Radical product yields from the ozonolysis of short chain alkenes under atmospheric boundary layer conditions

Article

pubs.acs.org/JPCA

Radical Product Yields from the Ozonolysis of Short Chain Alkenes

under Atmospheric Boundary Layer Conditions

Mohammed S. Alam,^†Andrew R. Rickard,^‡,∥Marie Camredon,^†,⊥Kevin P. Wyche,^§,#Timo Carr,^§

Karen E. Hornsby,^§,○Paul S. Monks,^§and William J. Bloss*^,†

^†School of Geography, Earth & Environmental Sciences, University of Birmingham, Edgbaston, Birmingham B15 2TT, U.K.

^‡School of Chemistry, University of Leeds, Leeds LS2 9JT, U.K.

^§Department of Chemistry, University of Leicester, Leicester LE1 7RH, U.K.

ABSTRACT: The gas-phase reaction of ozone with unsaturated volatile

organic compounds (VOCs), alkenes, is an important source of the critical

atmospheric oxidant OH, especially at night when other photolytic radical

initiation routes cannot occur. Alkene ozonolysis is also known to directly

form HO₂radicals, which may be readily converted to OH through

reaction with NO, but whose formation is poorly understood. We report a

study of the radical (OH, HO₂, and RO₂) production from a series of small

alkenes (propene, 1-butene, cis-2-butene, trans-2-butene, 2-methylpropene,

2,3-dimethyl-2-butene (tetramethyl ethene, TME), and isoprene). Experi-

ments were performed in the European Photoreactor (EUPHORE)

atmospheric simulation chamber, with OH and HO₂levels directly

measured by laser-induced ﬂuorescence (LIF) and HO₂+ ΣRO₂levels

measured by peroxy-radical chemical ampliﬁcation (PERCA). OH yields

were found to be in good agreement with the majority of previous studies

performed under comparable conditions (atmospheric pressure, long time scales) using tracer and scavenger approaches. HO₂

yields ranged from 4% (trans-2-butene) to 34% (2-methylpropene), lower than previous experimental determinations. Increasing

humidity further reduced the HO₂yields obtained, by typically 50% for an RH increase from 0.5 to 30%, suggesting that HO_x

production from alkene ozonolysis may be lower than current models suggest under (humid) ambient atmospheric boundary

layer conditions. The mechanistic origin of the OH and HO₂production observed is discussed in the context of previous

experimental and theoretical studies.

Gas-phase alkene ozonolysis is believed to proceed via the

Criegee mechanism,⁵illustrated in Figure 1. Ozonolysis is

highly exothermic, initiated by the electrophilic cycloaddition of

ozone across the CC double bond to form an unstable 1,2,3

trioxolane (hereafter referred to as a primary ozonide, POZ)

(R1). This intermediate is high in energy and rapidly

decomposes at the central C−C bond and one of the O−O

bonds. Given that the O−O bond can break at two diﬀerent

sites, two alternative pairs of carbonyl oxides (hereafter referred

to as Criegee intermediates, CIs) and stable carbonyl molecules

can be formed (R2a and R2b).

The CI and carbonyl coproduct produced from the

exothermic decomposition of the POZ possess a signiﬁcant

amount of vibrational excitation. This energy enables further

unimolecular reactions of the excited CI to occur but is not

suﬃcient for the decomposition of the carbonyl molecule;¹

Figure 2. The distribution of decomposition products of the

POZ is dependent upon the chemical conditions and

substitution of the alkene. Diﬀerent CIs behave as distinct

1. INTRODUCTION

Alkenes, unsaturated hydrocarbons, are emitted to the

atmosphere from a range of natural and anthropogenic sources,

notably biogenic emissions of isoprene, C₅H₈, and the

isoprenoid terpenes (C₁₀H₁₆) and sesquiterpenes (C₁₅H₂₄).

Alkenes can contribute up to 30% of the total OH sink in urban

regions,¹and a higher proportion in forested environments;²

atmospheric degradation of alkenes contributes to the

production of ozone in the presence of nitrogen oxides and

leads to the production of multifunctional oxygenated

degradation products,³which may act as precursors to, or

contribute to the formation of, secondary organic aerosol

(SOA).

In addition to degradation driven by reaction with OH and

NO₃, alkene oxidation may be initiated by reaction with ozone,

a process that leads to the nonphotolytic production of HO_x

radical intermediates; detailed analyses of measurements from

atmospheric ﬁeld campaigns have shown ozonolysis to account

for up to 30% of the total OH radical production.⁴

Understanding the yields of OH, HO₂, and RO₂radicals and

their dependence upon atmospheric conditions is essential to

quantify this important contribution to atmospheric oxidizing

capacity.

Received: September 1, 2013

Revised: October 29, 2013

Published: October 30, 2013

12468

dx.doi.org/10.1021/jp408745h | J. Phys. Chem. A 2013, 117, 12468−12483

The Journal of Physical Chemistry A

Article

Figure 1. Cycloaddition of ozone across the alkene double bond and subsequent decomposition of the POZ: the Criegee mechanism.

Figure 2. Potential fates of the Criegee intermediate.

chemical entities as demonstrated by the range of detected

experimental products, dependent upon the extent of the

substitution of the CI and distribution of energy following

decomposition of the POZ.³Substituted CIs can be formed in a

syn (i.e., with the alkyl substituent on the same side of the CI as

the terminal O atom) conﬁguration or anti conﬁguration, with a

substantial barrier to interconversion between the two.⁶

Brieﬂy, syn- and disubstituted CIs are thought to

predominantly decompose through isomerization via a ﬁve-

membered transition state to give an excited vinyl hydro-

peroxide intermediate, which subsequently decomposes to give

OH and a vinoxy radical (R3a; the hydroperoxide mecha-

nism).³The proportion of the vibrationally excited CIs that

have insuﬃcient energy to isomerize/decompose may form so-

called stable Criegee intermediates (SCIs), either directly

following decomposition of the primary ozonide or through

collisional stabilization of the CI (R4), which can subsequently

undergo bimolecular reactions⁷or decomposition (leading to

further, delayed production of OH, again via a vinyl

hydroperoxide intermediate (R3b), or potentially to HO₂, via

a dioxirane/hot acid intermediate (R6)). The prevailing

conditions (pressure; experimental time scale) have been

12469

dx.doi.org/10.1021/jp408745h | J. Phys. Chem. A 2013, 117, 12468−12483

The Journal of Physical Chemistry A

Article

shown to be critical in controlling the observed formation of

OH, SCIs, and other products.⁸

times; or (iii) turnover experiments in which the additional

removal of the alkene (due to reaction with OH) or ozone (due

primarily to reaction with HO₂) is used to infer OH and/or

HO₂production, again at long (minutes +) time scales.²⁸

The majority of OH radical yields reported have been

obtained through tracer^21,22and scavenger methods.^23,24In

some cases these approaches can be problematic, for example,

with the cyclohexane scavenger approach, production of the

observed products (usually cyclohexanol and/or cyclohex-

anone) depends upon the radical population actually present

(through dependencies upon RO₂vs. HO₂competition for the

cyclohexylperoxy radical) and hence can vary signiﬁcantly

between chemical systems and within the same system as the

chemical composition evolves.²⁹Other OH scavenger

techniques have included the use of excess 2-butanol³⁰and

CO,³¹monitoring the reaction products 2-butanone and CO₂,

respectively. Yields of OH from alkene-ozone reactions have

also been derived by adding relatively small quantities of tracer

compounds (e.g., 1,3,5-trimethylbenzene, m-xylene, and di-n-

butyl ether) to scavenge OH radicals. Such tracers exhibit

varying reactivity to OH, allowing the determination of OH

yields in an ozonolysis system by comparing the relative decay

ratios of tracer pairs, through knowledge of their OH reaction

rate coeﬃcients.²¹Indirect OH yield measurements using both

the scavenger and tracer techniques agree reasonably well, and

an extensive, generally self-consistent database now exists.^1,3

The ﬁrst direct observation of OH production from alkene

ozonolysis was obtained by Donahue et al.,²⁵who employed

LIF to monitor OH production from a range of small alkenes at

low pressure (4−6 Torr). In subsequent work,³²⁻³⁴OH yields

were found to be pressure-dependent for substituted alkenes,

with fast OH formation (time scales of a few ms) eﬀectively

quenched as pressures approach ca. 400 Torr; however, on long

time scales (ca. 1 s), the OH yield increased, approaching

values consistent with other longer-time scale atmospheric

pressure measurements from radical scavenger/tracer experi-

ments, indicating that the majority of the observed OH for

substituted alkenes under tropospheric conditions arises from

the decomposition of thermalized carbonyl oxides and/or vinyl

hydroperoxides, with a contribution from prompt decom-

position (although the latter will dominate for ethene and the

CH₂OO Criegee intermediate formed from terminal alkenes).⁸

These results indicated that OH formation from alkene

ozonolysis can result from both prompt formation, from a

vibrationally excited CI, and from the decomposition of a

partially stabilized CI on longer time scales,³therefore opening

the possibility for bimolecular reactions to occur with the SCI,

interrupting the decomposition process and reducing the OH

yields. This possibility has been investigated for various SCI

scavengers, including H₂O, SO₂, butanone, and acetic acid for

2-methyl-2-butene;¹⁴CH₃CHO for trans-2-butene;¹⁰and NO₂

for 2,3-dimethyl-2-butene, trans-5-decene, cyclohexene, and α-

pinene,³⁵with little reduction in (inferred) OH yield found,

indicating that production of OH through the unimolecular

decomposition of partially stabilized CIs is signiﬁcantly larger

than the rate of reaction with the potential SCI scavengers

studied, at least for those species and experimental conditions

employed. Subsequent direct measurements of steady-state OH

abundance and hence overall production, using LIF systems

incorporated within simulation chambers, have been reported

for 2,3-dimethyl-2-butene, 2-methyl-2-butene, trans-2-butene,

and α-pinene,³⁶for isoprene,²⁴and for ethene.²⁹Overall, the

substantial experimental database for OH yields on long

The vinoxy radical formed alongside OH (Figure 2) will

react with oxygen in the atmosphere to form an excited β-oxo

peroxy radical, which may be stabilized or undergo decom-

position forming CO, a (secondary) stable carbonyl species and

OH.⁹However, this pathway to OH formation is only thought

to be signiﬁcant if an aldehydic hydrogen is present.³The

stabilized β-oxo peroxy radical may then undergo self- or cross-

reaction with other peroxy radicals to form stable species such

as glyoxal, methylglyoxal, glycolaldehyde, peroxides, and further

secondary carbonyls.

