FTIR Kinetic and Mechanistic Study of the Atmospheric Chemistry of Methyl Thiolformate

Article abstract of DOI:10.1021/jp961452u

Some aspects of the atmospheric chemistry of methyl thiolformate (CH3SCHO), a recently detected intermediate in the oxidation of dimethyl sulfide, have been investigated at 298 K and 1000 mbar total pressure in large reaction chambers using long path in situ FTIR absorption spectroscopy for the analysis.Rate coefficients of (1.11 +/- 0.22)E-11 and (5.80 +/- 0.80)E-11 cm³ molecule^-1 s^-1 have been determined for its reaction with OH radicals and Cl atoms, respectively.The UV spectrum of CH3SCHO has been measured in the range 220-355 nm and a lower limit of 5.4 days determined for its atmospheric photolytic lifetime.Detailed product analyses have made for the OH and Cl initiated photooxidation of CH3SCHO.Strong SO absorption bands observed in both systems are tentatively assigned to CH3SOCHO in the OH system and to CH3SOCl in the Cl system.The first gas-phase spectra of CH3SCl and CH3SOCl are also presented.The results are discussed with respect to the atmospheric chemistry of CH3SCHO and possible consequences for the photooxidation mechanism of dimethyl sulfide.

Full text of DOI:10.1021/jp961452u

J. Phys. Chem. 1996, 100, 17207-17217
17207
FTIR Kinetic and Mechanistic Study of the Atmospheric Chemistry of Methyl Thiolformate
Iulia V. Patroescu, Ian Barnes,* and Karl H. Becker
Physikalische Chemie/Fachbereich 9, Bergische UniVersita¨tsGH Wuppertal, D-42097 Wuppertal, F.R.G.
ReceiVed: May 21, 1996; In Final Form: August 6, 1996X
Some aspects of the atmospheric chemistry of methyl thiolformate (CH3SCHO), a recently detected intermediate
in the oxidation of dimethyl sulfide, have been investigated at 298 K and 1000 mbar total pressure in large
reaction chambers using long path in situ FTIR absorption spectroscopy for the analysis. Rate coefficients
of (1.11 ( 0.22) × 10-11 and (5.80 ( 0.80) × 10-11 cm3 molecule-1 s-1 have been determined for its reaction
with OH radicals and Cl atoms, respectively. The UV spectrum of CH3SCHO has been measured in the
range 220-355 nm and a lower limit of 5.4 days determined for its atmospheric photolytic lifetime. Detailed
product analyses have made for the OH and Cl initiated photooxidation of CH3SCHO. Strong SO absorption
bands observed in both systems are tentatively assigned to CH3SOCHO in the OH system and to CH3SOCl
in the Cl system. The first gas-phase spectra of CH3SCl and CH3SOCl are also presented. The results are
discussed with respect to the atmospheric chemistry of CH3SCHO and possible consequences for the
photooxidation mechanism of dimethyl sulfide.
Introduction
will be quantitatively converted to CH3SCH2O, this is presently
being interpreted as indicating that MTF can only be formed in
mutual or cross reactions of CH3SCH2O2 and that the other
possible formation channel
Emissions of dimethyl sulfide (DMS: CH3SCH3), principally
from the world’s oceans, account for nearly half of the biogenic
sulfur emitted to the atmosphere and approximately 25% of the
total global sulfur emissions.1,2 Reactions with OH and NO3
radicals represent the major atmospheric sinks for DMS during
the daytime and nighttime, respectively.3-8 Although the exact
nature of the mechanisms of the reactions of these radicals with
DMS is still uncertain, production of (methylthio)methyl radicals
(CH3SCH2) via an overall abstraction mechanism is thought to
represent the major reaction channel for both radicals at room
temperature. In the atmosphere the further reactions of these
radicals are thought to result in the formation of methylthiyl
radicals and HCHO:5-8
CH3SCH2O + O2 f CH3SCHO + HO2
(7)
is negligible. However, simulations of the above system and
also current work on the oxidation of DMS in this laboratory
using a value of k ) 5 × 10-15 cm3 molecules-1 s-1 for reaction
7, which is typical for reactions of alkoxy radicals with O2,12
and a value of 1 × 105 s-1 for the thermal decomposition
channel, reaction 4, as suggested by recent work in the literature
on the oxidation of DMS, show that the concentration of MTF
would not be formed at levels very much above the MTF
experimental detection limit (50 ppb) because of its fast removal
via UV photolysis. Higher rate coefficients for reaction 7 and
lower thermal decomposition rates for reaction 4 raise the MTF
level significantly above the detection limit. Therefore, given
the complexity of the reaction systems and until rate coefficients
are available for reactions 4 and 7, it cannot presently be
completely excluded that reaction 7 is not operative to some
extent and that the failure to detect MTF is simply its rapid
further oxidization to levels below the detection limit of the
experimental setup. Experiments are ongoing to clarify this
point. Other recent studies in the literature, however, support
that thermal decomposition of CH3SCH2O is the major path-
way.7,9
If we assume that reaction 7 is negligible the laboratory results
imply that MTF will only have a possibility of being formed in
the atmospheric oxidation of DMS in areas of very low NO
such as is the case in the tropical and subtropical remote marine
boundary layer where NO is typically 2-8 ppt.13,14 Under such
conditions the major fate of CH3SCH2OO radicals will be
reaction with other peroxy radicals, in particular HO2 radicals
and to a lesser extent CH3OO radicals. It is well established
that a major channel in the reactions of HO2 with other peroxy
radicals is formation of hydroperoxides:
CH3SCH3 + OH/NO3 f CH3SCH2 + H2O/HNO3 (1)
CH3SCH2 + O2 f CH3SCH2O2
CH3SCH2O2 + NO f CH3SCH2O + NO2
CH3SCH2O f CH3S + HCHO
(2)
(3)
(4)
The results from recent laboratory investigations support that
in the presence of NOx this will be the major fate of
CH3SCH2.6-10 However, a recent study from this laboratory11
has shown that reaction of these radicals results in the formation
of methyl thiolformate (MTF, CH3SCHO) either via mutual
reaction of methylthiylperoxy radicals (CH3SCH2O2) and cross
reaction with HO2 or other RO2:
CH3SCH2O2 + CH3SCH2O2 f
CH3SCHO + CH3SCH2OH + O2 (5)
CH3SCH2O2 + HO2 f CH3SCHO + OH + O2 (6)
In the experiments the photolysis of bis(methylthio)methane
(CH3SCH2SCH3) was used to produce the CH3SCH2 radicals
and a formation yield of ∼15% was observed for methyl
thiolformate in this system. In the study it was found that when
NOx was present in the system formation of MTF was not
observed. Since in the presence of NO any CH3SCH2O2 radicals
CH3SCH2O2 + HO2 f CH3SCH2OOH + O2 (8a)
The atmospheric fate of hydroperoxides is wet and dry deposi-
tion, photolysis, and reaction with OH radicals. Photolysis will
probably result in the formation of CH3SCH2O radicals. There
X Abstract published in AdVance ACS Abstracts, October 1, 1996.
S0022-3654(96)01452-9 CCC: $12.00 © 1996 American Chemical Society

17208 J. Phys. Chem., Vol. 100, No. 43, 1996
Patroescu et al.
are several possible channels for reaction with OH such as re-
formation of CH2SCH2OO radicals and formation of species
such as CH2SCH2OOH and CH3SOH radicals and CH3S(O)CH2-
OOH. None of the reaction products is likely to result in the
formation of MTF. However, there is direct experimental
evidence that the reaction of HO2 radicals with substituted
peroxy radicals can also proceed via a second channel producing
carbonyl products. For example, the reaction of HO2 with
methoxymethylperoxy radical (CH3OCH2O2), the oxygenated
analogue of CH3SCH2O2, has been shown to produce methyl
formate (CH3OCHO) with a yield in the range 0.3-0.5.12 Thus,
it is not unreasonable to expect that the yield of MTF from the
reaction of HO2 with CH3SCH2O2 will be of the same order of
magnitude.
concentration-time behavior of both reactants and products.
