Research Collection · 2020. 3. 26. · CMC que les trois acides organiques precit´ es peuvent´ etre formˆ es´ a partir de l’acide pinonique.` Un quatrieme acide organique avec

Research Collection

Doctoral Thesis

Chemical composition of gas phase and aerosol particles fromenvironmental chambers with ion chromatography and massspectrometry

Author(s): Praplan, Arnaud Patrick

Publication Date: 2012

Permanent Link: https://doi.org/10.3929/ethz-a-007624295

Rights / License: In Copyright - Non-Commercial Use Permitted

This page was generated automatically upon download from the ETH Zurich Research Collection. For moreinformation please consult the Terms of use.

ETH Library

https://doi.org/10.3929/ethz-a-007624295

http://rightsstatements.org/page/InC-NC/1.0/

https://www.research-collection.ethz.ch

https://www.research-collection.ethz.ch/terms-of-use

Diss. ETH No. 20387

Chemical composition of gas phase and aerosol particles from environmental chambers with ion chromatography and mass spectrometry

Arnaud Patrick Praplan

DISS. ETH NO. 20387

CHEMICAL COMPOSITION OF GAS PHASE AND AEROSOL PARTICLESFROM ENVIRONMENTAL CHAMBERS WITH ION CHROMATOGRAPHY

AND MASS SPECTROMETRY

A dissertation submitted to

ETH ZURICH

for the degree of

Doctor of Sciences

presented by

ARNAUD PATRICK PRAPLAN

MSc ETH Chemistry, ETH Zurich

born on 14.07.1984

citizen of Icogne VS

accepted on the recommendation of

Prof. Dr. Alexander Wokaun (examiner)Prof. Dr. Urs Baltensperger (co-examiner)

Dr. Josef Dommen (co-examiner)Prof. Ive Hermans (co-examiner)

2012

Abstract

Air pollution has adverse health effects and influences the climate. It consists of gaseouscompounds present at trace levels (well below 1 % fraction of volume) such as for examplevolatile organic compounds (VOCs), nitrogen oxides (NOx), sulphur dioxide (SO2) or ozone (O3)and of particles in suspension (aerosols) with diameters from a few nanometres up to a fewtens of micrometres. Polluting species can be emitted directly into the atmosphere by anthro-pogenic activities, but they also can be formed through chemical reactions in the atmosphere.

Measuring the chemical composition of trace gases and of sub-micrometre aerosol parti-cles in the atmosphere is a challenging task, due to the great variety of compounds present.Laboratory experiments with environmental chambers are used to observe the evolution of agiven mixture of compounds under controlled conditions, typically close to the atmosphericconditions. Ion chromatography (IC) is a technique that can be used to analyse various kindof samples. Here, it was used to detect ammonia (NH3) and amines (as cations), as well asinorganic anions and organic acids (as anions) from environmental chamber experiments.

NH3 and amines participate in the nucleation processes with sulphuric acid (H2SO4) andwater (H2O) molecules. The environmental chamber of the CLOUD project, which investigatesthe relation between galactic cosmic rays (GCR) and the climate, was used to study the detailsof such nucleation processes. An IC method was used to quantify NH3 and dimethylamine(DMA) at low part per trillion by volume (pptv) levels. With a time resolution of 70 to 210minutes, DMA was measured up to 60 pptv. This species was injected into the chamber inorder to study its effect on nucleation and growth rates. On the other hand, NH3 was notinjected intentionally and was present only as a contaminant. The background mixing ratio ofthis species at 278 K remained mainly below 20 pptv. With such a low level of contaminants,nucleation rates similar to the ones observed in the atmosphere could be reproduced in thepresence of a few pptv of DMA.

Due to their higher oxidation state compared to VOCs, organic acids with similar or highermolar masses are less volatile. Therefore, they can partition into the particle phase, contribut-ing to the formation of secondary organic aerosol (SOA). They may also play a role in the for-mation of new particles. In the last few years, the chemical structure of several organic acidsin α-pinene SOA were elucidated. α-Pinene is a biogenic VOC from the class of monoterpenes(C10H16): First, 3-methyl-1,2,3-butanetricarboxylic acid (MBTCA), which has a high oxygen tocarbon ratio, then terpenylic acid, which displays a lactone ring in its structure, and diater-penylic acid acetate, which comprise an ester functionality. In order to study the formationof those acids from pinonic acid, a primary ozonolysis product of α-pinene oxidation, it wassuggested for this work to use a model compound, cyclobutyl methyl ketone (CMK), which cor-responds to the substructure of interest in pinonic acid. Previous studies demonstrated thatMBTCA is formed from pinonic acid. With IC coupled to a mass spectrometer (MS), the organicacids formed by oxidation of CMK were measured. Even though the formation mechanismof MBTCA remains unclear, it could be shown by analogy with the products formed by CMKoxidation that the three organic acids previously mentioned can be formed from pinonic acid.

iii

A fourth organic acid with nominal mass 188, the analogue of 4-oxobutanoic acid, should alsobe observed in α-pinene SOA and two structures were proposed for this compound. There wasan organic acid detected with nominal mass 146 (possibly diaterpenylic acid acetate), whichwas not formed in the presence of NOx; however, this effect could not be interpreted mechanis-tically. Despite the uncertainties remaining concerning the formation mechanisms of organicacids in the gas phase, evidence was found that even dicarboxylic acids can be formed from aprecursor containing no carboxylic acid functionality. Traditional gas phase chemistry cannotaccount for a reaction pathway, so that new alternative pathways (very likely intramolecularreactions) have to be explored.

With the same IC/MS system, organic acids from the 1,3,5-trimethylbenzene (TMB) photo-oxidation were measured in both gas and particle phases. They were sampled with water bya wet effluent diffusion denuder (WEDD) and an aerosol collector (AC), respectively. Up to 24masses showed one peak or more, but only a small amount of those species could be identifiedunambiguously. This is due to the lack of available standards and knowledge about reactionmechanisms. The amount of water-soluble organic acid present in the particle phase was com-prised between 6 and 14 % in the absence of SO2, but dropped below 3 % when 2 part per billionby volume (ppbv) of SO2 were added to the system. It remains unclear if the reduction is due togas or condensed phase reactions. Partitioning coefficients could be derived and were foundto be higher than the expected theoretical values, which is an indication of reactive uptake orof a sampling artefact.

Various issues remained unsolved, but this work demonstrated the importance of methodsthat can measure trace components of the atmosphere to very low levels and that the actualchemical knowledge is not sufficient to account for the formation of various organic acids inthe gas phase. New reaction mechanisms should be investigated.

iv

Resume

La pollution de l’air a des effets negatifs sur la sante et influence le climat. Elle est constitueede composes gazeux presents sous forme de traces (bien en dessous de 1 % de fraction volu-mique) comme par exemple les composes volatils organiques (COVs), les oxydes d’azote (NOx),le dioxyde de soufre (SO2) ou l’ozone (O3) et de particules en suspension (aerosols) avec desdiametres allant de quelques nanometres a plusieurs dizaines de micrometres. Ces especespolluantes peuvent etre emises directement dans l’atmosphere ou resulter de reactions chim-iques.

Mesurer ces gaz et la composition chimique des particules microscopiques dans l’atmo-sphere est une tache exigeante, etant donne l’immense variete de composes qui s’y trouvent.Des experiences en laboratoire avec des chambres environnementales sont utilisees pour ob-server l’evolution d’un melange donne de composes sous des conditions bien definies, typique-ment proche des conditions atmospheriques. La chromatographie ionique (CI) est une tech-nique qui peut etre utilisee pour analyser toutes sortes d’echantillons. Dans le cas present, ellea ete utilisee pour detecter l’ammoniaque (NH3) et des amines (en tant que cations) ainsi quepour detecter des anions inorganiques et des acides organiques (en tant qu’anions) provenantd’experiences dans des chambres environnementales.

Le NH3 et les amines participent, avec des molecules d’acide sulfurique (H2SO4) et d’eau(H2O), au processus de nucleation. La chambre environnementale du projet CLOUD qui etudiela relation entre les rayons cosmiques et le climat a ete utilisee pour examiner de tels proces-sus. Une methode de CI a ete utilisee pour quantifier le NH3 et la dimethylamine (DMA) a desniveaux de concentration tres bas (partie par billion en volume (pptv), 10−12). Avec un tempsde resolution de 70 a 210 minutes, la DMA a pu etre mesuree jusqu’a 60 pptv. Ce composea ete injecte dans la chambre afin d’etudier son effet sur les vitesses de nucleation et d’ac-croissement. Le NH3, en revanche, n’a pas ete injecte intentionnellement et la proportion demelange residuelle de ce compose est en general situee en dessous de 20 pptv a 278 K. Avec unniveau de contamination de bas, des vitesses de nucleation similaires a celles observees dansl’atmosphere ont pu etre reproduites en presence de seulement quelques pptv de DMA.

De par leur degre d’oxydation plus eleve en comparaison des COVs, les acides organiquesqui ont des masses molaires similaires ou plus elevees sont moins volatils. Ils peuvent doncse condenser sur des particules preexistantes, contribuant ainsi a la formation d’aerosols or-ganiques secondaires (AOS). Ils pourraient egalement jouer un role dans la formation de nou-velles particules. Ces dernieres annees, les structures chimiques de quelques acides organiquesdans l’AOS d’α-pinene ont ete elucidees. L’α-pinene est un COV biogenique de la classe desmonoterpenes (C10H16). Tout d’abord, l’acide 3-methyl-1,2,3-butanetricarboxylique (AMBTC) quipossede un rapport oxygene/carbone eleve, puis l’acide terpenylique qui presente un cycle lac-tone dans sa structure et l’acetate d’acide diaterpenylique, qui comprend un groupe fonction-nel ester. Afin d’etudier la formation de ces acides a partir de l’acide pinonique, un produitprimaire de l’ozonolyse de l’α-pinene, ce travail suggere qu’un compose modele soit utilise : lacyclobutyl methyl cetone (CMC), qui correspond a la sous-structure d’interet de l’acide pinon-

v

ique. Des etudes precedentes ont demontre que l’AMBTC est forme a partir de l’acide pinon-ique. Avec la CI couplee a un spectrometre de masse (SM), les acides organiques formes parl’oxydation de la CMC ont ete mesures. Bien que le mecanisme de formation de l’AMBTC resteobscur, il a pu etre demontre par analogie avec les produits formes lors de l’oxydation de laCMC que les trois acides organiques precites peuvent etre formes a partir de l’acide pinonique.Un quatrieme acide organique avec un masse nominale de 188, le produit analogue a l’acide 4-oxobutanoıque, devrait egalement etre observe dans l’AOS de l’α-pinene et deux structures ontete proposees pour ce compose. Un acide organique a ete detecte avec la masse nominale 146(et pourrait etre l’acetate d’acide diaterpenylique), mais il n’est pas forme en presence de NOx.Cependant, cet effet n’a pas pu etre interprete au niveau du mecanisme de reaction. Malgreles incertitudes restantes concernant le mecanisme de formation des acides organiques enphase gazeuse, des observations demontrent que meme des acides dicarboxyliques peuventetre formes a partir d’un precurseur ne contenant aucun groupe carboxyle. Les reactions chim-iques traditionnelles ne peuvent expliquer une voie de reaction, si bien que de nouvelles voies(tres probablement des reactions intramoleculaires) doivent etre explorees.

Avec la meme configuration CI/SM, les acides organiques formes durant l’oxydation pho-tolytique du 1,3,5-trimethylbenzene (TMB) ont ete analyses dans les phases gazeuse et partic-ulaire. Elles ont ete echantillonnees respectivement grace a un decomposeur et a un collecteurd’aerosol. Un ou plusieurs pics ont pu etre detectes pour vingt-quatre masses differentes, maisseule une petite partie de ces substances a pu etre clairement identifiee. Cela est du au manquede substances etalons disponibles et de connaissances des mecanismes de reaction. La quan-tite d’acides organiques solubles dans l’eau presents dans la phase particulaire etait compriseentre 6 et 14 % en l’absence de SO2, mais s’est reduite a moins de 3 % lorsque 2 parties parmilliard de volume (ppbv, 10−9) de SO2 ont ete ajoutees au systeme. Il est incertain si cettedifference est due a des reactions en phase gazeuse ou particulaire. Des coefficients de par-tition ont ete derives et se trouvaient a des valeurs plus elevees que les valeurs theoriquesescomptees, ce qui indique une absorption reactive ou un artefact d’echantillonnage.

Plusieurs questions restent en suspens, mais ce travail a demontre l’importance des methodescapables de detecter des traces de composes atmospheriques a des niveaux tres bas et que lesconnaissances actuelles ne permettent pas d’expliquer la formation de differents acides or-ganiques en phase gazeuse. De nouveaux mecanismes reactionnels doivent etre etudies.

vi

a David...

Grief causes you to leave yourself. You step outside your narrow little pelt.And you can’t feel grief unless you’ve had love before it — grief is the final

outcome of love, because it’s love lost.

Flow My Tears, The Policeman Said — Philip K. Dick

Acknowledgements

During the time I worked on my Ph.D. thesis, many people helped me in many different ways.I would like to express my gratitude to all those people here. Even if I fail to establish an ex-haustive list, I would like to mention the people that were of particular importance for me inthese last few years.

First, I would like to thank my examiner, Prof. Alexander Wokaun, as well as my co-examiners,Prof. Urs Baltensperger, Dr. Josef Dommen and Prof. Ive Hermans for the interesting and stim-ulating discussions about my work, the nice work conditions at the Paul Scherrer Institute andthe many opportunities they gave me to network with other atmospheric scientists.

A special thank you to Dr. Markus Kalberer, for introducing me to the world of aerosols duringmy master studies at ETH Zurich and for pointing me to available PhD positions at PSI.

Also, I would like to particularly thank Dr. Kathrin Hegyi-Gaggeler, who taught me everythingshe knew about ion chromatography, mass spectrometry, organic acids, the PSI smog chamberand many other things. I really enjoyed sharing an office with her for a few months.

A particular thank you to Mr. Peter Barmet, who joined me in the office after Kathrin gradu-ated. For the work at the smog chamber, for our discussion about free software (as in freedom,not free beer!), chemistry, PTR-MS, politics, music, life, nothing and everything, I could not havehoped to meet a person as nice as him!

To Mr. Federico Bianchi, who first joined our office OFLA/009 as a visiting Ph.D. student, butliked it so much that he decided to stay for a longer period. I am glad that he was interestedin positive ion chromatography for the ammonia measurements at CERN and I am profoundlygrateful that he brought me there together with “my” instrument.

I need to thank all the other “smoggers” that I had the pleasure to work with for the experi-ments at our smog chamber: Dr. Torsten Tritscher, Dr. Peter DeCarlo, Mr. Peter Mertes, Ms. LisaPfaffenberger, and Dr. Jay Slowik.

Merci to my French-speaking friends and colleagues: Dr. Cindy Follonier, Ms. Marie Laborde,Dr. Nolwenn Perron, Dr. Nicolas Leplat, Prof. Michel Rossi, Mr. Michel Tinguely, Mr. Pascal Wyss,Mr. Mael van der Woude, and Dr. Mathieu Hursin. It is always a good feeling to express myselfin French from time to time. I also express my deepest gratitude to all the members of thetheatre group “Les Tretaux de l’Aar”. I had so much fun with you on stage and during therehearsals.

Many thanks to all of the current and former members of the Laboratory for AtmosphericChemistry (students and non-students) that I have not yet mentioned, and who contributedto the nice working atmosphere, the meaningful coffee breaks, and the fun after or beforework (and maybe during work sometimes): Mr. Francesco Riccobono, Mr. Daniel Oderbolz,Mr. Emanuel Hammer, Mr. Peter Zotter, Ms. Monica Crippa, Mr. Riccardo Iannarelli, Mr. Gian-carlo Ciarelli, Dr. Imad El Haddad, Mr. Piotr Kupiszewski, Ms. Bernadette Rosati, Mr. FrancescoCanonaco, Ms. Suzanne Visser, Mr. Robert Wolf, Mr. Stephen Platt, Dr. Paul Zieger, Dr. KatjaRinne, Dr. Nicolas Bukowiecki, Mr. Roman Frohlich, Dr. Martin Gysel, Dr. Johannes Keller, Dr. Olga

ix

Sidorova, Dr. Matthias Saurer, Dr. Markus Furger, Dr. Sebnem Aksoyoglu Sloan, Mr. Rene Richter,Ms. Ineke Lotscher, Mr. Gunther Wehrle, Dr. Andre Prevot, Dr. Ernest Weingartner, and Dr. RolfSiegwolf.

I should not forget a few more alumni that also contributed to the great atmosphere: Dr. Clau-dia Mohr, Dr. Lukas Kammermann, Dr. Iakovos Barmpadimos, Dr. Zsofia Juranyi, Dr. MaartenHeringa, Dr. Agnes Richard, Dr. Rahel Fierz-Schmidhauser, and Dr. Anne Kress.

I want to express my deep gratitude to the CLOUD crowd for welcoming me (and the IC)in the summer of 2010, especially Dr. Jasper Kirkby, but also Dr. Jonathan Duplissy for theadministrative, technical, and logistical support, as well as Prof. Joachim Curtius, Mr. JoaoAlmeida, Ms. Linda Rondo, Ms. Agnieszka Kupc, Mr. Siegfried Schobesberger, Ms. Daniela Wim-mer, Mr. Alessandro Franchin, Mr. Daniel Hauser, Dr. Martin Breitenlechner, Mr. Andrew Dow-nard, Mr. Sebastian Ehrhart, Dr. Katrianne Lehtipalo, Ms. Eimear Dunne, Dr. Andreas Kurten andall the of the people that I may have forgotten, but who made the time in Meyrin very exciting.

Also, many thanks to the co-authors and collaborators in my various publications: Prof. NeilDonahue, Prof. Rainer Volkamer, Ms. Eleanor Waxman, Dr. Mia Frosch, Dr. Yao Liu, Dr. AnneMonod.

I also want to kindly thank Ms. Doris Hirsch-Hoffmann, Ms. Hannelore Kruger, and Ms. EstherSchmid for their support with PSI and ETH administrative tasks.

Sports were a good way of balancing all the cerebral activity, and so I would like to thank allthe members of the badminton section of the PSI-Sportclub, and especially the people respon-sible, Ms. Heide Beer and Mr. Markus Boschung. I also would like to thank all the people thatcame with me to run and/or exercise in the forest around PSI, to play squash, and even wentswimming in the Aare on the hottest summer days.

To my friends, Mr. Martin Muhlheim, Dr. Frederico Lima, Ms. Nadine Bohni, Mr. David Lazaro,Mr. Carlos Garrido, Mr. Tim Dallatomasina, Mr. Normand Beaudry, Mr. Oliver Senn, Mr. TobiasWackernagel, Ms. Paula Martins, Dr. Roland Flury, Ms. Beatrice Marti, Dr. Corey Rice, my friendlycurrent and former neighbours in Ennetturgi, and any other people that I may have forgottento mention here but that contributed to so many interesting discussions with a coffee or tea,to great dinners at home or in restaurants, to nice evenings with movies, drinks, games, hikes,and other activities.

Many thanks to the members of my family that showed regular interest in my work. Manythanks to my sister Ms. Isabelle Praplan for being there when I needed her. For their infinitesupport in so many aspects, I would like to thank my parents, Mr. Serge and Ms. Ginette Pra-plan. Finding the right words to express everything I owe them is such a difficult task...

I would also like to thank Mr. Patrick Fricker for all the great moments that we shared on theground and in the air.

And finally, many thanks to Mr. David Branchina to whom this work is dedicated. He taughtme so many things that one cannot learn at a university. I will never forget the tragic call I goton that Monday morning in August 2008 on my way to PSI just a few weeks after starting myPh.D. and I will never forget him. “Sache que je...”

x

Contents

List of Tables xv

List of Figures xviii

List of Abbreviations xix

1. Introduction 11.1. Chemical composition of the troposphere . . . . . . . . . . . . . . . . . . . . . . . 2

1.1.1. Volatile organic compounds (VOCs) . . . . . . . . . . . . . . . . . . . . . . 21.1.2. Ammonia (NH3) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 41.1.3. Amines . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 41.1.4. Organic acids . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 41.1.5. Tropospheric aerosol . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5

1.2. Chemical reactions in the troposphere . . . . . . . . . . . . . . . . . . . . . . . . . 71.2.1. Photochemical (Los Angeles type) smog formation . . . . . . . . . . . . . 71.2.2. Sulphuric acid formation and nucleation . . . . . . . . . . . . . . . . . . . 91.2.3. Heterogeneous reactions, condensed phase reactions and cloud process-

ing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 101.3. Thesis outline . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10

2. Methods 132.1. Environmental chambers . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13

2.1.1. CLOUD chamber . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 142.1.2. PSI smog chamber . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14

2.2. Sampling methods . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 142.2.1. CLOUD sampling line . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 152.2.2. Wet effluent diffusion denuder (WEDD) . . . . . . . . . . . . . . . . . . . . 172.2.3. Aerosol collector (AC) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17

2.3. Analytical methods . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 182.3.1. Absorption spectroscopy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 192.3.2. Chemiluminescence . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 192.3.3. Mass spectrometers (MS) . . . . . . . . . . . . . . . . . . . . . . . . . . . . 192.3.4. Chromatographic methods . . . . . . . . . . . . . . . . . . . . . . . . . . . 202.3.5. Particle number concentration and size distribution . . . . . . . . . . . . 21

3. DMA and NH3 measurements with IC during CLOUD4 233.1. Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 233.2. Experimental . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 24

3.2.1. CLOUD chamber . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 24

xi

Contents

3.2.2. Sampling of NH3 and DMA . . . . . . . . . . . . . . . . . . . . . . . . . . . 253.2.3. Ion chromatography (IC) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 26

3.3. Results and discussion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 263.3.1. Overall sampling efficiency and corrections . . . . . . . . . . . . . . . . . 263.3.2. IC method . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 283.3.3. NH3 and DMA mixing ratios . . . . . . . . . . . . . . . . . . . . . . . . . . . 29

3.4. Conclusions and outlook . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 30

4. Molecular understanding of amine-sulphuric acid particle nucleation 334.1. Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 344.2. Nucleation rates . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 364.3. Molecular composition of clusters . . . . . . . . . . . . . . . . . . . . . . . . . . . 384.4. Atmospheric implications . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 45

5. CMK as a model compound to elucidate oxidation mechanisms 495.1. Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 505.2. Experimental . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 50

5.2.1. Smog chamber and experiments . . . . . . . . . . . . . . . . . . . . . . . . 505.2.2. Instruments . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 51

5.3. Results and discussion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 535.3.1. Hydrogen abstraction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 535.3.2. Carbon-carbon bonds dissociation . . . . . . . . . . . . . . . . . . . . . . . 545.3.3. Organic acids formation mechanism . . . . . . . . . . . . . . . . . . . . . 575.3.4. Non-traditional chemistry . . . . . . . . . . . . . . . . . . . . . . . . . . . . 62

5.4. Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 62

6. Online measurement of organic acids from the photo-oxidation of TMB 656.1. Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 666.2. Experimental . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 66

6.2.1. PSI smog chamber . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 666.2.2. Ion chromatography system . . . . . . . . . . . . . . . . . . . . . . . . . . . 676.2.3. High resolution MS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 696.2.4. Partitioning theory . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 696.2.5. Chemicals . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 70

6.3. Results and discussion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 706.3.1. SOA formation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 706.3.2. Identified organic acids . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 716.3.3. Partitioning coefficients, Kp . . . . . . . . . . . . . . . . . . . . . . . . . . . 74

6.4. Summary and conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 80

7. Conclusions and outlook 817.1. Trace gases and new particle formation . . . . . . . . . . . . . . . . . . . . . . . . 827.2. Chemical mechanisms in the troposphere . . . . . . . . . . . . . . . . . . . . . . . 83

7.2.1. Organic acids formation mechanisms . . . . . . . . . . . . . . . . . . . . . 837.2.2. Effect of SO2 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 837.2.3. Ring-closure mechanisms . . . . . . . . . . . . . . . . . . . . . . . . . . . . 83

7.3. Outlook . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 84

xii

Contents

References 85

A. Supplementary material 99A.1. CLOUD experiment at CERN . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 99A.2. Determination of the nucleation and growth rates . . . . . . . . . . . . . . . . . . 100A.3. Cluster evaporation and fragmentation in the atmospheric pressure interface

time-of-flight (APi-TOF) . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 101A.4. Experimental errors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 102A.5. ACDC model . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 103A.6. Supplementary tables . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 105

xiii

List of Tables

1.1. Volume fraction of the main gases in dry tropospheric atmosphere at a pressureof 1 atm. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3

1.2. Global aerosol particles emissions and production estimates for the year 2000. 6

2.1. Selected environmental chambers for the study of secondary organic aerosols. . 16

5.1. List of performed CMK oxidation experiments. . . . . . . . . . . . . . . . . . . . . 525.2. Reaction rate constants for hydrogen atom abstraction from CMK by a hydroxyl

radical (OH)). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 52

6.1. Description of the performed TMB photo-oxidation experiments. . . . . . . . . . 676.2. Organic acids detected with WEDD/AC-IC/MS during TMB photo-oxidation. . . 786.3. Expected organic acids and related species from the MCM model for the oxida-

tion of TMB. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 79

A.1. Electronic energies, enthalpies, Gibbs free energies and entropies of formationfrom monomers for all clusters. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 105

A.2. Summary of neutral clusters explicitly simulated in the ACDC model. . . . . . . . 107A.3. Summary of the negatively charged clusters explicitly simulated in the ACDC

model. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 107A.4. Summary of the positively charged clusters explicitely simulated in the ACDC

model. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 108A.5. Dipole moments and polarizabilities of all studied clusters at 298 K. . . . . . . . 108

xv

List of Figures

1.1. Chemical structures of α-pinene and TMB. . . . . . . . . . . . . . . . . . . . . . . 41.2. Chemical structures of formic acid, acetic acid, pinonic acid, MBTCA, terpenylic

acid, and diaterpenylic acid acetate. . . . . . . . . . . . . . . . . . . . . . . . . . . . 51.3. Tropospheric chemistry reactions scheme. . . . . . . . . . . . . . . . . . . . . . . . 71.4. Formation and decomposition of Criegee intermediate. . . . . . . . . . . . . . . . 81.5. Gas phase degradation scheme of VOCs by reaction with OH, NO

3 or O3 andpartitioning of low volatility products onto the particle phase followed by con-densed phase reactions. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9

2.1. Scheme of the PSI smog chamber. . . . . . . . . . . . . . . . . . . . . . . . . . . . . 152.2. Schematic of the WEDD used for this work. . . . . . . . . . . . . . . . . . . . . . . 172.3. Schematic of the AC used for this work. . . . . . . . . . . . . . . . . . . . . . . . . 18

3.1. Unscaled schematic of the sampling system used for CLOUD4. . . . . . . . . . . 253.2. Schematic of the concentrating system with two columns and a 10-port valve. . 263.3. Calibration plots for NH3 and DMA. . . . . . . . . . . . . . . . . . . . . . . . . . . . 283.4. Liquid to air flow ratio from the sampling line. Normalised signal varying the liq-

uid to air flow ratio while sampling a constant DMA concentration in the chamber. 293.5. Overview of NH3 mixing ratios. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 303.6. MFC settings and summary of DMA mixing ratio measured. . . . . . . . . . . . . 31

4.1. Nucleation rate of new particles at 1.7 nm mobility diameter as a function ofH2SO4 concentration. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 39

4.2. Contribution of ions and dimethylamine to ternary nucleation. . . . . . . . . . . 404.3. Molecular composition of charged clusters measured by the APi-TOF. . . . . . . 434.4. Neutral sulphuric acid dimer vs. monomer concentrations measured by the

CIMS in the CLOUD chamber. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 44

5.1. Chemical structures of CMK and measured of expected products. . . . . . . . . . 515.2. Chemical mechanism following hydrogen atom abstraction from CMK by a OH. 535.3. Signals of m/z 131 and 113 from the PTR-MS. . . . . . . . . . . . . . . . . . . . . . . 545.4. Ring opening chemical mechanism of the alkoxy radical A from Fig. 5.2. . . . . . 555.5. Signals of the PTR-MS for m/z 147, 129 and 111. . . . . . . . . . . . . . . . . . . . . . 565.6. Signals of the PTR-MS for m/z 87 and 73. . . . . . . . . . . . . . . . . . . . . . . . . 575.7. Mixing ratios and molar yields of acids observed. . . . . . . . . . . . . . . . . . . . 585.8. Chemical mechanism of the reaction between the peroxyacyl radical with the

HO2 forming either a peracid or an organic acid and O3. . . . . . . . . . . . . . . . 59

5.9. Suggested formation mechanism of succinic acid adapted from the MBTCA for-mation mechanism of Muller et al. (2012). . . . . . . . . . . . . . . . . . . . . . . . 60

xvii

List of Figures

5.10. Chromatograms of m/z 143 for the four different experiments. . . . . . . . . . . 61

6.1. No breakthrough of the gas phase to the AC was observed when a filter wasplaced in the aerosol sampling line. . . . . . . . . . . . . . . . . . . . . . . . . . . . 68

6.2. Example of IC/MS calibration (methylmaleic acid). . . . . . . . . . . . . . . . . . . 696.3. Mobility mean diameter, total number concentration and aerosol mass concen-

tration from the SMPS measurements. . . . . . . . . . . . . . . . . . . . . . . . . . 726.4. Gas phase concentration profiles of eight measured organic acids. . . . . . . . . 746.5. Aerosol concentration profiles of eight measured organic acids. . . . . . . . . . . 756.6. Stacked normalized aerosol concentrations of the main detected organic acids

for the different experimental conditions. . . . . . . . . . . . . . . . . . . . . . . . 766.7. Time-dependent partitioning coefficients Kp . . . . . . . . . . . . . . . . . . . . . . 776.8. No partitioning of acetic acid onto the aerosol is observed after injection of a

large amount of acetic acid. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 79

1a. Schematic diagram of the CLOUD experiment at the CERN Proton Synchrotron. 100

xviii

List of Abbreviations

AC aerosol collectorACDC Atmospheric Cluster

Dynamics CodeAPi-TOF atmospheric pressure

interface time-of-flightCCN cloud condensation nucleiCD conductivity detectorCH4 methaneCIMS chemical ionisation mass

spectrometerCMK cyclobutyl methyl ketone,

CH3C(O)C4H7

CO carbon monoxideCO2 carbon dioxideCPC condensation particle

counterDMA dimethylamine, NH(CH3)2DMAH+ dimethylaminium,,

NH2(CH3)+2DMS dimethylsulfide,, S(CH3)2DU degree of unsaturationESI electrospray ionisationFEP fluoroethylene propylene,

(CF2CF2)n(CF2CF(CF3))mGC gas chromatographyGCR galactic cosmic raysGR growth rateHNO3 nitric acidHO

2 hydroperoxyl radicalH2O waterH3O+ hydronium ion

HONO nitrous acidHSO−4 bisulphateH2SO4 sulphuric acidIC ion chromatographyIPCC Intergovernmental Panel

on Climate ChangeIR infraredIUPAC International Union of Pure

and Applied ChemistryLC liquid chromatographyMBTCA 3-methyl-1,2,3-

butanetricarboxylic acid,C8H12O6

MCM Master ChemicalMechanism

MDL method detection limitMFC mass flow controllerMS mass spectrometerMSA methanesulfonic acid,

CH3SO3HN2 molecular nitrogenNAIS neutral cluster and air ion

spectrometerNH3 ammoniaNH+

4 ammonium ionNMHC non-methane

hydrocarbonsN2O nitrous oxideNO nitrogen monoxideNO2 nitrogen dioxideNO

3 nitrate radicalNO−3 nitrate

xix

List of Abbreviations

NOx nitrogen oxidesNOy total reactive nitrogenO2 molecular oxygenO−2 superoxideO3 ozoneOH− hydroxy anionOH hydroxyl radicalOOA oxygenated organic aerosolPAH polycyclic aromatic

hydrocarbonPAN peroxyacyl nitrate,

RC(O)OONO2ppbv part per billion by volume,

10−9

ppm part per million, 10−6

ppmv part per million by volume,10−6

pptv part per trillion by volume,10−12

PS Proton SynchrotronPSI Paul Scherrer InstitutPSM particle size magnifierPTFE polytetrafluoroethylene,

(CF2CF2)n

PTR-MS proton-transfer-reactionmass spectrometer

PTR-TOF proton-transfer-reactiontime-of-flight

R alkyl radicalRH relative humidityRO alkoxy radicalRO

2 peroxyl radicalRT retention timeSMPS scanning mobility particle

sizerSO2 sulphur dioxideSO3 sulphur trioxideSOA secondary organic aerosolTMB 1,3,5-trimethylbenzene,

C9H12

TME tetramethylethene,(CH3)2C=C(CH3)2

UV ultra-violetVOC volatile organic compoundWEDD wet effluent diffusion

denuderWSOC water soluble organic

compounds

xx

1Introduction

The Earth’s atmosphere is an extremely complex system; it developed on a geologicaltime scale in parallel with the Earth’s formation from a cloud of gas and dust (started around4.6 billions years ago). Oceans and Earth’s crust formation were also strongly correlated withchanges in the early atmosphere. Around 2.3 billions years ago, a dramatic increase of oxygenmodified the mildly reducing chemical mixture composing the atmosphere into a strongly ox-idising one (Seinfeld and Pandis, 2006). The reasons for this are still subject to debate (Kasting,2001).

Traditionally, the atmosphere is divided into several regions according to temperature vari-ations (Wallace and Hobbs, 2006). The lower atmosphere comprises the troposphere (0 to10–15 km altitude, depending on latitude and time of year) and the stratosphere (up to ∼45–55 km). The boundary between those two layers is called tropopause and the one betweenthe stratosphere and the upper atmosphere is named stratopause. The temperature decreasesin the troposphere almost linearly with height (9.7 K km−1 for dry air) up to the tropopause.There, the temperature stabilises around 217 K before increasing again due to the absorptionof solar ultra-violet (UV) radiation by ozone (O3). Finally, because the troposphere is subject tocontinuous turbulence and mixing, its composition is rather homogeneous within each Earthhemisphere.

This work focuses on the chemical analysis of trace gases and particles present in thetroposphere. The first chapter presents some basic tropospheric chemistry concepts, as wellas the structure of the work presented.

