Rates and mechanism of carbonyl sulfide oxidation by peroxides in concentrated sulfuric acid

Article abstract of DOI:10.1021/jp002333o

We measured the rates of carbonyl sulfide (OCS) oxidation by hydrogen peroxide H₂O₂ (HP) and peroxymonosulfuric acid HOOSO₂OH (PSA) in 13.5-18.0 M (76-96 wt%) sulfuric acid (SA) between 290 and 306 K. The reaction is first

Full text of DOI:10.1021/jp002333o

Evaluation Only. Created with Aspose.PDF. Copyright 2002-2021 Aspose Pty Ltd.
10794
J. Phys. Chem. A 2000, 104, 10794-10796
Rates and Mechanism of Carbonyl Sulfide Oxidation by Peroxides in Concentrated Sulfuric
Acid
N. F. Dalleska, A. J. Colussi,* A. M. Hyldahl, and M. R. Hoffmann*
W. M. Keck Laboratories, California Institute of Technology, Pasadena, California 91125
ReceiVed: July 3, 2000; In Final Form: September 12, 2000
We measured the rates of carbonyl sulfide (OCS) oxidation by hydrogen peroxide H2O2 (HP) and
peroxymonosulfuric acid HOOSO2OH (PSA) in 13.5-18.0 M (76-96 wt %) sulfuric acid (SA) between 290
and 306 K. The reaction is first order in both [OCS] and [peroxide], having nearly identical second-order rate
constants k1 for HP or PSA as oxidants. k1 increases exponentially with Hammett acidity Ho in the range
investigated: log(k1/M-1 s-1) ) - (9.57 ( 0.41) - (0.80 ( 0.05) Ho, (Ho < 0), at 306 K. OCS is, however,
inert toward HP in concentrated perchloric acid at equivalent Ho values. We infer that OCS oxidation by HP
+
in SA proceeds by an acid-catalyzed process involving the intermediacy of PSA rather than H3O2 . k1 depends
on temperature according to log(k1/M-1 s-1) ) (6.64 ( 1.58) - (2606 ( 472)/T, in 18 M SA. The observed
kinetic behavior and parameters are typical of sulfide oxidations by PSA in very acidic media. Present data,
in conjunction with data on ambient HP and sulfate aerosol levels in the lower stratosphere, lead to reaction
lifetimes that are many orders of magnitude longer than the currently estimated OCS atmospheric residence
time of about 4 years.
Introduction
valves. In a typical experiment, 5 mL of sulfuric acid-water
mixtures were introduced into thoroughly cleaned bulbs, and
degassed in three freeze-pump-thaw cycles. The bulb reactors
were then filled with a gas mixture slightly above 1 atm total
pressure, with a typical mixing ratio of OCS:Ar:He ) 40:60:
700 Torr, and later immersed in a controlled-temperature water
bath. The reaction was initiated by injection of a given amount
of H2O2 solution of known concentration through the septum.
We verified that OCS was indefinitely stable in the dark or under
room fluorescent lighting in the absence of H2O2, in accord with
previous results on OCS hydrolysis,7,8 and with negligible
photochemical effects. Gas aliquots (20 µL) were sampled at
regular intervals and analyzed for OCS and CO2 as described
below. Oxidations with OXONE (2KHSO5‚KHSO4‚K2SO4)
were carried out by equilibrating OXONE with the sulfuric
acid-water mixture at the reaction temperature prior to the
introduction of the gas mixture into the reactor by means of a
100 mL gas syringe.
Headspace gas analysis was performed with a Hewlett-
Packard 5890 series II gas chromatograph equipped with a 60
m, 0.32 mm fused silica column coated with a 5 µm, 100%
dimethyl polysiloxane film (Rtx-1 Crossbond, Restek), and a
Hewlett-Packard 5972 MSD mass spectrometer detector tuned
to m/z ) 44 (CO2+) and 60 (OCS+) ions. The injector, column,
and detector interface were held constant at 110 °C. Gas samples
(20 µL) were split in the ratio 30:1 with a 2 mL/min He carrier
flow through the column. Argon was used as an internal
standard.
