Temperature dependence of the heterogeneous reaction of carbonyl sulfide on magnesium oxide

Article abstract of DOI:10.1021/jp711302r

The experimental determination of rate constants for atmospheric reactions and how these rate constants vary with temperature remain a crucially important part of atmosphere science. In this study, the temperature dependence of the heterogeneous reaction of carbonyl sulfide (COS) on magnesium oxide (MgO) has been investigated using a Knudsen cell reactor and a temperature-programmed reaction apparatus. We found that the adsorption and the heterogeneous reaction are sensitive to temperature. The initial uptake coefficients (γ_t(Ini)) of COS on MgO decrease from 1.07 ± 0.71 × 10^-6 to 4.84 ± 0.60 × 10^-7 with the increasing of temperature from 228 to 300 K, and the steady state uptake coefficients (γ_t(SS)) increase from 5.31 ± 0.06 × 10^-8 to 1.68 ± 0.41 × 10^-7 with the increasing of temperature from 240 to 300 K. The desorption rate constants (k_des) were also found to increase slightly with the enhancement of temperature. The empirical formula between the uptake coefficients, desorption rate constants and temperature described in the form of Arrhenius expression were obtained. The activation energies for the heterogeneous reaction and desorption of COS on MgO were measured to be 11.02 ± 0.34 kJ·mol^-1 and 6.30 ± 0.81 kJ-mol^-1, respectively. The results demonstrate that the initial uptake of COS on MgO is mainly contributed by an adsorption process and the steady state uptake is due to a catalytic reaction. The environmental implication was also discussed.

Full text of DOI:10.1021/jp711302r

2820
J. Phys. Chem. A 2008, 112, 2820-2826
Temperature Dependence of the Heterogeneous Reaction of Carbonyl Sulfide on
Magnesium Oxide
Yongchun Liu, Hong He,* and Qingxin Ma
State Key Laboratory of EnVironmental Chemistry and Ecotoxicology, Research Center for Eco-EnVironmental
Science, Chinese Academy of Sciences, Beijing 100085, China
ReceiVed: NoVember 29, 2007; In Final Form: January 9, 2008
The experimental determination of rate constants for atmospheric reactions and how these rate constants vary
with temperature remain a crucially important part of atmosphere science. In this study, the temperature
dependence of the heterogeneous reaction of carbonyl sulfide (COS) on magnesium oxide (MgO) has been
investigated using a Knudsen cell reactor and a temperature-programmed reaction apparatus. We found that
the adsorption and the heterogeneous reaction are sensitive to temperature. The initial uptake coefficients
(γt(Ini)) of COS on MgO decrease from 1.07 ( 0.71 × 10-6 to 4.84 ( 0.60 × 10-7 with the increasing of
temperature from 228 to 300 K, and the steady state uptake coefficients (γt(SS)) increase from 5.31 ( 0.06
× 10-8 to 1.68 ( 0.41 × 10-7 with the increasing of temperature from 240 to 300 K. The desorption rate
constants (kdes) were also found to increase slightly with the enhancement of temperature. The empirical
formula between the uptake coefficients, desorption rate constants and temperature described in the form of
Arrhenius expression were obtained. The activation energies for the heterogeneous reaction and desorption
of COS on MgO were measured to be 11.02 ( 0.34 kJ‚mol-1 and 6.30 ( 0.81 kJ‚mol-1, respectively. The
results demonstrate that the initial uptake of COS on MgO is mainly contributed by an adsorption process
and the steady state uptake is due to a catalytic reaction. The environmental implication was also discussed.
1. Introduction
surface, as well as the season and the time of day. For example,
the tropospheric temperature for latitude 40 ° N during June
decreases from room temperature (r.t.) to 220 K with the
increase of altitude.15 In reality, the temperature at the tropo-
pause can reach values much lower than 220 K, down to 180
K above the Antarctic in winter.15,16 Therefore, the experimental
determination of rate constants for important atmospheric
reactions and how these rate constants vary with temperature
remain a crucially important part of atmosphere science.15 In
the past decades, many researchers have investigated the
heterogeneous reactions of N2O5, ClONO2, NO2, HNO3, SO2,
O3, and so forth, on ice, sea salt, soot, and mineral oxides in
laboratory,17-20 while most of the work was performed at one
temperature point, generally, at room temperature. Additionally,
the mineral dust particle represents one of the largest natural
mass fractions of the global aerosol with an annual emission of
about 1000-3000 Tg.21 Even at altitudes of 10-12 km, mineral
ions including Mg2+, Ca2+, Na+, K+, and so forth were also
detected in the aerosol samples. It indicates that mineral dust
particles are also transported into the upper troposphere by
convection.22,23 Therefore, it is significant to investigate the
dependence of rate constants on the temperature for the
heterogeneous reactions of trace gases on mineral dust particles.
