The Formation of Highly Oxidized Multifunctional Products in the

Subscriber access provided by UNIVERSITY OF SASKATCHEWAN LIBRARY

Article

The formation of highly oxidized multifunctional products in the ozonolysis of cyclohexene Matti P. Rissanen, Theo Kurtén, Mikko Sipilä, Joel A. Thornton, Juha Kangasluoma, Nina Sarnela, Heikki Junninen, Solvejg Jørgensen, Simon Schallhart, Maija K. Kajos, Risto Taipale, Monika Springer, Thomas F. Mentel, Taina Ruuskanen, Tuukka Petäjä, Douglas R. Worsnop, Henrik G. Kjaergaard, and Mikael Ehn J. Am. Chem. Soc., Just Accepted Manuscript • DOI: 10.1021/ja507146s • Publication Date (Web): 06 Oct 2014 Downloaded from http://pubs.acs.org on October 11, 2014

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a free service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are accessible to all readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

Journal of the American Chemical Society is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 24

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Journal of the American Chemical Society

1

The formation of highly oxidized multifunctional products in the ozonolysis of

2

cyclohexene

3

Matti P. Rissanen*1, Theo Kurtén 2, Mikko Sipilä1, Joel A. Thornton3, Juha Kangasluoma1, Nina Sarnela1, Heikki

4

Junninen1, Solvejg Jørgensen 4, Simon Schallhart1, Maija K. Kajos1, Risto Taipale1, Monika Springer5, Thomas F.

5

Mentel5, Taina Ruuskanen1, Tuukka Petäjä1, Douglas R. Worsnop1, 6, Henrik G. Kjaergaard4 and Mikael Ehn1

6

1

Department of Physics, P.O. Box 64, 00014 University of Helsinki, Finland.

7

2

Department of Chemistry, P.O. Box 55, 00014 University of Helsinki, Finland.

8

3

Department of Atmospheric Sciences, University of Washington, Seattle, WA 98195, USA.

9

4

Department of Chemistry, University of Copenhagen, Universitetsparken 5, 2100 Copenhagen Ø, Denmark.

10

5

Institute for Energy and Climate Research (IEK-8), Forschungszentrum Jülich, 52425 Jülich, Germany.

11

6

Aerodyne Research Inc., 45 Manning Road, Billerica, MA 01821, USA.

12 13

Abstract

14

The prompt formation of highly oxidized organic compounds in the ozonolysis of cyclohexene (C 6H10) was

15

investigated by means of laboratory experiments together with quantum chemical calculations. The

16

experiments were performed in borosilicate glass flow tube reactors coupled to a Chemical Ionization

17

Atmospheric Pressure interface Time-of-Flight mass spectrometer (CI-APi-TOF) with a nitrate ion (NO 3-)

18

based ionization scheme. Quantum chemical calculations were performed at the CCSD(T)-F12a/VDZ-

19

F12//ωB97XD/aug-cc-pVTZ level, with kinetic modeling using multiconformer transition state theory (MC-

20

TST), including Eckart tunneling corrections. The complementary investigation methods gave a consistent

21

picture of a formation mechanism advancing by peroxy radical (RO 2) isomerization through intramolecular

22

hydrogen shift reactions, followed by sequential O2 addition steps, i.e., RO2 autoxidation, on a timescale of

23

seconds. Dimerization of the peroxy radicals by recombination and cross-combination reactions is in

24

competition with the formation of highly oxidized monomer species and is observed to lead to peroxides,

25

potentially diacyl peroxides. The molar yield of these highly oxidized products (having O/C > 1 in

26

monomers and O/C > 0.55 in dimers) from cyclohexene ozonolysis was determined as (4.5 ± 3.8)%. Fully

27

deuterated cyclohexene and cis-6-nonenal ozonolysis, as well as the influence of water addition to the system

28

(either H2O or D2O), were also investigated in order to strengthen the arguments on the proposed mechanism.

29

Deuterated cyclohexene ozonolysis resulted in a less oxidized product distribution with a lower yield of

30

highly oxygenated products and cis-6-nonenal ozonolysis generated the same monomer product distribution,

31

consistent with the proposed mechanism and in agreement with quantum chemical modeling.

32

Keywords: Cyclohexene ozonolysis, ELVOC, Atmospheric oxidation, Ambient aerosol precursors

33 1

ACS Paragon Plus Environment


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

34

Page 2 of 24

1. Introduction

35

Understanding the gas-phase oxidation of hydrocarbons is a fundamental prerequisite for controlling and

36

optimizing many important physicochemical processes. For example, the amount of utilizable energy from

37

combustion of a fossil fuel, or any other organic compound, is ultimately governed by the amount of energy

38

released by breaking its carbon-carbon (C−C) and carbon-hydrogen (C−H) bonds while oxidizing the carbon

39

to CO2. In the atmosphere, the oxidation of biogenic and anthropogenic hydrocarbon emissions can lead to

40

the formation and growth of aerosol particles by gas-to-particle conversion of the lower volatility products,

41

and thus influence climate, ecosystem function, and human health 1-13.

42

Autoxidation of organic compounds by peroxy radical isomerization through intramolecular hydrogen-shift

43

reactions is a well-known reaction pathway in low-temperature combustion chemistry 14-16, and is known to

44

occur in various other environments too17-25. Recently, the autoxidation phenomenon was proposed to be a

45

significant route for gas-phase oxidation of volatile organic compounds (VOCs) in the ambient atmosphere

46

in low-NOx conditions, especially for VOCs related to the formation of secondary organic aerosol 19, 20.

47

Ehn et al.12 reported a large source of low-volatility secondary organic aerosol (SOA) generated from the

48

ozonolysis of a-pinene and other endocyclic monoterpenes. Detection of highly oxygenated and high

49

molecular weight products in the gas-phase during ozonolysis under atmospheric conditions explained a

50

significant fraction of the condensed-phase mass. Also Zhao et al. 26 reached into similar conclusions in their

51

chamber investigation of a-pinene ozonolysis system. These Extremely Low Volatility Organic Compounds

52

(ELVOCs9, 12, 27) were previously detected as natural ions measured at a boreal forest measurement station

53

(SMEAR II) in Hyytiälä, Finland28, but the formation mechanisms were unknown. Ehn et al. 12 proposed,

54

based on the behavior of several of the detected peroxy radical intermediates and products that the

55

mechanism of ELVOC formation was driven by RO 2 autoxidation.

56

Due to its structural similarity with abundant biogenic monoterpenes, cyclohexene (unsaturated, endocyclic

57

C6H10) has been extensively studied as a monoterpene surrogate for inferring oxidation mechanisms and

58

aerosol formation characteristics 29-57. Ozonolysis is a common removal pathway for unsaturated compounds

59

in the atmosphere, and due to the high reaction exothermicities, is known to lead to an ensemble of reactive

60

intermediates17, 39, 58, 59. From these the most attention has been centered on stabilized Criegee Intermediates

61

(sCI; a Criegee Intermediate capable of reacting in a bimolecular reaction) due to their potential important

62

role as an oxidant and source of low volatility carboxylic acids and acid esters 17, 59-61. The cyclohexene

63

ozonolysis reaction, however, is not expected to lead to a sCI35, 36, but instead, is expected to isomerize

64

promptly to another reactive intermediate: the vinylhydroperoxide (VHP) 30,

65

generally susceptible to dissociation to an OH and an organic oxygenated radical. This oxygenated radical is

66

the potential precursor to a sequence of reactions leading to ELVOC species, and hence, is in the focus of

67

this work. 2


36, 39

. The VHP formed is

Page 3 of 24

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


68

In previous experiments the cyclohexene ozonolysis was shown to produce an ELVOC product distribution 12,

69

and in the present investigation we aimed to explore more completely how these highly oxidized ELVOCs

70

are formed - in a step-by-step manner. We show by detailed experimental investigation, together with high-

71

level quantum chemical calculations, how the oxygen is infused to the organic carbon structures by

72

sequential O2 addition + peroxy radical isomerization steps. The radical chain process is eventually

73

terminated by ejection of an OH or HO 2 radical, or by bimolecular reactions with other peroxy radicals,

74

forming the ELVOC product. In addition to the cyclohexene model system (unsubstituted, cyclic C 6H10), we

75

also studied the fully deuterated isotopologue (C6D10) as well as H/D-exchange reactions by adding D2O to

76

the gas flow, to further clarify the mechanism. As a final test on the general applicability of the proposed

77

mechanism, we performed cis-6-nonenal (linear CH3CH2CH=CH(CH2)4CHO) ozonolysis experiments,

78

which should produce the same VHP intermediate, and hence, the same ELVOC product distribution.

79 80

2. Experimental Section

81

2. 1. Laboratory investigations

82

ELVOC detection was performed with a CI-APi-TOF mass spectrometer using a nitrate ion (NO 3-) based

83

chemical ionization scheme where ELVOCs form adducts with NO 3-. The organic precursor concentrations

84

were retrieved by a PTR-TOF-MS (Proton Transfer Reaction Time-of-Flight Mass Spectrometer) or were

85

calculated from the measured gas flow rates. Both of the mass spectrometers and their basic measurement

86

routines have been described previously62, 63-66 and thus only a brief overview is given in the Supporting

87

Information (SI).

