Kinetics of the reaction of CH3S with NO2 as a function of temperature

Article abstract of DOI:10.1016/S0009-2614(99)00579-5

The laser photolysis/laser-induced fluorescence technique was used to study the atmospheric reaction of the methylthiyl radical (CH₃S) with NO₂ as a function of temperature (263-381 K), in the pressure range 30-300 Torr. These measurements were performed using a newly constructed system of which a detailed description is given. The resulting Arrhenius expression, with an uncertainty of ±2σ, was k₁(T)=(3.8±0.3)×10^-11 exp[(160±22)/T] cm³ molecule^-1 s^-1. The measured value of k₁ was independent of pressure over the limited range employed. The absolute rate data obtained here are compared with previous studies carried out, at single temperature and as a function of temperature, by different techniques.

Full text of DOI:10.1016/S0009-2614(99)00579-5

16 July 1999
Ž
.
Chemical Physics Letters 308 1999 37–44
www.elsevier.nlrlocatercplett
Kinetics of the reaction of CH3S with NO2 as a function of
temperature
)
E. Martınez , J. Albaladejo, E. Jimenez, A. Notario, A. Aranda
´
´
UniÕersidad de Castilla-La Mancha, Facultad de Ciencias Quımicas, Departamento de Quımica Fısica, Campus UniÕersitario srn,
´
´
´
E-13071 Ciudad Real, Spain
Received 22 February 1999; in final form 29 April 1999
Abstract
The laser photolysisrlaser-induced fluorescence technique was used to study the atmospheric reaction of the methylthiyl
Ž
.
Ž
.
radical CH3S with NO2 as a function of temperature 263–381 K , in the pressure range 30–300 Torr. These
measurements were performed using a newly constructed system of which a detailed description is given. The resulting
y
11
3
exp 160"22 rT cm moleculey1
Ž .
Ž
.
wŽ
.
x
Arrhenius expression, with an uncertainty of "2s , was k1 T s 3.8"0.3 =10
sy1. The measured value of k1 was independent of pressure over the limited range employed. The absolute rate data
obtained here are compared with previous studies carried out, at single temperature and as a function of temperature, by
different techniques. q 1999 Elsevier Science B.V. All rights reserved.
1. Introduction
and new particles which depend on the distribution
of the likely major end products, SO2 , SO3 and
Organo-sulfur compounds from natural sources
account for ;60% of the total sulfur flux into the
atmosphere in the southern hemisphere and ;15%
in the northern hemisphere, mostly in the form of
Ž
.
CH3SO3H MSA . The SO2 , SO3 and MSA distri-
bution results from a complex DMS photooxidation
mechanism and is not well characterised as far as
parameters such as temperature, atmospheric ozone
and NOx levels are concerned. This oxidation of
DMS is principally initiated by OH and NO3 radi-
cals, the OH reaction being predominant in remote
Ž
x
.
dimethyl sulfide CH3SCH3, DMS from oceanic
w
phytoplankton 1,2 . It is postulated that if DMS is
oxidized to gas-phase H2 SO4, new condensation nu-
Ž
.
clei CN can be formed which leads to the forma-
tion of aerosols and cloud condensation nuclei
w x
oceanic atmosphere 4 . Besides, different studies
Ž
.
have indicated that the methylthiyl radical CH3S is
an important intermediate in the atmospheric oxida-
Ž
.
CCN . These, in turn, produce clouds. Both aerosols
and clouds can affect the heat balance of the atmo-
Ž
tion of DMS, dimethyl disulfide CH3SSCH3,
w x
sphere and possibly influence climate 3 .
The effect of DMS on the climate is critically
dependent on the production of gas-phase H2 SO4
.
Ž
. w
x
DMDS and methyl mercaptan CH3SH 5–10 . If
correct, its chemistry should play a large role in
determining the final oxidation products, and it is
necessary to understand the reaction of CH3S with
its most likely atmospheric oxidants, which include
O2 , NO2 and O3.
)
Corresponding author. E-mail: jalba@qifi-cr.uclm.es; fax:
q34 26 295318
0009-2614r99r$ - see front matter q 1999 Elsevier Science B.V. All rights reserved.