1

Chapter 1. Introduction

1.1. Chemical composition of the troposphere

Even though the troposphere represents only a small portion of the atmosphere’s totalheight, it contains about 80 % of its mass (Seinfeld and Pandis, 2006). The main constituentof tropospheric air is the chemically inert molecular nitrogen (N2) which accounts for∼78 % ofthe air by volume. The second most important constituent, molecular oxygen (O2), accountsfor ∼21 % and is, in contrast, reactive towards other compounds present in the atmosphere.Argon (Ar) is the most abundant noble gas (roughly 1 %). Table 1.1 gives an overview of thechemical composition of dry tropospheric air, including trace gases. The water (H2O) amountpresent in the atmosphere is roughly 0.25 % but varies over several orders of magnitude from10 part per million by volume (ppmv) in the coldest regions of the atmosphere up to 5 % byvolume in hot, humid air masses (Wallace and Hobbs, 2006).

There is a virtually infinite amount of other compounds containing carbon (Goldstein andGalbally, 2007), nitrogen and sulphur that are present at trace levels (ppmv and orders of mag-nitude lower) in the troposphere. Some of those species such as the greenhouse gases (e. g.carbon dioxide (CO2), O3, methane (CH4), nitrous oxide (N2O)) can have a disproportionateimpact on the atmosphere’s radiative balance, in regard to their low concentrations. Othercompounds (e g. aromatics, peroxyacyl nitrates (PANs)) can have a negative impact on health,due to their irritating or carcinogenic properties. Due to an increasing influence of anthro-pogenic activities on a global scale to the tropospheric chemistry, the concentrations of thetrace constituents of the atmosphere changed rapidly during the last two centuries.

1.1.1. Volatile organic compounds (VOCs)

Chemical compounds with a high volatility at room temperature are named volatile or-ganic compounds (VOCs) and represent a large class of compounds emitted to the atmosphere.Biogenic VOCs (mostly from vegetation) represent the major part of the total emissions. Guen-ther et al. (1995) estimated that isoprene (C5H8) represents 44 % and monoterpenes (C10H16,e. g.α-pinene) 11 % of an annual global biogenic VOC total flux of 1150 Tg C yr−1. For the remain-ing biogenic VOCs (45 %), large uncertainties exist in the emissions estimates. AnthropogenicVOCs emissions were estimated to be on the order of 142 Tg yr−1 in the early 1990s (Middleton,1995). The largest part of the emissions arise from fuel consumption (34 %), 25 % is due to roadtransport, 14 % to solvent use, and 13 % to fuel production/distribution. The rest (14 %) is due touncontrolled burning and other minor sources. Aromatic compounds (e. g. polycyclic aromatichydrocarbons (PAHs), benzene, toluene, xylenes) contribute to a large extent to the anthro-pogenic emissions. Figure 1.1 shows the chemical structures of a biogenic VOC (α-pinene) andof an anthropogenic VOC (1,3,5-trimethylbenzene (TMB)). Human activities also have an im-pact on biogenic VOC emissions (Guenther et al., 2000).

2


Table 1.1.: Volume fraction of the main gases in dry tropospheric atmosphere at a pressure of1 atm, adapted from Wallace and Hobbs (2006).

Gas Chemical formula Volume fraction of air

Nitrogen N2 78.084%Oxygen O2 20.946%Argon Ar 0.934%Neon Ne 18.18 ppmvHelium He 5.24 ppmvHydrogen H2 0.56 ppmvOzone O3 10–100 ppbvHydrogen peroxide H2O2 0.1–10 ppbv

Carbon-containing compoundsCarbon dioxide CO2 379 ppmv a

Carbon monoxide CO 40–200 ppbvMethane CH4 1.7 ppmvNon-methane hydrocarbons(NMHC)

5–20 ppbv

Halocarbons 3.8 ppbvFormaldehyde HCHO 0.1–1 ppbvVolatile organic compounds(VOCs): e. g. carbonyls, or-ganic sulphur compounds,and alcohols

ppbv range or lower

Nitrogen-containing compoundsNitrous oxide N2O 0.31 ppmvNitrogen species NOy = NO + NO2

(= NOx ) + NO3 + N2O5 +HNO3 + PAN

10 pptv–1 ppmv

Ammonia NH3 10 pptv–1 ppbv

Sulphur-containing compoundsSulphur dioxide SO2 10 pptv–1 ppbvDimethylsulphide (DMS) S(CH3)2 10–100 pptvHydrogen sulphide H2S 5–500 pptvCarbon disulphide CS2 1–300 pptv

Free radicalsHydroxyl radical OH· <0.4 pptvHydroperoxyl radical HO·

2 <5 pptv

a. upward trend

3


Figure 1.1.: Chemical structures of (a) α-pinene, a biogenic volatile organic compound (VOC),and (b) 1,3,5-trimethylbenzene (TMB), an anthropogenic VOC.

1.1.2. Ammonia (NH3)

The most abundant basic gas in the troposphere is ammonia (NH3). Bouwman et al. (1997)estimated its emission as 54 Tg N yr−1 for 1990. Sources are principally related to agriculturalactivities related to food production (e. g. fertilizers, domestic animals, crops), which accountsfor 70 % of the annual emissions. Oceans, undisturbed soils, wild animals and biomass burningcontribute to a lesser extent to the NH3 emissions. As a basic compound, NH3 can neutralisesulphuric acid (H2SO4) and nitric acid (HNO3) to form ammonium salts (Sect. 1.2.2).

1.1.3. Amines

There is a very large variety of amines in the atmosphere. A recent review by Ge et al.(2011) identified approximately 150 amines and about 30 amino acids. They possess a similarchemical structure as NH3, where one, two or all the hydrogen atoms were replaced by alkyl oraryl moieties. Therefore, they have also basic properties and can neutralise acidic compoundssimilarly to NH3. Furthermore, Murphy et al. (2007) reported that amines react with O3 andnitrogen oxides (NOx) and, by reaction with organic acids, form amides (Barsanti and Pankow,2006).

1.1.4. Organic acids

Organic acids (or carboxylic acids) are ubiquitous in the atmosphere (Chebbi and Carlier,1996). They can be emitted from soil micro-organisms and vegetation but also from biomasscombustion and motor exhaust. Organic acids can be formed through photochemical re-actions of precursors in the atmosphere (e. g. hydrocarbons degradation, ozone oxidation ofolefins, reaction of peroxyacyl radicals, aqueous reactions). While organic acids with a lowermolar mass (e. g. formic acid, acetic acid) can be formed from the reaction pathways of severalprecursors, pinonic acid, 3-methyl-1,2,3-butanetricarboxylic acid (MBTCA), terpenylic acid, anddiaterpenylic acid acetate (Szmigielski et al., 2007; Iinuma et al., 2008; Claeys et al., 2009) resultall from the oxidation of α-pinene. For example, the chemical structures of these compoundsare depicted in Fig. 1.2.

As many oxidised compounds, organic acids can partition onto particles present in theatmosphere, either because their vapour pressure is low or if the particle phase is primarilyaqueous, due to their solubility. Organic acids dissociate in solution into ions, whereby theeffective solubility depends on dissociation constants, Ka (Pun et al., 2002). Partitioning theory

4


Figure 1.2.: Chemical structures of (a) formic acid, (b) acetic acid, (c) pinonic acid, (d) 3-methyl-1,2,3-butanetricarboxylic acid (MBTCA), (e) terpenylic acid, and (f) diaterpenylic acid acetate.

will be discussed in Sect. 6.2.4. It remains however unclear if organic acids can be responsiblefor the formation of new particles or are only responsible for their growth.

1.1.5. Tropospheric aerosol

As already mentioned, not only gaseous species but also small liquid or solid particles insuspension in the air (called aerosols) with diameters from a few nanometres (nm) to hun-dreds of micrometres (µm) can be found in the troposphere. Due to their important impact onhealth, visibility, and climate, aerosol particles have been subject to many scientific investiga-tions (Kondratyev et al., 2006).

Aerosols can be emitted directly to the troposphere either by biogenic or anthropogenicsources (termed primary particles) or can be produced in situ through chemical reactions ofgaseous precursors towards low volatility products (named secondary particles). Table 1.2 pre-sents estimates of the global emission of particles emitted and produced in the atmospherefor the year 2000. While the effects of greenhouse gases on climate is fairly well understood,the role played by aerosols is yet not fully characterised due to the variety of particles existingin the atmosphere with different physical and chemical properties affecting their absorptionand scattering properties (Forster et al., 2007). Without considering primary emissions of seasalt and mineral dust particles, which are usually large and have a shorter residence time inthe atmosphere (Esmen and Corn, 1971), the amount of secondary formed particles (∼120–440 Tg yr−1) is as large or even larger than the primary emissions (∼100–350 Tg yr−1). For thisreason and in order to be able to better predict atmosphere behaviour, research has been con-ducted to better understand the chemistry leading to gas-to-particle conversion. This mayhappen either by condensation of low volatility products formed through oxidation processesonto pre-existing particles or by the formation of new particles, where the latter process isnucleation.

5


Table 1.2.: Global aerosol particles emissions and production estimates for the year 2000,adapted from Penner et al. (2001).

Aerosol type Emissions/Production estimates

Primary particles emissionsCarbonaceous aerosols

Organic matter (0–2 µm)Biomass burning 45–80 Tg yr−1

Fossil fuel 10–30 Tg yr−1

Biogenic (>1 µm) 0–90 Tg yr−1

Black carbon (0–2 µm)Biomass burning 5–9 Tg yr−1

Fossil fuel 6–8 Tg yr−1

Aircraft 0.006 Tg yr−1

Industrial dust, etc. (>1 µm) 40–130 Tg yr−1

Sea salt (mostly >1 µm) 1000–6000 Tg yr−1

Mineral (Soil) Dust (mostly >2 µm) 1000–3000 Tg yr−1

Secondary (in situ) aerosol sourcesSulphate (as NH4HSO4) 100–374 Tg yr−1

Anthropogenic 69–214 Tg yr−1

Biogenic 28–118 Tg yr−1

Volcanic 9–48 Tg yr−1

Nitrate (as NO−3 )

Anthropogenic 9.6–19.2 Tg yr−1

Natural 1.9–7.6 Tg yr−1

Organic compoundsAnthropogenic 0.3–1.8 Tg yr−1

VOCBiogenic 8–40 Tg yr−1

6

1.2. Chemical reactions in the troposphere


Dry and wet deposition are important sinks of chemicals and particles in the troposphere.Additionally, chemical reactions can be in situ sources or sinks for tropospheric components.Some of which and their relevance on particle formation or growth are described hereafter.

1.2.1. Photochemical (Los Angeles type) smog formation

The reaction mechanism between hydrocarbons and NOx under the influence of solar irra-diation producing irritating and harmful compounds present in the smog of Los Angeles wereidentified in the early 1950s (Haagen-Smit, 1952; Haagen-Smit et al., 1953). The term smogderives from smoke and fog and designs the severe air pollution that restricts visibility, partic-ularly in urban areas (Hobbs, 2000).

Figure 1.3.: Tropospheric chemistry reactions scheme, adapted from Atkinson (2000). See textfor details.

The mechanism of smog formation is as follows: With solar irradiation at wavelength(λ) shorter than 430 nm, nitrogen dioxide (NO2) photolyses to an energetically excited oxygenatom (O∗) and nitrogen monoxide (NO). O∗ recombines very rapidly with O2 to form O3, which

7


usually reacts with NO to form NO2 and O2 again (Fig. 1.3). This would lead to a steady state ofO3 of ∼0.03 ppmv in urban-polluted air, but values well above this concentration are typicallyreached (up to 0.5 ppmv).

Solar UV radiation (λ < 320 nm) can decompose O3, producing O∗. Most of these ex-cited atoms dissipate their excess energy as heat and eventually recombine with O2 to formO3 again. A small fraction (typically ∼1 %) can react with H2O vapour to form a hydroxyl rad-ical (OH). OH is the most powerful oxidant present in the atmosphere and can react withalmost all trace gases (e. g. SO2, hydrocarbons, halocarbons, NH3, carbon monoxide (CO), hy-drogen halides, NO2). Despite an OH concentration in the troposphere of just a few tenths of apart per trillion by volume (pptv), OH is one of the most important species in the atmospheredue to its high reactivity. If OH reacts with a VOC, some hydrocarbon for instance, it will ab-stract an hydrogen atom to produce H2O and an alkyl radical (R). Such radicals react almostinstantaneously with O2 to form a peroxyl radical (RO

2), which will react with NO to form NO2

and an alkoxy radical (RO) (Fig. 1.3) and compete with O3. RO will react with O2 to form car-bonyl compounds and hydroperoxyl radical (HO

2), which will react with NO to regenerate OHand form NO2, competing as well against O3. Therefore, O3 will accumulate in polluted regionswhen strong solar UV radiation is present and measured concentrations often exceed the al-lowed limit values (Klumpp et al., 2006). RO

2 and RO can also undergo other reactions. RO2

may also react with NO2 to form a PAN or, if the NOx concentration is low, with HO2 to form

a hydroperoxide. RO can also decompose to form a carbonyl compound and a new alkyl radi-cal. If RO has more than four carbon atoms, it may also undergo intramolecular isomerisation,resulting in the formation of an hydroxyalkyl radical.

Depending on the kind of VOC that react with OH, the radicals formed will have differentreaction pathways (Atkinson, 1997). Oxygenated VOCs (e. g. carbonyl compounds, alcohols)have also different reactivities. Aldehydes form peroxyacyl radicals (RC(O)OO), which reactwith NO2 to form eye-irritating PANs. Unsaturated VOCs do not only react with OH but withO3. Their carbon-carbon double (or triple) bonds react in the atmosphere with O3 to forma Criegee intermediate (Herron and Huie, 1977; Atkinson, 2000). This species decompose in acarbonyl compound and a diradical (Fig. 1.4). The fate of this diradical can lead to the formationof organic acids and other oxygenated compounds (Orzechowska and Paulson, 2005).

Figure 1.4.: Formation and decomposition of Criegee intermediate, adapted from Atkinson(2000).

8


The products formed through these reactions are oxidised and therefore less volatile thantheir parent compounds, so that they can condense (partly or completely) on existing particlesor form new particles (Fig. 1.5). The newly formed particulate mass is called secondary organicaerosol (SOA).

Figure 1.5.: Gas phase degradation scheme of volatile organic compounds (VOCs) by reactionwith hydroxyl radical (OH), nitrate radical (NO

3) or ozone (O3) and partitioning of low volatilityproducts onto the particle phase (grey area) followed by condensed phase reactions (Hallquistet al., 2009).

1.2.2. Sulphuric acid formation and nucleation

Sulphate is one of the main constituent of secondary aerosol. H2SO4 is produced throughreaction of sulphur dioxide (SO2) with O2 or OH to form sulphur trioxide (SO3) (Eq. 1.1 to 1.3).SO3 reacts in fog droplets with H2O to form H2SO4 (Eq. 1.4).

2 SO2 + O2 → 2 SO3 (1.1)SO2 + OH + M → HOSO

2 + M (1.2)HOSO

2 + O2 → HO2 + SO3 (1.3)

SO3 + H2O + M → H2SO4 + M (1.4)

Due to its very low-volatility, H2SO4 condenses immediately to the particle phase and wasfound to be a key compound in the formation of new particles by nucleation with H2O (binary

9


nucleation). However, the classical binary nucleation theory does not seem to apply to lowerand warmer altitudes (Weber et al., 1999). To explain the nucleation rates observed, a ternarymechanism involving NH3 has been postulated (Korhonen et al., 1999; Kulmala et al., 2000).H2SO4 remains the limiting species for this nucleation, but the presence of 100 pptv of NH3

is sufficient to reduce by several order of magnitude the H2SO4 amount required for similarnucleation rates. However, Kirkby et al. (2011) showed that the ternary nucleation of H2SO4,NH3 and H2O was not sufficient to explain nucleation rates observed in the atmosphere. Othercompounds (e. g. organics or amines) must play an important role in the formation of newparticles.

1.2.3. Heterogeneous reactions, condensed phase reactions and cloud processing

On top of the gas phase tropospheric chemistry, surfaces of the particles are expectedto react with OH or other reagents. Within the condensed phase, oligomerisation reactionsand formation of organosulphate compounds is also expected (Barsanti and Pankow, 2005;Hallquist et al., 2009).

Many atmospheric compounds are soluble in H2O. Fog and cloud droplets scavenge thesecompounds which become available for aqueous chemistry. For example, H2SO4 can also beproduced through aqueous chemistry following the dissolution of SO2 in a droplet. WhenH2O evaporates, low volatility products formed this way become constituents of the secondaryaerosol.

1.3. Thesis outline

In order to better understand chemical reaction mechanisms and gas-to-particle conver-sion, laboratory experiments were performed under controlled conditions and with well de-fined mixture of gases (and sometimes particles). Studies were conducted at two differentenvironmental chambers that will be described in more detail in Chapter 2: The CLOUD cham-ber at CERN and the smog chamber at the Paul Scherrer Institut (PSI).

The CLOUD project investigates the possible link between galactic cosmic rays (GCR) andthe climate through the formation of new particles, their growth to cloud condensation nuclei(CCN) sizes and their influence on cloud properties. The ternary nucleation of H2SO4, H2Oand dimethylamine (DMA) was studied in the frame of the CLOUD4 campaign. The studiesperformed at the PSI smog chamber focused on secondary organic aerosol (SOA). Of particularinterest were the chemical characterisation of gas and particulate phase products, as well asgas phase reaction mechanisms.

Such a chemical analysis of air sampled from these environmental chambers can be donewith a set of various instrumentation, which is described in the Chapter 2. The present work fo-cuses on the use of ion chromatography (IC) for both cations and anions to measure DMA andNH3 in the CLOUD chamber (Chapter 3) and organic acids in the PSI smog chamber (Chapters 5and 6). These compounds play a role in the formation of new particles and the productionof secondary aerosols. Chapter 3 presents the IC method used to measure DMA and NH3 at

10

1.3. Thesis outline

very low pptv levels. DMA was injected intentionally into the CLOUD chamber during CLOUD4while NH3 was present only as a contaminant species. Chapter 4 present the atmosphericimplications of the measured nucleation rates for the ternary nucleation of H2SO4, H2O andDMA.

For the experiments conducted at the PSI smog chamber, IC was used for analysis of an-ions and coupled to a mass spectrometer (MS) in order to identify organic acids. Chapter 5presents a model system designed to understand the OH oxidation mechanism of pinonicacid, a primary ozonolysis product of α-pinene. Using a compound with higher volatility andhaving only the substructure of interest, cyclobutyl methyl ketone (CMK), smog chamber ex-periments could be performed at higher precursor concentrations so that the products con-centrations were higher than the detection limits. IC/MS and proton-transfer-reaction massspectrometer (PTR-MS) were used for this mechanistical study. By analogy, conclusions aboutthe pinonic acid system could be drawn. Chapter 6 presents the results on the oxidation ofTMB, a compound representative of anthropogenic VOC emissions. With IC/MS, organic acidswere quantified in both gas and particulate phase, so that partitioning coefficients could becalculated and compared to theoretical estimations. Also the effect of SO2 on the organic acidsformation was investigated. Finally, Chapter 7 summarises the overall findings of this work andpresents the future challenges of the chemical analysis that can be done in the laboratory withenvironmental chambers.

11

2Methods

During the second half of the 20th century, growing interest in air pollution triggered thedevelopment of new instrumentation in order to detect the trace components in the atmo-sphere at very low levels (pptv and lower). The same instrumentation that is used for fieldmeasurements can be used for laboratory experiments with flow tubes and environmentalchambers to gain knowledge of the atmospheric mechanisms and processes in the absenceof meteorological effects. Even if discrepancies still appear between field observations andlaboratory experiments, the latter ones allow investigation of key processes in order to fur-ther elaborate the knowledge about gaseous chemistry, new particle formation and chemicalcomposition of secondary organic aerosol (SOA) under well-defined conditions.

2.1. Environmental chambers

Environmental chambers became since the 1940s unavoidable tools allowing scientiststo study atmospheric processes with a handle on key parameters (e. g. temperature, relativehumidity, gas phase composition, number and composition of seed aerosol particles, light ex-posure, etc.). Since the end of the 20th century, studies on the formation of new particles andsecondary organic aerosol became relevant for laboratory chamber experiments.

Table 2.1 presents a selection of environmental chambers available worldwide (Hallquistet al., 2009). Very large chambers are available, but many laboratories perform experiments insmaller volume chambers. However, a high surface-to-volume ratio leads to higher wall lossesof particles, which limits the duration of the experiments performed. Indoor chambers uselamps to simulate sunlight (very reproducible), while outdoor chambers use the natural sun-light (dependant on latitude, time of the day, period of the year and cloud coverage). Cham-bers can be made out of different materials. Teflon® (fluoroethylene propylene (FEP), poly-tetrafluoroethylene (PTFE)) is a commonly used material which is transparent and allow light

13

Chapter 2. Methods

to penetrate into the chamber volume to trigger photochemistry. Some other chambers aremade of stainless steel.

The work presented in the following chapters was performed at two different environ-mental chambers. Although both chambers have a similar volume, their construction and op-eration are completely different. They are briefly described below.

2.1.1. CLOUD chamber

The CLOUD chamber at CERN (Meyrin, Switzerland) was designed to study the influenceof galactic cosmic rays (GCR) on the formation of new particles. Figure 1a presents a detailedschematic of the chamber and the instrumentation used for analysis. The 26.1 m3 chamberis made of electropolished stainless steel and is equipped with two fans for a rapid homoge-neous mixing of the compounds throughout the chamber. A high-voltage electric field can beapplied to remove all the present ions in the chamber in order to study neutral nucleation. Inthe absence of this electric field, the natural GCR from space reach the CLOUD chamber andproduce ion pairs in the chamber. On top of the neutral nucleation channel, charged nucle-ation occurs. An adjustable proton beam from the CERN proton synchrotron can be used tosimulate different intensity of the GCR (i. e. different altitudes). A fibre-optic ultra-violet (UV)system described by Kupc et al. (2011) is used to produce hydroxyl radical (OH) in the cham-ber from ozone (O3) and water (H2O). The temperature, the relative humidity as well as thetrace gases concentrations (e. g. of sulphur dioxide (SO2), O3, ammonia (NH3)) can be preciselycontrolled.

2.1.2. PSI smog chamber

The smog chamber at the Paul Scherrer Institute (PSI, Switzerland) is described in Paulsenet al. (2005). Figure 2.1 presents a schematic of the chamber. It consists of a 27 m3 (3 m× 3 m× 3 m) Teflon® bag suspended in a temperature-regulated housing. Four Xe-arc lamps (4 ×4 kW) simulate sun light. The lamps are not directed towards the bag, but to the walls of thehousing which are covered with aluminium plates. The light is reflected and is homogeneousthroughout the bag. Recently, black lights were placed below the chamber to allow for strongerUV irradiation of the chamber.

2.2. Sampling methods

Sampling of the gas and/or particles from a chamber is similar to that in the field. Sam-ples can be analysed offline or online. Sampling for offline analysis is usually prone to bothnegative and positive artefacts and the achieved time resolution is generally lower than withonline methods. The main advantage of offline sampling is the ability to store the samples andanalyse only part of each sample in order to be able to reanalyse samples of interest. Onlinemethods provide usually data with a high time resolution, but the data can only be acquiredonce without possible reanalysis.

14


Figure 2.1.: Scheme of the Paul Scherrer Institut (PSI) smog chamber (Paulsen et al., 2005).

2.2.1. CLOUD sampling line

The sampling line developed for gaseous NH3 analysis at the CLOUD chamber (Bianchiet al., 2012) is depicted in Fig. 3.1. Ultra pure water (18.2 MΩ cm) is introduced into the samplingline. A small orifice in the sampling line is used to pump air from the CLOUD chamber andmix it with the water passing through a coil with ten loops with a turbulent flow. A detaileddiscussion about the efficiency of the sampling for NH3 and dimethylamine (DMA) is part ofChapter 3.

This device was used for the analysis of the gas phase; however, it is unable to separatethe gas phase from the aerosol particles. Corrections may be needed depending on the totalaerosol mass loading and the aerosol chemical composition.

15

Chapter 2. Methods

Table 2.1.: Selected environmental chambers for the study of secondary organic aerosols,adapted from Hallquist et al. (2009).

Location Type Volume Material a Temperaturerange

California Institute of Tech-nology, USA

Indoorphotoreactor

28 m3 (dual) FEP 290–303 K

Carnegie Mellon Univer-sity, USA

Indoorphotoreactor

10 m3 PTFE/FEP 288–313 K

Forschungszentrum Julich,Germany (SAPHIR)

Outdoorphotoreactor

270 m3 FEP ambient

Forschungszentrum Julich,Germany

Dark chamber 250 m3 PTFE/FEP ambient

Karlsruhe Institute of Tech-nology, Germany (AIDA)

Dark chamber 4–84 m3 Metal 183–323 K

Fundacion Centro deEstudios Ambientalesdel Mediterraneo, Spain(EUROPHORE)

Outdoorphotoreactor

200 m3 FEP ambient

Leibniz Institute for Tropo-spheric Research, Germany

Indoorphotoreactor

19 m3 FEP 289–308 K

Paul Scherrer Institut,Switzerland

Indoorphotoreactor

27 m3 FEP 288–313 K

University College Cork, Ire-land

Indoorphotoreactor

6.5 m3 FEP 293–305 K

University of Manchester,UK

Indoorphotoreactor

18 m3 FEP 288–313 K

University of CaliforniaRiverside, USA (CE-CERT)

Indoorphotoreactor

90 m3 (dual) FEP 278–323 K

University of CaliforniaRiverside, USA (APRC)

Indoorphotoreactor

6–8 m3

(several)PTFE/FEP ambient

University of North Car-olina, USA

Outdoorphotoreactor

120, 137 (dual)and 150 m3

(dual)

FEP ambient

US Environmental Protec-tion Agency

Indoorphotoreactor

14.5 m3 FEP/TFE 293–298 K

CSIRO Energy Technology,Australia

Indoorphotoreactor

18 m3 FEP ambient

National Institute for Envi-ronmental Studies, Japan

Indoorphotoreactor

6 m3 PTFE/FEP ambient

a. (P)TFE: (poly)tetrafluoroethylene, FEP: fluorinated ethylene propylene

16


2.2.2. Wet effluent diffusion denuder (WEDD)

Figure 2.2.: Schematic of the wet effluent diffusion denuder (WEDD) used for this work(Takeuchi et al., 2004).

Since the late 1980s, wetted denuders have been used to sample gases of interest into so-lution for further analysis and quantification. The earlier ones had either an annular or a par-alell plate design and had to be operated in strict vertical or horizontal orientation according tothe design, for which they were made. The addition of gas-permeable membranes improvedthe stability of the liquid flows, so that even when it is slightly tilted, the denuder can still beoperated without losses.

The planar design of the membrane-based WEDD that was used for this work is describedin details by Takeuchi et al. (2004). The air flow (up to 1.7 l min−1) passes between two wettedcellulose acetate membranes (Fig. 2.2). Gases diffuse through the membranes and dissolveinto the water that wets the other side of the membrane. The water is then pumped by aperistaltic pump to the instrument for analysis.

2.2.3. Aerosol collector (AC)

Continuous sampling of aerosol particles for online analysis is a challenging task. The de-vice presented here and described in details in Takeuchi et al. (2005) consists of a mist chamberin which air sample containing particles are pumped. The mist is produced by the action of thesampling flow (up to 1.5 l min−1) on the water coming from a capillary placed at the end of thesampling line (Fig. 2.3). The particles take up water and are collected in the aqueous film cov-ering the hydrophilic filter. This water passes through the filter by the action of the samplingpump and reaches the bottom of the device. From there, it is continuously pumped by a peri-

17

Chapter 2. Methods

Figure 2.3.: Schematic of the aerosol collector (AC) used for this work (Takeuchi et al., 2005).

staltic pump to the instrument for analysis. By combining this AC with the denuder presentedin Sect. 2.2.2, it is possible to sample gas and particle phase separately, avoiding artefacts.

2.3. Analytical methods

A large variety of methods can be used to analyse the chemical composition of gas phaseor aerosol particles which rely on different properties of the analyte of interest. The method ofchoice usually depends on the measurement conditions, the precision required and the budgetdevoted to the analysis. Analytical methods used in atmospheric science cover a broad rangeof optical (e. g. absorption spectroscopy, chemiluminescence), chemical (requiring chemicalconversion prior to detection) and mass spectrometric techniques often combined with sepa-ration methods (e. g. chromatography).

18


2.3.1. Absorption spectroscopy

Some species like O3 or nitrogen oxides (NOx) absorb specific light wavelength (e. g. inthe UV or infrared (IR) spectrum region). Using the Beer-Lambert law (Eq. 2.1), it is possible tolink the decrease of the source light intensity I with the concentration c of the compound ofinterest i .

I (λ) = I0(λ)exp(εi (λ)Lc) (2.1)

I0(λ) corresponds to the initial light intensity, I (λ) to the light intensity after being ab-sorbed on a path of length L. Each substance has a different extinction coefficient εi (λ).

2.3.2. Chemiluminescence

Light production by a chemical reaction is called chemiluminescence (Weinheimer, 2007).This is the case for the reaction of nitrogen monoxide (NO) with O3, which produces nitrogendioxide (NO2) (Eq. 2.2) with a fraction of excited molecules (NO∗2, Eq. 2.3). These excited mol-ecules can either be collisionally quenched with inert molecules, which will absorb the excessmolecular energy (Eq. 2.4), or fluoresce (0.1 % fraction at most, Eq. 2.5). The recorded signalfrom photon (hν) emission can be correlated to the mixing ratio of NO after calibration of theinstrument.

NO + O3 → NO2 + O2 (2.2)NO + O3 → NO∗2 + O2 (2.3)NO∗2 + M → NO2 + M (2.4)

NO∗2 → NO2 + hν (2.5)

NO2 can also be measured this way after conversion to NO on heated catalysts. However,the conversion is not specific enough and can also transform NH3, peroxyacyl nitrates (PANs),nitric acid (HNO3), nitrous acid (HONO) into NO; therefore, the measured signal often corre-sponds to total reactive nitrogen (NOy), which can be used as an advantage. Photolytic con-version has been shown to be more specific, using UV irradiation. With this technique, it ispossible to measure a NO signal (without conversion of NO2) and a NOx signal (with conver-sion) and obtain the NO2 concentration from the difference between both signals.

2.3.3. Mass spectrometers (MS)

There is a large variety of mass spectrometers (MS) available for atmospheric scientists,which can be used to analyse both the gas phase and aerosol particles present in the atmo-sphere (Williams, 2007; Coe and Allan, 2007). They can be utilised online for in situ measure-ments or offline to analyse filter samples. The resolution and detection limits have been con-stantly improving over the last decades.

19

Chapter 2. Methods

Mass spectrometers use electric or magnetic fields to separate ions according to theirmass-to-charge ratios, m/z . Different types of ion sources (e. g. electron ionisation, chemicalionisation, electrospray ionisation) and mass filters exist (e. g. quadrupole, time-of-flight, mag-netic sector, orbitrap). This section presents two examples of mass spectrometers used withinthis work.

Proton-transfer-reaction mass spectrometer (PTR-MS)

The proton-transfer-reaction mass spectrometer (PTR-MS) uses a specific chemical ion-isation reaction with hydronium ion (H3O+) as reagent. All compounds that have a protonaffinity larger than H2O are protonised and can be classified by a quadrupole or a time-of-flight mass filter according to their m/z . Because of the large variety of compounds presentin the atmosphere and the low mass resolution of the quadrupole PTR-MS (unit mass reso-lution), the data obtained can be difficult to interpret. Coupling with a chemical separationtechnique (Sect. 2.3.4) can be an advantage. Time-of-flight PTR-MS has a high resolution andgive the chemical formula of the detected compounds. The compounds analysed with PTR-MShave the tendency to fragment in the ion source. For example, alcohols (but also aldehydesand peroxides) eliminate water after protonation. These fragmentation patterns depend onthe chemical structure of the analytes.

Orbitrap mass spectrometer

The orbitrap device constrains ions radially between a central spindle electrode (Makarov,2000; Hu et al., 2005) and a coaxial outer barrel-like electrode. m/z values are measurednon-destructively from the frequency of harmonic ion oscillations undergone by the orbitallytrapped ions along the axis of the electric field. This kind of device can achieve very high massresolution (up to 150 000).

2.3.4. Chromatographic methods

Because the atmosphere is a complex mixture of many different constituents, techniquesare required to separate them for individual analysis and quantification. All chromatographytechniques use the same principle: The sample is dissolved in a fluid (mobile phase, i. e. gas,liquid or supercritical fluid) used to transport the analytes over or through an immiscible bedof material (stationary phase). Due to the different interactions of the analytes and the sta-tionary phase, a physical separation of the sample constituents occurs (Hamilton and Lewis,2007). The method of choice to analyse volatile compounds is gas chromatography (GC), whileliquid chromatography (LC) is often used for the offline analysis of aerosol samples collectedon filters. Enhancing the capabilities of detection with an MS is routinely done and GC/MS orLC/MS are very often used for atmospheric analysis. An MS can separate coeluting compoundswith different masses and can give, for example, structural information on organic compoundswith high molar masses.

20


Ion chromatography (IC)

A specific example of LC is ion chromatography (IC), which utilises chemically modifiedpolymers as a stationary phase in order to analyse highly polar organic compounds and ionicspecies. Cation-exchange systems can be used for the analysis of metallic cations, ammoniaand amines, while anion-exchange systems are suitable for separation and detection of or-ganic acids and inorganic anions (e. g. sulphate, nitrate, nitrite, chloride).

Detection of the analytes generally occurs either with a conductivity detector (CD) or witha UV absorption detector. An MS can also be used as a second detector, without the need of astrong ionisation technique.

For the CLOUD experiment (Chapter 3), IC was used with only a CD for cations analysisand was directly connected to the sampling device described in Sect. 2.2.1. For the other exper-iments performed at the PSI smog chamber (Chapters 5 and 6), a combined WEDD/AC system(Sect. 2.2.2 and 2.2.3) was used together with IC (for anions analysis) to sample separately gasphase and aerosol particles and analyse them online alternately. A quadrupole MS with nega-tive electrospray ionisation was used as a further separation device.

2.3.5. Particle number concentration and size distribution

Besides the chemical characterisation of aerosol particles, information on particle numberconcentration and size distribution in order to quantify the total aerosol mass are essentialfor the analysis of atmospheric processes in environmental chambers. This section describesbriefly the instrumentation used for this work.

Condensation particle counter (CPC)

The working principle of condensation particle counters (CPCs) is the following (Cheng,2001): Aerosol particles are exposed to an environment saturated with H2O or alcohol vapourand cooled down to induce supersaturation. The exposed particles grow by taking up somevapour and achieve sizes of several µm. These particles scatter laser light and are detectedaccordingly.