The reactor can be irradiated with either 32 low-pressure
mercury vapor lamps (Philips TUV 40, λmax ) 254 nm) or 32
fluorescent lamps (Philips Tl 40W/05, 320 < λ < 450 nm, λmax
) 365 nm). Details of the experimental setup can be found
elsewhere.19,20 The UV spectrum of methyl thiolformate was
measured in a glass reactor of 480-L volume which contains
two White mirror systems and allows simultaneous long-path
in situ measurements in the infrared and UV. The infrared
spectra were recorded at a total path length of 50 m with a
Nicolet 520 FTIR spectrometer and the UV spectra at a total
path length of 51.6 m in combination with a 22 cm spectrometer
(SPEX) equipped with a diode array detector (PAR 1412).
Details of this experimental setup can be found in references
21 and 22.
Irradiations were performed on MTF and MTF/CH3ONO,
MTF/reference hydrocarbon/CH3ONO, MTF/Cl2, and MTF/
hydrocarbon/Cl2 reaction mixtures at 1000 mbar total pressure
(N2 + O2). The initial concentrations of MTF and the reference
hydrocarbon were in the range 1-3 ppm (1 ppm ) 2.46 ×
1013 molecules cm-3 at 1000 mbar and 298 K) and those of
methyl nitrite (CH3ONO) and Cl2 were typically 2-4 and 20-
40 ppm, respectively. All the substances were injected into the
reactor using gas syringes. In all of the experiments, infrared
spectra were recorded with a resolution of 1 cm-1 in the range
600-4000 cm-1 using a KBr beamsplitter and a liquid nitrogen
cooled MCT detector. Typically, 32-128 interferograms were
co-added per spectrum and up to 10-20 spectra were recorded
per experiment.
CH3SCH2O2 + HO2 f CH3SCHO + O2 + H2O (8b)
The cross reactions of organic peroxy radicals generally have
two main pathways, reaction 9a which produces an alkoxy
radical and reaction 9b which produces an alcohol and carbonyl
product:15
RO + RO f 2RO + O2
(9a)
f ROH + R′CHO + O2 (9b)
Therefore, reaction of CH3SCH2O2 with other alkyl peroxy
radicals present in the atmosphere will certainly result, to some
extent, in the formation of MTF. There have not been very
many determinations of the carbonyl yields in the cross reactions
of organic peroxy radicals; however, from the reported values,15
yields of 0.5 are not uncommon.
The rate coefficients for the reactions of OH and Cl radicals
with methyl thiolformate were determined at 1000 mbar of
synthetic air at 298 ( 2 K using the relative kinetic method.23
The photolysis of methyl nitrite with the vis lamps was used as
the OH radical source and the photolysis of molecular Cl2, also
with the vis lamps, as the Cl atom source. Provided that gas
phase reaction with either OH or Cl represents the only loss
for methyl thiolformate and the reference hydrocarbon in the
reaction system
Nothing is presently known about the atmospheric chemistry
of MTF. Attempts have been made in this laboratory to detect
MTF in the OH-initiated oxidation of DMS using the UV
photolysis of DMS/H2O2/air reaction mixtures.16,17 Formation
of MTF was observed but only in the initial few minutes of the
irradiation, indicating that MTF is being rapidly further oxidized
in the reaction system. Addition of NOx to the system was
observed to suppress the formation of MTF. Tentative identi-
fication of MTF has been reported in a low-pressure discharge
flow-mass spectrometry study of the reaction of NO3 radicals
with DMS.4 Apart from this study, no other reports have been
made in the literature of the observation of MTF either in
laboratory or field investigations of DMS chemistry. This is
probably due to lack of a concerted effort by workers to look
for this compound both in laboratory and field measurements
and also because its reactivity may be such that its concentration
in both laboratory reaction systems and in the atmosphere may
be very low which would render its detection extremely difficult.
Since MTF is a potentially important intermediate in the
oxidation of DMS, some aspects of the atmospheric chemistry
of this compound have been investigated. The UV absorption
spectrum and photolysis products of methyl thiolformate have
been measured together with the kinetics and products of its
reaction with OH radicals. A recent study has indicated the
possible importance of the reaction of Cl atoms with DMS in
marine environments.18 The kinetics and products of the
reaction of Cl atoms with MTF have, therefore, also been
investigated.
CH3SCHO + OH/Cl f products, k1
reference + OH/Cl f products, k2
(10)
(11)
then the kinetic data can be treated using eq I:
[CH3SCHO]t
[Reference]t
0
k
0
ln
)
1 ln
(I)
(
)
(
)
k2
[CH3SCHOl]t
[Reference]t
where [CH3SCHO]t and [reference]t are the concentrations of
0
0
the reactant and reference organics, respectively, at time t0, [CH3-
SCHO]t and [reference]t are the corresponding concentrations
at time t, and k1 and k2 are the rate coefficients for the reaction
of either OH or Cl with the carbonyl and reference compounds,
respectively. Plots of ln([CH3SCHO]t /[CH3SCHO]t) against ln-
0
([reference]t /[reference]t) should yield a straight line of slope
0
k1/k2 and zero intercept. The experiments were performed
relative to the reaction of OH with ethene (k298 K ) 8.5 × 10-12
cm3 molecule-1 s-1 24,25
)
and Cl with ethane (k298 K ) 5.9 ×
10-11 cm3 molecule-1 s-1).26
Methyl thiolformate was synthesized according to a method
described in the literature.27,28 Briefly, methanethiol was added
slowly to a cooled equimolar mixture of formic acid and acetic
anhydride and the resulting solution was refluxed and distilled
into an acetone/CO2 cold trap. The distillate was washed with
sodium bicarbonate, extracted with ether, and washed with
water. The ether extract was then distilled at approximately
70 °C and purified by chromatography. The purity of the methyl
Experimental Section
The kinetic and photolysis studies on the reaction of methyl
thiolformate with OH and Cl radicals were performed in a quartz
glass reactor of 1080-L volume equipped with a built-in White
mirror system. Long-path in situ FTIR absorption spectroscopy
(492 m, Bruker IFS 88 spectrometer) was used to monitor the

Atmospheric Chemistry of Methyl Thioformate
J. Phys. Chem., Vol. 100, No. 43, 1996 17209
TABLE 1: UV Absorption Cross Sections for Methyl
Thiolformate in the Region 220-355 nm at 5 nm Intervals
absorpn cross
section (cm2
absorpn cross
section (cm2
wavelength
(nm)
wavelength
(nm)
molecule-1
)
molecule-1
)
220
225
230
235
240
245
250
255
260
265
270
275
280
285
2.137 × 10-18
2.836 × 10-18
3.075 × 10-18
2.637 × 10-18
1.752 × 10-18
9.133 × 10-19
3.978 × 10-19
1.631 × 10-19
7.323 × 10-20
4.381 × 10-20
3.322 × 10-20
2.876 × 10-20
2.575 × 10-20
2.299 × 10-20
290
295
300
305
310
315
320
325
330
335
340
345
350
355
2.031 × 10-20
1.642 × 10-20
1.063 × 10-20
6.217 × 10-21
3.074 × 10-21
1.642 × 10-21
1.024 × 10-21
6.958 × 10-22
4.909 × 10-22
4.278 × 10-22
2.982 × 10-22
2.863 × 10-22
2.384 × 10-22
2.057 × 10-22
Figure 1. Infrared spectrum of 0.84 ppm methyl thiolformate in the
spectral range 4000-700 cm-1 recorded in 1000 mbar of synthetic air
15-20% up to 325 nm, 30% over 330 nm, and 50% for 355
nm. The UV spectrum shows an apparent center at ap-
proximately 270 nm with a rise to a second center at 230 nm.