Carbonyl sulfide, the most abundant sulfur compound in the
atmosphere, has been long considered a potential source of
stratospheric sulfate aerosol.1,2 However, OCS is relatively inert
in the troposphere, where it is homogeneously distributed at a
constant tropospheric mixing ratio of 470 ( 30 pptv (parts per
1012 by volume). The photochemical decomposition of OCS
has its onset in the lower stratosphere under 300 < λ/nm <
388 radiation, but peaks above the ozone layer at about 25 km,
i.e., 5-10 km higher than the densest aerosol.3 Since strato-
spheric aerosol plays crucial roles in both ozone depletion and
in the global radiative balance,4,5 it is important to explore for
chemical pathways converting OCS into background sulfur
aerosol in the lower stratosphere.
One-dimensional photochemical models based on gas-phase
oxidations driven by O(3P) and OH radicals predict a strato-
spheric lifetime of about 10 years for OCS.1 This is considerably
longer than the global atmospheric lifetime of 4.0 years derived
from current estimates of OCS global fluxes and nonvolcanic,
background aerosol levels.6 Although there are several possible
causes for this discrepancy, such as an overestimation of
background levels, the existence of yet unidentified atmospheric
OCS sinks remains an open issue. In this paper, we consider
the oxidation of OCS by H2O2, a normal component of the lower
stratosphere, in highly acidic aerosol droplets as a possible
mechanism for the conversion of OCS into sulfate. We report
the results of experiments on OCS oxidation by H2O2 in
concentrated sulfuric acid solutions relevant to atmospheric
aerosol conditions. We find that H2O2, once activated by
conversion into peroxymonosulfuric acid HOOSO2OH (PSA),
is indeed able to oxidize OCS at appreciable rates.
Sulfuric acid 96% (GFS Chemicals; supplied in PTFE bottles
and certified as containing <0.00001% heavy metals and
<0.000002% iron levels) was used as received. The use of high-
purity sulfuric acid is intended to prevent possible catalysis by
trace metal contaminants. Sulfuric acid solutions in water (Milli-
Q, 18 MΩ cm-1) were prepared gravimetrically and their density
confirmed by pycnometry. Carbonyl sulfide (Aldrich, >97.5%)
was frozen at 77 K and degassed, prior to mixing it with argon
Experimental Section
The oxidation of OCS was studied in 50 mL round-bottom
Pyrex bulbs equipped with rubber septa and PTFE vacuum
10.1021/jp002333o CCC: $19.00 © 2000 American Chemical Society
Published on Web 10/21/2000

Evaluation Only. Created with Aspose.PDF. Copyright 2002-2021 Aspose Pty Ltd.
Carbonyl Sulfide Oxidation by Peroxides in H2SO4
J. Phys. Chem. A, Vol. 104, No. 46, 2000 10795
Figure 2. Arrhenius plot for k1 in 96 wt % SA.
Figure 1. Pseudo-first-order decay plot for OCS oxidation, reaction
1, of 51.7 mL of a OCS + Ar + He ) 40 + 60 + 700 Torr gas mixture
in 5.0 mL 80 wt % SA + 0.20 mL 7.5M H2O2 liquid phase at 306 K.
To probe whether sulfuric acid catalysis proceeds by proto-
nation of either OCS or H2O2 (but not of both) we investigated
the reaction in perchloric acid media of comparable acidity. In
these cases, OCS oxidation by H2O2 takes place at measurable
rates in 78 wt % sulfuric acid, which has a Hammett acidity Ho
) -7.03, while there is no reaction in 70 wt % HClO4 (Ho )
-7.72) after 72 h. This experiment clearly demonstrates that
the apparent acid catalysis is specific to sulfuric acid. In other
words, although H3O2+, with pKa ) -4.7 is certainly formed
at Ho < -5,10,11 we find evidence for the generation of a far
more reactive intermediate in HP/SA media.