Scanning mobility particle sizer (SMPS)

Particles size distribution can be measured with a scanning mobility particle sizer (SMPS).It consists of a CPC (Sect. 2.3.5) and a differential mobility analyser, which consists of a cylindri-cal capacitor in which charged particles are introduced (Cercl’Air, 2010). Because of the electricfield applied on the capacitor, the particles move towards the inner electrode and reach theelectrode at different positions due to their different (electrical) mobility diameters. Only thosewithin a narrow range of mobility diameters exits the differential mobility analyser throughthe output slit. (Particles with a larger diameter will be lost on the analyser walls.) Through

21

Chapter 2. Methods

variations of the electric field, particles of different sizes can be selected. By continuous scan-ning of the electric field, a particle size distribution can be derived (Wang and Flagan, 1990).

22

3Dimethylamine and ammonia measurements

with ion chromatographyduring the CLOUD4 campaign

Arnaud P. Praplan, Federico Bianchi, Josef Dommen, Urs BaltenspergerLaboratory of Atmospheric Chemistry, Paul Scherrer Institut, 5232 Villigen PSI, Switzerland

Published on 7 September 2012 in Atmospheric Measurement Techniques, 5, 2161–2167.

Abstract. The CLOUD project investigates the influence of galactic cosmic rays on the nucle-ation of new particles in an environmental chamber at CERN. Dimethylamine (DMA) was in-jected intentionally into the CLOUD chamber to reach atmospherically relevant levels awayfrom sources (up to 100 pptv) in order to study its effect on nucleation with sulphuric acid andwater at 278 K. Quantification of DMA and also background ammonia (NH3) was performedwith ion chromatography (IC). The IC method used together with the sampling line devel-oped for CLOUD in order to measure NH3 and DMA at low pptv levels is described; the overallsampling efficiency of the method is discussed; and, finally, mixing ratios of NH3 and DMAmeasured during CLOUD4 are reported.

3.1. Introduction

The CLOUD project investigates the influence of galactic cosmic rays (GCR) on the climatethrough their effect on cloud properties (Kirkby, 2007). A key process of these studies is thenucleation of new particles in the atmosphere from gaseous precursors and their growth todetectable sizes (diameter> 3 nm) (Kulmala et al., 2004; Hirsikko et al., 2011). These particles

23

Chapter 3. DMA and NH3 measurements with IC during CLOUD4

can eventually grow to cloud condensation nuclei (CCN) and influence indirectly the Earth’s cli-mate by modifying cloud properties. Recently published CLOUD data (Kirkby et al., 2011) on theternary sulphuric acid (H2SO4)/water (H2O)/ammonia (NH3) nucleation system (Coffman andHegg, 1995; Korhonen et al., 1999; Kulmala et al., 2000) demonstrated that NH3 (at a 100 pptv-level or less) increases nucleation rates by a factor higher than 100 to 1000 over the binary(H2SO4/H2O) system. However, NH3 was still not sufficient to explain the nucleation ratesobserved under typical tropospheric conditions (106–107 cm−3 H2SO4).

Because amines possess structural similarities with NH3 where one, two or all three hy-drogen atoms are replaced by organic moieties (RNH2, R2NH or R3N), their effect on secondaryorganic aerosol formation is increasingly subject of investigation. Murphy et al. (2007) showedthat amines can form secondary organic aerosol (SOA), by acting as bases and neutralisingacids present in the gas phase similar to NH3, but also by participating in gas phase chemistryinitiated by hydroxyl radical (OH) or ozone (O3) to form low volatility products (see also Tua-zon et al., 1994). Furthermore, a recent computational study suggests that amines are evenmore strongly bound to H2SO4 molecules than NH3 (Kurten et al., 2008) and can therefore en-hance even more neutral and ion-induced H2SO4/H2O nucleation in the atmosphere. Bzdeket al. (2010b) found experimental evidence for this effect: even at pptv-levels, complete dis-placement of NH3 by dimethylamine (DMA) occurs within seconds or minutes, changing thecomposition of sub-3 nm diameter bisulphate clusters. Not only substitution, but also addi-tion to those clusters can occur, influencing their growth rates (Bzdek et al., 2011). Makela et al.(2001) found that DMA was present at higher mixing ratios during nucleation events in borealforests compared to non-event periods, making it a potential nucleating species or a speciesincreasing growth rates of freshly formed particles, so that they can be detected faster at 3 nm.Yu et al. (2012) also reported enhancement effects on the nucleation rate by several amines.

CLOUD4 (June–July 2011) investigated the role of DMA in the formation and growth ofnew particles. Because DMA needed to be detected at trace concentration levels, an ion chro-matography (IC) method was deployed together with the sampling line developed for CLOUDby Bianchi et al. (2012). IC is often used to analyse amines and inorganic cations like ammoniumion (NH+

4 ) from atmospheric samples with various sampling methods (Makela et al., 2001;Murphy et al., 2007; VandenBoer et al., 2011). The overall sampling efficiency of the methodused for CLOUD4 for the simultaneous measurement of DMA and NH3 is discussed thereafterand the mixing ratios of both species are reported.

3.2. Experimental

3.2.1. CLOUD chamber

The CLOUD chamber is a 26.1 m3 electropolished stainless steel cylinder. A more detaileddescription can be found in Voigtlander et al. (2012) and Kirkby et al. (2011). The fibre-optic UVsystem for H2SO4 production with negligible thermal effect is described in Kupc et al. (2011).Although the chamber temperature can vary from 183 to 313 K, it was kept constant at 278 K(±0.01 K) with a relative humidity of 38 % (±0.1 %) during CLOUD4.

24

3.2. Experimental

Figure 3.1.: Unscaled schematic of the sampling system used for CLOUD4 (top panel) and themodified sampling system with two debubblers (bottom panel) for sampling efficiency tests.

3.2.2. Sampling of NH3 and DMA

The same sampling probe as the one described in Bianchi et al. (2012) was used (Fig. 3.1). Itwas specially designed to minimise the loss of NH3 and amines on surfaces. Briefly, it consistsof a 2-mm diameter stainless steel tubing of 140 cm length with a small orifice to the CLOUDchamber. Ultra pure water (18.2 MΩ cm) was introduced by a peristaltic pump in the stainlesssteel tubing (0.25–0.80 ml min−1). A cation trap column (CTC-2, Dionex) was additionally usedto remove the possibly interfering cations from this water prior to sampling. As the waterreached the orifice to the CLOUD chamber, it was mixed with air (0.8–2.1 l min−1) in a 10-loopcoil to allow for dissolution of the gaseous species into the water. With this setup only a sectionof 5 mm tubing remained unflushed by water, so that sampling losses became negligible.

A debubbler separates the air and the water, which is pumped from the bottom of thedebubbler by a peristaltic pump. This water was passed through a trace cation concentratorcolumn (TCC-2 or TCC-LP1, Dionex) where cations were retained. After 70 to 210 min of sam-pling, the 10-port valve described in Fig. 3.2 rotated to allow the elution of the cations to theanalytical column, while the sampling water was concentrated on a second cation concen-trator column to ensure continuous measurements. This automation was necessary as theCLOUD chamber cannot be accessed at all times (e.g. when it is irradiated by the pion beam)and it reduced the required maintenance to a minimum. No derivatisation was needed andthe samples could be directly eluted to the analytical column automatically at the end of thesampling period.

To test the sampling efficiency for NH3 and amines, a second scrubbing system was in-stalled in series at the end of the campaign as shown in the lower part of Fig. 3.1. It consisted

25


Figure 3.2.: Schematic of the concentrating system with two columns and a 10-port valve toensure continuous measurements. While one column concentrates the analytes coming fromthe chamber sampling line (red line), the other column can be eluted and the sample analysedby ion chromatography (blue line).

of a Teflon® tubing of 1 mm diameter and 140 cm length. The air flow used for this setup was1 l min−1 and both liquid flows were 0.5 ml min−1. The sampling was alternated from one de-bubbler to the other using the valve system described previously.

3.2.3. Ion chromatography (IC)

A Dionex DX600 system was used for the analysis of the collected cations with Ionpac®

CG10 and Ionpac® CS10 (Dionex) guard and analytical columns, respectively. The method usedwas similar to the one described in Chang et al. (2003): samples were eluted with 40 mM me-thanesulfonic acid (MSA) at 1 ml min−1 (isocratic). Retention times varied between 8.8 and10.5 min for NH3 and between 17.8 and 21.1 min for DMA. The peaks obtained from the con-ductivity detector were integrated manually. Calibration was performed by direct injection ofaqueous standards (no pre-concentration) of different concentration levels, corresponding toinjected amounts ranging from 0 to 30 ng for NH3 and 0 to 150 ng for DMA (Fig. 3.3).

The method detection limit (MDL) depends primarily on the noise of the chromatograms,but is influenced by the sample volume (based on sampling air flow and sampling time). De-pending on instrumental conditions and sampling time, the MDL ranged from 0.2 to 3.7 pptvfor NH3 and from 0.2 to 1.0 pptv for DMA (signal/noise = 3).

3.3. Results and discussion

3.3.1. Overall sampling efficiency and corrections

The use of pure water without addition of an acidic compound is assumed to be sufficientfor measurements of low mixing ratios of NH3 and DMA (< 100 pptv). From a theoretical pointof view, the overall sampling efficiency relies on

26


– the efficiency of the concentrator columns,– the stripping efficiency of the sampling device, and– the protonation degree of NH3 and DMA, because these species are detected as cations.

Because the calibration by direct injection does not take into account the efficiency ofthe concentrator column, it was tested separately. A solution of 0.1 µM NH3 and 0.1 µM DMAwas sampled on the concentrator columns. From the measured signals, a concentration of0.055 µM was calculated for both compounds, independent of the sampling time and the sam-pling flow. Therefore, a correction factor of 1.8 is applied to the mixing ratio of NH3 and DMA.Note that this effect is not due to the capacity of the concentrator column.

Indeed, the concentrator column capacity was found to be sufficient at those low levelsand sampling with ultra pure water at less than 1 ml min−1 for a long period of time. How-ever, due to this long sampling time, breakthrough of the analyte can happen. A sample wasacquired for roughly 10 h (13–14 July) and the DMA mixing ratio derived (0.86 pptv) was con-sistent with the previous (1.0 pptv, 3.51 h) and the following (0.59 pptv, 4.10 h) samples (withinuncertainties). For NH3, data from 14 and 15 July suggest that up to 4.5 h sampling time, theresults (11 pptv, corrected) remains consistent with data acquired for 3.5 hours (9.5 and 13 pptv,also corrected). Therefore, for the sampling time of 210 minutes (3.5 h), no breakthrough of theanalytes through the concentrator columns can be observed. A few NH3 mixing ratio valuesfor sampling times longer than 4.5 hours were discarded.

The stripping efficiency relies on the effective Henry’s law constant, H∗, which depends onthe degree of protonation of the species in solution and thus on pH (−log[H+]). Equation (1)defines H∗, taking into account the pH of the solution and the dissociation constant, Ka, of theanalytes. The negative logarithm of the dissociation constant (pKa) of NH+

4 is 9.90 at 278 K(Bates and Pinching, 1950) and for dimethylaminium (DMAH+)it is 10.64 at 293 K (Hall, 1957).

H∗ = H(1 + 10−pH+pKa) =[B] + [BH+]

pB(3.1)

where pB is the partial pressure of base (B) in the gas phase. From the review of Sander (1999),Henry’s law constants at 278 K are estimated to be in the range of 14.3–173 M atm−1 for NH3 and81.3–150 M atm−1 for DMA. At this temperature, Hawkes (1995) reports a negative logarithm ofthe self-ionisation constant of water (pKw =−log([H+][OH−])) of 14.7, corresponding to a pH of7.3 for pure water (H2SO4 can be neglected). This value can increase up to 7.5 by dissolving lowgas phase concentrations of DMA and NH3 (< 100 pptv each), assuming complete dissolution.Considering pure solutions of each species, the amount remaining in the gas phase (pB) isbetween 0.24 and 2.8 % for NH3 and between 0.08 and 0.04 % for DMA, assuming a liquid toair flow ratio of 0.3× 10−3.

This ratio influences the sampling efficiency as depicted in Fig. 3.4. The value of the ratiothroughout the experiment is presented on the left, while the normalised signals obtained byvarying the liquid to air flow ratio for a constant DMA mixing ratio in the chamber (around100 pptv) are depicted on the right. If the liquid to air flow ratio drops below 0.3× 10−3, thesampling efficiency decreases indicating that the stripping efficiency is governed by the res-idence time and not the gas-liquid equilibrium. DMA mixing ratios were corrected for thisobservation (see Sect. 3.3.3).

27


Figure 3.3.: Calibration plots for ammonia (NH3, left panel) and dimethylamine (DMA, rightpanel). Grey areas correspond to the range of experimental data.

Because NH3 was not injected at a constant mixing ratio into the chamber, the same testcould not be done during CLOUD4, but has been previously made by Bianchi et al. (2012).

Moreover, the degree of protonation of both species in solution is higher than 99 % (Eq. 2)at pH 7.5:

[B]

[BH+]= 10pH−pKa . (3.2)

Therefore, [B] can be neglected in Eq. (1) and because pB is equal to the difference between thetotal amount of B in the system (here 100 pptv) and [BH+], pB and [BH+] can be derived fromEq. (1).

Furthermore, at the end of the CLOUD4 campaign, a second sampling system was set inseries with the first one as shown in the lower part of Fig. 3.1. No DMA could be detected inthe water sampled from the second system, confirming the high overall sampling efficiency.Unfortunately, the second system was contaminated with NH3, so that no reliable conclusioncan be drawn for this species.

3.3.2. IC method

The use of the CS10 analytical column from Dionex allowed the separation of NH3 andDMA without the two peaks interfering with each other, even when both species had verydifferent concentration levels, which can often be problematic (VandenBoer et al., 2011). Thiswas possible because of the cleanliness of the CLOUD chamber and the absence of interferingpeaks. Only sodium and potassium peaks appeared in the chromatograms and were usuallywell resolved from the peaks of interest. However, during the period between 29 June and 7July, due to an instrumental contamination with sodium, its high and broad peak sometimesmasked the small NH+

4 peak and increased the conductivity background during this period.

28


Figure 3.4.: Liquid to air flow ratio from the sampling line used during the campaign (leftpanel). Normalised signal varying the liquid to air flow ratio while sampling a constantdimethylamine (DMA) concentration (around 100 pptv) in the chamber (right panel) with fitcurve (y = axb + c).

3.3.3. NH3 and DMA mixing ratios

Figures 3.5 and 3.6 summarise the measured mixing ratios for NH3 and DMA determinedwith IC. The instrument was not used at the very beginning of the campaign and no reliabledata could be acquired between 6 and 11 July, due to an elevated conductivity background,which increased the MDL.

Blank values were obtained by directly sampling water instead of flushing it through thesampling line. No peak for DMA could be observed from these measurements during thewhole campaign, so that no correction was applied to the obtained mixing ratio. On the otherhand, NH3 showed peaks in the blank samples during the second phase of the campaign (after11 July, when the sampling time was extended) and the measured mixing ratios were correctedproportionally to the water amount concentrated on the column (which depends on waterflow and sampling time) for this period.

Figure 3.5 shows the uncorrected mixing ratios of NH3, as well as the ones corrected forefficiency of the concentrator column. In addition, background levels from the sampling wa-ter were substracted for the measurement period after 11 July. The NH3 mixing ratio rangedmostly below 25 pptv, which corresponds to mixing ratios found at remote locations (Krupa,2003). This corresponds to the background level of NH3 in the CLOUD chamber at the presentconditions.

The top panel of Fig. 3.6 shows the mass flow controller settings for continuous DMAinjection into the CLOUD chamber. It was expected that the DMA levels were proportional tothese settings. Indeed, the mixing ratios shown on the bottom panel of Fig. 3.6 followed thistrend. However, after switching back from higher settings to lower ones, the measured mixingratio did not drop to previous levels but showed a background of a few pptv, slowly decreasing.Data were multiplied by a correction factor derived from the fit curve of Fig. 3.4 and by the

29


Figure 3.5.: Overview of ammonia (NH3) mixing ratios (in pptv): uncorrected (grey), correctedonly for concentrator column efficiency (filled light blue), and corrected for background levelsas well as concentrator columns efficiency (filled dark blue). The continuous line is a smoothingfunction through the final data.

factor 1.8 to take the concentrator column efficiency into account.

The values reported for both species were in some cases very close to the MDL, in partic-ular in the period from 21 to 24 July. With the sampling system used, no separation of the gasand aerosol phase occurred. For total aerosol mass loadings higher than 1 µg m−3, correctionsmay need to be considered, to take into account the ammonium and dimethylamine from theparticle phase.

DMA mixing ratios correspond to atmospherically relevant levels away from the sources(Ge et al., 2011). On the other hand, NH3 atmospheric levels range from a few hundreds of pptvto several ppbv (Li et al., 2006; Benson et al., 2010). Lower values could usually not be reportedbecause of detection limitations.

3.4. Conclusions and outlook

The analysis of trace gases at pptv-level (ca. 107 cm−3) is crucial for the understanding ofnucleation because such levels of certain contaminants (e.g. NH3, organics) can be sufficient toenhance nucleation rates by several orders of magnitude. The method presented here, basedon IC and making use of an efficient sampling line, could provide data down to sub-pptv levelswith 70 to 210 min time resolution.

Studies at low temperature also need to be performed. However, for CLOUD campaignsbelow 273 K, the sampling line needs to be adapted so that water does not freeze in the sam-pling line.

Gronberg et al. (1992) reported 0.5 and 1.8 pptv of DMA in urban and rural environmentsin Sweden. Chang et al. (2003) measured a broader range and slightly higher mixing ratios

30

3.4. Conclusions and outlook

Figure 3.6.: Mass flow controllers (MFC) settings (top panel) and summary of dimethylamine(DMA) mixing ratio (in pptv) measured (lower panel): uncorrected (grey), corrected for sam-pling efficiency (orange), and corrected for sampling efficiency and concentrator column ef-ficiency (filled red). The continuous line is a smoothing function through the final data. Alogarithmic scale was used to visualise the data of the second half of the campaign (around1 pptv).

of DMA (1.9–34 pptv) at a suburban site in Seoul, Korea. Closer to agricultural sources, Schadeand Crutzen (1995) found outdoor mixing ratios of DMA of 21 pptv in the afternoon and 76 pptvjust before sunrise. The levels of DMA injected into the CLOUD chamber represented well thatrange. Moreover, in all those studies, methylamine and trimethylamine were also present atsimilar levels (low pptv levels to a few hundreds of pptv). Akyuz (2007) found DMA levels ofapproximately 10 pptv, without a large variation between summer and winter, in six samplingsites in the Zonguldak province, Turkey, as well as strongly varying concentrations of variousother amines.

Usually, NH3 is also present at several orders of magnitude higher levels (ppbv) than theindividual amines. VandenBoer et al. (2011) reported amines to NH3 ratios in Toronto between1.6–20× 10−3, while this ratio ranged from 9.4× 10−3 to 23 during the CLOUD4 campaign. Theinfluence of such a level of DMA on nucleation and growth rates in the presence of a low NH3

mixing ratio will be the subject of other publications.

Acknowledgements This work was supported by the Swiss National Science Foundation (projectnos. 200020 135307 and 206620 130527) and by the EC Seventh Framework project CLOUD-ITN (MC-Initial Training Network No. 215072).

Edited by: V.-M. Kerminen

31

4Molecular understanding of amine-sulphuric

acid particle nucleation in the atmosphere

Joao Almeida1, Siegfried Schobesberger2, Andreas Kurten1, Ismael K. Ortega 2, Oona Kupiai-nen2, Arnaud P. Praplan3, Antonio Amorim4, Federico Bianchi3, Martin Breitenlechner5, AndreDavid6, Josef Dommen3, Neil M. Donahue7, Andrew Downard8, Eimear Dunne9, JonathanDuplissy2, Sebastian Ehrhart1, Richard C. Flagan8, Alessandro Franchin2, Roberto Guida6, JaniHakala2, Armin Hansel5, Martin Heinritzi5, Tuija Jokinen2, Heikki Junninen2, Maija Kajos2,Helmi Keskinen10, Agnieszka Kupc11, Theo Kurten12, Alexander N. Kvashin13, Ari Laaksonen10,Katrianne Lehtipalo2, Markus Leiminger1, Johannes Leppa13, Ville Loukonen2, Vladimir Makh-mutov14, Serge Mathot6, Matthew J. McGrath15, Tuomo Nieminen2,16, Tinja Olenius2, AnttiOnnela6, Tuukka Petaja2, Francesco Riccobono3, Ilona Riipinen17, Linda Rondo1, Taina Ruuska-nen2, Filipe D. Santos4, Simon Schallhart2, Ralf Schnitzhofer5, John H. Seinfeld8, Mario Simon1,Mikko Sipila2,16, Yuri Stozhkov14, Frank Stratmann18, Antonio Tome4, Jasmin Trostl3, GeorgiosTsagkogeorgas18, Petri Vaattovaara10, Yrjo Viisanen12, Annele Virtanen10, Aron Vrtala11, PaulE. Wagner11, Ernest Weingartner3, Heike Wex18, Christina Williamson1, Daniela Wimmer1,2,Penglin Ye7, Taina Yli-Juuti2, Kenneth S. Carslaw9, Markku Kulmala2,16, Joachim Curtius1, UrsBaltensperger3, Douglas R. Worsnop2,19, Hanna Vehkamaki2, and Jasper Kirkby61 Goethe-University of Frankfurt, Institute for Atmospheric and Environmental Sciences, 60438 Frankfurt am Main,Germany2 University of Helsinki, Department of Physics, FI-00014 Helsinki, Finland3 Paul Scherrer Institute, Laboratory of Atmospheric Chemistry, CH-5232 Villigen, Switzerland4 SIM, University of Lisbon and University of Beira Interior, 1749-016 Lisbon, Portugal5 Ionicon Analytik GmbH and University of Innsbruck, Institute for Ion and Applied Physics, 6020 Innsbruck, Austria6 CERN, CH-1211 Geneva, Switzerland7 Carnegie Mellon University, Center for Atmospheric Particle Studies, Pittsburgh, PA 15213, USA8 California Institute of Technology, Division of Chemistry and Chemical Engineering, Pasadena, CA 91125, USA9 University of Leeds, School of Earth and Environment, LS2-9JT Leeds, United Kingdom

33

Chapter 4. Molecular understanding of amine-sulphuric acid particle nucleation

10 University of Eastern Finland, FI-70211 Kuopio, Finland11 University of Vienna, Faculty of Physics, 1090 Vienna, Austria12 University of Helsinki, Department of Chemistry, FI-00014 Helsinki, Finland13 Finnish Meteorological Institute, FI-00101 Helsinki, Finland14 Lebedev Physical Institute, Solar and Cosmic Ray Research Laboratory, 119991 Moscow, Russia15 Department of Biophysics, Graduate School of Science, Kyoto University, 6068502 Kyoto, Japan16 Helsinki Institute of Physics, University of Helsinki, FI-00014 Helsinki, Finland17 University of Stockholm, Department of Applied Environmental Science, SE-10961 Stockholm, Sweden18 Leibniz Institute for Tropospheric Research, 04318 Leipzig, Germany19 Aerodyne Research Inc., Billerica, MA 01821, USA

Submitted to Science

Summary

Nucleation of particles from trace atmospheric vapours is thought to provide up to half ofglobal cloud condensation nuclei but, despite its importance for climate, the process is poorlyunderstood. It was recently shown that the presence of ammonia cannot explain the highrates of nucleation of sulphuric acid particles in the lower atmosphere. Using the CLOUDchamber at CERN, we show here that atmospheric mixing ratios of dimethylamine of a fewparts per trillion by volume enhance the formation rate of sulphuric acid particles by morethan a factor 100 compared with ammonia, reaching atmospheric rates. Molecular analysisof the clusters reveals that the faster nucleation is due to a strong base-stabilisation mecha-nism involving amine-acid pairs. The ion-induced contribution is generally small but grows inimportance as the total nucleation rate decreases, exceeding 20 % below 0.5 cm−3 s−1. Thisindicates that the influence of galactic cosmic rays on the nucleation of amine-sulphuric acidparticles is significant only when the overall formation rate is low. Our experimental measure-ments are well reproduced by a dynamical model based on quantum chemical calculationsof molecular cluster binding energies, without any free parameters. These results underscorethe importance of considering the influence of amine emissions—as well as sulphur dioxide—when assessing the impact of anthropogenic activities on aerosol radiative forcing of presentand future climate.

4.1. Introduction

Clouds have an important influence on climate by reflecting incoming solar radiation andby absorbing and emitting long-wave radiation. Clouds form on aerosol particles—tiny liquidor solid particles suspended in the air—above a size of around 50 nm. Increases of atmosphericaerosol particle concentrations cause a net cooling of climate by scattering sunlight and byleading to smaller but more numerous cloud droplets, which makes clouds brighter and ex-tends their lifetime (Lohmann and Feichter, 2005). The increased amount of aerosol in the at-mosphere caused by human activities is thought to have offset a large fraction of the warmingcaused by greenhouse gases (IPCC, 2007). By current estimates, up to half of all cloud droplets

34

4.1. Introduction

are formed on cloud condensation nuclei (CCN) that were nucleated from the clustering oftrace vapours rather than being directly emitted into the atmosphere as particles (Merikantoet al., 2009). However, despite its importance to climate radiative forcings and feedbacks, theprocess is poorly understood. In particular, an improved physical and chemical understandingis required of nucleation rates and growth rates to CCN sizes, and of the trace atmosphericvapours that drive these processes.

Sulphuric acid (H2SO4) is considered to be the primary vapour responsible for nucleationof particles in the atmosphere since it has an extremely low saturation vapour pressure and isproduced effectively in one step from a ubiquitous volatile vapour (by photo-oxidation of sul-phur dioxide, SO2). However, peak daytime H2SO4 concentrations in the atmospheric bound-ary layer are typically around 106–3·107 cm−3 (0.04–1.2 parts per trillion by volume, pptv) (Ker-minen et al., 2010) which results in negligible H2SO4–H2O binary homogeneous nucleation(Kirkby et al., 2011). Consequently a third vapour is required to stabilise the embryonic clusters,whose identity has widely been assumed to be ammonia (NH3). However, recent measure-ments have shown that NH3–H2SO4–H2O ternary nucleation can account for only one-tenthto one-thousandth of the rates observed in the lower atmosphere (Kirkby et al., 2011).

Ammonia enhances the nucleation rate of sulphuric acid particles by means of a base-stabilisation mechanism involving the formation in the cluster of strongly bound acid-basepairs (Kirkby et al., 2011), thereby reducing evaporation rates and lowering the energy barrierto nucleation (Napari et al., 2002). This suggests that stronger base vapours present in theatmosphere may further enhance nucleation. An important class of such vapours is amines,which are derivatives of ammonia in which one or more hydrogen atoms are replaced by analkyl or aryl group, while retaining a nitrogen atom with a lone electron pair and base prop-erties. Amine-enhanced nucleation of sulphuric acid aerosol has been proposed theoretically(Kurten et al., 2008; Loukonen et al., 2010; Paasonen et al., 2009) and is supported by numerousatmospheric and laboratory measurements (Angelino et al., 2001; Makela et al., 2001; Murphyet al., 2007; Facchini et al., 2008; Barsanti et al., 2009; Berndt et al., 2010; Smith et al., 2010;Zhao et al., 2011; Erupe et al., 2011; Yu et al., 2012; Zollner et al., 2012; Chen et al., 2012). Howeverall laboratory studies of amine-sulphuric acid nucleation so far have involved amine and/orsulphuric acid mixing ratios well above atmospheric values. Amine emissions are dominatedby anthropogenic activities (mainly from animal husbandry), but around 30 % of emissionsarise from the breakdown of organic matter in the oceans and 20 % from biomass burningand the soil of boreal forests (Facchini et al., 2008; Ge et al., 2011). Atmospheric mixing ratiosvary considerably (Ge et al., 2011; Yu and Lee, 2012). Samples of individual aliphatic amines ob-tained at urban and rural sites in southern Sweden (Gronberg et al., 1992) ranged between 0.5and 40 pptv, although these and other measurements may be biased high by inclusion of theparticulate fraction.

Despite extensive studies over the last decade, important open questions remain on therole of amines in atmospheric nucleation. The present work addresses four of these. Firstly,it is not yet known whether amines and sulphuric acid at atmospheric concentrations can re-produce observed nucleation rates. Secondly, the molecular mechanism of amine-sulphuricacid nucleation under atmospheric conditions has not yet been measured. Thirdly, the role ofions in amine-sulphuric acid nucleation is unknown. Since ions in the free troposphere and themarine boundary layer result mainly from galactic cosmic rays (GCR), their role in atmosphericnucleation is of considerable interest as a possible physical mechanism for solar-climate vari-

35


ability (Kirkby, 2007). Finally, state-of-the-art theoretical expectations for amine ternary nucle-ation based on quantum chemistry have not yet been tested with well-controlled experimentaldata.

4.2. Nucleation rates

We report here results from the CLOUD (Cosmics Leaving OUtdoor Droplets) experimentat CERN (for experimental details see supplementary material, Fig. 1a, and the supplemen-tary information of Kirkby et al. (2011)). The data were obtained during three campaigns atthe CERN Proton Synchrotron (PS) between October 2010 and November 2012, and representthe first measurements—in either the laboratory or atmosphere—of amine-sulphuric acid nu-cleation at vapour concentrations characteristic of the atmospheric boundary layer. Dimethy-lamine (DMA; C2H7N) was selected for this study since it is expected to have cluster bindingenergies representative of other light alkyl amines (Kurten et al., 2008; Paasonen et al., 2012).All measurements were performed at 1 atm, 278 K, and 38 % relative humidity (RH).

The nucleation rates (cm−3 s−1) were measured under neutral (Jn), ground-level galacticcosmic ray (Jgcr ), or charged pion beam (Jch) conditions. Neutral nucleation rates were mea-sured without any beam and with an internal electric field of about 20 kV/m which clears ionsfrom the chamber in about one second. For GCR and beam conditions, the electric field wasset to zero, leading to ion pair concentrations around 400 cm−3 for Jgcr , representative of theboundary layer, and around 3000 cm−3 for Jch, representative of the top of the troposphere.Both Jgcr and Jch comprise the sum of neutral and ion-induced nucleation rates, at their re-spective ion concentrations, whereas Jn measures the neutral rate alone. The nucleation ratesare determined at 1.7 nm mobility diameter (1.4 nm mass diameter).

In order to derive nucleation rates from the formation rates measured with a particlecounter, corrections are applied to account for losses between 1.7 nm and the detection sizethreshold (see supplementary material). The corrections use experimentally measured wallloss rates, dilution rates and particle growth rates in this size range (coagulation losses are neg-ligible). The growth rates are measured with a scanning particle size magnifier (PSM) and areindependently verified by other counters. In order to minimise the sensitivity to clusters below1.7 nm, the nucleation rates reported here are based on a condensation particle counter (CPC)with 50 % detection threshold at 3.2 nm mobility diameter. The derived nucleation rates wereverified with independent measurements from other CPCs at different detection thresholds.

The nucleation rates are shown in Fig. 4.1 as a function of [H2SO4] at 278 K and 38 % rela-tive humidity. Following previous conventions, the H2SO4 concentration is derived from chan-nel 97 (Th) of the chemical ionisation mass spectrometer (CIMS), which measures the puremonomer signal, HSO−4 , after charging in the CIMS ion source. The measurements fall intothree groups according to the vapours added to the chamber: a) ”binary” (H2SO4–H2O), b) am-monia ternary (NH3–H2SO4–H2O) and c) amine ternary (DMA–H2SO4–H2O). Here ”binary” in-cludes previously measured nucleation rates (Kirkby et al., 2011) under conditions of contam-inant NH3 and DMA which are estimated from later campaigns to be <2 pptv and <0.1 pptv,respectively, for the conditions of (Kirkby et al., 2011). These contaminant mixing ratios are be-low the minimum directly-measurable values (5 pptv for NH3 and and 0.2 pptv for DMA) and

36

4.2. Nucleation rates

are determined from precise calibration of the trace gas delivery systems and molecular anal-ysis of the nucleating clusters.

The measurements show that the presence of only 5 pptv DMA gives rise to nucleationrates more than 100 times faster than those with ammonia (Fig. 4.1). Additional DMA up to140 pptv results in less than a factor three further rate increase, indicating that amine mix-ing ratios of about 5 pptv are sufficient to reach the rate limit for amine ternary nucleationunder atmospheric conditions ([H2SO4]<∼3·10 cm−3). The measured H2SO4–DMA nucleationrates fall within the range of atmospheric observations (Sihto et al., 2006; Kuang et al., 2008;Paasonen et al., 2010), although the atmospheric data show significant variability and havea shallower slope than the CLOUD measurements. Part of the ambient variability may bedue to variations of (unmeasured) amine concentrations, variations of particle condensationsinks and temperature, or differences in calibration of the individual CIMS instruments usedin the various atmospheric studies. Amine ternary nucleation alone cannot explain all of theatmospheric observations below [H2SO4]=2·106 cm−3. However at higher concentrations themaximal amine ternary nucleation rates are at the upper edge of the atmospheric values; nu-cleation rates at lower amine levels will thus sweep through the ambient data. Thus amineternary nucleation can explain a significant fraction of atmospheric nucleation.

In order to compare our measurements with theoretical expectations we have explicitlysimulated all possible collision, coagulation, evaporation and fragmentation reactions for acertain set of clusters (see supplementary material for further details). Collision and coagula-tion rates are computed from kinetic gas theory, while evaporation and fragmentation ratesare obtained from quantum chemistry (Ortega et al., 2012). Dynamical simulations are carriedout using the Atmospheric Cluster Dynamics Code model (ACDC) (McGrath et al., 2012). Wehave calculated the formation of neutral and both positively- and negatively-charged clusterscontaining sulphuric acid, ammonia and DMA. Due to computing limitations, we have so farmodelled the formation and evaporation of clusters containing up to four sulphuric acid andfour base molecules (mobility diameters 1.2 to 1.4 nm). The diameters of our largest computedclusters are therefore smaller than the 1.7 nm size at which the CLOUD formation rates (J1.7)are determined. The results are shown by the coloured bands in Fig. 4.1. The model reproducesall qualitative features of the experimental measurements and even the quantitative rates arein good agreement, although somewhat higher. Part of this discrepancy is due to the smallersize—and hence higher formation rate—of the modelled clusters, and another part is due tothe poorly-known sticking probability for neutral-neutral collisions (the model assumes unitsticking probability for all charged-neutral collisions). Nevertheless, the agreement is impres-sive considering that the model uses first-principles evaporation rates with no fitted parame-ters. Computational studies (not presented here) indicate that the nucleation rates are ratherinsensitive to temperature or relative humidity, so the experimental measurements obtainedat 278 K and 38 % RH may be considered representative of a wide range of boundary layer con-ditions.

