&EPA
United States
Environmental Protection
Agency
Environmental Sciences Research
Laboratory
Research Triangle Park NC 27711
EPA 600 7 79-088
April 1979
Research and Development
Surface Reactions of
Oxides of Sulfur
Interagency
Energy/Environment
R&D Program
Report
-------
RESEARCH REPORTING SERIES
Research reports of the Office of Research and Development, U.S. Environmental
Protection Agency, have been grouped into nine series. These nine broad cate-
gories were established to facilitate further development and application of en-
vironmental technology. Elimination of traditional grouping was consciously
planned to foster technology transfer and a maximum interface in related fields.
The nine series are:
1. Environmental Health Effects Research
2. Environmental Protection Technology
3. Ecological Research
4. Environmental Monitoring
5. Socioeconomic Environmental Studies
6. Scientific and Technical Assessment Reports (STAR)
7. Interagency Energy-Environment Research and Development
8. "Special" Reports
9. Miscellaneous Reports
This report has been assigned to the INTERAGENCY ENERGY-ENVIRONMENT
RESEARCH AND DEVELOPMENT series. Reports in this series result from the
effort funded under the 17-agency Federal Energy/Environment Research and
Development Program. These studies relate to EPA's mission to protect the public
health and welfare from adverse effects of pollutants associated with energy sys-
tems. The goal of the Program is to assure the rapid development of domestic
energy supplies in an environmentally-compatible manner by providing the nec-
essary environmental data and control technology. Investigations include analy-
ses of the transport of energy-related pollutants and their health and ecological
effects; assessments of, and development of, control technologies for energy
systems; and integrated assessments of a wide range of energy-related environ-
mental issues.
This document is available to the public through the National Technical Informa-
tion Service, Springfield, Virginia 22161.
-------
EPA-600/7-79-088
April 1979
SURFACE REACTIONS OF OXIDES OF SULFUR
by
J. H. -Lunsford
Department of Chemistry
Texas A&M University
College Station, Texas 77843
Grant No. R-801136
Project Officer
Jack L. Durham
Atmospheric Chemistry and Physics Division
Environmental Sciences Research Laboratory
Research Triangle Park, North Carolina 27711
ENVIRONMENTAL SCIENCES R£SEARCH LABORATORY
OFFICE OF RESEARCH AND DEVELOPMENT
U.S. ENVIRONMENTAL PROTECTION AGENCY
RESEARCH TRIANGLE PARK, NORTH CAROLINA 27711
-------
DISCLAIMER
This report has been reviewed by the Environmental Sciences Research
Laboratory, U.S. Environmental Protection Agency, and approved for publica-
tion. Approval does not signify that the contents necessarily reflect the
views and policies of the U.S. Environmental Protection Agency, nor does
mention of trade names or commercial products constitute endorsement or
recommendation for use.
n
-------
ABSTRACT
Surface reactions of several sulfur-containing molecules have been studied
in order to understand the mechanism by which sulfate ions are formed on atmos-
pheric aerosols. At 25°C the heterogeneous oxidation of S02 by N02 to sulfuric
acid and sulfate ions occurred on hydrated silica and silica-alumina surfaces.
Nitrosonium ions, probably in the form of NOHSOit, underwent surface hydrolysis
to form H2SOi* and NO. The oxidation of S02 by 02 was undetectably slow on
these surfaces; however, the reaction was catalyzed by Mn2+ ions in a Y-type
zeolite, which is a crystalline aluminosolicate. In this case the only reac-
tion product was a SOi,2" species which was strongly bound in the zeolite micro-
pores. Substantial activity was observed only on partially and fully hydrated
zeolites. Molecular oxygen also reacted with H2S on zeolites and magnesium
oxide, but essentially no reactions were observed on silica-alumina. The pri-
mary reaction on zeolites resulted in the formation of elemental sulfur; where-
as, on MgO a variety of products were observed including elemental sulfur, sul-
fide ions, sulfite ions and a number of paramagnetic ions such as $3, S20~ and
Sn. Sulfate ions were not formed from H2S following the reactions at 25°C.
Experiments were also carried out to identify and to determine the mecha-
nism by which S20~ and H2S2 ions are formed on MgO. The S20" ion was observed
following the surface reaction of H2S with S02, and formation of the ion was
promoted by ultraviolet light. A mechanism has been proposed in which S2 mole-
cules react with oxide ions of the lattice. The H2S2 ion was formed upon
reacting H2S with trapped electrons on the surface. Information on the elec-
tronic structure of both paramagnetic species has been obtained from sulfur-33
hyperfine data.
In order to establish the geometric structure of the S20" ion and other
paramagnetic sulfur-containing molecules on MgO, CNDO semi empirical calcula-
tions were carried out to determine theoretical values of g tensors and spins
densities as a function of bond angles and bond length. By comparing the cal-
culated and experimental values, one could establish the structure of the mole-
cule. This represents one of the first attempts to determine the structure of
adsorbed paramagnetic ions.