Ž
.
PII: S0009-2614 99 00579-5

(
)
38
E. Martınez et al.rChemical Physics Letters 308 1999 37–44
´
The present Letter reports absolute rate constant
determination, as a function of temperature, for the
reaction:
k1
CH3SqNO2 ™ products.
1
Ž .
This rate coefficient has been measured, using a
newly constructed system, by the technique of laser
Ž
.
photolysisrlaser-induced fluorescence LPrLIF .
Only two different laboratories have previously
w
x
measured this rate constant 11–14 , showing the
results a significant discrepancy. Balla et al. and
Turnipseed et al. investigated this reaction as a func-
tion of temperature using LPrLIF systems and they
y11
Ž .
Ž
.
wŽ
exp 80 "
obtained k1 T s 8.3 " 1.4 = 10
60 rT cm3 moleculey1 sy1 and k1 T s 2.1"
.
x
Ž .
Ž
y11
.
0.5 = 10
exp 320 " 40 rT cm3 moleculey1
wŽ . x
sy1, respectively. The room temperature values ob-
y10
Ž
.
tained from these expressions are 1.1"0.1 =10
y11
cm3 moleculey1 sy1, re-
Ž
.
and 6.1"1.1 =10
spectively. In light of the error limits given by these
authors, the agreement between these two values is
relatively poor, and a further measurement, from a
different laboratory, may prove useful. In this con-
text, we have reinvestigated the temperature depen-
dence of the reaction of CH3S with NO2.
Fig. 1. Schematic diagram of the LPrLIF system used in this
kinetic study.
2. Experimental
Ž
was a frequency-doubled pulsed dye laser Con-
.
tinuum ND60; dye used: LDS751 from Exciton
pumped by a Nd:YAG laser Continuum NY 81CS-
10 which provided pulses of 6 ns with 0.08 cm
Ž
The experiments were performed using a newly
constructed LPrLIF system that is shown schemati-
cally in Fig. 1. The reaction cell consisted of a 200
cm3 temperature-jacketed Pyrex cell with orthogonal
y1
.
bandwidth resolution. Energies of -3 mJ pulsey1
of doubled light were used to avoid any saturation
effects. The probe beam, operated at 4–10 Hz, was
passed through the cell perpendicular to the photoly-
sis beam, intersecting it at the center of the reaction
cell. Light baffles were placed just inside the en-
trance and exit windows to reduce scattered laser
light. The red-shifted fluorescence resulting from
laser excitation, was collected by a photomultiplier
Ž
.
sealed with Viton O-rings quartz windows. CH3S
Ž
radicals were produced always in excess bath gas,
.
He along one axis in the reaction cell by either
photolysis of DMS at 193 nm using an ArF excimer
y1
Ž
laser Lambda Physik MINex, 0.9–6.0 mJ pulse
cmy2 , 9 ns, 4–10 Hz or photolysis of DMDS at 248
.
nm with the same laser but with a KrF gas mixture
1.5–4.0 mJ pulse
cmy2, 19 ns, 4–10 Hz . The
tube Thorn EMI 9813B, Bialk. perpendicular to
both the photolysis and probe beams, after passing
through two 7.5 cm focal length lenses and a cutoff
y1
Ž
.
Ž
.
energy of the photolysis beam was kept below 4 mJ
per pulse to minimize secondary chemistry. The
w
x
Ž .
the gated integrator of a boxcar averager B2 Stan-
ford Research Systems, SR250 and finally trans-
ferred to a computer via a SR245 computer interface
methylthiyl radical concentration, CH3S , was
probed by monitoring the fluorescence resulting from
filter lG400 nm . The signal was then input into
Ž
. Ž
Ž2
.
Ž2
.
.
laser excitation of the A A1 §X E electronic
transition at 371.4 nm. The LIF excitation source

(
)
E. Martınez et al.rChemical Physics Letters 308 1999 37–44
39
´
Ž
.
Stanford Research Systems, SR245 for subsequent
analysis.