88

The ozonolysis reactions of C 6H10, C6D10 and C9H16O (=cis-6-nonenal) were investigated in two different

89

borosilicate glass flow tube reactors: a 205 cm long with a 4.7 cm i.d. and a 63 cm long with a 4.0 cm i.d.,

90

respectively. All experiments were performed under laminar flow conditions, at room temperature (T = 293

91

± 3 K) and at ambient pressure using nitrogen (N 2) or synthetic air (N2 and O2) as the bath gas. Ozone and

92

organic precursor concentrations, and their residence time in the reactor, were varied between different

93

experiments. Measurements were also conducted with different amounts of water (H 2O) and deuterated water

94

(D2O) in the reaction mixture. The O3 was produced from synthetic air by an ozone generator (Dasibi 1008-

95

PC) and the concentration was quantified by an ozone analyzer (Thermo Scientific Model 49). Water (H 2O

96

or D2O) was added to the gas stream by bubbling a variable part of the bath gas flow through a water

97

reservoir. The amount of water in the gas flow was not quantified further. For more details about the setup

98

and description of the different experiments performed see the SI (Figure S1 and Table S1).

99 3



1 2 3 100 4 5 6 101 7 102 8 9 103 10 11 104 12 105 13 14 106 15 107 16 17 108 18 19 109 20 21 110 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 111 39 40 112 41 42 43 113 44 45 114 46 115 47 48 116 49 117 50 51 52 118 53 54 119 55 120 56 57 121 58 122 59 60 123

Page 4 of 24

2.2. Quantum chemical computations Quantum chemical calculations were used to investigate the first three sequential hydrogen shifts for the peroxy radical formed after the VHP decomposition in the ozonolysis of cyclohexene (C 6H9O4, see Scheme 1). As discussed e.g., in Ehn et al. 12 and references therein, the first hydrogen shift very likely takes place at the carbonyl group opposite the peroxy group. The second hydrogen shift can then occur either at the remaining carbonyl (a 1,8 H-shift) or at the COOH carbon (a 1,7 H-shift), the latter leading to a termination of the radical reaction chain through OH elimination19, 20, 67. If the second H-shift takes place at the carbonyl group, a third H-shift can then occur. For this third step we studied all four possible abstraction sites (though only the fastest two were selected for further analysis as the conformational sampling of the products was extremely time-consuming).

C6H10O 3 C6H10

O3

+

O

C6H10O 3

POZ CI

O

O

O O CH

O

O

H-Shift

O

O C6H10O3 C6H9O2

VHP

OH

O

HO

CH

O

C6H9O4 O

+

O

O2

O

O O

H-Shift O

...

Scheme 1 Ozonolysis of cyclohexene produces a primary ozonide (POZ) that instantly decomposes to one of two identical Criegee intermediates (CI). The CI promptly isomerizes to a vinylhydroperoxide (VHP), which subsequently dissociates, losing an OH-radical. This creates a carbon-centered radical (C 6H9O2) which is able to add an O2 molecule to form an oxygen-centered peroxy radical (C 6H9O4), which can isomerize yet further. An initial set of conformers for each reactant, transition state and product structure were generated with the systematic conformer search algorithm of Spartan (versions 08, 12 and 14) 68, using the MMFF and Sybyl force-fields. The systematic algorithm explores all possible combinations of torsional angles, leading to over 100 000 trial structures for the larger systems in this study. Two different force-field methods were used as they predicted different hydrogen bonding patterns, and we wanted to make sure all possible conformers are included in the sampling. For some structures (the carbon-centered radical products of the second and third 4


Page 5 of 24

1 2 3 124 4 5 125 6 126 7 8 127 9 10 128 11 12 129 13 14 130 15 131 16 17 132 18 19 133 20 134 21 22 135 23 24 136 25 26 137 27 28 138 29 139 30 31 140 32 33 141 34 142 35 36 143 37 144 38 39 145 40 41 146 42 147 43 44 148 45 46 149 47 48 150 49 50 151 51 152 52 53 153 54 55 154 56 155 57 58 156 59 157 60


peroxy radical H-shifts, as well as the second CO loss transition state, see Scheme 2 below) the Sybyl force field predicted chemically unreasonable structures, e.g., with C-C•-H bond angles below 90 degrees (C• is the radical center carbon atom). In these cases, additional conformers were generated either by reoptimizing the Sybyl-generated structures using the AM1 semi-empirical method, or by generating a set of new conformers using the Monte Carlo sampling algorithm with the AM1 method if the previous approach failed. The energies of the force-field optimized conformers (which numbered between 100 and 6000 per system, as many initial combinations of torsion angles did not lead to distinct minima) were subsequently evaluated at the B3LYP/6-31+G(d) level using the Spartan program. The conformers were then ordered by the B3LYP single-point energy, and those within either 3, 4 or 5 kcal/mol of the lowest-energy structure were chosen for B3LYP/6-31+G(d) optimizations, again using the Spartan program69. The cut-off was selected to keep the number of structure optimizations reasonable (i.e., less than 300). We emphasize that whenever a cut-off lower than 5 kcal/mol was used, several hundred different conformers were still optimized at the B3LYP/631+G(d) level, providing a quite extensive conformational sampling. For sampling the transition state conformers, key parameters of the reacting groups (-C-H…O-O-) were constrained ("frozen") in the force field and B3LYP optimizations. These were the C-H and H…O and O-O distances for the hydrogen shifts, and the C-C=O distance for the CO loss reactions. The values for the parameters were determined based on an initial B3LYP/6-31+G(d) transition state optimization for a randomly chosen conformer for each transition state. Test calculations on the first H-shift transition state conformers indicate that the relevant distances and angles change very little from one conformer (of the same transition state type) to another, and the B3LYP relative energy changed by less than 0.25 kcal/mol (and typically less than 0.1 kcal/mol) upon relaxation of the constraints in a subsequent transition state optimization. Using constraints was necessary for the force-field conformational search to work, as these methods are unable to treat bond breaking and formation. Using them also in the B3LYP optimization stage allowed us to use minimization (rather than transition state search) algorithms in the conformational sampling. This significantly reduces the computational effort, and allowed us to sample a much larger number of transition state conformers. Next, the B3LYP-optimized structures were again ordered by energy, and those within 2 kcal/mol of the lowest energy structure were chosen for subsequent optimization at the ωB97XD/aug-cc-pVTZ level 70 using the Gaussian 09 program suite71, tight optimization criteria and the ultrafine integration grid. The total number of structures optimized at this level in this study was over 500. For each system (reactant, transition state or product), the conformer with the lowest zero-point corrected energy at the ωB97XD/aug-cc-pVTZ level, was then selected, and a final single-point energy calculation was carried out at the ROHF-RCCSD(T)F12a/VDZ-F12 level72,

73

, using the Molpro 2012 program74. Intrinsic Reaction Coordinate (IRC)

calculations on the lowest zero-point corrected energy transition state conformers were carried out at the 5



1 2 3 158 4 5 159 6 160 7 8 9 161 10 11 162 12 163 13 14 15 164 16 17 18 165 19 20 166 21 167 22 23 168 24 25 169 26 27 170 28 29 171 30 172 31 32 33 173 34 35 36 174 37 38 175 39 40 176 41 42 177 43 178 44 45 179 46 47 180 48 181 49 50 182 51 183 52 53 184 54 55 185 56 57 186 58 59 60 187

Page 6 of 24

ωB97XD/aug-cc-pVTZ level for at least 40 points in each direction to verify that the transition states connect the correct reactant and product systems. See section S2.1. in the SI for further validation of the conformational sampling algorithm. The rate coefficients for the forward and reverse hydrogen shift as well as the forward rate coefficients for the CO loss reactions (assumed to be irreversible) were estimated using multiconformer transition state theory (MC-TST)75. The MC-TST rate coefficient is given by

where

=Γ

exp −

and

,

,

(eq.1)

are the partition functions for the reactant, R, and the transition state, TS, respectively,

defined in eq. 2. The energy

,

(

,

) corresponds to the zero-point corrected ROHF-RCCSD(T)-

F12a/VDZ-F12//ωB97XD/aug-cc-pVTZ electronic energy for the lowest-energy conformer. The constants h, kB, R and T are the Planck constant, the Boltzmann constant, the gas constant and the temperature, respectively. The temperature is set to 298.15 K. We extracted the partition function, electronic energy and zero point vibrational energy from all the conformers optimized at the ωB97XD/aug-cc-pVTZ level. The calculations of the partition functions are based on the standard rigid rotor and harmonic oscillation approximations. The partition function is given as =∑

,

,

exp −

(eq.2)

where N is the number of conformers included, and

is the zero-point corrected energy of conformer i at

the ωB97XD/aug-cc-pVTZ level. The energy of the lowest lying conformer is denoted by

.

The quantum tunneling correction, Γ, is computed with a one-dimensional Eckart correction model76. The tunneling corrections have only been computed for the conformers (reactant, transition state and product)

with the lowest zero-point corrected energy. We have used the ROHF-RCCSD(T)-F12a/VDZ-F12 energies to compute the barrier heights for the forward and reverse barriers, and the ωB97XD/aug-cc-pVTZ imaginary frequencies of the transition states. For a comparison of conventional transition state theory (TST) and MC-TST rate coefficients and partition functions, see section S2.2. in the SI.