The low- and high-intensity absorptions are probably due to
carbonyl and sulfur chromophores, respectively. The extension
of the absorption spectrum into the near-UV can be attributed
to the presence of the carbonyl chromophore. The UV
absorption cross sections from the present study can be used to
estimate the photolysis frequency of MTF in the atmosphere.
Assuming a quantum yield for photodissociation of Φ ) 1, for
noontime solar fluxes at 40 °N, July 1,29,30 a photolysis
frequency of J ) (2.14 ( 0.20) × 10-6 s-1 is obtained. Over
the Equator the corresponding values for the same day and time
would be J ) (3.08 ( 0.29) × 10-6 s-1 and τ ) 3.7 days.
Since the quantum yield for photodissociation is probably much
less than unity, a lower limit of 5.4 days can be put on the
photolytic lifetime of methyl thiolformate at 40° N. Using
ultraactinic fluorescent lamps, which approximately simulate
sunlight, the photolysis loss of MTF in the reactor was negligibly
slow. This lack of activity allows a limit of e10-5 s-1 to be
put on the MTF photolysis frequency which is in agreement
with the calculated frequency from the UV spectrum. It is
presently difficult to speculate on possible limits for the
photolysis quantum yield of MTF in the atmosphere.
at 298 K with a resolution of 1 cm-1
.
Figure 2. Gas-phase UV absorption spectrum of methyl thiolformate.
thiolformate, based on its infrared spectrum, was generally better
than 98%. Typical impurities included minor amounts of formic
and acetic acid.
(b) Kinetics of the Reactions of OH and Cl Radicals with
MTF. Loss of MTF in the reactor due to absorption on the
walls was found to be negligible over the time period of the
experiments. Similarly, photolysis with the vis lamps used for
the generation of the OH radicals and Cl atoms was also
negligibly small. Figure 3, panels a and b, shows the data from
a minimum of four experiments obtained from irradiation of
MTF/ethene/CH3ONO/air and MTF/ethane/Cl2/air reaction mix-
tures plotted according to eq I. From the slopes of these plots
the following rate coefficients have been derived for the reaction
of OH radicals and Cl atoms with MTF:
Results and Discussion
(a) Infrared and UV Spectrum of Methyl Thiolformate.
Figure 1 shows the gas phase infrared spectrum of 0.84 ppm
methyl thiolformate recorded in 1000 mbar of synthetic air. The
spectrum is in good agreement with a vapor phase spectrum
published in the literature.28 The three absorptions with band
centers at 2838, 1698, and 767 cm-1 can be assigned to the CH
stretch (ν3), CO stretch (ν4), and C-S-C symmetric stretch
(ν10), respectively. From a series of calibration experiments,
absorption coefficients of ꢀ ) (2.0 ( 0.1) × 10-18 cm2
molecule-1 and ꢀ ) (9.0 ( 0.3) × 10-19 have been determined
for the bands with maxima at 1698 and 767 cm-1, respectively.
Calibrated infrared spectra of MTF were obtained by injecting
weighed amounts of MTF into heated inlet manifold and
flushing it into the reactor with N2. This was performed for
several different quantities of MFT and the reproducibility was
generally better than (5%.
k(OH + MTF, 298 K) ) (1.11 ( 0.22) ×
10-11 cm3 molecule-1 s-1
k(Cl + MTF, 298 K) ) (5.80 ( 0.80) ×
10-11 cm3 molecule-1 s-1
Figure 2 shows the UV spectrum of MTF in the region 220-
355 nm. The measured cross sections in units of cm2
molecule-1 are given for 5 nm intervals in Table 1 and are
represented in Figure 2 by open circles. The measured
absorption values were converted to cross sections by determin-
ing the absolute concentrations from the simultaneously recorded
infrared spectra. The 2σ uncertainties of the cross sections are
The errors are estimates for the uncertainty at the 95%
confidence level.
The rate coefficient for the reaction of OH with MTF is of
the same order of magnitude as those reported in the literature
for the reactions of OH with simple aldehydes and R-dicarbonyls
and intermediate between those reported for dimethyl sulfide
and methanethiol.24-26 The dominant reaction pathway for

17210 J. Phys. Chem., Vol. 100, No. 43, 1996
Patroescu et al.
for the reactions of Cl with DMS and methanethiol.31-33 As
for OH, the reaction of Cl with MTF can proceed by three
channels:
CH3SCHO + Cl f CH3SCO + HCl
CH3SCHO + Cl f CH2SHCO + HCl
CH3SCHO + Cl T CH3S(Cl)CHO
(18)
(19)
(20)
Both the reactions of Cl and Br atoms with DMS are known to
involve two channels, a pressure-independent H abstraction and
a pressure-dependent channel involving a collisionally stabilized
adduct.
The reactions of Cl and Br with CH3SH are thought to
proceed via an addition-elimination reaction mechanism rather
than a direct H abstraction.33,35
A comparison of the magnitude of the rate coefficients for
the reactions of OH and Cl with MTF with those of the literature
rate coefficients for the reactions of OH and Cl with aldehydes,
CH3SH and DMS, leads to the conclusion that abstraction of
the aldehyde H atom will be the major channel for these
reactions; this is also supported by the product analyses
described below. The same comparison suggests that abstraction
from the methyl group will probably not be significant,
particularly for the Cl reaction. The presence of the carbonyl
group next to the S atom in MTF will have a deactivating effect
toward the addition of electrophilic reactants such as OH and
Cl and, therefore, it is to be expected that, although the addition
channel is probably occurring, it will not have the same
significance as in the reactions of OH and Cl with DMS. The
relative importance of the various possible reaction channels is
discussed further below in conjunction with the product analyses
of the UV photolysis and the OH and Cl reactions.
(c) Reaction Product Studies. The products identified in
the various reaction systems, UV photolysis, OH radical
reaction, and Cl atom reaction, are collected in Table 2. For
the UV photolysis and Cl atom systems, the product analyses
were performed in both synthetic air and N2. For the purpose
of the following discussion it should, however, be borne in mind
that when using N2 as a bath gas in large-volume chambers, O2
concentrations of up to 100 ppm can be present due to minor
leaks in the system. The product analysis of the OH reaction
system was only carried out in synthetic air since the OH radical
source, photolysis of methyl nitrite, requires the presence of an
adequate concentration of O2. Although yields are given for
CO and HCHO in the OH-initiated oxidation of MTF, the values
contain contributions from of the photooxidation of CH3ONO,
the OH radical precursor, which also produces these species.
As a consequence, quantification of the yields of CO and HCHO
due solely to the photooxidation of MTF has not been possible.