The entire experiment was controlled by a pulse
reactor prior to preparation of gaseous reagents was
achieved by the combined use of a rotatory vacuum
3
Ž
.
pump Telstar RD-18, 18 m rh and a home-made
Ž
delay generator Stanford Research Systems, DG
535 in combination with a double-pulse generator
oil diffusion pump with liquid-nitrogen traps. The
.
reactor, during the experiments, was evacuated with
3
Ž
.
Ž
.
Pulsetek 233 used to fire externally the probe laser
a small rotatory pump Telstar 2G-6, 6 m rh after
Ž
.
Ž
.
and another SR250 boxcar averager B1 used to
scan automatically the probe laser with respect to the
photolysis laser. The timing sequence for pulsed
irradiation, laser probing and the procedure for pulse
control was similar to that described in Ref. 15 .
The delay time was scanned, to construct the tempo-
ral profile of CH3S, from a few microseconds after
passing through a needle valve Edwards LV10K in
order to achieve the required total pressure in the
reactor. Concentrations of the different gases in the
cell were calculated from measurements of the ap-
propriate mass flow rates and the total pressure,
which was measured by a capacitance manometer
w
x
Ž
.
Leybold CM100 or CM1000 . Corrections to the
Ž
the photolysis pulse to avoid effects due to vibra-
NO2 concentration due to the presence of N2O4,
w
x
tional relaxation, prompt emission of photolysis
using the calculated equilibrium constant 16 , were
typically -1% and neglected.
.
products and NO2 fluorescence to ;100 ms, de-
pending on the lifetime of the CH3S radicals formed
The temperature was regulated by flowing cooled
Ž
in the photolysis. Typically, each bin of this profile
methanol or a heated fluid water or polyethylene
Ž
.
Ž
corresponding to the CH3S fluorescence at a given
glycol from a temperature-regulated bath Selecta
Ultraterm 200 through the jacket of the cell. The
reaction zone temperature "1 K was monitored by
a calibrated thermocouple Ni–Cr–Ni, type K in-
serted through a movable injector opposite the
photomultiplier.
The chemicals had the following purities: He
Carburos Metalicos, 99.999% used without further
purification. NO2 Praxair, 99.5% was degassed
several times before use, diluted in He 1% and
stored in 10 l glass bulb. DMS Aldrich, )99%
and DMDS Aldrich, 99% were further purified by
.
delay time between the photolysis and the probe
.
Ž
.
laser pulses was averaged for 5 laser shots; the LIF
Ž
.
decay curves were divided into 200 bins and usually
averaged five times to improve the signal-to-noise
Ž
.
ratio. The detection limit for CH3S SrNs1 was
;7=109 molecules cmy3 in 60 Torr of He. The
energy of the photolysis and probe beam, that should
remain constant during the acquisition of the tempo-
ral profile data, was monitored by a disk calorimeter
to check for stability. In general, the energies of the
Ž
.
´
Ž
.
Ž
.
Ž
.
Ž
.
Ž
.
two beams were acceptably stable "10% over the
time necessary for data acquisition.
trap-to-trap distillation before use, and then, diluted
Ž
.
All experiments were carried out in a large excess
in He 0.5–1% and stored in a glass bulb. ArF and
KrF gas mixtures from Praxair were prepared with
the required specifications of Lambda Physik.
Ž
.
of buffer gas He under slow gas flow conditions
Ž
.
50–300 sccm . The flow rate was fast enough to
prevent the accumulation of photolysis or reaction
products, but slow enough that the mixture could be
considered static during the measurement of a kinetic
3. Results and discussion
Ž
decay curve. The mass flow controllers used He:
Qualiflow AFC260; NO2 and precursor: Qualiflow
Ž .
1 were
All kinetic measurements of reaction
.
AFC 50.00 were conveniently calibrated. The car-
made under pseudo-first-order conditions over a
range of temperatures 263–381 K, using an excess
Ž
.
rier gas and the precursor of CH3S DMS or DMDS
were mixed prior to entering the reaction cell and
flowed into the reaction vessel. NO2 was added
down a movable injector, to minimize contact with
the precursor and suppress heterogeneous reactions.