To assess the maximum effect of hindered rotations on the rate coefficients, we used the hindered rotor module of the Gaussian 09 program package with the McClurg model77-79, as well as the MS-Tor program80, 81

(see section S2.3. in the SI for more details).

6


Page 7 of 24

1 2 3 188 4 5 6 189 7 190 8 9 191 10 11 192 12 13 14 193 15 16 194 17 195 18 19 196 20 197 21 22 198 23 24 199 25 26 200 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 201 46 202 47 48 203 49 50 204 51 52 53 205 54 55 206 56 57 58 207 59 60 208


3. Results & Discussion For the purposes of this work, we use a practical definition of ELVOC as a very oxidized (at least six Oatoms in the structure, i.e., an O/C ratio of >1 in the monomer products, see below) and multifunctional organic compound capable of clustering with the NO3- ion. 3.1. Cyclohexene (C6H10) ELVOC spectrum The nitrate chemical ionization mass spectrum of ELVOCs produced in cyclohexene ozonolysis is shown in Figure 1a. The spectrum consist of three highly oxygenated C 6-product species, C6H8O7, C6H8O8 and C6H8O9 (i.e., monomers), below mass to charge ratio (m/z) 300 Th, separated from each others by mass of an O-atom (15.995 Th) and detected as clusters with NO 3-. Figure 1b illustrates these same peaks after addition of D 2O to the bath gas flow (more below on section 3.2.3.). Additionally, smaller peaks are seen above 300 Th and are referred to hereafter as “dimers”, owing to their elemental composition. The major peaks observed and their compositions determined are given in Table 1.

Figure 1a) Nitrate CI-APi-TOF mass spectrum showing the ELVOC peaks detected in the cyclohexene ozonolysis system. b) ELVOC peaks shift after adding D2O to the gas flow (only monomer range shown).

7



1 2 3 209 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 210 32 33 211 34 212 35 36 213 37 38 214 39 40 215 41 216 42 43 217 44 218 45 46 219 47 48 49 220 50 51 221 52 53 222 54 223 55 56 224 57 58 59 60

Page 8 of 24

Table 1 Major ELVOC species observed in the ozonolysis experiments. Cyclohexene, C6H10 C6H10O6 a, c 240.0361 b

C6H8O7 254.0154

C6H8O8 270.0103

C6H8O9 286.0052

C12H20O7 338.1093

C12H20O9 370.0991

C12H18O11 400.0733

C11H18O12 404.0682

C12H18O12 416.0676

C12H18O13 432.0631

C12H18O14 448.0580

C12H18O15 464.0529

C6D5O8H3 C6D7O8H C6D8O8 275.0417 277.0542 278.0605

C6D5H3O9 C6D7O9H C6D8O9 291.0366 293.0491 294.0554

C12D18H2O7 356.2223

C12D17HO8 369.1968

C12H20O10 386.0940

Cyclohexene-d10, C6D10 C6D8H2O6 c 248.0863

C6D6H2O7 260.0530

d

C11D15H3O10 387.1725 C12D17H3O9 387.2058

C12D16H2O10 400.1772

C11D16H2O9 372.1839

C12D15H3O12 431.1623

cis-6-nonenal, C9H16O − a

C6H8O7 254.0154

C6H8O8 270.0103

C6H8O9 286.0052

Identified elemental composition of the peak detected, and bits mass to charge ratio given in Thomson units; all the

peaks detected as clusters with NO3-. cThe only monomer peak that is not explained by the mechanisms of Schemes 1 and 2, most likely produced by OH-oxidation of cyclohexene. dLabile D atoms have been exchanged to H atoms prior to detection.

It should be emphasized that the determination of final product structures from a mass spectrum is generally ambiguous – multiple isomers could be detected at the same m/z-signal. Thus, in order to understand the nature of the highly oxidized products formed, we utilized quantum chemistry, together with a suite of further laboratory experiments, to unravel the chemical identities of the products observed. Evidence gathered from these separate experiments and calculations enabled us to propose product structures for the monomer ELVOCs detected. 3.2. Proposed cyclohexene ELVOC formation scheme We propose a formation pathway for the monomer ELVOCs detected in C 6H10 ozonolysis (Figure 1, Table 1) starting from the C6H9O4 peroxy radical (Scheme 2). This scheme is motivated using quantum chemical calculations (section 3.2.1.), supporting experiments (sections 3.2.2. to 3.2.4.), and previously suggested autoxidative formation pathways 12, 19.

8


Page 9 of 24

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 225 51 226 52 53 227 54 55 228 56 229 57 58 230 59 231 60


O O

C 6H 9O 4

O O

k r(1a)

k f (1a)

HO

HO

O

O

C 6H 9O 4

+

CO C

C 5 H 9O 3

kO2 HO

HO

C 6 H 9O 6

O

k f (3b)

O

O

O

O

HO

O

O

k f (3a)

C 6H 8 O 5

C 6 H 9O 6

HO

HO O

O

O

O

CO HO

HO HO

O O

CH

O

k C O (4B)

C 5H 9O 5

HO

HO O

O

O

O

O

HO

H-Shift HO

HO

O

C

O

O

k f (5a)

k r(5a) HO

HO

O

HO

O

+

CH

O2

O

O

CH

HO

HO

HO O

...

O

O

k O2

HO

HO O

O

O2

O

C 6H 8O 7

HO

k f (5b) k r(5b)

O

O

O O

O

C 6H 9O8

O

O

HO

O

O

C 6H 9O 8

O

C 5H 8O4

kO2

O2

C 6H 9 O 8

HO

O

C

+

O O

k r (3a)

O

C 6H 9O6

O

HO

O

k r (3b)

+

O

HO

C

O

C 6 H 9 O 10

...

H2C

k C O (2B)

+ O2

O2

O

O

O

HO

O

HO 2

HO

O

O

O HO

O

O O

O

O O

OH

O

C 6H 8O 8

OH

H-Shift HO O HO

C

O

O O

O

HO

O O O

O OH

O

HO

O

OH

O OH

OH

C 6 H 8O 9

Scheme 2 Proposed ELVOC formation pathways starting from the C6H9O4 peroxy radical in the cyclohexene ozonolysis system. The rate coefficients for the different reversible hydrogen shift reactions (labelled by kf for forward and kr for reverse rates), and for the CO loss pathways (kCO), have been obtained by quantum chemical computations and are given in Table 2. The radical chain reaction is propagated until a termination occurs by loss of OH or HO 2 (thick arrows) or by reaction with other peroxy radicals (HO2 and RO2, not shown in this scheme). The ELVOCs observed in

232

the nitrate CI-APi-TOF spectra have been marked with bold font.

9



1 2 3 233 4 5 234 6 235 7 8 9 236 10 11 237 12 13 238 14 15 239 16 240 17 18 241 19 20 242 21 243 22 23 24 244 25 245 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 10 of 24

No experimental rate coefficients were determined during these investigations. Nevertheless, the shortest residence time of 4 s in the flow tube constrains the whole pseudo-unimolecular autoxidation process to terminate in less than this time. 3.2.1.

Computational Results

The most likely H-shift reaction pathways as predicted by our quantum calculations are shown in Scheme 2. The computed reaction rate coefficients for the most important H-shifts of the C6H9O4, C6H9O6 and C6H9O8 peroxy radicals and for the CO loss reactions of the C6H9O4, C6H9O6 acylic radicals are given in Table 2. For C6H9O8, forward rates of two additional minor H-shift channels are given in the Supporting Information. Due

to computational limitations, we were unable to treat the C6H9O10 radical at the level of theory chosen for this study. The structures, energetic parameters and Cartesian co-ordinates of the lowest-energy conformers are given in section S2.4. of the SI. Table 2 Calculated MC-TST rate coefficients including Eckart tunneling corrections for the main H-shift and CO loss reactions. a kMCTST (s-1) Label

Reaction type

Forward, kf

Reverse, kr

kf(1a), kr(1a)

1,7 H-shift in C6H9O4

7.5

86

kO2

O2 addition to 1,7 H-shift

2.5×107

b

-

product of C6H9O4 kCO(2B)

CO loss from 1,7 H-shift

5.1×102

-

product of C6H9O4

kf(3a), kr(3a)


0.5

1.3×10-6

kf(3b) kr(3b)


3.8

1.5×10-9

kO2

O2 addition to 1,8 H-shift

2.5×107

b

-

product of C6H9O6 kCO(4B)

CO loss from 1,8 H-shift

3.3×106

-

product of C6H9O6

10


Page 11 of 24

1 2 3 4 5 6 7 8 246 9 10 247 11 248 12 13 249 14 15 250 16 17 251 18 252 19 20 253 21 22 254 23 24 255 25 26 256 27 257 28 29 258 30 31 259 32 260 33 34 261 35 36 262 37 38 263 39 40 264 41 265 42 43 266 44 267 45 46 268 47 48 269 49 270 50 51 271 52 53 272 54 273 55 56 57 274 58 275 59 60 276


kf(5a), kr(5a)


4.5×10-2

3.8×10-4

kf(5b), kr(5b)


0.1

1.4

a

The relative ROHF-RCCSD(T)-F12a/VDZ-F12 energies (energy barriers) between the lowest energy conformers of R,

TS and P and the ωB97XD/aug-cc-pVTZ imaginary frequency have been used for Eckart tunneling correction. Hindered rotor corrections are not included; see the SI for discussion of these. bEstimated using an O2 concentration of 0.2 atm and a literature value of 5 × 10−12 cm3 molecule−1 s−1 for the O2 addition reaction 82.