The nucleation rates Jn , Jgcr and Jch vs. DMA are shown in Fig. 4.2a. All measurementsin this figure have been scaled to [H2SO4]=2.0·106cm−3 using the fitted slopes, n, from Fig. 4.1.Addition of only 5 pptv DMA enhances the nucleation rate of sulphuric acid particles by almost6 orders of magnitude, but further DMA up to 140 pptv produce no further increase of nucle-ation rate. The theoretical expectations are indicated by the yellow band for a range of stickingprobabilities between 0.1 and 1.0. The measured neutral, GCR, and charged (beam) nucleation

37


rates are indistinguishable, within experimental uncertainties. However a more sensitive de-termination of the ion-induced nucleation rate, Jiin = J+iin + J−iin, is obtained directly from ionmeasurements with the neutral cluster and air ion spectrometer (NAIS) rather than from acomparison of Jn with Jgcr or Jch (which measure the combined neutral and ion-induced rates).The ion-induced fraction, Jiin/J1.7, (Fig. 4.2b) averages about 20 % at 0.5 cm−3 s−1 but grows inrelative importance as the total nucleation rate decreases. This indicates that the influenceof galactic cosmic rays on the nucleation of amine-sulphuric acid particles is significant onlywhen the overall formation rate is low. No difference is measured for the ion-induced frac-tion under GCR or beam conditions (Fig. 4.2b). Although the ion pair concentration is largerfor beam conditions, this is compensated by a shorter ion-ion recombination lifetime, whichreduces the available time for monomers to arrive before neutralisation.

4.3. Molecular composition of clusters

The molecular composition of nucleating charged clusters in the presence of DMA is shownin Fig. 4.3 for a) negative and b) positive ions, measured in the atmospheric pressure interfacetime-of-flight mass spectrometers (APi-TOF, TOFWERK AG and Aerodyne Research, Inc.) (Junni-nen et al., 2010). The predominant negative clusters include an HSO−4 or HSO−5 ion. The latter isdeprotonated peroxysulphuric acid, a more highly oxidised sulphuric acid ion whose presencevaries with the ozone concentration in the chamber (it is absent when no ozone is present).We found no indication that the nucleation rates would be sensitive to the relative contribu-tion of these sulphuric acid ion species. Contaminant NO−3 ions are also detected, but at muchlower concentrations. The predominant positive clusters include a protonated dimethylamineion, DMAH+ (C2H7N·H+), in association with H2SO4 and DMA. The remaining positive ions areprotonated light organic contaminants, mostly also N-containing.

38


Figure 4.1.: Nucleation rate of new particles at 1.7 nm mobility diameter as a function of[H2SO4] measured by CLOUD compared with observations in the atmospheric boundary layer(small coloured squares) (Sihto et al., 2006; Kuang et al., 2008; Paasonen et al., 2010). TheCLOUD data, which were recorded at 38 % relative humidity and 278 K, show: a) Jgcr with onlyH2SO4, water and contaminant vapours (<0.1 pptv dimethylamine, DMA, and <2 pptv NH3)in the chamber (open blue circles and curve 1); b) Jgcr with NH3 at mixing ratios of 2–10 pptv(green triangles and curve 2) and 10–170 pptv (brown triangles and curve 3); and c) Jn, Jgcr andJch with 10 pptv contaminant NH3 and with DMA at mixing ratios of 1–5 pptv (filled purple cir-cles and curve 4), 5–13¡,pptv (filled blue-cyan circles and curve 5) and 13–140 pptv (filled green-yellow-orange-red circles and curve 6). The straight parts of the fitted curves indicate powerlaws, J ∝[H2SO4]n, with fitted slopes n of 1) 3.1±0.6, 2) 3.1±0.2, 3) not measured, 4) 4.1±0.7,5) 3.4±0.1 and 6) 3.8±0.2. The flattening of curves 1–3 at higher [H2SO4] results both from sat-uration of the ion production rate and also from a smaller contribution of ammonia ternarynucleation at high acid concentrations. The bars indicate 1σ total errors, although the over-all ±35 % scale uncertainty on [H2SO4] is not shown. Theoretical expectations (ACDC model)are indicated for ”binary” nucleation with 1 pptv contaminant NH3 (dashed blue line) and for10 pptv DMA plus 10 pptv NH3 (dashed brown line), assuming 0.5 neutral-neutral sticking prob-ability. The theoretical prediction for 150 pptv NH3 (not shown in the figure for clarity) is con-sistent with the experimental measurements within the uncertainties. The coloured bandscorrespond to the uncertainty range of the theory: a) +1 and -1 kcal/mol binding energy (blueband) and c) sticking probabilities between 0.1 and 1.0 (orange band), for the lower and upperlimits, respectively.

39

Figure 4.2.: Contribution of ions and dimethylamine to ternary nucleation at 278 K and 38 %RH: a) neutral, GCR and charged (beam) nucleation rates as a function of dimethylaminemixing ratio measured by the ion chromatograph (IC) and b) ion-induced fractions, Jiin/J1.7,where Jiin is measured by the neutral cluster and air ion spectrometer (NAIS), as a function ofJ1.7 for GCR and charged (beam) conditions with DMA=5–140 pptv. For panel a), all measure-ments have been scaled to [H2SO4]=2.0×106 cm−3 (0.08 pptv) using the fitted slopes in Fig. 4.1.The point shown at 0.1 pptv DMA corresponds to the mean projected measurement of Jgcrat contaminant-level DMA and NH3. The bars indicate 1σ total errors, and include correlatedsystematic contributions. The theoretical expectations (ACDC model, assuming 0.5 neutral-neutral sticking probability) are indicated by the dashed brown lines. The orange bands corre-spond to the model uncertainty range (sticking probabilities between 0.1 and 1.0; the experi-mental data lie below 0.5).


Negative cluster nucleation (Fig. 4.3a) proceeds as follows. Here we use the label (n, m) toindicate the number of sulphuric acid (nSA) and dimethylamine (mDMA) molecules in pure SA-DMA clusters, where n and m include both neutral and charged species. The first step is aciddimer formation (2,0): HSO−4 · H2SO4 (for simplicity the ”HSO−4 ” ion implies either HSO−4 orHSO−5 ). This constitutes an acid-base pair since HSO−4 is a Lewis base (an electron pair donor).Hence the first charged cluster to which DMA can bind to form an acid-base pair is the acidtrimer. The most abundant acid trimer is found to contain two DMA molecules (3,2). There-after, each additional acid molecule is stabilised by one additional DMA molecule, followinga sequence of acid-base pairs: (3,2)→ (4,3)→ (5,4)→ (6,5)→ (n,n − 1). Our calculations sug-gest that the process involves mainly the accretion of SA·DMA (dimethylaminium bisulphate)clusters, but the process may also involve stepwise addition of an SA molecule followed by aDMA molecule. Beyond (7, 6) clusters, there is evidence for further neutralisation of the acid byadditional DMA (partial formation of dimethylaminium sulphate). Positive cluster nucleation(Fig. 4.3b) proceeds in a similar way. Here DMAH+ is a Lewis acid and so binds only weaklywith H2SO−4 . Hence the first positive cluster is a DMAH+ ion together with a single SA·DMAacid-base pair (1,2). Thereafter, the cluster grows by accretion of SA·DMA pairs, exactly as seenfor negative clusters.

The amine-sulphuric acid nucleation process follows the same base-stabilisation mecha-nism previously found for ammonia ternary nucleation (Kirkby et al., 2011). However the char-acteristic 1:1 SA:DMA pattern seen in Fig. 4.3 appears already at the lowest experimental valuesof DMA near 3 pptv. Furthermore, 3 pptv DMA was found to completely displace all ammo-nia molecules from the nucleating clusters under measured NH3 contaminant conditions of10 pptv. This has been anticipated both experimentally (Bzdek et al., 2010a) and theoretically(Kupiainen et al., 2012). Pure tetramers (4,0) are not seen, in agreement with previous observa-tions that the tetramer is relatively unstable and evaporates rapidly (Kirkby et al., 2011). How-ever, it is interesting to note that DMA stabilises the negatively charged trimer, whereas NH3

is first seen in the charged tetramer (Kirkby et al., 2011). This observation explains the largerate enhancement measured for amine-sulphuric acid nucleation compared with ammonia-sulphuric acid: amines form stronger acid-base bonds and so stabilisation of charged sulphuricacid clusters commences at smaller sizes.

Since both HSO−4 and DMA are Lewis bases, each can form an acid-base bond with H2SO4.In fact HSO−4 is the stronger base, as demonstrated by its much stronger binding energy withH2SO4 (Table A.1 and Kurten et al. (2008)). The only fundamental difference is that not morethan one HSO−4 ion can be present in the cluster because of electrostatic repulsion. So, al-though the APi-TOF measures only charged clusters in the CLOUD chamber, it suggests thatneutral nucleation proceeds by the same mechanism, namely first the formation of an acid-base pair (SA·DMA)—equivalent to the acid-base pair (SA·HSO−4 ) seen in charged nucleation(Fig. 4.3a)—and subsequently the accretion of additional SA·DMA pairs (or else an SA moleculefollowed by a DMA molecule). This is also indicated by the ACDC model.

There is direct experimental evidence to support this picture of the neutral nucleationmechanism. The CIMS measures both neutral H2SO4 monomers and dimers by chemical ion-isation (higher masses are outside the operating range of the instrument). The neutral dimervs. monomer concentrations are shown in Fig. 4.4 before and after addition of DMA. Allmeasurements are consistent with a dimer concentration that depends quadratically on themonomer concentration, as expected. These measurements were all performed in the pres-

41


ence of the electric clearing field, ensuring only neutral clusters are present in the CLOUDchamber. Furthermore, the experiments allowed sufficient time for the dimer concentrationsto reach steady state between formation rates and loss rates to the walls and to larger clusters,assuming collision-rate-limited dimer formation times (1600–160 s for [H2SO4] in the range106–107 cm−3, respectively).

The measured H2SO4 dimer concentrations during both binary and ammonia ternary nu-cleation events are, within experimental uncertainties, consistent with the estimated back-ground production of dimers by the CIMS ion source (grey band in Fig. 4.4). However, whenDMA is added, there is a marked enhancement of acid dimers, reaching concentrations aboutsix orders of magnitude higher than those expected from the pure binary system (Hansonand Lovejoy, 2006). We infer from the observed absence of DMA on the negatively chargedmonomer or dimer (Fig. 4.3a) that, after charging in the CIMS, all clusters containing one H2SO4

molecule will be detected as a free charged monomer and all clusters containing two H2SO4

molecules will be detected as a free charged dimer—regardless of whether or not they wereoriginally clustered with base molecules. Consequently, DMA-enhanced neutral dimers in theCLOUD chamber are detected in the CIMS as free charged dimers. Furthermore, the sulphuricacid concentration measured by the CIMS represents the sum of the free monomers and allclusters containing exactly one H2SO4 molecule.

In the presence of a few pptv DMA, neutral clusters containing two H2SO4 molecules ap-proach concentrations as high as 1–10 % of the H2SO4 monomers at 106–107 cm−3, respectively.This corresponds to about one tenth the kinetic limit for dimer formation (indicated by thebrown line in Fig. 4.4). These observations show that the neutral H2SO4 dimers are stronglystabilised by DMA, as predicted by quantum chemical calculations (orange band in Fig. 4.4, andKurten et al. (2008)), providing direct experimental support for the neutral nucleation mecha-nism inferred above.

A previous experiment (Petaja et al., 2011) has measured unexpectedly high dimer concen-trations in a laminar flow tube and concluded that a stabilising contaminant must be present,although none was measured. This has been proposed (Petaja et al., 2011) to explain the high”binary” nucleation rates previously measured by Sipila et al. (2010) in the same flow tube. An-other experiment (Chen et al., 2012) has measured high sulphuric acid dimer formation rateslinked to amine mixing ratios of around 1 ppbv and higher. However, our results are the first tolink high neutral H2SO4 dimer concentrations with amines at atmospheric levels.

42

Figure 4.3.: Molecular composition of charged clusters measured by the APi-TOF for a) nega-tive and b) positive particles at Jgcr=1.2 cm−3 s−1, [H2SO4]=4.0×106 cm−3, 11 pptv NH3, 9.4 pptvDMA, 38 % relative humidity and 278 K. The figures show the mass defect vs. clustermass/charge; each circle represents a distinct molecular composition and its area representscounts/s. The labels (n,m) indicate the number of sulphuric acid (nSA) and dimethylamine(mDMA) molecules in pure SA-DMA clusters, including both neutral and charged species. Thehigh resolution measurement of cluster mass resolves differences in the nuclear binding ener-gies of the constituent atoms (”mass defect” with respect to integer cluster mass), providingunambiguous identification of the atomic composition of most clusters. The addition of a sin-gle SA (H2SO4) or DMA (C2H7N) molecule, or SA·DMA pair, to any cluster displaces it on the plotby a vector distance indicated by the grey arrows. The negative ions arise mostly from sulphuric(HSO−4 ) or peroxysulphuric (HSO−5 ) acid; the positive ions mostly involve protonated dimethy-lamine (C2H7NH+). The red circles represent the charged pure monomer, dimer and trimer ofsulphuric acid; green circles represent clusters containing sulphuric acid and dimethylaminealone; cyan circles are clusters containing ammonia in addition; other clusters are grey or blackcircles. Since water molecules are lightly bound on the clusters and evaporate rapidly, they arenot detected by the APi-TOF (see supplementary material).


Figure 4.4.: Neutral sulphuric acid dimer vs. monomer concentrations measured by the CIMSwithout DMA in the CLOUD chamber (open circles) and with 3–140 pptv DMA (filled rainbow-coloured circles), at 10 pptv contaminant NH3, 38 % relative humidity and 278 K. The dimer con-centrations are derived assuming the same CIMS calibration factor as that of the monomers.Ions are absent from the CLOUD chamber (clearing field on). The bars indicate 1σ counting er-rors. The curves show the expected neutral H2SO4 dimer concentrations for the binary H2SO4–H2O system (short-dashed black line) (Hanson and Lovejoy, 2006); for production by the CIMSion source (dashed black line and grey band indicating the estimated uncertainty in the cal-culation); and the theoretical expectation of the ACDC model for 10 pptv DMA, assuming 0.5sticking probability (dot-dashed black line). The orange band corresponds to the model un-certainty range (neutral-neutral sticking probabilities between 0.1 and 1.0). The brown lineindicates the kinetic limit for the dimer formation rate (negligible evaporation). The modelexceeds the kinetic limit at high [H2SO4] since it includes cluster evaporation and predicts ahigh fission rate for 4SA·4DMA→ 2SA·2DMA + 2SA·2DMA. The fitted red line through the DMAmeasurements shows a quadratic dependency on monomer concentration.

44

4.4. Atmospheric implications


Particle nucleation rates observed in the lower atmosphere show a large scatter and a rel-atively mild dependence on [H2SO4], suggesting the influence of other vapours and additionalvariables such as particle condensation sinks. The results reported here show that the nucle-ation of sulphuric acid particles is highly sensitive to the presence of dimethylamine, and thatpart-per-trillion levels of DMA give rise to nucleation rates comparable to atmospheric obser-vations. The results can be generalised to include other atmospheric amines that are expectedto have similar cluster binding energies (Kurten et al., 2008). Our molecular studies establishthat amine sulphuric acid nucleation proceeds by the same base-stabilisation mechanism aspreviously observed for ammonia (Kirkby et al., 2011), in which each additional acid moleculein the cluster is stabilised by one (or occasionally, two) base molecules. This nucleation mech-anism with amines has been theoretically proposed (Kurten et al., 2008) but not previouslymeasured.

Since amine mixing ratios of a few pptv are frequently present in the boundary layer, nu-cleation of amine-sulphuric acid particles is likely to be an important atmospheric process.However it cannot account for the relatively high nucleation rates observed in the atmosphereat sulphuric acid concentrations below 2·106cm−3. Indeed, it is important to stress that ourmeasurements do not exclude the possibility that nucleation of sulphuric acid particles in theatmosphere may also proceed with other vapours, such as highly-oxidised organic species ofvery low volatility, operating via different mechanisms. In such cases, ultra low amine concen-trations may still enhance the nucleation process by forming stable acid-base pairs with somefraction of the sulphuric acid molecules in an embryonic cluster. Atmospheric observationssuggest that amines are an important component of freshly-nucleated aerosol particles in awide range of environments that include pristine boreal forests, marine regions and pollutedurban environments (Makela et al., 2001; Facchini et al., 2008; Smith et al., 2010). Direct atmo-spheric observation of amine-sulphuric acid nucleation at the molecular level has not yet beenreported. However, the same is true of all other atmospheric nucleation owing to the com-plexity of atmospheric molecular cluster spectra and the difficulty of identifying the nucleat-ing clusters in the presence of high ’spectator’ backgrounds (Ehn et al., 2010). Furthermore,amine-enhanced nucleation of sulphuric acid particles in the atmosphere may be obscured(and, indeed, further enhanced) by the presence of additional molecular species on the clus-ters, such as oxidised organic compounds.

The ion-induced contribution to amine ternary nucleation is generally small but growsin importance as the total nucleation rate decreases, exceeding 20 % below 0.5 cm−3 s−1 (cor-responding to [H2SO4]= 2·106cm−3). Taken together with previous CLOUD measurements ofion-induced binary- and ammonia ternary nucleation of sulphuric acid particles (Kirkby et al.,2011), this suggests that galactic cosmic rays play a significant role in atmospheric nucleationwhen the overall nucleation rate is low and below the ion-pair production rate (2–20 cm−3 s−1,depending on altitude and latitude (Usoskin et al., 2004)). It is quite likely that this observa-tion is a general feature of atmospheric nucleation, regardless of the chemical species involved,since it arises from the enhanced arrival rate of polar molecules (such as H2SO4 or DMA-H2SO4

pairs) onto a charged cluster and also from the stronger molecular binding energy within acharged embryonic cluster (and hence reduced evaporation rate). However, since ion-ion re-combination lifetimes in the atmosphere are around 10 minutes or less, there is a lower limit

45


on the sulphuric acid concentration at which ion-induced nucleation can proceed, due to theslow monomer arrival rate on the cluster (around 5 minutes at [H2SO4]=106 cm−3). Field obser-vations suggest that neutral nucleation dominates in the continental boundary layer (Kulmalaet al., 2007; Manninen et al., 2010), although other analyses conclude that ion-mediated nu-cleation may be important (Yu and Turco, 2011).

The results of the dynamical molecular cluster model reported here differ from those ofprevious studies in three significant ways: firstly the molecular species are chosen from directexperimental observations of the participating species; secondly the evaporation rates are ob-tained from first principles, without any free parameters; and, finally, the predictions are testedagainst the first controlled laboratory measurements to reproduce atmospheric observations.The good agreement with the experimental measurements demonstrates that ab initio theo-retical calculations now constitute a powerful tool to investigate atmospheric particle forma-tion processes, and can be used in future studies to explore chemical species or conditions thathave so far not been experimentally accessible. The findings reported here represent an impor-tant experimental and theoretical advance towards a fundamental molecular understandingof atmospheric particle formation.

The Intergovernmental Panel on Climate Change (IPCC) considers that the increased amountof aerosol in the atmosphere from human activities constitutes the largest present uncertaintyin climate radiative forcing (IPCC, 2007). The effects of anthropogenic aerosols have createdgreat uncertainty in our knowledge of climate sensitivity to increasing greenhouse gases and,in turn, of projected climate change this century (Andreae et al., 2005). The results reportedhere show that the uncertainty is even greater than previously thought since extremely lowamine emissions—which have substantial anthropogenic sources and have not been hithertoconsidered by the IPCC—have a large influence on the nucleation of sulphuric acid particles.Moreover, amine scrubbing is likely to become the dominant technology for CO2 capture fromcoal-fired power plants, so anthropogenic amine emissions are expected to increase in thefuture (Rochelle, 2009; Nielsen et al., 2012). Should ambient amine concentrations increasesubstantially, an increase in nucleation rates, leading to an increase in CCN levels, could be an-ticipated. The effect of such an increase in CCN under present global conditions would needto be considered in future climate projections. This underscores the importance of monitoringamine emissions—as well as sulphur dioxide—when assessing the impact of anthropogenicactivities on aerosol radiative forcing of both regional and global climate.

Acknowledgements

We would like to thank CERN for supporting CLOUD with important technical and fi-nancial resources, and for providing a particle beam from the CERN Proton Synchrotron. Wealso thank J.-L. Agostini, P. Carrie, L.-P. De Menezes, F. Josa, I. Krasin, R. Kristic, O.S. Maksumov,S.V. Mizin, R. Sitals, A. Wasem and M. Wilhelmsson for their important contributions to the ex-periment. We thank the CSC Centre for Scientific Computing in Espoo, Finland for computertime, and J. Pierce and P. Paasonen for helpful discussions. This research has received fundingfrom the EC Seventh Framework Programme (Marie Curie Initial Training Network ”CLOUD-ITN” no. 215072, MC-ITN ”CLOUD-TRAIN” no. 316662, ERC-Starting ”MOCAPAF” grant no. 57360and ERC- Advanced ”ATMNUCLE” grant no. 227463), the German Federal Ministry of Educa-

46


tion and Research (project nos. 01LK0902A and 01LK1222A), the Swiss National Science Foun-dation (project nos. 200020 135307 and 206620 130527), the Academy of Finland (Center ofExcellence project no. 1118615), the Academy of Finland (135054, 133872, 251427, 139656, 139995,137749, 141217, 141451), the Finnish Funding Agency for Technology and Innovation, the NesslingFoundation, the Austrian Science Fund (FWF; project no. P19546 and L593), the PortugueseFoundation for Science and Technology (project no. CERN/FP/116387/2010), the Swedish Re-search Council, Vetenskapsradet (grant 2011-5120), the Presidium of the Russian Academy ofSciences and Russian Foundation for Basic Research (grants 08-02-91006-CERN and 12-02-91522-CERN), and the U.S. National Science Foundation (grants AGS1136479 and CHE1012293).

47

5Cyclobutyl methyl ketone as a

model compound for pinonic acidto elucidate oxidation mechanisms

Arnaud P. Praplan∗, Peter Barmet, Josef Dommen, Urs BaltenspergerLaboratory of Atmospheric Chemistry, Paul Scherrer Institut, 5232 Villigen PSI, Switzerland∗now at: Department of Physics, University of Helsinki, Helsinki, Finland

Published on 16 November 2012 in Atmospheric Chemistry and Physics, 12, 10749–10758.

Abstract. Although oxidation of the atmospherically relevant compound α-pinene has beenextensively studied, chemical mechanisms leading to the formation of later generation oxi-dation products remain poorly understood. The present work uses cyclobutyl methyl ketone(CMK) to study the oxidation mechanism of pinonic acid, an α-pinene reaction product, byhydroxyl radical (OH) CMK has a similar but simpler chemical structure compared to pinonicacid. Succinic acid, 4-hydroxybutanoic acid and 4-oxobutanoic acid were identified as first gen-eration products of CMK. These observed organic acids were compared to compounds foundin secondary organic aerosol formed from the oxidation of α-pinene. Results suggest that 3-methyl-1,2,3-butanetricarboxylic acid (MBTCA) terpenylic acid and diaterpenylic acid acetateare first generation products of OH oxidation of pinonic acid. Therefore, there is strong evi-dence that OH oxidation greatly increases the oxygenation of organic compounds (e.g. mono-carboxylic acid to tricarboxylic acid) through radical mechanisms, without requiring a stableintermediate. These observations cannot be explained by traditional atmospheric chemistrymechanisms.

49

Chapter 5. CMK as a model compound to elucidate oxidation mechanisms

5.1. Introduction

Monoterpenes are biogenic volatile organic compounds (VOCs) emitted into the atmo-sphere by vegetation. They represent 11 % of the total biogenic VOC emissions (Guenther et al.,1995). α-Pinene is a major representative of this class of compounds and can form secondaryorganic aerosol (SOA). However, the understanding of its oxidation mechanisms and SOA for-mation yield is still subject of research.

Szmigielski et al. (2007) identified 3-methyl-1,2,3-butanetricarboxylic acid (MBTCA) as a prod-uct from α-pinene SOA. Because of its high oxygen-to-carbon (O : C) ratio and its distinctstructure, it was suggested as a tracer compound for aged α-pinene SOA. Muller et al. (2012)demonstrated experimentally that MBTCA is generated from pinonic acid oxidation by hydroxylradical (OH), where pinonic acid is a primary product fromα-pinene ozonolysis. The structuresof further products of α-pinene SOA were identified recently: terpenylic acid and diaterpenylicacid acetate (Iinuma et al., 2008; Claeys et al., 2009).

Laboratory studies of the pinonic acid oxidation in the gas phase are difficult, becauseof its relatively low volatility. Thus, high pinonic acid mixing ratios cannot be achieved ina smog chamber. Even if the main products (such as MBTCA) could be detected, other poten-tial products would be present at too low levels to be detected. Therefore, cyclobutyl methylketone (CMK) was tested as a surrogate of pinonic acid to understand the oxidation mecha-nisms and the chemical structure of the products formed. Pinonic acid contains a four-carbonring structure which is common to many monoterpenes. A better understanding of the reac-tions following its opening can help to identify structures of products observed from the oxi-dation of α-pinene and other monoterpenes. Figure 5.1 depicts the chemical structure of CMKand of products from oxidation by OH: succinic acid, butyrolactone and 4-acetoxybutanoicacid. The presented analogy with pinonic acid (grey moieties accounting for structural differ-ences between the two systems) assumes that terpenylic acid and diaterpenylic acid acetatein α-pinene SOA are formed as second generation products from the oxidation of pinonic acid,similarly to MBTCA. Butyrolactone (in the same way as terpenylic acid) is expected to be hydrol-ysed during sampling with water (Sect. 5.2), so that it is detected as 4-hydroxybutanoic acid.It is however not possible to determine which amount of this acid is formed directly in thegas phase and which amount results from the hydrolysis of the lactone. Because diaterpenylicacid (hydrolysed form) is usually detected in small amounts in ambient samples compared toterpenylic acid (lactone), one may assume that it is formed from the hydrolysis of the lactone.Because 4-oxobutanoic acid was observed during CMK oxidation experiments, it is included inFig. 5.1 and two possible structures for analogous products from the α-pinene (or pinonic acid)oxidation are suggested.

5.2. Experimental

5.2.1. Smog chamber and experiments

The experiments presented in Table 5.1 were performed at the smog chamber of the PaulScherrer Institute (Paulsen et al., 2005). It consists of a 27-m3 Teflon® bag irradiated by Xe-arc

50

5.2. Experimental

Figure 5.1.: Chemical structures of cyclobutyl methyl ketone (CMK) and measured or expectedproducts. The added grey moieties represent the analogous products in the pinonic acidsystem. Butyrolactone (and terpenylic acid) are expected to hydrolyse during sampling,so that 4-hydroxybutanoic acid (or diaterpenylic acid) are detected. No equivalent prod-uct of 4-oxobutanoic acid was identified yet, so that two possible structures are proposedfor the pinonic acid system: 2-(2-methyl-1-oxopropan-2-yl)succinic acid and 3-formyl-2,2-dimethylpentanedioic acid. The numbers on CMK/pinonic acid correspond to carbon atomsbound to hydrogen atom(s) available for abstraction by an hydroxyl radical (OH).

lamps (4× 4 kW) and additional black light tubes emitting mainly between 320 and 400 nm(manufactured by Cleo Performance; 80× 100 W).

Two different ways of producing OH were used: (1) production by tetramethylethene(TME) ozonolysis (”Dark OH”), and (2) production by ozone (O3) photolysis in the absence(”Lights + O3”) or in the presence of nitrogen oxides (NOx, ”Lights + NOx). In the NOx exper-iment, the lights were turned off after 3 h of OH exposure in order to observe the decompo-sition of peroxyacyl nitrates (PANs). CMK was injected through a heated sample bulb (80 °C).The experiments performed with 400 ppbv CMK were also seeded with wet ammonium sul-phate aerosol particles, while the experiment with 1600 ppbv CMK (“high CMK”) was not. Therelative humidity was around 50 % in all experiments.

5.2.2. Instruments

In addition to the usual monitors (for O3 and NOx) and sensors (temperature and rela-tive humidity), a proton-transfer-reaction mass spectrometer (PTR-MS) was used to analyse

51


Table 5.1.: List of performed cyclobutyl methyl ketone (CMK) oxidation experiments.Date CMK1 O3

1 NOx1 ∆CMK2 Seed Comments

concentration 3

29 Jul 2010 400 ppbv 500 ppbv – 30 ppbv 40 µgm−3 TME ozonolysis (“Dark OH”)18 Oct 2010 1600 ppbv 400 ppbv – 145 ppbv no seed lights on for 7.5 h

(“Lights + O3, high CMK”)20 Oct 2010 400 ppbv 400 ppbv – 56 ppbv 38 µgm−3 lights on for 7.25 h (“Lights + O3”)22 Oct 2010 400 ppbv – 400 ppbv NO 27 ppbv 17 µgm−3 lights on for 3 h

400 ppbv NO2 (“Lights + NOx”)1 Nominal concentrations 2 after 3 h of oxidation 3 at the beginning of the oxidation.

Table 5.2.: Reaction rate constants for hydrogen atom abstraction from cyclobutyl methyl ke-tone (CMK) by a hydroxyl radical (OH)). Indices correspond to the carbon atom numbers inFig. 5.1.

cyclobutyl methyl ketone (CMK)

k1 0.102× 10−12 cm3 s−1k2 0.616× 10−12 cm3 s−1k3 (twice) 1.255× 10−12 cm3 s−1k4 0.396× 10−12 cm3 s−1

pinonic acid

k ′1 0.102× 10−12 cm3 s−1k ′2 0.616× 10−12 cm3 s−1k ′3 1.255× 10−12 cm3 s−1k ′4 1.011× 10−12 cm3 s−1k ′5 (twice) 0.167× 10−12 cm3 s−1k ′6 0.862× 10−12 cm3 s−1

gas phase organic compounds. PTR-MS data were corrected for background. Organic acidswere sampled in the gas phase by a wet effluent diffusion denuder (WEDD) and in the aerosolphase by an aerosol collector (AC) described in Takeuchi et al. (2004) and (2005), respectively.The sampling time was 30 min and anions were concentrated on two trace anion concentratorcolumns (TAC-LP1, Dionex). Sampling and analysis of gas and aerosol phase were performed al-ternatingly on each TAC (one for each phase). The separation of the analytes was performed byion chromatography (Dionex DX600 with guard column NG1 and analytical column AS11-HC)with a gradient eluent (0–60 mM OH−). Detection was performed after background suppres-sion (Anion Self-Regenerating Suppressor, ASRS 300 2 mm, Dionex) by conductivity detectionand by a quadrupole mass spectrometer (ThermoScientific MSQ) with electrospray ionisationin negative mode (Fisseha et al., 2004).

52


Figure 5.2.: Chemical mechanism following hydrogen atom abstraction from cyclobutyl methylketone (CMK) by a hydroxyl radical (OH)). The numbers in brackets correspond to the molarmass of the compounds. See text for details.


For all the experiments performed, no significant aerosol formation was observed, so thatthe discussion focuses on the gas phase observations. Furthermore, no mass increase in theparticle phase was measured when seed aerosol was used.

Because MBTCA could be observed in α-pinene oxidation experiments in the absence ofNOx (Muller et al., 2012), our interpretation of the oxidation mechanism is mainly based on thelow-NOx regime. The experiment with NOx was performed in order to compare the results andto confirm that certain m/z of the PTR-MS were correctly attributed to organic hydroperoxides,because organic hydroperoxides do not form at conditions with high NOx.

5.3.1. Hydrogen abstraction

CMK reacts only with OH, which abstracts a hydrogen atom (H), forming an alkyl radical.Table 5.2 presents the reaction rates for the different possible hydrogen atom abstraction re-actions from CMK by OH, derived from structure-activity relationship estimations (Kwok andAtkinson, 1995; Atkinson, 1997). The indices correspond to the atom numbers in Fig. 5.1. Thefastest abstraction occurs at position 3 for CMK and two such positions are available due to thesymmetry of the molecule. Therefore, this route is considered to dominate the gas phase oxi-dation of CMK. Note that there is only one equivalent position in pinonic acid (3′) and that po-sitions 4′ and 6′ (due to the presence of the carboxylic group) show competitive reaction rates.Therefore, the analogous products from hydrogen abstraction at position 3′ in the pinonic acidsystem are expected to have lower yields. In addition, other products are formed in the pinonicacid system which cannot be studied with CMK.

53


Figure 5.3.: Signals of m/z 131 (left) and 113 (right) from the proton-transfer-reaction mass spec-trometer (PTR-MS). The hydroperoxide C6H10O3 is mainly detected at m/z 113 since it loosesa water molecule during protonation in the ionisation region of the PTR-MS. However, at thism/z there is an interference with the 2-acetylcyclobutanone (C6H8O2), which is still measuredin the experiment with NOx, while the hydroperoxide formation channel is suppressed.

The mechanism following the hydrogen atom abstraction at position 3 is depicted inFig. 5.2. The alkyl radical (R) formed reacts immediately with molecular oxygen (O2) to forma peroxyl radical (RO

2). This radical can then react with a hydroperoxyl radical (HO2) to form

a hydroperoxide (C6H10O3) or, in the absence of NOx, it can react with another RO2 to form an

alkoxy radical (RO), labelled “alkoxy A” in Fig. 5.2) as well as alcohol and carbonyl compounds(not shown). In the high NOx case, hydroperoxide formation is suppressed by either the forma-tion of a peroxynitrate (ROONO2) or RO

2 reaction with NO to form RO and NO2 or a nitrate.

Figure 5.3 showsm/z 131 and 113, respectively, as measured with the PTR-MS. The hydroper-oxide C6H10O3 should appear at m/z 131, however, as it looses one water molecule after proto-nation in the PTR-MS, it is not detected at m/z 131 ([M + H]+), but at m/z 113 ([M + H−H2O]+).At thism/z , 2-acetylcyclobutanone (C6H8O2) also appears as an interference. In the NOx exper-iment, only the latter compound contributes to the signal because the hydroperoxide forma-tion is hindered by the presence of NO, lowering the HO

2 concentration. However, Fig. 5.3bshows that the signal at m/z 113 remains relatively low, indicating only little 2-acetylcyclo-butanone formation. This is explained by the fact that the ring opening reaction is favouredover reaction with O2, due to the release of the ring strain (∼27 kcal mol−1, Peeters et al., 2004).