The details of the vacuum system and associated
Ž
.
of NO2 and buffer gas 30–300 Torr of He over
CH3S radicals. The CH3S temporal profile was gov-
erned by the following reactions:
DMS or DMDSqhn
w
x
monitoring equipment have been given hitherto 15 .
The evacuation of the gas line, storage bulbs and
™ CH3Sqother photoproducts,
2
Ž .

(
)
40
E. Martınez et al.rChemical Physics Letters 308 1999 37–44
´
k1
CH3SqNO2 ™ products,
1
Ž
.
kdiff
CH3S ™ loss .
3
Ž .
Under these conditions, with a large excess of
Žw
x
w
x.
NO2 NO2 G100 CH3S and in the absence of
secondary reactions, the decay of CH3S should fol-
low a simple exponential rate law:
CH3S s CH3S exp ykX t ,
4
Ž .
w
x
w
x
Ž
.
t
0
with
X
w
x
k sk1 NO2 qkdiff
.
5
Ž .
kX obtained from the exponential fittings of the LIF
signals is the pseudo-first-order rate constant, k1 the
Ž
.
absolute rate coefficient for the reaction of CH3S
with NO2 , and kdiff , the CH3S loss rate coefficient
in the absence of reactant, due mostly to diffusion
out of the detection zone and, to a lesser extent,
reactions with impurities or other radicals in the
system. The bimolecular rate coefficient for reaction
Fig. 2. Dependence of kX on reactant concentration for CH3Sq
NO2 reaction with 193 nm photolysis of DMS as the source of
CH3S radical. Experimental conditions are T s263 K, pT s60
Torr and a flow rate of 200 sccm. Solid line is obtained from a
linear least-squares analysis of the data and gives the bimolecular
y11
Ž
.
rate constant, k1s 7.12"0.22 =10
cm3 moleculey1 sy1
,
where the uncertainty is 2s and represents precision only.
Ž .
1 , k1, was then obtained from the least-squares fit
X
w
x
of the straight lines of k vs. NO2 plots.
Profiles of CH3S were exponential in all cases,
14
13
Ž
.
Ž
.
varied in the range 2–9 =10 and 3–24 =10
indicating a negligible regeneration of CH3S by
X
w
x
molecule cmy3, respectively. The corresponding
secondary reactions. A typical plot of k vs. NO2 is
shown in Fig. 2. The values of kdiff obtained from
Ž
.
range for the initial CH3S concentration was 4–35
Ž
.
=1011 molecule cmy3, although in most cases the
concentration of CH3S was kept below 1012 molecule
cmy3 to minimize recombination. When DMS was
the photolytic precursor some experiments were per-
formed with 1–5 Torr of O2 added to the reaction
mixtures. This addition is sufficient to rapidly scav-
the intercepts of these plots "2s are in the order
of magnitude of those obtained in experiments car-
ried out in the absence of any reactant at different
total pressures of He kdiff s130–17 sy1 for a total
Ž
.
pressure range of 20–200 Torr . A summary of the
experimental conditions and the measured values in
the present work for the rate constant, k1, is given in
Table 1.
Ž
.
enge CH3 formed in the photolysis into CH3O2
and prevent the reaction of CH3 with DMS, that may
w
x
regenerate CH3S 14 . No difference was observed
in the linearity of the CH3S decays or the rate
coefficients obtained.
The reaction was firstly studied by photolyzing
y18
Ž
.
DMDS at 248 nm where s 248 nm s1.24=10
cm2 moleculey1
Ž .
, F DMDS CH3S s 1.80 " 0.21
w
x
As Table 1 shows, the experiments were per-
formed in He at a total pressure range of 30–300
Torr and flow rates between 50 and 300 sccm.
Room-temperature measurements of k1 were made
using DMDS as source of CH3S at different total
14,17 . In this region, NO2 does not absorb appre-
y20
Ž
Ž
.
ciably s 248 nm s2.75=10
cm2 moleculey1
.
w
x
18,19 . We also employed the photolysis of DMS at
y 17
Ž
Ž
.
.