Table 2 shows that in contrast to reactions of unsubstituted peroxy radicals, the hydrogen shift reactions of cyclohexene ozonolysis products are reasonably fast, occurring on a timescale of seconds or less. This is in agreement with recent work on H-shifts in other oxygen-containing molecules 19. Also, the reactions are thermodynamically either close to thermoneutral or exothermal, leading to low reverse rate coefficients especially for the second and third peroxy radical H-shifts. The computed results can be used to compare the main unimolecular loss mechanisms. CO loss from the aldehydic carbon-centered radicals competes with irreversible O2 addition. The rate of the latter is likely close to that of the analogous CH3CO + O2 reaction, which is around 5 × 10−12 cm3 molecule−1 s−1 82. At an O2 concentration of 0.2 atm, this implies an effective unimolecular rate coefficient of around 2.5 × 10 7 s−1 for the O2 addition. Thus, CO loss from the C6H9O4 carbon-centered radical is completely negligible, and even the very rapid CO loss reaction from C6H9O6 is still an order of magnitude slower than the O 2 addition pathway. Based on the computed results, the main unimolecular loss mechanism preventing the formation of C 6compounds with O/C ratios above 1 is OH loss from the C 6H9O6 radical following an 1,7 H-shift, and leading to a C6H8O5 product. (Test calculations at the ωB97XD/aug-cc-pVTZ level indicate that the barrier for OH loss is on the order of 5 kcal/mol, implying that OH loss occurs almost immediately after the H-shift.) The predicted branching ratio for the 1,8 H-shift permitting O 2 addition and further autoxidation is only about 10%. However, the subsequent hydrogen shifts for the C 6H9O8 peroxy radical favor further oxidation, as the 1,4 H-shift leading to OH loss is predicted to be slower than the H-shifts given in Table 2 (see Table S3 in the SI for a comparison of all four forward rates). Also, the net overall rates for the two H-shifts for C6H9O8 shown in Table 2 are likely a factor of 2 higher than the numbers given, due to the fact that there are two abstractable hydrogens in each CH2 group (unlike the other H-shifts in the table). Most of the C 6H9O8 radicals will therefore likely survive to form C6H9O10 radicals, and thus yield C6H8O9 or C6H8O8 closed-shell products after loss of OH and HO2, respectively. Our calculations thus predict that significant amounts of C 6H8O5 products should be formed compared to the more highly oxidized ELVOCs.

Unfortunately, C 6H8O5is probably not detectable by NO3- chemical

ionization, as it does not bond strongly enough to the NO3- ion with its single OOH group83. 11



1 2 3 277 4 5 278 6 279 7 8 280 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 281 42 43 282 44 45 283 46 284 47 48 285 49 50 286 51 52 287 53 54 288 55 289 56 57 290 58 291 59 60 292

Page 12 of 24

All in all, the computed rate coefficients are consistent with ELVOC formation proceeding as proposed in scheme 2. The whole pseudo-unimolecular sequence of reactions is calculated to complete at timescales ofseconds, and is expected to give an ELVOC formation yield on the order of some percent. Figure 2 illustrates the potential energy surface for the proposed ELVOC formation pathway from C 6H9O4 to C6H9O8.

Figure 2 Potential energy surface (zero-point corrected electronic energies at the ROHF-RCCSD(T)F12a/VDZ-F12//ωB97XD/aug-cc-pV(T+d)Z level, in kcal/mol) for the ELVOC-forming pathway. The key transition states are shown schematically. The reference energy level is that of the C 6H9O4 peroxy radical and free oxygen molecules. The computed rate coefficients are highly sensitive to the level of quantum chemical theory used. As described in the experimental section, relative energies of different conformers are reasonably reliably predicted already by density functional theory with modest basis sets. However, the absolute reaction energetics are not: the CCSD(T)-F12 energy corrections changed barrier heights by up to 5 kcal/mol (implying roughly a factor of 5000 in the rate coefficient). For the hydrogen shift reactions, accounting for tunneling is also important. As shown in Table S3, tunneling increases the net rate coefficients of H-shifts by between 30 and 700. Accounting for the presence of multiple conformers (by using MC-TST rather than 12


Page 13 of 24

1 2 3 293 4 5 294 6 295 7 8 296 9 10 11 297 12 13 298 14 15 299 16 300 17 18 301 19 20 302 21 303 22 23 304 24 305 25 26 306 27 28 307 29 308 30 31 309 32 33 310 34 35 311 36 37 312 38 313 39 40 314 41 42 315 43 316 44 45 317 46 318 47 48 319 49 50 320 51 321 52 53 322 54 55 56 57 58 59 60


conventional TST) usually decreases, but in a few cases may also increase the rate coefficient. This effect is always less than a factor of 30, and usually less than a factor of 10. Accounting for hindered rotations tends to decrease both the forward and reverse rate coefficients, depending on the number of internal rotations constrained in the transition states, but this effect is less than a factor of 6 (see section S2.3. in the SI). 3.2.2.

Deuterated cyclohexene (C6D10) ELVOC spectrum

Using fully deuterated cyclohexene (C 6D10), instead of unsubstituted cyclohexene (C 6H10), the postulated hydrogen shift reactions should be seen to slow down due to the primary kinetic isotope effect caused by the heavier D-atoms 84. Transition state theory calculations on the lowest-energy conformers of the three main ELVOC-forming H-shifts indicates that deuteration decreases the H-shift rate coefficients by about a factor of 30 for C6H9O4 and C6H9O6 peroxy radicals, and more than 200 for C6H9O8. The decrease is due both to a reduction of the tunneling factor, and to an increase in the zero-point corrected energy barrier. This should lead to at least four different observable outcomes: (i) the yield of deuterated ELVOC should be smaller than the yield of unsubstituted ELVOC, (ii) the product distribution should show less oxidized species (lower O/C) than with only hydrogen containing cyclohexene ozonolysis, (iii) the dimer distribution should be narrower in the deuterated case, as there are fewer different peroxy radicals in the reaction mixture and, (iv) the dimer to monomer ratio should be higher due to slower unimolecular termination paths. These effects arise as the hydrogen shifts leading to larger oxygen contents are slower with deuterium atoms, and the infusion of O 2 to the carbon structure requires the H/D-shift to create a place for the next O 2 addition. Indeed these influences are seen in the product distributions (Figure 1, 3a and 3b, and Table 1). The intensities of the monomer ELVOC peaks observed have dropped roughly by a factor of 100 to 1000 in the fully deuterated system, even though the vapor pressures of C 6H10 and C6D10 are similar, and their rate coefficients with O3 are comparable29 (note that the deuteration of cyclohexene does not change the O 3 reaction rate significantly, as the C-D bonds do not take part in the primary reaction step). Also the dimer distribution has narrowed and shows less oxidized species (i.e., the heaviest and most oxidized dimer in the C6H10 + O3 system is found at 464.0529 Th with 15 O-atoms attached, whereas the heaviest dimer in the C6D10 + O3 system has only 12 O-atoms and is found at 431.1623 Th; see Table 1). In addition, the dimer to monomer ratio is significantly higher in the deuterated system (compare Figures 1a and 3a), indicating that monomer formation through hydrogen shifts followed by OH or HO 2 elimination, is less efficient than dimer formation through peroxy radical recombination in the ozonolysis of C 6D10 compared to C6H10 case (more on dimers below and in the SI).

13



a)

60

Ion counts (s-1)

50 40 30 20 10 0 240

280

320

360

400

Mass / charge (Th)

b) C6D10+ H2O

3.0

a) c)

3H 3D

C6D10+ D2O

80

2.0

2H

1.5

2D 3H

1.0

3D

Ion counts (s-1)

2.5 Ion counts ((s-1)

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 323 32 33 324 34 35 325 36 326 37 38 327 39 328 40 41 329 42 43 330 44 45 331 46 47 332 48 49 333 50 334 51 52 335 53 54 336 55 337 56 57 58 338 59 339 60

Page 14 of 24

60

40

20

0.5 0.0

0 250

260

270

280

290

300

260

Mass / charge (Th)

270

280

290

Mass / charge (Th)

Figure 3a-3c ELVOC spectra from supporting experiments: a) ELVOC spectrum obtained from the C6D10 + O3 reaction. Peaks below 300 Th are monomers, and above 300 Th, dimers; see Table 1 for the elemental compositions of the peaks. b) Monomer ELVOC spectrum of Figure 3a magnified. The deuterated ELVOC products exchange D→H prior to detection, but adding D2O to the gas flow exchanges them H→D; the C6D10+D2O (+O3) peaks have been scaled by multiplying them with a factor of three. c) ELVOC spectrum obtained from the cis-6-nonenal + O3 experiment, which is nearly identical to the monomer spectrum of cyclohexene (Figure 1).

3.2.3.