This report was submitted in fulfillment of Grant No. R-801136 by Texas
A&M University under the sponsorship of the U.S. Environmental Protection
Agency. This report covers the period December 15, 1973 to May 15, 1978 and
the work was completed as of November 15, 1978.
m
-------
CONTENTS
Abstract iii
Figures vi
Tables vii
Acknowledgment viii
1. Introduction 1
2. Experimental 3
3. Results 4
4. Discussion 27
5. List of Publications 30
References 31
-------
FIGURES
Number
1 S(2p) and N(ls) ESCA spectra produced by reacting S02 with N02. 5
2 Infrared spectra of S02 and N02 on hydrated Si02 7
3 ESCA spectra of sulfur and manganese 2p levels 9
4 Initial rate of SO^ formation versus S02 partial pressure. . . 13
5 XPS spectrum of sulfur 2p lines 17
6 EPR spectrum of S2(T on MgO 18
7 EPR spectrum of S20" containing sulfur-33 on MgO 20
8 EPR spectra of H2S2 on MgO. 22
9 EPR spectra of H233S2 on MgO 23
VI
-------
TABLES
Number
1
2
3
4
C
6
7
8
Binding EnergiGS (eV) of Adsorbed Species
Observed Frequencies of Adsorbed Species
Reactive Capacities of MnY Zeolite Cat?lysts. . .
n-\/alijp«; of thp Isoelectroni c SOo SoO" and S^ Ions ,
A Comparison of the Experimental and CNDO/2-Calculated
A Comparison of Spin Densities Determined from CNDO/2
Calculations and Experimental Hyperfine Data
Spin Density of the Isoelectronic SOI, $20" and $3 Ions on
MqO
Pacje
4
8
10
14
19
25
26
28
VI1
-------
ACKNOWLEDGMENT
The contributions of Dr. Jack L. Durham in keeping the author abrest of
current views on the oxidation of S02 in the atmosphere was a significant
factor in this work. The author also wishes to express appreciation to
Prof. R. M. Hedges for the double-precision version of the CNDO/2 program
and to Dr. T. T. Yu for assistance in programming.
vm
-------
SECTION 1
INTRODUCTION
Expanding utilization of coal and lignite as primary fuel sources has
prompted concern over the fate of sulfur dioxide in the atmosphere. This major
pollutant is produced during combustion at a rate in excess of seventy billion
pounds per year in the United States (1). In addition, sulfur, mainly in a
more reduced state such as found in hydrogen sulfide, is introduced into the
atmosphere during the decay of organic matter. Aerosols which contain adsorbed
oxides of sulfur have been implicated as causative agents for vegetation dam-
age, reduced visibility, and chronic pulmonary diseases (2,3). For lack of
conclusive evidence, the harmful form of sulfur is believed to be the sulfate
ion. Of particular importance in this respect are the poorly understood mecha-
nisms through which sulfur dioxide or hydrogen sulfide may be transformed into
surface sulfate ions. Potential oxidizing agents include molecular oxygen,
ozone, and nitrogen dioxide.
Catalysts for the oxidation of S02 have long been known. In 1871, Deacon
(4) demonstrated that a solution of CuSOit was capable of producing sulfuric
acid from a gaseous mixture of SOa and air. Since that time, solutions of many
other salts have been shown to display similar catalytic activity. More re-
cently, several research groups (4-6) have explored the oxidation of S02 in
water droplets containing dissolved metal salts. The objective of these
studies has been to more closely approximate atmospheric conditions by using
droplets of a size comparable to those occuring naturally in atmospheric fogs.
These studies of S02 oxidation have shown that the catalysts in the droplets
can display appreciable activity under ideal conditions. Furthermore, the
experiment repeatedly indicate that Mn2+ is one of the most active catalysts in
such systems at ambient temperatures.
No significant results are available regarding the reactivity of adsorbed
S02 with oxides of nitrogen. This is unfortunate from a practical viewpoint
because NO and N02 are also common constituents of stack gases, along with S02.
The related gas phase S02-N02 reactions have attracted continued interest
(7,8). The results generally imply that the photoxidation of S02 to S03 may
occur at high concentration levels; however, at atmospheric levels, which are
on the order of lO^-lO"2 ppm, quenching by molecular oxygen essentially pre-
cludes SOs production. Recent evidence (9,10) suggests that photolysis of S02-
N02-air mixtures yields NOHSOit aerosol which slowly hydrolyzes in air to form
sulfuric acid and NO.