MgO is one of the typical components of mineral dust, whose
content is about 6.1 wt % of the inorganic components in the
Beijing area.11,24 On the other hand, the reaction mechanism of
COS on MgO indicates that the heterogeneous reaction of COS
on MgO may have important effect on the global sink of COS
in the troposphere.13 Therefore, in this work, we investigated
the temperature dependence in the range of 300-228 K for the
heterogeneous reaction of COS on MgO. The uptake coefficients
of adsorption and heterogeneous reaction, and the desorption
rate constants were measured with a Knudsen cell-mass
Carbonyl sulfide (COS) is the most abundant sulfur compound
in the atmosphere, with a rather uniform mixing ratio of about
500 part per trillion by volume (pptv) in the troposphere.1-3 Its
tropospheric lifetime is greater than 1 year,1,2 and the global
atmospheric lifetime is from 2 to 8.9 years, calculated by
different researchers.2,4 Chin et al.2 have reported that about
0.64 Tg‚yr-1 of COS is transported to the stratosphere from
the troposphere, where its photolysis is the important source
for the stratospheric sulfate layer.5 Therefore, the sinks of COS
in the troposphere should affect the stratospheric sulfate source.
Several atmospheric modeling studies have shown that atmo-
spheric particles can potentially alter the chemical partitioning
of gas-phase molecules, often acting as a sink for certain
species.6-8 Recently, a few studies have reported the hetero-
geneous reactions of COS on atmospheric particles, and mineral
oxides including Al2O3, SiO2, Fe2O3, CaO, MgO and MnO2 as
well.9-14 Hydrogen thiocarbonate (HSCO2-) was found as a key
intermediate. Gaseous carbon dioxide (CO2) and surface sulfate
(SO42-) were found to be the gaseous and surface products,11-14
respectively. Surface sulfite11,12 and elemental sulfur9,10 were
also observed as surface sulfur species. Additionally, hydrogen
sulfide (H2S) was detected as one of the gaseous products for
the heterogeneous reaction of COS on MgO.13 The previous
work demonstrates that heterogeneous reaction of COS on
mineral dust is a potential sink for COS in the troposphere.
However, little attention was given to the reaction kinetic of
COS on mineral oxides.
On the other hand, the temperature in the Earth’s atmosphere
varies with latitude, longitude, and altitude above the earth’s
* Corresponding author. Phone: +86-10-62849123; fax: +86-10-
62923563; e-mail: honghe@rcees.ac.cn.
10.1021/jp711302r CCC: $40.75 © 2008 American Chemical Society
Published on Web 02/27/2008

Temperature Dependence of COS-MgO Reaction
J. Phys. Chem. A, Vol. 112, No. 13, 2008 2821
spectrometer (KCMS). The energy change in the reaction
coordinate was also discussed. The results are helpful for further
understanding the heterogeneous reaction of COS on mineral
dust particles as well as the spatial distribution of activity for
this reaction in the troposphere. In addition, the uptake coef-
ficients at different temperatures will supply the basic data for
model studies.
2. Experimental Details
2.1. Materials. The MgO sample (A.R.) used in this
experiment was supplied by Haizhong Chemical Plant in Tianjin.
2-
The impurities in the MgO sample include Cl (0.01%), SO4
3-
2-
(0.02%), PO4 (0.003%), CO3 (1.5%), Na (0.05%), K
(0.005%), Ca (0.02%), Fe (0.005%), Cu (0.001%), Zn (0.005%),
As (0.0001%), Ba, and Sr (0.005%). The following chemicals
were used without further purification: carbonyl sulfide (COS,
2%, COS/N2, Scott Specialty Gases, Inc.); N2 and O2 (99.99%
purity, Beijing AP BEIFEN Gases, Inc.).
Figure 1. The uptake of COS and desorption of surface products on
100.0 mg MgO at 300 K. The average COS partial pressure in the
Knudsen cell reactor was 5.3 ( 0.3 × 10-6 Torr. (A), (B), (C) are the
temporal concentration profiles for COS (m/e ) 60), CO2 (m/e ) 44),
and H2S (m/e ) 34) in the uptake experiment; (D), (E), (F) are these
in the in situ desorption experiment.
2.2. Characterization of Sample. Nitrogen Brunauer-
Emmett-Teller (BET) physisorption measurement was per-
formed using a Micromeritics ASAP 2000 automatic equipment.
It was shown that the MgO sample has a total surface area of
14.6 m2‚g-1. The X-ray powder diffraction pattern was collected
from 10 to 90° 2θ on a D/max-RB automatic powder X-ray
diffractometer using Cu KR irradiation. The MgO sample was
identified to be periclase with the three main 2θ peaks at
42.918°, 62.300° and 78.622°.
× 10-7 Torr. After the system was cooled to a given temperature
(228-300 K), the sample cover was closed. 1.51% COS
balanced with simulated air (21% O2 and 79% N2) was
introduced into the reactor chamber through a leak valve. The
pressure in the reactor was measured using an absolute pressure
transducer (BOC Edward). Prior to the experiments, the reactor
chamber was passivated with COS in air for 150 min until a
steady state of QMS signal was established while the oxide
samples were isolated from the gas by the sample cover. Uptake
measurements on all samples were obtained with an average
COS partial pressure of 5.3 ( 0.3 × 10-6 Torr, which was
equivalent to 1.7 ( 0.2 × 1011 molecules‚cm-3 or 7.0 ( 0.3
ppb in the atmosphere. The uptake coefficients and desorption
rate constants were calculated based on the KCMS data.