5.3.2. Carbon-carbon bonds dissociation

The two possible ring opening reactions from the “alkoxy A” and the following mechanis-tic steps are depicted in Fig. 5.4. These reactions are expected to occur faster than the bimolec-ular reaction with O2 as previously mentioned. According to structure-activity relationships(Peeters et al., 2004), but without considering the ring strain release discussed above, the 2–3carbon-carbon bond dissociation is expected to be favoured because carbon 2 is more substi-

54


Figure 5.4.: Ring opening chemical mechanism of the alkoxy radical A from Fig. 5.2. Carbon-carbon bond 2–3 or 3–4 can dissociate, leading to the formation of different compounds. Theformed “alkoxy B” can also undergo carbon-carbon bond dissociation. The numbers in bracketscorrespond to the molar mass of the compounds. See text for more details.

tuted than carbon 4. The energy barrier is 3.1 kcal mol−1 lower in the former case. Peeters et al.(2004) discussed the effect of resonance stabilisation and, because the alkyl radical formed bydissociation of the 2–3 carbon-carbon bond is resonance-stabilised by the carbonyl group, itsformation is even more favoured and this route is expected to dominate over the other.

Following the ring opening reaction, the formed alkyl radical reacts rapidly with O2 toform a peroxy radical. This radical can react with HO

2 to form a hydroperoxide with a nominalmass of 146 or it can also react with either RO

2 or NO (in the NOx experiment) to form anotheralkoxy radical (“alkoxy B”). Upon reaction with O2, this alkoxy radical forms 4,5-dioxohexanal(nominal mass 128). Figure 5.5 shows the PTR-MS signals for m/z 147, 129 and 111. m/z 147would correspond to the hydroperoxide, but no signal is seen. m/z 129 corresponds to ei-ther 4,5-dioxohexanal or the hydroperoxide fragment ([M + H−H2O]+) with a signal below1.5 ppbv. Because 4,5-dioxohexanal is an aldehyde, it can also dehydrate in the PTR-MS result-ing in m/z 111. Figure 5.5 shows that this signal is lower than the one at m/z 129 and remainsbelow 1 ppbv.

The previously described alkoxy radical (B) can also undergo carbon-carbon bond dissoci-ation. Due to the presence of the carbonyl group at position 5 (see Fig. 5.4), the energy barrierfor the dissociation of the 2–0 bond is reduced by 8 kcal mol−1 according to structure-activity

55


Figure 5.5.: Signals of the proton-transfer-reaction mass spectrometer (PTR-MS) for m/z 147(top), 129 (middle) and 111 (bottom). The hydroperoxides C6O4H10 are not detected at m/z 147due to a water molecule loss during protonation but rather at m/z 129, where the carbonylcompounds C6O3H8 are detected. Because these carbonyl compounds are aldehydes, they alsopossibly loose a water molecule during protonation and some signal appears at m/z 111.

56


Figure 5.6.: Signals of the proton-transfer-reaction mass spectrometer (PTR-MS) for m/z 87(left) and 73 (right). m/z 87 is the signal for succinic aldehyde and m/z 73 for methylglyoxal.

relationships (Peeters et al., 2004) compared to the 2–3b bond dissociation. For this reason, theelimination of the acetyl radical is favoured. This is confirmed by the PTR-MS with the signalsfor m/z 87 (corresponding to succinic aldehyde, C4H6O2) and m/z 73 (corresponding to methyl-glyoxal, C3H4O2) as shown in Fig. 5.6. The step increase at 0 h for m/z 87 is due to a fragmentfrom CMK (or an impurity). If the signal comes from an impurity and does not stay constantbut decreases during the experiment, it may be that the net signal increase from new prod-ucts formed is higher than one would expect from a simple comparison of the mixing ratio at0 h and at the end of the experiment. In any case, the signals increase at least twice as muchfor m/z 87 compared to m/z 73 in the O3 experiments. Because methylglyoxal is also formedin the dark OH experiment from the ozonolysis of TME, the interpretation of these results ismore difficult, but no change in the m/z 73 increase rate can be observed while the increase ofm/z 87 is small due to the small amount of CMK reacted. The most intriguing observation isthe rapid signal increase for m/z 87 in the NOx experiment while the lights are on. This may bedue to the fragmentation of organonitrates of higher mass. However, no structure and forma-tion mechanism can be proposed for this observation. It would also not make sense that thepresence of NOx simply favours one carbon-carbon bond dissociation of the alkoxy radical overthe other. However, the presence of NOx suppresses the reaction of “alkoxy B” with anotherRO

2 to form alcohol and carbonyl compounds (not shown), so that more “alkoxy B” is producedand, subsequently, more succinic aldehyde.

5.3.3. Organic acids formation mechanism

The reactions discussed so far do not lead to the formation of organic acids. However,organic acids are indeed formed as shown in Fig. 5.7. The mixing ratios of all organic acidspresented in Fig. 5.1 increase immediately after the start of the oxidation, indicating that theseorganic acids are first generation products (first non-radical species). This suggests that ter-penylic acid and diaterpenylic acid acetate could be products of the pinonic acid oxidation by

57


Figure 5.7.: Mixing ratios (left axis) and molar yields (right axis) of acids observed: (a) succinicacid, (b) 4-hydroxybutanoic acid, (c) 4-oxobutanoic acid, and (d) monocarboxylic acid with mo-lar mass 146.

OH. This also implies that it is possible to add two carboxylic acid functionalities to a com-pound within one generation (from CMK to succinic acid or from pinonic acid to MBTCA).This reaction may proceed via a sequence of several radical intermediates. Succinic acid, 4-oxobutanoic acid and 4-hydroxybutanoic acid were quantified based on calibration with stan-dards. The monocarboxylic acid with nominal mass 146 could correspond to 4-acetoxybutanoicacid, but this could not be confirmed due to the lack of an available standard. Therefore, thisspecies was quantified based on the calibration of 4-hydroxybutanoic acid. This unknown acidcould also contribute to the PTR-MS m/z 129 signal (Fig. 5.5).

Overall, in the dark OH and NOx experiments the mixing ratios are lower, but this is dueto the lower amount of precursor reacted. Succinic acid has a molar yield of a few percent,while the other organic acids presented in Fig. 5.7 have molar yields of less than 1 %.

The formation mechanisms of these organic acids remain unclear, but their low molaryields indicate that they may arise from minor processes. Hydrogen atom migration fromthe aldehyde moiety of the radical “alkoxy B” presented in Fig. 5.4 would lead to the forma-

58


Figure 5.8.: Chemical mechanism of the reaction between the peroxyacyl radical with the HO2

forming either a peracid or an organic acid and ozone (O3). In the presence of nitrogen oxides(NOx), the HO

2 concentration will be reduced by nitrogen monoxide (NO) and a peroxyacylnitrate (PAN) can be formed from the reaction of the peroxyacyl radical with nitrogen dioxide(NO2). The numbers in brackets correspond to the molar mass of the compounds.

tion of a peroxyacyl radical (R(O)OO), which by reaction with HO2 can form either a peracid

(RC(O)OOH) or a carboxylic acid (and O3) as depicted in Fig. 5.8. Note that 4-hydroxy-5-oxo-hexanoic acid has a nominal mass 146 and could be the one measured and attributed to 4-acetoxybutanoic acid, as no standard was available to confirm the retention time of eitherof these species. This product with nominal mass 146 is not formed in the presence of NOx,which would be compatible with the formation of peroxyacylnitrate C6H9NO7 as the hydroper-oxide formation channel is suppressed in the presence of NO2. Moreover, it may be that 4-acetoxybutanoic acid is hydrolysed during sampling (similarly to lactone) and is detected as4-hydroxybutanoic acid. The analogous compound of 4-hydroxy-5-oxohexanoic in the pinonicacid system would be an isomer of diaterpenylic acid acetate with the same chemical structure(C10H16O6) and nominal mass 232 (depicted in Fig. 5.1).

Muller et al. (2012) suggested several mechanisms for the formation of MBTCA from pinonicacid. One of them includes an intermediate stable compound (i.e. non-radical compound), sothat it cannot be used to interpret the present results. Because succinic acid is formed as a firstgeneration compound in the experiments presented here, a mechanism without any stable(non-radical) intermediate is required. Muller et al. (2012) also suggested pathways withoutstable intermediate products. However these mechanisms require an intramolecular hydrogenatom shift to an acyloxy radical (see Fig. 5.9). This kind of unimolecular reaction was alreadysuggested by Jenkin et al. (2000) for the formation of pinic acid. However, based on structure-activity relationships, Vereecken and Peeters (2010) reported acyloxy migration rates at 298 Kand 1 atm in the order of 9.3× 103 to 1.1× 105 s−1, while the derived carbon dioxide (CO2) lossrate for acyloxy radicals (e.g. ethylacyloxy radical) based on structure-activity relationships andquantum chemical calculations is in the range of 6.7–9.3× 1010 s−1 (Vereecken and Peeters,2009, 2010). It is difficult to find a reason for such an enormous stabilisation of the acyloxyradical. If the migration happens, it is expected to occur only to a very minor extent.

59


Figure 5.9.: Suggested formation mechanism of succinic acid adapted from the 3-methyl-1,2,3-butanetricarboxylic acid (MBTCA) formation mechanism of Muller et al. (2012) (black path). Theintermediate alkyl radical marked in red is expected to react with O2 preferentially by hydro-gen atom abstraction to form 4,5-dioxohexanoic acid. For simplification, alkyl radicals are notdepicted as they react immediately with O2 to form peroxy radicals. The numbers in bracketscorrespond to the molar mass of the compounds.

Furthermore, the alkyl radical marked in red in Fig. 5.9 is expected to react preferentiallywith O2 by hydrogen atom abstraction and formation of a carbonyl functional group, leadingto 4,5-dioxohexanoic acid (C6O4H8) with a nominal mass of 144. It was not possible to iden-tify any first-generation monocarboxylic acid with ion chromatography (IC)/MS at m/z 143 (seethe chromatogram of Fig. 5.10). The peak at retention time (RT)∼19 min is present in the back-ground and does not increase during photooxidation of CMK. Only in the experiment withhigh CMK (Fig. 5.10b) another peak appears around 16 min. However, because this RT is higherthan 15 min, it is expected to be a dicarboxylic acid and from its time trend (data not shown), itseems to be a second generation product.

Similar to this minor dicarboxylic acid peak at m/z 143, other peaks were identified (e.g.at m/z 129 with RT∼ 13 min) which were not included in the present discussion. Nevertheless,they illustrate the mechanistic complexity of the CMK oxidation and the need to better under-stand organic acid formation in the gas phase. Also small organic acids such as formic acid,acetic acid, lactic acid and pyruvic acid were observed. Because these can arise from severalreaction pathways, they were not used to interpret the mechanism of the present discussion.

60


Figure 5.10.: Chromatograms of m/z 143 for the four different experiments: (a) dark OH pro-duction (b) ozone (O3) photolysis with 1600 ppbv cyclobutyl methyl ketone (CMK), (c) O3 pho-tolysis, and (d) O3 photolysis in the presence of nitrogen oxides (NOx). The grey chromatogramin each plot corresponds to the gas phase background measured before the lights were turnedon.

61


5.3.4. Non-traditional chemistry

With the so-called traditional atmospheric chemistry, gas-phase formation of the observedacids (and lactones) as first generation products cannot be explained. With larger molecules,many unimolecular reactions are however possible, including peroxy radical isomerisation (H-shift), which were not discussed previously. Such shifts have been published elsewhere (Zhuet al., 2007; Neuenschwander and Hermans, 2010).

As already mentioned, a key compound seems to be the peroxy radical marked in orangein Fig. 5.4 because after elimination of the acyl moiety, it remains with four carbon atoms thatare common to the organic acids presented in Fig. 5.7. Unknown reaction pathways are there-fore symbolised by an orange question mark. Quantum chemistry calculations estimate thatthe rate constant for that 1,6-H shift at 298 K in this peroxy radical (Figs. 5.4 and 5.9) is about100 s−1 (Neuenschwander, 2012), which is orders of magnitude faster than the intermolecularreactions considered. Further exploration of this possible mechanism would be highly specu-lative and beyond the scope of the present work.

5.4. Conclusions

Using CMK as a model compound, it is possible to study the gas phase reactions follow-ing the opening of the four-carbon ring present in the chemical structures of many terpenoids.Muller et al. (2012) already demonstrated that MBTCA can be formed from the OH oxidationof pinonic acid. In the CMK system, the analogue of MBTCA, succinic acid, was measuredwith molar yields of 2 to 5 %. 4-Hydroxybutanoic acid could also be identified as analogueof diaterpenylic acid, resulting from the hydrolysis of butyrolactone (analogue of terpenylicacid). A monocarboxylic acid with nominal mass 146 which could be the analogue of diater-penylic acid acetate was measured in the absence of NOx. However, due to the lack of anavailable standard, this analogy could not be confirmed. Moreover, it remains unclear if 4-acetoxybutanoic acid would hydrolyse similarly to butyrolactone during sampling with waterand be detected as 4-hydroxybutanoic acid. Its formation suppression by NOx could not beinterpreted mechanistically. The signal at m/z 146 could also be attributed to 4-hydroxy-5-oxohexanoic acid.

4-Oxobutanoic acid was identified and two analogue structures are proposed that wouldcorrespond to 2-(2-methyl-1-oxopropan-2-yl)succinic acid and 3-formyl-2,2-dimethylpentane-dioic acid. Claeys et al. (2009) identified in α-pinene SOA a compound with the same nominalmass (188), but tentatively attributed a slightly different structure to it. By understanding thereaction mechanisms, structures can be proposed in the context of a precursor oxidation, sothat out of the many possible compounds derived from a chemical formula, the ones that arerelevant for the system observed can be selected. For this, the simplified CMK system is a usefultool, because it focuses on the reactive substructure of the terpenes.

All the compounds measured in this work are first-generation oxidation products. Thisdemonstrates that two carboxylic acid functionalities can be added in one oxidation step, in-creasing rapidly the O : C ratio and the potential of SOA formation from VOC oxidation. Veryoften, offline filter analyses do not provide information on the temporal evolution of a com-

62

5.4. Conclusions

pound, so that mechanisms involving many steps are suggested, which can be correct for latergenerations products. In the present case, a mechanism is required to explain the immedi-ate formation of succinic acid from CMK without the formation of any stable intermediatecompound. Muller et al. (2012) proposed such a mechanistic scheme, starting from the mostreactive H abstraction (position 3 in Fig. 5.1) by OH. This scheme includes a 1,5-H shift to anacyloxy radical, which is expected to be negligible compared to the loss of CO2. Moreover, theradical resulting from this shift (in red in Fig. 5.9) should react with O2 to form in the CMK sys-tem either a monocarboxylic acid with nominal mass 144 (H abstraction, red path) or a peroxylradical towards the formation of succinic acid (black path). No acid with nominal mass 144could be identified, while succinic acid is measured; however, there is no reason why the latterreaction should be favoured over the first one.

Because of the complexity and the various branching possibilities of the gas phase ox-idation mechanism, the products detected show molar yields below 1 %, except for succinicacid. This allows the possibility of formation mechanisms through minor pathways that werenot discussed here (e.g. from the H-abstraction by OH at other positions than 3 or from non-traditional chemistry). To maintain the mass balance, dozens of oxygenated compounds shouldbe detected at very low concentrations. Traditional gas phase chemistry cannot explain theformation of the observed products, so that new reaction mechanisms are required to under-stand the atmospheric oxidation of VOCs with a large number of carbon atoms. Unimolecularreactions (H-shifts and ring closure) seem to be important in this regard.

Acknowledgements This work was supported by the Swiss National Science Foundation.

Edited by: F. Keutsch

63

6Online measurements of water-soluble organic

acids in the gas and aerosol phase from thephoto-oxidation of 1,3,5-trimethylbenzene

Arnaud P. Praplan, Kathrin Hegyi-Gaggeler, Peter Barmet, Josef Dommen, Urs BaltenspergerLaboratory of Atmospheric Chemistry, Paul Scherrer Institut, 5232 Villigen PSI, Switzerland

In preparation for Atmospheric Chemistry and Physics Discussions

Abstract. The formation of organic acids during photo-oxidation of 1,3,5-trimethylbenzene(TMB) was investigated with an online ion chromatography (IC) instrument coupled to a massspectrometer (MS) at the Paul Scherrer Institut (PSI) smog chamber. Both gas and aerosolphases were sampled. A molecular formula could be attributed to twelve compounds with thehelp of high resolution MS data from filter extracts and seven of those species could be iden-tified unambiguously: formic acid, acetic acid, glycolic acid, butyric acid, pyruvic acid, lacticacid and methylmaleic acid. The influence of the precursor concentration (1200 and 600 ppbv)and of the presence of 2 ppbv of sulphur dioxide (SO2) were further investigated. While theorganic acid fraction present in the aerosol phase does not strongly depends on the precursorconcentration (6 to 14 %), the presence of SO2 reduces this amount to less than 3 % for bothhigh and low precursor concentration scenarios. The partitioning of small acids has furtherbeen investigated by addition of acetic acid.

65

Chapter 6. Online measurement of organic acids from the photo-oxidation of TMB

6.1. Introduction

Aromatic compounds are volatile organic compounds (VOCs) emitted into the atmosphereby fuel combustion and evaporation, where they are oxidised by species such as hydroxyl radi-cal (OH) or nitrate radical (NO

3). Thus, they play a major role in urban areas and can represent16 – 44 % of the total hydrocarbon mass emitted into the atmosphere (Calvert, 2002; Dommenet al., 2003; Molina et al., 2007). As a result, oxygenated products with lower volatility areformed, which contribute to the formation of secondary organic aerosol (SOA).

The organic fraction of the atmospheric aerosol, to which SOA contributes, is a complexmixture of many different compounds. Decesari et al. (2000) reported that 20 – 70 % of thesecompounds are water soluble organic compounds (WSOC). However, only 10 % of this organicfraction of the aerosol can typically be chemically identified.

Organic acids represent an important class of atmospheric compounds (Chebbi and Car-lier, 1996) and a large fraction of WSOC. Even though they can be directly emitted into theatmosphere by traffic (Kawamura et al., 2000), biogenic emissions (Servant et al., 1991) andbiomass burning (Gaeggeler et al., 2008), they also can be produced by ozonolysis of alkenesthrough electrophilic addition and formation of Criegee intermediates and their subsequentstabilization by water (H2O), by reaction of peroxyl radical (RO

2) with hydroperoxyl radical(HO

2) (Madronich et al., 1990), by reaction of RO2 with an acyloxy radical as well as by ester

rearrangement. Aqueous oxidation in cloud droplets is an important source of organic acids inthe atmosphere (Altieri et al., 2008; Lim et al., 2005).

Measurement of organic acids is a challenging task. Offline methods are prone to arte-facts and are lab intensive, while online methods providing high time resolution and low de-tection limits are scarce. Although proton-transfer-reaction mass spectrometer (PTR-MS) candetect acids, several compounds and/or fragments can have the same m/z , which makes dataanalysis and interpretation difficult. The ion chromatography (IC) method presented here al-lows selective collection of organic acids and their separation prior to detection. By using amass spectrometer (MS) as an additional detector, further separation based on molar masscan be performed on coeluting peaks.

Fisseha et al. (2004) identified 20 different acids formed during 1,3,5-trimethylbenzene(TMB) photo-oxidation experiments at the Paul Scherrer Institut (PSI) smog chamber. The maingoal of this work is to investigate further organic acids formation and evolution over time dur-ing TMB photo-oxidation under different experimental conditions.

6.2. Experimental

6.2.1. PSI smog chamber

A detailed description of the smog chamber at PSI can be found elsewhere (Paulsen et al.,2005). Experiments are carried out in a 27-m3 transparent fluoroethylene propylene (FEP) bagplaced in a temperature-controlled housing (∼20 ° C). The chamber is first humidified to arelative humidity (RH) of approximately 50 % before injecting nitrogen oxides (NOx). A known

66

6.2. Experimental

amount of liquid TMB is evaporated in a heated glass sampling bulb (80 ° C) and flushed withpure air into the chamber approximately 30 minutes before the lamps are switched on, to al-low homogeneous mixing. To simulate the solar spectrum, four 4 kW xenon-arc lamps areused. Since 2010, supplementary black lights (eighty 100 W tubes, Cleo Performance) can beused to increase the ultra-violet (UV) light intensity. Light is reflected by aluminium plates cov-ering the housing walls of the chamber. For the experiments described here (Table 6.1), theratio VOC:NOx was set to 2. For some experiments, approximately 2 part per billion by vol-ume (ppbv) of sulphur dioxide (SO2) were also injected into the chamber in order to enhancethe nucleation rate, to increase the SOA particle number concentration and to reduce vapourwall losses.

Table 6.1.: Description of the performed 1,3,5-trimethylbenzene (TMB) photo-oxidationexperiments.

Date Nominal concentrations [ppbv] RH CommentsTMB nitrogen monoxide (NO) nitrogen dioxide (NO2) SO2

15 Nov 2006 600 150 150 2 ∼ 50%10 Jul 2009 1200 300 300 - ∼ 50% injection of acetic acid (500 µl) after ap-

proximately 6 hours of photo-oxidation27 Jul 2009 600 150 150 - ∼ 50% injection of acetic and formic acids

(0.4 µl each) after approximately 5.3hours of photo-oxidation

10 Dec 2010 1200 300 300 - ∼ 50%17 Dec 2010 0 0 0 - ∼ 50% “blank” experiment (with black lights)11 Apr 2011 1200 300 300 2 ∼ 50%13 Apr 2011 600 150 150 2 ∼ 50% with black lights

The standard instrumentation of the smog chamber consists of ozone (O3) and NOx mon-itors, a condensation particle counter (CPC) with a cut-off of 3 nm and a scanning mobilityparticle sizer (SMPS) for particles with mobility diameter between 14 and 698 nm. Addition-ally, a PTR-MS (Ionicon Analytical GmbH, Austria) for the analysis of gas phase compounds andIC coupled to a MS for the selective analysis of organic acids (Sect. 6.2.2) were used.

6.2.2. Ion chromatography system

The sampling of the gas and particle phase for the IC system was done with a wet effluentdiffusion denuder (WEDD) and an aerosol collector (AC), respectively (Takeuchi et al., 2004,2005). The WEDD part is connected through a Teflon® inlet to the smog chamber. It consistsof two parallel cellulose acetate membranes between which continuous sampling from thesmog chamber gas phase occurs. Purified water flows on the other side of the membranes,allowing gases to dissolve. The AC part is connected by a silcosteel® inlet to the smog chamber.An activated charcoal denuder and a WEDD are placed upward the sampling line, to removethe gas phase compounds. Figure 6.1 shows that this is sufficient to avoid breakthrough of thegas phase to the aerosol collector, even with high precursor concentrations. The AC consistsof a small chamber, the bottom of which is covered by a filter paper placed on a hydrophilicfilter. Water is introduced into this chamber through a capillary having its tip close to thenozzle where the airflow enters. A fine mist is formed, which condenses on the filter wherethe soluble compounds of the particles are dissolved.

67


Figure 6.1.: No breakthrough of the gas phase to the aerosol collector (AC) was observed when afilter was placed in the aerosol sampling line during a 1200 ppbv 1,3,5-trimethylbenzene (TMB)photo-oxidation experiment.

Effluents from both sampling devices pass through concentration columns (TAC-LP1, Dio-nex) where the (organic and inorganic) anions are retained. Those samples are then elutedalternately to the guard and analytical columns (NG1 and AS11-HC, Dionex) for separation. Theeluent consists of a hydroxy anion (OH−) gradient: 0 min 0.95 mM OH−, 3 min 0.95 mM OH−,18 min 12 mM OH−, 22 min 60 mM OH−, 24 min 60 mM OH−, 24.1 min 0.95 mM OH− and 29min 0.95 mM OH−. After elution, the OH− are suppressed by an anion self-regenerating sup-pressor (ASRS 300 2mm, Dionex). The analytes are detected as deprotonated species by aconductivity detector (CD) first and, after negative electrospray ionisation (ESI), by a quadru-pole MS (MSQ, ThermoFinnigan) with unity mass resolution.

The retention time (RT) of the analytes depends on the strength of the ion exchange withthe analytical column. Deprotonated organic acids with only one carboxylic functional groupwill elute earlier, followed by acids with two and three carboxylic functional groups at laterRTs.

Calibration is performed by direct injection of aqueous multi-compound standards of dif-ferent concentrations onto the analytical column. Non-linear least squares regression of apower function (y = axb + c) is used as calibration curve for MS, in order to take into accountthe slight curvature of some calibration curves (Fig. 6.2), as suggested by Kirkup and Mulhol-land (2004). This is due to the formation of dimers and aggregates with water during ESI,especially at higher concentrations, and possibly varying fragmentation (Grossert et al., 2005)at different concentration levels. Due to the presence of numerous unknown compounds, it isnot possible to tune the instrument for each individual substance to prevent the formation ofdimers, for example.

If a molecular formula could be assigned to an unknown compound, a surrogate com-pound with a similar degree of unsaturation (expected similar functionalisation, see Sect. 6.2.3)and a similar mass was used for calibration.

68

6.2. Experimental

Figure 6.2.: Example of IC/MS calibration (methylmaleic acid).

6.2.3. High resolution MS

The high resolution Orbitrap-MS instrument of the Functional Genomics Center Zurichis described in details elsewhere (Olsen et al., 2007). The high mass accuracy (2 ppm) allowsdeduction of the accurate elemental composition of the detected organic acids. From the ob-tained chemical formula, it is then possible to compute their degree of unsaturation (DU), alsoknown as “double bond equivalent”, using Eq. (6.1), where C, H and N represent the number ofcarbon, hydrogen and nitrogen atoms, respectively.

DU = (C + 1)− (H− N)

2(6.1)

It provides the number of double bonds or rings of a compound and can help for the struc-ture elucidation of an unknown compound: For example, monocarboxylic acids with DU>1contain a ring structure or a carbon-carbon double bond if O=2 (O represents the number ofoxygen atoms) or may contain one or more ketone or aldehyde functionalities if O>2. Thedeductions are similar for the dicarboxylic acids with DU>2. Furthermore, if O>2 for mono-carboxylic acids with DU=1 respectively if O>4 for dicarboxylic acids with DU=2, the oxygenatoms that are not in the carboxylic functional groups are most likely present as alcohol func-tional groups. In this study, N=0 is assumed as no even m/z was detected and it is unlikelythat species with two nitrogen atoms are detected but none with only one N-atom.

For these measurements, SOA particles were collected on filters and extracted with water.The solutions were finally directly injected into the instrument, whose measurements rangedfrom m/z 50 to 700.

6.2.4. Partitioning theory

One important feature of the WEDD/AC-IC/MS setup presented is that it is possible tomeasure almost simultaneously gas phase and aerosol concentrations of the organic acids.

69


Therefore, partitioning coefficients Kp [m3µg−1] can be determined experimentally by Eq. (6.2)and estimated theoretically by Eq. (6.3), using partitionning theory (Pankow, 1994b,a).

Kp,i =Fi

fm · TSP · Ai(6.2)

Kp,i =fm · 760 · R · T

MWm · p0L · ξ · 106(6.3)

TSP [µg m−3] is the total suspended particulates mass and fm represents its absorbingmass fraction (assumed to be unity for SOA). Fi and Ai [µg m−3] are the measured aerosol andgas phase concentrations of species i , respectively. R (8.2× 10−5 m3 atm mol−1 K−1) is the idealgas constant, T [K] is the temperature, MWm is the mean molecular weight of the absorbingmaterial and is assumed here to be 130 g mol−1, ξ is the activity coefficient of the species in thecondensed phase (assumed to be unity) and p0L [Torr] is the saturation vapor pressure. A the-oretical range for p0L can be determined with structure-based estimations 1 (Stein and Brown,1994; Myrdal and Yalkowsky, 1997; Nannoolal et al., 2004; Moller et al., 2008; Nannoolal et al.,2008; Compernolle et al., 2011).

6.2.5. Chemicals

Water used for this work was delivered by a Milli-Q water purification system and had aresistivity of 18 MΩ cm. TMB was supplied by Fluka (99.5 %). Gas cylinders were provided byCarbagas: NO 2.8 in N2 5.0, NO2 1.8 in synthetic air 5.0 and SO2 3.8 in N2 5.0.


6.3.1. SOA formation

The lowest panel of Fig. 6.3 presents the aerosol mass concentration measured by theSMPS assuming a particle density ρp of 1.4 g cm−3 (Alfarra et al., 2006). The SOA particlesreached mobility diameters up to ∼ 700 nm during the photo-oxidation of 1200 ppbv TMB inthe absence of SO2. This is outside of the SMPS measurement range for singly charged parti-cles. Therefore, the particle number size distribution was recovered from the raw signal of thedoubly charged particles using a custom-made data inversion routine, which works in a smi-lar way as the standard SMPS data inversion (Hagen and Alofs, 1983). This approach was onlypossible because the size distribution was very narrow such that the mode of singly chargedparticles was clearly separated from the multiply charged particles in the raw data and that alldoubly charged particles fell into the electrical mobility diameter range of the SMPS.

In the absence of SO2, more SOA is produced at higher precursor concentration but itsformation is slower and the particles formed are bigger than with a lower precursor concen-tration (Fig. 6.3, upper panel). From the CPC data (not shown), particles larger than 3 nm

1. http://www.aim.env.uea.ac.uk/aim/ddbst/pcalc main.php

70

http://www.aim.env.uea.ac.uk/aim/ddbst/pcalc_main.php


can be observed after approximately 3.2 hours after lights on during the experiments with1200 ppbv TMB while in the case with 600 ppbv nucleation occurs already after 2 hours ofphoto-oxidation. Nucleation only occurs when the NO levels become low which is later in thecase of the high concentration experiment (Wyche et al., 2009).

The presence of SO2 increases the nucleation rate as described in Metzger et al. (2010):it is possible to observe particles in the CPC (>3 nm) already after 1 hour of photo-oxidation(1200 ppbv TMB). The slightly higher final aerosol mass concentration produced in the pres-ence of SO2 is probably explained by the earlier particle formation, resulting in lower vapourwall losses of low-volatility oxidation products.

Black lights increase the nucleation rate the most, due to the higher OH exposure andfaster oxidation of the gaseous compounds. 12 minutes of photo-oxidation are sufficient todetect particles with the CPC in the presence of SO2. The maximum aerosol mass producedincreases by roughly a factor of 2 because of the higher amount of TMB reacted, but also dueto dramatical reduction of the vapour wall losses in the early nucleation stage.

6.3.2. Identified organic acids

Table 6.2 gives an overview of the organic acids detected during TMB photo-oxidation.Twenty-five masses showed at least one chromatographic peak and a chemical formula couldbe attributed to twelve species with the help of the high resolution MS data. Chemical formu-las containing only one oxygen atom were not considered, as carboxylic acids contain at leasttwo oxygen atoms. A name is also mentioned for seven compounds that could be identified:formic acid, acetic acid, glycolic acid, butanoic acid, pyruvic acid, lactic acid and methylmaleicacid. Fisseha et al. (2004) did not identified glycolic acid and butanoic acid, but reported detec-tion of oxalic acid, malonic acid, succinic acid, malic acid and citric acid, that were not detectedin the present work. Except for the unknown compounds with nominal mass 178, 190 and 234,the unidentified species are different for these two studies.

Sato et al. (2012) could also identify pyruvic acid with liquid chromatography coupled to atime-of-flight MS, as well as other organic acids. They found several peaks for each (deproto-nated) masses 127 (C6H7O−3 ), 161 (C6H9O−5 ), 215 (C9H11O−6 ) and 233 (C9H13O−7 ) as well as singlepeaks of the (deprotonated) masses 189 (C6H7O−3 ) and 217 (C9H13O−6 ), similar to this study.While this study report several possible formulae for these compounds, as it was not possi-ble to unambiguously identify which peaks in the high resolution spectra corresponded to theunity mass observed with IC/MS, Sato et al. (2012) report unique chemical composition foreach mass. They also report other masses, which were not found in this work, either becausethey are not organic acids (even though a structure with a carboxylic acid functionality wasproposed), or because the sensitivity of our method was not sufficient to detect them.

The Master Chemical Mechanism (MCM, Jenkin et al. (2003); Saunders et al. (2003)) con-tains a degradation scheme for TMB and Table 6.3 presents the expected organic acids. Onlyacetic acid and pyruvic acid could be identified in our experiments. Also acids with nominalmass 128, 144 and 160 were detected, but it was not possible to assign the species proposed byMCM to IC peaks, mainly because more than one peak were found for each of these masses.

71


Figure 6.3.: Mobility mean diameter, total number concentration and aerosol mass concentra-tion from the scanning mobility particle sizer (SMPS) measurements. The aerosol mass con-centration is derived from volume data, assuming ρ=1.4 g cm−3. These data are not correctedfor wall losses.

Without going into the detail of each detected species, it is interesting to note that thereis a large variety of compounds with a molar mass higher than TMB itself (functionalisation,oligomerisation) while the compounds with smaller molar mass (fragmentation) are less innumber but seem to contribute more to the aerosol mass. However, a quantitative comparisonbetween both pathways is difficult to address, as the contribution of many compounds cannotbe quantified properly, due to a lack of standards for calibration.

From the compounds to which a chemical formula could be attributed, two have a DU

72


equal to 4 (DU of TMB), while ten have a lower DU. This indicates that fragmentation and dou-ble bond opening reactions happen on top of the functionalisation with one or two carboxylicgroups (1 DU for each group).

Figures 6.4 and 6.5 show the concentration profiles in the gas and the aerosol phase, re-spectively, for the main measured species: formic acid, acetic acid, glycolic acid, butanoic acid,pyruvic acid, lactic acid, methylmaleic acid and an unknown dicarboxylic acid with molar mass234 g mol−1 (M234). Some of these species were close to the detection limit of the method andtherefore gaps can be observed in the plots. Methylmaleic acid is a dicarboxylic acid and isnot expected with such high concentrations in the gas phase (without being detected accord-ingly in the aerosol phase at the same time). Most probably it is formed from the expectedmethylmaleic anhydride in the sampling device. Note that other detected species could alsobe formed by similar artefacts from other compounds.

The production of glycolic acid in both gas and aerosol phase remains small and does notseem to depend on the precursor concentration. Butyric acid, pyruvic acid and methylmaleicacid gas phase concentrations show a precursor concentration dependence, confirming thatthey are products of TMB oxidation. For lactic acid, a marked difference is seen comparing the1200 ppbv TMB experiments. The presence of SO2 decreases its concentration by roughly a fac-tor 3. Studies (Hatakeyama and Akimoto, 1994; Kurten et al., 2011a; Boy et al., 2012; Mauldin IIIet al., 2012) suggest that SO2 reacts with Criegee radicals produding OH and influencing thegas phase chemistry. The differences in the experiments with and without SO2 could be ex-plained by reaction of the precursor Criegee radical with that species, hindering the formationof lactic acid.