The fate of the anti-CI and of the CH₂OO CI formed from

terminal alkenes diﬀers in that the hydroperoxide rearrange-

ment is not available. Brieﬂy, the anti-CI (and CH₂OO) can

undergo rearrangement through a dioxirane structure, which

can decompose to various products including OH, HO₂, CO,

CO₂, H₂O, and alkyl molecules via a hot acid/ester

intermediate.¹The syn- and anti-CIs can also undergo

stabilization followed by bimolecular reaction, but studies

suggest that stabilization is a minor process for disubstituted

and syn-monosubstituted CIs, as their lifetime with respect to

the vinyl hydroperoxide mechanism is thought to be

substantially shorter than the time required for bimolecular

processes to occur,^10,11although recent experiments have

shown that the vinyl hydroperoxide intermediate itself may

have an appreciable lifetime.^8,12Collisional stabilization (or

possible direct formation) of SCIs may be followed by a range

of bimolecular reactions with atmospherically relevant species

such as H₂O, NO₂, SO₂, and CO,^1,13,14alongside thermal

decomposition. Very recent results indicate that reactions of the

SCIs with SO₂and NO₂proceed much faster than previously

thought and that the reaction with water is comparatively slow,

suggesting an important role for SCIs as atmospheric

oxidants.¹⁵⁻¹⁷

Laboratory Measurements of HO_xRadical Production

from Alkene Ozonolysis. The ﬁrst indications for the

production of radical species, including OH, from gas-phase

alkene ozonolysis reactions arose through observations of

chemiluminescence, attributed to emission from vibrationally

and electronically excited OH, observed at low pressures (1 −

10 Torr) from the reaction between ozone and a range of short

chain alkenes (C₂− C₆);¹⁸the observed OH was attributed to

the decomposition of an excited Criegee intermediate,

generating H atoms that would rapidly be converted to OH

19,20

radicals via reaction with O_3.

Direct formation of OH

through ozonolysis reactions at atmospheric pressure via a vinyl

hydroperoxide channel was ﬁrst proposed for 2,3-dimethyl-2-

butene by Niki et al.,⁷who measured the reaction stoichiometry

of Δ[2,3-dimethyl-2-butene]/Δ[O₃] to derive an OH radical

yield of 0.70 0.10.

A number of studies have followed, reporting quantitative

indirect and direct OH yield measurements for the ozonolysis

of various alkenes, using one of three generalized approaches:

(i) indirect measurements, where a radical scavenger or

tracer(s), reactive toward OH but not toward O₃or other

reactants, are used to infer the cumulative OH production in

the system at long (minutes +) time scales;²¹⁻²⁴(ii) direct

observations using techniques such as laser-induced ﬂuores-

cence (LIF),²⁵peroxy radical chemical ampliﬁcation

(PERCA),²⁶or matrix isolation−electron spin resonance

(MIESR)²⁷are used to quantify HO_xradicals under steady-

state conditions or (in laboratory settings) at short reaction

12470

dx.doi.org/10.1021/jp408745h | J. Phys. Chem. A 2013, 117, 12468−12483

The Journal of Physical Chemistry A

Article

(seconds +) time scales relevant to the impact of ozonolysis

reactions upon the atmosphere is broadly consistent across a

range of direct and indirect approaches.

for a quantitative understanding of atmospheric oxidation.

Improved understanding is also necessary to address the

potential for other atmospheric reactants, such as water vapor,

to aﬀect the OH (and in particular HO₂) radical yields.

Atmospheric chemical mechanisms employ radical (HO₂, RO₂)

yields, which are largely inferred through the observation of

associated stable products on the basis of the mechanistic

understanding outlined above³⁹and, in the case of HO₂, are

substantially lower than those suggested by some of the studies

noted above.^28,38In this article, we present systematic

measurements of radical yields from the ozonolysis of propene,

1-butene, 2-methylpropene, cis-2-butene, trans-2-butene, 2,3-

dimethyl-2-butene, and isoprene, obtained through simulation

chamber experiments performed in the EUPHORE (European

Photoreactor) facility, and employing LIF and PERCA

techniques to monitor OH (directly), HO₂, and RO₂radicals.

In contrast to the substantial body of experimental data on

OH production, the observational database for direct HO₂

production from alkene ozonolysis is sparse and inconsistent;

even more so for RO₂. In the absence of experimental data,

some modeling studies have necessarily assumed a 1:1

stoichiometry for production of OH and a radical coproduct;³⁷

however, the existence of secondary channels for the

production of OH⁹and for the production of two HO₂radicals

from the CH₂OO CI formed from terminal alkenes^1,29indicate

that this supposition may not hold. Mihelcic et al.²⁷used the

matrix isolation−electron spin resonance (MIESR) approach to

measure the OH and HO₂yields from ethene ozonolysis,

obtaining yields of 20 2 and 39 3%, respectively. Radical

production in the ethene−ozone system has also been

investigated using a laboratory PERCA methodology combined

with detailed modeling, to extract yields of HO₂and HO₂+

RO₂of 38 2 and 45 3%, respectively;²⁶subsequently, this

approach was extended to propene, with yields of 39 8%, 19

2. APPROACH: EUPHORE CHAMBER EXPERIMENTS

Detailed accounts of EUPHORE, the monitoring techniques

used and experimental procedure followed can be found

elsewhere;^40,41,29a brief overview is provided here. The

EUPHORE facility comprises two ca. 200 m³reaction

chambers, formed from FEP ﬁlm, with housings that may be

closed to exclude light. Chamber pressure is maintained ca. 100

Pa above ambient and chamber temperatures at near ambient

levels (cooled through a chilled ﬂoor). For all the experiments

described here, the chamber covers were closed, excluding

ambient sunlight (j(NO₂) < 2 × 10⁻⁶s⁻¹). The chambers are

mixed by fans on a time scale of 3 min. The chamber is ﬁlled

with scrubbed dry air, to which reactants of interest are added

volumetrically by direct injection, and the evolving composition

monitored as a function of time. Dilution (arising from sample

withdrawal, as total chamber pressure is maintained by addition

of scrubbed air) is monitored by following the decay of SF₆,

added as a tracer prior to each experiment, and during the work

described here typically corresponded to a ﬁrst-order decay rate

of ca. 10⁻⁵s⁻¹, accounting for between 2 and 42% of the alkene

removal across all experiments (values for 2,3-dimethyl-2-

butene and propene, respectively; balance through chemical

processes). Alkene and stable product concentrations were

monitored by Fourier-transform infrared absorption spectros-

copy (FTIR), and by chemical ionization reaction time-of-ﬂight

mass spectrometry (CIR-TOF-MS) in the case of oxygenated

products. Levels of O₃(UV photometry), NO_x(chemilumi-

nescence with photolytic converter; below detection limit for all

(dark, NO_x-free) experiments described here), and CO

(infrared absorption) were followed by conventional monitors.

Chamber humidity was measured using a chilled-mirror dew-

point hygrometer. OH radicals were observed directly using

laser induced ﬂuorescence spectroscopy (LIF), while HO₂

radicals were detected following their titration to OH by

added NO within the LIF instrument.^36,42The sum of hydro-

and organic peroxy radicals, HO₂+ ΣRO₂, was measured using

a peroxy radical chemical ampliﬁer (PERCA).⁴³

4%, and 39

8% obtained for OH, HO₂, and CH₃O₂,

respectively.³⁸Radical yields have been deduced from experi-

ments in the SAPHIR simulation chamber, in which the decay

of the additional turnover of the alkene and/or ozone, in

experiments performed with/without an OH scavenger (excess

CO) were used to infer OH and HO₂yields.²⁸In contrast to

previous tracer/scavenger studies, the OH yields obtained

varied with humidity, increasing in the presence of H₂O, and

the inferred (dry) HO₂yields (propene, 1.50 0.75; 1-butene,

1.60

0.80; and isobutene, 2.00

1.00) were substantially

greater than those reported in the direct measurements (and

assumed in most atmospheric mechanisms). Most recently, the

LIF/chemical conversion technique was used to determine an

HO₂yield of 0.26 0.03 for isoprene ozonolysis in a small

simulation chamber.²⁴

The existing experimental data set is therefore in reasonable

agreement with regard to OH production (yields) from alkene

ozonolysis, which are broadly consistent with the dominant OH

production channel for substituted alkenes being the hydro-

peroxide mechanism for the syn-CI, with additional OH

production potentially arising from further decomposition of

the coformed alkoxy radical and from decomposition of the

anti-CI conformer.³In the case of terminal alkenes and ethene,

decomposition of the C₁CI, CH₂OO, also leads to direct OH

production.²⁹The mechanistic origin and yield of HO₂is rather

less clear; in the ethene system, HO₂is produced through

decomposition of the CH₂OO CI (potentially via a hot acid

intermediate), while for substituted alkenes the literature

observations of substantial HO₂production noted above

indicate that further decomposition of one or both CIs (in

addition to OH production via the hydroperoxide mechanism)

may occur. The experimental yields of HO₂reported are

substantially greater than the OH yields in the limited number

of studies reported (e.g., ethene; ca. 40% for HO₂compared

with 10−20% for OH across a number of measurements), with

substantial inconsistency apparent between studies; this may in

part reﬂect the greater complexity of interpretation of HO₂data

in the presence of multiple chemical source and sink processes.

As NO-driven RO₂−HO₂−OH radical cycling within the

atmosphere is likely to be signiﬁcant in most environments

where the alkene abundance is substantial, characterizing non-

OH radical production from alkene ozonolysis is of importance

Instrumental Accuracy. The results reported in this study

were derived using a comprehensive suite of measurement

instrumentation. Alkene and ozone concentrations were

determined by FTIR and converted to mixing ratios through

the measured chamber temperature and pressure. Systematic

uncertainties arising from the literature cross-sections were

conservatively estimated to be 10% (the FTIR O₃data agreed

very well with measurements made by UV photometry). The

radical measurements were performed by LIF (OH and HO₂)

12471

dx.doi.org/10.1021/jp408745h | J. Phys. Chem. A 2013, 117, 12468−12483

The Journal of Physical Chemistry A

Article

and PERCA (HO₂+ ΣRO₂). The LIF system was calibrated

before, during, and after each measurement campaign using the

H₂O photolysis/ozone actinometry approach,⁴⁴and calibra-

tions were constant to within a few percent. The estimated

uncertainty in the LIF data from a single calibration is 27%

(combined systematic error and precision)⁴²although this may

be regarded as conservative as the precision component of this

uncertainty within a single calibration will, to a certain extent,

average out across the experiments. The instrument was

calibrated as a function of humidity; in keeping with previous

results⁴⁵a reduction in the instrument sensitivity is observed as

humidity increases compared with that measured under dry

conditions, but the response ﬂattens out (and was fully

characterized) within the humidity range employed here. The

potential for interference from substituted peroxy radicals in the

LIF HO₂measurement is considered explicitly in the analysis

section below. The PERCA was calibrated using the photolysis

of methyl iodide/air mixtures to form methylperoxy radicals,

thus determining the chain length of the chemical ampliﬁca-

tion.⁴⁶An NO₂permeation device was used to determine the

sensitively of the Scintrex (Luminol chemiluminescence

detector) to the NO₂product of the ampliﬁcation. The overall

accuracy of the peroxy radical measurements was estimated to

be 38% (2 SD) from the combined uncertainties associated

with the radical calibration, NO₂quantiﬁcation, and humidity

correction.⁴⁷

Experimental Procedure. All experiments were initiated in

a clean (ﬂushed overnight) dry chamber. Ozone (generated

online by corona discharge) and (if necessary) water vapor

and/or OH scavengers (CO or cyclohexane) were added to the

required levels, followed by an aliquot of SF₆(as a tracer for

dilution). The ozonolysis reaction was initiated by injection of

the relevant alkene using a graduated syringe, with reactant and

product concentrations monitored over a time scale of 1−2 h,

at a time resolution ranging from 1 min (LIF, PERCA) to 10

min (FTIR scan period), which may be compared with the

chamber mixing time of 3 min. Three sets of experimental

conditions were employed: (i) simple alkene−ozone only

systems; (ii) experiments with added excess CO (ca. 500 ppm)

as an OH scavenger, to suppress the OH + alkene reaction; and

(iii) experiments performed with added water vapor (to 30%

RH); the dry chamber humidity was approximately 0.3%

(measured dew point −45 °C). The experimental conditions

and chemical systems studied are summarized in Table 1.

detailed chemical model, within which the yields of OH and

HO₂were optimized to match the observed data. The model

reaction scheme was based upon the relevant subset of the

Master Chemical Mechanism (MCM) version 3.2^39,48for the

degradation of each alkene in question, including inorganic

reactions and updated cyclohexane degradation chemistry as

described in ref 29. Model conditions (temperature, pressure,

and dilution rate) were set to the corresponding mean value for

each experiment (as variability in these parameters was

negligible during each run) and simulations initialized with

the relevant experimental conditions as outlined in Table 1.