TPR Experiment. The instantaneous experiment for the
heterogeneous reaction of COS on MgO was performed with a
temperature-programmed reaction (TPR) apparatus. It consists
of a temperature-programmed tube oven and a QMS (Hiden
HPR20). MgO samples (0.5 g) were pretreated in simulated air
at 573 K for 1 h. After the sample was cooled to 298 K, it was
exposed to 0.1% COS balanced with simulated air with a total
flow rate of 100 mL‚min-1. The temperature was increased from
298 to 430 K at 20 K‚min-1. The mass channels including m/e
) 60 (COS), 44 (CO2), 34 (H2S), and 64 (SO2) were monitored
with the QMS online. The blank experiment with a quartz tube
was also carried out under the same conditions.
2.3. Experimental Methods. KCMS Experiments. A Knudsen
cell reactor coupled to a quadrupole mass spectrometer (QMS,
Hiden, HAL 3F PIC) was used to determine the uptake
coefficients of COS on MgO. The mass spectrometer was
housed in a vacuum chamber equipped with a 300 L‚s-1
turbomolecular pump (Pfeiffer) and an ion gauge (BOC
Edward). The vacuum chamber between the QMS and the
Knudsen cell reactor was pumped by a 60 L‚s-1 turbomolecular
pump for differential pumping of the mass spectrometer and
an ion gauge (both from BOC Edward). The Knudsen cell
reactor consisted of a stainless steel chamber with a gas inlet
controlled by a leak valve, an escape aperture whose area can
be adjusted with an adjustable iris in the range of 0-300 mm2,
and a sample holder attached to the top ceiling of a circulating
fluid bath. The effective area of escape aperture was measured
in each independent experiment according to the attenuation of
the N2 signal from one steady state to another.25 It was about
0.40 mm2 in our experiments. The geometric area of the sample
holder (As) was 5.26 cm2. A blank flange, serving as sample
cover, attached to a linear translator made it possible to either
isolate or expose the sample to reactant gas. No corrections were
needed because the volume change upon opening the sample
cover was less than 0.52 cm3, and the total volume of reactor
chamber was 94.0 cm3. The seal between the cover and sample
holder was made with a Teflon O-ring. All exposed interior
surfaces including the surface of the sample holder were coated
with Teflon to provide a chemically inert surface. Blank
experiment revealed that there was no uptake due to the sample
holder and the sample cover. The temperature of the sample
holder, which was measured with an embedded Pt resistance
thermometer, could be controlled from 220 to 373 K within
(0.10 K by using a super thermostat and a cryofluid pump
(DFY 5/80, Henan Yuhua laboratory instrument factory).
The 40.0-120.0 mg of MgO samples were dispersed evenly
on a sample holder with alcohol and then dried at 393 K for 2
h. The pretreated samples and reactor chamber were evacuated
at 323 K for 6 h to reach a base pressure of approximately 5.0
3. Results and Discussion
3.1. Reaction Pathway and Kinetics for the Heterogeneous
Reaction of COS on MgO. Figure 1 shows a typical KCMS
profile for the uptake of COS on 100.0 mg of MgO at 300 K.
The average COS partial pressure in the Knudsen cell reactor
was 5.3 ( 0.3 × 10-6 Torr, which is equivalent to 1.7 ( 0.2 ×
1011 molecules‚cm-3 or 7.0 ( 0.3 ppb in the atmosphere. It
can be seen that the mass signal intensity of COS (m/e ) 60)
decreased dramatically after opening the sample cover (Figure
1A). The uptake of COS was accompanied by the production
of CO2 (m/e ) 44) and H2S (m/e ) 34) on MgO (Figure 1B,C).
In our previous work,13 we also observed the formation of
gaseous CO2 at 2341 and 2361 cm-1, and surface HS at 2578
cm-1 using in situ DRIFTS to study the heterogeneous reaction
of COS on MgO at 303 K. The KCMS results support the in
situ DRIFTS experiment results very well. In Figure 1A, the

2822 J. Phys. Chem. A, Vol. 112, No. 13, 2008
Liu et al.
initial uptake of COS was followed by a quick saturation
process, leading to an apparent steady-state uptake on the time
scale of this experiment. The quick saturation can be ascribed
to the adsorption process. It is very interesting that the maximal
yields of CO2 and H2S lagged behind the maximal consumption
of COS. This phenomenon is related to the adsorption of CO2
and H2S on the basic site of MgO. In Figure 1, the maximal
yield of H2S was also later than that of CO2, which could be
explained by the reasons that the acidity of H2S is stronger than
that of CO2, and thus H2S is more readily to be adsorbed on
the basic site of MgO.