UV Photolysis Products. Figure 4, panels a and b, shows
the concentration-time profiles of the products measured in
the 254 nm photolysis of ppm levels of methyl thiolformate in
1000 mbar of synthetic air and N2, respectively. As noted
above, traces of up to 100 ppm O2 can be present due to minor
leaks in the system. The same products, CO, HCHO, HCOOH,
CH3OOH, SO2, and OCS, were observed in both systems. The
individual yields of the products and the overall sulfur and
carbon balances are given in Table 2. Apart from CH3OOH
and OCS, whose yields were higher in air than in N2, the yields
of the other measurable products were generally somewhat
higher in N2 than in air. Very significant differences in
formation yields were observed for CH3OH, CH3OOH, and OCS
between synthetic air and N2. In the N2 system the sulfur and
carbon balance were both approximately 100%, indicating that
virtually all of the carbon and sulfur is accounted for. In
synthetic air, however, the sulfur and carbon balances were
Figure 3. Kinetic data plotted according to eq I for the reactions of
(a) OH radicals and (b) Cl atoms with methyl thiolformate.
simple aldehydes5,25-27 is a direct H abstraction; however, for
sulfur compounds the situation is more complex. In the case
of CH3SH the reaction proceeds via a CH3S(OH)H adduct which
decomposes to give CH3S and H2O.24-26 For DMS it is now
known that the reaction proceeds via two channels, (i) an overall
H atom abstraction to form CH3SCH2 radicals and H2O and
(ii) formation of an CH3S(OH)CH3 adduct which rapidly
decomposes back to reactants but in the presence of O2 will
react to give products.5 Because of the chemical structure of
MTF, three reaction pathways are possible: (i) direct abstraction
of the aldehydic H atom, (ii) abstraction of an H atom from the
methyl group, and (iii) OH addition to the sulfur atom, reactions
12, 13, and 14, respectively.
CH3SCHO + OH f CH3SCO + H2O
CH3SCHO + OH f CH2SCHO + H2O
CH3SCHO + OH T CH3S(OH)CHO
(12)
(13)
(14)
There are several possibilities for the fate of the OH adduct
which are analogous to those which have been discussed for
the OH + DMS reaction:5-8 (i) it could react with O2 to form
products (reaction 15), (ii) undergo intramolecular H-atom
abstraction (reaction 16) which would be indistinguishable from
reaction 12, or (iii) thermally decompose (reaction 17).
CH3S(OH)CHO + O2 f products (possibly CH3SOCHO)
(15)
CH3S(OH)CHO + ∆ f CH3SCO + H2O
CH3S(OH)CHO + ∆ f
(16)
CH3S + HCOOH and CH3SOH + HCO (17)
The rate coefficient for the reaction of Cl atoms with MTF
is also of the same order of magnitude as those reported for the
reaction of Cl with simple aldehydes and R-dicarbonyls;26
however, it is an order of magnitude slower than that reported

Atmospheric Chemistry of Methyl Thioformate
J. Phys. Chem., Vol. 100, No. 43, 1996 17211
TABLE 2: Product Formation Yields (% C or S) Measured in the UV Photolysis, OH Radical, and Cl Atom Initiated
Oxidation of MTF in 1000 mbar of N2 and Synthetic Air at 298 K by ∼70% Conversion of MTF
UV photolysis
Cl reaction
OH reaction
air
product
N2
air
N2
air
CO
COCl2
HCHO
HCOCl
CH3OH
HCOOH
CH3OOH
SO2
51.6% C
47.2% C
42.2% C
0.4% C
9.8% C
0.2% C
0.5% C
0.2% C
39.3% C
0.3% C
15.1% C
0.8% C
0.3% C
1.3% C
74% C
28.1% C
21.8% C
300% C
12.2% C
5.4% C
2.0% C
96.7% S
0.24% S
1.2% C
4.4% C
12.3% C
86.7% S
0.84% S
7% C
13% C
15.8% S
n.d.
56.6% S
0.6% S
40% S
n.d.
OCS
CH3SO2(O2)NO2
CH3SCl
CH3S(O)Cl
CH3S(O2)Cl
SOCl2
(20% S)
(10% S)
(18.5% S)
3.7% S
(4% S)
(23% S)
4.3% S
(4% S)
(4% S)
total % S
total % C
97
99
86.5
87.0
44
65
83
72
60
certainly in the form of CO2; however, saturation of its infrared
absorption band due to background CO2 prevented quantification
of its yield.
Figure 5, panel a, shows the residual infrared spectrum
obtained for the photolysis of methyl thiolformate in air after
subtraction of all the identified products. The spectrum is
dominated by strong absorption bands at ∼1190 and ∼1746
cm-1; other weaker but significant absorption features are
centered around 1400, 1245, and 1050 cm-1. Strong absorptions
in the regions around ∼1190 cm-1 and ∼1746 cm-1 are typical
for the sulfoxide -SO-36,37 and -CO- groups, respectively.36
There are two possible channels for the 254 nm photolysis
of methyl thiolformate involving either cleavage of the C-H
bond or the S-C bond, pathways 21 and 22, respectively:
CH3SCHO + hν f CH3SCO + H
f CH3S + HCO
(21)
(22)
Further reactions of either CH3SCO or HCO could be respon-
sible for the large measured yields of CO, 47 and 51% C in
synthetic air and N2, respectively. Although the end formation
yield of CO does not show a large dependence on the nature of
the bath gas, there are differences in its behavior as a function
of time between N2 and synthetic air. In N2 the CO yield is
independent of time whereas in synthetic air the yield increases
with time (Figure 6, panel a). If reaction 22 was the major
channel, then the yield of CO in synthetic air would be expected
to be independent of time due to the fast reaction:
HCO + O2 f HO2 + CO
(23)
The difference in behavior of the CO yield between air and N2
supports that reaction 21 is the dominant UV photolysis pathway
for MTF. The thermal decomposition of CH3SCO, reaction 24,
is probably the main source of CO in the N2 photolysis system:
CH3SCO + ∆ f CH3S + CO
(24)
In synthetic air, addition of O2 to CH3SCO will be the dominant
fate of CH3SCO radicals:
CH3SCO + O2 f CH3SCO(O2)
(25)
Figure 4. Concentration-time profiles of the products formed in the
UV photolysis of methyl thiolformate in 1000 mbar of synthetic air
and N2 containing trace amounts of O2 (up to 100 ppm), panels a and
b, respectively.
The further reaction of CH3SCO(O2) will eventually yield CH3S
radicals and CO2. Further secondary reactions of CH3S will
produce CO, i.e., through formation of HCHO and is further
oxidation, and result in the observed time dependence of the
CO yield.
somewhat poorer with yields of ∼87% S and not all the reaction
products were identified. Some of the missing carbon is

17212 J. Phys. Chem., Vol. 100, No. 43, 1996
Patroescu et al.
Figure 5. Residual infrared spectra obtained after subtraction of all identified products from (a) a MTF/air UV photolysis system, (b) a MTF/
CH3ONO/air visible photolysis system, (c) a MTF/Cl2/air visible photolysis system, and (d) a MTF/Cl2/N2 visible photolysis system.
The CH3SCO radical could also thermally decompose to
produce OCS:
It is known that OH radicals can be formed in the photooxi-
dation of organic sulfur compounds, particularly in the presence
of oxygen,19 and it is probable that the generation of OH radicals
in the UV photolysis of MTF in air is mainly responsible for
many of the differences in the product distributions between
air and N2 and also for the unidentified products. In synthetic
air, self-reactions and cross reactions of peroxy radicals will
also be much more important than in the “pure” N2 system.
The residual spectrum obtained from the UV photolysis of MTF
as shown in Figure 5, panel a, has many similarities with the
residual spectrum obtained from the OH-initiated oxidation of
MTF (Figure 5, panel b), particularly in the -CO- and -SO-
absorption regions 1700-1800 and 1100-1200 cm-1, respec-
tively. As will be discussed below, the dominant spectral
features of Figure 5, panels a and b, are ascribed to methylformyl
sulfoxide (CH3S(O)-C(O)-H) formed via the addition of OH
to MTF and subsequent reaction of the adduct with O2, reactions
14 and 15.