Within the model, the actual ozonolysis process (i.e., Criegee

decomposition reactions) were updated from the MCM

representation to reﬂect a more detailed description of the

actual mechanism: The primary ozonide (POZ) was assumed

to decompose rapidly, and then each initially formed (excited)

Criegee intermediate to further decompose to form radical

products and stable species or to undergo collisional

stabilization forming a stabilized Criegee intermediate (SCI),

from which no further radical formation occurred. These

reactions (2−6 in Figure 2) were not assigned individual rate

constants, as the process occurred eﬀectively instantaneously on

the time scale of the experiments (of the order of minutes), but

rather a branching ratio for each potential decomposition/

stabilization product was used. This approach implicitly groups

together CI, SCI, and vinyl hydroperoxide decomposition

routes, irrespective of the diﬀering time scales (microsecond vs

millisecond) over which they may occur, as long as these are

shorter than the experimental time scale (2−3 min, determined

by the chamber mixing time). A consequence of this approach

is that processes such as SCI thermal decomposition would be

interpreted here as formation of the resulting decomposition

products, rather than of SCIs, as long as such transformations

occurred on a time scale of the order of tens of seconds or

faster; this consideration must be noted when comparing the

results obtained here to those reported from (e.g.) millisecond

time scale laboratory measurements, considered further below.

Similarly, the decomposition/O₂reaction converting HCO into

HO₂+ CO and H atoms into HO₂were implicitly adopted,

following the MCM protocol.³⁹Rate constants for the

subsequent bimolecular reactions of the SCI with CO and

H₂O were taken directly from the MCM v3.2. The impact of

recent measurements, suggesting some SCI reactions may be

much faster than the values used within the MCM,¹⁵is

discussed later.

An iterative approach was adopted to derive the OH and

HO₂yields, to account for the interdependence between these

quantities. It was not possible to derive both yields from a

single experiment, due to interferences in the HO₂measure-

ment from substituted organic peroxy radicals, formed

following the OH + alkene reaction, in the absence of an

OH scavenger (see discussion below). Three stages were

followed: (i) OH yields were optimized, using measured OH

data obtained from experiments performed without an OH

scavenger (Table 1). The HO₂concentrations measured in

these runs were found to include a substantial interference

component from substituted peroxy radicals (measurements

hereafter referred to as HO₂*, see below) and were not used to

constrain the optimization; rather HO₂yields were ﬁxed at

MCM v3.2 values. (ii) HO₂yields were obtained using

measured HO₂data obtained from experiments performed

with an OH scavenger present (in which the interferant organic

peroxy radicals are absent). Finally, (iii) the OH yields were

3. APPROACH: DATA ANALYSIS

To account for secondary chemical processes (radical sinks and

additional sources), the measurements were interpreted using a

Table 1. Range of Initial Conditions Used for EUPHORE

a

Experiments

species

propene

alkene (ppb)

O₃(ppb)

H₂O range (RH %)

300

0.3−30%

0.3%

1-butene

300

cis-2-butene

100

200

trans-2-butene

2-methylpropene

2,3-dimethyl-2-butene

isoprene

100−200

30−50

100−200

200−500

25−30

200−300

0.3−30%

0.3%

a

Additional reagents SF₆(dilution tracer) and CO (OH scavenger)

added for some experiments.

12472

dx.doi.org/10.1021/jp408745h | J. Phys. Chem. A 2013, 117, 12468−12483

The Journal of Physical Chemistry A

Article

Figure 3. (a) Temporal proﬁle of OH (red circles) plus model simulations before (dashed line, base case MCM v3.2 chemistry) and after (solid line)

optimizing the OH yield to the LIF data (see Table 1), for cis-2-butene ozonolysis. (b) Equivalent plot for HO₂measurements and model

predictions (MCM v3.2 and following yield optimization) for 1-butene ozonolysis.

reoptimized, from nonscavenger experiments, using the actual

HO₂yields obtained in the second stage. Optimization was

performed within the FACSIMILE solver.⁴⁹In principle one

could repeat the iteration to continue reﬁnement; however, in

practice, the change in OH yields resulting from this single

repetition was <5% in all cases. Sensitivity analyses were

performed to assess the cross-dependence of the OH and HO₂

yields obtained following this approach. For the worst case

system considered (propene, slowest ozonolysis reaction rate,

hence largest contribution from secondary HO₂/OH inter-

conversion via reaction with O₃) varying the HO₂channel yield

from 0.08 (as shown for reaction R10d, below) to 0.77 (the

upper limit possible considering observations of stable product

channels R10b and R10c)⁵⁰corresponded to a reduction in OH

yield from 0.36 to 0.29, reﬂecting the relatively limited

contribution of HO₂+ O₃to OH production.

collisional stabilization was optimized in order to obtain an OH

yield, taking into account any secondary OH formation through

inclusion of the full degradation chemistry within the model.

Decomposition yields for the most simple (C₁) CI, CH₂OO*,

formed from all terminal alkenes, were ﬁxed at the values

determined by Alam et al.²⁹Stable product yields for the

substituted CIs were taken from the literature¹and have

minimal impact upon the derived radical yields under the

conditions of these experiments. The ratio of the two CIs

formed (i.e., the value of α) was determined from stable

product yields;²⁹the values used for propene, 1-butene, and 2-

methylpropene were 51:49 for [CH₃CHOO]* + HCHO

(R8a)/[CH₂OO]* + CH₃CHO (R8b), 59:41 for

[C₂H₅CHOO]* + HCHO/[CH₂OO]* + CH₃CH₂CHO, and

34:66 for [CH₂OO]* + CH₃COCH₃/[(CH₃)₂COO]* +

HCHO, respectively. In the cases of cis- and trans-2-butene,

50:50 yields of the relevant CIs were used.²²The optimization

approach adopted to obtain HO₂yields for each system studied

was as follows. As noted above, degradation products of the

Within the modeled reaction scheme, the branching ratio

between isomerization/decomposition of the syn-CI producing

OH (via the hydroperoxide mechanism) and its potential

12473

dx.doi.org/10.1021/jp408745h | J. Phys. Chem. A 2013, 117, 12468−12483

The Journal of Physical Chemistry A

Article

Table 2. OH Yields, All under Dry Conditions (RH < 0.4%) unless Noted

parent alkene; CIs formed

OH yield

0.17 0.09

method

LIF

ref

ethene

Alam et al.²⁹

CH₂OO

0.16

IUPAC⁵⁰

MCM³⁹

0.13

propene

0.36 0.10

0.34

LIF

this study

IUPAC⁵⁰

CH₂OO

0.36

MCM³⁹

CH₃CHOO

0.33 (+0.17/−0.11)

0.18 0.01

0.34 0.06

0.35 0.07

0.32 0.08

0.40 0.06

0.10 0.07

0.30 0.08 (RH ≈ 32%)

0.39 0.08

0.56 0.15

0.36

cyclohexane

CO

Atkinson & Aschmann⁶⁸

Gutbrod et al.³¹

Neeb & Moortgat⁶⁹

Paulson et al.²¹

Rickard et al.²²

Aschmann et al.⁷⁰

Wegener et al.²⁸

Qi et al.³⁸

cyclohexane

tracer

2,3-butandiol

stoichiometry

PERCA

1-butene

LIF

this study

MCM³⁹

CH₂OO

0.41 (+0.21/−0.14)

0.29 0.04

0.26

cyclohexane

tracer

Atkinson & Aschmann⁶⁸

Paulson et al.²¹

Fenske et al.¹⁰

Wegener et al.²⁸

this study

CH₃CH₂CHOO

tracer

0.00 0.08

0.30 0.09 (RH ≈ 32%)

0.67 0.18

0.62

stoichiometry

LIF

2-methylpropene

IUPAC⁵⁰

CH₂OO

0.82

MCM³⁹

(CH₃)₂COO

0.84 (+0.42/−0.28)

0.60 (+0.05/−0.07)

0.72 0.12

0.60 0.15

0.30 0.14

0.80 0.10 (RH ≈ 32%)

0.26 0.07

0.33

cyclohexane

tracer

Atkinson & Aschmann⁶⁸

Neeb & Moortgat⁶⁹

Paulson et al.²¹

Rickard et al.²²

Wegener et al.²⁸

this study

tracer

stoichiometry

cis-2-butene

LIF

IUPAC⁵⁰

CH₃CHOO

0.57

MCM³⁹

0.41 (+0.21/−0.14)

0.14 0.03

0.17 0.02

0.33 0.07

0.33 0.05

0.18 0.09

0.40 0.05 (RH ≈ 32%)

0.63 0.17

0.64

cyclohexane

stoichiometry

CO

Atkinson & Aschmann⁶⁸

Horie et al.⁷¹

Gutbrod et al.³¹

McGill et al.⁷²

Orzechowski & Paulson⁷³

Wegener et al.²⁸

this study

IUPAC⁵⁰

MCM³⁹

Atkinson & Aschmann⁶⁸

Horie et al.⁷¹

tracer

stoichiometry

trans-2-butene

LIF

CH₃CHOO

0.57

0.64 (+0.32/−0.21)

0.24 0.05

0.24 0.02

0.68 0.09

0.54 0.11

(ca. 0.60)

cyclohexane

stoichiometry

CO

Gutbrod et al.³¹

Donahue et al.²⁵

McGill et al.⁷²

Kroll et al.³²

LIF (5 Torr)

tracer

LIF (low P)

LIF

0.75 0.19

0.64 0.12

0.54 0.05

0.52 0.04 (RH ≈ 32%)

0.70 0.12 (RH ≈ 32%)

0.60 0.12 (RH ≈ 32%)

0.83 0.22

0.90

Siese et al.³⁶

tracer

Orzechowski & Paulson⁷³

Hasson et al.⁷⁴

Wegener et al.²⁸

this study

tracer

stoichiometry

LIF

2,3-dimethyl-2-butene

(CH₃)₂COO

IUPAC⁵⁰

MCM³⁹

Atkinson & Aschmann⁶⁸

Chew & Atkinson³⁰

1.00

1.00 (+0.5/−0.33)

0.80 0.12

cyclohexane

2-butanol

12474

dx.doi.org/10.1021/jp408745h | J. Phys. Chem. A 2013, 117, 12468−12483

The Journal of Physical Chemistry A

Table 2. continued

Article

parent alkene; CIs formed

OH yield

0.36 0.02

method

CO

ref

Gutbrod et al.³¹

Donahue et al.²⁵

Rickard et al.²²

Fenske et al.¹⁰

Kroll et al.³²

Siese et al.³⁶

Orzechowski & Paulson⁷³

Aschmann et al.⁷⁰

0.70 0.03

0.89 0.22

0.99 0.18

(ca. 1.00)

LIF (5 Torr)

tracer

LIF (low P)

LIF

1.00 0.25

0.91 0.14

1.07 0.16

tracer

2,3-butandiol

CH₂OO* CI were ﬁxed at the values determined for ethene²⁹

for all systems, and OH yields derived as described above were

used for all systems. The modeled HO₂treatment necessarily

varied between alkenes: For propene ozonolysis (see example

below), any additional HO₂formation (over and above that

from CH₂OO) was assumed to result from the decomposition

of CH₃CHOO*; accordingly, the branching ratio between

decomposition (via the hot acid channel, producing HCHO +

HO₂+ HO₂+ CO) and stabilization was optimized. The

overall HO₂yields reported here are then the sum of

contributions from both CI channels. The data for 1-butene

were analyzed in an equivalent manner. For cis-2-butene and

trans-2-butene, CH₂OO* is not formed, and HO₂formation

within the model was constrained via the competition between

decomposition through the hot acid route and CI stabilization.

