (Bowman et al. 2003; Ciccioli et al. 2002; Helmig et al. 1996; McClenny et al. 1998). It is obtained by increasing the temperature of the trap up to 200–250◦ C. In this step, the direction of the flow rate of the carrier gas is opposite to that used during sampling. Desorbed compounds are cryofocussed on a capillary tube kept at very low temperature (typically −150◦ C), before they are injected into the analytical column. This step is necessary to avoid band broadening of the sample, which drastically reduces the efficiency and selectivity of capillary columns.

330


The use of ozone scrubbers is particularly important with Tenax adsorbents because ozonolysis of the sorbent matrix produces large amounts of benzaldehyde, phenol and acetophenone (Ciccioli et al. 2002; Helmig 1997 and reference therein). Formation of n-alkanals by ozonolysis of some Tenax adsorbents has also been reported (Roberts et al. 1984). As far as carbon materials are concerned, the removal of ozone is mandatory only with certain types of traps, because partial decomposition of n-alkanals has been observed (McClenny et al. 2001). Traps exhibiting these features are those in which a graphitic carbon adsorbent (Carbopack B) is placed in series with a carbon molecular sieve (usually Carbosieve S III). Although recommended by the U.S. EPA for certain applications, this combination was already found not suitable for the collection of polar VOCs in air, especially in humid environments (Ciccioli et al. 1992, 2002). Better results can be obtained with traps filled with two or three graphitic carbon adsorbents with surface areas ranging from 15 to 250 m2 /g (Ciccioli et al. 2002). VOCs retained on this type of traps show a better resistance to ozone and atmospheric contaminants present in atmospheric water. The quality of the sample can be preserved at levels higher than 90 ppbv (Ciccioli et al. 2002). Above these values, the use of scrubbers is recommended, especially in very humid environments. Since ozone scrubbers can affect the quality of the sample, special care should be paid in their selection (Helmig 1997). At the moment, the simplest ozone scrubber that better preserves the quality of the sample is the one obtained by impregnating a glass-coated filter with a dilute solution of sodium thiosulfate (Helmig 1997). Although less practical, the addition of NO to the air stream entering the trap is also efficient, and preserves well the quality of the sample (Helmig 1997). With adsorption traps filled with porous polymers and graphitic carbon materials, volumes ranging from 2 to 5 L are sufficient for the analysis of high-molecular-weight compounds because the whole sample can be injected into the capillary column. Positive identification of eluted compounds is achieved by knowing the retention times of VOCs on the capillary column used and by selecting mass spectrometry (MS) as detection system (Ciccioli et al. 2002). The mass spectrometer can be run in the scan mode or in selected ion detection. Relative retention indices and selective ions for the identification and quantification of higher-molecular-weight carbonyls and acids by GC–MS have been recently published by Ciccioli et al. (2002). The database contains information for the identification and quantification of more than 600 VOCs analysed on a DB-1 capillary column. Elution of all VOCs from C4 to C14 is achieved by increasing the column temperature from 30◦ C to 250◦ C. Providing that the same column is used, the database can help in the identification of VOCs when other GC detectors such as a flame ionisation detector (FID) are used. Methods based on 2,4-dinitrophenylhydrazine (DNPH)-coated cartridges are also widely used for the selective collection of high-molecular-weight carbonyls (Cecinato et al. 2002; Bakeas et al. 2003; Grosjean et al. 1996, 1999, 2002; Hellén et al. 2004; Ho et al. 2002; Muller 1997; Zhang et al. 1994). With respect to those using uncoated material, they offer the advantage that all carbonyl compounds, including the very volatile ones, can be detected at once. This method does not provide, however, any information on the other VOCs present in the sample, because selective retention is achieved by converting carbonyls into hydrazone derivatives. With these cartridges, the use of an ozone scrubber is mandatory −C11 alkanals, occur to a large because sampling artefacts, simulating the presence of C4 − extent even at low ozone levels (Pires and Carvalho 1998). The most widely used ozone scrubbers are those made by glass or copper tubes internally coated with potassium iodide


331

(Pires and Carvalho 1998). In some instances, annular denuders coated with the same material also have been successfully used for ozone removal (Cecinato et al. 2002). With conventional DNPH-coated cartridges, sample recovery is accomplished by liquid extraction, using acetonitrile or methanol as eluants. Since only small aliquots of the liquid extract can be injected in the HPLC column, high sampling volumes are needed to meet the sensitivity requirements of the detection system. Depending on the concentrations of carbonyls in air, sampled volumes range from 0.15 to 1.5 m3 (Cecinato et al. 2002; Bakeas et al. 2003; Grosjean et al. 1996, 1999, 2002; Hellén et al. 2004; Ho et al. 2002; Muller 1997; Zhang et al. 1994). High sampling flow rates must be used when diurnal cycles of carbonyls need to be followed. Gradient elution is necessary to quantify the high molecular fraction of carbonyls by HPLC. Since the best selectivity is obtained on columns working in reversed-phase liquid chromatography, the analysis starts with organic eluants (usually acetonitrile or methanol) containing high percent of water (usually 64% v/v). The water content is then gradually decreased until 100% of the organic solvent is passed through the column (Grosjean et al. 1996, 2002; Pires and Carvalho 1998). For a complete elution of carbonyls containing 11 carbon atoms in the molecule, the flow of the pure organic solvent must be maintained for some time. With conventional HPLC columns, with an internal diameter of 4.6 mm and a length of 250 mm, flow rates of the eluant typically range between 1 and 1.5 ml/min. UV-visible absorption is the cheaper and commonly used method for the detection of hydrazones. It can be performed at various wavelengths (360, 385 and 430 nm), but the most intense absorption band occurs at around 360 nm. Pure hydrazone standards are needed for identification and quantification purposes. After the pioneering work made by Grosjean and co-workers (Grosjean et al. 1996, 1999), MS has become more and more used for the positive identification of DNPH-derivatives eluted from HPLC columns. Selective identification can be achieved by atmospheric pressure mass spectrometry working in negative ion recording (Grosjean et al. 1999; Kolloker et al. 1998). Particularly useful for identification purposes is the use of MS–MS techniques, allowing the selective fractionation of parent and secondary ions into fragments providing structural information on the eluted compound (Kolloker et al. 1998). With the advent of instruments equipped with ion-trap sources connected to quadrupole analysers, this method has become affordable to many research laboratories. An instrument of this type has been used for the analysis of samples collected in Helsinki and Hyytiälä (Hellén et al. 2004). Very recently, attempts have been made to develop DNPH-coated cartridges suitable for thermal desorption. This task has been successfully accomplished by depositing small amounts of DNPH on Tenax GC (Ho and Yu 2004). The analysis of hydrazones is carried by GC–MS. Very good profiles of carbonyls from acetaldehyde to hexanal have been obtained with this method (Ho and Yu 2004). However, the electron beam of the MS must be shut off during the elution of unreacted DNPH. Since the reagent generates a large peak overlapping with formaldehyde, this compound cannot be quantified by GC–MS. To avoid overloading of the capillary column and the appearance of DNPH impurities in the chromatogram, the reagent must be extremely pure and the amount deposited on the Tenax surface accurately calculated (Ho and Yu 2004). Another derivatising agent allowing the selective collection of carbonyl compounds in air is perfluorophenylhydrazine (PFPH) (Cecinato et al. 2001, 2002; Schlomski et al. 1997).

332


It has the same efficiency as DNPH but hydrazones formed have a much higher volatility and a better thermal stability. This makes their analysis by GC–MS easier. Sample volumes needed for the analysis are smaller than those required with conventional DNPH-coated cartridges (Cecinato et al. 2001). The only limitation associated with the use of PFPH is that reaction with carbonyls often gives rise to two different hydrazones (the E and Z isomers) (Cecinato et al. 2001). Only compounds that are symmetrical with respect to the functional group (such as formaldehyde and acetone) form single products. Studies performed on the ozone interferences (Cecinato et al. 2001) have shown that an ozone scrubber is needed to prevent artefacts formation (Cecinato et al. 2001, 2002). Only one attempt has been made to combine conventional DNPH-coated cartridges with uncoated adsorption traps filled with graphitic carbon materials (Possanzini et al. 2000). The system performed quite well in the analysis of carbonyls and high-molecularweight VOCs, but the identification and quantification of the whole sample required the use of different analytical systems. Moreover, front traps, made by tubes filled with carbon materials, needed to be changed when 10 L of air were passed through the sampling system. Denuders coated with o-benzylhydroxyl ammonium chloride have been successfully used for the gas-phase collection of 4-OPA, nonanal, decanal, pinonic acid and few other volatile carbonyls (Matsunaga et al. 2003, 2004a, 2004b; OSOA-Final report and Hoffmann 1999). The benzylhydroxyl oximes formed by reaction with carbonyl compounds were extracted with ethyl acetate, and the solution analysed by GC-FID. To prevent degradation of oximes by ozone, a flow of NO was added to the air stream entering the denuder system. So far direct analysis of higher-molecular-weight carbonyls has been performed only with PTR-MS (Karl et al. 2001). Basically, the instrument is a mass spectrometer working with chemical ionisation. High sensitivity is achieved by using a flow tube to obtain very high ionisation efficiencies of the molecules to be identified. H3 O+ ions generated in the source are mixed with the air sample inside the flow tube. Efficient oxidation of carbonyl compounds with more than one carbon atom takes place through an acid-base reaction, in which a proton is transferred to the neutral molecule. From this reaction, a parent ion with a mass of (M + 1) is generated. Since the energy transferred by the proton to the neutral molecule is on the order of a few eV, fragmentation of the parent ions is very small or lacking. The main limit of this instrument is the impossibility to distinguish isobaric compounds, as they generate parent ions of the same mass/charge ratio. Therefore, carbonyls such as MAC and MVK cannot be individually quantified. The same holds true for hexanals and hexanols. So far, this instrument has been successfully applied to the direct determination of alkanals only in air sheds in which no interfering compounds were present (Karl et al. 2001). Even with these limitations, monitoring with PTR-MS is extremely advisable because the measurements are instantaneous and do not require any ozone scrubber. The ideal situation is to use PTR-MS in parallel with conventional methods to check the levels of carbonyls and the possible occurrence of sampling artefacts.

8.5.2

Carboxylic acids

For high-molecular-weight alkanoic acids, the simplest and reliable sampling method remains the one based on filters coated with alkaline agents (usually KOH), placed after the filter used for particle collection (Satsumabayashi et al. 1995). Acids can be analysed by


333

GC-FID or GC–MS, after derivatisation. Denuder sampling has also been used for the collection of these compounds, but only in emission sources (Schauer et al. 1999a, 2001, 2002a, 2002b). Although PTR-MS has a great potential in the analysis of acids, no applications have been reported so far.

8.5.3

Calibration methods

Calibration of the instrumentation is a critical step in the analysis of high-molecular-weight carbonyls and carboxylic acids as it affects both the accuracy and precision of determinations. It has been shown that good gaseous standards of alkanals up to nonanal can be obtained (Ciccioli et al. 2002) by using the system proposed by Gautrois and Koppmann (1999). A constant source of VOCs is obtained by diffusion of vapours through a capillary tube in equilibrium with the liquid. Vapours emitted are mixed with the flow of an inert gas (usually helium or nitrogen) to generate gas mixtures at known concentrations. A reliable source is obtained by keeping the vials at constant temperature and pressure. A schematic diagram of the apparatus for the generation of standard mixtures of VOCs, including the systems needed for the addition of ozone and water vapours, can be found in Ciccioli et al. (2002), Gautrois et al. (1999) and Larsen et al. (1997). It is worth noting that this calibration method generates primary standards because the mass of VOCs released by capillary diffusion can be accurately determined by measuring the weight losses of liquid from the vials. With preconcentration devices, adsorption of few μL of liquid standard solutions, containing the compounds of interest at ppmv levels, is an alternative calibration method. The solvent, usually a chlorinated solvent, can be removed by passing adequate volumes of clean air (or helium) through the trap. Very good consistency has been reported between this method and the one based on diffusion tubes (Ciccioli et al. 2002). The use of liquid solutions is particularly suitable for carboxylic acids (Ciccioli et al. 2002). Gas cylinders filled with standard gaseous mixtures are not reliable for the calibration of high-molecular-weight carbonyls and carboxylic acids, because selective adsorption and condensation can take place in the container walls. Concentrations can drastically change as a function of the time and ambient temperature.

8.6

Conclusions

After 15 years of data collection and analysis, the hypothesis of the ubiquitous occurrence of high-molecular-weight carbonyls in the troposphere is now a consolidated reality. It is supported by reliable data on the emission and a better knowledge of photochemical processes leading to their formation, either in the gas phase or in condensed phases. These compounds definitely contribute to the production of ozone, PANs, TAN and, very likely, to the mass of SOA. Although their sampling is not performed on a routine basis, more and more laboratories now start to recognise their importance, and actions are taken to monitor them. However, research is lacking on new methods for their collection and detection in air. Data on alkanoic acids are still extremely limited. This is only partly justified by their low atmospheric reactivity and aerometric levels. Also, in this case, almost no research has been performed in the development of new methods for their collection and analysis.

334


References Alvarado, A., Arey, J. and Atkinson, R. (1998) Kinetics of the gas-phase reactions of OH and NO3 radicals and O3 with the monoterpene reaction products pinonaldehyde, caronaldehyde, and sabinaketone. Journal of Atmospheric Chemistry, 31: 281–97. Andreae, M.O. and Merlet, P. (2001) Emission of trace gases and aerosols from biomass burning. Global Biogeochemical Cycles, 15: 955–66. Arey, J., Crowley, D.E., Crowley, M., Resketo, M. and Lester, J. (1995) Hydrocarbons emissions from natural vegetation in California’s South coast air basin. Atmospheric Environment, 29: 2977–88. Aschmann, S.M., Reissell, A., Atkinson, R. and Arey, J. (1998) Products of the gas phase reactions of the OH radical with α- and β-pinene in the presence of NO. Journal of Geophysical Research, 103: 25553–61. Atkinson, R. and Arey, J. (2003) Gas-phase tropospheric chemistry of biogenic volatile organic compounds: A review. Atmospheric Environment, 37: S197–219. Atkinson, R., Baulch, D.L., Cox, R.A., et al. (2005) Evaluated kinetic and photochemical data for atmospheric chemistry: Volume II – reactions of organic species. Atmospheric Chemistry and Physics Discussion, 5: 6295–7168. Atkinson, R. and Aschmann, S.M. (1993) Atmospheric chemistry of the monoterpene reaction products nopinone, camphenilone, and 4-acetyl-1-methylcyclohexene. Journal of Atmospheric Chemistry, 16: 337–48. Audino Gonzalez, P., Villaverde, R., Alfaro, R. and Zerba, E. (2005) Identification of volatile emissions from Platypus mutatus (sulcatus) (Coleoptera: Platypodidae) and their behavioral activity. Journal of Economic Entomology, 98: 1506–9. Bakeas, E.B., Argyris, I.D. and Siskos, A.P. (2003) Carbonyl compounds in the urban environment of Athens, Greece. Chemosphere, 52: 805–13. Baker, J., Arey, J. and Atkinson, R. (2004) Kinetics of the gas-phase reactions of OH radicals, NO3 radicals and O3 with three C7 -carbonyls formed from the atmospheric reactions of myrcene, ocimene and terpinolene. Journal of Atmospheric Chemistry, 48: 241–60. Birkett, M.A., Agelopoulos, N., Jensen, K.M., et al. (2004) The role of volatile semiochemicals in mediating host location and selection by nuisance and disease-transmitting cattle flies. Medical and Veterinary Entomology, 18: 313–22. Bloss, C., Wagner, V., Jenkin, M.E., et al. (2005) Development of a detailed chemical mechanism (MCMv3.1) for the atmospheric oxidation of aromatic hydrocarbons. Atmospheric Chemistry and Physics, 5: 641–64. Bonn, B., von Kuhlmann, R. and Lawrence, M.G. (2004) High contribution of biogenic hydroperoxides to secondary organic aerosol formation. Geophysical Research Letters, 31: L10108. Bowman, J.H., Barket, D.J. Jr, and Shepson, P.B. (2003) Atmospheric chemistry of nonanal. Environmental Science and Technology, 37: 2218–25. Boy, M., Petäjä, T., Dal Maso, M., et al. (2004) Overview of the field measurement campaign in Hyytiälä, August 2001 in the framework of the EU project OSOA. Atmospheric Chemistry and Physics, 4: 657–78. Calogirou, A., Jensen, N.R., Nielsen, J.C., Kotzias, D. and Hjorth, J. (1999) Gas-phase reactions of nopinone, 3-Isopropenyl-6-oxo-heptanal, and 5-Methyl-5-vinyltetrahydrofuran-2-ol with OH, NO3 , and Ozone. Environmental Science and Technology, 33: 453–60. CATOME-Final Report and Dye, C.M. (ed.) (1999) CEC Contract No. ENV4-CT-97-0416-CATOME, NILU, Lillestrom, Norway. Cecinato, A., Di Palo, V., Mabilia, R. and Possanzini, M. (2001) Pentafluorophenylhydrazine as a coating reagent for the HRGC-MS determination of semi-volatile carbonyl compounds in air. Chromatographia, 54: 263–7.


335

Cecinato, A., Yassaa, N., Di Palo, V. and Possanzini, M. (2002) Observation of volatile and semivolatile carbonyls in an Algerian urban environment using dinitrophenylhydrazine/silica-HPLC and pentafluorophenylhydrazione/silica-GC-MS. Journal of Environmental Monitoring, 4: 1–7. Chung, S.H. and Seinfeld, J.H. (2002) Global distribution and climate forcing of carbonaceous aerosols. Journal of Geophysical Research, 107: 4407. Ciccioli, P., Brancaleoni, E., Cecinato, A., Sparapani, R. and Frattoni, M. (1993b) Identification and determination of anthropogenic volatile organic compounds in forest areas of northern and southern Europe and a remote site of the Himalaya region by high resolution gas chromatography-mass spectrometry. Journal of Chromatography A, 643: 55–69. Ciccioli, P., Brancaleoni, E., Frattoni, M., Cecinato, A. and Brachetti, A. (1993a) Ubiquitous occurrence of semi-volatile carbonyl compounds in tropospheric samples and their possible source. Atmospheric Environment, 27A: 1891–901. Ciccioli, P., Cecinato, A., Brancaleoni, E., Brachetti, A., Frattoni, M. and Sparapani, R. (1994) Composition and distribution of polar and non-polar VOCs in urban, rural, forest and remote areas. In: A. Angeletti and G. Restelli (Eds) Proceedings of the sixth European Symposium, EUR Report15609 EN, European Communities, Brussels, pp. 549–67. Ciccioli, P., Brancaleoni, E. and Frattoni, M. (1999) Reactive hydrocarbons in the atmosphere at urban and regional scales. In: C.N. Hewitt (Ed.) Reactive Hydrocarbons in the Atmosphere. Academic Press, San Diego, CA, pp. 159–207. Ciccioli, P., Brancaleoni, E., Frattoni, M., Cecinato, A. and Pinciarelli, L. (2001) Determination of volatile organic compounds (VOC) emitted from biomass burning of Mediterranean vegetation species. Analytical Letters, 34: 939–55. Ciccioli, P., Brancaleoni, E., Frattoni, M. and Maris, C. (2002) Sampling of atmospheric volatile organic compounds (VOCs) with sorbent tubes and their analysis by GC-MS. In: F.R. Burden, I. McKelvie, U. Forstner and A. Guenther (Eds) Environmental Monitoring Handbook. McGraw-Hill, New York, pp. 21.1–85. Ciccioli, P., Cecinato, A., Brancaleoni, E. and Frattoni, M. (1996) Occurrence of oxygenated volatile organic compounds (VOC) in Antartica. International Journal of Analytical Chemistry, 62: 245–53. Ciccioli, P., Cecinato, A., Brancaleoni, E. and Frattoni, M. (1992) Use of adsorption traps combined with high-resolution gas chromatography-mass spectrometry for the analysis of polar and non−C14 volatile organic compounds involved in photochemical smog formation. Journal of polar C4 − High Resolution Chromatography and Chromatography Communications, 15: 75–84. Ciccioli, P., Fabozzi, C., Brancaleoni, E., et al. (1997) Use of the isoprene algorithm for predicting the monoterpene emission from the Mediterranean holm oak Quercus ilex L.: Performances and limits of this approach. Journal of Geophysical Research, 102: 23319–28. Claeys, M., Graham, B., Vas, G., et al. (2004) Formation of secondary organic aerosols through photooxidation of isoprene. Science, 303: 1173–6. Cleary, P.A., Murphy, J.G., Wooldridge, P.J., et al. (2005) Observations of total alkyl nitrates within the Sacramento urban plume. Atmospheric Chemistry and Physics Discussion, 5: 4801–43. D’Anna, B., Andresen, O., Gefen, Z. and Nielsen, C.L. (2001) Kinetic study of OH and NO3 radical reactions with 14 aliphatic aldehydes. Physical Chemistry Chemical Physics, 3: 3057–63. Day, D.A., Dillon, M.B., Wooldridge, P.J., et al. (2003) On alkyl nitrates, O3, and the ‘missing NOy ’. Journal of Geophysical Research, 108 D16, 4501. Derwent, R.G. and Jenkin, M.E. (1991) Hydrocarbons and the long range transport of ozone and PAN across Europe. Atmospheric Environment, 25A: 1661–78. Derwent, R.G., Jenkin, M.E. and Saunders, S.M. (1996) Photochemical ozone creation potentials for a large number of reactive hydrocarbons under European conditions. Atmospheric Environment, 25A: 181–99.