Influence of water (H 2O or D2O) on the cyclohexene ELVOC spectra

To inspect the ELVOC functionality we added H 2O and D2O to the carrier gas flow. When H 2O was added, we did not observe changes in the cyclohexene ELVOC signals, (see Figure S2 in the SI), but when D 2O was used instead, we observed systematic shifts in the m/z of the detected ELVOC, as expected given that labile acidic hydrogens are easily exchanged (e.g., -OH and -OOH, i.e., H-atoms bonded to electronegative species such as O-atom; see Figure 1b and 3b and Figure S3 in the SI)29, 85-87. That is, a specific number of hydrogens in the ELVOC were substituted to deuterium (HàD), revealing important clues about the structures of the species formed. Given the high number density of D2O in the gas mixture, we expected H/D exchange to go to completion, such that the identified ELVOCs will shift one mass unit in the spectrum for every exchangeable hydrogen 14


Page 15 of 24

1 2 3 340 4 5 341 6 342 7 8 343 9 10 344 11 345 12 13 346 14 15 347 16 348 17 18 349 19 350 20 21 351 22 23 352 24 25 353 26 27 28 354 29 355 30 31 356 32 33 357 34 358 35 36 359 37 360 38 39 361 40 41 362 42 43 363 44 45 46 364 47 365 48 49 50 366 51 367 52 53 368 54 55 369 56 370 57 58 371 59 60 372


atom in the ELVOC molecule. In the presence of D 2O, the lightest, major ELVOC product (C 6H8O7 at 254.0154 Th, see Scheme 2), was observed to shift by two mass units, indicating that 2 H-atoms were exchanged to D (Figure 1b). The more oxygenated 270.0103 Th (C 6H8O8) and 286.0052 Th (C6H8O9) products, in contrast, shifted by 3 mass units (i.e., 3 H were exchanged to D). These observations are consistent with the mechanism in that the third hydroperoxide functionality is not yet formed in the parent compound producing the 254.0154 Th (C6H8O7) signal, and support the reaction sequence advancing by Hshifts and creating labile, exchangeable hydrogens (as –OOH) during the process, as described in Scheme 2. However, the effective formation rate of the C 6H8O7 product through 1,4-Hshift of the C 6H8O8 intermediate was calculated to be only about 0.001 s−1 (see Table S3 in the SI), and hence, should not give an appreciable yield in the experimental time-scale. Thus it seems more likely that C6H8O7 product is formed through bimolecular reactions of the intermediate peroxy radials (i.e., by RO 2 + RO2 reactions discussed below in Section 3.3.). Nevertheless, some product formation could be observed even with the calculated formation rate. 3.2.4.

cis-6-nonenal

To further test the general applicability of the proposed mechanism, we performed additional cis-6-nonenal ozonolysis experiments, previously reported to produce an ELVOC spectrum 12. According to general ozonolysis mechanisms of unsaturated organic compounds 60, 61, cis-6-nonenal produces the same CI as cyclohexene, but with less internal excitation due to the exocyclic double bond position in 6-nonenal, and with a lesser yield because another three-carbon CI is produced concomitantly (see Scheme S1 in the SI for structures of the nonenal CIs). The cis-6-nonenal + O3 reaction was observed to generate a nearly identical ELVOC monomer distribution to that from cyclohexene + O 3 (compare Figures 1a and 3c), thus lending credit to the suggested formation mechanism (Scheme 2). Also, the addition of D2O to the gas mixture resulted in similar mass shifts in the spectrum, further crediting the proposition (see Figure S4 in the SI). 3.3. Bimolecular termination reactions and dimer formation In addition to unimolecular termination reactions described above, the radical chain can also be terminated by recombination and cross combination reactions of RO 288, 89: RO2 + RO2 à ROOR + O2

(1)

RO2 + HO2 à ROOH + O2

(2)

RO2 + RO2 à RO + RO + O2

(3)

RO2 + RO2 à RCHO + ROH + O2

(4)

Channels (1) and (2) are likely the most relevant for ELVOC formation. However, as the HO 2/RO2 ratio is small in our flow tube experiments, in contrast to the ambient atmosphere where the opposite usually holds, we do not see products that could be unambiguously assigned toreaction (2). Nevertheless, the product 15



1 2 3 373 4 5 374 6 375 7 8 376 9 10 377 11 378 12 13 379 14 15 380 16 381 17 18 19 382 20 383 21 22 384 23 24 385 25 386 26 27 387 28 388 29 30 389 31 32 390 33 391 34 35 392 36 37 393 38 39 394 40 41 42 395 43 396 44 45 397 46 398 47 48 399 49 50 400 51 401 52 53 54 402 55 56 57 58 59 60

Page 16 of 24

C6H10O6 (shown in Table 1, but observed with too low intensity to be visible in the scale of Figure 1a), has the expected composition and could be produced through this channel. The alkoxy radicals formed in reaction (3) are potential/common chain branching agents in radical reaction sequences, and hence, are not generally expected to lead to molecular weight growth chemistry, although exceptions (e.g., by isomerization through 1,4 and 1,5 H-shifts) do exist 17, 87, 90. The alcohol and carbonyl species created in reaction (4) could in principle be ELVOC compounds, provided that the starting peroxy radicals were oxidized sufficiently before the reaction. The observed C 6H8O7 product, shortly discussed above, is potentially formed through reaction channel (4), as its formation rate was found too slow to be produced in appreaciable amounts from the autoxidation reaction pathway shown in Scheme 2. When the O3 and cyclohexene concentrations are increased enough, dimers appear in the spectra (Figure 1a and 3a, Table 1). These are not conventional dimers held together by Van der Waals forces, but instead distinct chemical species formed in reaction (1). The dimers generally have 12 C-atoms, either 18 or 20 Hatoms and lower O/C than the monomers (Table 1), which shows their formation being in competition with the unimolecular autoxidative, H-shift + O2 addition sequence, forming the monomer compounds. In experiments with D2O, the dimer species generally exchanged an equal or smaller number of hydrogens than the monomer species, and thus indicated different chemical structures (i.e., some of the acidic hydrogens have not formed through H-shifts before a termination has occured by a bimolecular reaction). The reaction channel (1) leads to peroxides, which are the likely dimer structures detected in this work. If the recombination or cross-combination reaction occurs between C 6H9O6 and C6H9O8 acylic peroxy radicals, diacyl peroxides are expected to form. More on the specific dimers, their formation and D-exchange reactions can be found in the SI. 3.4. ELVOC yield The yield of ELVOC from the cyclohexene ozonolysis reaction was determined by varying the reagent cyclohexene and ozone concentrations and measuring the produced ELVOC concentrations with the nitrate CI-APi-TOF. The yield increases linearly with the product of these concentrations (i.e., with [cyclohexene] × [O3]), as expected for products of a chain process that begins with, and is limited by, the cyclohexene ozonolysis reaction. At the highest reagent concentrations, the yield curve started to bend gradually, and thus only the lowest concentrations were used in the ELVOC yield determination (Figure 4), in order to avoid complication by uncertain second order terms.

16


Page 17 of 24

8 [ELVOC] / 1x109 molecule cm-3

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 403 23 24 25 404 26 405 27 28 406 29 30 407 31 32 408 33 34 409 35 410 36 37 411 38 39 412 40 413 41 42 414 43 415 44 45 416 46 47 417 48 49 418 50 51 419 52 420 53 54 421 55 56 422 57 58 59 423 60


6

4

2

0 0

4

8

12

16

k [CH]x[O3] / 1x1010 molecule cm-3 s-1

Figure 4 ELVOC yield plot obtained from cyclohexene ozonolysis. Measured ELVOC concentrations follow the product of k × [cyclohexene] × [O3] (k(C6H10 + O3) = 7.4 × 10−17 cm3 molecule−1 s−1 33). Only the identified peaks shown in Table 1 were included. The detection efficiency of the nitrate CI-APi-TOF to ELVOC was recently estimated to be similar to the detection efficiency of H2SO412. Taking into account the calibration factor for H 2SO4 determined in this work (see the SI), together with an estimated diffusion limited wall loss for the sticky ELVOC molecules, one arrives at an ELVOC yield of (4.5 ± 0.2)% from the C 6H10 + O3 reaction (Figure 4), with the uncertainty given as one standard error of the fit. The overall uncertainty of this value was estimated as ±80%, and was obtained by the propagation of error method. The agreement with the only previous determination of Ehn et al.12, (4 ± 2)%, obtained under very different experimental conditions in Jülich Plant Atmosphere Chamber (JPAC), is excellent. For the C6D10 + O3 reaction, appreciable ELVOC signals were obtained only with the highest reagent concentrations used, and hence, the ELVOC yield could not be determined similarly as in the C6H10 + O3 reaction. It can, however, be roughly estimated as about two orders of magnitude lower, leading to a value of about 0.04%. More details of the yield experiments and calculation are given in the SI. It must be emphasized that the yield determined in this work corresponds only to a few percent of the total C6H10 + O3 product yield and does not contradict the formation of already well-established major products of the reaction29-57. Nevertheles, even such a small yield can have significant consequences on the formation of SOA, as was shown by Ehn et al.12.