For 2-methylpropene, our basic understanding of the

ozonolysis mechanism indicates that HO₂formation is

restricted to [CH₂OO]*. Thus, the CH₂OO decomposition

products were optimized to derive an HO₂yield for 2-

methylpropene following the approach adopted previously for

ethene ozonolysis.²⁹While the hydroperoxide mechanism

suggests little direct HO₂production would be expected for

2,3-dimethyl-2-butene (all CIs are formed with a syn-hydrogen

atom), this assumption was tested by inclusion of an HO₂-

producing CI decomposition channel and optimizing the

competition between this and the alternative (OH production)

route.

CH₃CHOO* → HCHO + HO₂+ HO₂+ CO 0.08¹

(R10d)

CH₃CHOO* → CH₃CHOO (0.69 − β)⁵¹

(R10e)

Interpretation of Yields Obtained. The OH and HO₂

yields obtained represent the overall yields from the fast CI

decomposition chemistry (sum of both channels), relative to

the ﬂux through the initial alkene−ozone reaction, as

highlighted in the reaction scheme above. It is important to

note that interpretation of these yield values does not

necessarily imply the production mechanism above; the data

from this study do not determine the mechanistic origin of the

observed HO_xradicals. The model approach in principle

incorporates allowance for secondary chemical processes

(thermalizd bimolecular reactions), which aﬀect HO_xabun-

dance but does not account for potential rapid HO_xformation

following the ozonolysis processes; for example, some of the

detected OH production could originate from decomposition

of the excited β-oxo peroxy radical.⁹Secondary processes that

were explicitly included within the model include OH

formation via HO₂+ O₃(reaction R11) and from the reactions

of acyl peroxy radicals + HO₂(reactions R12, R13)⁵²⁻⁵⁴plus

further chemistry as included within the MCM v3.2

degradation mechanisms for the parent alkenes.

HO₂+ O₃→ 2O₂+ OH

(R11)

For example, in the case of propene ozonolysis, this approach

corresponds to

CH₃C(O)CH₂O₂+ HO₂

→ OH + CH₃C(O)CH₂O + O₂

(R12)

O₃+ C₃H₆→ POZ k7

(R7)

CH₃C(O)O₂+ HO₂→ OH + CH₃C(O)O + O₂

POZ → HCHO + CH₃CHOO*

α

(R8a)

(R8b)

(R9a)

(R9b)

(R9c)

(R9d)

(R13)

POZ → CH₃CHO + CH₂OO* (1 − α)

CH₂OO* → OH + HO₂+ CO 0.17²⁹

CH₂OO* → HO₂+ HO₂+ CO₂0.05²⁹

CH₂OO*→CO₂+ H₂0.24²⁹

4. RESULTS

Figure 3a shows a typical temporal proﬁle of the OH steady-

state concentration as measured by the LIF system and OH

model simulation comparisons for cis-2-butene ozonolysis.

These data illustrate that the MCM v3.2 overestimates (in the

case of cis-2-butene) the OH yield, compared to the

measurements obtained here; the MCM OH yield is 57%,

while the yield obtained here is 27 7%. [In this instance, it is

likely that this arises as a 50:50 split of the two CIs assumed in

the MCM 3.2 mechanism, while the lower yield obtained here

(and in other studies⁵⁵) suggests that the anti-conformer of the

CI is preferentially formed (alongside acetaldehyde) from the

primary ozonide decomposition in this system.⁶A further

contribution may in principle arise from diﬀering reactivity of

the syn- and anti-CIs with H₂O at the low (but not zero) RH

level of the dry experiments reported here (e.g., with H₂O)⁵⁶.]

The OH yields obtained from direct observation (LIF detection

CH₂OO*→CO + H₂O 0.24²⁹

CH₂OO* → CH₂OO 0.54²⁹

(R9e)

CH₃CHOO* → OH + HCOCH₂O₂

CH₃CHOO* → CH₃OH + CO 0.08¹

β

(R10a)

(R10b)

(R10c)

1

CH₃CHOO* → CH₄+ CO₂0.15

12475

dx.doi.org/10.1021/jp408745h | J. Phys. Chem. A 2013, 117, 12468−12483

The Journal of Physical Chemistry A

Article

of OH) are listed in Table 2, where the reported uncertainty for

the LIF data is 27%, representing combined systematic

calibration uncertainty and measurement precision ( 2 SD).

These values correspond to the overall formation of OH via the

(fast) direct decomposition/isomerization of the CI, through

the optimization approach described above, which takes into

account secondary sources as noted above.

Figure 3b shows a typical temporal proﬁle of the HO₂steady-

state concentration as measured by the LIF system, and HO₂

model simulation comparisons for 1-butene ozonolysis,

obtained in the presence of excess CO (OH scavenger). The

with the values measured in the presence of excess CO (which

eﬀectively suppressed the OH + alkene reaction). It is likely

that this reﬂects an interference in the LIF HO₂radical

measurements, arising from the presence of β-hydroxyalkyl

peroxy radicals, formed from the reaction of OH with the

parent alkene: Within the LIF sampling system, addition of NO

leads to the conversion of peroxy radicals to the corresponding

alkoxy species, which may then undergo further reaction

(primarily with O₂or NO as the dominant reagents present),

decompose or isomerize; if these reactions lead to signiﬁcant

production of HO₂, on the time scale (ms) and under the

conditions (<1 Torr, reduced temperature arising from the

supersonic gas expansion) of the LIF system sampling chamber,

a positive interference in the HO₂measurement will result.

While this conversion (e.g., via RO + O₂) has been shown⁵⁷to

be too slow to have a signiﬁcant eﬀect for small (<C₄) RO₂,

recent studies⁵⁸have shown that the interference can be

signiﬁcant for larger peroxy radicals (for which the reactions of

the derived alkoxy radical with O₂are faster), and in particular

for the β-hydroxyalkyl peroxy radicals, such as those formed

from addition of OH to alkenes; for these species, the

corresponding β-hydroxyalkoxy radical reacts rapidly with O₂,

forming HO₂, on the time scale of the LIF system. HO₂

interferences from peroxy radicals ranging from 0.04 (for

methyl peroxy radicals, CH₃O₂) to 0.95 for β-hydroxy-propyl-

peroxy radicals (CH₃CH(OH)CHO₂, formed from OH +

propene) have been reported.⁵⁸While the magnitude of the

interference has been shown to vary with the speciﬁc

instrument design and operational parameters (pressures, NO

concentration, inlet nozzle design, and hence, gas expansion

properties), the characteristics of the EUPHORE LIF system

are similar to those of (one of) the systems tested⁵⁸(0.4 mm

nozzle conﬁguration), so similar interference behavior is

expected. The implication for this work is that LIF HO₂

observations performed in the absence of a radical scavenger

are likely to signiﬁcantly overestimate the true HO₂

concentration present; we therefore disregard these HO₂*

observations from the subsequent analysis and discussion (the

values obtained are included in Figure 7 for information only).

Where a scavenger such as CO was used in excess (in all cases

here, such that ≥97% of the OH formed reacted with the

scavenger), formation of β-hydroxyalkyl peroxy radicals was

essentially suppressed.

The RO₂interference eﬀect is clearly illustrated in Figure 4,

which shows the temporal proﬁle of HO₂retrieved using the

LIF system during the ozonolysis of propene, and the response

of the measured levels to the sequential addition of CO and

H₂O. Sections 1−3 of Figure 4 indicate those stages of the

experiment for which only propene and ozone, and then excess

CO, and then additionally H₂O (RH 30%) were present,

respectively. The dashed line in Figure 4 represents the

modeled HO₂using a constant overall HO₂yield of 0.09

(obtained using section 2 of the experiment). In all cases, the

modeled OH yield was ﬁxed at 0.36. The ﬁrst stage of the

experiment illustrates the β-hydroxyalkyl peroxy radical HO₂

interference eﬀect noted above; the HO₂levels are over-

estimated. Following the addition of CO, the retrieved HO₂

increases slightly (increased through the conversion of OH to

HO₂, oﬀset by removal of the interferant RO₂species); this part

of the experiment represents the base case conditions for

retrieval of the actual HO₂yield. In the third part of the

experiment, water is added to increase the relative humidity to

ca. 30%, and the observed HO₂levels (shown corrected for the

yields of HO₂(Y_HO) obtained are listed in Table 3, together

2

with equivalent values from the recent literature. The inferred

Y_HOobtained in this work were found to be substantially larger,

2

for the same chemical systems, in the absence of an OH radical

scavenger (i.e., in alkene/ozone only experiments), compared

Table 3. HO₂Yields, All under Dry Conditions (RH < 0.4%)

unless Noted

parent alkene; CIs

formed

HO₂yield

method

LIF

refe

ethene

0.10 0.03

Alam et al.²⁹

[CO system]

0.05 0.01

(RH ≈ 30%)

0.13

LIF

Alam et al.²⁹

[CO system]

CH₂OO

MCM³⁹

0.39 0.03

0.38 0.02

0.50 0.25

MIESR

PERCA

Mihelcic et al.²⁷

Qi et al.²⁶

stoichiometry Wegener et al.²⁸

0.40 0.20

(RH ≈ 32%)

propene

0.09 0.02

LIF

this study

0.02 0.00

(RH ≈ 30%)

0.28

CH₂OO

MCM³⁹

CH₃CHOO

1.50 0.75

stoichiometry Wegener et al.²⁸

1.15 0.60

(RH ≈ 32%)

0.19 0.04

0.18 0.05

0.28

PERCA

LIF

Qi et al.³⁸

this study

MCM³⁹

1-butene

CH₂OO

1.60 0.80

stoichiometry Wegener et al.²⁸

CH₃CH₂CHOO

1.60 0.80

(RH ≈ 32%)

2-methylpropene

0.34 0.09

LIF

this study

0.38 0.10

(RH ≈ 30%)

0.41

CH₂OO

MCM

MCM³⁹

(CH₃)₂COO

2.00 1.00

stoichiometry Wegener et al.²⁸

1.60 0.80

(RH ≈ 32%)

cis-2-butene

CH₃CHOO

trans-2-butene

CH₃CHOO

0.12 0.03

LIF

this study

MCM³⁹

0.125

MCM

LIF

0.04 0.01

this study

0.00 0.01

(RH ≈ 30%)

0.125

LIF

MCM

LIF

MCM³⁹

2,3-dimethyl-2-

butene

0.18 0.05

this study

(CH₃)₂COO

isoprene

0.00

MCM

LIF

MCM³⁹

0.16 0.04

0.26 0.03

0.26

this study

Malkin et al.²⁴

MCM³⁹

CH₂OO,

MACROOA,

MVKOOA

LIF

MCM

12476

dx.doi.org/10.1021/jp408745h | J. Phys. Chem. A 2013, 117, 12468−12483

The Journal of Physical Chemistry A

Article

the observed concentrations of HO₂+ ΣRO₂as a function of

time, together with a model simulation (with OH and HO₂

yields optimized corresponding to the values given in Tables 2

and 3, respectively). As the PERCA observations were limited,

we have not included these data in the yield derivation process.