In the end of the uptake process, the in situ desorption of
surface species was carried out. After the sample was isolated
from the feed gas by the sample cover, the feed gases were
ceased from being introduced into the reactor, and then the
escape aperture area was increased to 300 mm2 from 0.40 mm2
using a variable iris to increase the flow rate of gases from the
Knudsen cell reactor into the QMS detector. When the pressure
in the Knudsen cell decreased to 5.0 × 10-7 Torr, the sample
cover was opened, and the desorbed species were monitored
online with the QMS. As shown in Figure 1D-F, as the sample
cover opened, desorption of CO2 and H2S was clearly observed.
When the sample cover was closed, the QMS signal for CO2
and H2S decreased to their baselines. In Figure 1D, no desorption
of COS was detected. Using in situ DRIFTS, we identified
hydrogen thiocarbonate (HSCO2-) as the key intermediate,13
which is formed by the reaction of COS with surface hydroxyl.
-
Therefore, we here appoint HSCO2 as one of the adsorbed
species for COS adsorbed on MgO. The above results suggest
that COS can easily react with OH- on MgO, and then quickly
-
translate to HSCO2-, and the adsorbed HSCO2 can hardly
desorb from the surface. The desorption of CO2 and H2S or the
decomposition of the intermediate (HSCO2-) should be a rate-
determining step (RDS) for the heterogeneous reaction of COS
on MgO. These results also suggest that the reverse reaction
for the heterogeneous reaction of COS on MgO is very slow.
However, it should be noted that the concentration of COS in
the Knudsen cell reactor is very low. According to the signal-
to-noise ratio (SNR) of QMS signal, the low limit of the QMS
is about 2 × 1014 molecules (3σ) for COS. The quantity of
desorbed COS molecule may be lower than the sensitivity of
QMS. Although the desorption of COS was not observed
directly, according to the reversible principle, the reverse
reaction must occur, but it is very slow. On the basis of our
previous work13 and KCMS results, the reaction can be
described as R1.
Figure 2. The temporal concentration profile for temperature-
programmed reaction of COS on MgO in the range of 298-430 K at
20 K‚min-1. (A) 0.5 g MgO was exposed to 0.1% COS balanced with
simulated air, and (B) is the control experiment.
of experimental temperature. It also indicates that gaseous COS
does not decompose into other products under the experiment
conditions. However, the case is not true when the reactant gases
passing through 0.5 g MgO sample. As shown in Figure 2A,
the concentration of COS decreased dramatically with the
temperature increasing, which was accompanied by the increas-
ing of the concentration of CO2 and H2S. When the temperature
increased to 380 K, the concentration of COS decreased to the
lowest level, which corresponded to the highest concentration
of CO2 and H2S. Because the background noise of the QMS
for COS is about 200 CPS; therefore, the conversion for COS
at 380 K should be 100%. According to reaction R1, the amount
of produced H2S ought to be the same as that of CO2. Therefore,
the increase rate (slope) for H2S should be the same as that of
CO2 in Figure 2A. However, the increase rate for H2S is slower
than that of CO2 slightly. As discussed in section 3.1, the
adsorption ability of H2S on MgO is stronger than that of CO2.
On the other hand, H2S can be oxidized to other surface sulfur-
ka
kh
COS(g) y
\
kz COS(ads) y
\
z CO2(g) + H2S(g)
(R1)
d
3.2. The Effect of Temperature on the Heterogeneous
Reaction of COS on MgO. The results from KCMS and in
situ DRIFTS13 demonstrate that the heterogeneous reaction of
COS on MgO belongs to catalytic reaction. Therefore, temper-
ature should have an important influence on the reaction. It is
also significant to determine the step rate constants in R1. Figure
2 shows a temporal concentration profile for the TPR experiment
in the range of 298-430 K. Because the reactor for TPR could
not be cooled to low temperature (below 298 K), the experiment
above 298 K was performed. It can be seen in Figure 2B when
0.1% COS balanced with simulated air passed through the quartz
reactor (the control experiment), the concentration of COS (m/e
) 60), CO2 (m/e ) 44), and H2S (m/e ) 34) did not change
with the increasing of temperature. It suggests that the quartz
reactor has no catalytic activity for the reaction (R1) in the range
containing species, including element sulfur (S), sulfur dioxides
2- 9-14
(SO2), sulfite (SO32-), bisulfite (HSO3-), and sulfate (SO4
)
under the reaction conditions. Therefore, TPR results are in good
agreement with the KCMS results. In Figure 2B, the signal
intensity of COS is much higher than that of CO2 and H2S at
298 K, while it is not this way in Figure 2A. This difference is
due to the production of CO2 and H2S by reaction R1 on MgO
at 298 K (Figure 2A). The TPR results indicate that the
heterogeneous reaction of COS on MgO is sensitive to tem-

Temperature Dependence of COS-MgO Reaction
J. Phys. Chem. A, Vol. 112, No. 13, 2008 2823
Figure 3. The linear mass dependence for the uptake coefficient of
COS on MgO at 300 K.