336


Di Lorenzo, A., Polletta, A., Ciccioli, P., Brancaleoni, E. and Cucinato, A. (1991) Emission of PAH from a light-duty diesel car as a function of the engine operating conditions: A possible approach for investigating the parameters affecting formation of toxic components in soot. In: M. Cooke, K. Loening and J. Merritt (Eds) Polynuclear Aromatic Hydrocarbons: Measurements, Means and Metabolism. Battelle Press, Columbus, OH, pp. 239–58. El-Sayed, A.M. (2005) The pherobase: Database of insect pheromones and semichemicals. http:/www. Pherobase/compound/15 November 2005. Ezquerro, O., Pons, B. and Tena, M.T. (2002) Development of a headspace solid-phase microextraction–gas chromatography–mass spectrometry method for the identification of odour-causing volatile compounds in packaging materials. Journal of Chromatography A, 963: 381–92. Fall, R. (1999) Biogenic emissions of volatile organic compounds from higher plants. In: C.N. Hewitt (Ed.) Reactive Hydrocarbons in the Atmosphere. Academic Press, San Diego, CA, pp. 41–96. Feng, Y., Wen, S., Chen, Y., et al. (2005) Ambient levels of carbonyl compounds and their sources in Guangzhou, China. Atmospheric Environment, 39: 1789–800. Feng, Y., Wen, S., Wang, X., et al. (2004) Indoor and outdoor carbonyl compounds in the hotel ballrooms in Guangzhou, China. Atmospheric Environment, 38: 103–12. Finalyson-Pitts, J.B. and Pitts, J.N. (2000) Chemistry of the Upper and Lower Atmosphere-Theory, Experiments and Applications. Academic Press, San Diego, CA. Fruekilde, P., Hjorth, J., Jensen, N.R., Kotzias, D. and Larsen, B. (1997) Ozonolysis at vegetation surface: A source of acetone, 4-oxopentanal, 6-methyl-5-heptene-2-one and geranyl acetone in the troposphere. Atmospheric Environment, 32: 1893–902. Gautrois, M. and Koppmann, R. (1999) Diffusion technique for the production of gas standards for atmospheric measurements. Journal of Chromatography A, 848: 239–49. Greenberg, J. and Zimmerman, P.R. (1984) Non methane hydrocarbons in remote, tropical, continental, and marine atmosphere. Journal of Geophysical Research, 89: 4767–78. Grosjean, D., Grosjean, E. and Gentler, A.W. (2001) On road emissions carbonyls from light-duty and heavy-duty vehicles. Environmental Science and Technology, 35: 45–53. Grosjean, D., Grosjean, E. and Moreira, L.F.R. (2002) Speciated ambient carbonyls in Rio de Janeiro, Brazil. Environmental Science and Technology, 36: 1389–95. Grosjean, E. and Grosjean, D. (1996) Carbonyl products from the gas-phase reaction of ozone with 1-alkenes. Atmospheric Environment, 24: 4107–13. Grosjean, E., Green, P.G. and Grosjean, D. (1999) Liquid Chromatography analysis of carbonyl (2,4-Dinitrophenyl)hydrazones with detection by diode array ultraviolet spectroscopy and by atmospheric pressure negative chemical ionization mass spectrometry. Analytical Chemistry, 70: 1851–61. Grosjean, E., Grosjean, D., Fraser, M.P. and Cass, G.R. (1996) Air quality model evaluation data for organics. 2. C1 -C14 carbonyls in Los Angeles air. Environmental Science and Technology, 36: 1389–95. Guenther, A. (1999) Modeling biogenic volatile organic compound emissions to the atmosphere. In: C.N. Hewitt (Ed.) Reactive Hydrocarbons in the Atmosphere. Academic Press, San Diego, CA, pp. 97–118. Hellén, H., Hakola, H., Reissell, A. and Ruuskanen, T.M. (2004) Carbonyl compounds in boreal coniferous forest air in Hyytiala, Southern Finland. Atmospheric Chemistry and Physics, 4: 1771–80. Helmig, D. and Greenberg, J.P. (1994) Automated in situ gas chromatographic-mass spectrometric analysis of ppt level volatile organic trace gases using multistage solid adsorbing trapping. Journal of Chromatography A, 677: 123–32. Helmig, D. (1997) Ozone removal techniques in the sampling of atmospheric volatile organic trace gases. Atmospheric Environment, 31: 3635–51.


337

Helmig, D., Pollock, W., Greemberg, J. and Zimmerman, P. (1996) Gas chromatography mass spectrometry analysis of volatile organic trace gases at Mauna Loa Obseervatory, Hawaii. Journal of Geophysical Research, 101: 14697–710. Ho, K.F., Lee, S.C., Louie, P.K.K. and Zou, S.C. (2002) Seasonal variation of carbonyl compound concentrations in urban area of Hong Kong. Atmospheric Environment, 36: 1259–65. Ho, S.S.A. and Yu, J.Z. (2004) Determination of airborne carbonyls: Comparison of a thermal desorption/GC method with the standard DNPH/HPLC method. Environmental Science and Technology, 38: 862–70. Iinuma, Y., Boge, O., Miao, Y., Siero, B., Gnauk, T. and Herrmann, H. (2005) Laboratory studies on secondary organic aerosol formation from terpenes. Faraday Discussions, 130: 279–94. Isidorov, V.A. (1992) Non-methane hydrocarbons in the atmosphere of boreal forests: Composition, emission rates, estimation of regional emission and photocatalytic transformation. Ecology Bulletins, 42: 71–6. Lelieveld, J., Crutzen, P.J., Ramanathan, V., et al. (2001) The Indian Ocean experiment: Widespread air pollution from South and Southeast Asia. Science, 291: 1031–6. Jang, M., Czoschke, N.M., Lee, S. and Kamens, R.M. (2002) Heterogeneous atmospheric aerosol production by acid-catalyzed particle-phase reactions. Science, 298: 814–17. Jensen, B., Wolkoff, P. and Willkins, C.K. (1995) Characterization of linoleum. Part 2: Preliminary odor evaluation. Indoor Air, 5: 44–9. Jurvelin, J.A., Edwards, R.D., Vartainen, M., Pasanen, P. and Jantunen, M. (2003) Residential indoor, outdoor, and workplace concentrations of carbonyl compounds: Relationships with personal exposure concentrations and correlation with sources. Journal of the Air and Waste Management Association, 53: 560–73. Karl, T., Crutzen, P.J., Mandl, M., et al. (2001) Variability-lifetime relationship of VOCs observed at the Sonnblick Observatory 1999 Estimation of HO-densities. Atmospheric Environment, 35: 5287–300. −C10 monoKawamura, K., Steinberg, S. and Kaplan, I.R. (2000) Homologous series of C1 − −C6 carbonyls in Los Angeles air and motor vehicle exhaust. Atmospheric carboxylic acids and C1 − Environment, 34: 4175–91. Kean, A.J., Grosjean, E., Grosjean, D. and Harley, A.R. (2001) On road emissions carbonyls from light-duty and heavy-duty vehicles. Environmental Science and Technology, 35: 45–53. Kesselmeier, J., Bode, K., Hofmann, U., et al. (1997) Emission of short chained organic acids, aldehydes and monoterpenes from Quercus ilex L. and Pinus Pinea L. in relation to physiological activity, carbon budget and emission algorithms. Atmospheric Environment, 31 (SI): 119–34. Kesselmeier, J., Kuhn, U., Rottembeger, S., et al. (2002) Concentrations and species composition of atmospheric volatile organic compounds (VOCs) as observed during the wet and dry season in Rondonia (Amazonia). Journal of Geophysical Research, 107: D20, 8053. Kesselmeier, J. and Staudt, M. (1999) Biogenic Volatile Organic Compounds (VOC): An overview on emission, physiology and ecology. Journal of Atmospheric Chemistry, 33: 23–88. Kessler, A. and Baldwin, I.T. (2001) Defensive function of herbivore-induced plant volatile emissions in nature. Science, 291: 2141–3. Kirstine, W., Galbally, I., Ye, Y. and Hooper, M. (1998) Emissions of volatile organic compounds (primarily oxygenated species) from pasture. Journal of Geophysical Research, 103: 10605–19. Knöppel, H. and Schauenburg, H. (1989) Screening of household products for the emission of volatile organic compounds. Environment International, 15: 413–18. Kolloker, S., Oehme, M. and Dye, C. (1998) Structure elucidation of 2-dinitrophenylhydrazone derivatives of carbonyl compounds in ambient air by HPLC-MS and multiple MS-MS using atmospheric chemical ionization in negative ion mode. Analytical Chemistry, 70: 1979–85. König, G., Brunda, M., Puxbaum, H., Hewitt, C.N., Duckham, S.C. and Rudolph, J. (1995) Relative contribution of oxygenated hydrocarbons to the total biogenic VOC emissions of

338


selected mid-European agricultural and natural plant species. Atmospheric Environment, 29: 861–74. Kotzias, D. and Hjorth, J. (1991) Sampling and analysis of selected volatile organic compounds (VOC) relevant to the formation of photochemical oxidants. Pollution Atmospherique, 131, numéro special: 209–16. Larsen, B., Bomboi-Mingarro, T., Brancaleoni, E., et al. (1997) Sampling and analysis of terpenes in air. An interlaboratory comparison. Atmospheric Environment, 31 (SI): 35–49. Larsen, B.R., Tudos, A., Slanina, J., Van der Borg, K. and Kotzias, D. (2001) Quantification of airborne fossiland biomass carbonylic carbon by combined radiocarbon and liquid chromatography mass spectrometry. Atmospheric Environment, 35: 5695–707. Lichtenthaler, H.K. (1999) The 1-deoxy-D-xylulose-5-phosphate pathway of isoprenoid biosynthesis in plants. Annual Review of Plant Physiology and Plant Molecular Biology, 50: 47–65. Loreto, F., Ciccioli, P., Brancaleoni, E., Cecinato, A., Frattoni, M. and Sharkey, T. (1996) Different sources of reduced carbon contribute to form three classes of terpenoid emitted by Quercus ilex L. leaves. Proceedings of the National Academy of Science USA, 93: 9966–9. Matsunaga, S., Mochida, M. and Kawamura, K. (2003) Growth of organic aerosols by biogenic semivolatile carbonyls in the forestal atmosphere. Atmospheric Environment, 37: 2045–50. Matsunaga, S., Mochida, M. and Kawamura, K. (2004) High Abundance of gaseous and particulate 4-oxopentanal in the forestal atmosphere. Chemosphere, 55: 1143–7. Matsunaga, S., Mochida, M. and Kawamura, K. (2004) Variation in the atmospheric concentration of biogenic carbonyl compounds and their removal processes in the Northern forest at Moshiri, Hokkaido Island Japan. Journal of Geophysical Research, 109: D04302. McClenny, W.A., Colon, M. and Oliver, K.D. (2001) Ozone reaction with n-aldehydes (n = 4–10), benzaldehyde, ethanol, isopropanol, and n-propanol adsorbed on a dual-bed graphitized carbon– carbon molecular sieve adsorbent cartridge. Journal of Chromatography A, 929: 89–100. McClenny, W.A., Daughtrey, E.H. Jr, Adams, J.R., Oliver, K.D. and Kronmiller, K.G. (1998) Volatile organic compound concentration patterns at the New Hendersonville monitoring site in the 1995 Southern Oxidants Study in the Nashville, Tennessee, area. Journal of Geophysical Research, 103: 22509. Micha, S.G., Stammel, J. and Holler, C. (1993) 6-Methyl-5-heptene-2-one, a putative sex and spacing pheromone of the aphid hyperparasitoid, Alloxysta victrix (Hymenoptera: Alloxystidae). European Journal of Entomology, 90: 439–42. Moortgat, G.K. (2001) Important photochemical processes in the atmosphere. Pure and Applied Chemistry, 73: 487–90. Moortgat, G.K., Grossmann, D., Boddemberg, A., et al. (2002) Hydrogen peroxide, organic peroxides and higher carbonyl compounds determined during the BERLIOZ campaign. Journal of Atmospheric Chemistry, 42: 443–63. Muller, K. (1997) Determination of aldehydes and ketones in the atmosphere: A comparative long time study at an urban and a rural site in Eastern Germany. Chemosphere, 35: 2093–106. Muller, K., van Pinxteren, D., Plewka, A., et al. (2005) Aerosol characterisation at the FEBUKO upwind station Goldlauter (II): Detailed organic chemical characterization. Atmospheric Environment, 39: 4219–31. OSOA-Final report and Hoffmann, T. (Ed.) (1999) Origin and Formation of Secondary Organic Aerosol. CEC Contract No. EVK2-CT-1999-00016, ISAS, Dortmund, Germany. Owen, S., Boissard, C., Street, R.A., Duckham, C., Csiky, O. and Hewitt, C.N. (1997) Screening of 18 Mediterranean plant species for volatile organic compound emissions. Atmospheric Environment, 31 (SI): 101–18. Parker, T., Dottridge, J. and Kelly, S. (2002) Investigation of the Composition and Emissions of Trace Components in Landfill Gas. R&D Technical Report P1-438/TR Environment Agency, Bristol, UK.


339

Peck, G.E. (2004) Potential Semiochemicals of Wheat (Triticum aestivum L.) Induced by Oviposition and Feeding of the Wheat Stem Sawfly Cephus Cinctus Norton (Hymenoptera: Cephidae). Master of Science Thesis, Montana State University, Bozeman, MO. Pires, M. and Carvalho, L.R.F. (1998) An artifact in air carbonyls sampling using C18 DNPH-coated cartridge. Analytica Chimica Acta, 367: 223–31. Plagens, H. (2001) Untersuschungen zum atmosparenchemischen Abbau langkettiger Aldehyde, Doctorat Thesis in Chemistry, Bergischen Universitat Gesamthochschule Wuppertal, Wuppertal, available on-line at: http://elpub.bib.uniwuppertal.de/edocs/dokumente/fb09/diss2001/ plagens;internal&action = buildframes.action. Possanzini, M., Di Palo, V., Brancaleoni, E., Frattoni, M. and Ciccioli, P. (2000) A train of DNPH-coated cartridges for the determination of carbonyls from C1 to C12 in air and emission samples. Atmospheric Environment, 34: 5311–18. Purdue, L.J., Reagan, J.A., Lonnemmand, W.A., et al. (1992) Atlanta Ozone Precursor Study. Report. EPA/ 600-R-92/157, U.S. Environmental Protection Agency, Research Triangle Park, NC. Puxbaum, H. (1997) Biogenic emissions of alcohols, ester, ether and higher aldehydes. In: G. Helas, S. Slanina and R. Steinbrecher (Eds) Biogenic Volatile Organic Compounds in the Atmosphere– Summary of Present Knowledge. SPB Academic Publishers, Amsterdam, The Netherlands, pp. 79–99. Quiroz, A., Wadhams, L.J., Niemeyer, H.M., Pettersson, J. and Pickett, J.A. (1997) Semiochemical mediating spacing behavior of bird cherry-oat aphid, Rhopalosiphum padi, feeding on cereals. Journal of Chemical Ecology, 23: 2599–607. Rainer, F. and Obermeier, A. (1999) Anthropogenic emissions of volatile organic compounds. In: C.N. Hewitt (Ed.) Reactive Hydrocarbons in the Atmosphere. Academic Press, San Diego, CA, pp. 1–39. Ramanathan, V., Crutzen, P.J., Kiehl, J.T. and Rosenfeld, D. (2001) Aerosols, climate, and the hydrological cycle. Science, 294: 2119–24. Reissell, A. and Arey, J. (2001) Biogenic volatile organic compounds at Azusa and elevated sites during the 1997 Southern California ozone study. Journal of Geophysical Research, 106: 1607–21. Reissell, A., Herry, C., Aschmann, S.M., Atkinson, R. and Arey, J. (1999) Formation of acetone from the OH-radical- and O3 -initited reactions from a series of monoterpenes. Journal of Geophysical Research, 104: 13869–79. Reissell, A., Arey, J. and Atkinson, R. (2001) Atmospheric chemistry of camphor. International Journal of Chemical Kinetics, 33: 56–63. Roberts, J.M., Fehsenfeld, F.C., Albritton, D.L. and Sievers, R.E. (1984) Sampling and analysis of monoterpene hydrocarbons in the atmosphere with Tenax gas chromatographic porous polymer. In: L.H. Keith (Ed.) Identification and Analysis of Organic Pollutants in Air. Butterworth: Wooburn, pp. 371–87. Roberts, J.M., Williams, J., Baumann, K., et al. (1998) Measurements of PAN, PPN and MPAN made during the 1994 and 1995 Nashville intensives of the southern oxidant study: Implications for regional ozone production from biogenic hydrocarbons. Journal of Geophysical Research, 103: 22473–90. Rudich, Y. (2003) Laboratory perspectives on the chemical transformations of organic matter in atmospheric particles. Chemical Reviews, 103: 5097–124. Sagebiel, J.C., Zielinska, B., Pierson, W.R. and Gertler, A.W. (1996) Real world emissions and calculated reactivities of organic species from motor vehicles. Atmospheric Environment, 12: 2287–844. Sartin, J.H., Halsall, C.J., Davison, B., Owen, S. and Hewitt, N.C. (2001) Determination of biogenic −C16 ) in the coastal atmosphere of Mace Head, Ireland. Analytica volatile organic compounds (C8 − Chimica Acta, 428: 61–72.

340


Satsumabayashi, H., Kurita, H., Chang, Y.S., Carnichael, M.R. and Ueda, H. (1995) Photochemical formation of lower aldehydes and lower fatty acids under long-range transport in central Japan. Atmospheric Environment, 29: 255–66. Schauer, J.J., Kleeman, M.J., Cass, G.R. and Simoneit, B.R.T. (2002a) Measurement of emissions −C32 organic compounds from gasoline-powered vehicles. from air pollution sources. 5. C1 − Environmental Science and Technology, 36: 1169–80. Schauer, J.J., Kleeman, M.J., Cass, G.R. and Simoneit, B.R.T. (2002b) Measurement of emissions from −C27 organic compounds from cooking with seed oils. Environmental air pollution sources. 4. C1 − Science and Technology, 36: 567–75. Schauer, J.J., Kleeman, M.J., Cass, G.R. and Simoneit, B.R.T. (2001) Measurement of emissions −C29 Organic compounds from fireplace combustion wood. from air pollution sources. 3. C1 − Environmental Science and Technology, 35: 1716–28. Schauer, J.J., Kleeman, M.J., Cass, G. and Simoneit, B.T. (1999a) Measurement of emissions from air pollution sources. 2. C1 through C30 organic compounds from medium duty diesel trucks. Environmental Science and Technology, 33: 1578–87. Schauer, J.J., Kleeman, M.J., Cass, G. and Simoneit, B.T. (1999b) Measurement of emissions from air pollution sources. 1. C1 through C29 organic compounds from meat charbroiling. Environmental Science and Technology, 33: 1566–77. Schlomski, S., Kibler, M., Ebert, P., et al. (1997) Multiphase chemistry of organic acids and carbonyl compounds. In: B. Larsen, B. Versino and G. Angeletti (Eds) Proceedings of the seventh European Symposium Physico-Chemical Behaviour of Atmospheric Pollutants, EUR 17428 EN, European Communities, Brussels, pp. 557–61. Shaughnessy, R.J., McDaniels, T.J. and Weschler, C.J. (2001) Indoor chemistry: Ozone and volatile organic compounds found in tobacco smoke. Environmental Science and Technology, 35: 2758–64. Smith, A.M., Rigler, E., Kwok, E.S.C. and Atkinson, R. (1996) Kinetics and products of the gas-phase reactions of 6-methyl-5-hepten-2-one and trans-cinnamaldehyde with OH and NO3 radicals and O3 at 296◦ K. Environmental Science and Technology, 30: 1781–5. Solberg, S. (2005) VOC measurements 2003. EMEP/CCC-Report 10/2005, NILU, Kjeller, Norway. Speijers, G.J.A. (1993) VOCs and the environment and public health-health effects. In: Th.H.J. Bloemen and J. Burn (Eds) VOCs in the Environment. Chemistry and Analysis. Blackie Academical & Professional: London, pp. 24–91. Staudt, M., Bertin, N., Hansen, U., et al. (1997) Seasonal and diurnal patterns of monoterpene emissions from Pinus pinea (L.) under field conditions. Atmospheric Environment, 31 (SI): 145–6. Tadic, M.J., Juranic, O.I. and Moortgat, G.K. (2002) Photooxidation of n-heptanal in air: Norrish type I and II processes and quantum yield total pressure dependency. Journal of Chemical Society Perkin Transaction, 2: 135–40. Tillman, A.J., Seybold, S.J., Jurenka, R.A. and Blomquist, G.J. (1999) Insect pheromones – an overview of biosynthesis and endocrine regulation. Insect Biochemistry and Molecular Biology, 29: 481–514. Tolocka, M.P., Myoseon, J., Ginter, J.M., Cox, J.F., Kamens, R.M. and Johnston, M.V. (2004) Formation of oligomers in secondary organic aerosols. Environmental Science and Technology, 38: 1428–34. UK PORG (1994) Ozone in the United Kingdom 1993. Third Report of the United Kingdom Photochemical Oxidants Review Group. Air Quality Division, Department of the Environment, London. Wadia, Y., Tobias, D.J., Stafford, R. and Finlayson-Pitts, B.J. (2000) Real-time monitoring of the kinetics and gas-phase products of the reaction of ozone with 1-oleoyl-2-palmitoyl-sn-glycero-3phosphocholine at the air/water interface. Langmuir, 16: 9321–30. Wildt, J., Kobel, K., Schuh-Thomas, G. and Heiden, C. (2003) Emissions of oxygenated volatile organic compounds from plants Part II: Emissions of saturated aldehydes. Journal of Atmospheric Chemistry, 45: 173–96.