It is well-known that acids generally catalyze hydrogen
peroxide oxidations via the formation of peracid interme-
diates.12-14 In this case, hydrogen peroxide seems to undergo
metathesis with sulfuric acid to produce peroxymonosulfuric
acid:
P0
is OCS initial partial pressure.
OCS
and helium (both Matheson, UHP grade). Hydrogen peroxide
(EM Science, 30%) was used as received. Its concentration was
verified by titration periodically. OXONE, 2KHSO5‚KHSO4‚K2-
SO4, (Aldrich) was used as received.
Results and Discussion
OCS is found to react with excess H2O2 in sulfuric acid media
greater than 78 wt % to produce CO2 in stoichiometric amounts.
We did not detect any gas-phase sulfur-containing products by
the analytical procedure described above. However, we found
traces of SO2 in the gases evolved halfway through the course
of a reaction run in the absence of He or Ar, and condensed in
a coldfinger. Hence, we propose the following reaction stoi-
chiometry:
H2O2 + HOSO2OH h HOOSO2OH + H2O
(3)
H2SO4
OCS + 4H2O2 + 2H+
8 CO2 + H2SO4 + 4H2O (1)
We actually confirmed that PSA is the reactive intermediate
by carrying out the oxidation of OCS using OXONE, a mixed
potassium peroxymonobisulfate/bisulfate of known stoichiom-
etry (see Experimental Section) in sulfuric acid. Second-order
rate constants, k1, based on the calculated [PSA] are identical
within experimental error to those measured for HP as oxidant.
The comparison between the rates measured for the two oxidants
must take into account the kinetics and equilibrium of reaction
3. We estimate that equilibrium between HP and PSA in SA
>13 M is established in less than a second15 and, therefore,
that step 3 can be treated as a rapid preequilibrium kinetic
process. However, only a fraction of the initial HP concentration,
[H2O2]o, is converted into reactive PSA. In other words, it is
necessary to multiply the second-order rate constants calculated
from eq 2 with [OXIDANT] ) [H2O2]o by the factor f )
[H2O2]o/[PSA]. This correction factor, evaluated from literature
data on K2 as a function of acidity,15 amounts to f ∼ 1.2 at
[SA] ) 13 M and monotonically approaches f ) 1.02 at [SA]
) 18 M.
Reaction 1 leads to the first-order decay of OCS (Figure 1),
with pseudo-first-order rate constants k1′ that are proportional
to [H2O2]. Second-order rate constants for reaction 1 were
calculated from k1′ values by assuming that dissolved OCS(s)
is in equilibrium with OCS(g), i.e., [OCS(s)] ) KH × POCS
,
where KH is Henry’s constant for OCS dissolution in the H2-
SO4-H2O media, and POCS is the partial pressure of OCS in
the reactor headspace. The overall rate of OCS(g) disappearance
is given by
dP
OCS ) -k1′POCS
)
dt
VLKHRT
-k1
[OXIDANT]POCS (2)
(
)
VG + VLKHRT
where typical values of VLand VG, the volumes of the gas and
liquid phases within the reactor, were 5 and 50 cm3, respectively.
Values for KH were taken from De Bruyn et al.9 who found
that RT ln(KH/55.5) ) 4195 - 29.6T for OCS in water and
sulfuric acid solutions up to 1 M. We made no attempt to adjust
KH for possible (and unknown) salting-out or salting-in effects
in more concentrated sulfuric acid solutions. Using the expres-
sion from De Bruyn, we adopted 0.0186, 0.024, and 0.0272 M
atm-1 as the value for KH at 306, 298, and 290 K, respectively.
k1 can be obtained from reaction 2 once the actual identity of
the oxidant is established.
The temperature dependence of k1 in 18 M SA is shown in
Figure 2. The Arrhenius parameters:
log(k1/M-1 s-1) ) (6.64 ( 1.58) - (2606 ( 472)/T (4)
are very similar to those measured for the oxidations of alkyl
aryl sulfides into the corresponding sulfoxides by PSA in sulfuric
acid media.16 These oxidations typically have low A-factors and
activation energies. The value of E2 ) 11.9 kcal mol-1 is

Evaluation Only. Created with Aspose.PDF. Copyright 2002-2021 Aspose Pty Ltd.