193 nm
s 193 nm s 1.13 = 10
cm2
moleculey1, FDMS CH3S s0.8"0.1 14 , where
Ž
. w x
sNO2 193 nm is ;2.4=10y19 cm2 moleculey1
Ž
.
Ž
.
pressures 30, 60, 90 and 200 Torr with a flow rate
of 50 sccm and, also using DMS at a total pressure
of 60 and 300 Torr, employing flow rates from 150
to 300 sccm. Clearly, k1 does not show any pressure
w
x
20 , and there was not difference in the rate coeffi-
cient with that obtained using 248 nm photolysis of
DMDS. Concentrations of DMDS and DMS were

(
)
E. Martınez et al.rChemical Physics Letters 308 1999 37–44
41
´
Table 1
Summary of experimental conditions and the obtained rate constants for the reaction studied in this work
w
x
T
pT
Flow rate
NO2
Ž
k1 "2s
Ž
Torr
sccm
10 molecule cm y3
cm moleculey1 sy1
14
3
Ž
K
.
Ž
.
Ž
.
.
.
y11
y11
y11
y11
y11
y11
y11
y11
y11
y11
y11
y11
y11
y11
y11
y11
y11
y11
y11
y11
y11
y11
y11
y11
y11
y11
y11
Ž
Ž
Ž
Ž
Ž
Ž
Ž
Ž
Ž
Ž
Ž
Ž
Ž
Ž
Ž
Ž
Ž
Ž
Ž
Ž
Ž
Ž
Ž
Ž
Ž
Ž
Ž
.
263
60
60
200
60
200
60
60
60
60
30
60
90
200
60
60
60
300
60
60
60
60
60
60
60
60
60
60
50
50
50
200
200
50
50
50
50
50
0.3–3.6
0.6–3.4
0.4–4.0
0.5–4.0
0.3–3.0
0.3–3.0
0.3–3.1
0.5–4.4
0.5–3.5
0.6–3.5
0.7–4.1
0.4–2.1
0.9–4.0
0.4–2.9
0.2–3.2
0.5–4.7
0.5–5.2
0.5–4.4
0.4–4.0
0.3–3.4
0.4–3.0
0.8–3.5
0.2–2.9
0.5–3.0
0.5–3.4
0.4–3.4
0.2–3.5
7.08"0.71 =10
263a
263a
263a
263a
269a
275
7.06"0.34 =10
.
.
7.08"0.30 =10
.
7.12"0.22 =10
.
7.05"0.14 =10
.
7.09"0.59 =10
.
6.72"0.42 =10
.
6.76"0.75 =10
286
295a
298
6.65"0.67 =10
.
.
6.67"0.66 =10
.
6.59"0.36 =10
298
298
298
50
50
50
.
6.79"0.43 =10
.
6.36"0.62 =10
298a
298a
298a
298a
303a
303a
303a
310
150
200
300
200
50
100
200
50
50
200
50
100
50
150
6.49"0.56 =10
.
.
6.69"0.21 =10
.
6.43"0.18 =10
.
6.16"0.48 =10
.
6.32"0.15 =10
.
6.37"0.15 =10
.
6.36"0.21 =10
.
6.33"0.28 =10
.
323
6.25"0.54 =10
333a
348
6.12"0.42 =10
.
.
6.11"0.43 =10
355a
363
6.00"0.55 =10
.
.
6.06"0.17 =10
381a
5.98"0.34 =10
.
a193 nm photolysis of DMS as the photolysis source of CH3S.
or flow rate dependence within the given uncertainty
ranges.
using the LPrLIF technique. At single temperature,
this last laboratory has also reported two additional
measurements by Tyndall et al. using a LPrLIF
Ž
The variation of k1 with temperature over the
.
w
x
range 263–381 K is shown in Fig. 3. Over this
system 13 and Domine et al. employing a discharge
´
narrow temperature range, our data fit the expression
flow tube with photoionization mass spectrometric
Ž
.
w
x
with standard errors, "2s :
detection 11 .
In the studies carried out at single temperature, by
k T s 3.8"0.3 =10y11 exp 160"22 rT
1Ž .
Ž
Ž
.
Ž
.