CH3SCO + ∆ f CH3 + OCS
(26)
This reaction could explain the formation of OCS observed in
the UV photolysis of MTF; however, the time behavior and
formation yields of OCS in synthetic air and N2 are very
different (Figure 6, panel b). The OCS yield is much higher in
air than in N2 and the OCS yield is independent of reaction
time whereas as in N2 there is a time dependence. If reaction
26 was the main source of OCS the yields in air and N2 should
both be independent of time. The yield in air would also be
expected to be lower than in N2 due to scavenging of CH3SCO
by O2. The results imply that, although reaction 26 may
possibly contribute to the OCS formation, it is not the main
source. The main pathway for OCS formation is currently
thought to be reaction of CH3S radicals with O2 to form
thioformaldehyde (H2CdS) which further oxidizes to form, in
part, OCS.16,17 The further reactions of CH3S radicals are
responsible for the formation of SO2, HCHO, and CH3OH
observed in the photolysis of MTF. The various pathways
possible for the formation of these compounds are described in
the literature and will not be discussed further here.7,8
OH Radical Reaction System. Table 2 lists the products
from the OH-initiated oxidation of MTF using the photolysis
of methyl nitrite as the OH source. The yield of SO2 (∼40%
S) is similar to that observed in the OH-initiated oxidation of
20,38
DMS in the presence of NOx
and is the only positively
identified S-containing compound in the reaction system. In
the presence of excess NO2 the formation of methanesulfonyl
peroxynitrate (CH3SO2O2NO2) is also observed; under such
conditions estimations indicate that the yield of this compound
is of the order of 20% S. Identified carbon-containing products
include CO, HCHO, CH3OH, and HCOOH. As mentioned
above, since the photolysis of the OH source, methyl nitrite,
The SO2 formation yield of ∼90% S obtained in air is very
similar in magnitude to that observed recently in the 254 nm
photolysis of DMDS in air 19 which produces exclusively CH3S
radicals. The results support the conclusions of the DMDS
photolysis study19 that in the absence of NOx the major fate of
the CH3S radical will be oxidation to SO2..

Atmospheric Chemistry of Methyl Thioformate
J. Phys. Chem., Vol. 100, No. 43, 1996 17213
this pathway would explain the “prompt” formation of CO
observed in the reaction system (i.e., the shape of the measured
concentration-time profile supports that CO is a primary
reaction product and is not formed in secondary reactions).
In principle, abstraction of an H atom from the methyl group
and further oxidation of the resulting CH2SCHO radical could
also produce high yields of CO. However, the CO would be
formed in secondary reactions which again would not be
consistent with the observed concentration-time profile. Further,
work on the Cl initiated oxidation of methyl chlorothiolformate
(CH3SCOCl) performed in this laboratory, where the only
H-abstraction route possible is from the methyl group to form
CH2SCOCl radicals, shows that although high yields of CO are
formed in this oxidation they are accompanied by relatively high
formation yields of OCS (∼50% S) probably via the series of
reactions:
2CH2SCOCl + O2 f 2O2CH2SCOCl f
2OCH2SCOCl + O2 (27)
OCH2SCOCl f HCHO + OCS + Cl
(28)
Since similar oxidation pathways can be expected for CH2-
SCOCl and CH2SCHO, formation of CH2SCHO in the OH-
initiated oxidation of MTF would be expected to give high yields
of OCS. Formation of OCS is not observed in the OH-initiated
oxidation of MTF and is taken as indicating that H abstraction
from the methyl group is probably negligible. The identified
products do not give any information on the possibility of an
addition channel such as has been observed for the reaction of
OH with DMS.5-8
The addition reaction of OH with DMS in the presence of
O2 is now known to lead to the formation of dimethyl sulfoxide
(DMSO).5,16,17 The mechanism of DMSO formation is not
presently known but probably involves the formation of an
intermediate OH adduct which further reacts with O2:
(CH3)2S + OH f (CH3)2S-OH
(29)
Figure 6. Yields of CO and OCS observed as a function of time in
the UV photolysis of methyl thiolformate in synthetic air and N2, panels
a and b, respectively.
(CH3)2S-OH + O2 f {(CH3)2S(O2)-OH} f
(CH3)2SO + HO2 (30)
also produces CO and HCHO, it is difficult to give quantitative
estimates of the yields of these compounds for the MTF
oxidation. The residual spectrum obtained after subtraction of
all the identified products is shown in Figure 5, panel b. The
major features are absorptions in the carbonyl region with sharp
features at 1758.7 and 1745.9 cm-1, a broad absorption in the
region typical for the SO-group with a maximum about 1192
cm-1, and a broad absorption in the 800-750 cm-1 region with
If the addition channel is operative in the case of MTF, the
OH-initiated oxidation process would lead to the formation of
methylformyl sulfoxide (CH3S(O)-C(O)-H) which would give
rise to infrared absorptions in the 1700-1800 and 1100-1200
cm-1 regions:
a maximum at 775.5 cm-1
.
CH3SCHO + OH (+ M) f CH3S(OH)CHO (+ M) (31)
CH3S(OH)CHO + O2 f CH3S(O)CHO + HO2 (32)
As discussed in the kinetic section there are three possible
pathways for the attack of the OH radical aldehyde H-atom
abstraction, methyl H-atom abstraction or addition to the sulfur
atom. On the basis of a comparison of the rate coefficients of
OH with various aldehydes and DMS,26 one would expect that
the aldehyde H-atom abstraction pathway probably dominates
with perhaps some contribution from the addition channel and
a minor contribution from the methyl group H-atom abstraction
channel. A very high formation yield of CO is observed in the
product studies (Table 2). The photooxidation of CH3ONO
certainly contributes to the CO yield; however, from a consid-
eration of the magnitude of the measured OH radical rate
coefficient and the concentration-time behavior of CO and
HCHO, it is concluded that a major proportion of the CO yield
must stem from the reaction of OH with MTF. Although this
evidence is by no means conclusive, it implies that abstraction
of the aldehyde H atom is a major reaction pathway since only
In the residual spectrum from the OH/MTF reaction system
(Figure 5, panel b) the strong absorption feature with maximum
at ∼1190 cm-1 is typical for a compound containing the
sulfoxide group, -SO-,36,37 and a sulfoxide compound is
probably responsible for at least part of this absorption. Because
of the presence of NOx in the system, absorptions from nitrate
compounds may also contribute in this region. The concentra-
tion-time behavior of the absorption at ∼1190 cm-1 correlates
with the carbonyl absorption in the 1750 cm-1 region, suggesting
that they probably belong to the same compound. As mentioned
above, the absorptions at ∼1750 and ∼1190 cm-1 observed in
the both the OH-initiated and UV photolysis of MTF are very
similar. Based on the recent observed formation of DMSO in
the OH-initiated oxidation of DMS,16,17 the absorptions are being

17214 J. Phys. Chem., Vol. 100, No. 43, 1996
Patroescu et al.
tentatively assigned to the formation of methylformyl sulfoxide
in the reaction system.
Another compound which could explain the absorptions is
diformyl sulfoxide (OHC-SO-CHO) which could be formed
either by the oxidation of MTF (reaction 13 followed by
reactions 33-36) or the further oxidation of methylformyl
sulfoxide (reaction 36):
CH2SCHO + O2 several steps8 OHC-S-CHO
(33)
OHC-S-CHO + OH (+ M) f
OHC-S(OH)-CHO (+M) (34)
OHC-S(OH)-CHO + O2 f OHC-SO-CHO (35)
CH3S(O)CHO + OH + O2 several steps8 OHC-SO-CHO
(36)
However, since H-atom abstraction from the methyl group is
considered to be very minor, formation of diformyl sulfoxide,
if occurring, is also expected to be very minor.