17



1 2 3 424 4 5 6 425 7 426 8 9 427 10 11 428 12 429 13 14 430 15 16 431 17 18 432 19 20 433 21 434 22 23 435 24 436 25 26 27 437 28 29 438 30 439 31 32 440 33 441 34 35 442 36 37 443 38 444 39 40 41 445 42 43 446 44 45 46 447 47 448 48 49 449 50 51 450 52 451 53 54 452 55 56 453 57 58 454 59 60 455

Page 18 of 24

4. Conclusions The formation of highly oxidized, extremely low volatility products from ozonolysis of cyclohexene was investigated by experimental and computational methods. A large amount of previously unidentified product species, with O/C ratios as high as 1.5, were detected in the nitrate chemical ionization mass spectra. Results of quantum chemical calculations, supplemented by additional specific laboratory experiments, enabled us to assign chemical structures for the measured elemental compositions. The ELVOC yield from the cyclohexene ozonolysis system was estimated to be (4.5 ± 3.8)%. Gas-phase oxidation chemistry advancing by peroxy radical H-shift + O 2 addition has been reported previously. However, due to the lack of experimental tools to detect the highly oxidized and extremely low volatility products of these reactions, the sequence of reactions has not been recognized to advance as far as what is found here (and implicated by very recent research). In addition, the time scale in which these reactions form closed-shell ELVOC products is much shorter than previously thought, the whole process starting and terminating in seconds or less. The results of the current study have potentially important consequences for the understanding of atmospheric SOA formation processes. In the current work, cyclohexene was specifically chosen as a surrogate for many biogenic VOCs with endocyclic double bonds, to gain insight into ELVOC formation from ozonolysis. The simpler structure of cyclohexene allowed high-level quantum chemical calculations of the detailed step-by-step progression of rapid oxygenation of the precursor compound. This progression, with rate coefficients for favorable H-abstraction sites and typical radical termination reactions, provides a starting point for interpreting the more complex ELVOC spectra produced from important biogenic emissions, such as mono- and sesquiterpenes, crucial precursors for atmospheric SOA.

Supporting Information Sections S1.1.-S1.9., additional information on performed laboratory experiments: S1.1. Laboratory investigations, S1.2. Chemicals, S1.3. Uncertainties, S1.4. Water addition, S1.5. Deuterated cyclohexene, S1.6. Dimers, S1.7. cis-6-nonenal ozonolysis, S1.8. ELVOC molar yield, S1.9. Relation to previous studies. Figures S1-S3: Figure S1 Experimental setup, Figure S2 H2O influence on ELVOC signals, Figure S3 D2O influence on ELVOC signals. Scheme S1 Criegee intermediates from cis-6-nonenal ozonolysis. Table S1 Description of the experiments performed. Sections S2.1.-S2.4., additional details on quantum chemical calculations: S2.1. Validation of configurational sampling algorithm, S2.2. TST versus MC-TST, S2.3.1 Assessing the effect of hindered rotors: uncoupled models, S2.3.2 Assessing the effect of hindered rotors: MS-T modeling, S2.4. Molecular 18


Page 19 of 24

1 2 3 456 4 5 457 6 458 7 8 459 9 10 460 11 461 12 13 462 14 15 463 16 17 18 464 19 20 465 21 22 466 23 24 467 25 468 26 27 469 28 29 470 30 471 31 32 472 33 473 34 35 474 36 475 37 476 38 39 477 40 478 41 42 479 43 480 44 45 481 46 482 47 48 483 49 484 50 51 485 52 486 53 54 487 55 488 56 57 489 58 490 59 60


structures and thermodynamic parameters of the systems studied. Figures S5-S7: Lowest-energy structures for the C6H9O4, C6H9O6 and C6H9O8. Tables S2-S22: Table S2 The partition functions of the reactants, transition states and products for each reaction., Table S3 Calculated rate coefficients for the different H-shift and CO loss reactions, Table S4 Thermodynamic properties of the structures shown in Figures S5-S7, Table S5 Cartesian co-ordinates of the structures shown in Figures S5-S7. Tables S6-S22 Conformers of the species used in the final quantum chemical calculations. This information is available free of charge via the Internet at http://pubs.acs.org/.

Acknowledgements T. K. and M. Sipilä thank the Academy of Finland for funding (Academy Research Fellow grant number 266388 and Post Doctoral Researcher grant number 251427, respectively). H. G. K. thanks the support from The Danish Council for Independent Research—Natural Sciences and the Danish Center for Scientific Computing. J.A.T. acknowledges support from the U.S. Department of Energy through grants DESC0011791 and DE-SC0006867. We thank Markku Kulmala for providing the instrumentation and laboratory facilities used in this work. References

[1] Kulmala, M.; Kontkanen, J.; Junninen, H.; Lehtipalo, K.; Manninen, H. E.; Nieminen, T.; Petaja, T.; Sipila, M.; Schobesberger, S.; Rantala, P.; Franchin, A.; Jokinen, T.; Järvinen, E.; Äijälä, M.; Kangasluoma, J.; Hakala, J.; Aalto, P.; Paasonen, P.; Mikkilä, J.; Vanhanen, J.; Aalto, J.; Hakola, H.; Makkonen, U.; Ruuskanen, T.; Mauldin, R. L. III.; Duplissy, J.; Vehkamaki, H.; Back, J.; Kortelainen, A.; Riipinen, I.; Kurten, T.; Johnston, M. V.; Smith, J. N.; Mikael, E.; Mentel, T. F.; Lehtinen, K. E. J.; Laaksonen, A.; Kerminen, V.-M.; Worsnop, D. R. Science, 2013, 339 ,943-946. [2] Zhang, Q.; Jimenez, J. L.; Canagaratna, M. R. ; Allan, J. D.; Coe, H.; Ulbrich, I.; Alfarra, M. R.; Takami, A.; Middlebrook, A. M.; Sun, Y. L.; Dzepina, K.; Dunlea, E.; Docherty, K.; DeCarlo, P. F.; Salcedo, D.; Onasch, T.; Jayne, J. T.; Miyoshi, T.; Shimono, A.; Hatakeyama, S.; Takegawa, N.; Kondo, Y.; Schneider, J.; Drewnick, F.; Borrmann, S.; Weimer, S.; Demerjian, K.; Williams, P.; Bower, K.; Bahreini, R.; Cottrell, L.; Griffin, R. J.; Rautiainen, J.; Sun, J. Y.; Zhang Y. M.; Worsnop, D. R. J. Geophys Res. 2007, 34, L13801-6. [3] Jimenez, J. L.; Canagaratna, M. R.; Donahue, N. M.; Prevot, A. S.; Zhang, Q.; Kroll, J. H.; DeCarlo, P. F.; Allan, J. D.; Coe, H.; Ng, N. L.; Aiken, A. C.; Docherty, K. S.; Ulbrich, I. M.; Grieshop, A. P.; Robinson, A. L.; Duplissy, J.; Smith, J. D.; Wilson, K. R.; Lanz, V. A.; Hueglin, C.; Sun, Y. L.; Tian, J.; Laaksonen, A.; Raatikainen, T.; Rautiainen, J.; Vaattovaara, P.; Ehn, M.; Kulmala, M.; Tomlinson, J. M.; Collins, D. R.; Cubison, M. J.; Dunlea, E. J.; Huffman, J. A.; Onasch, T. B.; Alfarra, M. R.; Williams, P. I.; Bower, K.; Kondo, Y.; Schneider, J.; Drewnick, F.; Borrmann, S.; Weimer, S.; Demerjian, K.; Salcedo, D.; Cottrell, L.; Griffin, R.; Takami, A.; Miyoshi, T.; Hatakeyama, S.; Shimono, A.; Sun, J. Y.; Zhang, Y. M.; Dzepina, K.; Kimmel, J. R.; Sueper, D.; Jayne, J. T.; Herndon, S. C.; Trimborn, A. M.;

19



1 2 3 491 4 5 492 6 493 7 8 494 9 495 10 11 496 12 497 13 498 14 15 499 16 500 17 18 501 19 502 20 21 503 22 504 23 24 505 25 506 26 27 507 28 508 29 30 509 31 510 32 33 511 34 512 35 36 513 37 514 38 39 515 40 516 41 42 517 43 518 44 45 519 46 520 47 521 48 49 522 50 523 51 52 524 53 525 54 55 526 56 527 57 58 528 59 529 60