341

Winer, A., Arey, J., Atkinson, R., et al. (1992) Emission rates of of organics from from vegetation in California’s central valley. Atmospheric Environment, 26A: 2647–59. Yassaa, N., Meklati, B.Y., Brancaleoni, E., Frattoni, M. and Ciccioli, P. (2001) Polar and non-polar volatile organic compounds (VOCs) in urban Algiers and saharian sites of Algeria. Atmospheric Environment, 35: 787–801. Yasuhara, A. and Shibamoto, T. (1995) Quantitative analysis of volatile aldehydes formed from various kinds of fish flesh during heat treatment. Journal of Agricultural and Food Chemistry, 43: 94–7. Yokouchi, Y., Mukai, H., Makajima, K. and Ambe, Y. (1990) Semi-volatile aldehydes as predominant organic gases in remote areas. Atmospheric Environment, 24: 439–42. Zhang, J., He, Q. and Lioy, P.J. (1994) Characteristics of aldehydes: Concentrations, sources and exposure for indoor and outdoor residential microenvironments. Environmental Science and Technology, 28: 146–52. Zhou, X. and Mopper, K. (1997) Photochemical production of low-molecular-weight carbonyl compounds in seawater and surface microlayer and their air-sea exchange. Marine Chemistry, 56: 201–13.


Chapter 9

Organic Aerosols Thorsten Hoffmann and Jörg Warnke

9.1

Introduction

Atmospheric aerosols interact both directly and indirectly with the Earth’s radiation budget and climate. As a direct effect, the aerosols scatter or absorb sunlight. As an indirect effect, aerosols in the lower atmosphere can modify number and size of cloud droplets, changing how the clouds reflect and absorb sunlight, thereby affecting the Earth’s radiation budget. Aerosols also can act as sites for chemical reactions to take place (heterogeneous chemistry). Hence, they play an important role in global climate and atmospheric chemistry. Furthermore, atmospheric aerosols affect our environment at the local and regional levels. Aerosols are now becoming recognised as a significant health problem, especially in regard to respiratory diseases. The formation of organic aerosols from the oxidation of hydrocarbons is only one but important pathway that determines the overall composition of atmospheric aerosols. In general, the volatile aerosol precursors are first degraded in the gas phase by bimolecular reactions with radicals or ozone or by photolysis, followed by the formation of products with a lower volatility. These products are higher functionalised compounds with hydroxyl, carbonyl, carboxyl groups or groups containing heteroatoms, which will either condense on existing particles or even form new aerosol particles. To distinguish this fraction of tropospheric aerosols from the direct input of particulate organics into the atmosphere, it is specified as secondary organic aerosol (SOA). Since primary and SOA particles are closely linked in several aspects about their potential role and significance in atmospheric chemistry as well as their analytical treatment, we consider both primary and secondary contributions in this chapter. However, emphasis lies on the aerosol components from hydrocarbon oxidation.

9.1.1

Historical

The optical effects of atmospheric aerosol particles formed from the oxidation of volatile organic compounds (VOCs) were recognised quite early by humans. Sha-co-na-qe ‘Place of Blue Smoke’ was the name given for the Great Smoky Mountains by the Cherokee Indians, of course, without knowing the cause of the haziness often observed in the summertime above the forested Appalachian Highlands. This phenomenon of forested regions – that they are frequently enveloped in a blue haze or smoke – is created by Rayleigh scattering of submicrometre particles. Numerous other forested mountain sites are named after this

Organic Aerosols

343

phenomenon: ‘Blue Mountains’, ‘Blue Ridges’ or ‘Smoky Mountains’ can be found, for example, in India close to the Burman border, in Jamaica, at the east coast of Australia and even several times in North America (Montana, Oregon, Idaho, Maine, Pennsylvania, Tennessee). The first connection between VOCs and the formation of atmospheric particles was probably made by Arie Haagen-Smit at Caltech in 1952 in a strongly anthropogenically influenced environment. Studying various aspects of the Los Angeles smog formation, he not only explained ozone and peroxide formation by the photochemistry of the released hydrocarbons and nitrogen oxides, but also related the decrease in visibility during smog episodes to the condensation of aldehydes and acids formed in the oxidation of the hydrocarbons. In 1960, F.W. Went, director of the Missouri Botanical Garden and former colleague of Haagen-Smit at Caltech, published an extensive article in Nature titled ‘Blue hazes in the atmosphere’. Based on his observations while staying in countryside and on his everyday experiences, as well as his knowledge about secondary plant products, he finally also connected the occurrence of the natural phenomena with the volatilisation and gas-phase oxidation of terpenes from terrestrial vegetation. However, during these first years of atmospheric chemistry and the following decades, the main interest in the VOC chemistry was focused on gas-phase photochemistry. At about 1990, increasing interest arose to understand the aerosol formation behaviour of hydrocarbons, driven by the awareness of the role of natural and anthropogenic aerosols in the radiative properties of the atmosphere and the Earth’s climate. Moreover, the ozone hole research in the Antarctic clearly showed that heterogeneous reactions on surfaces of air suspended matter can influence gas-phase processes. Another major driver to investigate the origin and formation of aerosol particles in the last decades has been their effect on human health. It has been shown that cardio-pulmonary diseases and mortality are related to the presence of fine particulate matter (Dockery et al. 1993; Laden et al. 2000; Mar et al. 2000; Tsai et al. 2000). As a consequence of the increasing scientific interest to understand aerosol-forming atmospheric processes, new instrumental techniques for particle analysis have been developed at a rapid rate in order to produce methods with lower detection limits, shorter temporal resolution and increased selectivity. Especially mass spectrometric online techniques have been developed during the last years, as reviewed by Sipin et al. (2003).

9.1.2

Sources and sinks of atmospheric particles

Particles in the atmosphere are divided into primary and secondary particles according to their formation processes. Primary particles are released directly into the atmosphere, whereas secondary particles are produced within the atmosphere as a consequence of the conversion of volatile precursors into low- or non-volatile substances that form new particles. Formation processes of primary particles are basically mechanical production (abrasion, suspension and sea spray) and production during combustion processes (condensation of hot vapours or formation inside flames, such as soot particles) (Seinfeld and Pandis 1998). In general, mechanical processes create coarse particles whereas combustion processes create fine particles. Particles are called “fine particle” if their diameter is below 1 or 2.5 μm, respectively, depending on the

344


Coarse particles

Fine particles

Primary (mechanical production)

Primay (combustion) Secondary (nucleation and condensational growth) Nucleation (Molecules)

0.1 nm

Cluster

1 nm

Condensational growth Nucleation Aitken Accumulation mode mode mode 10 nm

100 nm

1 μm

Coarse mode 10 μm 100 μm Particle diameter

Figure 9.1 Size modes of atmospheric particles and important sources.

definition. In atmospheric measurements the considered size class of atmospheric aerosols is often described by the maximum diameter of sampled particles. A size class is then characterised by e.g. PM10 which means the upper diameter of sampled “particulate matter” is 10 μm. This term will be used frequently in this chapter (refer also to Section 9.3.1). Secondary atmospheric particles belong to the fine particle fraction as well; they are created by the so-called ‘nucleation process’. The nucleation process in the troposphere is currently not completely understood, and several mechanisms are discussed (Kulmala 2003). One hypothesis about new particle production in the atmosphere assumes that the process is initialised by the formation of sulphuric-acid-containing clusters (thermodynamically stable clusters) in the size range of 1 nm, which grow under suitable conditions, for example, the availability of condensable vapours, into a size range of 3–20 nm, the so-called nucleation mode. If the concentration of condensable vapours is not high enough, the clusters will be rapidly lost by coagulation and no new particles will be formed. However, once formed the nucleation mode particles can continue to grow by uptake of condensable vapours into the Aitken mode (around 20 to 100 nm) and further to particles in the accumulation mode (100-nm range) (Kulmala et al. 2004). Condensable vapours mean low volatile compounds, which are produced during chemical reactions in the atmosphere from volatile precursors. Due to the Kelvin-Effect (smaller droplets have higher vapour pressures), these vapours cannot condense without a condensation nuclei (e.g. a cluster). Condensable vapours might be inorganic like sulphuric acid or organic like low volatile products of the terpene oxidation. These low volatile compounds are not only involved in the formation or growth of secondary particles but they can also condense onto pre-existing particles, leading to increased particle size and mass and to an alteration of the chemical composition. Changes in the chemical composition also alter the physical or physicochemical properties of the particles (light scattering, hygroscopicity, etc.), which affect the impact of aerosols on climate. Figure 9.1 illustrates size modes and sources of atmospheric particles. Processes that act as sinks of atmospheric particles are dependent on the particle sizes. Coarse particles are removed mainly by dry deposition, that is, by sedimentation. Particles in the accumulation mode are eliminated mostly by wet deposition (rainout, washout). The main sink for smaller particles (Aitken and nucleation mode) is by coagulation with other particles.

Organic Aerosols

9.1.3

345

General chemical composition of aerosol particles and aerosol mass in the troposphere

Particles can be characterised by their origin: soot from combustion, mineral dust from levitated crustal material, salt from sea spray, pollen from vegetation and so on. By chemical analysis, the source characteristics of the different particles are replaced by the chemical composition of the whole sample. One approach to characterise the aerosol sample by its chemical composition is to divide the aerosol mass into several compound classes, for example, organic matter (OM), elemental carbon (EC), non-sea-salt (NSS) sulphate, nitrate, ammonia, sea salt, mineral dust (Hueglin et al. 2005; Putaud et al. 2004; Rees et al. 2004; Sellegri et al. 2003; Ward et al. 2004). EC, nitrate and ammonium are defined chemical components whose concentration can be measured. In contrast, NSS-sulphate, sea salt, OM and mineral dust are no single chemical compounds, and, therefore, they have to be calculated from the measured content of selected tracer species. Sea salt can be estimated using the measured concentrations of Na+ and Cl− or Na+ and SO2− 4 and a standard sea water composition. NSS-sulphate is the difference between measured sulphate and calculated fraction of sea salt sulphate. OM can be estimated from measured organic carbon (OC) by multiplying with a certain factor (often 1.4, refer to Section 9.2.3) representing the elemental composition of organic substances present in the atmosphere. Metal species are used as tracers to calculate the mass of particulate minerals (mineral dust) suspended in the atmosphere. Mineral dust (sometimes also called crustal material) consists of oxides, silicates, SiO2 or other minerals containing iron, aluminium, calcium, sodium, potassium and other metal ions. Nitrate and NSS-sulphate are mainly secondary aerosol constituents. Nitrate is formed mostly from anthropogenic NOx -emissions by atmospheric oxidation to nitric acid. Sulphate is formed from SO2 -emissions (predominantly anthropogenic) or emissions of sulphur-containing VOCs, like dimethyl sulphide from marine environments, by oxidation to sulphuric acid (Finlayson-Pitts and Pitts 2000). The predominance of primary aerosol constituents in the coarse and secondary aerosol constituents in the fine-particle fraction can be seen in Figure 9.2 comparing the composition of the fine and coarse fraction at different locations in Europe. The overall mass of the aerosol can be measured directly by weighing. Another method is the calculation of mass from the measured particle size distribution. The size distribution can be used to calculate aerosol volume and, supposing a certain mean density for aerosol particles, aerosol mass can be estimated. Aerosol mass is in the μg/m3 range, depending strongly on the sampling site, meteorological conditions, seasonal and diurnal cycles. The annual mean concentration of PM10 , for example, in locations in Scandinavia, is below 10 μg/m3 , whereas concentrations in urban sites in central Europe can reach mean concentrations of >50 μg/m3 . Also, the aerosol-relative composition strongly depends on location and time.

9.2

Carbonaceous aerosols

The term ‘carbonaceous aerosols’ includes all aerosol constituents that are based on carbon, for example, the variety of different organic compounds, EC, bioaerosols and inorganic constituents. Except inorganic carbon, the characteristics of these different carbonaceous

346


Atmospheric concentration (μg/m3)

40 35

unacc.

30

NO3−

25

NH4+ NSS-SO42 −

20

OM

15

BC

10

Sea salt min. dust

5 0 Fine Coarse Natural

Fine Coarse Rural

Fine Coarse Urban

Fine Coarse Kerbside

Figure 9.2 Annual mean composition of fine (PM2.5 ) and coarse (PM2.5 –PM10 ) aerosol fraction at a natural (Sevettijarvi (FIN)), rural (Illmittz (A)), urban (Zuerich (CH)) and kerbside (Barcelona (E)) site. Abbreviations: unacc. = unacccounted mass, NSS = non-sea-salt, OM = organic matter, BC = black carbon, min. dust = mineral dust. Data from http://ies.jrc.cec.eu.int/Download/cc/6.xls, http://ies.jrc.cec.eu.int/Download/cc/5.xls.

fractions will be discussed in the following sections. The concentration of inorganic carbon, essentially as carbonate, is negligible on an average, at least when considering PM2.5 . If inorganic carbon occurs in atmospheric PM, it is limited to the form of carbonate-containing mineral dust, which is part of the coarse mode of PM.

9.2.1

Bioaerosols

Bioaerosols are primary organic aerosols with diameters from ∼10 nm to 100 μm, which are either alive (viruses, bacteria, fungi, algae), carry living organisms or are released from living organisms (pollen, spores, cell debris) (Ariya and Amyot 2004). Roughly, the size of bacteria is around 1 μm, pollen grains are mostly >10 μm and viruses are in the nanometer range. Each of these ‘particles’ is usually itself a complex mixture of various molecules. Bacteria may spread diseases, can act as cloud condensation nuclei and ice nuclei. They were found even at high altitudes in the atmosphere and remote regions. Bacteria and fungi can be suspended from soil or plants by wind and from water surfaces by bubble bursting processes or sea spray. They can also be released by anthropogenic sources such as farming, waste and waste water treatment. Bacteria can even live and grow in atmospheric water droplets like fog (Fuzzi et al. 1997) or even super-cooled cloud droplets (Sattler et al. 2001). The importance of bioaerosols for the atmospheric aerosol content is very unclear. Some authors report bioaerosols as major components, whereas other studies report only insignificant contribution of bacteria to the atmospheric particulate material. In Amazonian aerosols, the nocturnal increase of coarse size particle mass (PM10 –PM2 ) was assigned to fungi (Graham et al. 2003b). During the wet season, biogenic particles accounted for

Organic Aerosols

347

55–92% of the fine-particle mass and for 65–95% of the coarse-particle mass sampled in the Amazonian basin (Artaxo et al. 1988, 1990). For Russia (Lake Baikal) and Germany (Mainz) (Jaenicke 2005), contributions of primary biological aerosol particles (including plant fragments, pollen, etc.) to total atmospheric particles (>0.2 μm) were reported to be in the range of 20–30%, respectively. On the other hand, studies from the southern ocean report that bacteria represent only about 1% of the total number of particles >0.2 μm (Posfai et al. 2003) or did not even indicate the presence of bacteria in marine or continental samples (Bates et al. 1998). In Hong Kong, the dry mass load of Gram-negative bacteria was estimated to be in the range of 10–100 ng/m3 , which is small compared to the total OM present in the atmosphere (Lee et al. 2004). Nevertheless, bacteria were even found at high altitudes (5.4 km) in the same relative (low) concentration as near to the surface. There is also evidence that microbes can alter the composition of existing aerosol particles (Ariya et al. 2002) using aerosol constituents as nutrients. Dicarboxylic acids, for example, may be transformed into volatile products by airborne microorganisms.

9.2.1.1

Measurements of bacteria and fungi

One approach to measure bacteria and fungi is to count them directly on the collection media (filter, impactor plates), using light microscopes with or without staining procedures, epifluoresence microscopy (Griffin et al. 2001), electron microscopes or electron probe X-ray microanalysis (EPXMA) (Graham et al. 2003b). The direct counting can be used for the measurement of all types of bioaerosols; staining procedures usually need specific target molecules, such as proteins. Another approach to measure and identify living cells is to collect them onto culture media and count the forming colonies. Until now, this technique is the most widely used method for monitoring bacteria. However, not all species are able to grow on these media and therefore will not be detected. The identification of microbes can be done by their morphology or based on a genetic identification using the polymerase chain reaction (PCR) (Griffin et al. 2001). Also, specific markers can be used to measure microbes in the atmosphere. Applying this approach, living as well as dead microbes will be measured. Lee and co-workers (2004) used 3-hydroxy fatty acids as biomarkers for Gram-negative bacteria in aerosol samples from Hong Kong. The 3-hydroxy fatty acids are constituents of the endotoxins of Gram-negative bacteria. These endotoxins can have strong inflammatory properties. Chemically, endotoxins are lipopolysaccharides that are located in the outer membrane of the bacteria. Measured concentrations of 3-hydroxy fatty acids can be used to estimate the endotoxin concentration and, therefore, the mass load of Gram-negative bacteria in the aerosol. Ergosterol is used as a specific tracer for fungi in various matters (house dust, building materials) and bioaerosols (Pasanen et al. 1999; Saraf et al. 1997; Szponar and Larsson 2001). For Gram-positive bacteria, muramic acid serves as a marker compound. Muramic acid is part of the peptidoglycan structure of the Gram-positive bacteria cell wall (Szponar and Larsson 2001). Fatty acids with carbon atom numbers less than 20 are sometimes also used as markers for the microbial contribution to atmospheric aerosols (Guo et al. 2003; Zheng et al. 2000). However, they are not very specific, because these substances are also emitted by anthropogenic sources (Rogge et al. 1991). The occurrence of various sugars and sugar alcohols in the atmosphere is also linked to bioaerosols; see also Section 9.2.5.

348

9.2.2


EC, black carbon and soot

The term ‘black carbon’ is often used to describe carbonaceous residues produced by incomplete combustion of OM (Fernandes et al. 2003). However, the chemical composition of these residues is not very well defined. Black carbon includes soot and other particulate residues of various combustion processes like biomass burning, coal and oil burning, diesel engines, etc., which can be usually identified by their black colour. All of these combustion residues are primary aerosol constituents. Black carbon is not equal to EC; however, it might be a good approximation. Black carbon can be measured by optical methods using the absorption of light. Actually, nowadays this light absorption (black colour) is the basic definition of ‘black carbon’ (Gelencser 2004). Soot is generated in the gas phase (in the flame) from gaseous precursors, whereas carbonaceous residues formed on the surface of solid (biomass) fuels are called chars (charcoals). Soot is a form of appearance of EC, and, as a first approximation, is similar to graphite, consisting of layers of hexagonally arranged carbon atoms. Unlike graphite, these layers can be curved, forming spherical bodies. If the order of these layers is very low, carbon is called amorphous; if it is high, it is called graphitic carbon (Wal and Tomasek 2004). Also, fullerenes and carbon nanotubes can be formed under specific combustion conditions (Bang et al. 2004; Height et al. 2004; Takehara et al. 2005). The size of soot particles depends on the history of their formation (combusted substances, temperature, oxygen content in flame, etc.). Primary soot particles (from diesel engines or oil burning) have diameters between 10 and 50 nm and a spherical onionshell structure; however, they usually agglomerate during the combustion process, forming chain-aggregates with some 100 nm to about 1 μm length (Wentzel et al. 2003). Chars are much bigger than soot particles, having diameters usually in the micrometre range. Although soot and chars consist mostly of carbon they also contain various amounts of hydrogen atoms and other atoms, as well as some extractable OM, e.g. polycyclic aromatic hydrocarbons (PAHs) or alkanes. The OM may be adsorbed onto the soot surface or even enclosed inside the primary soot particles (Fernandes and Brooks 2003). In reality, there is no sharp border between elemental and organic carbon. Starting with an organic molecule such as a low-mass PAH (e.g. naphthalene), the addition of further aromatic rings will lead to polyaromatic systems. Systems consisting, for example, of some thousand carbon atoms might finally be better described as graphitic layers and thus might be counted as EC instead of OC.

9.2.3

OC and organic matter

EC constitutes only a small fraction of the carbonaceous matter present in atmospheric aerosols. The dominating fraction of carbonaceous aerosols is the so-called organic matter (OM), which includes all organic compounds present in the particle phase. Unfortunately, it is not possible to measure the mass of OM in aerosol samples directly. An indirect method to determine the OM is to measure the so-called OC, which represents all carbon contained in organic compounds. For the estimation of organic mass, the determined OC is multiplied by a certain value that represents the mean ratio of organic mass to OC of the organic substances present in the aerosol. Early estimations often used an OM/OC ratio of 1.4 to calculate OM from measured OC. More recent publications suggest that a value of 1.4 is more likely the lower limit. It was estimated that urban aerosols are best represented by an

Organic Aerosols

349

OM/OC ratio of 1.4 to 1.8, whereas non-urban aerosols usually have higher OM/OC ratios in the range of 1.9–2.3 (Turpin and Lim 2001). The value of the OM/OC ratio depends on the individual molecular composition of the respective aerosols. Organic molecules that consist only of hydrogen and carbon have a low OM/OC ratio, for example, alkanes or PAHs. Such compounds are usually non-polar and are insoluble or only slightly soluble in water. Highly oxygenated compounds have a higher OM/OC ratio. They are often water soluble due to the presence of polar functional groups, like dicarboxylic acids or sugars. Due to the high content of non-polar constituents in urban aerosols, for example, from fossil fuel combustion sources, their OM/OC ratio is rather low. Aerosols produced or influenced by photochemical activity are usually higher oxidised and, therefore, have a higher OM/OC ratio. The same is true for aerosols from combustion of biomass, which carry oxygen-containing constituents such as levoglucosan (see Section 9.2.4.1). Oxygen is the most important element besides carbon, adding to OM; however, also other elements, such as nitrogen, sulphur, phosphorous, and so on, contribute to OM.