10796 J. Phys. Chem. A, Vol. 104, No. 46, 2000
Dalleska et al.
forward, estimated global OCS production flows exceeded
destruction rates by about a factor of 2. Since then, however,
new findings lead to a budget that is nearly balanced within
the limits of uncertainty.23 Further, Hamill et al. recently
concluded that the life cycle of the sulfate aerosol particles is
well understood. Aerosol droplets are largely formed during the
ascent of air masses across the tropical tropopause, suggesting
that the stratospheric H2SO4 has a tropospheric origin.5 Assess-
ing the OCS budget and its contribution to stratospheric sulfate
involves assumptions about the level of the background aerosol.
In this regard, it should be emphasized that the last five years
were the longest volcanically quiescent period for which
measurements of stratospheric sulfate aerosol are available. A
new, lower limit to the background aerosol levels would be
consistent with slower OCS oxidation rates.24 Further investiga-
tion and more accurate budgets and lifetimes of the Junge
aerosol layer will ultimately settle the question of whether a
“substantial, yet unidentified, systematic error” remains in the
chemical kinetic database for OCS.
Figure 3. The second-order rate constant k1 vs the sulfuric acid
Hammett acidity function -Ho at 306 K.
comparable to the activation energy determined for MeSC6H4-
NO2 oxidation by PSA in 9.9 wt % SA.13
We also found, in accord with previous work on PSA
oxidations, that k1 is a strongly increasing function of overall
acidity. For example, plots of log k1 vs -Ho are linear, with
specific slopes that are substrate dependent. For weak bases,
such as OCS, the observed slopes are generally smaller than
one. In fact, the dependence of k1 on -Ho at 306 K follows the
expected behavior, as shown in Figure 3, which corresponds to
the following parameters:
Acknowledgment. This work was supported by the National
Science Foundation under Grant ATM-9711923
References and Notes
(1) (a) Chin, M.; Davis, D. J. Geophys. Res. 1995, 100, 8993-9005.
(b) Turco, R. P.; Whitten, R. C.; Toon, O. B.; Pollack, J. B.; Hamill, P.
Nature 1980, 283, 283-286. (c) Griffith, D. W. T.; Jones, N. B.; Matthews,
W. A. J. Geophys. Res. 1998, 103, 8447-8454. (d) Rinsland, C. P.; Zander,
R.; Mahieu, E.; Demoulin, P.; Goldman, A.; Ehhalt, D. H.; Rudolph, J. J.
Geophys. Res. 1992, 97, 5995-6002.
(2) Kjellstrom, E. J. Atmos. Chem. 1998, 29, 151-177.
(3) (a) Hofmann, D. J.; Stone, R. S.; Wood, M. E.; Deshler, T.; Harris,
J. M. Geophys. Res. Lett. 1998, 25, 2433-2436. (b) Hofmann, D. J. Nature
1991, 349, 659.
log(k1/M-1 s-1) ) - (9.57 ( 0.41) - (0.80 ( 0.05) Ho (5)
-
The marked solvent effect on oxidations by HSO5 is
consistent with the development of positive charge on both sulfur
and the transferred oxygen atoms.17,18 We did not attempt to
separate protonation from solvent effects. Equation 5 provides
an empirical indication of the expected dependence of k1 on
sulfuric acid concentrations at all temperatures.
(4) Junge, C. E.; Manson, J. E. J. Geophys. Res. 1961, 66, 2163.
(5) Hamill, P.; Jensen, E.; Russell, P.; Bauman, J. Bull. Am. Meteorol.
Soc. 1997, 78, 1395-1410.
(6) Kuhn, U. A., C.; Wolf, A.; Meixner, F. X.; Andreae, M. O.;
Kesselmeier, J. Atmos. EnViron. 1999, 33, 995.
(7) Elliott, S.; Lu, E.; Rowland, F. S. EnViron. Sci. Technol. 1989, 23,
458-461.