The modeled and observed total peroxy radical levels are in

agreement, within uncertainty, during the ﬁrst and second

stages of the experiment. Following the increase in relative

humidity from 1.3 to 20.4% (stage 3), the observed radical

concentrations decrease notably, while the modeled levels are

almost unaﬀected (the very slight change in gradient for the

model trace reﬂecting the variation in HO₂self-reaction rate

constant with humidity, which is included in the model and

indirectly aﬀects RO₂levels (through the RO₂+ HO₂sink) in

addition to HO₂directly); the modeled HO₂/RO₂ratio is

approximately equal (to within 4%) in stages two and three.

The change is consistent with the reduction in observed HO₂

levels (Figure 4) with increasing humidity, to a greater extent

than can be explained by the modeled secondary chemical

processes.

Figure 4. Observed temporal proﬁle of HO₂(open circles) plus model

simulation during a propene ozonolysis experiment. Vertical dotted

lines indicate periods during which CO and H₂O were added. During

period 1, the RO₂interference elevates the observed HO₂(HO₂*)

above that modeled. During period 2, production of the interferant

peroxy radicals is suppressed, and the measured data is used to obtain

the HO₂yield. This yield is used throughout the simulation to

generate the blue dashed model line. During period 3, after the

addition of H₂O, the observed HO₂falls to a greater extent than the

simulation predicts.

5. DISCUSSION

Figure 6 shows the direct OH yields obtained for ethene,

propene, 1-butene, 2-methylpropene, cis-2-butene, trans-2-

butene, and 2,3-dimethyl-2-butene obtained from this work,

in comparison with those from other previous indirect studies

plotted as a function of the equivalent IUPAC recommenda-

tions.⁵⁰The uncertainties in the results from this work

represent the combined (2σ) statistical uncertainty from

repeated determinations propagated with the corresponding

OH measurement calibration uncertainty. The results are well

correlated with the IUPAC values; as the literature studies

mainly exploit indirect methods to detect OH, by the use of

OH scavenger⁵⁹and tracer²²techniques, or indirect observa-

tion by matrix-isolation electron spin resonance²⁷and

PERCA,³⁸the agreement with the direct OH observations in

this work indicates that the overall OH radical yields from the

ozonolysis of small alkenes is well characterized and that the

radical data obtained here are representative of the ﬁnal, long-

time scale yields following relaxation of the CI system

applicable to chemical modeling of the ambient atmospheric

boundary layer.

LIF system calibration humidity dependence) decrease. The

inference from Figure 4 then is that interference eﬀects increase

the retrieved HO₂in the absence of CO (section 1); the HO₂

yield in the absence of H₂O is 9% (section 2), and upon the

addition of H₂O, the HO₂levels decrease, potentially to a

greater extent than can be accounted for by the humidity

dependence of the HO₂recombination reaction (which is

included in the model), corresponding to a reduction in the

HO₂yield from propene ozonolysis with increasing humidity

(section 3).

Experimental diﬃculties limited the PERCA observations to

two chemical systems, ethene²⁹and propene. Figure 5 shows

The diﬀerence between the OH yields as shown in Figure 6

and a value of 1.0 reﬂects (a lower limit to) the formation of

other products from CI decomposition and/or stabilization. In

the case of 2,3-dimethyl-2-butene, all CIs formed will have the

syn conﬁguration and might therefore be expected to undergo

the hydroperoxide mechanism rearrangement, corresponding

to an OH yield of unity (the basis of the OH yield structure−

activity relationship²²); however, a yield of 0.83

0.22 is

obtained here, in common with a number of other long time

scale studies (Table 2), and non-negligible HO₂formation is

also observed (0.18

0.05; Table 3). As OH may also be

produced through other routes (e.g., by decomposition of the

excited β-oxo-peroxy radical), the proportion of the CIs that do

not directly yield OH may actually be somewhat larger than

17% (inferred from the corresponding OH yield of 0.83). The

experimental conditions (pressure, time scale) are critical to

such interpretations: Drozd et al.⁸have illustrated the complex

variation in OH and SCI production from 2,3-dimethyl-2-

butene ozonolysis, showing that time- and pressure-dependent

OH yields arise from at least two decomposition vs.

stabilization routes, from the initial SCI and the vinyl-

Figure 5. Measured temporal proﬁle of HO₂+ ΣRO₂(PERCA data,

ﬁlled circles) during propene ozonolysis, plus model simulations after

optimizing HO_xyields using the LIF data (PERCA data not used in

optimization): blue long-dashed line, modeled HO₂; red short-dashed

line, modeled RO₂; solid black line, modeled HO₂+ RO₂. Dotted

vertical lines indicate periods during which CO and H₂O were

introduced to the chamber, as in Figure 4

12477

dx.doi.org/10.1021/jp408745h | J. Phys. Chem. A 2013, 117, 12468−12483

The Journal of Physical Chemistry A

Article

Figure 6. Comparison of measured OH yields from this work and selected literature studies vs. IUPAC OH yield recommendations. Filled green

circles, this work; ﬁlled blue squares, Atkinson et al.,⁵⁹cyclohexane scavenger; open magenta circles, Marston and co-workers,^22,72TMB scavenger;

ﬁlled black diamond, Qi et al.,³⁸PERCA; ﬁlled red triangle, Mihelcic et al.,²⁷MIESR. The dashed line shows the 1:1 correlation.

Figure 7. Comparison of HO₂yields for small chain alkenes investigated during this study with literature. Filled green circles and open green

triangles are HO₂yields calculated from this study (by LIF) for excess CO and nonscavenged (i.e., with RO₂interference) experiments, respectively.

Filled red triangles, Wegener et al.;²⁸open black squares, MCM v3.2; gray diamonds, Qi et al.;^26,38black diamond, Malkin et al.;²⁴black star,

Mihelcic et al.²⁷

hydroperoxide intermediate. Berndt et al.⁶⁰have recently

reported an SCI yield of 0.62 0.28 for 2,3-dimethyl-2-

The HO₂yields obtained here (Table 3) range from 0.01 for

trans-2-butene to 0.34 for 2-methylpropene (in experiments

where excess CO was present, and hence, LIF HO₂interference

eﬀects did not occur), suggesting that HO₂production can be a

signiﬁcant proportion of the total ozonolysis radical yield for

some systems but that it is not formed at yields approaching

unity or above, consistent with HO₂production from the anti-

CI, and in eﬀective competition with other fates such as

stabilization, other decomposition pathways and bimolecular

reaction. Nonzero HO₂production was observed for the 2,3-

dimethyl-2-butene system, where all CIs will form in a syn

conﬁguration amenable to the hydroperoxide rearrangement,

pointing to some scope for HO₂formation from syn-CIs, in

competition with OH, potentially via the hot acid intermediate

(Figure 2).

butene, obtained from measurement of the H₂SO₄formed from

SCI + SO₂reactions at atmospheric pressure and on a time

scale of 0.5−1.5 s. The diﬀerence between this value and (1 −

Y

_OH) values obtained from this work and other studies (Table

2) may therefore reﬂect the diﬀerent time scales probed, with

further OH production occurring from thermalizd SCI

decomposition occurring at longer time scales³(although a

humidity dependence to the OH yield might then be expected,

from competition between SCI decomposition and reaction

with H₂O, in contrast to the majority of literature studies, see

discussion below). This in turn would indicate that the yield of

SCIs formed in the atmospheric boundary layer from alkene

ozonolysis (and available to undergo bimolecular reactions with

other species) may be somewhat lower than measurements

performed in short-time scale laboratory studies indicate.

The literature database of experimental measurements of

production of HO₂from alkene ozonolysis is sparse in

12478

dx.doi.org/10.1021/jp408745h | J. Phys. Chem. A 2013, 117, 12468−12483

The Journal of Physical Chemistry A

Article

comparison with that for OH. The HO₂yield for propene

ozonolysis obtained here (0.09 0.02) is lower than the value

obtained by Qi et al.³⁸of 0.19 0.04, and substantially below

the values obtained by Wegener et al.²⁸of 1.50 0.75 and 1.15

0.60 (dry and humid, respectively). The HO₂yields for 1-

butene and 2-methylpropene obtained here are also substan-

tially lower than the corresponding values reported by Wegener

et al. (Table 3). Qi et al. comment that their results were

sensitive to the assumed distribution of radical products (OH/

HO₂/CH₃O₂) within the model used to interpret the PERCA

data, and analyzed their data assuming an SCI yield from

CH₂OO of 0.37, substantially lower than the 0.54 found

elsewhere.²⁹While full details of the 58-reaction chemical

scheme used to interpret the experimental data and the

sensitivity of the inferred yields to the mechanistic assumptions

are not given, it is possible that a corresponding increase in the

stabilization yield for CH₂OO would result in lower inferred

radical production in their study. For isoprene, the result

obtained here (0.16 0.04) is somewhat lower than the only

other value directly measured, 0.26 0.03,²⁴obtained using a

comparable chamber ozonolysis−LIF observation approach,

but in the absence of a scavenger, which may have led to an

interference contribution to the measured HO₂from hydroxyl-

substituted peroxy radicals.

It is diﬃcult to reconcile some very high HO₂yields from the

literature with the other product yields measured in this and

other studies; for example, in the case of 2-methylpropene, the

inferred yield²⁹of the two CI species is 34:66 for

(CH₃)₂COO*/CH₂OO*, while the measured HO₂yield

from ethene (i.e., from the CH₂OO* CI) ranges from 0.10

to 0.40.^27,29As the (CH₃)₂COO* CI would be expected to

always undergo a hydroperoxide rearrangement to produce

predominantly OH (and indeed OH yields in the range 0.6−0.8

are observed, see Figure 6), with limited HO₂coproduction (as

noted above), HO₂yields from 2-methylpropene (from

CH₂OO*, or from (CH₃)₂COO* undergoing alternative

minor decomposition pathways) would be expected to be of

the order of 0.3−0.5, rather than the (2.00 1.00) measured

by Wegener et al.²⁸Although the measurements made using

the additional turnover approach suﬀer from relatively low

precision in comparison with the LIF observations, the

discrepancy is greater than the (combined) uncertainties of

both studies.

The variation in HO_xyields with humidity was studied for a

subset of systems (Tables 2 and 3). Experimental diﬃculties

(partial failure of the LIF system pump laser) limited the data

obtained for OH; however, for HO₂, reduced production with

increasing humidity was observed for propene and trans-2-

butene (albeit from a very low level in the latter case), while no

change (within uncertainty) was observed for 2-methylpropene.