9.2.3.1

The OC/EC tracer method for the estimation of SOA

SOA is formed by the chemical conversation of volatile organic precursors to low volatile compounds and their subsequent condensation. The amount of SOA cannot be measured directly, because there is no possibility to separate SOA constituents from primary organic aerosol constituents. Therefore, several methods were developed to estimate the secondary organic contribution. One possible method is the mathematical modelling of SOA formation in connection with transport and deposition of SOA (Pandis et al. 1992; Strader et al. 1999). Another method is the use of a receptor model to estimate the primary organic aerosol and calculate SOA by subtracting the estimated primary organic aerosol from the measured total organic aerosol (Schauer and Cass 2000; Schauer et al. 1996). However, the mostly used method to estimate SOA is the OC/EC tracer method, which will be described subsequently. Primary organic aerosols are often linked to combustion processes, and these processes yield certain fractions of EC and OC. The ratio of OC to EC depends on several factors; however, on average, a specific OC/EC ratio, which is characteristic for the individual combustion source, can be estimated. The same counts for other primary organic aerosol sources. A certain location will be influenced usually by the same sources of primary emissions. However, these sources may change during the year, and the impact of single sources to local aerosols can vary with meteorology, for example, wind direction. If measurements of OC and EC are carried out during a longer time period, typical OC/EC ratios for primary emissions at this location can be obtained. Of course, the measurement of OC/EC ratios typical for primary emissions has to be carried out during periods in which the influence of primary emissions is dominating. On the other hand, this means that these periods should be influenced only little by photochemistry (Cabada et al. 2004; Strader et al. 1999; Turpin and Huntzicker 1995). Periods of mainly primary emissions might be identified using gaseous combustion tracers (e.g. CO, NOx ), whereas ozone can be used as an indicator of photochemical activity. Sometimes, simply the lowest measured OC/EC ratios of a time series are accounted to primary emissions (Castro et al. 1999). Also, primary emissions inventories of OC and EC were used to estimate the primary OC/EC ratio (Gray et al. 1986).

350


6 Influenced by SOA production 5

OC (μg/m3)

4 3 2 Lowest OC/EC ratio (= 0.81)

1

Mostly primary emissions

Linear regression: y = 0.71 × +0.41

0 0

0.5

1.0

1.5

2.0

2.5

3.0

3.5

EC (μg/m3) Figure 9.3 Plot of OC-concentrations vs EC-concentrations for measurements in Jülich, Germany.

If the measured OC/EC ratio exceeds the previous determined primary OC/EC ratio ((OC/EC)pri ), contribution of secondary OC to the organic aerosol is indicated. The difference between total OC (OCtot ) and primary OC (OCpri ) is secondary OC (OCsec ). This can be expressed by two simple equations (Turpin et al. 1991): OCpri = EC × (OC/EC)pri

(9.1)

OCsec = OCtot − OCpri

(9.2)

Usually, (OC/EC)pri is calculated as the slope of a linear regression from a plot of the measured OC vs. the measured EC, as indicated in Figure 9.3. The linear regression process normally results a certain value for the intercept, which can be explained by sources of primary OC that emit no EC (e.g. cooking or suspension of biological material). Therefore, Equation 9.1 has to be modified to (Turpin and Huntzicker 1995): OCpri = EC × (OC/EC)pri + b

(9.3)

where b is the intercept on the OC-axis. Typical values for (OC/EC)pri are in the range of 0.9 in urban areas to more than 3 in rural or remote areas (Table 9.1). Finally, SOA mass has to be calculated from the estimated secondary OC.

9.2.3.2

Concentrations of OC, EC and secondary OC in the ambient atmosphere

Typical mean concentrations of OC in the atmosphere are in the range of 1 μg/m3 in clean areas to >10 μg/m3 in polluted areas (refer to Table 9.2), whereas peak concentrations might reach >50 μg/m3 , for example, in the rainforest during the biomass-burning season (Artaxo et al. 2002). EC concentrations in clean areas are typically below 1 μg/m3 , in polluted areas the EC concentrations might exceed 5 μg/m3 . High primary OC/EC ratios point to a high input of primary OC from non-combustion sources (Table 9.1). As explained above, if

Table 9.1 Mean OC/EC ratios and OC/EC ratios for primary emissions (OC/EC prim.) measured at different locations. Calculated primary OC and secondary (sec.) OC concentration as well as the relative contribution of secondary OC to total OC (secondary OC (%)) are also displayed Location

Site description

Birmingham, UK1

Urban

Areao, Portugal1

Rural/Coastal

Cities, China†2

Oceanic Urban

United States3 , Northeast Central West Mira Loma, United States4 Helsinki, Finland5 Hyytiälä, Finland6 Jülich, Germany6

Time

January 1994 May 1993 May–August 1994 November–December 1993, February–March 1994 May–July 1993 2002 June–August 1999

Continental Rural/Urban Urban Remote Rural/Urban

September 2001–January 2002 July 2000–July 2001 March 2003 June 2002 July 2003

∗ In% of the total OC. † Average of different sites in the cities Hong Kong, Guangzhou, Shenzhen, Zhuhai. 1 Castro et al. (1999). 2 Cao et al. (2004). 3 Yu et al. (2004). 4 Na et al. (2004). 5 Viidanoja et al. (2002). 6 Warnke (2004).

PM

OC/EC

OC/EC prim.

Primary OC (μg/m3 )

Secondary OC (μg/m3 )

10 10 10 10

1.4 3.5 2.9–3.5 2.4–4.0

1.1 1.1 1.5 1.5

3.95 1.65 0.38 0.70

0.63 3.10 0.56 0.71

17 65 56 45

7.3 2.5 2.5 — — — 5.2 — 2.1 1.0 1.8

— 1.1 1.2 1.2 2.4 3.5 3.7 1.1 0.98 0.81 0.81

0.4 — — 0.39 0.47 0.51 6.2 — — — —

1.75 4.9 6.3 1.3 0.9 0.51 4.2 1.68 0.44 0.3 1.5

78 51 47 77 66 48 40 54 60 21 53

10 2.5 10 2.5 2.5 2.5 2.5 2.5 2.5 2.5 2.5

Secondary OC (%)∗

352


Table 9.2 Mean concentrations of OC and EC in the atmospheric fine particle phase (PM2 or PM2.5 ) measured at different locations Location

Description

Lennox, California1 Beijing, China2 Ulan-Bator, Mongolia2 Chongju, Korea3 Conroe, Texas4 Galveston, Texas4 Aldine, Texas4 Budapest, Hungary5

Urban Urban Rural Urban-city Rural Coastal Urban Kerbside (day) Urban (day)

OC (μg/m3 )

EC (μg/m3 )

6.3 12.4 2.3 5.0 2.4 3.1 4.3 6.8∗ 4.1∗

1.7 5.4 0.4 4.4 0.23 0.26 0.57 3.4∗ 0.33∗

∗ Median. 1 Turpin et al. (1991). 2 He et al. (2004b). 3 Lee and Kang (2001). 4 Russell and Allen (2004). 5 Salma et al. (2004).

the mean OC/EC ratios exceed the primary OC/EC ratios, formation of SOA is indicated. However, the measurement results of OC and EC can easily be influenced by different errors (refer to Section 9.3). Obviously, these errors also influence source attribution based on OC/EC analysis. Estimated secondary OC is in the range of below 20% to about 80% of the total OC. Its value is strongly dependent on the time of the year and on the origin of the air masses. It is less abundant, for example, in the European winter with low radiation and, therefore, low photochemical activity and simultaneously high primary emissions. Especially high secondary contributions to OC can be found during periods with strong photochemical activity (summer) and simultaneously low input of primary emissions, for example, air masses from the ocean. However, absolute concentration values of OC might be low during these periods. High absolute concentrations of secondary OC (e.g. around 5 μg/m3 in cities in China and Portugal) can be found in polluted areas with high photochemical activity. Typically, mean secondary contribution to OC seems to be around 50%, refer to Table 9.1.

9.2.3.3

C-14 measurements: A method to differentiate between biogenic and anthropogenic carbon

The measurement of the 14 C/12 C ratio of the organic fraction in aerosol samples can be used to estimate the relative contribution of anthropogenic and biogenic sources to the organic aerosol fraction (Currie 2000). 14 C works like a tracer for biogenic carbon. Due to the continuous production of radioactive 14 CO2 in the atmosphere by cosmic radiation, organic substances produced by plants have a certain 14 C/12 C ratio. Over the years, this ratio changes due to the radioactive decay of the 14 C, for example, in fossil fuels such as coal, oil or natural gas. By the measurement of the 14 C/12 C ratio in atmospheric aerosols, the contribution of sources originating from the combustion of 14 C-poor fossil fuel and the contribution of contemporary biogenic sources can be calculated. Obviously, at this point,

Organic Aerosols

353

the anthropogenic contribution to the atmospheric organic aerosol fraction is set equal to fossil fuel combustion, an assumption that is not completely valid (e.g. wood combustion – biomass burning – will be accounted for as biogenic contribution). Therefore, this technique is only effective when the combustion of biomass is not a major contributor to the regional aerosol burden. For these estimations, precise and sensitive measurements of the 14 C/12 C ratio are necessary. The measurements have to be carried out by accelerator mass spectrometry (AMS). These measurements are very complex, expensive and time consuming. Nevertheless, the results are very important since they can be directly used to differentiate between biogenic and anthropogenic contributions to the aerosol total carbon (TC). Results of these measurements indicate a substantial biogenic contribution to the organic aerosol even in urban or anthropogenic influenced regions. In south-east Texas, 27–73% of the aerosol TC collected at an urban/suburban site in summer 2000 was modern carbon (and thus biogenic). At a rural, forested site, even 44–77% of the TC was of biogenic origin (Lemire et al. 2002). In high alpine snow samples from the eastern Alps about 64% of the non-soluble carbon was of biogenic origin (Weissenbok et al. 2000). Biogenic aerosol contribution in Nashville, Tennessee and Zurich, Switzerland, ranged from 56–80% and from 51–80%, respectively (Lewis et al. 2004; Szidat et al. 2004). In a size resolved sample from Tokyo from April 2002 (Endo et al. 2004), the biogenic contribution ranged from about 40% in the fine-size class (7 μm).

9.2.4

Sources and molecular composition of organic aerosols

The organic fraction of atmospheric particulate material can contain a large number of diverse molecular species. The composition is mainly dependent on the aerosol source with possible modifications during atmospheric transport. The organic mixture ranges from non-polar hydrocarbons (alkanes) over highly polar and water soluble components, such as short dicarboxylic acids or sugars, to macromolecular organics. Therefore, it is helpful to characterise the different sources (or source types) of organic aerosols in terms of their individual chemical composition. Knowing the source compositions selected source specific compounds can be used as tracers for the origin of aerosols or to estimate the contribution of the different source types to the measured aerosol.

9.2.4.1

Sources and composition of primary organic aerosols

There is a variety of biogenic and anthropogenic sources of primary organic aerosol constituents. Source strength estimations indicate that globally the input from natural sources is prominent. However, for the aerosol composition on the local or regional scale even weak sources (like cooking or cigarette smoking) can be important (Schauer et al. 1996). The following processes constitute the most important sources of primary organic aerosols that have been characterised: • • • •

Biomass burning Fossil fuel burning Plant abrasion Suspension of soil and dust

354


R

O R

O HO

OH

OH Levoglucosan

O CH3O

OH

HO R = H, CH3–C6H13 17(H)-Hopanes

Cholesterol: R = H -Sitosterol: R = ethyl

HO Vanillic acid

Figure 9.4 Tracers utilised for petroleum use (hopanes), biomass burning (levoglucosan, vanillic acid) and cooking (sterols).

• • • •

Suspension and release of bacteria, fungi, viruses, pollen, algae, spores (e.g. fern) Bubble bursting and sea spray Cigarette smoking Cooking.

Traffic is a massive source of particles from fossil fuel burning. The composition of these exhaust-particles depends on the engine-type (gasoline or diesel) and possible treatment of the exhaust (catalyst) (Rogge et al. 1993b; Schauer et al. 2002). Major compound classes are alkanes, alkanoic acids and PAH. Hopanes (see Figure 9.4) and steranes are only minor compounds; nevertheless, they can be used as specific tracers for emissions of motor vehicles or the general use of petroleum products. Biomass burning is a strong source for atmospheric aerosols, producing about four times more than fossil fuel burning (Kuhlbusch 1998). A main product and a general tracer used for biomass burning is levoglucosan (1,6-anhydro-β-d-glucopyranose; see Figure 9.4), an anhydro-sugar derived from the thermal degradation of cellulose during the combustion process (Simoneit 1999). Anhydro-sugars seem to be the most abundant compounds produced during the combustion process of plant material. Other major compound groups identified in smoke particles from biomass burning are alkanes, alkenes, alkanoic acids, di- and triterpenoids, monosaccharides, methoxyphenols and PAHs (Simoneit 2002). Some of these constituents derive from thermally altered plant material (like anhydro-sugars); others are unchanged ingredients (like some wax-alkanes). Especially, lignin pyrolysis products, such as vanillic acid, can be used as tracers for certain plant species. Plant abrasion is mainly induced by wind-driven mechanical force, like the rubbing of leaves against each other (Rogge et al. 1993a). Identified substances in aerosols from plant abrasion (green and dead leaves) are mainly constituents of the epicuticular plant waxes: n-alkanes, n-alkanals, n-alkanols, n-alkanoic acids (fatty acids). These compound groups are not very specific for biological sources but, due to their biosynthesis, specific patterns in carbon numbers of plant-derived wax components can be observed. Leaf wax alkanes have a strong odd carbon number predominance with the dominant carbon numbers C29 , C31 and C33 , whereas alkanoic acids, alkanals and alkanols have a predominance of even carbon numbers. Fossil fuel constituents are showing no predominance in carbon numbers. Therefore, it is possible to use these specific patterns to identify contributions of plants to atmospheric aerosols (Simoneit et al. 1988).

Organic Aerosols

355

Bubble bursting and sea spray are processes important for the production of primary organic aerosols in the marine environment. Surface active organic substances, such as fatty acids (Mochida et al. 2002), are enriched in the oceanic surface layer. Therefore, these substances are transferred into the atmosphere especially easily by bubble bursting and sea spray processes. This surface active organic material may be produced, for example, by phytoplankton (O’Dowd et al. 2004). Also, microorganisms may be transported into the atmosphere by bubble bursting processes. Soil contains several percent of OM (e.g. 5–6% for agricultural soils) that consists of plant litter, microbes, microbial and animal residues, lipids, carbohydrates, peptides, cellulose, lignin and humic material (Simoneit et al. 2004a). Carbohydrates account for about 5–20% of the organic material. Saccharides may be appropriate tracers for soil input into the atmosphere, in particular α- and β-glucose, inositols, sucrose and mycose (trehalose). However, there are also indications that a direct release of bioaerosols could be a main source of sugars and sugar alcohols in the atmosphere. The main constituents identified in aerosols produced by cooking are alkanoic and alkenoic acids derived from thermal hydrolysis and thermal oxidation of fat and oil. Other identified compounds are dicarboxylic acids, alkanes, sterols, levoglucosan and PAHs. The detailed composition of the aerosols depends on the food preparation method and the ingredients used (He et al. 2004a; Rogge et al. 1991). Different sterols seem to be appropriate tracers for cooking derived aerosols. While cholesterol (Figure 9.4) indicates use of meat, stigmasterol and β-sitosterol are present in plant oils.

9.2.4.2

Sources and composition of SOAs

SOAs are produced (a) by gas-phase oxidation of VOCs that can either form new particles or condense onto pre-existing particles, (b) by heterogeneous reactions on particle surfaces or (c) by in-cloud processing. Precursors of organic SOA are mostly volatile reactive biogenic (e.g. terpenes) or anthropogenic (e.g. aromatics) hydrocarbons. Products formed can be relative low volatile organics, which convert almost completely to the particle-phase, or semivolatile organics, which partition between the gas and particle phase. This gasparticle partitioning of semivolatile (and also low volatile) compounds can be described by gas-particle-partitioning models introduced by Odum et al. (1996) and Pankow (1994a, 1994b). These models consider the dependence of the concentration of an individual organic compound i in the particle phase from the available absorbing organic aerosol mass (MO ), the partitioning coefficient of compound i and the concentration of i in the gas phase: caer = cgas × Kom × MO

(9.4)

where Kom , partitioning coefficient of i (m3 /μg) (temperature dependent); caer , concentration of compound i in the absorbing organic particle phase (ng/m3 ); cgas , concentration of i in the gas phase (ng/m3 ); MO , concentration of the absorbing organic phase in the aerosol (μg/m3 ). Apart from problems such as source strength of precursors, chemical conversation, atmospheric transport, etc., the temperature dependence of the partitioning coefficient leads to some additional uncertainty in the modelling of SOA (Takekawa et al. 2003). During the first years of SOA-research, attention was paid to condensation of non-volatile and partitioning of semi-volatile compounds regarding the formation of particle phase.

356


Volatile products formed during the chemical conversation of aerosol precursors seemed to play no role in the formation of particulate matter. In contrast, more recent studies show that volatile carbonylic products formed in the gas-phase oxidation of organics may contribute to SOA mass by the formation of low volatile oligomers. For example, acidcatalysed reactions of aldehydes or ketones on the particle surfaces or inside the particles, like aldol reaction/condensation or acetal formation, are able to form such oligomers. These processes result in increased particle mass and a lower volatility of the particles. Compounds that are able to contribute to particle mass by oligomerisation may be formed from biogenic as well as anthropogenic precursors (Gao et al. 2004; Iinuma et al. 2004; Jang et al. 2002, 2004; Kalberer et al. 2004). There are also evidences for the direct formation of oligomeric products by heterogeneous reactions of unsaturated gas-phase compounds (e.g. isoprene) on particle surfaces (Limbeck et al. 2003). Several groups speculate that these oligomeric products formed from gaseous precursors could represent a substantial fraction of the so-called ‘humic-like substances’ (HULIS) often identified in atmospheric aerosols. HULIS is a collective term for a group of particle phase compounds, which add to the water soluble organic carbon. However, the exact chemical structures are not known. The chemical properties of the HULIS seem to be similar to humic and fulvic acids, which are known from waters, sediments and soils. Humic and fulvic acids are collective terms for molecules in a large range of molecular weights. These molecules contain multiple functional groups, e.g. carboxylic acid and carbonyl groups, and also aromatic structures. The formation of new particle-phase products from gas-phase constituents during the atmospheric lifetime of aerosols is part of the so-called atmospheric ageing of organic particles, a process that is currently not well characterised. Beside the incorporation of reactive gas-phase species into the organic aerosol fraction by oligomer formation, ageing also includes the degradation or chemical modification of particle-phase constituents by atmospheric oxidants. Since these chemical modifications will result in alterations of the physical (volatility, light absorption, light scattering) and physico-chemical properties (water solubility, activity as cloud condensation nuclei (CCN)) of atmospheric aerosols, which are relevant for the climatic effect of the aerosols, the investigation of these processes has to be addressed in future research on organic aerosols. The incorporation of SOA formation into atmospheric models is not an easy task, since a variety of chemical and physico-chemical processes influence the SOA particle mass in the ambient atmosphere. A sensitivity analysis of SOA production and transport modelling (Tsigaridis and Kanakidou 2003) showed a factor of about 20 of uncertainty in predicting the SOA production, considering the different influences of partitioning, ageing and MO , excluding the uncertainties of precursor emissions and individual oxidation pathways. This results in a range for the annual global production of SOA from 2.55 to 47.12 Tg of OM per year. Another study yields annual SOA production of between 15.3 and 24.6 Tg/year using the partition method and bulk yield method, respectively (Lack et al. 2004).

9.2.4.3

Biogenic SOA

Precursors of biogenic SOA in the continental environment are mainly unsaturated hydrocarbons namely (mono-) terpenes, sesquiterpenes and isoprene. The SOA-forming potential of terpenes is well known and was intensively investigated (e.g. Griffin et al. 1999; Hoffmann et al. 1997, 1998; Kavouras et al. 1998; O’Dowd et al. 2002; Went 1960; Yokouchi and

Organic Aerosols

357

OH

HO OH

HO

OH

OH

OH Isoprene

Methyltetrol (erythro- and threo-)

O

2,3-Dihydroxy-4-oxobutanoic acid

Figure 9.5 Isoprene and selected products from its atmospheric oxidation.

Ambe 1985; Yu et al. 1999a; Zhang et al. 1992), whereas isoprene was only most recently found to form low volatile secondary products (Claeys et al. 2004a, 2004b). Known products of the atmospheric isoprene oxidation are polyols and acidic compounds such as 2-methyltetrols and 2,3-dihydroxymethacrylic acid (Figure 9.5). It was estimated that isoprene might add about 2 teragrams of the polyols to the atmospheric SOA. This is a substantial amount, although terpenes may add 10 times more to SOA. The most frequently studied and most important SOA-forming reactions of terpenes are gas-phase oxidations by ozone, OH- and NO3 -radicals. The detailed chemical mechanisms of alkene (terpene) oxidation by atmospheric oxidants can be found in detail elsewhere in this book. Oxidation of terpenes under different conditions (ozone, OH-radicals, photosmog, etc.) generates a variety of oxygenated gas-phase (Calogirou et al. 1999) and particle-phase products, which have been identified in chamber experiments (e.g. Christoffersen et al. 1998; Glasius et al. 2000; Hoffmann et al. 1998; Jaoui and Kamens 2003a, 2003b, 2003c; Koch et al. 2000; Larsen et al. 2001; Winterhalter et al. 2003; Yu et al. 1999a). Known terpene oxidation products relevant for SOA production mainly contain carbonyl, alcohol and carboxylic acid functions. Products bearing carboxylic acid functions are low volatile and, therefore, are especially interesting for SOA-formation. Figure 9.9 shows some important products from monoterpene oxidation. Recently, it was proposed that peroxides could also represent a major part of the SOA formed by terpene ozonolysis (Bonn et al. 2004), a suggestion that was recently confirmed by chamber studies (Docherty et al. 2005). As mentioned above, also oligomer formation from (semi-) volatile oxygenated terpene oxidation products might contribute to SOA formation from biogenic precursors.