Cl-Atom Reaction System. Figure 7, panels a and b, shows
the concentration-time profiles of the calibrated products
identified in the reaction of Cl atoms with MTF in synthetic air
and N2, respectively. The yields of all the identified products
determined for ∼70% conversion of MTF are listed in Table 2.
The list of products in Table 2 also contains chlorinated
compounds such as CH3SCl, CH3S(O)Cl, CH3S(O2)Cl, and
SOCl2 which have been identified or assigned in the reaction
system. The yields of these compounds should be treated as
preliminary due to difficulties in accurate calibration; because
of the uncertainty in the yields the concentration-time profiles
of these compounds are not shown in Figure 7. The analysis
of the results leading to the identification of these compound is
outlined below. In synthetic air the carbon and sulfur balances
are both approximately 70-80%. SO2 accounts for a large
fraction of the chlorine-free detected sulfur in synthetic air along
with low yields of OCS and the major carbon-containing product
is CO which accounts for approximately 55% of the detected
carbon. In nitrogen, SO2 accounts for only 15% S and a large
fraction the sulfur has been shown to be in the form of
chlorinated organosulfur compounds. The total carbon balance
(65% C) is a little higher than that obtained in air and again
CO is the major carbon containing product.
Figure 7. Concentration-time profiles of the products formed in the
reaction of Cl atoms with methyl thiolformate in 1000 mbar of synthetic
air and N2, panels a and b, respectively.
Figure 5, panels c and d, shows the residual spectra obtained
from the reaction of Cl atoms with MTF in air and N2,
respectively, after subtraction of all the non-chlorine-containing
compounds listed in Table 2. In both residual spectra the
dominant feature is a strong absorption feature with maxima at
1190, 1180, and 1175 cm-1. The relative intensities of the
maxima are different between air and N2 suggesting that more
than one compound is probably contributing to the absorption.
There is another band on the right-hand flank of this absorption
tion appears fairly late in the irradiation and increases with time.
It could be formed by reactions of Cl with CH3SO or SO radicals
formed in the system by oxidation of CH3S. However, the
intensity of the 1255 cm-1 absorption increases as that of the
absorption centered around 1180 cm-1 decreases, suggesting
that it is probably an oxidation product of the compound giving
rise to this absorption.
As already mentioned, strong absorption features in the region
1100-1200 cm-1 are typical for compounds containing the
sulfoxide group, -SO-,36,37 and a sulfoxide compound is
probably responsible for the major fraction the residual absorp-
tions in Figure 5. Inspection of the broad carbonyl region
1780-1700 cm-1 suggests that at least two compounds con-
tribute these CO absorptions. In the product analysis all of the
potential carbonyl-containing products formed in bond cleavage
reactions such as HCHO, HCOCl, COCl2, and HCOOH have
all been accounted for. Therefore, the compounds giving rise
to the residual carbonyl absorptions must come from carbonyl-
containing compounds produced in the MTF oxidation which
leave the MTF entity intact; i.e., C-S bond fission does not
occur.
with a center at approximately 1145 cm-1
. Other features
presence in both spectra are absorptions due to C-H stretching
vibrations at approximately 2998 and 2945 cm-1 (not shown),
a broad absorption in the carbonyl region from approximately
1700 to 1780 cm-1, and various absorptions in the fingerprint
region centered at 1402, 1255, 1025, 944, and 745 cm-1. The
absorption features at 2998, 2945 1255, 1145, 1025, 944, and
745 cm-1 are more intense in N2 compared to air.
The absorption centered at 1255 cm-1 has been identified as
thionyl chloride (SOCl2) from an authentic spectrum of the
substance.38 This compound is difficult to calibrate and as stated
above the preliminary yield should be treated with caution. The
mechanism of SOCl2 formation is presently unclear; its absorp-

Atmospheric Chemistry of Methyl Thioformate
J. Phys. Chem., Vol. 100, No. 43, 1996 17215
The high yields of CO observed in N2 support that abstraction
of the aldehydic H atom must be a major reaction pathway:
CH3SCHO + Cl f CH3SCO + HCl
CH3SCO + ∆ f CH3S + CO
(18)
(24)
However, as discussed in the kinetic section there are several
other reaction pathways for the Cl atom attack, an H atom
abstraction from the methyl group (reaction 19) and formation
of a weakly bound Cl-MTF adduct (reaction 20). Adduct
formation is known to occur between the reactions of Cl and
Br atoms with DMS.31,32,34 The fate of the Cl-MTF adduct
could be either decomposition for which a number of pathways
are possible
(CH3)(CHO)S-Cl + ∆ f CH3S + HClCO
(37)
(CH3)(CHO)S-Cl + ∆ f CH3S-Cl + HCO (38)
or further reaction with O2 to form products (reaction 39),
(CH3)(CHO)S-Cl + O2 f products
(39)
Formyl chloride is observed in the reaction system; however,
the formation yields are extremely low and thus reaction 37, if
occurring, is of negligible importance. No evidence was found
32
for a channel producing CH3SCl in a discharge flow-mass
spectroscopy investigation of the reaction of Cl atoms with DMS
at low pressure, and a laser flash photolysis-resonance fluo-
rescence study of the reaction of Br atoms with DMS was
inconclusive concerning the fate of the (CH3)2SBr adduct.39
However, the formation of CH3SF has been observed in the
32,40
reaction of F atoms with DMS
and there is evidence that
the formation of CH3SOH in the reactions of OH radicals with
DMS is possible.7,8 Further, ongoing work in this laboratory
on the reaction of Br atoms with DMS at room temperature
has shown that the fate of the (CH3)2SBr adduct in synthetic
air apart from reforming DMS and Br atoms is decomposition
to form methanesulfenyl bromide CH3SBr and CH3 radicals.
The results support that reaction of the adduct with molecular
oxygen is negligible.
Figure 8. (a) Infrared spectrum of 18 ppm of methanesulfenyl chlorine
(CH3SCl) produced by the reaction of Cl2 with dimethyl disulfide
(DMDS) in 1000 mbar of N2. (b) Residual spectrum obtained after 2
min irradiation of a DMDS/Cl2/N2 mixture in which the strongest
residual absorption with centered at 1180 cm-1 is tentatively assigned
to methanesulfinyl chloride (CH3SOCl). (c) Residual spectrum obtained
from the mixture in (b) after complete consumption of DMDS in which
absorption features attributable to methanesulfonyl chloride (CH3SO2-
Cl) and thionyl chloride (SOCl2) are clearly visible.
The infrared spectrum of methanesulfenyl chloride (CH3SCl)
is known,41 and it can also be easily generated by reacting
DMDS with molecular chlorine.37
CH3SSCH3 + Cl2 f CH3SCl + CH3SCl
(40)
Figure 8, panel a, shows a gas-phase reference spectrum of
approximately 18 ppm of CH3SCl produced in situ in the 1080
L reactor by mixing stoichiometric quantities of DMDS (9 ppm)
and Cl2 (9 ppm) in 1000 mbar of synthetic air. This is to our
knowledge the first calibrated gas-phase spectrum of CH3SCl.
Using this spectrum as a reference, formation of CH3SCl was
confirmed in the Cl-initiated oxidation of MTF. Methanesulfe-
nyl chloride is known to be very reactive37,42 and, if formed in
the present reaction system, its major fate is expected to be
oxidation initially to methanesulfinyl chloride (CH3SOCl) which
is itself very reactive and will rapidly further oxidize to
methanesulfonyl chloride (CH3SO2Cl).
a strong SO absorption at approximately 1150 cm-1. Figure 8,
panel b, shows the spectrum obtained on photolysis of a DMDS/
Cl2 mixture in synthetic air for 5 min. The dominant feature
here is an SO absorption with maxima at 1188, 1180, and 1175
cm-1. Since the major fate of DMDS in this system is initially
formation of CH3SCl, the absorptions entered around 1180 cm-1
are assigned to CH3SOCl. Figure 8, panel c, shows the product
spectrum obtained upon further irradiation. The SO absorption
diminishes with formation of SO2, CH3SO2Cl, and SOCl2.