In the aerosol phase the measured concentration are mostly lower than 1 µ g m−3, exceptfor M234. The presence of SO2 seems to decrease the aerosol concentration for this species,as well as for lactic acid and butyric acid. Due to the different aerosol mass concentrationsproduced, acid fractions were computed by dividing the particle bound acid concentration withthe aerosol mass concentration measured by SMPS. Figure 6.6 (c) and (d) show that the acidfraction of the main acids drops below 3 % after 5 hours of photo-oxidation if SO2 is present,while independently of the precursor concentration, its lower limit lies between 6 and 14 % inthe absence of SO2 (Fig. 6.6 (a) and (c)).

The compound M234 merits particular attention. It is the most abundant organic acidof the aerosol phase. It could be detected as well in small amounts in the gas phase and ispotentially important for the SOA formation. Sato et al. (2012) report three compounds withthis mass and the chemical formula (C9H14O7), while we could identify only one peak withthis mass in the range of dicarboxylic acids. Very likely, the two other compounds are hydrox-ycarbonyl compounds without carboxylic acid functional group, so that the isomer structureproposed by Sato et al. (2012) does not correspond to any of the compounds with nominal mass234 (one dicarboxylic acid and two hydroxycarbonyl compounds).

Also, the unknown compound at m/z 86 has a longer RT than those of methacrylic acid,3-butenoic acid and trans-2-butenoic acid. cis-2-Butenoic acid was not tested but has an ex-pected RT similar to that of the trans isomer. Therefore, this compound may not be a mono-carboxylic acid and cannot be a dicarboxylic acid, because the smallest carboxylic acid (oxalicacid) would appear later in the chromatogram and has a nominal mass of 89. No other higher

73


mass was identified at the same RT, so that it is unlikely a fragment of a carboxylic acid.

Due to the remaining uncertainties, more work is needed to identify the chemical struc-ture of the numerous compounds that are formed during photo-oxidation of TMB. A combina-tion of different analytical techniques would be necessary.

Figure 6.4.: Gas phase concentration profiles (in µg m−3) of eight measured organic acids.

6.3.3. Partitioning coefficients, Kp

Figure 6.7 depicts the time-dependent Kp values for the previously discussed species. Dueto the alternating gas phase and aerosol measurements, linear interpolation or extrapolationwas performed to estimate the corresponding aerosol or gas phase concentration for a giventime.

In most cases, the experimental Kp values for the different conditions are stable duringthe experiments with no clear influence from the presence of SO2. However, the values be-tween the experiments can vary over a few orders of magnitude. This uncertainty is also re-

74


Figure 6.5.: Aerosol concentration profiles (in µg m−3) of eight measured organic acids.

flected in the expected theoretical Kp range, though.

Only for glycolic acid and methylmaleic acid to some extend, a decreasing trend is ob-served in the earlier stages of the experiment. This is very likely due to measurements close tothe detection limits in the aerosol phase for these species, so that the decrease manly reflectsthe TSP increase in this case.

Interestingly, Kp values of methylmaleic acid fall in the range of the theoretical values formethylmaleic anhydride (light grey shaded area). This confirms that no methylmaleic acid (ora negligible amount) is formed in the chamber. It would also indicate that partitioning is notfurther driven by hydrolysis of the anhydride in the aerosol.

Overall, the experimental values obtained are higher than the theoretically estimated val-ues. Healy et al. (2008) found similar results for a wide range of dicarbonyl products, includingglyoxal and methylglyoxal. The interpretation given was that reactive uptake of acids (or theirprecursors) as well as oligomerisation reaction in the particle phase take place, shifting thepartitioning to the aerosol. Then, during sampling in water, the oligomers can reverse to their

75


Figure 6.6.: Stacked normalized aerosol concentrations of the main detected organic acids forthe different experimental conditions.

monomeric form, making the aerosol concentration higher than expected. However, this can-not apply to all species. First, the gas phase concentration already (e. g. fro lactic acid and M234)varies between the different experiments and second because the normalised concentrations(Fig. 6.6) are found to be smaller in the experiments with SO2, which contradicts a possiblereversion of acid oligomers to the monomeric form.

During the experiment of 10 July 2009, a very large amount of acetic acid (∼ 18 mg m−3)was injected into the chamber after roughly 6 hours of photo-oxidation. Fig. 6.8 shows thatthis results in almost no increase of the aerosol phase acetic acid concentration. This is a strongindication that the acetic acid (and probably the other volatile organic acids) detected in theparticle phase are not due to (enhanced) partitioning. We assume that their occurence is anartifact originating most likely from the hydrolysis of ester functionalities during sampling.

76


Figure 6.7.: Time-dependent partitioning coefficient, Kp (in m3 µg−1). Grey areas represent thetheoretical range values. In panel (e), the light grey area represent the theoretical range valueof methylmaleic anhydride.

77


Table 6.2.: Organic acids detected with WEDD/AC-IC/MS during 1,3,5-trimethylbenzene (TMB)photo-oxidation.

Trivial name (IUPAC name) Chemical DU Molar mass Type RT [min] m/zformula [g mol−1] WEDD AC

Formic acid (Methanoicacid)

CH2O2 1 46.03 Mono 11.4 – 13.3 11.6 – 13.4 45

Acetic acid (Ethanoic acid) C2H4O2 1 60.05 Mono 10.5 – 12.5 9.5 – 12.7 59, 41Glycolic acid (Hydroxy-ethanoic acid)

C2H4O3 1 76.05 Mono 10.7 – 12.6 10.5 – 12.8 75

Unknown 86 ? ? ? ? 15.1 – 16.4 n. d.b 85Butyric acid (Butanoic acid) C4H8O2 1 88.11 Mono 11.3 – 13.1 11.6 – 12.6 87Pyruvic acid (2-Oxo-propanoic acid)

C3H4O3 2 88.06 Mono 11.9 – 13.9 12.5 – 14.1 87, 105

Lactic acid and hydracrylicacid? (Hydroxypropanoicacid)

C3H6O3 1 90.08 Mono 10.0 – 12.1 10.9 – 12.4 89, 43

Unknown 116 C4H4O4 3 116.07 Di ? 16.5 – 18.5 16.5 – 18.4 115Unknowns 128 several compounds with formula C5H4O4, C6H8O3 or C7H11O2 127, 83, 101Methylmaleic/Methyl-fumaric? acid ((Z/E?)-2-Methylbutenedioic acid)

C5H6O4 3 130.10 Di 23.8 – 25.1 23.4 – 25.2 129, 85, 259

Unknown 144 several compounds with formula C6H8O4 143, 99Unknowns 148 several compounds with formula C5H8O5 or C4H3O6 147Unknown 156a C7H8O4 4 156.14 Di 17.6 – 18.9 n. d. 155, 111, 83Unknown 156b C7H8O4 4 156.14 Di 18.4 – 19.7 n. d. 155, 137Unknown 160 C6H8O5 or

C7H12O4

? ? Mono 10.6 – 12.8 11.1 – 12.6 159

Unknown 162a C6H10O5 orC5H6O6

? ? Mono 11.9 – 12.2 11.0 – 12.4 161, 143

Unknown 162b C6H10O5 orC5H6O6

? ? Di 21.0 – 22.7 n. d. 161

Unknown 164 C5H8O6 2 164.11 Di n. d. 22.5 – 24.2 163Unknowns 178 several compounds with forumla C6H10O6 177Unknown 190 C7H9O6 or

C8H14O5

? ? Di n. d. 20.2 – 21.6 189, 171, 145

Unknown 208 several compounds with formula C7H12O7 207, 189Unknowns 216 several compounds with formula C8H8O7 or C9H12O6 or C10H15O5 215, 153, 113Unknowns 218 several compounds with formula C8H10O7 or C9H14O6 217, 155Unknown 234c C9H14O7 or

C8H10O8

? ? Di 18.6 – 19.1 18.7 – 20.1 233, 117, 115, 73

Unknown 250 C9H14O8 3 250.20 Di n. d. 20.5 – 21.6 249, 231abutanoic acid as surrogate bnot detected cmethylmaleic acid as surrogate

78


Table 6.3.: Expected organic acids and related species from the Master Chemical Mechanism(MCM) model for the oxidation of 1,3,5-trimethylbenzene (TMB).

MCM name IUPAC name (Trivial name) Formula Molar mass

CH3CO2Hb Ethanoic acid (Acetic acid) C2H4O2 60.05HCOCO2H Oxoethanoic acid (Glyoxylic acid) C2H2O3 74.04

CH3COCO2H Oxopropanoic acid (Pyruvic acid) C3H4O3 88.06C5CODBCO2H 2-Methyl-4-oxo-pent-2-enoic acid C6H8O3 128.13EPXMKTCO2H 3-Acetyl-2-methyloxirane-2-carboxylic acid C6H8O4 144.13C23O3MCO2H 2-((2-Oxopropanoyl)oxy)propanoic acid C6H8O5 160.12

TMBCO2H 3,5-Dimethylbenzoic acid C9H10O2 150.17TM135MUO2H (Z)-2-Methyl-3-(4-oxopent-2-en-2-yl)oxirane-2-carboxylic acid C9H12O4 184.19MMALANHYc 3-Methylfuran-2,5-dione (Methylmaleic anhydride) C5H4O3 112.08

Figure 6.8.: No partitioning of acetic acid onto the aerosol is observed after injection of a largeamount of acetic acid.

79


6.4. Summary and conclusions

Despite the complexity of the chemical composition of photo-oxidation products fromTMB, we were able to selectively separate and detect several organic acids from both gas andaerosol phases with IC/MS from which seven could be identified unambiguously. A chemicalformula could be attributed to five more compounds.

Some of the detected compounds could also be produced during sampling, like methyl-maleic acid, by anhydride hydrolysis or other aqueous reactions, for example. However, manyof those compounds are present in low concentration and by focussing on the main detectedcompounds a general picture of the fate of organic acids in TMB photo-oxidation experimentscould be drawn. Overall, their fraction represent less than 3 % of SOA after 5 hours of photo-oxidation in the presence of SO2, while in it absence, they represent between 6 and 14 % of theSOA mass, independently on the precursor concentration.

Higher partitioning coefficients than predicted by theory were found without clear influ-ence from SO2. This effect was already observed for other dicarbonyl compounds and wasexplained by reactive uptake and oligomers reversing to the monomeric form in solution dur-ing sampling. However, this interpretation cannot apply to all the species discussed here, dueto differences comparing the effect of SO2 for some species in both gas phase and aerosolconcentrations.

Acknowledgements The Functional Genomics Center Zurich and Dr. Bertran Gerrits are thankedfor the opportunity to perform measurements at the Orbitrap XL, Rene Richter for hishelp in building the WEDD/AC, Dr. Martin Gysel for his support on the SMPS data analy-sis and Prof. Dr. Alexander Wokaun for helpful discussions. This work was supported bythe Swiss National Science Foundation as well as the EC FP6 projects EUROCHAMP (No.505968) and POLYSOA (No. 012719).

80

7Conclusions and outlook

A high level of complexity is found in the tropospheric chemical composition: 0.1 % of theatmosphere consists of a large number (thousands if not millions) of trace compounds thatcan be reactive towards ozone (O3), hydroxyl radical (OH) and other oxidants. Environmen-tal chamber studies are used to reduce this complexity to observe only a few processes. Thecontrolled condition of temperature, relative humidity and light intensity, for instance, alsocontributes to the simplification of the system studied.

Nevertheless, the analysis of gas and particle phase in environmental chambers remainsa challenging task due to the complex mixture of products from the oxidation of a precur-sor compound. Some of the products, but also contaminants, are present at extremely lowconcentration and cannot always be detected by standard instrumentation. However, theycan influence the results of the experiments performed. Therefore, instrumentation with highsensitivity is required in order to characterise those trace compounds and allow to performchamber experiments at atmospheric relevant concentrations.

In this work, ion chromatography (IC) proved to be a versatile tool to analyse ammonia(NH3), amines (as cations), and organic acids (as anions) at part per billion by volume (ppbv)levels and lower. The addition of a mass spectrometer (MS) as second detector provided fur-ther separation of the coeluting compounds from gas phase and particles samples. Parallelsampling of both phases allowed the determination of partitioning coefficients and the esti-mation of total yields for species partitioning. IC can be used online, combined with differentsampling devices using water, so that even “sticky” species like NH3 can be collected withoutbeing lost to surfaces. One drawback is that artefacts can occur, such as hydrolysis of analytesin the sampling system. IC could also be used offline with longer analysis times for betterseparation of the peaks.

81

Chapter 7. Conclusions and outlook

7.1. Trace gases and new particle formation

In laboratory experiments, the chemical analysis of the gas phase was for decades mainlyfocused on the analysis of the main oxidation products from a precursor and the monitoring ofthe used reactants. However, it appears that products and contaminants present at ppbv levelsand below may affect the results of the experiment (Killus and Whitten, 1990; Zador et al.,2006), even if they are not detected. Due to the challenging detection and quantification ofsuch species at very low levels with a sufficently high time resolution, many contaminants areusually not detected during environmental chamber experiments. Kirkby et al. (2011) were thefirst to demonstrate that the postulated ternary nucleation of sulphuric acid (H2SO4), water(H2O) and NH3 cannot explain atmospheric nucleation rates. Those results suggest that, dueto the lack of characterization of contaminants during laboratory experiments, the previousobserved nucleation rates were biased (Sipila et al., 2010).

Other trace gases at very low levels (e. g. amines, organics) could play an important role inthe new particle formation mechanism during laboratory experiments and in the atmosphere.Metzger et al. (2010) demonstrated already that organics are expected to be involved in the nu-cleation process. Amines were suggested to react similary to ammonia in a ternary nucleationmechanism with H2SO4 and H2O (Kurten et al., 2008). For this reason, the CLOUD project atCERN designed experiments to investigate the influence of dimethylamine (DMA) on ternarynucleation. DMA was monitored with a ion chromatography presented in this work, based onconcentration of water used for sampling onto a cation concentration column. Detection limitsas low as a few tenth of part per trillion by volume (pptv) could be achieved with a time resolu-tion of roughly 3 hours. This method allowed simultaneous monitoring of NH3 at very low lev-els, because NH3 was not added intentionally to the CLOUD chamber and was only present at acontaminant level. This work revealed that at 278 K the background mixing ratio of NH3 showmostly values below 20 pptv. However, this method is restricted to applications over 273 K dueto the sampling with water and because there is no separation of the gas phase from theaerosol particles, correction may be required for total mass loadings higher than 0.1 µg m−3.The chromatographic method used may be affected if other cations are present in the samples(e. g. sodium, potassium, magnesium, calcium in the aerosol phase).

The results of the CLOUD4 campaigns show that DMA increases the nucleation rates ofternary nucleation by around a factor 1000, compared to NH3. Only a few pptv of DMA are suf-ficient to reach similar nucleation rates to the ones observed in the atmosphere. The increaseof DMA mixing ratio up to 65 pptv results in less than a factor 3 higher nucleation rates. Satu-ration occurs already a low DMA pptv levels. Evidence was found that the negative clusters areformed by the same base-stabilisation mechanism previously seen for NH3 tertiary nucleation(Kirkby et al., 2011) and that in the presence of DMA and only contaminant levels of NH3, thelatter is not found in the clusters. The ion-induced contribution is strongly suppressed by DMA,in contrast to the experiments performed with NH3.

82

7.2. Chemical mechanisms in the troposphere

7.2. Chemical mechanisms in the troposphere

7.2.1. Organic acids formation mechanisms

The formation of organic acids remains unsatisfactorily understood. Although severalmechanisms are known to lead to the formation of the low molar mass organic acids (3 car-bon atoms or less), the understanding of the formation of organic acids with higher masses(4 carbon atoms or more) is still poor and may involve specific chemical mechanisms. Largermolecules are more subject to intramolecular reactions (e. g. hydrogen atom migrations) sothat it increases the complexity of the potential reaction routes. Therefore, the formation ofidentified products with large nominal masses often cannot be explained.

Interestingly, the formation as a primary oxidation product of a dicarboxylic acid froma precursor not containing any carboxylic functionality demonstrates the importance of theintramolecular reactions that allows the addition of many oxygen atoms to the primary com-pound. Succinic acid is observed from the oxidation of cyclobutyl methyl ketone (CMK) andsimilarly 3-methyl-1,2,3-butanetricarboxylic acid (MBTCA) is formed as a first generation prod-uct of the (monocarboxylic) pinonic acid (Chapter 5). This indicates that aerosol aging formingoxygenated organic aerosol (OOA) can be driven mainly by the formation and the condensa-tion of those highly oxidised products.

Studies on model compounds such as CMK offer the possibility to study simpler systemswith products easier to identify and quantify, due to the availibility of standards. Extrapola-tion of the results cannot always be made quantitatively, but at least qualitative results onexpected chemical structures can be formulated, even if the reaction mechanisms remain un-clear.

7.2.2. Effect of sulphur dioxide (SO2)

The presence of 2 ppbv of SO2 has a marked effect on the organic acids particle concentra-tion during 1,3,5-trimethylbenzene (TMB) photooxidation (Chapter 6). The detected amount ofwater soluble organic acids decreases from 6–14 % to below 3 %. This may be an indication forcondensed phase reaction as the formation of organosulfates or the acid-catalysed condensa-tion reactions forming oligomers (Tolocka et al., 2004), due to presence of sulfuric acid in theaerosol particles. Alternatively, the precursors of organic acids in the gas phase may react withSO2 or other sulphur species.

7.2.3. Ring-closure mechanisms

Sampling of gas phase and aerosol with water as performed in this work hydrolyses an-hydrides and very likely lactones too. Therefore, those species are detected as organic acids asdemonstrated with methylmaleic anhydride (Chapter 6) and terpenylic acid (Chapter 5). Priorknowledge from other measurements or from the oxidation mechanism can help to interpretthe data, but also estimation of a partitioning coefficient.

83

Chapter 7. Conclusions and outlook

In the Master Chemical Mechanism (MCM) degradation scheme of TMB, the precursorsof methylmaleic anhydride are acyloxy radicals. Those are expected to eliminate carbon diox-ide (CO2) to form a new alkyl radical. Vereecken and Peeters (2010) suggest from structure-activity relationships that an hydrogen atom migration to the acyloxy radical cannot competewith this process. On the other hand, as the structure of SOA products are identified, formationmechanisms are proposed involving that kind of intramolecular migrations (Jenkin et al., 2000;Muller et al., 2012). Furthermore, lactones are cyclic esters and esters are usually expected toform in the gas phase from the reaction of Criegee biradicals through the so-called ester chan-nel. However, our results suggest that terpenylic acid contains a lactone functionality and isformed by oxidation of pinonic acid by OH. Therefore, another ring-closure mechanism toexplain the formation of lactones (and anhydrides) must exist. Those examples illustrate theneed for identification of new reaction mechanisms. Unimolecular reactions are potentiallyvery important because the low concentrations of the products make the intermolecular reac-tion with other trace compounds less probable.

7.3. Outlook

In the future, powerful measurement techniques will be required to analyse gas phasecompounds at levels below 1 pptv, similarily to the chemical ionisation mass spectrometer(CIMS) used to measure sulphuric acid (Eisele and Tanner, 1993) as low as 106 cm−3 (0.04 pptv).Several mass spectrometry techniques are also suitable for chemical characterization of aerosolparticles (Baltensperger et al., 2010). Coupled with chromatography techniques like IC (butmore generally all liquid chromatography methods), they are the method of choice to analyse(online or offline) individual compounds from the gas phase and the aerosol particles. Due tothe complex mixture of oxidation products, a quadrupole MS may not be sufficient to identifyproducts due to the lack of available standards. More than performing an offline analysis witha high resolution MS, coupling of chromatography techniques with MS for online detectionof oxidation products can be more efficient. Moreover, tandem MS techniques would providedirectly structural information on the compounds.

Studies with simpler molecular structures should also be designed in order to better un-derstand gas phase oxidation mechanisms, even if the compound of interest is not directlyrelevant for the real atmosphere. The results of those studies can be extrapolated to morecomplex systems of interest. Because validation of reaction pathways is virtually impossible,indirect information about a mechanism can be found by confirming the presence of side prod-ucts. On the other hand, if expected products cannot be found, quantitative considerationsmay play a role, due to detection limitations. Finally, identification of non-traditional reactionmechanisms will be essential to explain formation pathways of products observed in chamberexperiments or in the field.

84

References

Ahlrichs, R., Bar, M., Haser, M., Horn, H., and Kolmel, C., Electronic structure calculations onworkstation computers: The program system turbomole, Chem. Phys. Lett., 162(3), 165–169,1989.

Akyuz, M., Simultaneous determination of aliphatic and aromatic amines in indoor and out-door air samples by gas chromatography-mass spectrometry, Talanta, 71(1), 486–492, 2007.

Alfarra, M. R., Paulsen, D., Gysel, M., Garforth, A. A., Dommen, J., Prevot, A. S. H., Worsnop, D. R.,Baltensperger, U., and Coe, H., A mass spectrometric study of secondary organic aerosolsformed from the photooxidation of anthropogenic and biogenic precursors in a reactionchamber, Atmos. Chem. Phys., 6(12), 5279–5293, 2006.

Altieri, K., Seitzinger, S., Carlton, A., Turpin, B., Klein, G., and Marshall, A., Oligomers formedthrough in-cloud methylglyoxal reactions: Chemical composition, properties, and mecha-nisms investigated by ultra-high resolution FT-ICR mass spectrometry, Atmos. Environ., 42(7),1476–1490, 2008.

Andreae, M. O., Jones, C. D., and Cox, P. M., Strong present-day aerosol cooling implies a hotfuture, Nature, 435(7046), 1187–1190, 2005.

Angelino, S., Suess, D. T., and Prather, K. A., Formation of aerosol particles from reactions ofsecondary and tertiary alkylamines: Characterization by aerosol time-of-flight mass spec-trometry, Environ. Sci. Technol., 35(15), 3130–3138, 2001.

Atkinson, R., Gas-phase tropospheric chemistry of volatile organic compounds: 1. Alkanes andalkenes, J. Phys. Chem. Ref. Data, 26(2), 215–290, 1997.

Atkinson, R., Atmospheric chemistry of VOCs and NOx, Atmos. Environ., 34(12-14), 2063–2101,2000.

Baltensperger, U., Chirico, R., DeCarlo, P. F., Dommen, J., Gaeggeler, K., Heringa, M. F., Li, M.,Prevot, A. S. H., Alfarra, M. R., Gross, D. S., and Kalberer, M., Recent developments in the massspectrometry of atmospheric aerosols, Eur. J. Mass Spectrom., 16(1), 389–395, 2010.

Barsanti, K. C., McMurry, P. H., and Smith, J. N., The potential contribution of organic salts tonew particle growth, Atmos. Chem. Phys., 9(9), 2949–2957, 2009.

Barsanti, K. C. and Pankow, J. F., Thermodynamics of the formation of atmospheric organic par-ticulate matter by accretion reactions—2. Dialdehydes, methylglyoxal, and diketones, Atmos.Environ., 39(35), 6597–6607, 2005.

85

References

Barsanti, K. C. and Pankow, J. F., Thermodynamics of the formation of atmospheric organic par-ticulate matter by accretion reactions—Part 3: Carboxylic and dicarboxylic acids, Atmos. En-viron., 40(34), 6676–6686, 2006.

Bates, D., Recombination of small ions in the troposphere and lower stratosphere, Planet. SpaceSci., 30(12), 1275–1282, 1982.

Bates, R. G. and Pinching, G. D., Dissociation constant of aqueous ammonia at 0 to 50° from e.m. f. studies of the ammonium salt of a weak acid, J. Am. Chem. Soc., 72(3), 1393–1396, 1950.

Becke, A. D., Density-functional thermochemistry. III. the role of exact exchange, J. Chem. Phys.,98(7), 5648–5652, 1993.

Benson, D. R., Markovich, A., Al-Refai, M., and Lee, S.-H., A chemical ionization mass spectrome-ter for ambient measurements of ammonia, Atmos. Meas. Tech., 3(4), 1075–1087, 2010.

Berndt, T., Stratmann, F., Sipila, M., Vanhanen, J., Petaja, T., Mikkila, J., Gruner, A., Spindler, G., LeeMauldin III, R., Curtius, J., Kulmala, M., and Heintzenberg, J., Laboratory study on new particleformation from the reaction OH + SO2: influence of experimental conditions, H2O vapour,NH3 and the amine tert-butylamine on the overall process, Atmos. Chem. Phys., 10(15), 7101–7116, 2010.

Bianchi, F., Dommen, J., Mathot, S., and Baltensperger, U., On-line determination of ammoniaat low pptv mixing ratios in the CLOUD chamber, Atmos. Meas. Tech., 5(7), 1719–1725, 2012.

Bouwman, A. F., Lee, D. S., Asman, W. A. H., Dentener, F. J., Hoek, K. W. V. D., and Olivier, J. G. J.,A global high-resolution emission inventory for ammonia, Global Biogeochem. Cycles, 11(4),561–587, 1997.

Boy, M., Mogensen, D., Smolander, S., Zhou, L., Nieminen, T., Paasonen, P., Plass-Dulmer, C.,Sipila, M., Petaja, T., Mauldin, L., Berresheim, H., and Kulmala, M., Oxidation of SO2 by sta-bilized criegee intermediate (sCI) radicals as a crucial source for atmospheric sulphuric acidconcentrations, Atmos. Chem. Phys. Discuss., 12(10), 27693–27736, 2012.

Bzdek, B. R., Ridge, D. P., and Johnston, M. V., Amine exchange into ammonium bisulfate andammonium nitrate nuclei, Atmos. Chem. Phys., 10(8), 3495–3503, 2010a.

Bzdek, B. R., Ridge, D. P., and Johnston, M. V., Size-dependent reactions of ammonium bisulfateclusters with dimethylamine, J. Phys. Chem. A, 114(43), 11638–11644, 2010b.

Bzdek, B. R., Ridge, D. P., and Johnston, M. V., Amine reactivity with charged sulfuric acid clusters,Atmos. Chem. Phys., 11(16), 8735–8743, 2011.

Calvert, J., The mechanisms of atmospheric oxidation of aromatic hydrocarbons, Oxford Univer-sity Press, USA, 2002.

Cercl’Air, Empfehlung Nr. 28: Messung der Partikelanzahlkonzentration und -grossen-verteilung, 2010.

Chang, I., Lee, C., and Lee, D. S., Development of an automated method for simultaneous deter-mination of low molecular weight aliphatic amines and ammonia in ambient air by diffusionscrubber coupled to ion chromatography, Anal. Chem., 75(22), 6141–6146, 2003.

86

References

Chebbi, A. and Carlier, P., Carboxylic acids in the troposphere, occurrence, sources, and sinks: Areview, Atmos. Environ., 30(24), 4233–4249, 1996.

Chen, M., Titcombe, M., Jiang, J., Jen, C., Kuang, C., Fischer, M. L., Eisele, F. L., Siepmann, J. I.,Hanson, D. R., Zhao, J., and McMurry, P. H., Acid–base chemical reaction model for nucleationrates in the polluted atmospheric boundary layer, Proc. Natl. Acad. Sci. USA, 109(46), 18713–18718, 2012.

Cheng, Y., Condensation detection and diffusion size separation techniques, in P. A. Baron andK. Willeke, editors, Aerosol measurement — Principles, techniques, and applications, John Wi-ley & Sons, 2nd edition, 2001.

Claeys, M., Iinuma, Y., Szmigielski, R., Surratt, J. D., Blockhuys, F., Van Alsenoy, C., Boge, O., Sierau,B., Gomez-Gonzalez, Y., Vermeylen, R., Van der Veken, P., Shahgholi, M., Chan, A. W. H., Her-rmann, H., Seinfeld, J. H., and Maenhaut, W., Terpenylic acid and related compounds from theoxidation of α-pinene: Implications for new particle formation and growth above forests,Environ. Sci. Technol., 43(18), 6976–6982, 2009.

Coe, H. and Allan, J. D., Mass spectrometric methods for arosol composition masurements,in Analytical techniques for atmospheric measurement, pages 267–310, Blackwell Publishing,2007.

Coffman, D. J. and Hegg, D. A., A preliminary study of the effect of ammonia on particle nucle-ation in the marine boundary layer, J. Geophys. Res., 100(D4), 7147–7160, 1995.

Compernolle, S., Ceulemans, K., and Muller, J.-F., EVAPORATION: a new vapour pressure estima-tion methodfor organic molecules including non-additivity and intramolecular interactions,Atmos. Chem. Phys., 11(18), 9431–9450, 2011.

Decesari, S., Facchini, M. C., Fuzzi, S., and Tagliavini, E., Characterization of water-soluble organiccompounds in atmospheric aerosol: A new approach, J. Geophys. Res., 105(D1), 1481–1489,2000.

Dommen, J., Prevot, A., Baertsch-Ritter, N., Maffeis, G., Longoni, M., Gruebler, F., and Thielmann,A., High-resolution emission inventory of the Lombardy region: development and compari-son with measurements, Atmos. Environ., 37(29), 4149–4161, 2003.

Dunning, T. H., Peterson, K. A., and Wilson, A. K., Gaussian basis sets for use in correlated molec-ular calculations. x. the atoms aluminum through argon revisited, J. Chem. Phys., 114(21),9244–9253, 2001.

Ehn, M., Junninen, H., Petaja, T., Kurten, T., Kerminen, V.-M., Schobesberger, S., Manninen, H. E.,Ortega, I. K., Vehkamaki, H., Kulmala, M., and Worsnop, D. R., Composition and temporal be-havior of ambient ions in the boreal forest, Atmos. Chem. Phys., 10(17), 8513–8530, 2010.

Ehn, M., Junninen, H., Schobesberger, S., Manninen, H. E., Franchin, A., Sipila, M., Petaja, T., Ker-minen, V.-M., Tammet, H., Mirme, A., Mirme, S., Horrak, U., Kulmala, M., and Worsnop, D. R., Aninstrumental comparison of mobility and mass measurements of atmospheric small ions,Aerosol Sci. Technol., 45(4), 522–532, 2011.

87

References

Eisele, F. L. and Tanner, D. J., Measurement of the gas phase concentration of H2SO4 andmethane sulfonic acid and estimates of H2SO4 production and loss in the atmosphere, J.Geophys. Res., 98(D5), 9001–9010, 1993.

Erupe, M. E., Viggiano, A. A., and Lee, S.-H., The effect of trimethylamine on atmospheric nucle-ation involving H2SO4, Atmos. Chem. Phys., 11(10), 4767–4775, 2011.

Esmen, N. A. and Corn, M., Residence time of particles in urban air, Atmos. Environ. (1967), 5(8),571–578, 1971.

Facchini, M. C., Decesari, S., Rinaldi, M., Carbone, C., Finessi, E., Mircea, M., Fuzzi, S., Moretti, F.,Tagliavini, E., Ceburnis, D., and O’Dowd, C. D., Important source of marine secondary organicaerosol from biogenic amines, Environ. Sci. Technol., 42(24), 9116–9121, 2008.

Fisseha, R., Dommen, J., Sax, M., Paulsen, D., Kalberer, M., Maurer, R., Hofler, F., Weingartner, E.,and Baltensperger, U., Identification of organic acids in secondary organic aerosol and thecorresponding gas phase from chamber experiments, Anal. Chem., 76(22), 6535–6540, 2004.

Forster, P., Ramaswamy, V., Artaxo, P., Berntsen, T., Betts, R., Fahey, D. W., Haywood, J., Lean, J.,Lowe, D. C., Myhre, G., Nganga, J., Prinn, R., Raga, G., Schulz, M., and Van Dorland, R., Changesin atmospheric constituents and in radiative forcing, in S. D. Solomon, M. Qin, Z. Manning,M. Chen, K. B. Marquis, M. Averyt, Tignor, and H. L. Miller, editors, Climate Change 2007: ThePhysical Science Basis. Contribution of Working Group I to the Fourth Assessment Report of theIntergovernmental Panel on Climate Change, Cambridge University Press, Cambridge, UnitedKingdom and New York, NY, USA, 2007.

Frisch, M. J., Trucks, G. W., Schlegel, H. B., Scuseria, G. E., Robb, M. A., Cheeseman, J. R., Scalmani,G., Barone, V., Mennucci, B., Petersson, G. A., Nakatsuji, H., Caricato, M., Li, X., Hratchian, H. P.,Izmaylov, A. F., Bloino, J., Zheng, G., Sonnenberg, J. L., Hada, M., Ehara, M., Toyota, K., Fukuda,R., Hasegawa, J., Ishida, M., Nakajima, T., Honda, Y., Kitao, O., Nakai, H., Vreven, T., Mont-gomery, J. A., Jr., Peralta, J. E., Ogliaro, F., Bearpark, M., Heyd, J. J., Brothers, E., Kudin, K. N.,Staroverov, V. N., Kobayashi, R., Normand, J., Raghavachari, K., Rendell, A., Burant, J. C., Iyen-gar, S. S., Tomasi, J., Cossi, M., Rega, N., Millam, J. M., Klene, M., Knox, J. E., Cross, J. B., Bakken,V., Adamo, C., Jaramillo, J., Gomperts, R., Stratmann, R. E., Yazyev, O., Austin, A. J., Cammi, R.,Pomelli, C., Ochterski, J. W., Martin, R. L., Morokuma, K., Zakrzewski, V. G., Voth, G. A., Sal-vador, P., Dannenberg, J. J., Dapprich, S., Daniels, A. D., Farkas, O., Foresman, J. B., Ortiz, J. V.,Cioslowski, J., and Fox, D. J., Gaussian 09 Revision A.1, 2009, Gaussian Inc. Wallingford CT2009.

Gaeggeler, K., Prevot, A. S., Dommen, J., Legreid, G., Reimann, S., and Baltensperger, U., Residen-tial wood burning in an alpine valley as a source for oxygenated volatile organic compounds,hydrocarbons and organic acids, Atmos. Environ., 42(35), 8278–8287, 2008.

Ge, X., Wexler, A. S., and Clegg, S. L., Atmospheric amines — Part I. A review, Atmos. Environ.,45(3), 524–546, 2011.

Goldstein, A. H. and Galbally, I. E., Known and unexplored organic constituents in the Earth’satmosphere, Environ. Sci. Technol., 41(5), 1514–1521, 2007.

Graus, M., Muller, M., and Hansel, A., High resolution PTR-TOF: quantification and formula con-firmation of VOC in real time, J. Am. Soc. Mass Spectrom., 21(6), 1037–1044, 2010.

88

References

Gronberg, L., Lovkvist, P., and Jonsson, J. A., Measurement of aliphatic amines in ambient airand rainwater, Chemosphere, 24(10), 1533–1540, 1992.