530

Page 20 of 24

Williams, L. R.; Wood, E. C.; Middlebrook, A. M.; Kolb, C. E.; Baltensperger, U.; Worsnop, D. R. Science, 2009, 326, 1525-1529. [4] IPCC (intergovernmental panel on climate change); 5th Assessement Report, 2013. Web address: http://www.ipcc.ch/report/ar5/ [5] Ng, N. L.; Canagaratna, M. R.; Zhang, Q.; Jimenez, J. L.; Tian, J.; Ulbrich, I. M.; Kroll, J. H.; Docherty, K. S.; Chhabra, P. S.; Bahreini, R.; Murphy, S. M.; Seinfeld, J. H.; Hildebrandt, L.; Donahue, N. M.; DeCarlo, P. F.; Lanz, V. A.; Prévôt, A. S. H.; Dinar, E.; Rudich, Y.; Worsnop, D. R. Atmos. Chem. Phys. 2010, 10, 4625-4641. [6] Aiken, A. C.; DeCarlo, P. F.; Kroll, J. H.; Worsnop, D. R.; Huffman, J. A.; Docherty, K. S.; Ulbrich, I. M.; Mohr, C.; Kimmel, J. R.; Sueper, D.; Sun, Y.; Zhang, Q.; Trimborn, A.; Northway, M.; Ziemann, P. J.; Canagaratna, M. R.; Onasch, T. B.; Alfarra, M. R.; Prévôt, A. S. H.; Dommen, J.; Duplissy, J.; Metzger, A.; Baltensperger, U.; Jimenez, J. L. Environ. Sci. Technol. 2008, 42, 4478-4485. [7] Kroll, J. H.; Seinfeld, J. H. Atmos. Environ. 2008, 42, 3593-3624. [8] Chacon-Madrid, H. J.; Donahue, N. M. Atmos. Chem. Phys. 2011, 11, 10553-10563. [9] Donahue, N. M.; Kroll, J. H.; Pandis, S. N.; Robinson, A. L. Atmos. Chem. Phys. 2012, 12, 615-634. [10] Donahue, N. M.; Trump, E. R.; Pierce, J. R.; Kulmala, M. Geophys. Res. Lett. 2011, 38, L16801. [11] Donahue, N. M.; Ortega, I. K.; Chuang, W.; Riipinen, I.; Riccobono, F.; Schobesberger, S.; Dommen, J.; Baltensperger, U.; Kulmala, M.; Worsnop, D. R.; Vehkamaki, H. Faraday Discuss. 2013, 165, 91-104. [12] Ehn, M.; Thornton, J. A.; Kleist, E.; Sipilä, M; Junninen, H.; Pullinen, I.; Springer, M.; Rubach, F.; Tillmann, R.; Lee, B.; Lopez-Hilfiker, F.; Andres, S.; Acir, I.-H.; Rissanen, M.; Jokinen, T.; Schobesberger, S.; Kangasluoma, J.; Kontkanen, J.; Nieminen, T.; Kurtén, T.; Nielsen, L. B.; Jørgensen, S.; Kjaergaard, H. G.; Canagaratna, M.; Dal Maso, M.; Berndt, T.;Petäjä, T.; Wahner, A.; Kerminen, V.-M.; Kulmala, M.; Worsnop, D. R.; Wildt, J.; Mentel, T. F. Nature, 2014, 506, 476-479. [13] Kulmala, M.; Toivonen, A.; Mäkelä, J. M.; Laaksonen, A. Tellus B 1998, 50, 449-462. [14] Glowacki, D. R.; Pilling, M. J. Chemphyschem 2010, 11, 3836-3843. [15] Taatjes, C. A. J. Phys. Chem. A 2006, 110, 4299-4312. [16] Pilling, M. J. Chem. Soc. Rev. 2008, 37, 676-685. [17] Vereecken, L.; Francisco, J. S. Chem. Soc. Rev. 2012, 41, 6259-6293. [18] Eddingsaas, N. C.; Loza, C. L.; Yee, L. D.; Seinfeld, J. H.; Wennberg, P. O. Atmos. Chem. Phys. 2012, 12, 6489– 6504. [19] Crounse, J. D.; Nielsen, L. B.; Jørgensen, S.; Kjaergaard, H. G.; Wennberg, P. O. J. Phys. Chem. Lett. 2013, 4, 3513−3520. [20] Crounse, J. D.; Knap, H. C.; Ørnsø, K. B.; Jørgensen, S.; Paulot, F.; Kjaergaard, H. G.; Wennberg, P. O. J. Phys. Chem. A 2012, 116, 5756-5762. [21] Crounse, J. D.; Paulot, F.; Kjaergaard, H. G.; Wennberg, P. O. Phys. Chem. Chem. Phys. 2011, 13, 13607-13613. [22] Mattill, H. A. Oil and Soap, 1941, 18, 73-76. [23] Amorati, R.; Foti, M. C.; Valgimigli, L. J. Agric. Food Chem. 2013, 61, 10835-10847 [24] Karlberg, A.-T.; Börje, A.; Johansen, J. D.; Lidén, C.; Rastogi, S.; Roberts, D.; Uter, W.; White, I. R. Contact Dermatitis, 2013, 69, 323-334 [25] Vereecken, L.; Muller, J. F.; Peeters, J. Phys. Chem. Chem. Phys. 2007, 9, 5241-5248. [26] Zhao, J.; Ortega, J.; Chen, M.; McMurry, P. H.; Smith, J. N. Atmos. Chem. Phys. 2013, 13, 7631–7644.

20


Page 21 of 24

1 2 3 531 4 5 532 6 533 7 8 534 9 535 10 11 536 12 537 13 538 14 15 539 16 540 17 18 541 19 542 20 21 543 22 544 23 24 545 25 546 26 27 547 28 548 29 30 549 31 550 32 33 551 34 552 35 36 553 37 554 38 39 555 40 556 41 42 557 43 558 44 45 559 46 560 47 561 48 49 562 50 563 51 52 564 53 565 54 55 566 56 567 57 58 568 59 569 60


[27] Ehn, M.; Kleist, E.; Junninen, H.; Petäjä, T.; Lönn, G.; Schobesberger, S.; Dal Maso, M.; Trimborn, A.; Kulmala, M.; Worsnop, D. R.; Wahner, A.; Wildt, J.; Mentel, Th. F. Atmos. Chem. Phys. 2012, 12, 5113-5127. [28] Ehn, M.; Junninen, H.; Petäjä, T.; Kurtén, T.; Kerminen, V. –M.; Schobesberger, S.; Manninen, H. E.; Ortega, I. K.; Vehkamäki, H.; Kulmala, M.; Worsnop, D. R. Atmos. Chem. Phys. 2010, 10, 8513-8530. [29] Aschmann, S. M.; Tuazon, E. C.; Arey, J.; Atkinson, R. J. Phys. Chem. A 2003, 107, 2247-2255. [30] Chuong, B.; Zhang, J.; Donahue, N. M. J. Am. Chem. Soc. 2004, 126, 12363-12373. [31] Keywood, M. D.; Kroll, J. H.; Varutbangkul, V.; Bahreini, R.; Flaglan, R. C.; Seinfeld, J. H. Environ. Sci. Technol. 2004, 38, 3343-3350. [32] Ziemann, P. J. Phys. Chem. A 2002, 106, 4390-4402. [33] Carlsson, P. T. M.; Dege, J. E.; Keunecke, C.; Krüger, B. C.; Wolf, J. L.; Zeuch, T. Phys. Chem. Chem. Phys. 2012, 14, 11695-11705. [34] Hamilton, J. F.; Lewis, A. C.; Reynolds, J. C.; Carpenter, L. J.; Lubben A. Atmos. Chem. Phys. 2006, 6, 4973-4984. [35] Drozd, G. T.; Donahue, N. M. J. Phys. Chem. A 2011, 115, 4381-4387. [36] Donahue, N. M.; Drozd, G. T.; Epstein, S. A.; Presto, A. A.; Kroll, J. H. Phys. Chem. Chem. Phys. 2011, 13, 10848-10857. [37] Atkinson, R.; Aschmann, S. M. Environ. Sci. Technol. 1993, 27, 1357-1363. [38] Fenske, J. D.; Kuwata, K. T.; Houk, K. N.; Paulson, S. E. J. Phys. Chem. A 2000, 104, 7246-7254. [39] Epstein, S. A.; Donahue, N. M. J. Phys. Chem. A 2010, 114, 7509-7515. [40] Stokes, G. Y.; Chen, E. H.; Walter, S. R.; Geiger, F. M. J. Phys. Chem. A 2009, 113, 8985-8993. [41] Warren, B.; Austin, R. L.; Cocker, D. R. III. Atmos. Environ. 2009, 43, 3548-3555. [42] Warren, B.; Malloy, Q. G. J.; Yee, L. D.; Cocker, D. R. III. Atmos. Environ. 2009, 43, 1789-1795. [43] Stokes, G. Y.; Buchbinder, A. M.; Gibbs-Davis, J. M.; Scheidt, K. A.; Geiger, F. M. J. Phys. Chem. A 2008, 112, 11688-11698. [44] Nørgaard, A. W.; Nøjgaard, J. K.; Clausen, P. A.; Wolkoff, P. Chemosphere 2008, 70, 2032-2038. [45] Shepherd, T. G.; Jonsson, A. I. Atmos. Chem. Phys. 2008, 8, 1435-1444. [46] Nøjgaard, J. K.; Nørgaard, A. W.; Wolkoff, P. Atmos. Environ. 2007, 41, 8345-8354. [47] Narukawa, M.; Matsumi, Y.; Matsumoto, J.; Takahashi, K.; Yabushita, A.; Sato, K.; Imamura, T. Anal. Sci. 2007, 23, 507-512. [48] Gao, S.; Ng, N. L.; Keywood, M.; Varutbangkul, V.; Bahreini, R.; Nenes, A.; He, J.; Yoo, K. Y.; Beauchamp, J. L.; Hodyss, R. P.; Flaglan, R. C.; Seinfeld, J. H. Environ. Sci. Technol. 2004, 38, 6582-6589. [49] Gao, S.; Keywood, M.; Ng, N. L.; Surratt, J.; Varutbangkul, V.; Bahreini, R.; Flagan, R. C.; Seinfeld, J. H. J. Phys. Chem. A 2004, 108, 10147-10164. [50] Presto A. A.; Donahue, N. M. J. Phys. Chem. A, 2004, 108, 9096-9104. [51] Ziemann, P. J. J. Phys. Chem. A 2003, 107, 2048-2060. [52] Bailey, P. S. Ind. Eng. Chem. 1958, 50, 993-996. [53] Bailey, P. S. J. Org. Chem. 1957, 22, 1548-1551. [54] Eastman, R. H.; Silverstein, R. M. J. Am. Chem. Soc. 1953, 75, 1493-1494. [55] Müller, L.; Reinnig, M.-C.; Warnke, J.; Hoffmann, Th. Atmos. Chem. Phys. 2008, 8, 1423-1433. [56] Kalberer, M.; Yu, J.; Cocker, D. R.; Flagan, R. C.; Seinfeld, J. H. Environ. Sci. Technol. 2000, 34, 4894-4901.