9.2.4.4

Anthropogenic SOA

Of the variety of anthropogenic VOC emissions, aromatic hydrocarbons are believed to be the most important compounds for the formation of SOA. The SOA forming potentials of various aromatic VOCs under photosmog conditions have been studied (e.g. Izumi and Fukuyama 1990; Odum et al. 1996, 1997; Stern et al. 1987; Wang et al. 1992). Detailed mechanisms for the atmospheric oxidation of anthropogenic VOC are given elsewhere in this book. Many particle-phase products have been identified in these studies (e.g. Bethel et al. 2000; Edney et al. 2001; Fisseha et al. 2004; Forstner et al. 1997a; Jang and Kamens 2001; Kleindienst et al. 2004; Smith et al. 1998, 1999), but only a few of these compounds could be measured in atmospheric aerosols. The products of the oxidation of aromatic hydrocarbons can be divided into three groups: (1) ring retaining aromatic, (2) ring retaining non-aromatic and (3) ring degradation products (Jang and Kamens 2001). Ring retaining

358


aromatic products can origin from the OH-addition to the aromatic ring that leads to the formation of OH- and NO2 -substituted rings or from the H-atom abstraction of alkyl substituents (toluene, xylenes, etc.) forming, for example, aromatic aldehydes or carboxylic acids. Ring retaining aromatic products are rather stable whereas ring retaining nonaromatic products containing double bonds can be easily further oxidised by ozone or nitrate radicals due to their high reactivity (Atkinson 2000). They finally form small multifunctional acids or carbonyls under atmospheric conditions. Therefore, the unsaturated products should be present only at very low concentration levels in atmospheric aerosols in contrast to the partly relative high yields determined in chamber studies. The often-high NOx levels in chamber studies compared to atmospheric conditions further favour the formation of aromatic nitro compounds because of the competitive reaction of NO2 and O2 with the intermediate benzyl radicals (Atkinson and Aschmann 1994; Bethel et al. 2000; Kleindienst et al. 2004). Nevertheless, nitrophenols as aromatic photooxidation products can be found in atmospheric aerosols, though they may also have primary sources such as automobile exhaust (Cecinato et al. 2005; Tremp et al. 1993). Most particle-phase constituents that derive from aromatic hydrocarbon oxidation in the atmosphere (Fisseha et al. 2004; Jang and Kamens 2001; Kleindienst et al. 2004) are ring degradation products: low volatile short carboxylic and dicarboxylic acids, such as succinic, pyruvic, malonic, maleic, methylmaleic, malic, glyoxylic and oxalic acid (Falkovich et al. 2005; Kawamura and Kasukabe 1996; Kawamura et al. 2005; Kerminen et al. 2000; Sempere and Kawamura 2003; Yao et al. 2004), see Figure 9.6 for structures. However, these substances may also be secondary products from other precursors and they can also be emitted by primary sources like automobile exhaust (Kawamura and Kaplan 1987) and biomass burning (Falkovich et al. 2005). Low molecular weight dicarboxylic acids are also suggested to derive from the degradation of higher dicarboxylic acids on particles (Kawamura and Ikushima 1993) and liquid-phase reactions like in-cloud processing (Crahan et al. 2004; Warneck 2003; Yu et al. 2005a). Oxalic acid is frequently found to be the most abundant dicarboxylic acid in aerosols, but also opposite observations exist (Yu et al. 2005b). Since oxalic acid seems to be the last step in the degradation pathway of a variety of gas and particle-phase constituents, it is not surprising that its concentration often dominates the composition of the aerosol particles. Figure 9.7 shows some possible pathways for the formation of oxalic acid from different sources. Also, malonic and succinic acid, which are usually second and third in particle-phase concentrations of dicarboxylic acids, have different primary and secondary sources. Malic acid is sometimes found to be more abundant than succinic acid. More reactive species such as glyoxylic, pyruvic and maleic acid are usually present at lower concentrations and may be seen as intermediate products of the degradation of different substances, including aromatic hydrocarbons. It would be favourable to identify reliable marker substances for the formation of secondary aerosols from aromatic precursors. As outlined above, oxalic acid as well as other shortchain diacids and ketoacids cannot be used as specific tracers for anthropogenic SOA because of their various other possible sources. Recent investigations (Kleindienst et al. 2004) found that 2,3-dihydroxy-4-oxobutanoic acid and 2,3-dihydroxy-4-oxopentanoic acid might be suitable markers for SOA from toluene or related aromatic compounds. Phthalic acid has been found to derive from degradation of polycyclic aromatic hydrocarbons (Jang and McDow 1997; Kawamura and Ikushima 1993). Measurements of phthalic acid concentrations, its concentration in different particle size ranges (Fine et al. 2004) and comparison with other diacids present in atmospheric aerosol particles (Fraser et al. 2003) confirm the

Organic Aerosols

359

COOH HOOC — COOH Oxalic aicd

HOOC

COOH

HOOC

Malonic acid O

Succinic acid OH

O

COOH

C HC — COOH Glyoxylic acid (wC2)

HOOC

COOH

HOOC Malic acid

Ketomalonic acid (kC3) COOH

O H 3C

O COOH

COOH

C

COOH Fumaric acid

Pyruvic acid

COOH

O HC

4-Oxobutanoic acid (wC4)

COOH

COOH

Phthalic acid

HC

HOOC

COOH

Maleic acid

Methylmaleic acid

O CH

Glyoxal

COOH

H3C

HOC

COOH

3-Oxopropanoic acid (wC3)

H3C

O

O

C

CH

Methylglyoxal

Figure 9.6 Dicarboxylic acids, oxocarboxylic acids and other oxygenated molecular species which can be found in the atmospheric particle phase. Compounds are either emitted directly into the atmosphere or are produced by chemical processes within the atmosphere.

suitability of phthalic acid as a specific tracer for anthropogenic SOA. Although in principle other compounds also seem to be suitable for tracing anthropogenic secondary aerosol formation, such as benzoic acid and substituted benzoic acids, they are relatively volatile and consequently a substantial portion of them exist in the gas phase.

9.2.5

Atmospheric concentrations of primary and secondary organic aerosols constituents

Until now, the molecular identity of the largest part of OM found in atmospheric aerosols remains unknown. This is illustrated by Figure 9.8 (Kubatova et al. 2002) showing the identified fraction vs. the unidentified in the extractable part of OM. In the last column, the most abundant identified classes of organics are displayed. Different acids (mostly fatty acids) dominate the identified fraction, but also alkanes are found as important constituents. The composition of the aerosol shown in Figure 9.8 is one possible example. Generally, the composition of aerosol particles is dependent by their sources as well as by the specific history of the investigated air mass. The concentrations of alkanes are relatively high in urban areas, typically around 100 ng/m3 . Under special conditions concentrations might be considerably higher, for example, up to several thousand ng/m3 in Kuala Lumpur during a pollution event

360


Aromatic HCs (toluene, benzene, etc.)

Pyruvic

Glyoxal methylglyoxal

Unsaturated fatty acids

Maleic methylmaleic

Malic

kC3

C2 Oxalic

Fumaric Monocarboxylic

wC3

wC2

C3 Malonic

n-Alkanes aldehydes

wC4 Mid-chain ketoand hydroxyacids

C4 Succinic

Primary sources (combustion processes)

Figure 9.7 Primary and secondary sources for oxalic, malonic and succinic acid. Intermediate and rather reactive products are in broken lines. Abbreviations are explained in Figure 9.6. Modified figure after Kawamura and Kasukabe (1996).

caused by air masses from biomass burning in Indonesia (Table 9.3). Usually, the use of fossil fuel is the main source of alkanes in urban areas. In more natural environments, concentrations of alkanes are considerably lower, especially in remote marine areas, where concentrations may be below 1 ng/m3 . In aerosols from remote marine areas, the input of alkanes from marine sources, which are weak in comparison to terrestrial or anthropogenic sources, can be substantial (Sicre and Peltzer 2004). Aldehyde and alcohol concentrations show a similar behaviour as the alkane concentrations. Aldehyde concentrations range from 0.01 in marine to 7 ng/m3 in urban environments; alcohols are also less abundant in remote marine areas (below 1 ng/m3 ) have higher concentrations in urban areas (up to about 100 ng/m3 ) and peak during, pollution events from biomass burning. For example, more than 700 ng/m3 were measured in Kuala Lumpur during pollution events due to air masses transported from tropical forest fires in Indonesia (Fang et al. 1999). Alkanoic acids show concentrations around 10 ng/m3 in remote marine aerosols, from 10 to 100 ng/m3 in terrestrial remote and rural areas and above 100 ng/m3 in urban areas that peak during biomass burning events at several thousand ng/m3 (max. 14 000 ng/m3 ). PAHs are typical for combustion aerosols from fossil fuel especially coal. Concentrations are well below 1 ng/m3 in remote marine environment growing with anthropogenic influence reaching concentrations of almost 100 ng/m3 in polluted urban/industrial areas.

Organic Aerosols

361

100 EC

Unident.

Percentage

80 Unextr. and/or unelut. OM

60

40

20

Unres. (UCM)

DTs

Acids

Alkanes

OM

Identif.

Fatty Ac. (palm., stearic, oleic)

0.66 μg/m3

0.44 μg/m3

Bases + neutrals Carbon. EEOM

Others Ox. deg. Lign. pyr. PAHs

Resolv.

0 54 μg/m3

17.8 μg/m3

14.5 μg/m3

2.70 μg/m3

0.66 μg/m3

Figure 9.8 Percentage contributions of different carbonaceous and organic compound classes in winter 1998 ‘total’ filter samples from Gent, Belgium (based on 22 samples from each season). The first column represents the contributions of the carbonaceous (OM + EC) and the inorganic aerosol to the particulate mass (PM). The atmospheric concentrations under the columns indicate the average concentration for the sum of the species in the column. From Kubatova et al., 2002 with permission. Abbreviations: EEOM = extractable and elutable organic matter, UCM = unresolved complex mixture, DTs = diterpenoic acids, Lign. pyr. = lingnin pyrolysis products, Ox. deg. = oxidative degradation products.

While these compounds represent mainly the lipid constituents, levoglucosan, sugars and short-chain dicarboxylic acids are major components representing water-soluble organic compounds (WSOC – water-soluble organic carbon) in the atmospheric particle phase. Table 9.4 shows atmospheric concentrations of selected sugars and levoglucosan in different regions. Levoglucosan is a degradation product of cellulose and almost exclusively produced by combustion of plant material. Therefore, it is not surprising that concentrations are low in remote areas (around 1 to about 10 ng/m3 ), higher in urban areas due to the use of wood as fuel (roughly 100 to >1 000 ng/m3 ) and reach highest concentrations in aerosols from massive biomass burning in tropical areas (around 1 000 to >10 000 ng/m3 ). Different sugars also show a wide range of concentrations from less than 1 ng/m3 in marine aerosols to around 1 000 ng/m3 in aerosol samples from Chile. Usually, concentrations of glucose, sucrose and mycose are lower than 100 ng/m3 . The source of sugars is believed to be mainly soil dust including suspended microorganisms (Simoneit et al. 2004a). Other observations in the Amazonian rainforest point to contributions from living plants. Mycose, arabitol and mannitol are well-known constituents of fungal spores, and sucrose, glucose and fructose are known to be present in pollen grains. Consequently, a strong dependence of the sugar concentrations from the daytime was observed with higher glucose, sucrose and fructose concentrations during daytime and higher levels of mycose (trehalose), arabitol and mannitol during the night when a strong release of fungal spores occurred (Graham et al. 2003a).

362


Table 9.3 Concentrations (ranges or mean) of lipid compounds measured in aerosol samples from different sites. Concentrations of compound classes sometimes include individual substances with different chain length of carbon atoms. All concentrations were obtained from TSP samples (= total suspended particulate matter), except those reported by Brown et al. (2002). Values are rounded off for better readability Location

South Pacific1 Cichi-jima Island2 Crete, Greece3 Texas4 Petrana, Greece5 Gosnan, Jeju Island, Korea2 Heraklion, Greece3 Kuala Lumpur, Malaysia6 Qingdao, China7 Kuala Lumpur, Malaysia8 Gent, Belgium9

Concentration (ng/m3 )

Site description Alkanes

PAHs

Fatty acids

Remote/Marine Remote/Marine Marine Remote/Continental Rural Rural/Marine

0.02–0.54 0.3–7.0 6.5–26 1.2–15 27 9.6–150

— 0.0–0.08 0.1–2.5 — 0.52 0.05–7.7

— 3.2–7.6 1.0–20 7.0–32 — 23–96

0.05–0.77 4.6–22 3–17 — — 5.0–130

Urban Urban/BMB Urban Urban Urban

65–320 290–9 300 220 26–340 77

20–60 7–46 88 — 11

110–200 320–14 000 650 4 000–7 400 140

17–32 31–740 — 160–720 —

n-Alkanols

1 Sicre and Peltzer (2004). 2 Simoneit et al. (2004b).

3 Kavouras and Stephanou (2002). 4 Brown et al. (2002). 5 Kalaitzoglou et al. (2004). 6 Fang et al. (1999). 7 Guo et al. (2003). 8 BinAbas and Simoneit (1996). 9 Kubatova et al. (2002).

Levoglucosan, sugars and the lipid constituents represent the primary fraction of organic aerosols, whereas short-chain dicarboxylic acids have primary and secondary sources. Oxocarboxylic acids derive mostly from secondary processes, either the oxidation of VOCs or the further oxidation of carbonyls, mono- or dicarboxylic acids. Oxalic acid (and other di- and oxocarboxylic acids) levels are rather low in marine and remote areas, although a secondary production can even be observed in remote arctic regions (Kawamura et al. 2005). Concentrations of oxalic acid are roughly on the order of around 100 ng/m3 in natural or remote regions, as shown in Table 9.5. Urban concentrations are on the order of a few hundred ng/m3 and highest concentrations are observed in biomass burning aerosols (>1 000 ng/m3 ). The same trend can be observed for malonic and succinic acid although concentrations are substantially lower. Pyruvic, glyoxylic and maleic acid are known products of the degradation of aromatic VOCs. Their concentrations are usually lower in remote regions and higher in polluted air masses. The main source of phthalic acid is probably the degradation of (polycyclic) aromatic hydrocarbons as explained above. Products from the oxidation of biogenic VOC (e.g. terpenes) are important contributors to SOAs. Mainly forested regions are influenced by this group of compounds. The acidic

Organic Aerosols

363

Table 9.4 Concentration of some sugars and levoglucosan in the atmospheric particle phase at different locations. Values are rounded off for better readability Location

Site description

PMa

Gosnan, Jeju Island, Marine/Continental TSP Korea1 North Pacific1 Marine/Remote TSP Rain-forest/Remote Fine Amazonia, Brazil2 Coarse Rondonia, Brazil3 Rain-forest/BMBb Fine Urban TSP Santiago, Chile4 Kuala-Lumpur, Urban/BMB TSP Malaysia4

Concentration (ng/m3 ) Glucose

Sucrose

Mycose

Levoglucosan

11–110

6–440

2.5–30

8–74

0–0.2 4.9–12 21–90 5–18 8–1 700 —

0.2–1.3 7.7–33 C7 Alkylbenzenes

Biomass burning Street tunnel Underground garage Gas station

−25.7 (0.5)f −25.3 (0.5)d ; −25.5 (0.7)e ; −27.1 (0.6)g −27.6 (0.1)g −26.9 (0.3)g

Styrene

Biomass burning

−23.3 (1.0)f

Isoprene

Vegetation (velvet bean)

−27.7 (0.4)h

a Numbers in parenthesis give the error of mean; if the available information is insufficient for a statistical evaluation the range of data is given; if less than three data points are available the individual values are listed. b From Rudolph et al. (1997). c From Komatsu et al. (2005). d Samples collected in summer, downtown Toronto (Canada), from Thompson (2003). e Samples collected in winter, downtown Toronto (Canada), from Thompson (2003). f From Czapiewski et al. (2002). g Measurements in the area of Toronto (Canada), from Rudolph et al. (2002). h From Rudolph et al. (2003).

Gas Chromatography–Isotope Ratio Mass Spectrometry

455

For aromatic compounds and alkenes in urban air 13 C is, compared to alkanes, generally slightly enriched. This most likely reflects the higher KIEs for their atmospheric reactions (Table 10.7). There is also some indication that alkenes in emissions from incomplete combustion sources are slightly heavier than alkanes or aromatic compounds. This difference is only small, but it can be found for transportation related emissions as well as for emissions from biomass burning. Overall, the presently available data suggest that the carbon isotope ratios of emitted NMHC are generally very close to that of the burnt fuel. There is some evidence for the existence of a systematic carbon isotope fractionation between fuel and emitted NMHC for incomplete combustion processes but these effects are small and in most cases the variations are comparable in magnitude to the uncertainty of the measurements (Czapiewski et al. 2002; Rudolph et al. 2003). Similar to the findings of little carbon isotope fractionation effects for combustion processes, evaporation of VOCs only results in a small enrichment of 13 C in the gas phase (Harrington et al. 1999; Irei et al. 2006; Smallwood et al. 2002). Typically, the VOCs in the gas phase are a fraction of a per mille heavier than in the liquid phase. Most of the reported fractionation effects are in the range of 0.2–0.5‰. However, the actual impact of this effect on the isotope ratio of atmospheric VOC depends on the type of evaporation process. For partial evaporation such as evaporative losses from storage containers or during fuelling evaporation will cause a small isotope fractionation. However, or in the case of complete evaporation, for example from gasoline spills, there is overall no isotope fractionation. There are an extremely large number of studies of isotope ratios of total crude oil as well as of individual NMHC in crude oil or natural gas. A detailed review of this extremely extensive body of information is beyond the scope of this chapter. In general, crude oil has stable carbon isotope ratios in the range of −2333‰ with most of the values between −25‰ and −31‰ (see, for example, Yeh and Epstein 1981). An effectively identical range of carbon isotope ratios was found by Smallwood et al. (2002) for a substantial number of individual compounds in 19 gasoline samples from different areas of the United States. Harrington et al. (1999) determined the stable carbon isotope ratio of 44 samples of commercially available light alkyl benzenes. The results ranged from −23.9‰ to −29.4‰ with an average of −27.1‰. The 10‰ and 90‰ were −28.9‰ and −25.4‰, respectively. Overall the observed isotope ratios of most anthropogenic NMHC sources fall into a tight range, which is consistent with the range of isotope ratios of the parent materials. Somewhat different isotope ratios are expected for emissions of light alkanes due to leakage of natural gas. Although for ethane and heavier alkanes the depletion of 13 C relative to crude oil is far less pronounced than for methane, ethane in natural gas is, depending on the origin of the natural gas, generally a few per mille lighter than the average of crude oil. Similar to methane, the isotope ratio of NMHC in natural gas depends on the origin of the natural gas; however, the variability is by far less pronounced for ethane and very small for propane and heavier alkanes. The available stable carbon isotope ratio measurements in urban areas (Table 10.10) show no evidence for a systematic difference between ethane and other alkanes. A possible explanation is that the presently available studies are conducted for areas, where the influence of natural gas loss is small compared to other fossil fuel derived emissions such as automotive exhaust. Furthermore, the background concentration of ethane is substantial and enriched in 13 C relative to ethane over urban areas. Thus the contribution of heavier ethane from background air may compensate for the influence of urban ethane emissions depleted in 13 C.

456


A large variability is found for the isotope ratios of ethyne. The presently available data show a very large variability for emissions as well as for ambient ethyne. Moreover all observations indicate that 13 C in ethyne is substantially enriched relative to fossil fuel. The reasons for this unusual behaviour of ethyne are not known. The isotope ratios of isoprene emitted by vegetation have been studied by several groups. The first study was conducted by Sharkey et al. (1991) who report that isoprene emitted from oak leaves was depleted by 2.8 ± 1.1‰ relative to photosynthetically fixed carbon. Rudolph et al. (2003) studied the carbon isotope ratio and its temperature and light dependence for isoprene emitted from velvet bean (Mucana pruriens) at ambient carbon dioxide concentrations and isotope ratios. The emitted isoprene was 2.6 ± 0.9‰ lighter than the leaf carbon. The observed light and temperature dependence of the isotope ratio was small. Funk et al. (2004) found that isoprene emitted from Populus deltoids was between 2.5‰ and 3.2‰ lower in 13 C than photosynthetically fixed carbon in the absence of heat or water stress. For plants under severe heat- or water-stress the fractionation effect increased to −8.5‰ and −9.3‰. Affek and Yakir (2003) studied the carbon isotope fractionation of isoprene emitted from myrtle (Myrtus communis), buckthorn (Rhamnus alaterus) and velvet bean (Mucana pruriens). They report that isoprene was depleted by 4–11‰ relative to photosynthetically fixed carbon. This is significantly larger than the 13 C depletion found by other groups in the absence of stress. Based on these results the carbon isotope ratios of isoprene emitted from vegetation in the absence of stress can be expected to range from −27‰ to −30‰. Under conditions of severe stress the isotope ratio may decrease to approximately −35‰. The measured carbon isotope ratios of isoprene in the atmosphere (Table 10.11) are often considerably heavier than the emissions, which is consistent with the carbon KIE for the reaction of isoprene with the OH-radical or ozone (Table 10.7) and the short atmospheric residence time of isoprene. In most cases, the carbon in atmospheric NMHC becomes heavier with increasing distance of the observation site from major NMHC sources. However, there are some exceptions. Several of the NMHC carbon isotope ratios measured at background sites such as Baring Head or Alert are very close to those of emissions. This can be explained by the combination of two factors. For remote locations the contribution to NMHC concentrations from long-range transport is very small for reactive NMHC. Therefore local or regional emissions can play a dominating role, even if the emission rates are only small. Furthermore, the currently applied measurements techniques for NMHC isotope ratios are not sufficient to reliably determine isotope ratios for the lower end of the range of NMHC concentrations at remote concentrations. The consequence is that for NMHC with very low concentrations in background air isotope ratio measurements at remote locations will be biased towards samples influenced by local or regional sources.