Figure 4. The desorption rate verse adsorption quantity.
perature. Consequently, the kinetic and the temperature depen-
dence of heterogeneous reaction for COS on MgO will be
discussed in the following section.
In atmospheric chemistry, uptake coefficient is the most
commonly used kinetic parameter. The uptake coefficient (γ)
is defined by eq 1:
dn
-
dt
γ )
(1)
ω
where -(dn/dt) is the number of molecules lost from the gas-
phase per second due to the collision between gas molecules
and solid surface (molecules‚s-1), and ω is the total number of
gas-surface collisions per second, namely, collision frequency.26
For steady-state uptake, the uptake coefficient can be calculated
using the Knudsen cell equation:26-29
Figure 5. The relationship between the uptake coefficients of COS
on MgO and temperature.
the repetitive experiments. The true uptake coefficients for COS
on MgO were also measured at 228, 240, 248, and 265 K.
Ah (I0 - I)
γ )
(2)
When the temperature decreased to 228 K, only adsorption
of COS on MgO was observed. The saturated adsorptive
capacity for COS on MgO was measured to be 7.1 ( 1.2 ×
1018 molecules‚g-1. Taking into account the BET surface area
of MgO of 1.46 × 105 cm2‚g-1 and the size of an adsorbed
COS molecule around 10-15 cm2, the COS saturated adsorption
on MgO is close to one monolayer at 228 K. At other
temperatures, the concentration profiles for the uptake of COS
were similar to that at 300 K. H2S and CO2 were also found as
the gaseous products. Therefore, the uptake of COS is contrib-
uted by both adsorption and heterogeneous reaction. The values
of γt(Ini) and γt(SS) are given in Table 1. The rate constants
for adsorption, desorption, and heterogeneous reaction (catalytic
reaction) of COS on MgO are also shown in Table 1 and will
be discussed later. As can be seen in Table 1, the values of
γt(Ini) decreased with the increasing of temperature, while it
was reversed for γt(SS). It means that the uptake at the initial
stage is an exothermic process, and the uptake at the steady-
state needs an active energy. Therefore, we can conclude that
the initial uptake of COS on MgO is mainly due to an adsorption
process, while, at the steady state, the uptake is mainly
contributed by catalytic reaction.
As
I
where Ah is the effective area of the escape aperture (m2), As is
the geometric area of the sample holder (m2); I0 and I are the
mass spectral intensities with the sample holder closed and open,
respectively. As for the multilayer powder sample, the linear
mass dependent (LMD) region of the observed uptake coef-
ficient has been reported by different groups in various reaction
systems, and has been ascribed to the diffusion of reactant
molecules into the underlying layers of multilayer powder
samples.29-33 Therefore, the true uptake coefficients, γt, can be
calculated from
As
γt ) slope‚
(3)
(
)
SBET
where “slope” is the slope of the plot of γ versus sample mass
in the linear regime (kg-1), As is the geometric area (m2) of the
sample holder, and SBET is the specific surface area of the
particle sample (m2‚kg-1).34
As shown in Figure 3, there was a strong linear dependence
of uptake coefficients at the initial time γ(Ini) and at steady
state γ(SS) versus sample mass. The correlation coefficients
were 0.990 and 0.966, respectively. When the sample mass
increased to 120 mg, the uptake coefficients reached a constant
value. According to eq 3, γt(Ini) and γt(SS) at 300 K were
In our previous work, we have found that the heterogeneous
reaction of COS on MgO is a first-order reaction.14 Therefore,
the reaction rate constant can be derived from eq 4:
calculated to be 4.84 ( 0.60 × 10-7 and 1.68 ( 0.41 × 10-7
,
dc
dt
γ‚ω
V
-
) k‚c )
(4)
respectively. The error bar in Figure 3 was 15.0% obtained from

2824 J. Phys. Chem. A, Vol. 112, No. 13, 2008
Liu et al.
TABLE 1: The Kinetic Parameters for the Heterogeneous Reaction of COS on MgO
T (K)
228
240
248
265
300
γt(Ini)
γt(SS)
kads (s-1
1.07 ( 0.71 × 10-6
7.13 ( 0.19 × 10-7
5.31 ( 0.06 × 10-8
8.11 ( 0.22 × 10-1
1.39 ( 0.09 × 10-2
6.04 ( 0.07 × 10-2
5.46 ( 0.33 × 10-7
6.69 ( 0.46 × 10-8
6.31 ( 0.38 × 10-1
1.59 ( 0.03 × 10-2
7.73 ( 0.53 × 10-2
6.68 ( 0.12 × 10-7
9.09 ( 0.12 × 10-8
7.98 ( 0.14 × 10-1
4.84 ( 0.60 × 10-7
1.68 ( 0.41 × 10-7
6.15 ( 0.76 × 10-1
)
)
1.19 ( 0.79
kdes (s-1
1.21 ( 0.01 × 10-2
kh (s-1
)
1.09 ( 0.01 × 10-1
2.14 ( 0.52 × 10-1
Here, c is the molecule concentration of COS in the Knudsen
cell reactor (mol‚m-3); V is the volume of the Knudsen cell
reactor (m3); k is the first-order reaction rate constant (s-1).