21



1 2 3 570 4 5 571 6 572 7 8 573 9 574 10 11 575 12 576 13 577 14 15 578 16 579 17 18 580 19 581 20 21 582 22 583 23 24 584 25 585 26 27 586 28 587 29 30 588 31 589 32 33 590 34 591 35 36 592 37 593 38 39 594 40 595 41 42 596 43 597 44 45 598 46 599 47 600 48 49 601 50 602 51 52 603 53 604 54 55 605 56 606 57 58 607 59 608 60

609

Page 22 of 24

[57] Hatakeyama, S.; Tanonaka, T.; Weng, J.; Bandow, H.; Takagi, H.; Akimoto, H. Environ. Sci. Technol. 1985, 19, 935-942. [58] Johnson, D.; Marston, G. Chem. Soc. Rev. 2008, 37, 699-716. [59] Taatjes, C. A.; Shallcross, D. E.; Percival, C. J. Phys. Chem. Chem. Phys. 2014,16, 1704-1718. [60] Criegee, R. Angew. Chem. Int. Ed. Engl. 1975, 14, 745-752. [61] Marston, G. Science, 2012, 335, 178-179. [62] Jokinen, T.; Sipilä, M.; Junninen, H.; Ehn, M.; Lönn, G.; Hakala, J.; Petäjä, T.; Mauldin III, R. L.; Kulmala, M.; Worsnop, D. R. Atmos. Chem. Phys. 2012, 12, 4117-4125. [63] Junninen, H.; Ehn, M.; Petäjä, T.; Luosujärvi, L.; Kotiaho, T.; Kostiainen, R.; Rohner, U.; Gonin, M.; Fuhrer, K.; Kulmala, M.; Worsnop, D. R. Atmos. Meas. Tech. 2010, 3, 1039-1053. [64] Graus, M.; Müller, M.; Hansel, A. J. Am. Soc. Mass Spectr. 2010, 21, 1037-1044. [65] Jordan, A.; Haidacher, S.; Hanel, G.; Hartungen, E.; Märk, L.; Seehauser, S.; Schottkowsky, R.; Sulzer, P.; Märk, T. D. Int. J. Mass Spectrom. 2009, 286, 122-128. [66] Taipale, R.; Ruuskanen, T. M.; Rinne, J.; Kajos, M. K.; Hakola, H.; Pohja, T.; Kulmala, M. Technical Note: Atmos. Chem. Phys. 2008, 8, 9435-9475. [67] Vereecken, L.; Nguyen, T. L.; Hermans, I.; Peeters, J. Chem. Phys. Lett. 2004, 393, 432-436. [68] Spartan'08, Wavefunction Inc., Irivine CA (2008); Spartan'10, Wavefunction Inc., Irivine CA (2011); Spartan'14, Wavefunction Inc., Irivine CA (2014). [69] Garden, A. L.; Paulot, F.; Crounse, J. D.; Maxwell-Cameron, I. J.; Wennberg, P. O.; Kjaergaard, H. G. Chem. Phys. Lett. 2009, 474, 45-50. [70] Chai, J.-D.; Head-Gordon, M. Phys. Chem. Chem. Phys. 2008, 10, 6615-6620. [71] 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.; Montgomery, 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.; Iyengar, S. S.; Tomasi, J.; Cossi, M.; Rega, N.; Millam, N. J.; 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.; Salvador, P.; Dannenberg, J. J.; Dapprich, S.; Daniels, A. D.; Farkas, Ö.; Foresman, J. B.; Ortiz, J. V.; Cioslowski, J.; Fox, D. J. Gaussian 09, Revision D.01, Gaussian, Inc., Wallingford CT, 2009. [72] Adler, T. B.; Knizia, G.; Werner, H.-J. J. Chem. Phys. 2007, 127, 221106(1-4). [73] Peterson, K. A.; Adler, T. B.; Werner, H.-J. J. Chem. Phys. 2008, 128, 084102(1-12). [74] Werner, H.-J.; Knowles, P. J.; Manby, F. R.; Schütz, M.; Celani, P.; Knizia, G.; Korona, T.; Lindh, R.; Mitrushenkov, A.; Rauhut, G.; Shamasundar, K. R.; Adler, T. B.; Amos, R. D.; Bernhardsson, A.; Berning, A.; Cooper, D. L.; Deegan, M. J. O.; Dobbyn, A. J.; Eckert, F.; Goll, E.; Hampel, C.; Hesselmann, A.; Hetzer, G.; Hrenar, T.; Jansen, G.; Köppl, C.; Liu, Y.; Lloyd, A. W.; Mata, R. A.; May, A. J.; McNicholas, S. J.; Meyer, W.; Mura, M. E.; Nicklaß, A.; O'Neill, D. P.; Palmieri, P.; Peng, D.; Pflüger, K.; Pitzer, R.; Reiher, M.; Shiozaki, T.; Stoll, H.; Stone, A. J.; Tarroni, R.; Thorsteinsson, T.; Wang, M., MOLPRO, a package of ab initio programs, version 2010.1; 2010; see http://www.molpro.net.

22


Page 23 of 24

1 2 3 610 4 5 611 6 612 7 8 613 9 614 10 11 615 12 616 13 617 14 15 618 16 619 17 18 620 19 621 20 21 622 22 623 23 24 624 25 625 26 27 626 28 627 29 30 628 31 32 629 33 630 34 35 631 36 632 37 38 633 39 634 40 41 635 42 636 43 44 637 45 638 46 639 47 48 640 49 641 50 51 642 52 643 53 54 644 55 645 56 57 646 58 647 59 60 648


[75] Vereecken, L.; Peeters, J. J. Chem. Phys. 2003, 119, 5159-5170. [76] Eckart, C. Phys. Rev. 1930, 35, 1303-1309. [77] Ayala, P. Y.; Schlegel, H. B. J. Chem. Phys. 1998, 108, 2314-2325. [78] McClurg, R. B.; Flagan, R. C.; Goddard III, W. A. J. Chem. Phys. 1997, 106, 6675-6680. [79] McClurg, R. B. Comment on: [J. Chem. Phys. 106, 6675 (1997)] and [J. Chem. Phys. 108, 2314 (1998)], J. Chem. Phys. 1999, 111, 7163. [80] Zheng, J.; Mielke, S. L.; Clarkson, K. L.; Truhlar, D. G. Comp. Phys. Comm. 2012, 183, 1803-1812. [81] Zheng, J.; Meana-Pañeda, R.; Trular, D. G. Comp. Phys. Comm. 2013, 184, 2032-2033. [82] Atkinson, R.; Baulch, D. L.; Cox, R. A.; Hampson, R. F., Jr.; Kerr, J. A.; Rossi, M. J.; Troe, J. J. Phys. Chem. Ref. Data 1997, 26, 521-1011. [83] Hyttinen, N.; Rissanen, M. P.; Muuronen, M.; Kupiainen-Määttä, O.; Ehn, M.; Kurtén, T. in preparation [84] Benson, S. W. Thermochemical Kinetics, 2nd ed.; John Wiley & Sons: New York, U.S.A., 1976. [85] Wine, P. H.; Astalos, R. J.; Mauldin, R. L. III J. Phys. Chem. 1985, 89, 2620-2624. [86] Vaghjiani, G. L.; Ravishankara, A. R. J. Phys. Chem. 1989, 93, 1948-1959. [87] Atkinson, R.; Kwok, E. S. C.; Arey, J.; Aschmann, S. M. Faraday Discuss. 1995, 100, 23-37. [88] Orlando, J. J.; Tyndall, G. S. Chem. Soc. Rev. 2012, 41, 6294-6317. [89] Hasson, A. S.; Tyndall, G. S.; Orlando, J. J. J. Phys. Chem. A 2004, 108, 5979-5989. [90] Orlando, J. J.; Tyndall, G. S.; Wallington, T. J. Chem. Rev. 2003, 103, 4657-4689.

23



1 2 3 649 4 5 650 6 7 8 9 10 11 12 13 14 15 16 651 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

TOC Graphic:

24


Page 24 of 24

The Formation of Highly Oxidized Multifunctional Products in the

Recommend Documents