10.5.3

Oxygenated VOC

Table 10.12 summarises measurements of the stable carbon isotope ratios of oxygenated VOCs in the atmosphere. Most of the studies targeted identification of the sources, sometimes using measurements of 14 C as indicator for a biogenic origin of the VOCs. Glasius et al. (2000) concluded from their 14 C measurements that formic acid found in the atmosphere


457

Table 10.12 Stable carbon isotope ratio measurements of atmospheric oxygenated VOC Number of data points

Rangea δ 13 C, ‰

Formaldehydeb

2

−26.69; −28.51

Formaldehydeb Formaldehydec Formaldehydec Formaldehyded Acetaldehydeb

2 6 3 7 2

−16.68; −18.99 −17 −28.3 −22.5 (0.8) −29.2; −29.28

Acetaldehydeb Formic acide Formic acidf

2 4 52

−21; −21 −27.6 (0.4) −18‰ to −25‰

VOC

Formic acidf Acetic Acidg

5 5

−30.1 (0.6) −20.5 (0.7)

Location of study

Industrial, near petrochemical factory Ghuangzhou China Bus station in Ghuangzhou China Mountain, continental US Clean air site, coastal NZ Coastal, Nova Scotia, Canada Industrial, near petrochemical factory Ghuangzhou China Bus station in Ghuangzhou China Rural, Denmark Several sites continental and coastal United States Rainwater Los Angeles, California (USA) Rainwater Los Angeles, California (USA)

a The number in brackets give the standard deviation and the standard error of the mean. If no statistical evaluation is

available the range of observations is given. If less than three data points are available the individual values are listed.

b From Wen et al. (2005). c From Johnson and Dawson (1990). d From Tanner et al. (1996). e From Glasius et al. (2000). f From Johnson and Dawson (1993).

g From Sagukawa and Kaplan (1995).

in a rural area of Denmark is predominantly derived from biological material. This is consistent with the observation of a narrow range of δ 13 C values with an average value of −27.6‰. This also suggests that for these conditions isotope fractionation due to the formation processes or reactions in the atmosphere is marginal. A nearly identical type of study was conducted by Johnson and Dawson (1993) for the continental and coastal United States. Similar to the findings of Glasius et al. (2000) the measured 14 C content strongly suggested a predominantly biogenic origin of gas-phase formic acid. However, the corresponding δ 13 C values range from −18‰ to −25‰, indicating the existence of a process resulting in 13 C enrichment. Qualitatively this is compatible with isotope fractionation associated with atmospheric loss processes. However, presently neither the KIEs for atmospheric loss processes nor the carbon isotope ratio of formic acid formed by biological processes are known. The available stable carbon isotope ratio measurements for formaldehyde and acetaldehyde show a considerable variability. The δ 13 C values range from −29‰ to −17‰. Sources of formaldehyde are the atmospheric oxidation of VOC and methane as well as emissions from industrial sources and incomplete combustion. It has been argued that 13 C in atmospheric formaldehyde is enriched relative to the emissions due to isotope fractionation associated with formaldehyde photolysis (Johnson and Dawson 1990). Qualitatively this can indeed explain the observations. Photolysis is one of the most important atmospheric removal formaldehyde processes and its KIE is substantial (Section 10.4.4). Unfortunately,

458


the necessary information for quantitative considerations is not available. The KIE for another very important atmospheric loss process, the reaction with OH-radicals, is only poorly known (Table 10.6), and the isotope ratios of formaldehyde emissions and formation processes have not been measured.

10.6

Conclusions

There is a substantial set of data available for sources, sinks, atmospheric reactions and measured ambient isotope ratios of atmospheric VOCs. It is, therefore, somewhat unexpected that the quantitative interpretation of atmospheric measurements VOC isotope ratios is still seriously limited by the lack of data for comparison. For those who remember the early days of studies of VOC concentrations, this will sound familiar. It took about 20–30 years to obtain a reasonable level of basis for the interpretation of atmospheric VOC concentrations in general and, as can be seen from other chapters in this book, there are still very significant gaps. One of the problems is the very large number of different chemical substances summarised under the heading of VOCs. The consequence is that the available information on VOCs is spread over a large number of different compounds, with often only a limited amount of knowledge for individual compounds. It is easy to predict that once the necessary basic information is available, isotope ratio measurements will belong to the established tools for studying atmospheric VOC chemistry. However, the number of research groups studying VOC stable isotope ratios is presently small, although there has been considerable increase during the past few years. Consequently, we have to accept that developing a broad, general basis for studies of atmospheric VOC isotope ratio measurements will take some time. The present lack of quantitative information makes it very speculative to predict the accuracy, and therefore also the usefulness of constraints which can be derived from VOC isotope ratio studies. Nevertheless, it is possible to identify several areas where it is extremely likely that isotope ratio studies will be particularly valuable in the near future or where their usefulness has already been demonstrated. The possibility to use measurements of NMHC isotope ratios for studies of atmospheric processing and identification of the origin of NMHC has been shown in a small number of studies. The main impediment for a wider application is the substantial experimental effort required for such measurements. However, the continuing progress in GC–IRMS instrumentation and the decreasing costs of such instrumentation make it likely that such measurements will become a routine tool in the near future. Due to their important role for stratospheric ozone halogenated VOCs belong to the most important organic compounds in the atmosphere. Therefore, additional tools to constrain the budget, such as isotope ratio studies, can be extremely valuable. Indeed, there already is substantial progress towards establishing the isotope ratio budgets for chloromethane and bromomethane. Although these isotope budgets still have gaps and substantial uncertainties, the missing pieces of information can be identified, and the necessary experimental studies are well within the reach of existing methods. It is somewhat surprising that there has been little effort towards studying the isotope ratio budgets for other halogenated VOCs. Especially for 1,1,1-trichloroethane (methyl chloroform) and the hydrochlorofluorocarbons (HCFC), isotope ratio studies have the potential


459

of providing valuable constraints for their budgets. For these compounds, removal in the troposphere by reaction with OH-radicals is an important sink. This can and has been used to derive estimates for the average tropospheric OH-radical concentration. One of the obvious problems is that discrepancies in the mass budget may be explained by errors in the emission rates as well as changes in the OH-radical concentration. Isotope ratio budgets can supply the additional constraints needed to differentiate between these possibilities. The experimental techniques for measurements of the isotope ratios for halogenated VOCs in the atmosphere are available, and, due to the long atmospheric lifetime of these compounds, the number of atmospheric measurements needed to derive accurate averages for their isotope ratios is within reasonable limits. However, accurate measurements of the KIEs for the dominant tropospheric loss reaction of these compounds, the reaction with OH-radicals, will present some experimental challenges due to the low reactivity of these compounds. Nevertheless, such measurements are within the reach of existing KIE measurement techniques. There is one area where it can be expected that isotope ratio studies will have a number of valuable applications, although presently there is little information available. The chemistry of the oxidation products of atmospheric VOCs is extremely complex. Therefore, any additional constraints, including isotope ratio studies, will be highly welcome. Possible applications range from studies of formaldehyde to methacrolein and methyl vinyl ketone and secondary POM. Inherent in the complexity of the atmospheric chemistry of secondary VOCs and POM is the need for laboratory studies of isotope fractionation effects associated with the formation and loss reactions and measurement techniques for their isotope ratios in the atmosphere. Both tasks are experimentally challenging. Nevertheless, it has been demonstrated that such measurements are possible. Kawamura and Watanabe (2004) described a method for compound-specific carbon isotope ratio measurements of dicarboxylic acids and ketocarboxylic acids in atmospheric POM and very recently ambient and laboratory studies of methacrolein and methyl vinyl ketone formed by the oxidation of isoprene have been conducted (R. Iannone, R. Koppmann and J. Rudolph, unpublished results). The examples for applications of VOC isotope ratio measurements given above should not be considered as a complete list of all possibilities. With more detailed quantitative information it is very likely that there will be additional applications. Especially further development of experimental and interpretative tools have the potential of leading to a variety of new and valuable applications of VOC isotope ratio studies. An obvious development in experimental techniques is measurement of VOC isotope ratios for other elements but carbon. Methods for the measurement of VOC hydrogen isotope ratios are within the range of existing state of the art methods. With very few exceptions, all atmospheric VOCs contain hydrogen. The value of combining measurements of isotope ratios for different elements in one compound has been demonstrated in numerous applications in various fields of science. There can be little doubt that such an approach will be equally valuable for studies of atmospheric VOCs. Finally, it should be remembered that the diagnostic value of VOC stable isotope ratio measurements depends on the conceptual and numerical tools available for interpretation. Numerical model simulations belong to the most versatile and advanced methods to evaluate atmospheric observations and thus have a large potential for interpreting observations of VOC isotope ratios. Similarly, comparison between VOC isotope ratio measurements

460


and observations can play an important role for model testing and validation. Presently, there are very few modelling studies of VOC isotope ratios, but it has already been shown that VOC isotope ratio predictions can be added to models without changing the model chemistry. This opens a wide range of possibilities to use VOC isotope ratios as diagnostic tracers for the performance of numerical models.

References Affek, H.P. and Yakir, D. (2003) Natural abundance carbon isotope composition of isoprene reflects incomplete coupling between isoprene synthesis and photosynthetic carbon flow. Plant Physiology, 131: 1727–6. Allan, W., Manning, M.R., Lassey, K.R., et al. (2001) Modeling the variation of δ 13 C in atmospheric methane: Phase ellipses and the kinetic isotope effect. Global Biogeochemical Cycles, 15: 467–81. Anderson, R.S. (2005) Carbon Kinetic Isotope Effects in the Gas-Phase Reactions of Nonmethane Hydrocarbons with Hydroxyl Radicals and Chlorine Atoms. PhD Thesis, York University, Toronto, Canada. Anderson, R.S., Czuba, E., Ernst, D., et al. (2003) Method for measuring carbon kinetic isotope effects of gas-phase reactions of light hydrocarbons with the hydroxyl radical. Journal of Physical Chemistry, 107: 6191–9. Anderson, R.S., Iannone, R., Thompson, A.E., et al. (2004a) Carbon kinetic isotope effects in the gasphase reactions of aromatic hydrocarbons with the OH radical at 296 ± 4 K. Geophysical Research Letters, 31. Anderson, R.S., Iannone, R., Thompson, A.E., et al. (2004b) Carbon kinetic isotope effects in the gasphase reactions of light alkanes and ethene with the OH radical at 296 ± 4 K. Journal of Physical Chemistry A, 108(52), 11537–44 . Archbold, M.E., Redeker, K.R., Davis, S., et al. (2005) A method for carbon stable isotope analysis of methyl halides and chlorofluorocarbons at pptv concentrations. Rapid Communications in Mass Spectrometry, 19: 337–42. Aschmann, S.M. and Atkinson, R. (1995) Rate constants for the gas-phase reactions of alkanes with Cl-atoms at 296 ± 2 K. International Journal of Chemical Kinetics, 27: 613–22. Atkinson, R. (1989) Kinetics and mechanisms of the gas-phase reactions of the hydroxyl radical with organic compounds. Journal of Physical and Chemical Reference Data, 1: 1–246. Atkinson, R. (1994) Gas phase tropospheric chemistry of organic compounds. Journal of Physical Chem. Ref. Data Monogr, 2: 11–216. Atkinson, R. (1997) Gas phase tropospheric chemistry of organic compounds. 1. Alkanes and Alkenes. Journal of Physical Chemistry Reference, 26: 215–19. Beukes, J.A., D’Anna, B., Bakken, V., et al. (2000) Experimental and theoretical study of the F, Cl and Br reactions with formaldehyde and acetaldehyde. Journal of Physical Chemistry Chemical Physics, 2: 4049–60. Biegeleisen, J. and Wolfsberg, M. (1958) Theoretical and experimental aspect of isotope effects in chemical kinetics. In: I. Progogine (Ed.) Advances in Chemical Physics. Interscience Publishers, Inc., NY, pp. 15–76. Bill, M., Miller, L.G. and Goldstein, A.H. (2002a) Carbon isotope fractionation of methyl bromide during agricultural fumigation. Biogeochemistry, 60: 181–90. Bill, M., Rhew, R.C., Weiss, R.F., et al. (2002b) Carbon isotope ratios of methyl bromide and methyl chloride emitted from coastal salt marshes. Geophysical Research Letters, 29: 1045–8. Bill, M., Conrad, M.E. and Goldstein, A.H. (2004) Stable carbon isotope composition of atmospheric methyl bromide. Geophysical Research Letters, 31: L0410.


461

Brenninkmeijer, C.A.M., Lowe, D.C., Manning, M.R., et al. (1995) The 13 C, 14 C, and 18 O istopic composition of CO, CH4 , and CO2 in the higher southern latitudes lower stratosphere. Journal of Geophysical Research, 100: 26163–72. Brenninkmeijer, C.A.M. and Röckmann, T.T. (1997) Principal factors determining the 18 O/16 O ratio of atmospheric CO as derived from observations in the southern hemispheric troposphere and lowermost stratosphere. Journal of Geophysical Research, 102: 25477–85. Burgoyne, T.W. and Hayes, J.M. (1998) Quantitative production of H2 by pyrolysis of gas chromatographic effluents. Analytical Chemistry, 70: 5136–41. Chiltz, G., Eckling, R., Goldfinger, P., et al. (1963) Kinetic isotope effect in photochlorination of H2 , CH4 , CHCl3 , and C2 H6 . Journal of Physical Chemistry, 38 (5): 1053–61. Ciais, P., Tans, P.P., Trolier, M., et al. (1995) A large northern hemisphere terrestrial CO2 sink indicated by the 13 C/12 C ratio of atmospheric CO2 . Science, 269: 1098–102. Ciais, P., Tans, P.P., Denning, A.S., et al. (1997) A three-dimensional synthesis study of δ 18 O in atmospheric CO2 , simulations with the TM2 transport model. Journal of Geophysical Research, 102: 5873–83. Conny, J.M. and Currie, L.A. (1996) The isotopic characterization of methane, non-methane hydrocarbons and formaldehyde in the troposphere. Atmospheric Environment, 30: 621–38. Conny, J.M., Verkouteren, R.M. and Currie, L.A. (1997) Carbon 13 composition of tropospheric CO in Brazil: A model scenario during the biomass burn season. Journal of Geophysical Research, 102: 10683–93. Craig, H. (1957) Isotopic standards for carbon and oxygen and correction factors for mass-spectrometric analysis of carbon dioxide. Geochimica et Cosmochimica Acta, 12: 133–49. Craig, H., Chou, C.C., Welhan, J.A., et al. (1988) The isotopic composition of methane in polar ice cores. Science, 242: 1535–9. Crawford, M.A., Li, Z., Heuerman, H.A., et al. (2004) A kinetic and product study of reaction of chlorine atom with CH3 CH2 OD. International Journal of Chemical Kinetics, 36: 584–90. Czapiewski, K.V., Czuba, E., Huang, L., et al. (2002) Isotopic composition of non-methane hydrocarbons in emissions from biomass burning. Journal of Atmospheric Chemistry, 43: 45–60. Czuba, E. (1999) Development of a Technique to Study Stable Carbon Isotope Composition of NMHCs in Ambient Air; MSc. Thesis, York University, Toronto. D’Anna, B., Bakken, V., Are Beukes, J., et al. (2003) Experimental and theoretical studies of gas phase NO3 and OH radical reactions with formaldehyde, acetaldehyde and their isotopomers. Journal of Physical Chemistry Chemical Physics, 5 (9): 1790–805. Dobis, O., Benson, S.W. and Mitchell, T.J. (1994) Kinetics of the reactions of C1 atoms with C2 D6 and C2 D5 and the disproportionation of 2C2 D5 . The deuterium isotope effect. Journal of Physical Chemistry, 98: 12284–93. Donahue, N.M. (2001) Revisiting the Hammond postulate: The role of reactant and product ionic states in regulating barrier heights, locations, and transition state frequencies. Journal of Physical Chemistry A, 105: 1489–97. Donahue, N.M., Anderson, J.G. and Demerjian, K.L. (1998) New rate constants for ten OH-Alkane reactions from 300–400 K: An assessment of accuracy. Journal of Physical Chemistry A, 102: 3121–6. Droege, A.T. and Tully, F.P. (1986a) Hydrogen atom abstraction from alkanes by hydroxyl. 5. n-Butane. Journal of Physical Chemistry, 90: 5937–41. Droege, A.T. and Tully, F.P. (1986b) Hydrogen-atom abstraction from alkanes by hydroxyl. 3. propane. Journal of Physical Chemistry, 90: 1949–54. Droege, A.T. and Tully, F.P. (1987a) Hydrogen-atom abstraction from alkanes by OH. 6. cyclopentane and cyclohexane. Journal of Physical Chemistry, 91: 1222–5. Droege, A.T. and Tully, F.P. (1987b) Kinetics of the reaction of the hydroxyl radical with dimethyl ether and diethyl ether. International Journal of Chemical Kinetics, 19: 251–9.

462


Enghoff, M.B., Von Hessberg, P., Nielsen, C.J., et al. (2003) The kinetic isotope effects in the reactions of four ethene isotopologues with chlorine and bromine atoms. Journal of Physical Chemistry A, 107 (39): 7667–70. Farkas, E., Szilágyi, I., Dóbé, S., et al. (2003) Kinetic isotope effect in the reaction of OH radical with acetone-d6. Reaction Kinetics and Catalysis Letters, 80: 351–8. Feilberg, K.L., Johnson, M.S. and Nielsen, C.J. (2004) Relative reaction rates of HCHO, HCDO, DCDO, H13 CHO, and HCH18 O with OH, Cl, Br, and NO3 radicals. Journal of Physical Chemistry A, 108 (36): 7393–8. Feilberg, K.L., Griffith, D.W.T., Johnson, M.S., et al. (2005a) The 13 C and D kinetic isotope effects in the reaction of CH4 with Cl. International Journal of Chemical Kinetics, 37 (2): 110–18. Feilberg, K.L., D’Anna, B., Johnson, M.S., et al. (2005b) Relative tropospheric photolysis rates of HCHO, H13 CHO, HCH18 O, and DCDO measured at the European photoreactor facility. Journal of Physical Chemistry A, 109: 8314–19. Francey, R.J., Tans, P.P., Allison, C.E., et al. (1995) Changes in oceanic and terrestrial carbon uptake since 1982. Nature, 373: 326–30. Funk, J.L., Mak, J.E. and Lerdau, M.T. (2004) Stress-induced changes in carbon sources for isoprene production in populus deltoides. Plant, Cell and Environment, 24: 747–55. Fung, I., Field, C.B., Berry, J.A., et al. (1997) Carbon 13 exchanges between the atmosphere and biosphere. Global Biogenochemical Cycles, 11 (4): 507–33. Gerst, S. and Quay, P. (2000) The deuterium content of atmospheric molecular hydrogen: Method and initial measurements. Journal of Geophysical Research, 105 (D21): 26433–45. Gerst, S. and Quay, P. (2001) Deuterium component of the global molecular hydrogen cycle. Journal of Geophysical Research, 106 (D5): 5021–31. Gierczak, T., Talukdar, R.K., Herndon, S., et al. (1997) Rate coefficients for the reactions of hydroxyl radicals with methane and deuterated methanes. Journal of Physical Chemistry A, 101 (17): 3125–34. Glasius, M., Wessel, S., Christensen, C.S., et al. (2000) Sources to formic acid studied by carbon isotopic analysis and air mass characterization. Atmospheric Environment, 34: 2471–9. Gola, A.A., D’Anna, B., Feilberg, K.L., et al. (2005) Kinetic isotope effects in the gas phase reactions of OH and Cl with CH3 Cl, CD3 Cl, and 13 CH3 Cl. Atmospheric Chemistry and Physics Discussion, 5: 3873–98. Goldstein, A.H. and Shaw, S.L. (2003) Isotopes of volatile organic compounds: An emerging approach for studying atmospheric budgets and chemistry. Chemical Reviews, 102: 5025–48. Gray, J.R., Lacrampe-Couloume, G., Gandhi, D., et al. (2002) Carbon and hydrogen isotopic fractionation during biodegradation of methyl tert-butyl ether. Environmental Science and Technology, 36: 1931–8. Greiner, N.R. (1970) Comparison of kinetics of alkane H-atom abstraction by methyl and hydroxyl radicals. Journal of Chemical Physics, 53: 1285–7. Gupta, M., Tyler, S. and Cicerone, R. (1996) Modeling atmospheric δ 13 CH4 and the causes of recent changes in atmospheric CH4 amounts. Journal of Geophysical Research, 101: 22923–32. Hagemann, R., Nief, G. and Roth, E. (1970) Absolute isotopic scale for deuterium analysis of natural waters. Absolute D/H ratio for SMOW. Tellus, 22: 712–15. Harrington, R.R., Poulson, S.R., Drever, J.I., et al. (1999) Carbon isotope systematics of monoaromatic hydrocarbons: Vaporization and adsorption experiments. Organic Geochemistry, 30: 765–75. Harper, D.B., Kalin, R.M., Hamilton, J.T.G., et al. (2001) Carbon isotope ratios of chloromethane of biological origin: Potential tool in determining biological emissions. Environmental Science and Technology, 35: 3616–19. Harper, D.B., Hamilton, J.T.G., Ducrocq, V., et al. (2003) The distinctive isotopic signature of plantderived chloromethane: Possible application in constraining the atmospheric chloromethane budget. Chemosphere, 52: 433–6.