The reduction in yield for propene was replicated in the

(independent) LIF and PERCA measurements (Figures 4 and

5), consistent with the behavior observed previously for the

ethene system²⁹producing the CH₂OO CI in common with

other terminal alkenes. Both the LIF and PERCA instruments

used in this work exhibit known humidity-dependent

sensitivities/calibrations, which were taken into account in

the data analysis; an error in this factor is considered unlikely

for the LIF HO₂data, which would otherwise be expected to

show a systematic humidity bias across all the chemical/alkene

systems considered, but cannot be precluded on the same basis

for the PERCA instrument, for which only very limited

observations (on a single system) were possible. One further

caveat is that the interpretation presented here is that the

humidity dependence lies in the HO₂yields. Equivalent

measurements of the OH yields with humidity were not

available from this study, consequently, this mechanistic

interpretation is based upon the balance of the literature

regarding OH production (discussed below); in principle a

change in the OH yields with humidity could also contribute to

the observed trend.

The majority of published indirect experimental tracer/

scavenger studies ﬁnd OH yields from alkene ozonolysis to

remain unaﬀected under enhanced humidity,^14,23indicating

that either the water reaction is not competitive with

decomposition or forms products that also decompose to

yield OH, e.g., an unstable hydroperoxide.⁶¹Kuwata et al.⁶²

predict that the hydroperoxide mechanism (thought to be the

principal source of OH) proceeds 3−8 orders of magnitude

faster than the thermalized CI reaction with water for stabilized

syn-CIs, under typical tropospheric conditions (T, P, and

humidity), but that unimolecular rearrangement to dioxirane

occurs at a comparable rate to the bimolecular reaction with

water for the anti-CI conformer of methyl carbonyl oxide

(CH₃CHOO), raising the possibility for water to aﬀect radical

production from this conformer. Kroll et al.⁶³investigated the

possibility of OH formation from the anti-CI by measuring

yields of OD and OH radicals from deuterated (and

undeuterated) cis- and trans-3-hexene, concluding that OH

formation from anti-CIs may also play a signiﬁcant role in the

total OH yield, as approximately one-third of the total OH yield

comes from the anti-CI in the ozonolysis of cis-3-hexene. Drozd

et al.⁸have shown that increasing alkene/CI size/carbon

number substantially increases SCI stabilization for internal

alkenes, indicating that competition between stabilization and

OH production is likely to be strongly species/conformer

dependent; indeed, an increase in OH yield with humidity

(0.02% to 45% RH) for α-pinene using the cyclohexane

scavenger approach has been reported,⁶⁴attributed to

decomposition of the SCI + H₂O formed hydroxyl-hydro-

peroxide, but subject also to assumptions regarding the

contribution of HO₂(potentially also changing with humidity)

and variations in the cyclohexanone/cyclohexanol yields in the

presence of varying HO₂/RO₂populations.²⁹Wegener et al.²⁸

reported positive humidity dependence for OH formation

(increasing production with increasing H₂O) for a number of

small alkenes, where yields were obtained through assessment

of the additional turnover of the parent alkene. It is hard to

reconcile these results with the tracer/scavenger studies for

small alkenes reported above; Wegener et al. speculate that the

humidity dependence observed in their study, which corre-

sponds to better agreement between their humid OH yields

and those observed in (literature) dry experiments, may reﬂect

ambient water vapor entering the experimental systems used in

other literature studies, which are eﬀectively therefore some-

what wet; an eﬀect minimized in the double-walled SAPHIR

chamber (in the present study, where the single-walled

EUPHORE chamber was used, but maintained at above-

ambient pressure, leaks led to a reduction in RH (for elevated

H₂O experiments) as the chamber air was replaced by dry

scrubbed air). Future studies of ozonolysis product yields

should directly measure the humidity within the experimental

chamber in question to address this possibility.

The observed variation in HO₂yields with humidity from

this work is consistent with a humidity-dependent radical

production from the CH₂OO CI formed from all terminal

12479

dx.doi.org/10.1021/jp408745h | J. Phys. Chem. A 2013, 117, 12468−12483

The Journal of Physical Chemistry A

Article

alkenes but suggest that H₂O exerts little inﬂuence in HO₂

production from the C₂CH₃CHOO CIs formed in the cis- and

trans-2-butene systems. In the case of 2-methylpropene, the

relative yields of the (CH₃)₂CHOO and CH₂OO CIs are 66/

34%,⁵¹such that the variation in the small contribution to HO₂

from the CH₂OO moiety would lie within the uncertainty of

the measurements for this parent alkene overall (Table 3). This

interpretation is consistent with theoretical predictions,⁶⁵which

found that reaction of the anti-CI CH₃CHOO (the postulated

predominant precursor to the HO₂observed) with water

occurs several orders of magnitude faster than the equivalent

reaction for the syn-CH₃CHOO, or (CH₃)₂CHOO and is

supported by recent direct measurement of SCI + H₂O rate

constants for the syn- and anti-conformers of CH₃CHOO of <4

× 10⁻¹⁵and 1 × 10⁻¹⁴molec⁻¹cm³s⁻¹, respectively.⁵⁶The

only other reported measurement of the humidity dependence

of HO₂production, from Wegener et al., found no change in

HO₂production for ethene, propene, 1-butene, and isobutene

between 100 and ca. 10000 ppm H₂O (RH levels of

approximately 0.3 and 32%, respectively) but with substantial

uncertainty (50%), which would encompass the variation

reported here (Table 3).

The results of this work indicate a generally consistent

understanding of OH yields from the ozonolysis of small

alkenes, consistent with the predominance of the hydro-

peroxide mechanism, but with some indications (e.g., subunity

yield for 2,3-dimethyl-2-butene) that other fates are possible for

syn-CIs and (from the observation of OH production from

ethene) that other primary OH formation routes occur. It

seems unlikely that there is a substantial humidity dependence

to ozonolysis OH yields, at least for the small (≤C₅) alkenes

considered here, but the contradictory results of recent studies

require further deﬁnitive conﬁrmation of this aspect. The HO₂

yields obtained here (in experiments where excess CO was

present and hence LIF HO₂interference eﬀects eliminated)

range from 0.01 (trans-2-butene) to 0.34 (2-methylpropene),

suggesting that HO₂production can be a signiﬁcant proportion

of the total ozonolysis radical yield for some systems but that it

is not formed at yields approaching unity or above, consistent

with HO₂production from the anti-CI, through the hot acid

intermediate, in competition with stabilization and other

degradation products. The reduction in HO₂formation with

increasing humidity observed for the ethene, propene, and

trans-2-butene systems are consistent with theoretical studies

indicating that reaction with H₂O can be competitive with

dissociation for the CH₃CHOO anti-CI. Following the

argument of Wegener et al. regarding the water level present

in the chamber under nominally dry conditions, considering the

humidity range probed here (where dry conditions correspond

to [H₂O] = 2.7 × 10¹⁵molecules cm⁻³; RH = 0.35%; and wet

to [H₂O] = 2.4 × 10¹⁷molecules cm⁻³; RH = 30%), it may be

that higher HO₂yields would be observed under totally dry

conditions and that the yields applicable to the ambient

atmospheric boundary layer (where higher RH is frequently

encountered) will be lower than those reported here, which

could then be regarded as upper limits.

ethanol, formaldehyde, and acetaldehyde. Initial concentrations

of these parent volatile organic compounds (VOCs) were set to

the mean values observed during the TORCH ﬁeld campaign in

Chelmsford, a suburban location downwind of London, during

summer 2003;⁶⁶the parent VOC species used accounted for

92% of the total calculated OH reactivity from all measured

species. NO_x, O₃, CO, CH₄, temperature, and RH conditions

were set to median values of those observed, while photolysis

rates were calculated for clear sky conditions. Following spin-

up, the model was used to quantify mean radical production

rates over two three-hour periods, from 12.00 to 15.00 and

24.00−03.00 h local time. During daytime, 71% of OH primary

production arose from O₃photolysis/O(¹D) + H₂O reaction,

while 29% arose from alkene ozonolysis, consistent with

previous results,⁴as might be expected considering the

similarity in measured OH yields to those implemented within

the MCM. Daytime primary HO₂production was dominated

by photolysis of HCHO (84%) with photolysis of other

carbonyl species accounting for 12% and alkene ozonolysis

contributing 4% of the total. At night, ozonolysis dominated

OH and HO₂production. Under both conditions, reaction of

2,3-dimethyl-2-butene was the single most signiﬁcant ozonol-

ysis contribution, accounting for ca. 45% of the total ozonolysis

contribution, at least under the conditions of the TORCH

measurements. These calculations do not include a contribu-

tion from HONO photolysis, which was not measured during

TORCH but has been shown to make a very important

contribution to photolytic OH production in the boundary

layer.⁶⁷

6. CONCLUSIONS

The OH and HO₂yields from the ozonolysis of a range of small

C₂−C₅alkenes have been measured using large simulation

chamber experiments incorporating direct measurement of OH

and HO₂by laser-induced ﬂuorescence (LIF) and HO₂+ RO₂

by peroxy radical chemical ampliﬁcation (PERCA). The direct

yields of OH obtained here are consistent with those reported

in the literature as obtained indirectly by tracer/scavenger

studies on long time scales relevant to the atmospheric

boundary layer and are similar to those currently used in

atmospheric models. The LIF HO₂measurements were found

to be substantially inﬂuenced by an interference from hydroxyl-

peroxy radicals, formed from OH addition to the parent alkene,

other than in experiments where an OH scavenger (CO) was

employed. The yields of HO₂obtained in such experiments are

lower than the limited number of previous studies suggest and

are somewhat lower than those used in the Master Chemical

Mechanism (MCM v3.2), indicating that total radical

production from alkene ozonolysis may be somewhat less

than currently calculated using this and other comparable

mechanisms. A reduction in HO₂formation with increasing

humidity was observed, indicating that the values obtained

might be regarded as upper limits under ambient atmospheric

boundary layer conditions, with the caveat that this conclusion

depends in part upon applying humidity-independent OH

yields to the data, based upon past literature results. The

observed OH and HO₂formation is consistent with the

hydroperoxide mechanism dominating (but not exclusively) the

evolution of syn-CIs, and with HO₂formation occurring

primarily through decomposition of anti-CIs and the CH₂OO

CI formed from terminal alkenes, in a competition with

reaction with water vapor, and potentially other atmospheric

trace gases.

The impact of the measured yields for atmospheric OH and

HO₂production was evaluated using a photochemical box

model simulation of boundary layer composition under

polluted suburban conditions. A zero-dimensional box model

simulation was performed using the MCM (v3.2) mechanism,

with radical yields updated following Tables 2 and 3, for ﬁve

C₂−C₅alkanes, seven C₂−C₆alkenes, isoprene, methanol,

12480

dx.doi.org/10.1021/jp408745h | J. Phys. Chem. A 2013, 117, 12468−12483

The Journal of Physical Chemistry A

Article

(12) Kurten

́

, T.; Donahue, N. M. MRCISD Studies of the

AUTHOR INFORMATION

■

Dissociation of Vinylhydroperoxide, CH₂CHOOH: There Is a Saddle

Point. J. Phys. Chem. A 2012, 116, 6823−6830.

Present Addresses

^∥(A.R.R.) National Centre for Atmospheric Science, Wolfson

Atmospheric Chemistry Laboratory, Department of Chemistry,

University of York, York YO10 5DD, U.K.

(13) Hatakeyama, S.; Akimoto, H. Reactions of Criegee Inter-

mediates in the Gas Phase. Res. Chem. Intermed. 1994, 20, 503−524.

(14) Johnson, D.; Lewin, A. G.; Marston, G. The Effect of Criegee-

Intermediate Scavengers on the OH Yield from the Reaction of Ozone

with 2-Methylbut-2-ene. J. Phys. Chem. A 2001, 105, 2933−2935.