Tyndall et al. and Domine et al., they obtained rate
´
cm3 moleculey1 sy1
,
6
Ž .
constant values of in cm3 moleculey1 sy1
.
Ž
.
.
y11
Ž .
Ž
.
Ž
Ž .
which indicates that reaction 1 exhibits a slight
negative temperature dependence with a value of
k1 298 K s 6.1"0.9 =10
and k1 297 K s
5.1"0.9 =10y11, respectively. As seen in Table
.
Ž
y11
k1 298 K s 6.6 " 1.0 = 10
cm3 moleculey1
2, our value of k1 298 K is in good agreement with
these two room-temperature measurements.
Ž
.
Ž
.
Ž
.
sy1
.
Table 2 summarizes the result obtained in this
work together with those found in the literature for
CH3SqNO2. The bimolecular coefficient, k1, has
been previously measured as a function of tempera-
On the other hand, as it has been previously
noted, the two previous measurements of k1 as a
function of temperature made by Balla et al. and
Ž
.
Ž
Turnipseed et al. give, respectively, k1 298 K s 1.1
y10
y11
w
x
w
x
.
Ž . Ž .
and k1 298 K s 6.1"1.1 =10
ture by Balla et al. 12 and Turnipseed et al. 14
"0.1 =10

(
)
42
E. Martınez et al.rChemical Physics Letters 308 1999 37–44
´
Fig. 3. Arrhenius plot of the experimental results obtained. Errors are chosen so as to bracket all measured k1 "2s values.
3
y1
in cm molecule sy1 . In light of the error limits
coefficient measured in our laboratory was also inde-
Ž
.
w x
pendent of the CH3S , which was varied in the
0
given by these authors, there is not agreement be-
tween these two values whereas the bimolecular rate
constant reported in this work at Ts298 K, agrees
with that of Turnipseed et al. Furthermore, our value
of k1 298 K and the rate coefficient obtained by
Turnipseed et al. at the same temperature agree with
those measured at single temperature see Table 2 .
The absolute rate coefficient reported by Balla et
al. at room temperature is higher than ours and all
the others published data for k1. Possible reasons for
this higher value have been discussed by Tyndall et
11
range 4–35 =10 molecule cmy3, and unaffected
Ž
.
by changes in the flow velocity in the cell in the
y1
Ž
.
range studied 1–6 cm s . This variation of tres
Ž
.
w x
and CH3S is important to ascertain that the correct
0
w
x
homogeneous rate coefficient was measured 13 .
Balla et al. found a dependence of their constant on
the flow rate and they interpreted this dependence
due to the thermal reaction between the photolytic
Ž
.
Ž
.
precursor DMDS and NO2 ,
CH SSCH q4NO ™ CH SO 2 q4NO ;
7
Ž .
Ž
.
3
3
2
3
2
w
x
al. 13 . The measured constant, k1, obtained by
Tyndall et al. was independent of the residence time
however, Tyndall et al. did not observe substantial
production of NO when DMDS and NO2 were flowed
Ž
.
w
x
in the vessel tres and the CH3S . The absolute
0
Table 2
Ž
.
Absolute second-order rate constants "2s for the reaction of CH3S with NO2
k298 K
Ž
A
Ž
EarR
Technique
Authors
3
3
cm moleculey1 sy1
cm moleculey1 sy1
.
.
Ž .
K
y10
y11
Ž
Ž
Ž
Ž
Ž
.
.
.
.
Ž
.
1.1"0.1 =10
8.3"1.4 =10
y80"60
LPrLIF
Balla et al.a
y11
6.1"0.9 =10
–
–
Ž
–
–
LPrLIF
Tyndall et al.b
y11
5.1"0.9 =10
discharge flow tube
LPrLIF
LPrLIF
Domine et al.c
´
y11
y11
y11
y11
Turnipseed et al.d
this Letter
.
.
6.1"1.1 =10
2.1"0.5 =10
3.8"0.3 =10
y320"40
y160"22
.
Ž
6.6"1.0 =10
a
b
c
d
w
x
w
x
w
x
w x
Ref. 12 . Ref. 13 . Ref. 11 . Ref. 14 .