463

Harry, C., Arey, J. and Atkinson, R. (1999) Rate constants for the reactions of OH radicals and Cl atoms with di-n-propyl ether and di-n-butyl ether and their deuterated analogs. International Journal of Chemical Kinetics, 31: 425–31. Hitsuda, K., Takahashi, K., Matsumi, Y., et al. (2001a) Kinetics of the reactions of Cl(2 P1/2 ) and Cl(2 P3/2 ) atoms with C2 H6 , C2 D6 , CH3 F, C2 H5 F, and CH3 CF3 at 298 K. Journal of Physical Chemistry A, 105: 5131–6. Hitsuda, K., Takahashi, K., Matsumi, Y., et al. (2001b) Kinetics of the reactions of Cl(2 P1/2 ) and Cl(2 P3/2 ) atoms with C3 H8 , C3 D8 , n-C4 H10 , and i-C4 H10 at 298 K. Chemical Physics Letters, 346: 16–22. Iannone, R., Anderson, R.S., Rudolph, J., et al. (2003) The carbon kinetic isotope effects of ozonealkene reactions in the gas-phase and the impact of ozone reactions on the stable carbon isotope ratios of alkenes in the atmosphere. Journal of Geophysical Letters, 30: 1684. Iannone, R., Anderson, R.S., Vogel, A., et al. (2004) Laboratory studies of the hydrogen kinetic isotope effects (KIES) of the reaction of non-methane hydrocarbons with the OH radical in the gas phase. Journal of Atmospheric Chemistry, 47: 192–208. Iannone, R., Anderson, R.S., Vogel, A., et al. (2005) The hydrogen kinetic isotope effects of the reactions of n-alkanes with chlorine atoms in the gas phase. Journal of Atmospheric Chemistry, 50: 121–38. Irei, S., Huang, L., Collin, F., et al. (2006) Flow reactor studies of the stable carbon isotope composition of secondary particulate organic matter generated by OH-radical induced reactions of toluene. Atmospheric Environment, 40: 5857–67. Johnson, B.J. and Dawson, A.G. (1990) Collection of formaldehyde from clean air for carbon isotopic analysis. Environmental Science and Technology, 24: 898–902. Johnson, B.J. and Dawson, G.A. (1993) A preliminary study of the carbon-isotopic content of ambient formic acid and two selected sources: Automobile exhaust and formicine ants. Journal of Atmospheric Chemistry, 17 : 123–40. Kalin, R.M., Hamilton, J.T.G., Harper, D.B., et al. (2001) Continuous flow stable isotope method for study of fractionation of δ 13 C fractionation during halomethane production an degradation. Rapid Communications in Mass Spectrometry, 15: 357–63. Kawamura, K. and Watanabe, T. (2004) Determination of stable carbon isotopic compositions of low molecular weight dicarboxylic acids and ketocarboxylic acids in atmospheric aerosol and snow samples. Analytical Chemistry, 76: 5762–8. Keeling, C.D., Mook, W.G. and Tans, P.P. (1979) Recent trends in the 13 C/12 C ratio of atmospheric carbon dioxide. Nature, 277: 121–3. Keppler, F., Kalin, R.M., Harper, D.B., et al. (2004) Carbon isotope anomaly in the major plant C1 pool and its global biogeochemical implications. Biogeosciences, 1: 123–31. Keppler, F., Harper, D.B., Röckmann, T., et al. (2005) New insight into the atmospheric chloromethane budget gained using stable carbon isotope ratios. Atmospheric Chemistry and Physics Discussion, 5: 2403–11. Komatsu, D.D., Tsunogai, U., Yamaguchi, J., et al. (2004) The budget of atmospheric methyl chloride using stable carbon isotopic mass-balance, Poster A51C-0784, AGU Fall Meeting, San Francisco, 2004. Cited from Keppler et al. (2005). Komatsu, D.K., Tsunogai, U., Yamaguchi, J., et al. (2005) A selective unsaturated hydrocarbon subtraction technique for stable carbon isotopic analysis of atmospheric methyl chloride, methyl bromide, and C2 –C5 saturated hydrocarbons using continuous-flow isotope ratio mass spectrometry. Rapid Communications in Mass Spectrometry, 19: 477–83. Kwok, E.S.C. and Atkinson, R. (1995) Estimation of hydroxyl radical reaction rate constants for gas-phase organic compounds using a structure-reactivity relationship: An update. Atmospheric Environment, 29: 1685–95. Lowe, D.C., Schmidt, U. and Ehhalt, D.H. (1980) A new technique for measuring atmospheric formaldehyde [CH2 O]. Geophysical Research Letters, 7: 825–8.

464


Lowe, D.C., Brenninkmeijer, C.A.M., Brailsford, G.W., et al. (1994) Concentration and 13 C records of atmospheric methane in New Zealand and Antarctica: Evidence for changes in methane sources. Journal of Geophysical Research, 99: 16913–25. Lowe, D.C., Allan, W., Manning, M.R., et al. (1999) Shipboard determinations of the distribution of 13 C in atmospheric methane in the pacific. Journal of Geophysical Research, 104: 26125–35. Mancini, S.A., Ulrich, A.C., Lacrampe-Couloume, G., et al. (2003) Carbon and hydrogen isotopic fractionation during anaerobic biodegradation of benzene. Applied and Environmental Microbiology, 69: 191–8. Manning, M.R. (1999) Characteristic modes of isotopic variations in atmospheric chemistry. Geophysical Research Letters, 26: 1263–6. Manning, M.R., Brenninkmeijer, C.A.M. and Allan, W. (1997) Atmospheric carbon monoxide budget of the southern hemisphere: Implications of 13 C/12 C measurements. Journal of Geophysical Research, 102: 10673–82. Matthews, D.E. and Hayes, A.J. (1978) Isotope ratio monitoring gas chromatography-mass spectrometry. Analytical Chemistry, 50: 1465–73. McCauley, S.E., Goldstein, A.H. and DePaolo, D.J. (2001) An isotopic approach for understanding the CH3 Br budget of the atmosphere. Proceedings of National Academy of Science of USA, 96: 10006–9. McKeen, S.A., Trainer, M., Hsie, E.-Y., et al. (1990) On the indirect determination of atmospheric OHradical concentrations from reactive hydrocarbon measurements. Journal of Geophysical Research, 95: 7493–500. McKeen, S.A., Liu, S.C., Hsie, E.-Y., et al. (1996) Hydrocarbon ratios during PEM-WEST A: A model perspective. Journal of Geophysical Research, 101: 2087–109. McKenna, D.S. (1997) Analytic solution of reaction diffusion equations and implications for the concept of an air parcel. Journal of Geophysical Research, 102: 19719–13725. McKenna, D.S., Hord, C.J. and Kent, J.M. (1995) Hydroxyl radical concentrations and kuwait oil fire emission rates for March 1991. Journal of Geophysical Research, 100: 26005–26. Miller, L.G., Kalin, R.M., McCauley, S.E., et al. (2001) Large carbon isotope fractionation associated with oxidation of methyl halides by methylotrophic bacteria. Proceedings of the National Academy of Science of USA, 98: 5833–7. Miller, L.G., Warner, K.L., Baesman, S.M., et al. (2004) Degradation of methyl bromide and methyl chloride in soil microcosms: Use of stable C isotope fractionation and stable isotope probing to identify reactions and the responsible microorganisms. Geochemica Cosmochemica Acta, 68: 3271–83. Mook, W.G., Koopmans, M., Carter, A.F., et al. (1983) Seasonal latitudinal, and secular variations in the abundance and the isotope ratios of atmospheric carbon dioxide. Journal of Geophysical Research, 88: 10915–33. Niki, H., Maker, P.D., Savage, C.M., et al. (1978) Relative rate constants for the reaction of hydroxyl radical with aldehydes. Journal of Physical Chemistry, 82: 132–4. Niki, H., Maker, P.D., Savage, C.M., et al. (1984) Fourier transform infrared study of the kinetics and mechanism for the reaction of hydroxyl radical with formaldehyde. Journal of Physical Chemistry, 88: 5342–4. Parrish, D.D., Hahn, C.J., Williams, E.J., et al. (1992) Indications of photochemical histories of pacific air masses from measurements of atmospheric trace species at Point Arena, California. Journal of Geophysical Research, 97: 15883–901. Rieley, G. (1994) Derivatization of organic compounds prior to gas chromatographic-combustionisotope ratio mass spectrometric analysis: Identification of isotope fractionation processes. Analyst, 119: 915–19. Roberts, J.M., Fehsenfeld, F.C., Liu, S.C., et al. (1984) Measurements of aromatic hydrocarbon ratios and NOx concentrations in the rural troposphere: Observation of air mass photochemical aging and NOx removal. Atmospheric Environment, 18: 2421–32.


465

Rudolph, J. (2002) Stable carbon ratio measurements: A new tool to understand atmospheric processing of volatile organic compounds. Proceedings of the NATO Advanced Research Workshop on Global Atmospheric Change and its impact on regional air quality, Irkutsk, Russian Federation, 21–27 August 2001. In: I. Barnes (Ed.) Global Atmospheric Change and Its Impact on Regional Air Quality, NATO Science Series, IV. Earth and Environmental Sciences – Vol. 16. Kluwer Academic Publishers, Dordrecht. Rudolph, J. and Czuba, E. (2000) On the use of isotopic composition measurements of volatile organic compounds to determine the ‘photochemical age’ of an air mass. Geophysical Research Letters, 27: 3865–8. Rudolph, J. and Johnen, F.J. (1990) Measurements of light atmospheric hydrocarbons over the Atlantic in regions of low biological activity. Journal of Geophysical Research, 95: 20583–91. Rudolph, J., Lowe, D.C., Martin, R.J. and Clarkson, T.S. (1997) A novel method for compound specific determination of C13 in volatile organic compounds at ppt levels in ambient air. Geophysical Research Letters, 24: 659–62. Rudolph, J., Czuba, E. and Huang, L. (2000) The stable carbon isotope fractionation for reactions of selected hydrocarbons with OH-radicals and its relevance for atmospheric chemistry. Journal of Geophysical Research, 10: 29329–46. Rudolph, J., Czuba, E., Norman, A.L., et al. (2002) Stable carbon composition of nonmethane hydrocarbons in emissions from transportation related sources and atmospheric observations in an urban atmosphere. Atmospheric Environment, 36: 1173–81. Rudolph, J., Anderson, R.S., Czapiewski, K.V., et al. (2003) The stable carbon isotope ratio of biogenic emissions of isoprene and the potential use of stable isotope ratio measurements to study photochemical processing of isoprene in the atmosphere. Journal of Atmospheric Chemistry, 44: 39–55. Saito, T., Tsunogai, U., Kawamura, K., et al. (2002) Stable carbon isotopic compositions of light hydrocarbons over the western north pacific and implication for their photochemical ages. Journal of Geophysical Research, 107: ACH 2/1–ACH 2/9. Sakugawa, H. and Kaplan, R.I. (1995) Stable carbon isotope ratio measurements of atmospheric organic acids in Los Angeles, California. Geophysical Research Letters, 22: 1509–12. Santrock, J., Studley, S.A. and Hayes, J.M. (1985) Isotopic analyses based on the mass spectrum of carbon dioxide. Analytical Chemistry, 57: 1444–8. Saueressig, G., Crowley, J.N., Bergamaschi, P., et al. (2001) Carbon 13 and D kinetic isotope effects in the reactions of CH4 with O(1 D) and OH: New laboratory measurements and their implications for the isotopic composition of stratospheric methane. Journal of Geophysical Research, 106: 23127–38. Sharkey, T.D., Loreto, F., Delwiche, C.F., et al. (1991) Fractionation of carbon isotopes during biogenesis of atmospheric isoprene. Plant Physiology, 97: 463–6. Singleton, D.L., Paraskevopoulos, G., Irwin, R.S., et al. (1988) Rate and mechanism of the reaction of hydroxyl radicals with formic and deuterated formic acids. Journal of American Chemical Society, 110: 7786–90. Sirju, A.-P. and Shepson, P.B. (1995) Laboratory and field investigation of the DNPH cartridge technique for the measurement of atmospheric carbonyl compounds. Environmental Science Technology, 29 (2): 384–92. Slater, G.F., Lollar, B.S., Sleep, B.E., et al. (2001) Variability in carbon isotopic fractionation during biodegradation of chlorinated ethenes: Implications for field applications. Environmental Science Technology, 35: 901–7. Smallwood, B.J., Philp, R.P. and Allen, J.D. (2002) Stable carbon isotopic composition of gasolines determined by isotope ratio gas chromatography mass spectrometry. Organic Geochemistry, 33: 149–9. Stevens, C.M. and Wagner, A.F. (1989) The role of isotope fractionation in atmospheric chemistry. Zeitschrift für Naturforschung, 44a: 376–84.

466


Stutz, J., Ezell, M.J. and Finlayson-Pitts, B.J. (1997) Inverse kinetic isotope effect in the reaction of atomic chlorine with C2 H4 and C2 D4 . Journal of Physical Chemistry A, 101: 9187–90. Stutz, J., Ezell, M.J., Ezell, A.A., et al. (1998) Rate constants and kinetic isotope effects in the reactions of atomic chlorine with n-butane and simple alkenes at room temperature. Journal of Physical Chemistry A, 102: 8510–19. Taatjes, C.A., Christensen, L.K., Hurley, M.D., et al. (1999) Absolute and site-specific abstraction rate coefficients for reactions of Cl with CH3 CH2 OH, CH3 CD2 OH, and CD3 CH2 OH between 295 and 600 K. Journal of Physical Chemistry A, 103: 9805–14. Tanner, R.L., Ziclinska, B., Uberna, E., et al. (1996) Concentration of carbonyl compounds and the carbon isotope ratios of formaldehyde at a coastal site in nova scotia during the NARE summer intensive. Journal of Geophysical Research, 101: 28961–70. Thompson, A., Rudolph, J., Rohrer, F., et al. (2003) Concentration and stable carbon isotopic composition of ethane and benzene using a global 3D isotope inclusive chemical tracer model. Journal of Geophysical Research, 108: D13. Thompson, A.E. (2003) Stable Carbon Isotope Ratios of Nonmethane Hydrocarbons and Halocarbons in the Atmosphere. PhD. Thesis, York University, Toronto. Thompson, A.E., Anderson, R.S., Rudolph, J., et al. (2002) Stable carbon isotope signatures of background tropospheric chloromethane and CFC113. Biogeochemistry, 60: 191–211. −C5 hydrocarbons Tsunogai, U., Yoshida, N. and Gamo, T. (1999) Carbon isotopic composition of C2 − and methyl chloride in urban, coastal, and maritime atmospheres over the western north pacific. Journal of Geophysical Research, 104: 16033–9. Tully, F.P., Koszykowski, M.L. and Binkley, J.S. (1985) Hydrogen-atom abstraction from alkanes by hydroxyl radical. I. Proceedings of the 20th Neopentane and Neooctane. Symposium (International) on Combustion, 715–21. Tully, F.P., Goldsmith, J.E.M. and Droege, A.T. (1986) Hydrogen atom abstraction from alkanes by hydroxyl. 4. Isobutane. Journal of Physical Chemistry, 90: 5932–7. Tyler, S.C., Ajie, H.O., Rice, A.L., et al. (2000) Experimentally determined kinetic isotope effects in the reaction of CH4 with Cl: Implications for atmospheric CH4 . Geophysical Research Letters, 27: 1715–18. Vairavamurthy, A., Roberts, J. and Newman, L. (1992) Methods for determination of low-molecularweight carbonyl-compounds in the atmosphere. Atmospheric Environment A, 26: 1965–93. Wallington, T.J., Dagaut, P. and Kurylo, M.J. (1988) Correlation between gas-phase and solutionphase reactivities of hydroxyl radicals toward saturated organic compounds. Journal of Physical Chemistry, 92: 5024–8. Ward, J.A.M., Ahad, J.M.E., Lacrampe-Couloume, G., et al. (2000) Hydrogen isotope fractionation during methanogenic degradation of toluene: Potential for direct verification of bioremediation. Environmental Science and Technology, 34: 4577–81. Wen, S., Feng, Y., Wang, X., et al. (2004) Development of a compound-specific isotope analysis method for acetone via 2,4-dinitrophenylhydrazine derivatization. Rapid Communications in Mass Spectrometry, 18: 2669–72. Wen, S., Feng, Y., Yu, Y., et al. (2005) Development of a compound-specific isotope analysis method for atmospheric formaldehyde and acetaldehyde. Environmental Science and Technology, 39: 6202–7. Yeh, H.-W. and Epstein, S. (1981) Hydrogen and carbon isotopes of petroleum and related matter. Geochimica and Cosmochimica Acta, 45: 753–62.


Chapter 11

Comprehensive Two-Dimensional Gas Chromatography Jacqueline F. Hamilton and Alastair C. Lewis

11.1

Introduction

One of the greatest difficulties in analysing the organic content of the atmosphere is the sheer number of individual species present. The range of organic compounds in the atmosphere is vast, with emission sources that are complex and varied, being both anthropogenic and biogenic in origin. In urban areas, the troposphere is dominated by petrochemical emissions, ranging in size from 2 to upwards of 30 carbon atoms, and their oxidation products formed through reactions with OH, O3 and NO3 . As molecular weight increases, the number of possible compounds increases exponentially. An example of this is the monoaromatic compounds; with one carbon substituent there is only one possible isomer, toluene. However, with five carbon substituents the number of possible isomers rises to 90 and to ∼450 when there are six carbons on the ring. In areas influenced by biogenic emissions, the number and type of compounds emitted are dependent on factors such as plant type, temperature, season and light conditions. Gas chromatography (GC) coupled with flame ionisation detection (FID) or mass spectrometry are the most common methods for the analysis of complex organic mixtures, such as those found in atmospheric samples, and have been described in the previous chapter. However, the separation power of GC is limited, and even with extremely long, narrow bore columns, the total number of species that can be isolated is considerably lower than the vast range of volatile organic compounds (VOCs) present in the atmosphere. A recent development in complex mixture analysis has been the introduction of comprehensive two-dimensional gas chromatography (GC × GC), which was pioneered by J. Phillips in the early 1990s (Liu and Phillips 1991; Phillips and Xu 1995), but not applied routinely to atmospheric samples until 2000 (Lewis et al. 2000). GC × GC is a hyphenated chromatographic technique, involving the coupling of two GC columns, providing greater resolution of organic mixtures than ever before. It has found use in a number of branches of atmospheric science, including field measurements, smog chamber studies and aerosol analysis. In this chapter, a description of the fundamentals of comprehensive two-dimensional gas chromatography will be presented, followed by a number of examples of its use in the analysis of VOCs in the atmosphere.

468


Primary column separation Second column separation Figure 11.1 Peak profiles of primary and secondary peaks in comprehensive GC.

11.2

Fundamentals of comprehensive gas chromatography

In comprehensive two-dimensional gas chromatography, generally abbreviated as GC × GC, two columns of different selectivity are coupled together via a mid-point modulator. The modulator sends discrete bundles or ‘plugs’ of eluent from the first column to the second, effectively re-injecting small partially separated aliquots of the original sample. The two columns have different stationary phases and should separate chromatographically according to different analyte properties. One of the most common column combinations is a non-polar primary column, which separates on the basis of volatility and a polar secondary column, which separates on the basis of polarity. The modulator must pre-concentrate a bundle of eluent and then launch it onto the second column for GC × GC to function; otherwise, the peak capacity of the system is simply the sum of the two columns. The second column’s chromatographic separation must also be short so that the analysis is complete before the next bundle of eluent is launched from the modulator. This produces a series of short second-dimension chromatograms. Peak widths of analytes eluting from the first column should be at least three to four times the modulation frequency. Thus, each analyte peak eluting from the first dimension will be present in at least three secondary chromatograms. Figure 11.1 shows the relationship between the peak profile of an analyte as it elutes from the primary and secondary columns. In GC × GC, it is essential that the eluent is injected onto the second column as a narrow band. This most obviously aids resolution on the second column, but can also greatly improve the signal/noise ratio and, thus, sensitivity of the technique. Peaks reaching the detector are much narrower and much higher than the primary peak due to full mass conservation, as can be seen in Figure 11.1. A recent study (Lee et al. 2001) has shown that signal enhancement using GC × GC can be modelled to be up to a factor 70 over one-dimensional separations, albeit neglecting the influence of additional sources of noise arising from data capture at high speeds.

11.2.1

Peak capacity

Unlike heart-cut GC, where a small section of the eluent is transferred to a second column, in GC × GC the entire eluent is transferred in discrete bundles to the second column. The overall resolving power of a chromatographic column can be described in terms of peak capacity, n. This is the maximum number of component peaks that can be theoretically resolved on a given column. The peak capacity can be related to the number of theoretical plates (N ) as shown in Equation 11.1 (Giddings 1990 and references therein).

Comprehensive Two-Dimensional Gas Chromatography

469

A chromatographic column, of length, L, can be separated into a series of identical segments called theoretical plates, N . The process of partitioning can be thought of as a series of absorption and desorption events. Each interaction with the stationary phase can be thought of as a ‘plate’ for separation. n ≈ 1 + N 1/2

(11.1)

This equation assumes that the peaks will be evenly spaced throughout the chromatogram. In real samples, however, this is not true, and the actual peak capacity is much lower than that calculated using Equation 11.1. The statistical model of overlap (Davis and Giddings 1983) takes into account the random way in which component peaks fall over the separation space. This model showed that in chromatograms lacking uniform spacing, resolution was reduced due to overlapping peaks. The number of single-component peaks was actually no more than 18% of the potential peak capacity. In heart-cut GC, the maximum peak capacity is given by the sum of the two columns’ respective peak capacity, as shown below: n1 + n2 = ntotal

(11.2)

where n1 and n2 are the peak capacities of columns one and two, respectively, and ntotal is the peak capacity of the entire system. In GC × GC, the theoretical peak capacity for orthogonal separations approaches that of the product of the two columns respective peak capacities (Liu et al. 1995). n1 × n2 ≈ ntotal

(11.3)

In GC × GC, there is generally a degree of retention correlation between dimensions (there is an element of volatility selectivity in all GC), and the actual peak capacity is somewhat lower than the theoretical value. However, using two columns with only modest peak capacities can yield very high total peak capacities due to this product relation, which is simply not possible in heart-cut GC. If the primary and secondary column peak capacities are 50 and 20, respectively, typical values for this type of separation, the capacity of the system is 1 000. This is already well beyond the limits of conventional GC. Optimising the system can improve this figure dramatically, giving peak capacities in the tens of thousands (Dalluge et al. 2002).

11.2.2

Orthogonality

One of the key features of GC × GC is the orthogonality of the system. The separations performed on the two columns must be independent of each other to ensure maximum use of the two-dimensional separation plane. The retention of an analyte on one column must not be related to its retention on the other column. In GC × GC, the primary column separation is long, and the temperature in the oven is ramped throughout the run. However, the second column separation is very fast in comparison, on the order of a few seconds. Using conventional ramping rates, for example 1–3◦ C/min, the second column separation can essentially be considered as isothermal. Where both columns are based on similar stationary phases, that is, a 100% dimethyl-polysiloxane primary column and a 50% phenyl dimethyl-polysiloxane secondary column, this difference is of utmost importance. When

470


Signal A B

5.0

0

2.5 30 60

Primary retention time (s)

0.0

Secondary retention time (s)

Figure 11.2 Relationship between one-dimensional, two-dimensional and the contour plot in GC × GC.

columns are properly tuned, the volatility part of the second column’s retention mechanism is cancelled out as the entire second column separation occurs at the same temperature. This allows a polarity-based secondary column separation.