It has been suggested that it is possible to obtain sulfate ions from H2S
under ambient conditions, prov/ided a catalyst (an aerosol) is employed. The
-------
oxidation of H?S by molecular oxygen has indeed been observed, but the major
product is elemental sulfur. Moreover, the catalyzed reaction usually requires
elevated temperatures and high humidity. Catalysts include activated carbon,
molecular sieves, alumina, bauxite and Al-Mg alloy (11-17).
The purpose of this study has been to explore the surface reactions of S02
with 02 (18) and N02 (19,20) as well as the reactions of H2S with 02 (21).
The solids included both acidic alumino-silicates and a basTC metal oxide. MgO.
Hydrated silica, which is generally considered to be a rather inert solid, was
also used during the study with N02. The oxidized forms of sulfur were de-
tected both spectroscopically (electron paramagnetic resonance, inf™re°.a™coc
X-ray photoelectron spectroscopy) and by wet chemical techniques. In all cases
temperatures near ambient were used, although the concentrations of S02, H2S
and N02 were much higher than those found under atmospheric conditions.
In addition to studying the reactions involving the oxidation of S02 or
H2S, research has also been carried out on the formation and structure of para-
magnetic sulfur-containing molecules. These molecules include H2S2, which is
formed by allowing H2S to react with trapped electrons on MgO, and S20 . wh ch
is formed by reacting H2S and S02 on the same metal oxide (22, 23). In addi-
l on to yielding S2C)-, the reaction between H2S and S02 produces elemental sul-
fur and water (the Claus reaction). The geometric structure of S20 and other
sulfur-containing radicals has been determined by comparing calculated magnetic
oarameters with those which have been observed experimentally (24).
-------
SECTION 2
EXPERIMENTAL
Details of the experiments may be found in refs. 18-23; however, a sum-
mary of the materials and techniques will be described here. Amorphous silica
gel and silica-alumina (9.3% alumina) were used as supplied from a commercial
source. The BET surface areas of the silica and silica-alumina, obtained from
nitrogen adsorption data, were 310 and 360 m2/g, respectively. The MnY zeo-
lites were obtained by exchanging a sodium form in aqueous solutions of MnCl2
which yielded Mn2+ concentrations between 0.4 and 23 ions per unit cell. High
surface area magnesium oxide was prepared by reacting reagent grade powder
with water at 100°C. The resulting Mg(OH)2 was decomposed under vacuum at
300°C and then 500°C giving MgO which had surface area ca_. 200 m2/g.
With the exception of isotopically enriched S02 and H2S, all of the gases
were obtained from commercial sources. Sulfur-33 enriched S02 was prepared by
allowing 3 mg of sulfur containing 25.5 or 59.2% 33S to react with an excess of
pure oxygen at 450°C for 2 hr. Oxygen-17 labeled S02 was prepared in a similar
manner by using 41.7% enriched 1702- Sulfur-33 enriched H2S was prepared by
allowing 5 mg of sulfur containing 25.5% 33S to react with an excess of pure
hydrogen at 300°C for 10 hr. The unreacted gases were removed by the freeze-
pump technique prior to adsorption.
The S02 uptake and reaction products were determined by gravimetrically
measuring the adsorption, by analyzing for the sulfate formed and by several
spectroscopic techniques. The sulfate formed in the zeolite was determined by
using a barium titration procedure, following the digestion of the zeolite in
HC1. The ESCA spectra were obtained on a photoelectron spectrometer using alu-
minum Ka x-radiation. From the binding energies of the emitted electrons the
oxidation state of the sulfur could be determined. Binding energies are re-
ferenced relative to the Au 4f7/2 level (84.0 eV) of gold which was vacuum
sputtered onto the samples. Transmission infrared spectra were obtained with
the samples in the form of thin self-supporting wafers. The paramagnetic ions
were identified by electron paramagnetic resonance (EPR) spectroscopy.
-------
SECTION 3
RESULTS
SURFACE REACTIONS OF S02 WITH N02
Using ESCA it was determined that S02 does not chemisorb on silica or
silica-alumina at 25°C, and neither does it react on the surface with 02 or
air. Markedly different behavior was observed when S02 and N02 were allowed to
react at 25°C on the hydrated surfaces. Curve a of Figure 1 was recorded after
admitting 25 Torr S02 to a silica sample for 4 hr, expanding the excess gas
phase to about 2 Torr, and introducing 25 Torr of N02 for 4 hr. An intense
S(2p) line demonstrates that a strongly bound sulfur oxoanion was formed in
the surface reaction. Likewise, curve b illustrates spectra collected from
a hydrated Si02-Al203 sample after admitting 8 Torr S02 for 20 min, expanding
the excess to less than 1 Torr, and subsequently adding 8 Torr N02 for 4 addi-
tional hr. Both samples were in the spectrometer about 12 hr at 60°C before
collecting the N(ls) spectrum, and desorption of the residual species was not
detected.