When the mean free path of molecules in the Knudsen cell is
greater than the cell dimensions, the collision frequency between
the gas-phase molecules and the reactive surface can be
calculated by eq 5:34
which is in the linear range between the observed uptake
coefficient and the sample mass. When the temperature was
higher than 248 K, the effect of the heterogeneous reaction
became prominent and the correlation between Fdes(t) and Nads
-
(t) was very bad. Therefore, the desorption rate constants at
265 and 300 K were not given in Table 1. As shown in Table
1, the desorption rate constants increased with the increasing
of temperature, which means the desorption process is also an
endothermic process. From 228 to 248 K, kads, kdes, and kh are
in the order of kads > kh > kdes. The desorption rate constants
are lower than both the adsorption rate constants and the
heterogeneous reaction rate constants, which means that the
desorption process is a slow step. It well supports the above
assumption that the reverse reaction or desorption of COS is
very slow. On the other hand, for the positive reaction, the
heterogeneous reaction rate constants are lower than the
adsorption rate constants, even at 300 K. The fast adsorption
process suggests that the adsorption should be a relatively
important heterogeneous process for COS on the surface of
mineral dust particles in the atmosphere, especially, at low
temperature. However, when compared with the heterogeneous
reaction (catalytic reaction), the total amount of COS adsorbed
on the mineral dust particle should be a limited quantity on a
long time scale.
Vh‚c‚As
4
ω )
(5)
where Vh is the average molecular speed (m‚s-1). Therefore, the
rate constant can be calculated by eq 6:35
Vh‚γ‚As
4V
k )
(6)
When the true uptake coefficient (γt) is used in eq 6, the
geometric area of the sample holder (As) should be replaced
with the effective collision area (ABET). The rate constants for
the heterogeneous reaction (kh) in Table 1 were calculated on
the basis of the steady state uptake coefficients. At the moment
the sample was exposed to the reactant gas, the lost of gaseous
molecules due to the heterogeneous reaction may be neglected
when compared with adsorption. Therefore, the adsorption rate
constants can be calculated based on the initial uptake coef-
ficients according to eq 6.22,36
Figures 5 and 6 show the linear regression of γt(Ini), γt(SS),
and kdes with the reciprocal value of temperature (1/T). The
correlation coefficients are 0.813, 0.999, and 0.992, respectively.
Therefore, γt(Ini), γt(SS), and kdes can be described in the form
of the Arrhenius expression. Namely, ln γt(Ini) ) -16.58 +
601.39/T; ln γt(SS) ) -11.21 - 1325.04/T; ln kdes ) -1.10 -
758.50/T. A linear regression of these data yields an activation
energy for R1 on MgO of 11.02 ( 0.34 kJ‚mol-1, and the
As for the desorption rate constants, Seisel et al.22,36 have
proposed a calculation method based on the flux balance
equation according to the uptake profile using KCMS. That is,
Fdes(t) ) kdesNads(t)
(7)
(8)
activation energy for desorption is 6.30 ( 0.81 kJ‚mol-1
.
kini
Usually, for a given reaction in the atmosphere, when the
activation energy is greater than 20 kJ‚mol-1, the role of this
reaction in atmospheric chemistry is regarded as unimportant.15
It suggests that the heterogeneous reaction of COS on mineral
oxides should not be neglected in the troposphere. The energy
Fdes(t) ) F(t) 1 +
- F0
(
)
kesc
where Fdes(t) is the flow out of the reactor at time t
(molecules‚s-1); kdes is the desorption rate constant (s-1); Nads
-
(t) is the number of adsorbed molecule on the surface; F0 is the
flow into the reactor (molecules‚s-1); kini is the initial rate
constant and equals kads (s-1); kesc is the escape rate constant of
the Knudsen cell (s-1), and kesc ) ω‚Ah. The number of
molecules adsorbed on the surface at time t, Nads(t), can be
determined by integrating the MS signal between t ) 0 and the
desired time t. If the heterogeneous process includes a catalytic
reaction, eq 8 should be corrected with eq 9:
change in the reaction coordinate was shown in Figure 7, where
-
the energy gap between the adsorbed intermediate (HSCO2
)
kini
Fdes(t) ) F(t) 1 +
+ khNads(t) - F0
(9)
(
)
kesc
However, in the most cases for the heterogeneous reaction
of COS on MgO at low temperature, the value of F(t)(1 + (kini/
k
esc)) is much greater than that of khNads. Therefore, the
desorption rate constants given in Table 1 were approximately
calculated from eq 8. The typical relationship between Fdes(t)
and Nads(t) is shown in Figure 4. And the desorption rate
constants at 228, 240, and 248 K were given in Table 1. The
sample mass in these experiments was fixed to be 100.0 mg,
Figure 6. The relationship between desorption rate constants and
temperature.