Grossert, J. S., Fancy, P. D., and White, R. L., Fragmentation pathways of negative ions producedby electrospray ionization of acyclic dicarboxylic acids and derivatives, Can. J. Chem., 83(11),1878–1890, 2005.

Gronberg, L., Lovkvist, P., and Jonsson, J., Determination of aliphatic amines in air by membraneenrichment directly coupled to a gas chromatograph, Chromatographia, 33(1), 77–82, 1992.

Guenther, A., Geron, C., Pierce, T., Lamb, B., Harley, P., and Fall, R., Natural emissions of non-methane volatile organic compounds, carbon monoxide, and oxides of nitrogen from NorthAmerica, Atmos. Environ., 34(12–14), 2205–2230, 2000.

Guenther, A., Hewitt, C. N., Erickson, D., Fall, R., Geron, C., Graedel, T., Harley, P., Klinger, L., Ler-dau, M., Mckay, W. A., Pierce, T., Scholes, B., Steinbrecher, R., Tallamraju, R., Taylor, J., and Zim-merman, P., A global model of natural volatile organic compound emissions, J. Geophys. Res.,100(D5), 8873–8892, 1995.

Haagen-Smit, A. J., Chemistry and physiology of Los Angeles smog, Ind. Eng. Chem., 44(6), 1342–1346, 1952.

Haagen-Smit, A. J., Bradley, C. E., and Fox, M. M., Ozone formation in photochemical oxidationof organic substances, Ind. Eng. Chem., 45(9), 2086–2089, 1953.

Hagen, D. E. and Alofs, D. J., Linear inversion method to obtain aerosol size distributions frommeasurements with a differential mobility analyzer, Aerosol Sci. Technol., 2(4), 465–475, 1983.

Hall, H. K., Correlation of the base strengths of amines, J. Am. Chem. Soc., 79(20), 5441–5444,1957.

Hallquist, M., Wenger, J. C., Baltensperger, U., Rudich, Y., Simpson, D., Claeys, M., Dommen, J.,Donahue, N. M., George, C., Goldstein, A. H., Hamilton, J. F., Herrmann, H., Hoffmann, T.,Iinuma, Y., Jang, M., Jenkin, M. E., Jimenez, J. L., Kiendler-Scharr, A., Maenhaut, W., McFiggans,G., Mentel, T. F., Monod, A., Prevot, A. S. H., Seinfeld, J. H., Surratt, J. D., Szmigielski, R., andWildt, J., The formation, properties and impact of secondary organic aerosol: current andemerging issues, Atmos. Chem. Phys., 9(14), 5155–5236, 2009.

Hamilton, J. F. and Lewis, A. C., Chromatographic methods, in Analytical Techniques for Atmo-spheric Measurement, pages 361–405, Blackwell Publishing, 2007.

Hanson, D. R. and Lovejoy, E. R., Measurement of the thermodynamics of the hydrated dimerand trimer of sulfuric acid, J. Phys. Chem. A, 110(31), 9525–9528, 2006.

Hatakeyama, S. and Akimoto, H., Reactions of criegee intermediates in the gas phase, Res.Chem. Intermed., 20(3), 503–524, 1994.

Hawkes, S. J., pKw is almost never 14.0: Contribution from the Task Force on the General Chem-istry Curriculum, J. Chem. Educ., 72(9), 799–802, 1995.

Haynes, W. M. and Lide, D. R., CRC handbook of chemistry and physics a ready-reference book ofchemical and physical data : 2010-2011, CRC Press, Boca Raton (Fla.); London; New York, 2010.

89

References

Healy, R. M., Wenger, J. C., Metzger, A., Duplissy, J., Kalberer, M., and Dommen, J., Gas/particlepartitioning of carbonyls in the photooxidation of isoprene and 1,3,5-trimethylbenzene, At-mos. Chem. Phys., 8(12), 3215–3230, 2008.

Herron, J. T. and Huie, R. E., Stopped-flow studies of the mechanisms of ozone-alkene reactionsin the gas phase. ethylene, J. Am. Chem. Soc., 99(16), 5430–5435, 1977.

Hirsikko, A., Nieminen, T., Gagne, S., Lehtipalo, K., Manninen, H. E., Ehn, M., Horrak, U., Kermi-nen, V.-M., Laakso, L., McMurry, P. H., Mirme, A., Mirme, S., Petaja, T., Tammet, H., Vakkari,V., Vana, M., and Kulmala, M., Atmospheric ions and nucleation: a review of observations,Atmos. Chem. Phys., 11(2), 767–798, 2011.

Hobbs, P. V., Introduction to Atmospheric Chemistry, Cambridge University Press, 2000.

Hu, Q., Noll, R. J., Li, H., Makarov, A., Hardman, M., and Graham Cooks, R., The orbitrap: A newmass spectrometer, J. Mass Spectrom., 40(4), 430–443, 2005.

Hattig, C. and Weigend, F., CC2 excitation energy calculations on large molecules using theresolution of the identity approximation, J. Chem. Phys., 113(13), 5154–5161, 2000.

Iida, K., Stolzenburg, M. R., and McMurry, P. H., Effect of working fluid on sub-2 nm particledetection with a laminar flow ultrafine condensation particle counter, Aerosol Sci. Technol.,43(1), 81–96, 2009.

Iinuma, Y., Boge, O., Keywood, M., Gnauk, T., and Herrmann, H., Diaterebic acid acetate and di-aterpenylic acid acetate: Atmospheric tracers for secondary organic aerosol formation from1,8-cineole oxidation, Environ. Sci. Technol., 43(2), 280–285, 2008.

IPCC, Climate Change 2007: The Physical Science Basis. Contribution of Working Group I to theFourth Assessment Report of the Intergovernmental Panel on Climate Change, Cambridge Uni-versity Press, Cambridge, United Kingdom and New York, NY, USA, 2007.

Jenkin, M. E., Saunders, S. M., Wagner, V., and Pilling, M. J., Protocol for the development of themaster chemical mechanism, MCM v3 (Part B): tropospheric degradation of aromatic volatileorganic compounds, Atmos. Chem. Phys., 3(1), 181–193, 2003.

Jenkin, M. E., Shallcross, D. E., and Harvey, J. N., Development and application of a possiblemechanism for the generation of cis-pinic acid from the ozonolysis of α- and β-pinene, At-mos. Environ., 34(18), 2837–2850, 2000.

Junninen, H., Ehn, M., Petaja, T., Luosujarvi, L., Kotiaho, T., Kostiainen, R., Rohner, U., Gonin, M.,Fuhrer, K., Kulmala, M., and Worsnop, D. R., A high-resolution mass spectrometer to measureatmospheric ion composition, Atmos. Meas. Tech., 3(4), 1039–1053, 2010.

Kasting, J. F., The rise of atmospheric oxygen, Science, 293(5531), 819–820, 2001.

Kawamura, K., Steinberg, S., and Kaplan, I. R., Homologous series of C1-C10 monocarboxylicacids and C1-C6 carbonyls in Los Angeles air and motor vehicle exhausts, Atmos. Environ.,34(24), 4175–4191, 2000.

Kerminen, V.-M. and Kulmala, M., Analytical formulae connecting the “real” and the “apparent”nucleation rate and the nuclei number concentration for atmospheric nucleation events, J.Aerosol Sci., 33(4), 609–622, 2002.

90

References

Kerminen, V.-M., Petaja, T., Manninen, H. E., Paasonen, P., Nieminen, T., Sipila, M., Junninen, H.,Ehn, M., Gagne, S., Laakso, L., Riipinen, I., Vehkamaki, H., Kurten, T., Ortega, I. K., Dal Maso,M., Brus, D., Hyvarinen, A., Lihavainen, H., Leppa, J., Lehtinen, K. E. J., Mirme, A., Mirme, S.,Horrak, U., Berndt, T., Stratmann, F., Birmili, W., Wiedensohler, A., Metzger, A., Dommen,J., Baltensperger, U., Kiendler-Scharr, A., Mentel, T. F., Wildt, J., Winkler, P. M., Wagner, P. E.,Petzold, A., Minikin, A., Plaß-Dulmer, C., Poschl, U., Laaksonen, A., and Kulmala, M., Atmo-spheric nucleation: highlights of the EUCAARI project and future directions, Atmos. Chem.Phys., 10(22), 10829–10848, 2010.

Killus, J. P. and Whitten, G. Z., Background reactivity in smog chambers, Int. J. Chem. Kinet., 22(6),547–575, 1990.

Kirkby, J., Cosmic rays and climate, Surv. Geophys., 28(5-6), 333–375, 2007.

Kirkby, J., Curtius, J., Almeida, J., Dunne, E., Duplissy, J., Ehrhart, S., Franchin, A., Gagne, S., Ickes, L.,Kurten, A., Kupc, A., Metzger, A., Riccobono, F., Rondo, L., Schobesberger, S., Tsagkogeorgas, G.,Wimmer, D., Amorim, A., Bianchi, F., Breitenlechner, M., David, A., Dommen, J., Downard, A.,Ehn, M., Flagan, R. C., Haider, S., Hansel, A., Hauser, D., Jud, W., Junninen, H., Kreissl, F., Kvashin,A., Laaksonen, A., Lehtipalo, K., Lima, J., Lovejoy, E. R., Makhmutov, V., Mathot, S., Mikkila,J., Minginette, P., Mogo, S., Nieminen, T., Onnela, A., Pereira, P., Petaja, T., Schnitzhofer, R.,Seinfeld, J. H., Sipila, M., Stozhkov, Y., Stratmann, F., Tome, A., Vanhanen, J., Viisanen, Y., Vrtala,A., Wagner, P. E., Walther, H., Weingartner, E., Wex, H., Winkler, P. M., Carslaw, K. S., Worsnop,D. R., Baltensperger, U., and Kulmala, M., Role of sulphuric acid, ammonia and galactic cosmicrays in atmospheric aerosol nucleation, Nature, 476(7361), 429–433, 2011.

Kirkup, L. and Mulholland, M., Comparison of linear and non-linear equations for univariatecalibration, J. Chromatogr. A, 1029(1-2), 1–11, 2004.

Klumpp, A., Ansel, W., Klumpp, G., Calatayud, V., Pierre Garrec, J., He, S., Penuelas, J., Ribas, A.,Ro-Poulsen, H., Rasmussen, S., Sanz, M. J., and Vergne, P., Ozone pollution and ozone biomon-itoring in European cities. Part I: Ozone concentrations and cumulative exposure indices aturban and suburban sites, Atmos. Environ., 40(40), 7963–7974, 2006.

Kondratyev, K. Y., Ivlev, L. S., Krapivin, V. F., and Varotsos, C. A., Atmospheric Aerosol Properties,Springer Praxis Books, Springer Berlin Heidelberg, 2006.

Korhonen, P., Kulmala, M., Laaksonen, A., Viisanen, Y., McGraw, R., and Seinfeld, J. H., Ternarynucleation of H2SO4, NH3, and H2O in the atmosphere, J. Geophys. Res., 104(D21), 26349–26353, 1999.

Krupa, S. V., Effects of atmospheric ammonia (NH3) on terrestrial vegetation: A review, Environ.Pollut., 124(2), 179–221, 2003.

Kuang, C., McMurry, P. H., McCormick, A. V., and Eisele, F. L., Dependence of nucleation rateson sulfuric acid vapor concentration in diverse atmospheric locations, J. Geophys. Res., 113,D10209, 2008.

Kulmala, M. and Kerminen, V.-M., On the formation and growth of atmospheric nanoparticles,Atmos. Res., 90(2–4), 132–150, 2008.

91

References

Kulmala, M., Pirjola, L., and Makela, J. M., Stable sulphate clusters as a source of new atmo-spheric particles, Nature, 404(6773), 66–69, 2000.

Kulmala, M., Riipinen, I., Sipila, M., Manninen, H. E., Petaja, T., Junninen, H., Maso, M. D., Mordas,G., Mirme, A., Vana, M., Hirsikko, A., Laakso, L., Harrison, R. M., Hanson, I., Leung, C., Lehtinen,K. E. J., and Kerminen, V.-M., Toward direct measurement of atmospheric nucleation, Science,318(5847), 89–92, 2007.

Kulmala, M., Vehkamaki, H., Petaja, T., Dal Maso, M., Lauri, A., Kerminen, V.-M., Birmili, W., andMcMurry, P. H., Formation and growth rates of ultrafine atmospheric particles: a review ofobservations, J. Aerosol Sci., 35, 143–176, 2004.

Kupc, A., Amorim, A., Curtius, J., Danielczok, A., Duplissy, J., Ehrhart, S., Walther, H., Ickes, L.,Kirkby, J., Kurten, A., Lima, J., Mathot, S., Minginette, P., Onnela, A., Rondo, L., and Wagner, P., Afibre-optic UV system for H2SO4 production in aerosol chambers causing minimal thermaleffects, J. Aerosol Sci., 42(8), 532–543, 2011.

Kupiainen, O., Ortega, I. K., Kurten, T., and Vehkamaki, H., Amine substitution into sulfuric acid– ammonia clusters, Atmos. Chem. Phys., 12(8), 3591–3599, 2012.

Kurten, T., Lane, J. R., Jørgensen, S., and Kjaergaard, H. G., A computational study of the oxida-tion of SO2 to SO3 by gas-phase organic oxidants, J. Phys. Chem. A, 115(31), 8669–8681, 2011a.

Kurten, T., Loukonen, V., Vehkamaki, H., and Kulmala, M., Amines are likely to enhance neu-tral and ion-induced sulfuric acid-water nucleation in the atmosphere more effectively thanammonia, Atmos. Chem. Phys., 8(14), 4095–4103, 2008.

Kurten, T., Petaja, T., Smith, J., Ortega, I. K., Sipila, M., Junninen, H., Ehn, M., Vehkamaki, H.,Mauldin, L., Worsnop, D. R., and Kulmala, M., The effect of H2SO4 – amine clustering on chem-ical ionization mass spectrometry (CIMS) measurements of gas-phase sulfuric acid, Atmos.Chem. Phys., 11(6), 3007–3019, 2011b.

Kwok, E. and Atkinson, R., Estimation of hydroxyl radical reaction rate constants for gas-phaseorganic compounds using a structure-reactivity relationship: An update, Atmos. Environ.,29(14), 1685–1695, 1995.

Kurten, A., Rondo, L., Ehrhart, S., and Curtius, J., Performance of a corona ion source for mea-surement of sulfuric acid by chemical ionization mass spectrometry, Atmos. Meas. Tech., 4(3),437–443, 2011.

Kurten, A., Rondo, L., Ehrhart, S., and Curtius, J., Calibration of a chemical ionization mass spec-trometer for the measurement of gaseous sulfuric acid, J. Phys. Chem. A, 116(24), 6375–6386,2012.

Laakso, L., Anttila, T., Lehtinen, K. E. J., Aalto, P. P., Kulmala, M., Horrak, U., Paatero, J., Hanke, M.,and Arnold, F., Kinetic nucleation and ions in boreal forest particle formation events, Atmos.Chem. Phys., 4(9/10), 2353–2366, 2004.

Li, Y., Schwab, J. J., and Demerjian, K. L., Measurements of ambient ammonia using a tunablediode laser absorption spectrometer: Characteristics of ambient ammonia emissions in anurban area of new york city, J. Geophys. Res., 111, D10S02, 2006.

92

References

Lim, H.-J., Carlton, A. G., and Turpin, B. J., Isoprene forms secondary organic aerosol throughcloud processing: Model simulations, Environ. Sci. Technol., 39(12), 4441–4446, 2005.

Lohmann, U. and Feichter, J., Global indirect aerosol effects: a review, Atmos. Chem. Phys., 5(3),715–737, 2005.

Loukonen, V., Kurten, T., Ortega, I. K., Vehkamaki, H., Padua, A. A. H., Sellegri, K., and Kulmala, M.,Enhancing effect of dimethylamine in sulfuric acid nucleation in the presence of water – acomputational study, Atmos. Chem. Phys., 10(10), 4961–4974, 2010.

Madronich, S., Chatfield, R. B., Calvert, J. G., Moortgat, G. K., Veyret, B., and Lesclaux, R., A photo-chemical origin of acetic acid in the troposphere, Geophys. Res. Lett., 17(13), 2361–2364, 1990.

Makarov, A., Electrostatic axially harmonic orbital trapping: A high-performance technique ofmass analysis, Anal. Chem., 72(6), 1156–1162, 2000.

Manninen, H. E., Nieminen, T., Asmi, E., Gagne, S., Hakkinen, S., Lehtipalo, K., Aalto, P., Vana,M., Mirme, A., Mirme, S., Horrak, U., Plass-Dulmer, C., Stange, G., Kiss, G., Hoffer, A., Toro, N.,Moerman, M., Henzing, B., de Leeuw, G., Brinkenberg, M., Kouvarakis, G. N., Bougiatioti, A.,Mihalopoulos, N., O’Dowd, C., Ceburnis, D., Arneth, A., Svenningsson, B., Swietlicki, E., Tarozzi,L., Decesari, S., Facchini, M. C., Birmili, W., Sonntag, A., Wiedensohler, A., Boulon, J., Sellegri,K., Laj, P., Gysel, M., Bukowiecki, N., Weingartner, E., Wehrle, G., Laaksonen, A., Hamed, A.,Joutsensaari, J., Petaja, T., Kerminen, V.-M., and Kulmala, M., EUCAARI ion spectrometer mea-surements at 12 european sites – analysis of new particle formation events, Atmos. Chem.Phys., 10(16), 7907–7927, 2010.

Mauldin III, R. L., Berndt, T., Sipila, M., Paasonen, P., Petaja, T., Kim, S., Kurten, T., Stratmann, F.,Kerminen, V.-M., and Kulmala, M., A new atmospherically relevant oxidant of sulphur diox-ide, Nature, 488(7410), 193–196, 2012.

McGrath, M. J., Olenius, T., Ortega, I. K., Loukonen, V., Paasonen, P., Kurten, T., Kulmala, M., andVehkamaki, H., Atmospheric cluster dynamics code: a flexible method for solution of thebirth-death equations, Atmos. Chem. Phys., 12(5), 2345–2355, 2012.

Merikanto, J., Spracklen, D. V., Mann, G. W., Pickering, S. J., and Carslaw, K. S., Impact of nucle-ation on global CCN, Atmos. Chem. Phys., 9(21), 8601–8616, 2009.

Metzger, A., Verheggen, B., Dommen, J., Duplissy, J., Prevot, A. S. H., Weingartner, E., Riipinen,I., Kulmala, M., Spracklen, D. V., Carslaw, K. S., and Baltensperger, U., Evidence for the roleof organics in aerosol particle formation under atmospheric conditions, Proc. Natl. Acad. Sci.USA, 107(15), 6646–6651, 2010.

Middleton, P., Sources of air pollutants, in H. B. Singh, editor, Composition chemistry, and climateof the atmosphere, Van Nostrand Reinhold, New York [etc.], 1st edition, 1995.

Molina, L. T., Kolb, C. E., de Foy, B., Lamb, B. K., Brune, W. H., Jimenez, J. L., Ramos-Villegas, R.,Sarmiento, J., Paramo-Figueroa, V. H., Cardenas, B., Gutierrez-Avedoy, V., and Molina, M. J.,Air quality in north america’s most populous city – overview of the MCMA-2003 campaign,Atmos. Chem. Phys., 7(10), 2447–2473, 2007.

93

References

Moller, B., Rarey, J., and Ramjugernath, D., Estimation of the vapour pressure of non-electrolyteorganic compounds via group contributions and group interactions, J. Mol. Liq., 143(1), 52–63,2008.

Montgomery, J. A., Frisch, M. J., Ochterski, J. W., and Petersson, G. A., A complete basis set modelchemistry. VI. use of density functional geometries and frequencies, J. Chem. Phys., 110(6),2822–2827, 1999.

Murphy, S. M., Sorooshian, A., Kroll, J. H., Ng, N. L., Chhabra, P., Tong, C., Surratt, J. D., Knipping,E., Flagan, R. C., and Seinfeld, J. H., Secondary aerosol formation from atmospheric reactionsof aliphatic amines, Atmos. Chem. Phys., 7(9), 2313–2337, 2007.

Myrdal, P. B. and Yalkowsky, S. H., Estimating pure component vapor pressures of complex or-ganic molecules, Ind. Eng. Chem. Res., 36(6), 2494–2499, 1997.

Makela, J. M., Yli-koivisto, S., Hiltunen, V., Seidl, W., Swietlicki, E., Teinila, K., Sillanpaa, M., Ko-ponen, I. K., Paatero, J., Rosman, K., and Hameri, K., Chemical composition of aerosol duringparticle formation events in boreal forest, Tellus B, 53(4), 380–393, 2001.

Muller, L., Reinnig, M., Naumann, K. H., Saathoff, H., Mentel, T. F., Donahue, N. M., and Hoff-mann, T., Formation of 3-methyl-1,2,3-butanetricarboxylic acid via gas phase oxidation ofpinonic acid — a mass spectrometric study of SOA aging, Atmos. Chem. Phys., 12(3), 1483–1496, 2012.

Nadykto, A. B. and Yu, F., Uptake of neutral polar vapor molecules by charged clusters/particles:Enhancement due to dipole-charge interaction, J. Geophys. Res., 108(D23), 4717–4723, 2003.

Nannoolal, Y., Rarey, J., and Ramjugernath, D., Estimation of pure component properties: Part 3.Estimation of the vapor pressure of non-electrolyte organic compounds via group contribu-tions and group interactions, Fluid Phase Equilib., 269(1-2), 117–133, 2008.

Nannoolal, Y., Rarey, J., Ramjugernath, D., and Cordes, W., Estimation of pure component prop-erties: Part 1. Estimation of the normal boiling point of non-electrolyte organic compoundsvia group contributions and group interactions, Fluid Phase Equilib., 226, 45–63, 2004.

Napari, I., Noppel, M., Vehkamaki, H., and Kulmala, M., An improved model for ternary nucle-ation of sulfuric acid–ammonia–water, J. Chem. Phys., 116(10), 4221–4227, 2002.

Neuenschwander, U., Dft calculation on UB3LYP/6-311++G(df,pd)//UB3LYP/6-31G(d,p) includingZPE, 2012, unpublished results.

Neuenschwander, U. and Hermans, I., Autoxidation of α-pinene at high oxygen pressure, Phys.Chem. Chem. Phys., 12(35), 10542–10549, 2010.

Nielsen, C. J., Herrmann, H., and Weller, C., Atmospheric chemistry and environmental impactof the use of amines in carbon capture and storage (CCS), Chem. Soc. Rev., 41(19), 6684–6704,2012.

Olsen, J. V., Macek, B., Lange, O., Makarov, A., Horning, S., and Mann, M., Higher-energy C-trapdissociation for peptide modification analysis, Nat. Meth., 4(9), 709–712, 2007.

94

References

Ortega, I. K., Kupiainen, O., Kurten, T., Olenius, T., Wilkman, O., McGrath, M. J., Loukonen, V., andVehkamaki, H., From quantum chemical formation free energies to evaporation rates, Atmos.Chem. Phys., 12(1), 225–235, 2012.

Orzechowska, G. E. and Paulson, S. E., Photochemical sources of organic acids. 1. Reaction ofozone with isoprene, propene, and 2-butenes under dry and humid conditions using SPME,J. Phys. Chem. A, 109(24), 5358–5365, 2005.

Paasonen, P., Nieminen, T., Asmi, E., Manninen, H. E., Petaja, T., Plass-Dulmer, C., Flentje, H.,Birmili, W., Wiedensohler, A., Horrak, U., Metzger, A., Hamed, A., Laaksonen, A., Facchini,M. C., Kerminen, V.-M., and Kulmala, M., On the roles of sulphuric acid and low-volatilityorganic vapours in the initial steps of atmospheric new particle formation, Atmos. Chem.Phys., 10(22), 11223–11242, 2010.

Paasonen, P., Olenius, T., Kupiainen, O., Kurten, T., Petaja, T., Birmili, W., Hamed, A., Hu, M., Huey,L. G., Plass-Duelmer, C., Smith, J. N., Wiedensohler, A., Loukonen, V., McGrath, M. J., Ortega,I. K., Laaksonen, A., Vehkamaki, H., Kerminen, V.-M., and Kulmala, M., On the formation ofsulphuric acid – amine clusters in varying atmospheric conditions and its influence on at-mospheric new particle formation, Atmos. Chem. Phys., 12(19), 9113–9133, 2012.

Paasonen, P., Sihto, S.-L., Nieminen, T., Vuollekoski, H., Riipinen, I., Plaß-Dulmer, C., Berresheim,H., Birmili, W., and Kulmala, M., Connection between new particle formation and sulphuricacid at Hohenpeissenberg (Germany) including the influence of organic compounds., BorealEnviron. Res, 14, 616–629, 2009.

Pankow, J. F., An absorption model of gas/particle partitioning of organic compounds in theatmosphere, Atmos. Environ., 28(2), 185–188, 1994a.

Pankow, J. F., An absorption model of the gas/aerosol partitioning involved in the formation ofsecondary organic aerosol, Atmos. Environ., 28(2), 189–193, 1994b.

Paulsen, D., Dommen, J., Kalberer, M., Prevot, A. S. H., Richter, R., Sax, M., Steinbacher, M.,Weingartner, E., and Baltensperger, U., Secondary organic aerosol formation by irradiationof 1,3,5-trimethylbenzene–NOx–H2O in a new reaction chamber for atmospheric chemistryand physics, Env. Sci. Technol., 39(8), 2668–2678, 2005.

Peeters, J., Fantechi, G., and Vereecken, L., A generalized structure-activity relationship for thedecomposition of (substituted) alkoxy radicals, J. Atmos. Chem., 48(1), 59–80, 2004.

Penner, J. E., Andreae, M. O., Annegarn, H., Barrie, L., Feichter, J., Hegg, D., Jayaraman, A., Leaitch,R., Murphy, D., Nganga, J., and Pitari, G., Aerosols, their direct and indirect effects, in J. T.Houghton, Y. Ding, D. J. Griggs, M. Noguer, P. J. van der Linden, X. Dai, K. Maskell, and C. A.Johnson, editors, Climate Change 2001: The Scientific Basis. Contribution of Working Group Ito the Third Assessment Report of the Intergovernmental Panel on Climate Change, page 881,Cambridge University Press, Cambridge, United Kingdom and New York, NY, USA, 2001.

Petaja, T., Sipila, M., Paasonen, P., Nieminen, T., Kurten, T., Ortega, I. K., Stratmann, F., Vehkamaki,H., Berndt, T., and Kulmala, M., Experimental observation of strongly bound dimers of sulfuricacid: Application to nucleation in the atmosphere, Phys. Rev. Lett., 106(22), 228302, 2011.

95

References

Praplan, A. P., Bianchi, F., Dommen, J., and Baltensperger, U., Dimethylamine and ammoniameasurements with ion chromatography during the CLOUD4 campaign, Atmos. Meas. Tech.,5(9), 2161–2167, 2012.

Pun, B. K., Griffin, R. J., Seigneur, C., and Seinfeld, J. H., Secondary organic aerosol 2. Ther-modynamic model for gas/particle partitioning of molecular constituents, J. Geophys. Res.,107(D17), 4333, 2002.

Rochelle, G. T., Amine scrubbing for CO2 capture, Science, 325(5948), 1652–1654, 2009.

Sander, R., Compilation of Henry’s law constants for inorganic and organic species of potentialimportance in environmental chemistry (Version 3), http://www.henrys-law.org, 1999.

Sato, K., Takami, A., Kato, Y., Seta, T., Fujitani, Y., Hikida, T., Shimono, A., and Imamura, T., AMSand LC/MS analyses of SOA from the photooxidation of benzene and 1,3,5-trimethylbenzenein the presence of NOx: effects of chemical structure on SOA aging, Atmos. Chem. Phys.,12(10), 4667–4682, 2012.

Saunders, S. M., Jenkin, M. E., Derwent, R. G., and Pilling, M. J., Protocol for the developmentof the Master Chemical Mechanism, MCM v3 (Part A): tropospheric degradation of non-aromatic volatile organic compounds, Atmos. Chem. Phys., 3(1), 161–180, 2003.

Schade, G. W. and Crutzen, P. J., Emission of aliphatic amines from animal husbandry and theirreactions: Potential source of N2O and HCN, J. Atmos. Chem., 22(3), 319–346, 1995.

Sedo, G., Schultz, J., and Leopold, K. R., Electric dipole moment of sulfuric acid from fouriertransform microwave spectroscopy, J. Mol. Spectrosc., 251(1–2), 4–8, 2008.

Seinfeld, J. H. and Pandis, S. N., Atmospheric chemistry and physics: From air pollution to climatechange, John Wiley & Sons, 2nd edition, 2006.

Servant, J., Kouadio, G., Cros, B., and Delmas, R., Carboxylic monoacids in the air of Mayombeforest (Congo): Role of the forest as a source or sink, J. Atmos. Chem., 12, 367–380, 1991.

Sihto, S.-L., Kulmala, M., Kerminen, V.-M., Dal Maso, M., Petaja, T., Riipinen, I., Korhonen, H.,Arnold, F., Janson, R., Boy, M., Laaksonen, A., and Lehtinen, K. E. J., Atmospheric sulphuric acidand aerosol formation: implications from atmospheric measurements for nucleation andearly growth mechanisms, Atmos. Chem. Phys., 6(12), 4079–4091, 2006.

Sipila, M., Berndt, T., Petaja, T., Brus, D., Vanhanen, J., Stratmann, F., Patokoski, J., Mauldin, R. L.,Hyvarinen, A., Lihavainen, H., and Kulmala, M., The role of sulfuric acid in atmospheric nucle-ation, Science, 327(5970), 1243–1246, 2010.

Smith, J. N., Barsanti, K. C., Friedli, H. R., Ehn, M., Kulmala, M., Collins, D. R., Scheckman, J. H.,Williams, B. J., and McMurry, P. H., Observations of aminium salts in atmospheric nanoparti-cles and possible climatic implications, Proc. Natl. Acad. Sci. USA, 107(15), 6634–6639, 2010.

Stein, S. E. and Brown, R. L., Estimation of normal boiling points from group contributions, J.Chem. Inf. Comput. Sci., 34(3), 581–587, 1994.

Su, T. and Bowers, M. T., Theory of ion-polar molecule collisions. comparison with experimen-tal charge transfer reactions of rare gas ions to geometric isomers of difluorobenzene anddichloroethylene, J. Chem. Phys., 58(7), 3027–3037, 1973.

96

References

Szmigielski, R., Surratt, J. D., Gomez-Gonzalez, Y., Van der Veken, P., Kourtchev, I., Vermeylen,R., Blockhuys, F., Jaoui, M., Kleindienst, T. E., Lewandowski, M., Offenberg, J. H., Edney, E. O.,Seinfeld, J. H., Maenhaut, W., and Claeys, M., 3-Methyl-1,2,3-butanetricarboxylic acid: An at-mospheric tracer for terpene secondary organic aerosol, Geophys. Res. Lett., 34, L24811, 2007.

Takeuchi, M., Li, J., Morris, K. J., and Dasgupta, P. K., Membrane-based parallel plate denuderfor the collection and removal of soluble atmospheric gases, Anal. Chem., 76(4), 1204–1210,2004.

Takeuchi, M., Ullah, S. M. R., Dasgupta, P. K., Collins, D. R., and Williams, A., Continuous collectionof soluble atmospheric particles with a wetted hydrophilic filter, Anal. Chem., 77(24), 8031–8040, 2005.

Tolocka, M. P., Jang, M., Ginter, J. M., Cox, F. J., Kamens, R. M., and Johnston, M. V., Formation ofoligomers in secondary organic aerosol, Environ. Sci. Technol., 38(5), 1428–1434, 2004.

Tuazon, E. C., Atkinson, R., Aschmann, S. M., and Arey, J., Kinetics and products of the gas-phasereactions of O3 with amines and related compounds, Res. Chem. Intermed., 20, 303–320, 1994.

Usoskin, I., Gladysheva, O., and Kovaltsov, G., Cosmic ray-induced ionization in the atmosphere:spatial and temporal changes, J. Atmos. Sol.-Terr. Phy., 66(18), 1791–1796, 2004.

VandenBoer, T. C., Petroff, A., Markovic, M. Z., and Murphy, J. G., Size distribution of alkyl aminesin continental particulate matter and their online detection in the gas and particle phase,Atmos. Chem. Phys., 11, 4319–4332, 2011.

Vanhanen, J., Mikkila, J., Lehtipalo, K., Sipila, M., Manninen, H. E., Siivola, E., Petaja, T., and Kul-mala, M., Particle size magnifier for nano-CN detection, Aerosol Sci. Technol., 45(4), 533–542,2011.

Vereecken, L. and Peeters, J., Decomposition of substituted alkoxy radicals—Part I: A gener-alized structure–activity relationship for reaction barrier heights, Phys. Chem. Chem. Phys.,11(40), 9062–9074, 2009.

Vereecken, L. and Peeters, J., A structure-activity relationship for the rate coefficient of H-migration in substituted alkoxy radicals, Phys. Chem. Chem. Phys., 12(39), 12608–12620, 2010.

Voigtlander, J., Duplissy, J., Rondo, L., Kurten, A., and Stratmann, F., Numerical simulations ofmixing conditions and aerosol dynamics in the CERN CLOUD chamber, Atmos. Chem. Phys.,12(4), 2205–2214, 2012.

Wallace, J. M. and Hobbs, P. V., Atmospheric science: An introductory survey, Academic Press, 2ndedition, 2006.

Wang, S. C. and Flagan, R. C., Scanning electrical mobility spectrometer, Aerosol Sci. Technol.,13(2), 230–240, 1990.

Weber, R. J., McMurry, P. H., III, R. L. M., Tanner, D. J., Eisele, F. L., Clarke, A. D., and Kapustin, V. N.,New particle formation in the remote troposphere: A comparison of observations at varioussites, Geophys. Res. Lett., 26(3), 307–310, 1999.

97

References

Weinheimer, A. J., Chemical methods: Chemiluminscence, chemical amplification, electro-chemistry, and derivation, in Analytical Techniques for Atmospheric Measurement, pages 311–360, Blackwell Publishing, 2007.

Williams, J., Mass spectrometric methods for atmospheric trace gases, in Analytical Techniquesfor Atmospheric Measurement, pages 229–264, Blackwell Publishing, 2007.

Wyche, K. P., Monks, P. S., Ellis, A. M., Cordell, R. L., Parker, A. E., Whyte, C., Metzger, A., Dom-men, J., Duplissy, J., Prevot, A. S. H., Baltensperger, U., Rickard, A. R., and Wulfert, F., Gasphase precursors to anthropogenic secondary organic aerosol: detailed observations of 1,3,5-trimethylbenzene photooxidation, Atmos. Chem. Phys., 9(2), 635–665, 2009.