(
)
E. Martınez et al.rChemical Physics Letters 308 1999 37–44
43
´
together in the dark, and they thought that the varia-
tion of the rate coefficient was due to depletion of
NO2 by reaction, since, in the mechanism proposed,
up to three molecules of NO2 can be consumed for
between the two previous studies. These results to-
gether with previous studies that have shown that
Ž
Ž
.
CH3SO and NO are the primary products F NO s
Ž
.
.
0.8"0.2 and F CH3SO s1.07"0.15 of this re-
action 11,13 are consistent with the following rapid
atom transfer reaction as being the major reaction
channel:
Ž
w
x.
w
x
every CH3S formed see Ref. 13 . It is quite possi-
ble that the high radical concentrations used by Balla
Ž
et al. in their study photolysis energy of 18 mJ
pulsey1 could account for this difference in their
.
CH3SqNO2 ™ CH3SOqNO .
8
Ž .
rate constant and, also, the formation of a product
that may interfere the detection of CH3S. Tyndall et
al. actually observed such an interference and at-
tributed it possibly to the fluorescence of CH3SO2
The absolute rate constant was also independent
Ž
.
of the precursor used DMS or DMDS , the CH3S
precursor concentration, laser fluence and flow rate
in the ranges studied. At room temperature, our
results for the reaction of CH3S with NO2 are in
good agreement with the two previous single temper-
ature measurements for the same reaction and with
the value reported by Turnipseed et al. These values
are sensibly lower than the bimolecular coefficient
obtained by Balla et al. at room temperature.
Ž
formed in the reaction of CH3SO produced in the
Ž ..
reaction 1 with NO2 , that it has been investigated
w
x
by several authors 11,13,21 . Furthermore, Balla et
al. and Tyndall et al., used photolysis of DMDS as
precursor of CH3S as was the case for the majority
of measurements made by Turnipseed et al., but 193
nm photolysis of DMS was also used by this last
author. Likewise, both DMS and DMDS, were used
to produce CH3S in our study, being k1 independent
of the precursor. As in Turnipseed et al. investiga-
For atmospheric applications, the present study
Ž .
confirms that reaction 1 can be an important oxida-
tion route of CH3S with NO2 concentrations at ppb
levels or lower and it indicates that this reaction is
also of potential atmospheric importance. In the clean
marine troposphere, the reactions of CH3S with NOx
tion, rate constant obtained here was also unaffected
y1
Ž
by changes in laser fluence 0.9–6.0 mJ pulse
cmy2
.
.
w
x
species are negligible 22 . However, in coastal re-
gions where polluted continental air can be trans-
ported over the oceans, NOx concentrations can reach
a few parts per billion, and oxidation of CH3S via
The published data by Balla et al., Turnipseed et
al. and the measurements presented in this work,
demonstrate that k1 exhibits a slight negative tem-
perature dependence. Nevertheless, the temperature
dependence reported by Balla et al. and that found in
our work is weaker than the dependence observed by
Ž .
reaction
reaction with O3, the major atmospheric loss of
1 can become comparable to that via
w
x
CH3S 22,23 .
Ž
.
Turnipseed et al. see Table 2 . The pre-exponential
Finally, this study may contribute to extend the
database for this reaction that is very limited.
factor A obtained here is two times higher than that
y11
ŽŽ
.
of Turnipseed et al. 3.8"0.3 =10
compared
y11
to 2.1"0.5 =10
cm3 moleculey1 sy1 while
.
EarR is two times lower y160"22 and y320"
40 K, respectively . Although, no definite reason for
this difference has been identified 14 , it may be due
to a different influence of the product formed in
Ž
.
Ž
Acknowledgements
.
This work was supported by the European Com-
mission DOMAC project No. ENV4-CT97-0410.
w
x
Ž .
reaction 1 , that, as noted above, may interfere the
detection of CH3S. The temperature dependence re-
ported by Balla et al. is weaker than ours, being both
of them weaker than that observed by Turnipseed et
References
w x
1
T.S. Bates, B.K. Lamb, A. Guenther, J. Dignon, R.E. Stoiber,
J. Atmos. Chem. 14 1992 315.
Ž
.