(15) Welz, O.; Savee, J. D.; Osborn, D. L.; Vasu, S. S.; Percival, C. J.;

Shallcross, D. E.; Taatjes, C. A. Direct Kinetic Measurements of

Criegee Intermediate (CH₂OO) Formed by Reaction of CH₂I with

O₂. Science 2012, 335, 204−207.

^⊥(M.C.) LISA, UMR CNRS/INSU 7583, Universite

́

Paris Est

́ ́

Creteil et Universite Paris Diderot, Institut Pierre Simon

́

Laplace, 94010 Creteil Cedex, France.

^#(K.P.W.) School of Environment and Technology, University

of Brighton, Brighton BN2 4GJ, U.K.

^○(K.E.H.) Department of Geological Sciences, Indiana

University, Bloomington, Indiana 47405, United States.

(16) Mauldin, R. L., III; Berndt, T.; Sipila, M.; Paasonen, P.; Petaja,

̈

́

T.; Kim, S.; Kurten, T.; Stratmann, F.; Kerminen, V.-M.; Kulmala, M.

A New Atmospherically Relevant Oxidant of Sulphur Dioxide. Nature

2012, 488, 193−197.

(17) Ouyang, B.; McLeod, M. W.; Jones, R. L.; Bloss, W. J. NO₃

Radical Production from the Reaction between the Criegee

Intermediate CH₂OO and NO₂. Phys. Chem. Chem. Phys. 2013, 15,

17070−17075.

(18) Finlayson, B. J.; Pitts, J. N.; Atkinson, R. Low-Pressure Gas

Phase Ozone−Olefin Reactions: Chemiluminescence, Kinetics and

Mechanisms. J. Am. Chem. Soc. 1974, 96, 5356−5367.

(19) Herron, J. T.; Huie, R. E. Stopped-Flow Studies of the

Mechanisms of Ozone−Alkene Reactions in the Gas Phase Propene

and Isobutene. Int. J. Chem. Kinet. 1978, 10, 1019−1041.

(20) Martinez, R. I.; Herron, J. T.; Huie, R. E. The Mechanism of

Ozone−Alkene Reactions in the Gas Phase. A Mass Spectrometric

Study of the Reactions of Eight Linear and Branched-Chain Alkenes. J.

Am. Chem. Soc. 1981, 103, 3807−3820.

(21) Paulson, S. E.; Chung, M. Y.; Hasson, A. S. OH Radical

Formation from the Gas-Phase Reaction of Ozone with Terminal

Alkenes and the Relationship between Structure and Mechanism. J.

Phys. Chem. A 1999, 103, 8125−8138.

(22) Rickard, A. R.; Johnson, D.; McGill, C. D.; Marston, G. OH

Yields in the Gas-Phase Reactions of Ozone with Alkenes. J. Phys.

Chem. A 1999, 103, 7656−7664.

(23) Atkinson, R.; Aschmann, S. M.; Arey, J.; Shorees, B. Formation

of OH Radicals in the Gas-Phase Reactions of O₃with a Series of

Terpenes. J. Geophys. Res. 1992, 97, 6065−6073.

(24) Malkin, T. L.; Goddard, A.; Heard, D. E.; Seakins, P. W.

Measurements of OH and HO₂Yields from the Gas Phase Ozonolysis

of Isoprene. Atmos. Chem. Phys. 2010, 10, 1441−1459.

(25) Donahue, N. M.; Kroll, J. H.; Anderson, J. G.; Demerjian, K. L.

Direct Observation of OH Production from the Ozonolysis of Olefins.

Geophys. Res. Lett. 1998, 25, 59−62.

(26) Qi, B.; Sato, K.; Imarnura, T.; Takami, A.; Hatakeyama, S.; Ma,

Y. Production of the Radicals in the Ozonolysis of Ethene: A Chamber

Study by FT-IR and PERCA. Chem. Phys. Lett. 2006, 427, 461−465.

(27) Mihelcic, D.; Heitlinger, M.; Kley, D.; Musgen, P.; Volz-

Thomas, A. Formation of Hydroxyl and Hydroperoxy Radicals in the

Gas-Phase Ozonolysis of Ethane. Chem. Phys. Lett. 1999, 301, 559−

564.

(28) Wegener, R.; Brauers, T.; Koppmann, R.; Bares, S. R.; Rohrer,

F.; Tillmann, R.; Wahner, A.; Hansel, A.; Wisthaler, A. Simulation

Chamber Investigation of the Reactions of Ozone with Short-Chained

Alkenes. J. Geophys. Res. 2007, 112, D13301.

(29) Alam, M. S.; Camredon, M.; Rickard, A. R.; Carr, T.; Wyche, K.

P.; Hornsby, K. E.; Monks, P. S.; Bloss, W. J. Total Radical Yields from

Tropospheric Ethene Ozonolysis. Phys. Chem. Chem. Phys. 2011, 13,

11002−11015.

(30) Chew, A. A.; Atkinson, R. OH Radical Formation Yields from

the Gas Phase Reactions of O₃with Alkenes and Monoterpenes. J.

Geophys. Res. 1996, 101, 28649−28653.

Notes

The authors declare no competing ﬁnancial interest.

ACKNOWLEDGMENTS

■

We would like to thank co-workers from Fundaci

Valencia and Dr. Roland Leigh, Dr. Zoe Fleming, and Dr. Iain

White from the University of Leicester for technical support

during the TRAPOZ experiments performed at EUPHORE

and subsequent data analysis. The TRAPOZ project was

funded by the U.K. Natural Environment Research Council

(Grant ref NE/E016081/1).

̧

on CEAM,

̈

REFERENCES

■

(1) Calvert, J. G.; Atkinson, R.; Kerr, J. A.; Madronich, S.; Moortgat,

G. K.; Wallington, T. J.; Yarwood, G. The Mechanism of Atmospheric

Oxidation of the Alkenes; Oxford University Press: New York, 2000.

(2) Di Carlo, P.; Brune, W. H.; Martinez, M.; Harder, H.; Lesher, R.;

Ren, X.; Thornberry, T.; Carroll, M. A.; Young, V.; Shepson, P. B.;

et al. Missing OH Reactivity in a Forest: Evidence for Unknown

Reactive Biogenic VOCs. Science 2004, 302, 722−725.

(3) Johnson, D.; Marston, G. The Gas-Phase Ozonolysis of

Unsaturated Volatile Organic Compounds in the Troposphere.

Chem. Soc. Rev. 2008, 37, 699−716.

(4) Emmerson, K. M.; Carslaw, N.; Carslaw, D. C.; Lee, J. D.;

McFiggans, G.; Bloss, W. J.; Gravestock, T.; Heard, D. E.; Hopkins, J.;

Ingham, T.; et al. Free Radical Modelling Studies during the UK

TORCH Campaign in Summer 2003. Atmos. Chem. Phys. 2007, 7,

167−181.

(5) Criegee, R. Mechanism of Ozonolysis. Angew. Chem., Int. Ed.

1975, 14, 745−752.

(6) Rathman, W. C. D.; Claxton, T. A.; Rickard, A. R.; Marston, G. A

Theoretical Investigation of OH Formation in the Gas-Phase

Ozonolysis of E-but-2-ene and Z-but-2-ene. Phys. Chem. Chem. Phys.

1999, 1, 3981−3985.

(7) Niki, H.; Maker, P. D.; Savage, C. M.; Breitenbach, L. P.; Hurley,

M. FTIR Spectroscopic Study of the Mechanism for the Gas-Phase

Reaction between Ozone and Tetramethylethylene. J. Phys. Chem.

1987, 91, 941−946.

(8) Drozd, G. T.; Kroll, J.; Donahue, N. M. 2,3-Dimethyl-2-butene

(TME) Ozonolysis: Pressure Dependence of Stabilized Criegee

Intermediates and Evidence of Stabilized Vinyl Hydroperoxides. J.

Phys. Chem. A 2011, 115, 161−166.

(9) Kuwata, K. T.; Hasson, A. S.; Dickinson, R. V.; Petersen, E. B.;

Valin, L. C. Quantum Chemical and Master Equation Simulations of

the Oxidation and Isomerization of Vinoxy Radicals. J. Phys. Chem. A

2005, 109, 2514−2524.

(10) Fenske, D.; Hasson, A. S.; Paulson, S. E.; Kuwata, K. T.; Ho, A.;

Houk, A. N. The Pressure Dependence of the OH Radical Yield from

Ozone−Alkene Reactions. J. Phys. Chem. A 2000, 104, 7821−7833.

(11) Olzmann, M.; Kraka, E.; Cremer, D.; Gutbrod, R.; Andersson, S.

Energetics, Kinetics, and Product Distributions of the Reactions of

Ozone with Ethene and 2,3-Dimethyl-2-butene. J. Phys. Chem. A 1997,

101, 9421−9429.

(31) Gutbrod, R.; Meyer, S.; Rahman, M. M.; Schindler, R. N. On the

Use of CO as Scavenger for OH Radicals in the Ozonolysis of Simple

Alkenes and Isoprene. Int. J. Chem. Kinet. 1997, 29, 717−723.

(32) Kroll, J. H.; Hanisco, T. F.; Donahue, N. M.; Demerjian, K. L.;

Anderson, J. G. Accurate, Direct Measurements of OH Yields from

12481

dx.doi.org/10.1021/jp408745h | J. Phys. Chem. A 2013, 117, 12468−12483

The Journal of Physical Chemistry A

Article

Gas-Phase Ozone−Alkene Reactions using an in Situ LIF Instrument.

Geophys. Res. Lett. 2001, 28, 3863−3866.

(52) Dillon, T. J.; Crowley, J. N. Direct Detection of OH Formation

in the Reactions of HO₂with CH₃C(O)O₂and other Substituted

Peroxy Radicals. Atmos. Chem. Phys. 2008, 8, 4877−4889.

(53) Hasson, A. S.; Tyndall, G. S.; Orlando, J. J. A Product Yield

Study of the Reaction of HO₂Radicals with Ethyl Peroxy (C₂H5O₂),

Acetyl Peroxy (CH₃C(O)O₂), and Acetonyl Peroxy (CH₃C(O)-

CH₂O₂) Radicals. J. Phys. Chem. A 2004, 108, 5979−5989.

(54) Jenkin, M. E.; Hurley, M. D.; Wallington, T. J. Investigation of

the Radical Product Channel of the CH₃C(O)O₂+ HO₂Reaction in

the Gas Phase. Phys. Chem. Chem. Phys. 2007, 9, 3149−3162.

(55) Atkinson, R.; Baulch, D. L.; Cox, R. A.; Crowley, J. N.;

Hampson, R. F.; Hynes, R. G.; Jenkin, M. E.; Rossi, M. J.; Troe, J.

Evaluated Kinetic and Photochemical Data for Atmospheric

Chemistry: Volume II, Gas Phase Reactions of Organic Species.

Atmos. Chem. Phys. 2006, 6, 3625−4055.

(56) Taatjes, C. A.; Welz, O.; Eskola, A. J.; Savee, J. D.; Scheer, A. M.;

Shallcross, D. E.; Rotavera, B.; Lee, E. P. F.; Dyke, J. M.; Mok, D. K.

W.; et al. Direct Measurements of Conformer-Dependent Reactivity of

the Criegee Intermediate CH₃CHOO. Science 2013, 340, 177−180.