Yu, F. and Turco, R. P., The size-dependent charge fraction of sub-3-nm particles as a key diag-nostic of competitive nucleation mechanisms under atmospheric conditions, Atmos. Chem.Phys., 11(18), 9451–9463, 2011.

Yu, H. and Lee, S.-H., Chemical ionisation mass spectrometry for the measurement of atmo-spheric amines, Environ. Chem., 9(3), 190–201, 2012.

Yu, H., McGraw, R., and Lee, S.-H., Effects of amines on formation of sub-3 nm particles andtheir subsequent growth, Geophys. Res. Lett., 39, L02807, 2012.

Zhao, J., Smith, J. N., Eisele, F. L., Chen, M., Kuang, C., and McMurry, P. H., Observation of neutralsulfuric acid-amine containing clusters in laboratory and ambient measurements, Atmos.Chem. Phys., 11(21), 10823–10836, 2011.

Zhu, L., Bozzelli, J. W., and Kardos, L. M., Thermochemical properties, ∆fH°(298), S°(298), andCp°(T ), for n-butyl and n-pentyl hydroperoxides and the alkyl and peroxy radicals, transitionstates, and kinetics for intramolecular hydrogen shift reactions of the peroxy radicals, J. Phys.Chem. A, 111(28), 6361–6377, 2007.

Zollner, J. H., Glasoe, W. A., Panta, B., Carlson, K. K., McMurry, P. H., and Hanson, D. R., Sulfuricacid nucleation: power dependencies, variation with relative humidity, and effect of bases,Atmos. Chem. Phys., 12(10), 4399–4411, 2012.

Zador, J., Turanyi, T., Wirtz, K., and Pilling, M. J., Measurement and investigation of chamber rad-ical sources in the european photoreactor (EUPHORE), J. Atmos. Chem., 55(2), 147–166, 2006.

98

ASupplementary material

A.1. CLOUD experiment at CERN

The key features of the CLOUD experiment (Fig. 1a) are a large volume (26 m3) stainlesssteel chamber; precise (±0.01 K) temperature control at any tropospheric temperature; pre-cise delivery of selected trace gases and ultra-pure humidified synthetic air; precise and uni-form adjustment of the H2SO4 concentration by means of ultra-violet (UV) illumination froma fibre-optic system; suppression of contaminant vapours at the technological limit; an ad-justable π+ beam from the CERN Proton Synchrotron (PS) to simulate cosmic rays; and theability to simulate an ion-free environment by applying an electric field to sweep ions fromthe chamber. A comprehensive array of state-of-the-art instruments continuously samplesand analyses the contents of the chamber. For the results reported here, the instrumentsincluded a chemical ionisation mass spectrometer (CIMS) for H2SO4 concentration (Kurtenet al., 2011), two atmospheric pressure interface time-of-flight (APi-TOF) mass spectrometersfor molecular composition of positive and negative charged clusters (Junninen et al., 2010), sev-eral condensation particle counters (CPCs) with 50 % detection efficiency thresholds near 2 nm(two Airmodus A09 particle size magnifiers, PSM (Vanhanen et al., 2011), two diethylene glycolCPCs, DEG-CPC (Iida et al., 2009), a TSI 3776 and TSI 3786 CPC), a scanning mobility particlesizer (SMPS), a neutral cluster and air ion spectrometer (NAIS) (Kulmala et al., 2007), a proton-transfer-reaction time-of-flight (PTR-TOF) mass spectrometer for organic vapours (Graus et al.,2010), and an ion chromatograph (IC) for measurements of NH3 and dimethylamine (DMA)concentrations (Praplan et al., 2012). Two particle counters were operated in a continuously-stepped scanning mode to provide measurements of particle growth rates at small sizes: a) aPSM whose detection threshold was varied between about 1 and 2.5 nm, and b) the TSI 3786with a variable laminar flow rate through its sampling probe producing a cutoff size betweenabout 2.5 and 5 nm.

99

Appendix A. Supplementary material

Figure 1a.: Schematic diagram of the CLOUD experiment at the CERN Proton Synchrotron.

A.2. Determination of the nucleation and growth rates

The nucleation rates are obtained from the formation rates, dNd th/dt (where the subscriptdth refers to the detection threshold diameter of the particle counter). The nucleation rates,J1.7, are determined at 1.7 nm mobility diameter after correcting for losses between 1.7 nm andthe detection size threshold (Kerminen and Kulmala, 2002; Kulmala and Kerminen, 2008). A di-ameter of 1.7 nm corresponds to a cluster considered to be above the critical size and there-fore thermodynamically stable. The critical size, which corresponds to equal evaporation andgrowth rates of the cluster, varies with temperature, chemical species and concentrations, andmay even be absent when evaporation rates are extremely low as in the case of DMA-sulphuricacid clusters. Since the loss rate of particles to the chamber walls is comparable to the rate atwhich particles are lost in the atmosphere to background aerosols under pristine boundarylayer conditions, the reported formation rates at 1.7 nm size should correspond reasonably wellto atmospheric observations of new particle formation.

Before calculating J1.7 the measured particle number concentrations vs. time are cor-rected in a two-step process (Kerminen and Kulmala, 2002; Kulmala and Kerminen, 2008):(1) for the loss of particles above dth due to wall loss, dilution and coagulation, and (2) for thesame loss processes during the growth from 1.7 nm to dth. The wall loss rate is 1.5·10−3 s−1 forH2SO4 monomers and decreases with increasing cluster diameter as 1/d . The dilution lifetime

100

A.3. Cluster evaporation and fragmentation in the APi-TOF

is 3–5 hours, depending on the total sampling rate of all instruments attached to the cham-ber. The second correction above requires knowledge of the particle growth rate (GR). This isexperimentally determined from the different rise times measured in a scanning PSM, whichdetects particles over a range of threshold diameters between 1 and 2.5 nm. The GRs were veri-fied with several other instruments, including a fixed-threshold PSM, two DEG-CPCs, a TSI 3776,two APi-TOFs, an NAIS and an SMPS. Since instrumentally-determined GRs were not availablefor all runs, a parameterisation was derived to allow the growth rate to be calculated for everyexperimental run.

The detection thresholds of the particle counters do not represent perfect step functions,so particles with smaller diameters are detected to some extent, leading to an over-estimationof the nucleation rate. For this reason, the nucleation rates reported here are based on a TSI3776 CPC with dth=3.2 nm since that instrument is least sensitive to detection of particles be-low 1.7 nm. In order to verify the nucleation rates obtained with the 3.2 nm CPC, they wereindependently derived from the other CPCs with lower detection thresholds.

Neutral nucleation rates are measured with no pion beam and with the field cage elec-trodes set to ±30 kV. This completely suppresses ion-induced nucleation since, under theseconditions, small ions or molecular clusters are swept from the chamber in about 1 s. Since allof the nucleation processes under consideration take place on substantially longer time scales,neutral nucleation rates can be measured with zero background from ion-induced nucleation.

The ion-induced nucleation rate, Jiin, for positive and negative particles is measured withthe NAIS. This provides the most sensitive determination of the ion-induced fraction, Jiin/Jtotalsince the NAIS measures only charged clusters. Loss corrections are applied to the chargedcluster spectra to account for wall losses, dilution and ion-ion recombination. In addition asource correction is applied to account for diffusion charging of neutral clusters by small ions.The latter correction requires knowledge of the number concentrations of small ions and ofneutral clusters vs. particle diameter. The neutral cluster concentrations are measured withthe 3.2 nm TSI 3776 CPC and their size spectra are measured with the SMPS. The small ionconcentrations are measured with the AIS+ and AIS−. The charging (collision) probabilities aredetermined using the collision kernels vs. diameter from Laakso et al. (2004).

A.3. Cluster evaporation and fragmentation in the APi-TOF

Evaporation (loss of single molecules) and fragmentation (fission) of clusters inside theAPi-TOF is not yet well characterised in this newly-developed instrument. Nevertheless we canmake the following observations.

The entry section of the APi-TOF is characterised by several chambers at successively lowerpressures and the presence of an extraction voltage that accelerates the cluster ions at a re-duced pressure, thereby increasing their mean free path and kinetic energy. Therefore, whencharged clusters enter the APi-TOF, they are rapidly cooled by adiabatic expansion of the airand, at the same time, accelerated in the electric fields and heated by collisions with air mo-le-cules.

Although these conditions are different from the equilibrium conditions that exist when

101


the cluster is in the CLOUD chamber, it is instructive to consider some simple estimates that ig-nore collision heating in the APi-TOF. The time interval between a cluster entering the APi-TOFand its detection is a few µs. This implies that molecules that evaporate from the cluster at arate above about 105 s−1 will not be detected, whereas molecules that evaporate at a slowerrate will be detected. For the high-evaporation-rate molecules, X, to be in equilibrium on thecluster inside the CLOUD chamber, the X arrival rate must be at least 105s−1. This sets a min-imum vapour concentration in the chamber, [X] = 105/k = 3×1013 cm−3 (∼1 ppmv), where k =3×10−9 cm3 s−1 is the charged-neutral collision kernel.

Under these assumptions, the only molecules that could be present on the charged clus-ters in the CLOUD chamber and yet escape detection in the APi-TOF must be present as avapour with a mixing ratio of at least 1 ppmv. In the CLOUD experiments the only such vapouris water (mixing ratio ∼1 %). Other molecules such as H2SO4 and DMA—present at the partper trillion by volume (pptv) level—evaporate so slowly that there would be negligible evapo-ration losses in the APi-TOF. Our theoretical calculations, summarised in Table A.1, indicate thatthe charged DMA-SA clusters have a large negative Gibbs free energy of formation, so are notexpected to fragment in the APi-TOF.

These observations suggest that water can be expected to evaporate rapidly from the clus-ters in the APi-TOF, but not H2SO4 or DMA. The experimentally observed spectra (Fig. 4.3)support this picture. Firstly, no H2O molecules are found on the clusters. Secondly, largeDMA·H2SO4 clusters that contain more than 10 acid-base pairs are observed in the APi-TOF,with no evidence for strong fragmentation (which would produce pileup at low cluster masses).

Nevertheless, these are only qualitative observations and it is likely that cluster evapora-tion and fragmentation in the APi-TOF is quite sensitive to several variables, including ambienttemperature, cluster type (binary, NH3 ternary, DMA ternary, etc.) and accelerating voltagesettings. These processes are presently under active study and will be reported in a futurepublication.

A.4. Experimental errors

The error on J1.7 has three components that are added together in quadrature to esti-mate the total error indicated in Figs. 4.1 and 4.2a. The first is a statistical measurement errorderived from the scatter of the particle counter measurements, evaluated separately for eachnucleation event; the second is an estimated +50%/-33 % uncertainty on the calculated cor-rection factor, J1.7/Jd th, where Jd th is the nucleation rate at size dth, obtained after correctingdNd th/dt for detection losses. The third is a ± 30% systematic uncertainty on Jd th estimatedfrom the run-to-run reproducibility of dNd th/dt under nominally identical chamber conditions.

The error on Jiin has two main components. The first is a statistical measurement errorderived from the scatter of the NAIS measurements, evaluated separately for each nucleationevent. The second is an estimated± 50 % error to account for the uncertainty in the correctionfor diffusion charging of neutral clusters by small ions. These errors are added together inquadrature with the error on J1.7 to estimate the error on the ion-induced fraction, Jiin/J1.7,shown in Fig. 4.2b.

102

A.5. ACDC model

The overall experimental uncertainty on [H2SO4] measured by the CIMS is estimated tobe± 35 %, based on three independent measurements: particle growth rate under conditionsof binary nucleation, the depletion rate of [SO2] by photo-oxidation, and an external calibra-tion source Kurten et al. (2012). However, the run-to-run relative experimental uncertainty on[H2SO4] is smaller, typically ± 10 %. The concentrations of SO2, O3 are measured with cali-brated instruments and are known to± 10 %. The overall uncertainty on the NH3 mixing ratiois estimated to be +100 %/-50 %. The point-to-point uncertainty on the DMA mixing ratio isestimated to be±(11 % + 12 %/[DMA](pptv)), with an overall scale uncertainty of +50 %/-33 %.

A.5. ACDC model

We used the Gibbs free energies calculated in our previous work (Ortega et al., 2012; Kupi-ainen et al., 2012) for neutral and positively charged clusters containing sulphuric acid, ammo-nia and DMA. These were then converted into evaporation rates as described by Ortega et al.(2012). We used the same method, B3LYP/CBSB7//RI-CC2/aug-cc-pV(T+d)Z, to compute the for-mation Gibbs free energies of negatively charged clusters as well as some remaining neutraland positive clusters. The geometry optimisations and frequency calculations were performedwith the Gaussian09 program (Frisch et al., 2009) using the B3LYP hybrid functional (Becke,1993) and a CBSB7 basis set (Montgomery et al., 1999). A single-point electronic energy wasthen calculated with the TURBOMOLE program (Ahlrichs et al., 1989) using the RI-CC2 method(Hattig and Weigend, 2000) and an aug-cc-pV(T+d)Z basis set (Dunning et al., 2001). Ther-modynamic parameters for all studied clusters not previously published at this level of theoryare given in Table A.1 at 298.15 K. In the actual simulations, the temperature dependence ofthe free energies was taken into account by recomputing them directly from the vibrationalfrequencies and rotational constants.

In principle, we treat all clusters with up to four sulphuric acid molecules and four basemolecules. However, in order to reduce the computational effort, we omitted some clustersthat have extremely high evaporation rates. Tables A.2–A.5 show clusters explicitly included inthe model. Incorporating water molecules in the simulations is at the moment computation-ally impossible and has not been included in the model. According to our preliminary teststhe effect of water on the simulated formation rates will be relatively small compared to othererror sources. Moreover, estimations based on classical thermodynamics suggests that thegrowing clusters contain very little—if any—water under atmospheric conditions.

The ACDC model has been described and tested in detail by McGrath et al. (2012). Forcomparison with the CLOUD measurements, which correspond to steady-state conditions, wefixed the base monomer concentrations and the total concentration of acid (see below), andran the simulation until concentrations and formation rates no longer changed with time.

The [H2SO4] value (measured as HSO−4 ions by the CIMS instruments, with a typical charg-ing efficiency on the order of 10−3) may potentially contain contributions from free H2SO4

molecules, hydrated H2SO4 molecules, H2SO4 clustered with base molecules, as well as pos-sibly H2SO4 dimers (with or without water and bases) (Kurten et al., 2011b). When bases suchas DMA are present, the concentration of free H2SO4 may be only a small fraction of the to-tal H2SO4. However, from the CLOUD measurements (Figs. 4.3a and 4.4), after charging in

103


the CIMS, all clusters containing one H2SO4 molecule should be detected as a free chargedmonomer and all clusters containing two H2SO4 molecules should be detected as a free chargeddimer—regardless of whether or not they were originally clustered with base molecules. There-fore, following the experimental convention, we have modelled the H2SO4 concentration asthe sum of the free H2SO4 monomers and all clusters containing exactly one H2SO4 moleculeand any number of DMA molecules.

Negative and positive charges are introduced into the system simulations as constantsource terms of representative primary or secondary ions, here modelled as O−2 and H3O+.When an O−2 ion (mass 32.00 u, radius 2.23 A) collides with a sulphuric acid molecule or anyneutral cluster containing sulphuric acid, it charges the molecule or cluster negatively by re-moving one proton. It can also collide with a positive ion or cluster and neutralise it by re-moving one proton. Similarly, H3O+ ions (mass 19.02 u, radius 1.96 A) can either charge basemolecules and base-containing neutral clusters positively or neutralise negatively-charged ionsand clusters by donating a proton. Finally, the O−2 and H3O+ ions can neutralise each other incollisions, or be lost to the walls of the CLOUD chamber. Every time an O−2 or H3O+ ion losesits charge in a collision, it is removed from the system. Ionic species containing acids and/orbases cannot lose their charge by colliding with neutral air molecules.

While the enhancement of ion-molecule collisions due to electric charge was set to a con-stant (size-independent) factor in McGrath et al. (2012), we used a mass- and temperature de-pendent expression from Su and Bowers (1973) (Eq 6, with parameter values recommendedin that reference). This approach was shown to agree well with experiments for collisions ofDMA with positively charged sulphuric acid/ammonia/DMA clusters (Kupiainen et al., 2012).The dipole moments and polarizabilities of all neutral molecules and clusters, computed atthe B3LYP/CBSB7 level, are presented in Table A.5. Collisions of two ions of the same polaritywere not allowed, and the recombination rate of ions or ionic clusters with opposite polaritywas set to 1.6×10−6 cm−3 s−1 (Bates, 1982). In the CLOUD chamber simulations, we used a di-lution loss 1.06×10−4 s−1 and a size-dependent wall loss Si=(1.76×10−3/dp) nm/s, where themobility diameter, dp=(dm+0.3 nm)

√1 + (28.8u/m), depends on the mass diameter, dm, the

mass of the molecule or cluster, m, and the atomic mass unit, u (Ehn et al., 2011).

As our simulations are limited to clusters with up to four acids and four bases, we have as-sumed that clusters leaving the 4×4 simulation ”box” cannot re-evaporate. ACDC simulationsshow that, for neutral clusters, the dominant route for growth within the simulation box fallsalong the diagonal. In other words, the most stable (and populous) clusters are those with asimilar number of acids and bases, and they grow either by alternating addition of acid andbase molecules or addition of clusters with one acid and one DMA or else two acids and eitherone or two DMAs. To account for this effect, we only allow neutral clusters to leave the systemif they contain at least five acids and four bases or, in the case of DMA, at least four acids andfive DMAs (as summarised below).

Our ACDC simulations show that the negative clusters grow first by adding acids to abisulphate ion, but since the cluster with one bisulphate ion and four acids is not stable (evap-oration rate γ=5.7×102 s−1) we only let negative clusters grow out of the system if they haveat least five acids (including the bisulphate ion) and one base. The growth of positive clustersstarts by first charging a base molecule that subsequently grows by colliding with monomersor DMA-acid clusters. Since the most stable and abundant positively charged clusters contain

104

A.6. Supplementary tables

more bases (including the protonated ion) than acids, we require the positive clusters to con-tain at least four acids and five bases in order to leave the simulation box.

When a collision results in a cluster that is outside the 4×4 ”box” but does not satisfythe conditions for it to leave the simulation, we force it back to the nearest boundary of thesimulation box by evaporating molecules out of it. These extra molecules are returned to thesimulation as free monomers. In summary, the simulated formation rate thus includes allcollisions (including ion-ion recombination) leading to formation of the following clusters:

neutral: (H2SO4)a·(NH3)b·((CH3)2NH)c with a=5–8 and b + c=4–8neutral: (H2SO4)a·((CH3)2NH)c with a=4–8 and c=5–8negative: (HSO−4 )·(H2SO4)a·(NH3)b·((CH3)2NH)c with a=4–7 and b + c=1–8positive: (H2SO4)a·(NH3)b·((CH3)2NH)c ·H+ with a=4–8 and b + c=5–8


Table A.1.: Electronic energies (∆Eelec ), enthalpies (∆H298K ), Gibbs free energies (∆G298K ) andentropies (∆S298K ) of formation from monomers for all clusters not previously published atthis level of theory. All values are in kcal/mol, at 298.15 K and 1 atm reference pressure.Cluster ∆Eelec ∆H298K ∆G298K ∆S298K

(kcal/mol) (kcal/mol) (kcal/mol) (cal/K·mol)

a) Negative clustersH2SO4·HSO−4 -49.08 -48.32 -34.51 -46.32(H2SO4)2·HSO−4 -78.95 -76.45 -52.13 -81.57(H2SO4)3·HSO−4 -103.92 -100.02 -64.76 -118.26(H2SO4)4·HSO−4 -125.62 -120.92 -74.55 -155.53NH3·HSO−4 -10.94 -9.10 1.23 -34.67H2SO4·NH3·HSO−4 -61.93 -58.81 -35.06 -79.64(H2SO4)2·NH3·HSO−4 -99.92 -96.71 -59.64 -124.32(H2SO4)3·NH3·HSO−4 -129.49 -123.14 -77.39 -153.45(H2SO4)2·(NH3)2·HSO−4 -119.15 -111.62 -67.32 -148.59(H2SO4)3·(NH3)2·HSO−4 -157.37 -149.17 -90.57 -196.55(H2SO4)3·(NH3)3·HSO−4 -180.90 -168.72 -100.60 -228.46(CH3)2NH·HSO−4 -14.18 -12.52 -0.85 -39.13H2SO4·(CH3)2NH·HSO−4 -73.33 -69.37 -45.13 -81.29(H2SO4)2·(CH3)2NH·HSO−4 -109.61 -104.29 -68.18 -121.10(H2SO4)3·(CH3)2NH·HSO−4 -137.14 -130.91 -83.92 -157.61H2SO4·((CH3)2NH)2·HSO−4 -87.24 -82.64 -47.74 -117.08(H2SO4)2·((CH3)2NH)2·HSO−4 -139.63 -132.10 -83.27 -163.77(H2SO4)3·((CH3)2NH)2·HSO−4 -176.80 -167.93 -107.39 -203.04(H2SO4)2·((CH3)2NH)3·HSO−4 -161.44 -150.39 -91.58 -197.26(H2SO4)3·((CH3)2NH)3·HSO−4 -204.73 -192.99 -119.65 -245.97(H2SO4)3·((CH3)2NH)4·HSO−4 -230.44 -214.82 -131.37 -279.91(CH3)2NH·NH3·HSO−4 -24.22 -20.90 0.17 -70.70

105


Cluster (continued) ∆Eelec ∆H298K ∆G298K ∆S298K(kcal/mol) (kcal/mol) (kcal/mol) (cal/K·mol)

H2SO4·(CH3)2NH·NH3·HSO−4 -82.03 -76.21 -43.10 -111.06(H2SO4)2·(CH3)2NH·NH3·HSO−4 -130.67 -124.00 -75.86 -161.46(H2SO4)3·(CH3)2NH·NH3·HSO−4 -164.82 -155.65 -96.72 -197.63(H2SO4)2·(CH3)2NH·(NH3)2·HSO−4 -144.87 -136.67 -78.92 -193.70(H2SO4)3·(CH3)2NH·(NH3)2·HSO−4 -190.15 -178.07 -108.19 -234.38(H2SO4)2·((CH3)2NH)2·NH3·HSO−4 -147.41 -137.34 -80.47 -190.73(H2SO4)3·((CH3)2NH)2·NH3·HSO−4 -201.82 -188.88 -117.01 -241.06

b) Neutral clusters with DMA and NH3

(H2SO4)4·(NH3)3 † -156.25 -145.17 -78.88 -222.33H2SO4·(CH3)2NH -27.22 -24.65 -15.40 -31.01(H2SO4)2·((CH3)2NH)2 ‡ -95.06 -87.57 -54.26 -111.73(CH3)2NH·NH3 -3.83 -2.43 2.99 -18.20(H2SO4)4·(CH3)2NH·NH3 -136.94 -127.53 -72.25 -185.38(H2SO4)4·(CH3)2NH·(NH3)2 -167.61 -155.34 -88.71 -223.48(H2SO4)4·((CH3)2NH)2·NH3 -181.47 -168.67 -99.70 -231.34(H2SO4)2·(CH3)2NH·(NH3)3 -105.64 -96.19 -44.70 -172.68(H2SO4)3·(CH3)2NH·(NH3)3 -160.06 -147.10 -79.59 -226.43(H2SO4)4·(CH3)2NH·(NH3)3 -191.12 -176.59 -100.49 -255.24(H2SO4)2·((CH3)2NH)2·(NH3)2 -117.27 -106.71 -55.31 -172.39(H2SO4)3·((CH3)2NH)2·(NH3)2 -172.37 -158.53 -89.85 -230.37(H2SO4)4·((CH3)2NH)2·(NH3)2 -208.57 -192.41 -112.12 -269.31(H2SO4)2·((CH3)3NH)2·NH3 -133.88 -122.51 -66.79 -186.88(H2SO4)3·((CH3)2NH)3·NH3 -181.90 -167.40 -97.25 -235.26(H2SO4)4·((CH3)2NH)3·NH3 -213.15 -196.36 -116.61 -267.49

c) Positive clusters(CH3)2NH+

2 + NH3 (CH3)2NH + NH+4 18.67 18.25 18.54 -0.98

(NH3)2·NH+4 -46.54 -43.94 -29.38 -48.81

H2SO4·(NH3)2·NH+4 -77.48 -73.76 -48.07 -86.14

(H2SO4)3·(NH3)2·NH+4 -141.94 -133.67 -84.26 -165.72

(H2SO4)4·(NH3)2·NH+4 -166.03 -156.00 -93.64 -209.15

(H2SO4)2·(NH3)3·NH+4 -138.34 -130.13 -81.77 -162.20

(H2SO4)3·(NH3)3·NH+4 -175.50 -164.31 -103.27 -204.72

(H2SO4)4·(NH3)3·NH+4 -199.17 -186.85 -114.94 -241.17

(CH3)2NH·(CH3)2NH+2† -26.33 -25.68 -16.89 -29.45

((CH3)2NH)2·(CH3)2NH+2 -46.41 -43.77 -25.80 -60.28

H2SO4·((CH3)2NH)2·(CH3)2NH+2 -98.15 -91.60 -61.05 -102.45

(H2SO4)3·((CH3)2NH)2·(CH3)2NH+2 -164.50 -154.59 -99.82 -183.70

(H2SO4)4·((CH3)2NH)2·(CH3)2NH+2 -193.82 -182.75 -115.35 -226.07

(H2SO4)2·((CH3)2NH)3·(CH3)2NH+2 -171.25 -160.17 -104.95 -185.20

(H2SO4)3·((CH3)2NH)3·(CH3)2NH+2 -210.29 -196.67 -130.14 -223.16

(H2SO4)4·((CH3)3NH)3·(CH3)2NH+2 -229.88 -215.90 -139.67 -255.68

(NH3)2·(CH3)2NH+2 -40.12 -37.19 -21.67 -52.05

NH3·(CH3)2NH·(CH3)2NH+2 -43.27 -40.52 -23.83 -55.98

106


Cluster (continued) ∆Eelec ∆H298K ∆G298K ∆S298K(kcal/mol) (kcal/mol) (kcal/mol) (cal/K·mol)

(H2SO4)3·(NH3)2·(CH3)2NH+2 -140.04 -131.18 -78.81 -175.65

(H2SO4)3·NH3·(CH3)2NH·(CH3)2NH+2 -152.75 -143.09 -89.88 -178.44

(H2SO4)3·(NH3)3·(CH3)2NH+2 -169.48 -158.07 -96.05 -208.02

(H2SO4)4·(NH3)3·(CH3)2NH+2 -194.89 -182.24 -108.82 -246.25

H2SO4·(NH3)2·(CH3)2NH·(CH3)2NH+2 -102.18 -95.14 -57.93 -124.80

(H2SO4)3·(NH3)2·(CH3)2NH·(CH3)2NH+2 -182.26 -170.06 -107.74 -209.04

(H2SO4)4·(NH3)2·(CH3)2NH·(CH3)2NH+2 -210.17 -196.84 -121.66 -252.15

(H2SO4)3·NH3·((CH3)2NH)2·(CH3)2NH+2 -195.81 -182.87 -119.13 -213.78

(H2SO4)4·NH3·((CH3)2NH)2·(CH3)2NH+2 -220.37 -206.83 -130.32 -256.61

† Better configuration than that of our previous work (Ortega et al., 2012; Kupiainen et al., 2012).‡ Different configuration than that of our previous work (Ortega et al., 2012); the earlier valuewas incorrect.

Table A.2.: Summary of neutral clusters explicitly simulated in the ACDC model. NH3 indicatesammonia-containing, DMA indicates DMA-containing and M indicates mixed clusters (withboth types of bases, all combinations yielding the indicated number of bases are included inthe model), pure sulphuric acid clusters are also included.

1 base 2 bases 3 bases 4 bases0 acids NH3 DMA M NH3 DMA NH3 DMAH2SO4 NH3 DMA NH3 DMA M NH3 DMA M NH3 DMA

(H2SO4)2 NH3 DMA NH3 DMA M NH3 DMA M NH3 DMA M(H2SO4)3 NH3 DMA NH3 DMA M NH3 DMA M NH3 DMA M(H2SO4)4 NH3 DMA NH3 DMA M NH3 DMA M NH3 DMA M

Table A.3.: Summary of the negatively charged clusters explicitly simulated in the ACDC model.NH3 indicates ammonia-containing, DMA indicates DMA-containing and M indicates mixedclusters (with both types of bases, all combinations yielding the indicated number of bases areincluded in the model), negatively charged pure sulphuric acid cluster are also included.

1 base 2 bases 3 bases 4 basesHSO−4 NH3 DMA

(H2SO4)·HSO−4 NH3 DMA DMA M(H2SO4)2·HSO−4 NH3 DMA NH3 DMA M DMA M(H2SO4)3·HSO−4 NH3 DMA NH3 DMA M NH3 DMA M DMA

107


Table A.4.: Summary of the positively charged clusters explicitly simulated in the ACDC model.NH3 indicates ammonia-containing, DMA indicates DMA-containing and M indicates mixedclusters (with both types of bases, all combinations yielding the indicated number of bases areincluded in the model), positively charged pure sulphuric acid clusters are not included.

H+ + 1 base H+ + 2 bases H+ + 3 bases H+ + 4 bases0 acid NH3 DMA NH3 DMA M NH3 DMA MH2SO4 NH3 DMA NH3 DMA M NH3 DMA M

(H2SO4)2 NH3 DMA NH3 DMA M NH3 DMA M NH3 DMA M(H2SO4)3 NH3 DMA M NH3 DMA M(H2SO4)4 NH3 DMA NH3 DMA M

Table A.5.: Dipole moments and polarizabilities of all studied clusters at 298 K. For moleculeswe used experimental values, except for the sulphuric acid polarizability, for which only a the-oretical value is available, and for clusters we used values calculated for the minimum-free en-ergy clusters (at the corresponding temperature) with the Gaussian09 program (Frisch et al.,2009) using the B3LYP hybrid functional (Becke, 1993) and a CBSB7 basis set (Montgomery et al.,1999).

Dipole moment (D) Polarizability (A3)

H2SO4 2.96a 6.2b(H2SO4)2 0.002 9.061(H2SO4)3 3.692 13.710(H2SO4)4 3.252 18.759

NH3 1.47c 2.81cH2SO4·NH3 5.259 6.073(H2SO4)2·NH3 9.309 10.733(H2SO4)3·NH3 4.580 15.482(H2SO4)4·NH3 8.006 20.171

(NH3)2 0.000 3.106H2SO4·(NH3)2 3.990 7.826(H2SO4)2·(NH3)2 6.762 12.455(H2SO4)3·(NH3)2 10.065 17.121(H2SO4)4·(NH3)2 9.769 21.828

(NH3)3 0.000 4.964H2SO4·(NH3)3 7.430 9.633(H2SO4)2·(NH3)3 4.006 14.184(H2SO4)3·(NH3)3 7.700 18.719(H2SO4)4·(NH3)3 † 7.163 23.637

(NH3)4 0.000 6.797H2SO4·(NH3)4 5.798 11.458

108


(continued) Dipole moment (D) Polarizability (A3)

(H2SO4)2·(NH3)4 4.208 15.980(H2SO4)3·(NH3)4 2.150 20.630(H2SO4)4·(NH3)4 4.011 25.383

(CH3)2NH 1.01c 6.37cH2SO4·(CH3)2NH 8.755 9.367(H2SO4)2·(CH3)2NH 7.842 13.941(H2SO4)3·(CH3)2NH 9.113 18.640(H2SO4)4·(CH3)2NH 7.938 23.277

((CH3)2NH)2 0.000 10.170H2SO4·((CH3)2NH)2 6.823 14.733(H2SO4)2·((CH3)2NH)2 ‡ 5.843 18.876(H2SO4)3·((CH3)2NH)2 7.606 23.345(H2SO4)4·((CH3)2NH)2 0.411 28.419

((CH3)2NH)3 0.012 15.451H2SO4·((CH3)2NH)3 5.909 19.762(H2SO4)2·((CH3)2NH)3 7.030 24.141(H2SO4)3·((CH3)2NH)3 6.338 28.669(H2SO4)4·((CH3)2NH)3 8.378 33.305

((CH3)2NH)4 0.005 20.512H2SO4·((CH3)2NH)4 1.415 25.246(H2SO4)2·((CH3)2NH)4 9.154 29.324(H2SO4)3·((CH3)2NH)4 5.046 33.774(H2SO4)4·((CH3)2NH)4 4.802 38.239

NH3·(CH3)2NH 3.197 6.609H2SO4·NH3·(CH3)2NH 8.537 11.038(H2SO4)2·NH3·(CH3)2NH 6.154 15.666(H2SO4)3·NH3·(CH3)2NH 10.409 20.347(H2SO4)4·NH3·(CH3)2NH 7.863 25.074

H2SO4·(NH3)2·(CH3)2NH 6.823 13.062(H2SO4)2·(NH3)2·(CH3)2NH 4.132 17.499(H2SO4)3·(NH3)2·(CH3)2NH 4.861 22.072(H2SO4)4·(NH3)2·(CH3)2NH 2.349 26.891

H2SO4·NH3·((CH3)2NH)2 5.752 16.342(H2SO4)2·NH3·((CH3)2NH)2 5.184 20.657(H2SO4)3·NH3·((CH3)2NH)2 6.294 25.371(H2SO4)4·NH3·((CH3)2NH)2 3.510 29.987

(H2SO4)2·(NH3)3·(CH3)2NH 3.981 19.412

109


(continued) Dipole moment (D) Polarizability (A3)

(H2SO4)3·(NH3)3·(CH3)2NH 8.128 24.095(H2SO4)4·(NH3)3·(CH3)2NH 6.897 28.610

(H2SO4)2·(NH3)2·((CH3)2NH)2 8.824 22.457(H2SO4)3·(NH3)2·((CH3)2NH)2 9.046 27.355(H2SO4)4·(NH3)2·((CH3)2NH)2 3.496 31.680

(H2SO4)2·NH3·((CH3)2NH)3 5.966 25.872(H2SO4)3·NH3·((CH3)2NH)3 8.487 30.550(H2SO4)2·NH3·((CH3)2NH)2 3.140 35.001a Nadykto and Yu (2003).b Better configuration than that of our previous work (Ortega et al., 2012; Kupiainen et al., 2012).c Haynes and Lide (2010).† Sedo et al. (2008).‡ Different configuration than that of our previous work (Ortega et al., 2012); the earlier valuewas incorrect.

110

Documents

Research Collection · 2020. 3. 26. · CMC que les trois acides organiques precit´ es peuvent´ etre formˆ es´ a partir de l’acide pinonique.` Un quatrieme acide organique avec