Temperature Dependence of COS-MgO Reaction
J. Phys. Chem. A, Vol. 112, No. 13, 2008 2825
of mineral particles is very high, and the atmospheric temper-
ature may reach 330 K. Therefore, the role of the catalytic
reactions of COS on mineral particles in its sinks will become
prominent. Our results also indicate that it should be cautious
when the uptake coefficients for trace gases on mineral particles
used in the model studies, especially obtained at one temperature
point, and the temperature factor for the uptake coefficients
should be considered.
Acknowledgment. This research was funded by National
Natural Science Foundation of China (40775081, 50621804)
andtheMinistryofScienceandTechnology,China(2007CB407301).
References and Notes
(1) Turco, R. P.; Whitten, R. C.; Toon, O. B.; Pollack, J. B.; Hamill,
P. Nature 1980, 283, 283-286.
Figure 7. The energy change in the reaction coordinate of COS on
MgO.
(2) Chin, M.; Davis, D. D. J. Geophys. Res. 1995, 100 (DS), 8993-
9005.
(3) Watts, S. F. Atmos. EnViron. 2000, 34, 761-799.
(4) Kjellstro¨m, E. J. Atmos. Chem. 1998, 29, 151-177.
(5) Crutzen, P. J. Geophys. Res. Lett. 1976, 3, 73-76.
(6) Jacob, D. J. Atmos. EnViron. 2000, 34, 2131-2159.
(7) Dentener, F. J.; Carmichael, G. R.; Zhang, Y.; Lelieveld, J.; Crutzen,
P. J. J. Geophys. Res. 1996, 101, 22869-22889.
(8) Usher, C. R.; Michel, A. E.; Stec, D.; Grassian, V. H. Atmos.
EnViron. 2003, 37, 5337-5347.
and the final products was calculated to be 44.62 kJ‚mol-1 using
Gaussian 98 (DFT-6-31G-B3LYP). As discussed above, the
reverse reaction is less likely than the positive reaction.
Therefore, the activation energy for the reaction between CO2
and H2S should be higher than that for the inversed process.
Therefore, the energy of CO2 and H2S was given below that of
the adsorbed species. Figure 7 shows that, when COS is
adsorbed on MgO, the heterogeneous reaction is more sensitive
to temperature than the desorption process, and the heteroge-
neous reaction of COS on MgO is in favor of a high temperature.
(9) Wu, H. B.; Wang, X.; Cheng, J. M.; Yu, H. K.; Xue, H. X.; Pan,
X. X.; Hou, F. Q. Chin. Sci. Bull. 2004, 49, 1231-1235.
(10) Wu, H. B.; Wang, X.; Cheng, J. M. Sci. China, Ser. B: Chem.
2005, 48, 31-37.
(11) He, H.; Liu, J. F.; Mu, Y. J.; Yu. Y. B.; Chen, M. X. EnViron. Sci.
Technol. 2005, 39, 9637-9642.
(12) Liu, J. F.; Yu, Y. B.; Mu, Y. J.; He, H. J. Phys. Chem. B 2006,
110, 3225-3230.
4. Conclusions and Atmospheric Implications
(13) Liu, Y. C.; He, H.; Xu, W. Q.; Yu, Y. B. J. Phys. Chem. A 2007,
111, 4333-4339.
In this work, the uptake of COS on MgO has been
investigated at a temperature scope from 228 to 300 K. The
initial adsorption rate constants increase with the decreasing of
temperature, while the desorption rate constants and the
heterogeneous reaction rate constants increase with the enhance-
ment of temperature. The empirical formula between the uptake
coefficients and desorption rate constants and temperature can
be described in the form of the Arrhenius equation. That is, ln
γt(Ini) ) -16.58 + 601.39/T; ln γt(SS) ) -11.21 - 1325.04/
T; ln kdes ) -1.10 - 758.50/T. The activation energy for the
heterogeneous reaction and desorption of COS on MgO were
measured to be 11.02 ( 0.34 kJ‚mol-1 and 6.30 ( 0.81
kJ‚mol-1, respectively. The relatively low activation energy
indicates that the heterogeneous reactions of COS on mineral
dust particles may have an important role in the sink of COS
in the atmosphere. The empirical formula between the uptake
coefficients and temperature can be used to predict the uptake
coefficient at any temperature in the atmosphere.
(14) Liu, Y. C.; Liu, J. F.; He, H.; Yu, Y. B.; Xue, L. Chin. Sci. Bull.
2007, 52, 2063-2071.
(15) Simth, I. W. M. Chem. ReV. 2003, 103, 4549-4564.
(16) Toon, O. B.; Hamill, P.; Turzo, P. P.; Pinto, J. Geophys. Res. Lett.
1986, 13, 1284-1287.
(17) Tolbert, M. A.; Rossi, M. J.; Golden, D. M. Science 1988, 240,
1018-1021.
(18) Finlayson-Pitt, B. J. Chem. ReV. 2003, 103, 4801-4822.