(57) Heard, D. E.; Pilling, M. J. Measurement of OH and HO₂in the

Troposphere. Chem. Rev. 2003, 103, 5163−5198.

(33) Kroll, J. H.; Clarke, J. S.; Donahue, N. M.; Anderson, J. G.;

Demerjian, K. L. Mechanism of HO_xFormation in the Gas-Phase

Ozone-Alkene Reaction. 1. Direct, Pressure-Dependent Measurements

of Prompt OH Yields. J. Phys. Chem. A 2001, 105, 1554−1560.

(34) Kroll, J. H.; Sahay, S. R.; Anderson, J. G.; Demerjian, K. L.;

Donahue, N. M. Mechanism of HO_xFormation in the Gas-Phase

Ozone-Alkene Reaction. 2. Prompt versus Thermal Dissociation of

Carbonyl Oxides to form OH. J. Phys. Chem. A 2001, 105, 4446−4457.

(35) Presto, A.; Donahue, N. Ozonolysis Fragment Quenching by

Nitrate Formation: The Pressure Dependence of Prompt OH Radical

Formation. J. Phys. Chem. A 2004, 108, 9096−9104.

(36) Siese, M.; Becker, K. H.; Brockmann, K. J.; Geiger, H.;

Hofzumahaus, A.; Holland, F.; Milhelcic, M.; Wirtz, K. Direct

Measurement of OH Radicals from Ozonolysis of Selected Alkenes:

A EUPHORE Simulation Chamber Study. Environ. Sci. Technol. 2001,

35, 4660−4667.

(37) Paulson, S. E.; Orlando, J. J. The Reactions of Ozone with

Alkenes: An Important Source of HO_xin the Boundary Layer.

Geophys. Res. Lett. 1996, 23, 3727−3730.

(38) Qi, B.; Yang, B.; Wang, Z. Q.; Yang, H. Y.; Liu, L. Production of

Radicals in the Ozonolysis of Propene in Air. Sci. China, Ser. B: Chem.

2009, 52, 356−361.

(39) Jenkin, M. E.; Saunders, S. M.; Pilling, M. J. The Tropospheric

Degradation of Volatile Organic Compounds: A Protocol for

Mechanism Development. Atmos. Environ. 1997, 31, 81−104.

(40) Becker, K. H. In Situ Euphore Radical Measurement

(EUROPHAM): Report to the European Commission, Contract

(58) Fuchs, H.; Bohn, B.; Hofzumahaus, A.; Holland, F.; Lu, K. D.;

Nehr, S.; Rohrer, F.; Wahner, A. Detection of HO₂by Laser-Induced

Fluorescence: Calibration and Interferences from RO₂Radicals. Atmos.

Meas. Tech. 2011, 4, 1209−1225.

(59) Atkinson, R. Gas-Phase Tropospheric Chemistry of Volatile

Organic Compounds: 1. Alkanes and Alkenes. J. Phys. Chem. Ref. Data

1997, 26, 215−290.

(60) Berndt, T.; Jokinen, T.; Mauldin, R. L., III; Petaja, T.;

̈

ENV4-CT95-0011; Bergische Universitat Wuppertal: Germany, 1999.

Herrmann, H.; Junninen, H.; Paasonen, P.; Worsnop, D.; Sipila, M.

̈

(41) Camredon, M.; Hamilton, J. F.; Alam, M. S.; Wyche, K. P.; Carr,

T.; White, I. R.; Monks, P. S.; Rickard, A. R.; Bloss, W. J. Distribution

of Gaseous and Particulate Organic Composition during Dark

Alphapinene Ozonolysis. Atmos. Chem. Phys. 2010, 10, 2893−2917.

(42) Bloss, W. J.; Lee, J. D.; Bloss, C.; Heard, D. E.; Pilling, M. J.;

Wirtz, K.; Martin-Reviejo, M.; Siese, M. Validation of the Calibration

of a Laser-Induced Fluorescence Instrument for the Measurement of

OH Radicals in the Atmosphere. Atmos. Chem. Phys. 2004, 4, 571−

583.

Gas-Phase Ozonolysis of Selected Olefins: The Yield of Stabilized

Criegee Intermediate and the Reactivity toward SO₂. J. Phys. Chem.

Lett. 2012, 3, 2892−2896.

(61) Anglada, J. M.; Aplincourt, P.; Bofill, J. M.; Cremer, D.

Atmospheric Formation of OH Radicals and H₂O₂from Alkene

Ozonolysis under Humid Conditions. ChemPhysChem 2002, 3, 215−

221.

(62) Kuwata, K. T.; Hermes, M. R.; Carlson, M. J.; Zogg, C. K.

Computational Studies of the Isomerization and Hydration Reactions

of Acetaldehyde Oxide and Methyl Vinyl Carbonyl Oxide. J. Phys.

Chem. A 2010, 114, 9192−9204.

(63) Kroll, J. H.; Donahue, N. M.; Cee, V. J.; Demerjian, K. L.;

Anderson, J. G. Gas-Phase Ozonolysis of Alkenes: Formation of OH

from Anti Carbonyl Oxides. J. Am. Chem. Soc. 2002, 124, 8518−8519.

(64) Tillmann, R.; Hallquist, M.; Jonsson, Å. M.; Kiendler-Scharr, A.;

Saathoff, H.; Iinuma, Y.; Mentel, T. F. Influence of Relative Humidity

and Temperature on the Production of Pinonaldehyde and OH

Radicals from the Ozonolysis of Alpha-Pinene. Atmos. Chem. Phys.

2010, 10, 7057−7072.

(65) Ryzhkov, A. B.; Ariya, P. A. The Importance of Water Clusters

(H₂O)_n(n = 2, ..., 4) in the Reaction of Criegee Intermediate with

Water in the Atmosphere. Chem. Phys. Lett. 2006, 419, 479−485.

(66) Lee, J. D.; Lewis, A. C.; Monks, P. S.; Jacob, M.; Hamilton, J. F.;

Hopkins, J. R.; Watson, N. M.; Saxton, J. E.; Ennis, C.; Carpenter, L. J.;

et al. Ozone Photochemistry and Elevated Isoprene during the UK

Heatwave of August 2003. Atmos. Environ. 2006, 40, 7598−7613.

(67) Ren, X.; Harder, H.; Martinez, M.; Lesher, R. L.; Oliger, A.;

Simpas, J. B.; Brune, W. H.; Schwab, J. J.; Demerjian, K. L.; He, Y.;

et al. OH and HO₂Chemistry in the Urban Atmosphere of New York

City. Atmos. Environ. 2003, 37, 3639−3651.

(43) Green, T. J.; Reeves, C. E.; Fleming, Z. L.; Brough, N.; Rickard,

A. R.; Bandy, B. J.; Monks, P. S.; Penkett, S. A. An Improved Dual

Channel PERCA Instrument for Atmospheric Measurements of

Peroxy Radicals. J. Environ. Monit. 2006, 8, 530−536.

(44) Aschmutat, U.; Hessling, M.; Holland, F.; Hofzumahaus, A.

Physico-Chemical Behaviour of Atmospheric Pollutants; Angeletti, G.;

Restelli, C.; Proc. EUR 15609; Wiley: New York, 1994; pp 811−816.

(45) Creasey, D. J.; Halford-Maw, P. A.; Heard, D. E.; Pilling, M. J.;

Whitaker, B. J. Implementation and Initial Deployment of a Field

Instrument for Measurement of OH and HO₂in the Troposphere by

Laser-Induced Fluorescence. J. Chem. Soc., Faraday Trans. 1997, 93,

2907−2913.

(46) Clemitshaw, K. C.; Carpenter, L. J.; Penket, S. A.; Jenkin, M. J. A

Calibrated Peroxy Radical Chemical Amplifier for Ground-Based

Tropospheric Measurements. J. Geophys. Res. 1997, 102, 25405−

25416.

(47) Fleming, Z. L.; Monks, P. S.; Rickard, A. R.; Heard, D. E.; Bloss,

W. J.; Seakins, P. W.; Still, T. J.; Sommariva, R.; Pilling, M. J.; Morgan,

R.; et al. Peroxy Radical Chemistry and the Control of Ozone

Photochemistry at Mace Head, Ireland during the Summer of 2002.

Atmos. Chem. Phys. 2006, 6, 2193−2214.

(48) Saunders, S. M.; Jenkin, M. E.; Derwent, R. G.; Pilling, M. J.

Protocol for the Development of the Master Chemical Mechanism,

MCM v3 (Part A): Tropospheric Degradation of Non-Aromatic

Volatile Organic Compounds. Atmos. Chem. Phys. 2003, 3, 161−180.

(49) Curtis, A. R.; Sweetenham, W. P. Facsimile/Checkmat User’s

Manual; Harwell Laboratory: Oxfordshire, U.K., 1987.

(50) IUPAC: Subcommittee for Gas Kinetic Data Evaluation, 1999.

http://www.iupac-kinetic.ch.cam.ac.uk.

(51) Alam, M. S. Ph.D. Thesis, University of Birmingham, 2011.

(68) Atkinson, R.; Aschmann, S. M. OH Radical Production from the

Gas-Phase Reactions of O₃with a Series of Alkenes under

Atmospheric Conditions. Environ. Sci. Technol. 1993, 27, 1357−1363.

(69) Neeb, P.; Moortgat, G. K. Formation of OH Radicals in the Gas-

Phase Reaction of Propene, Isobutene, and Isoprene with O₃: Yields

and Mechanistic Implications. J. Phys. Chem. A 1999, 103, 9003−9012.

(70) Aschmann, S. M.; Tuazon, E. C.; Arey, J.; Atkinson, R. Products

of the Gas-Phase Reaction of O₃with Cyclohexene. J. Phys. Chem. A

2003, 107, 2247−2255.

12482

dx.doi.org/10.1021/jp408745h | J. Phys. Chem. A 2013, 117, 12468−12483

The Journal of Physical Chemistry A

Article

(71) Horie, O.; Neep, P.; Moortgat, G. K. Ozonolysis of trans-2-

Butenes and cis-2-Butenes in Low Parts-Per-Million Concentration

Ranges. Int. J. Chem. Kinet. 1994, 26, 1075−1094.

(72) McGill, D. D.; Rickard, A. R.; Johnson, D.; Marston, G. Product

Yields in the Reactions of Ozone with Z-but-2-ene, E-but-2-ene and 2-

Methylbut-2-ene. Chemosphere 1999, 38, 1205−1212.

(73) Orzechowska, G. E.; Paulson, S. E. Production of OH Radicals

from the Reactions of C4−C6 Internal Alkenes and Styrenes with

Ozone in the Gas Phase. Atmos. Environ. 2002, 36, 571−581.

(74) Hasson, A. S.; Chung, M. Y.; Kuwata, K. T.; Converse, A. D.;

Krohn, D.; Paulson, S. E. Reaction of Criegee Intermediates with

Water Vapor: An Additional Source of OH Radicals in Alkene

Ozonolysis? J. Phys. Chem. A 2003, 107, 6176−6182.

12483

dx.doi.org/10.1021/jp408745h | J. Phys. Chem. A 2013, 117, 12468−12483

Article Doi

Radical product yields from the ozonolysis of short chain alkenes under atmospheric boundary layer conditions

DOI: 10.1021/jp408745h

Source and publish data:

Authors:

Article abstract of DOI:10.1021/jp408745h

Full text of DOI:10.1021/jp408745h

Products guided by the article

R&D Labs maybe for 3352-57-6

Relevant to this article

Hot Product