11.2.3

Visualisation

GC × GC produces a series of very short second-dimension chromatograms. These can be reconstructed and visualised as a contour plot where the retention time on columns one and two represent the X- and Y-axis, respectively, and the peak height is represented as a coloured contour. The process of producing a GC × GC chromatogram is shown in Figure 11.2. Computer software is used to convert a single-stream one-dimensional chromatogram, into a two-dimensional matrix, which can be viewed using a visualisation software package, a number of which are now commercially available.

11.2.4

Ordered chromatograms

One of the most useful features of GC × GC is the production of ordered chromatograms. GC × GC can create very high peak capacities; however, the effectiveness of this increased peak capacity is highly dependant on the distribution of component peaks. If compounds with similar structures are ordered within the separation space, then it is more likely that the effective peak capacity will be high (Davis and Giddings 1983). Giddings (1990) introduced the idea of sample dimensionality to describe the complexity of a sample with respect to predicting order/disorder in multidimensional separations. A series of n-alkanes require only one chemical variable to sufficiently specify each member, for example, volatility. Giddings demonstrated that the degree of order within a chromatogram was determined by the sample dimensionality and the separation system dimensionality. The system dimensionality refers to the number of separation mechanisms, with different selectivity used in the system. Thus, the dimensionality in conventional GC and GC × GC is one and two, respectively. In the case of complex hydrocarbon mixtures, the dimensionality is also two, and thus the system and sample dimensionality match. Ledford et al. (1996) showed the analysis of


471

Retention time 2 (s)

0.5 1.0

2

1

1.5 3 2.0 2.5

4

3.0 3.5 0

10

20

30

40

50

Retention time 1 (min)

Figure 11.3 Structured GC × GC chromatogram of gasoline. Red square is expanded in lower section. 1, alkanes; 2, alkenes and cycloalkanes; 3, monoaromatics; 4, polyaromatics. This image appears in full colour in the plate section that follows page 268 as Plate 7.

petroleum products by GC × GC produced structured chromatograms, where groups of similar analytes formed bands across the separation plane. This allows maximum use of the retention space, which in turn, maximises effective peak capacity. This structured nature is shown in Figure 11.3, for the analysis of a gasoline sample. As retention time on column one increases, a corresponding decrease in analyte volatility is seen. Thus, within a specific group, for example, the alkanes in group 1, moving from left to right of the chromatogram generally indicates an increasing carbon number. This is mirrored in the Y-axis, where an increasing retention time corresponds to an increase in polarity of the separated analytes, thus moving from the alkanes (group 1) to the polyaromatics (group 4). This greatly aids peak identification, with compounds arranged according to their chemical and structural properties. The red square in Figure 11.3 is expanded in the lower section and shows a common feature of structured GC × GC chromatograms. The ‘roof tile’ structure is apparent where each tile corresponds to an isobaric group, in this case monoaromatics isomers with C4 , C5 , C6 and C7 substitution. These visual representations can simplify pattern matching within a series of samples, in comparison to one-dimensional chromatograms.

11.3

Modulators

The key component of the GC × GC system is the modulator, which delivers discrete bundles of eluent from the primary to the secondary column. The modulator must concentrate a fraction and launch it onto the second column in a narrow band to ensure maximum resolution on the second column. A number of possible modulation devices have

472


been introduced, including thermal methods, both heated and cryogenic and valve-based modulators. Almost all types of GC × GC modulators have been used in the analysis of VOCs in the atmosphere, with the choice highly dependent on the analytes of interest.

11.3.1

Thermal modulators

The original modulator introduced by Liu and Phillips (1991), the thermal desorption modulator, incorporated a piece of capillary tubing coated with thin layers of gold or aluminium. The modulator is resistively heated using a pulse of electrical current. The analytes are trapped on a thick film of stationary phase in the modulator tube and are subsequently desorbed when the modulator is heated. The analytes are then swept onto the secondary column by the carrier gas. This modulator was found to be unreliable as the application of the thin layer of metallic paint can be non-uniform and can cause local overheating and a temperature gradient within the modulator. Electrically heated modulators were eventually replaced by mechanical movement modulators with two different approaches, both being highly successful. The first is based on rotating mechanical heater, termed ‘the sweeper’, and the second based on cryogenic cooling. Heated modulators have not found usage in atmospheric analysis as a result of the high maintenance required, for example, the repeated heating cycle can cause degradation of the stationary phase. The heater must also be approximately 100◦ C above the analyte boiling point, producing a volatility range restriction. The second type of thermal mechanical modulator involves the use of a cryogenic trap and was pioneered by Marriott and Kinghorn (1997) for the analysis of essential oils, and its use extended to atmospheric samples in 2000 (Lewis et al. 2000). The chromatographic column, usually the second dimension, fits through the centre of a small chamber or trap cooled using CO2 . The trap is physically moved back and forth along the column. The effluent from the first column is condensed as it enters the region cooled by the trap and accumulates into a small band of solutes. The cryogenic trap then moves position, and the oven heats up the short length of column, which was under the trap. The plug of solutes rapidly heats up, vaporises and is swept onto the second column by the carrier gas. The trap can be moved by a number of means including pneumatic, electrical or stepper motor. The most modern design is the Longitudinally Modulated Cryogenic System (LMCS) (Kinghorn et al. 2000), which incorporates a fast-acting solenoid valve and a pneumatic ram, or stepper motor, mounted on a stand. Rubber spacers are placed between the mounting plate and the GC oven chassis to minimise vibrations. Liquid CO2 is used as the coolant, and dry nitrogen or air purge is used to prevent ice build up. A dual-stage jet modulator has also been introduced, where two jets of cryogen are sprayed onto the capillary column, to trap, focus and release analytes (Beens et al. 2001; Ledford and Billesbach 2000). The release of analytes can be achieved using the ambient oven temperature alone or by the use of hot air jets. Original jet modulators used liquid CO2 as the cryogen, but recent improvements allow the use of liquid N2 or N2 gas that has been cooled by passing through tubing within a liquid N2 Dewar. A secondary oven may also be housed within the GC oven, to hold the second column. It is usually held at a higher temperature than the main oven to ensure no thermal cold spots. The sequential trapping mechanism of the cryojet modulator is shown in Figure 11.4. The


473

Column 1: Non-polar phase Column 2: Polar phase Heater 1 Off and Heater 2 Off

Heaters Off

1

4

1

2

1

2 Analytes partially resolved on Column 1

Cryogenic Cooling On

1

2

Stage 1 returns to thermal trapping mode and the analytes are trapped on Stage 2 of the thermal modulator Cryo 1 On and Cryo 2 On 1

2

Heater 1 Off and Heater 2 On

2

1

2

1

2

5 Analytes trapped on Stage 1 of the thermal modulator

1

2

1

2

Analytes released to Column 2

Cryo 1 On and Cryo 2 Off Heater 1 On and Heater 2 Off

3

1

2

1

2

Heater 1 Off and Heater 2 Off

6 Analytes released to Stage 2 of the thermal modulator

Cryo 1 Off and Cryo 2 On

1

2

Analytes separate on Column 2 next bands enter the thermal modulator Cryo 1 On and Cryo 2 On 1

2

Figure 11.4 Modulation using LECO Thermal Quad Jet Modulator. Figure reproduced courtesy of LECO Corporation, St. Joseph, MI. Licensed by Zeox.

jets use dual stage trapping to ensure minimum bandwidth and breakthrough of volatile analytes. Two-stage modulation can also be achieved using a single jet and a delay loop (http://www.zoex.com/TechnicalNote_word2.doc). The column passes the cold jet and is looped around and passes under the cold jet again. Analytes are trapped as they enter the cooled region as previously. When the hot jets are activated, the analytes are released and travel through the loop, usually 0.6–1 m. The cold jets are then reactivated and the analytes trapped a second time, as they pass through the jets. Cryogenic modulations allow the range of volatilities amenable to GC × GC to be increased, compared to heated thermal methods. The majority of atmospheric applications of GC × GC have involved the use of cryogenic modulators, particularly in aerosol analysis. However, there are limitations, the most important of which relates to field measurements, where the use of a cryogen is avoided wherever possible. In addition, highly volatile and polar compounds are prone to breakthrough from the trap, causing very wide peaks and a severe loss of chromatographic efficiency.

11.3.2

Valve modulators

In valve modulation, narrow pulses of eluent are diverted from the primary column to the secondary column using a switching valve. Opiteck et al. (1998) have demonstrated the use of valves in two-dimensional liquid chromatography, and valves are regularly used in

474


heart-cut GC. The first valve-based GC × GC system, incorporated a multi-port diaphragm valve, housed within the GC oven (Bruckner et al. 1998). Most of the primary eluent was vented to waste and small bundles transferred to the second column at a high frequency (2 Hz). This method of valve-based modulation was successful and, although not truly comprehensive, as most of the eluent is vented, can be considered as such. The second valve-based GC × GC system was introduced by Seeley et al. (2000) and incorporated a six-port diaphragm valve with sample loop, housed within a secondary oven, as the modulation device. In the sample position, primary eluent flowed through the sample loop, while a secondary carrier gas flow is simultaneously used to run the second column chromatography. In the load position, the secondary gas supply pushes the loop contents on the second column, and the primary eluent is vented. Seeley et al. used a high secondary column flow rate relative to the first (at least 20 times higher) to ensure both maximum transfer of analytes between columns and minimum injected plug thickness. The valve was in the sample position for 0.8 s and in the launch position for 0.2 s, and, thus, approximately 80% of the sample was transferred. Valves have found only very limited use in GC×GC, but have some significant advantages over thermal methods for atmospheric monitoring (Hamilton et al. 2003a). Valve modulators are independent of temperature, and, when the valve is housed within the GC oven, analytes are transferred from column one to column two at the ambient oven temperature. This is of particular importance when very high volatility or polar compounds are being investigated, such as those found in gas-phase atmospheric samples. In cryogenic modulation, breakthrough can occur in the high-volatility region of the analysis as the temperature is not sufficiently low to trap analytes and, in general, species with less than five carbons are not effectively trapped. The valve, however, suffers no such breakthrough problems and is independent of analyte volatility. In addition, the valve requires no cryogen, making it much more suited to field studies than cryogenic modulation. The main disadvantage of the valve system is the loss of sensitivity compared to thermal modulation methods, when venting most of the sample to waste. However, a recent study has shown that, provided the first-dimension peak is adequately sampled, valve modulation retains the peak height of the one-dimensional peak and, in some cases, may provide some degree of peak amplitude enhancement (Hamilton et al. 2003a).

11.4

Detectors

In GC × GC, the peaks produced during the secondary separation are inherently narrow, in the order of tens to hundreds of milliseconds. To properly define a peak, at least ten data points should be collected across the peak, with between 20 and 50 being optimum. Therefore, the detection system must be capable of collecting data in the tens to hundred hertz range. The most common detector used in GC × GC is the FID, which is ideal for hydrocarbon analysis (Phillips and Beens 1999), giving a universal response for hydrocarbons. The disadvantage of the FID is the lack of structural information given, although this is partially overcome by the structured nature of the chromatograms. Electron capture detectors have also been used in GC × GC for the selective detection of polychloro-biphenyl compounds (PCBs) in environmental matrices, but have yet to be used in atmospheric samples (Korytar et al. 2002).


475

The coupling of GC×GC to a mass spectrometer forms an extremely powerful technique, combining the improved resolution of the GC×GC with mass spectral information obtained across the peak. Early attempts used a quadrupole mass spectrometer (Frysinger and Gaines 1999), but the mass spectral acquisition rate was insufficient. A conventional quadrupole usually operates at a scan rate of 1 Hz but can be operated up to about 20 Hz although usually in a single ion mode or with reduced sensitivity. For a 1-s-wide peak, the quadrupole would give very poor peak definition and produce triangular-shaped peaks. In order to obtain faster acquisition rates than are possible using a quadrupole or ion trap, GC × GC can be coupled to a time-of-flight mass spectrometer (TOFMS). A TOFMS can be operated at up to 500 mass scans per second, each scan being the sum of ten pulses. The TOFMS would be able to fully define such a peak without a great loss in sensitivity. The mass analyser in a TOFMS is a cylindrical flight tube. Ions are accelerated by a negatively charged plate and achieve the same kinetic energy. The ions present will be of variable mass, and, thus, the speed imparted to them will also vary. The ions travel down the flight tube at different speeds and are separated according to their m/z ratio. Although coupling of GC×GC to TOFMS has been reported, this is very much a research lab tool rather than fieldwork tool at present. The fast mass spectral acquisition rates of the TOFMS allow adequate peak coverage across the narrow peaks inherent to GC × GC (van Deursen et al. 2000). One of the most interesting new applications of GC×GC-TOFMS is the analysis of organic aerosol (OA) content. GC × GC-TOFMS has been used to separate over 10 000 individual analytes in an urban PM10 sample, a resolution not possible using any current methodologies (Hamilton et al. 2004; Welthagen et al. 2003).

11.5

Examples of GC × GC use in atmospheric samples

In the early days of GC × GC, the primary uses were in the analysis of complex oil samples, including both petrochemical and essentials oils. The high resolving power and structured nature of GC × GC allowed previously impossible separation of complex samples in a short single analysis. In 2000, GC × GC was used for the first time in atmospheric analysis (Lewis et al. 2000). Lewis et al. used a Peltier-cooled multi-bed absorbent trap to pre-concentrate hydrocarbon species in the urban atmosphere of Melbourne, Australia. The trapped analytes were desorbed at high temperature into a GC × GC–FID system, with LMCS cryogenic modulation. The primary column was a 100% dimethyl polysiloxane column and the second a 50% phenyl dimethyl polysiloxane column, to provide a volatility- and polarity-based separation. Figure 11.5 shows the GC×GC chromatogram obtained, alongside a concurrent one-dimensional chromatogram. It is clear that there are significantly more peaks isolated in the GC × GC chromatogram, and, on close inspection, more than 110 peaks are visible, compared to only 20–30 using GC. Large numbers of peaks exist in the higher-boilingpoint ranges above C6 , which are isolated in the GC × GC chromatogram only. In the latter regions of the single column separation, no single component is present at concentrations sufficient to be raised above the baseline. This baseline is, in fact, not a true baseline, but the constant co-elutions of many analytes at low concentrations. The impact of this ‘missing’ carbon in the atmosphere, which was previously undetected, was investigated further by Hamilton and Lewis (2003), with particular emphasis on the monoaromatic fraction. Aromatic hydrocarbons are released into the atmosphere

476


1

2

3

4

40 H

G

30

F 25 E 20 D 15 C B

Retention time on BP-1 column (min)

35

10

A 5

30 20 10 Flame ionization detector response (pA)

1 2 3 4 Retention time on BP-50 columns (s)

Figure 11.5 Comprehensive and one-dimensional separations of VOCs in urban air. A, benzene; B, heptane; C, toluene; D, xylenes; E, C3 -benzenes; F, C4 -benzenes; G, C5 -benzenes; H, naphthalene; 1, aliphatic band; 2, carbonyl band; 3, monoaromatic band; 4, bi-aromatic band. Reproduced with permission of Nature Publishing Group, from: Lewis et al. (2000). A larger pool of ozone-forming carbon compounds in urban atmospheres. Nature, 405: 778–81. This image appears in full colour in the plate section which follows page 268 as Plate 8.

through automotive emissions and industrial and domestic solvent use and have been estimated to represent as much as 20% of anthropogenic non-methane hydrocarbons emissions. The ozone production potential for multi-substituted aromatics is high, increasing with the degree of ring substitution. A recent model calculation predicts that simple aromatics are responsible for about 40% of ozone formation in Northern Europe (Derwent et al. 1996). Previous studies of VOC composition in the atmosphere using GC–MS and GC–FID have provided detailed inventories, and typically monoaromatics with up to three carbons substituents are routinely reported. More detailed inventories have been reported by Grosjean et al. (1998), Sin et al. (2001) and Yassaa et al. (2001) (143 VOCs 26 monoaromatics, 150 VOCs 22 monoaromatics and 190 VOCs 17 monoaromatics respectively) using GC-quadrupole MS. All of these studies can be considered state-ofthe-art, and covered a wide range of not only hydrocarbon but also halocarbon and other


477

gases. The upper limit in these analyses for reporting speciated alkyl substitution has been between 3 and 11 C4 isomers (all three studies), 1–2 C5 isomers (Grosjean et al. and Sin et al.) and 1 C6 isomer by Sin et al. The GC × GC chromatogram obtained in Figure 11.5, indicated that these onedimensional GC studies would isolate only a small fraction of the monoaromatic fraction of the summer time urban atmosphere of Melbourne. Hamilton and Lewis (2003) used a cryogenic and a valve modulator to investigate the monoaromatic content of both atmospheric samples and gasoline vapours, the major emission source in urban areas. Analysis of air collected at a roadside location in the United Kingdom identified a total of 147 monoaromatic species, and a sample chromatogram is shown in Figure 11.6. The enhanced resolution is very clear, and, in the higher boiling regions that are successively expanded at higher gain, in Figure 11.6, there are many hundreds of components isolated.

400 300 200 100 0

0

10

20

30

40

50

60

70

80

90

100

Second column RT/s

3

1 2 3

5 6

A

B

C

D 4

8 9

H

11

Second column RT/s

5

5

10 15 20 25 30 35 40 45 50 55 60 65 70 75 80 First retention time/min

3 4 5D

E

G

F

7 8 9 50

55

61 66 72 Retention time/min

77

82

66

75

84

Figure 11.6 Comparison of single column (upper) and GC × GC separations (lower) of a Leeds urban air sample. Areas of the full chromatogram are successively extracted at higher gain to illustrate increasing isomeric complexity at higher boiling points. GC × GC chromatograms are annotated with start of individual Cx isomer band (running right to left) where A = C2 , B = C3 , C = C4 , D = C5 , E = C6 , F = C7 , G = C8 , H = naphthalene. Chemical banding assignments: 1, aliphatic; 2, olefins; 3, oxygenates; 4, monoaromatics; and 5, polyaromatics. Reproduced with permission of Elsevier, from: Hamilton et al. (2003). Mono aromatic complexity Atmospheric Environment, 37(5): 589–602. This image appears in full colour in the plate section which follows page 268 as Plate 9.

A

B

A

B

Increasing polarity


C

Increasing polarity

478

C

Increasing boiling point Figure 11.7 Comparison of GC × GC chromatograms of gasoline vapours and urban air. Upper: Leeds urban air chromatogram. Lower: Gasoline vapours at 20◦ C chromatogram. A, C3 -substituted monoaromatics; B, C4 -substituted monoaromatics; C, C5 -substituted monoaromatics.Reproduced with permission of Elsevier, from: Hamilton et al. (2003) Mono aromatic complexity. Atmospheric Environment, 37(5): 589–602. This image appears in full colour in the plate section which follows page 268 as Plate 10.

An advantage of the 3D contour plot is that they are much more amenable to visual interpretation than the analogous two-dimensional versions. Pattern matching or source fingerprinting using this technique is greatly enhanced, since it is analyte chemical characteristics rather than sheer abundance that drive visual identifications. Figure 11.7 illustrates this concept, showing a comparison of the GC × GC chromatograms obtained for 1 ml gasoline vapours at 20◦ C and 1 l urban air. Even though different modulation devices have been used (valve and LMCS, respectively), an identical set of separation conditions were employed, and as such the two chromatograms are directly comparable. There are obviously some differences between the two chromatograms, both in terms of content and peak shape. The air chromatogram has additional carbonyls and olefins visible, and the peak shape appears poorer due to working nearer the detection limit of the instrument. In some aromatic regions, the two chromatograms are remarkably similar, with almost identical distributions of C3 –C5 aromatics compounds. Although this is only a qualitative visual matching between source and receptor, it is clear that at this location urban air has a significant input from evaporative/unburnt fuel sources. Overall, the air chromatogram is more complex than the gasoline vapour, particularly in terms of the number of compounds in the slightly polar or oxygenated regions. While the C3 aromatics and C4 monoaromatics are clearly identifiable in both gasoline and air chromatograms and are present in similar relative abundances (when normalised to toluene), as substitution increases, the relative proportion of C5 and greater aromatics in air diminishes significantly when compared to fuel vapour.


479

It was calculated that the effect of these low-concentration aromatics (i.e.

Volatile Organic Compounds in the Atmosphere

Volatile Organic Compounds in the Atmosphere

VOLATILE ORGANIC COMPOUNDS

Volatile Sulfur Compounds in Food

Volatile

Pentavalent nitrogen in organic compounds

ELECTROCHEMISTRY OF ORGANIC COMPOUNDS

Small Ring Compounds in Organic Synthesis I

Small Ring Compounds in Organic Synthesis II

Small Ring Compounds in Organic Synthesis VI

Small Ring Compounds in Organic Synthesis V

Reactive Halogen Compounds in the Atmosphere (Handbook of Environmental Chemistry)

Sigma and Pi Electrons in Organic Compounds

Volatile

Small Ring Compounds in Organic Synthesis IV

Small Ring Compounds in Organic Synthesis VI

Spectrometric Identification of Organic Compounds

Oxidation of Organic Compounds vol.III

Oxidation of Organic Compounds vol.I

Spectra Interpretation of Organic Compounds

Important Biochemicals and Organic Compounds

Spectrometric Identification of Organic Compounds

Spectrometric Identification of Organic Compounds

Oxidation of Organic Compounds vol.II

Radiation in the atmosphere

The Chemistry of Organic Silicon Compounds

The chemistry of organic silicon compounds

The Systematic Identification of Organic Compounds

Turbulence in the Atmosphere

Turbulence in the Atmosphere

Turbulence in the Atmosphere

Volatile Organic Compounds in the Atmosphere