Ž
.
al. see Table 2 .
w x
2
M.O. Andreae, in: J.N. Galloway, M.O. Andreae, H. Rodhe
In conclusion, our data indicate that reaction of
CH3S with NO2 is essentially pressure independent
in the range studied and exhibits a small negative
temperature dependence with values of A and EarR
Ž
.
Eds. , Biogeochemical Cycling of Sulfur and Nitrogen in the
Remote Atmosphere, D. Reidel, Dordrecht, 1985, p. 5.
w x
3
R.J. Charlson, J.E. Lovelock, M.O. Andreae, S.G. Warren,
Ž
.
Ž
.
Nature London 326 1987 655.

(
)
44
w x
E. Martınez et al.rChemical Physics Letters 308 1999 37–44
´
Ž
.
w
x
4
w x
5
D.J. Cooper, J. Atmos. Chem. 25 1996 97.
16 M.W. Chase Jr., C.A. Davies, J.R. Downey Jr., D.J. Frurip,
F. Yin, D. Grosjean, J.H. Seinfeld, J. Atmos. Chem. 11
R.A. McDonald, A.N. Syverud, J. Phys. Chem., Suppl. 1 14
Ž
.
Ž
.
1990 309.
1985 .
w x
w
w
w
x
6
F. Yin, D. Grosjean, R.C. Flagan, J.H. Seinfeld, J. Atmos.
17 C.H. Hearn, E. Turcu, J.A. Joens, Atmos. Environ. 24A
Ž
.
Ž
.
Chem. 11 1990 365.
1990 1939.
w x
Ž
.
x
7
S. Hatakeyama, H. Akimoto, J. Phys. Chem. 87 1993 2387.
18 W. Scheneider, G.K. Moortgat, G.S. Tyndall, J.P. Burrows,
w x
Ž
.
Ž .
J. Photochem. Photobiol. 40 1987 195.
8
D. Grosjean, Environ. Sci. Technol. 18 1984 460.
w x
x
9
H. Niki, P.D. Maker, C.M. Savage, L.P. Breitenbach, Int. J.
19 W.B. DeMore, S.P. Sander, D.M. Golden, R.F. Hampson,
M.J. Kurylo, C.J. Howard, A.R. Ravishankara, C.E. Kolb,
M.J. Molina, Chemical kinetics and photochemical data for
use in stratospheric modeling, Evaluation No. 12, Jet Propul-
Ž
.
Chem. Kinet. 15 1983 647.
w
w
w
w
w
w
x
10 N.R. Jensen, J. Hjorth, C. Lohse, H. Slov, G. Restelli,
Ž
.
Atmos. Chem. 14 1992 95.
x
Ž
11 F. Domine, T.P. Murrells, C.J. Howard, J. Phys. Chem. 94
sion Lab., Pasadena, CA, 1997, JPL Publ. No. 97-4 and
.
references therein .
´
Ž
.
1990 5839.
x
w
w
w
w
x
12 R.J. Balla, H.H. Nelson, J.R. McDonald, Chem. Phys. 109
20 A.M. Bass, A.E. Ledford Jr., A.H. Laufer, J. Res. Natl. Bur.
Ž
.
Ž .
Stand. Sect. A 80 1976 143.
1986 101.
x
Ž
.
x
13 G.S. Tyndall, A.R. Ravishankara, J. Phys. Chem. 93 1989
21 A. Mellouki, J.L. Jourdain, G. Le Bras, Chem. Phys. Lett.
Ž
.
2426.
148 1988 231.
x
x
14 A.A. Turnipseed, S.B. Barone, A.R. Ravishankara, J. Phys.
22 G.S. Tyndall, A.R. Ravishankara, Int. J. Chem. Kinet. 23
Ž
.
Ž
.
Chem. 97 1993 5926.
1991 483.
x
x
Ž
.
15 E. Martinez, J. Albaladejo, A. Notario, D. Husain, J. Pho-
23 G.S. Tyndall, A.R. Ravishankara, J. Phys. Chem. 93 1989
4707.
Ž
.
tochem. Photobiol. 95 1996 103.