Comparison of the absorptions around 1180 cm-1 in Figure
5, panels c and d, with those in Figure 8, panel b, shows that
they are very similar, particularly for the Cl-initiated oxidation
of MTF in synthetic air. From this comparison it is suggested
that a major portion of the SO absorption observed in the
residual spectrum from the reaction of Cl with MTF (Figure 5,
panels c and d) is due to CH3SOCl formed by the oxidation of
CH3SCl produced via reactions 36 and 38.
According to Bellamy,36 CH3SOCl would be expected to have
an SO absorption at ∼1180 cm-1 (i.e., estimated by determining
the middle value of the sum of DMSO and SOCl2 SO absorption
frequencies: (1256 + 1103)/2 ) 1180 cm-1). This is in
agreement with the strong absorption observed at 1180 cm-1
in the Cl initiated oxidation of MTF (Figure 5, panels c and d);
however, it would not explain the carbonyl adsorption. Infrared
spectra of CH3SOCl recorded in an aqueous medium43,44 report

17216 J. Phys. Chem., Vol. 100, No. 43, 1996
Patroescu et al.
Low yields of methanesulfonyl chloride (CH3SO2Cl) are
observed in the MTF/Cl2 oxidation systems and also in the
DMDS/Cl2 system. Experiments carried out with high starting
concentrations of Cl2 led to an increase of the CH3SO2Cl yields
in the MTF/Cl2 systems. The source of CH3SO2Cl is thought
to be primarily further oxidation of CH3SOCl as has been
observed in liquid phase studies of CH3SCl chemistry37,45 and
is supported by the present gas-phase study of the DMDS/Cl2
reaction system. Under the present gas-phase conditions
reaction of CH3SOCl with ClO radicals is probably the most
likely process responsible for the oxidation:
Thus, no large temperature dependence is expected for the
reaction which could alter the importance of the OH sink
reaction for MTF. This is, however, speculation and needs to
be confirmed by experiment. MTF is only very sparingly
soluble in water and has a boiling point of 70-75 °C at 1 atm;
therefore, it is unlikely that heterogeneous processes such as
wet deposition or aerosol formation will be able to compete
with oxidation via OH radicals. Therefore, reaction of OH
radicals will most probably be the dominant atmospheric sink
for MTF.
Although not described here, investigations were also carried
out to determine the rate coefficient for the reaction of MFT
with NO3 radicals using the thermal decomposition of N2O5 as
the radical source. The decay of MTF under the experimental
conditions was very slow and did not allow an accurate
determination of the rate coefficient for the reaction with NO3
radicals. However, from the measurements it can be estimated
that the rate coefficient must be of the order of 2 × 10-16
molecules cm-3 s-1; i.e., of the same order of magnitude as for
CH3SOCl + ClO f CH3SO2Cl + Cl
(41)
As discussed above, the reaction of OH with MTF is believed
to produce methylformyl sulfoxide. A similar reaction sequence
is possible for the reaction between Cl atoms and MTF leading
to the formation of CH3SOCHO:
CH3SCHO + Cl + O2 f
26
the NO3 radical reaction with HCHO and thus of negligible
{(CH3)(CHO)S(O2)-Cl} f CH3-SO-CHO + ClO (42)
importance as a nighttime sink for MTF in the atmosphere.
Acknowledgment. Financial support of this work by the EC
(European Commission) is gratefully acknowledged. The
authors thank Dr. H. G. Libuda for the invaluable assistance in
recording the UV spectra
Formation of this compound would account partly for the
carbonyl absorption in the 1700 cm-1 region in Figure 5, panels
c and d, and also the differences observed for the SO absorptions
assigned to CH3SOCl in the MTF/Cl2 reaction system compared
to the DMDS/Cl2 system. There is another, and more probable,
pathway to formation of CH3SOCHO in the MTF/Cl2 system,
reaction of Cl atoms with alkyl peroxy radicals produced in
the oxidation could result in the formation of ClO radicals which
could react with MTF to produce methylformyl sulfoxide:
References and Notes
(1) Bates, T. S.; Lamb, B. K.; Guenther, A.; Dignon, J.; Stoiber, R. E.
J. Atmos. Chem. 1992, 14, 315.
(2) Spiro, P. A.; Jacob, D. J.; Logan, J. A. J. Geophys. Res. 1992, 97,
6023.
(3) Jensen, N. R.; Hjorth, J.; Lohse, C.; Skov, H.; Restelli, G. J. Atmos.
Chem. 1992, 14, 95.
Cl + organic peroxy f ClO + organic alkoxy (43)
ClO + CH3-S-CHO f CH3-SO-CHO + Cl (44)
(4) Butkovskaya, N. I.; LeBras, G. J. Phys. Chem. 1994, 98, 2582.
(5) a) Hynes, A. J.; Stoker, R. B.; Pounds, A. J.; McKay, T.; Bradshaw,
J. D.; Nicovich, J. M.; Wine, P. H. J. Phys. Chem. 1995, 99, 16967. (b)
Hynes, A. J.; Wine, P. H.; Semmes, D. H. J. Phys. Chem. 1986, 90, 4148.
(6) Stickel, R. E.; Zhao, Z.; Wine, P. H. Chem. Phys. Lett. 1993, 212,
312.
The reaction of Cl atoms with methylperoxy radicals (CH3O2)
26
is known to produce ClO
radicals and it is now well
established that reactions of ClO with DMS produce DMSO
with unit yield.46 Recent experience on the formation of DMSO
in both the Br- and Cl-initiated oxidation of DMS from ongoing
studies in this laboratory lead us to believe that formation of
CH3SOCHO via reaction sequence (43) and (44) is more
probable.
(7) Turnipseed, A. A.; Ravishankara, A. R. In Proceedings of the
International Symposium on Dimethylsulphide: Oceans, Atmosphere, and
Climate, Belgirate, Italy, 13-15, Oct. 1992; Restelli, G., Angeletti, G., Eds.;
Kluwer Academic Publishers: Dordrecht, 1993; pp 185-195.
(8) Barone, S. B.; Turnipseed, A. A.; Ravishankara, A. R. Faraday
Discuss. 1995, 100, 39.
(9) Wallington, T. J.; Ellermann, T.; Nielsen, O. J. J. Phys. Chem. 1993,
97, 8442.
(10) Nielsen, O. J.; Sehested, J.; Wallington, T. J. Chem. Phys. Lett.
1995, 236, 385.
Conclusions
(11) Barnes, I.; Becker, K. H.; Patroescu, I.; Ruppert, L. In Second
Progress Report, Contract No. EV5V-CT91-0038; Moortgat, G. K.,
coordinator; Commission European Communities, Brussels, 1994.
(12) (a) Wallington, T. J.; Hurley, M. D.; Ball, J. C.; Jenkin, M. E.
Chem. Phys. Lett. 211, 1993, 41. (b) Jenkin, M. E.; Hayman, G. D.;
Wallington, T. J.; Hurley, M. D.; Ball, J. C.; Nielsen, O. J.; Ellermann, T.