(19) Usher, C. R.; Michel, A. E.; Grassian, V. H. Chem. ReV. 2003,
103, 4883-4939.
(20) Ammann, M.; Kalberer, M.; Jost, D. T.; Tobler, L.; Ro¨ssler, E.;
Plguet, D.; Ga¨ggeler, H. W.; Baltensperger, U. Nature 1998, 395, 157-
160.
(21) Grassian, V. H. J. Phys. Chem. A 2002, 106, 860-877.
(22) Seisel, S.; Pashkova, A.; Lian, Y.; Zellner, R. Faraday Discuss.
2005, 130, 437-451.
(23) Tabazadeh, A.; Jacobon, M. Z.; Singh, H. B.; Toon, O. B.; Lin, J.
S.; Chatfield, R. B.; Thakur, A. N.; Talbot, R. W.; Dibb, J. E. Geophys.
Res. Lett. 1998, 25, 4185-4188.
(24) Zhang, Z. Q.; Friedlander, S. EnViron. Sci. Technol. 2000, 34,
4687-4694.
(25) Li, P.; Al-Abadleh, H. A.; Grassian, V. H. J. Phys. Chem. A 2002,
106, 1210-1219.
As pointed out in the Introduction section, the temperature
in the earth’s atmosphere varies with altitude above the earth’s
surface. At high troposphere, COS can adsorb on the mineral
dust particles quickly. The saturated adsorptive capacity for COS
(26) Underwood, G. M.; Li, P.; Usher, C. R.; Grassian, V. H. J. Phys.
Chem. A 2000, 104, 819-829.
(27) Beichert, P.; Finlayson-Pitts, B. J. J. Phys. Chem. 1996, 100,
15218-15228.
on MgO was measured to be 7.1 ( 1.2 × 1018 molecules‚g-1
.
(28) Tabor, K.; Gutzwiller, L.; Rossi, M. J. J. Phys. Chem. A 1994, 98,
6172-6186.
Mineral dust particles smaller than 10 µm can be transported
over thousands of kilometers and are found far away from their
sources.22,37 Therefore, the adsorbed COS can sink to the earth’s
surface or be transported to elsewhere with mineral dust
particles. With the assumption of the global flux of 1000-3000
Tg‚yr-1 for mineral dust in the troposphere, the fraction of MgO
in the mineral dust particles of 6.1%,19,38 and the saturated
adsorptive capacity for COS on MgO, the total sink flux of COS
due to adsorbing on mineral dust and then sink with the mineral
dust is estimated to be 0.04 - 0.13 Tg‚yr-1. As for low
troposphere, especially in the boundary layer, the concentration
(29) Ullerstam, M.; Johnson, M. S.; Vogt, R.; Ljungstro¨m, E. Atmos.
Chem. Phys. 2003, 3, 2043-2051.
(30) Al-Abadleh, H. A.; Grassian, V. H. J. Phys. Chem. A 2000, 104,
11926-11933.
(31) Frinak, E. K.; Wermeille, S. J.; Mashburn, C. D.; Tolbert, M. A.;
Pursell, C. J. J. Phys. Chem. A 2004, 108, 1560-1566.
(32) Gebel, M. E.; Finlayson-Pitts, B. J. J. Phys. Chem. A 2001, 105,
5178-5187.
(33) Underwood, G. M.; Li, P.; Al-Abadleh, H. A.; Grassian, V. H. J.
Phys. Chem. A 2001, 105, 6609-6620.
(34) Carlos-Cuellar, S.; Li, P.; Christensen, A. P.; Krueger, B. J.;
Burrichter, C.; Grassian, V. H. J. Phys. Chem. A 2003, 107, 4250-
4261.

2826 J. Phys. Chem. A, Vol. 112, No. 13, 2008
Liu et al.
(35) Ravishankara, A. R. Science 1997, 276, 1058-1065.
(36) Seisel, S.; Lian, Y.; Keil, T.; Trukhin, M. E.; Zellner, R. Phys.
Chem. Chem. Phys. 2004, 6, 1926-1932.
(37) Husar, R. B.; Tratt, D. M.; Schichtel, B. A.; Falke, S. R.; Li, F.;
Jaffe, D.; Gasso´, S.; Gill, T.; Laulainen, N. S.; Lu, F.; Reheis, M. C.; Chun,
Y.; Westphal, D.; Holben, B. N.; Gueymard, C.; McKendry, I.; Kuring,
N.; Feldman, G. C.; McClain, C.; Frouin, R. J.; Merrill, J.; Dubois, D.;
Vignola, F.; Murayama, T.; Nickovic, S.; Wilson, W. E.; Sassen, K.;
Sugimoto, N.; Malm, W. C. J. Geophys. Res. 2001, 106 (D16), 18317-
18330.
(38) Usher, C. R.; Al-Hosney, H.; Carlos-Cuellar, S.; Grassian, V. H.
J. Geophys. Res. 2002, 107, 4713-4721.