To assess the potential role of reaction 1 in the atmospheric
oxidation of OCS we proceeded to estimate an upper limit to
its lifetime under typical conditions. In the stratosphere at noon,
40° N, 20 km altitude, [H2O2] ∼ 109 molecules cm-3, T ∼ 225
K. Local aerosol typically consists of 0.1 µm diameter droplets
(8) Adewuyi, Y. G.; Carmicheal, G. R. EnViron. Sci. Technol. 1987,
21, 170.
(9) De Bruyn, W. J.; Swartz, E.; Hu, J. H.; Shorter, J. A.; Davidovits,
P.; Worsnop, D. R.; Zahniser, M. S.; Kolb, C. E. J. Geophys. Res. 1995,
100, 7245-7251.
(10) Gmelin, Sauerstoff; Weinheim, 1966; Vol. 6; p 2260.
(11) Curci, R.; Edwards, J. O. In Organic Peroxides; Swern, D., Ed.;
Wiley-Interscience: New York; Vol. 1; pp 199 and ff.
(12) Betterton, E. A.; Hoffmann, M. R. J. Phys. Chem. 1988, 92, 5962.
(13) Betterton, E. A.; Hoffmann, M. R. EnViron. Sci. Technol. 1990,
24, 1819.
(14) Strukul, G. Catalytic Oxidation with Hydrogen Peroxide as Oxidant;
Kluwer: Dordrecht, The Netherlands, 1992.
(15) Monger, J. M.; Redlich, O. J. Phys. Chem. 1956, 60, 797.
(16) Bunton, C. A.; Foroudian, H. J.; Gillitt, N. D.; Kumar, A. Can. J.
Chem. 1999, 77, 895.
with an average density of 10 cm-3 19
which amounts to a
,
reactive ratio of Vaerosol/Vair ∼ 5 × 10-16. The latter is consistent
with independent reports of about 107 SA molecules cm-3
.
Assuming a Henry’s law constant for H2O2 in concentrated
SA: KH ∼ 3 × 1010 M/atm,9,20,21 we obtain [H2O2] ∼ 1 M
within the aerosol droplets. If we further assume that all of the
HP is quantitatively converted into PSA, i.e., that [PSA] ∼ 1M,
and that [SA] ∼ 18 M at the aerosol interface,22 we can calculate
the OCS reaction lifetime as
(17) Lowry, T. H.; Richardson, K. S. Mechanism and Theory in Organic
Chemistry, 3rd ed.; Harper and Row: New York, 1987.
(18) Liler, M. Reaction Mechanisms in Sulphuric Acid; Academic
Press: London, 1971.
τ-1 ) k1[PSA]Vaerosol/Vair
(6)
(19) Drdla, K.; Pueschel, R. F.; Strawa, A. W.; Cohen, R. C.; Hanisco,
T. F. J. Geophys. Res. 1999, 104, 26737.
(20) Lind, J. A.; Kok, G. L. J. Geophys. Res. 1986, 91, 7889.
(21) O’Sullivan, D. W. L., M.; Noone, B. C.; Heikes, B. C. J. Phys.
Chem. 1996, 100, 3241.
With k1 ) 1 × 10-5 M-1 s-1 in SA 18 M at 225 K (from eq 4),
we get τ ∼ 6 × 1012 years! Therefore, reaction 1 is utterly
irrelevant regarding the atmospheric fate of OCS.
Given the slowness of reaction 1, there appears to be no viable
oxidation pathway for OCS by known stratospheric oxidants
that are fast enough to meet the kinetic constraints imposed by
current considerations on “missing sinks”. In retrospect, we
observe that when the “missing sink” hypothesis was first put
(22) Longfellow, C. A. I., T.; Ravishankara, A. R.; Hanson, D. R. J.
Phys. Chem. A 1998, 102, 3323.
(23) Andreae, M. O.; Crutzen, P. J. Science 1997, 276, 1052-1058.
(24) “Session A22E “Stratospheric Aerosols Under Historically Low
Conditions” ”; American Geophysical Union Fall Meeting, 1999, San
Francisco, CA.