J. Phys. Chem. 1993, 97, 11712.
(13) Berresheim, H.; Wine, P. H.; Davis, D. D. In Composition,
Chemistry, and Climate of the Atmosphere; Singh, H. B., Ed.; Van Nostrand
Reinhold: New York, 1995; pp 251-307.
(14) Davis, D. D.; Bradshaw, J. D.; Rodgers, M. O.; Sandholm, S. T.;
KeSheng, S. J. Geophys. Res. 1987, 92, 2049.
(15) (a) Lightfoot, P. D.; Cox, R. A.; Crowley, J. N.; Destriau, M.;
Hayman, G. D.; Jenkin, M. E.; Moortgat, G. K.; Zabel, F. Atmos. EnViron.
1992, 26A, 1805. (b) Wallinton, T. J.; Dagaut, P.; Kurylo, M. J. Chem.
ReV. 1992, 92, 667.
(16) Barnes, I.; Becker, K. H.; Patroescu, I. Geophys. Res. Lett. 1994,
21, 2389.
(17) Barnes, I.; Becker, K. H.; Patroescu, I. Atmos. EnViron. 1996, 30,
1805.
(18) Pszenny, A. A.; Keene, W. C.; Jacob, D. J.; Fan, S.; Maben, J. R.;
Zetwo, M. P.; Springer-Young, M.; Galloway, J. N. Geophys. Res. Lett.
1993, 20, 699.
(19) Barnes, I.; Becker, K. H.; Mihalopoulos, N. J. Atmos. Chem. 1994,
18, 267.
(20) Barnes, I.; Bastian, V.; Becker, K. H Int. J. Chem. Kinet. 1988,
20, 415.
The atmospheric chemistry of methyl thiolformate, an
intermediate in the oxidation of dimethyl sulfide, has been
investigated. Its UV absorption spectrum has been measured
in the range 220-355 nm at 298 K and a lower limit of 5.4
days determined for its atmospheric photolytic lifetime in the
summertime at 40° N. It is presently not known what effect
temperature will have on the UV spectrum; however, since the
absorption in the atmospheric actinic region is relatively weak
it is unlikely that significant changes will occur over the
tropospherically relevant temperatures. Rate coefficients have
been determined for its reaction with OH radicals and Cl atoms.
Since the concentration of Cl atoms in the atmosphere is, with
the possible exception of a few specific marine regions,
generally very low, loss by this reaction will be negligible.
However, OH radicals concentrations are much higher, using a
globally-averaged 24 h mean value for [OH] of the order of
7.7 × 105 molecules cm-3 47 results in an atmospheric lifetime
due to reaction with OH of approximately 1.4 days. Since the
reaction of OH with MTF is believed to be an overall abstraction
process similar to those of the simple aldehydes the temperature
dependence of the rate coefficient is also expected to be similar.

Atmospheric Chemistry of Methyl Thioformate
J. Phys. Chem., Vol. 100, No. 43, 1996 17217
(21) Barnes, I.; Becker, K. H.; Zhu, T. J. Atmos. Chem. 1993, 17, 353.
(22) Libuda, H. G.; Zabel, F.; Fink, E. H.; Becker, K. H. J. Phys. Chem.
1990, 94, 5860.
(23) Bierbach, A.; Barnes, I.; Becker, K. H.; Wiesen, E. EnViron. Sci.
Technol. 1994, 28, 715.
(24) Atkinson, R. Atmos. EnViron. 1990, 24A, 1.
(25) Atkinson, R. J. Phys. Chem. Ref. Data, Monograph No. 2, Gas-
Phase Tropospheric Chemistry of Organic Compounds, 1994, 1-216.
(26) Atkinson, R.; Baulch, D. L.; Cox, R. A.; Hampson, R. F., Jr.; Kerr,
J. A.; Troe, J. J. Phys. Chem. Ref. Data 1992, 21, 1125.
(27) Shire, D. Ph.D. Thesis, University of Sheffield, 1969.
(28) Jones, G. I. L.; Lister, D. G.; Owen, N. L.; Gerry, M. C. L.; Palmieri,
P. J. Mol. Spectros. 1976, 60, 348.
(37) (a) Sulfur in Organic and Inorganic Chemistry; SenningA., Ed.;
Marcel Dekker, Inc.: New York, 1971; Vol. 1. (b) Sulfur in Organic and
Inorganic Chemistry; Senning, A., Ed.; Marcel Dekker, Inc.: New York,
1982; Vol. 4, pp 193-267.
(38) The Aldrich Library of FT-IR Spectra, Edition I, Vol. 3, Vapor
Phase; Charles J. Pouchert, Aldrich Chemical Co., Inc.: Milwaukee, WI,
1989.
(39) Wine, P. H.; Nicovich, J. M.; Stickel, R. E.; Zhao, Z.; Shackelford,
C. J.; Kreutter, K. D.; Daykin, E. P.; Wang, S. In Proceedings of the NATO
AdVance Research Workshop Tropospheric, Chemistry of Ozone in the Polar
Regions; Niki, H., Becker, K. H., Eds.; NATO ASI Subseries, Vol. 17,
Global Environmental Change; Springer-Verlag: New York, 1993; pp 385-
395.
(29) Finlayson-Pitts, B. J.; Pitts, J., Jr. Atmospheric Chemistry; Wiley:
New York, 1986.
(30) Demerjian, K. L.; Schere, K. L.; Peterson, J. T. AdV. EnViron. Sci.
Technol. 1980, 10, 369.
(31) Stickel, R. E.; Nicovich, J. M.; Wang, S.; Zhao, Z.; Wine, P. H. J.
Phys. Chem. 1992, 96, 9875.
(32) Butkovskaya, N. I.; Poulet, G.; LeBras, G. J. Phys. Chem. 1995,
99, 4536.
(33) Nicovich, J. M.; Wang, S.; Wine, P. H. Int. J. Chem. Kinet. 1995,
27, 359.
(34) Jefferson, A.; Nicovich, J. M.; Wine, P. H. J. Phys. Chem. 1994,
98, 7128.
(35) Nicovich, J. M.; Kreutter, K. D.; van Dijk, C. A.; Wine, P. H. J.
Phys. Chem. 1992, 96, 2518.
(36) (a) Bellamy, L. J. The IR Spectra of Complex Molecules; Chapman
& Hall: London, 1975; Vol. 1. (b) Bellamy, L. J. The IR Spectra of Complex
Molecules: AdVances in Infrared Group Frequencies, Vol 2; Chapman &
Hall: London, 1980.
(40) (a) Baker, J.; Dyke, J. M. Chem. Phys. Lett. 1993, 213, 257. (b)
Baker, J.; Dyke, J. M. J. Phys. Chem. 1994, 98, 757.
(41) Winther, F.; Guarnieri, A.; Nielsen, O. F. Spectrochim. Acta 1975,
31A, 689.
(42) Norton, R. V.; Beverly, G. M.; Douglass, I. B. J. Org. Chem. 1967,
32, 3645.
(43) Grossert, J. S.; Langler, R. F. Can. J. Chem. 1977, 55, 407.
(44) Binder, G. E.; Schmidt, A. Spectrochim. Acta 1977, 33A, 815.
(45) Douglass, I. B., Farah, B. S.; Thomas, E. G. J. Org. Chem. 1961,
26, 1996.
(46) Barnes, I.; Becker, K. H.; Overath, R. D. In Proceedings of the
NATO AdVance Research Workshop, Tropospheric Chemistry of Ozone in
the Polar Regions; Niki, H., Becker, K. H., Eds.; NATO ASI Subseries,
Vol. 17, Global Environmental Change; Springer-Verlag, 1993; pp 371-
395.
(47) Crutzen, P. J.; Zimmermann, P. H. Tellus 1991, 43B, 136.
JP961452U