For comparison purposes, curve c depicts the S(2p) and N(ls) ESCA spectrum
for (NHit)2 SOt, . The 2p3/2 and 2pi/2 binding energies in each spectrum are
very close to 168.9 and 169.9 eV, respectively. Based upon the similarity in
binding energies following the surface reaction and those observed for (NH^
S0i», we conclude that sulfate ions are present on the surface in both cases.
The identification of this and other surface species is given in Table 1.
TABLE 1. BINDING ENERGIES (eV)a OF ADSORBED SPECIES
Adsorbate Orbital
background N(ls)
NO2 N(ls)
S02 + N02 S^2p3/2^
S(2p1/2)
S(2s)
N(ls)
a) ±0.2 eV
b) (Fwhm)
SiO2
-
407.2(1.8)b
400.6(1.9)
168.9(1.4)
169.9(1.4)
233.1(1.9)
401.2(1.8)
Si02-Al203
401.9(1.9)
407.3(1.8)
400.5(1.9)
168.8(1.4) ~J
169.9(1.4) >
233.0(1.9) J
401.3(1.8)
Assignment
+
NHit
N03~
phys. ads. N2Oit
SOi^
NO+ or -O-N=O
-------
174 170 166 162 408 404 400 396
BINDING ENERGY, eV
Figure 1. S(2p) and N(ls) ESCA spectra produced by reacting S02 with N02:
(a) Si02; (b) Si02-Al203-, (c) pure (NHJ2S(V
-------
Tt should be noted that the formation of N
-------
1400 2000 1800 1600 1400 1200
WAVENUMBER, cm"'
Figure 2. Infrared spectra of S02 and N02 on hydrated Si02: (a) background;
(b) 5 Torr S02 added 20 min with excess expanded to about 2 Torr; (c) 7
Torr NOa-Mu admitted 30 min [(--) 18 Torr]; (d) degassed at 25°C for
3 hr (6 hr); (e) 8 Torr H20 adsorbed.
-------
-IN a
TABLE 2. OBSERVED FREQUENCIES OF ADSORBED SPECIES (cm" )
AdsorbateSiCbSiO2-Al203 Assignment
H2o 1635 1635 U2 phys. ads. H20
S02 1337 1336 03 phys. ads. S02
N204 1735 1730 Ug-U(NO) phys. ads. N^
N02 1315(sh) 1320 u
H
H
690
1680 1672 Uit(B2);
^OiiiiN. phys. ads.
1420 1420(sh) 03
1545 VJ(AI) bidentate nitrato species
1590 UNQ antisym.
1650
N02 595 610
975 960
2460 2465 0 sym. H2SOif
U—M
2280 2270 \)\ NO
Tl
1400 1398 UN02>- Si-0-N-Si<
sh = shoulder
(a) ± cm'1
(b) only on Si02'Al203 degassed to 500°
SURFACE REACTIONS OF S02 WITH 02
The ESCA spectra of a representative MnY sample, both before and after
exposure to the S02-02 mixture, are shown in Figure 3. The spectrum of pure
Mnsoti ncluded for comparison purposes. From the observed binding energies
at 168 8 and 169.9 eV it is evident that sulfate ions were formed in the zeo-
lite during reaction. It also may be noted from the spectra that no other
strongly adsorbed sulfur species were present on the catalyst surface either
before or after reaction.
Adsorption experiments were undertaken to determine the reactive capacties
of the various zeolites. The zeolite samples were allowed 24 hr Jo react and
equilibrate with the reactant gases which contained 40 Torr 02 and 10 Torr S02.
To determine quantitatively the fraction of the total irreversible adsorption
that was due to the formation of sulfate ions, several samples of MnY used in
the adsorption experiments were analyzed for SO,2' via the barium titration
-------
-H 01
O O
-s
00 r+
O
to IQ
Oi
ai in
a- 3
-P» X
O d-
o fo
-s -•
-s
c
fD fD
CD
(£) CO
fD •
3 00
3" O
00 3>
O
-r 10
T3
-s fD
CO O
o o n> -s
N> --- -S OJ
n>
S = —
3 O O
-< n> o
<•• 3
N ft
(D --^-5
O CD OJ
n> 3 Q.
-h N n
ct fD O>
(DO—1
-S — ' fD
fD rt-*"^
X tt> O
-O -h
O O-
(/) fD
C -h C
-S O — •
fD -S -h
fD c
r+ -S
O fD
X cu
-S "O 3
fD O Q.
O) V)
O C 3
r+ -5 O)
o* ro 3
3 10
<-h c+ O)
O 3
3 fD
-". OO (/>
x o n>
c i ro
-5 OTD
ro ro
O -S fD
-h fD <
i n>
0)
z
UJ
H
Z
B
660
650
640 170
BINDING ENERGY (eV)
165
-------
technique. It was found in all cases that the irreversibly adsorbed material
observed gravimetrically was completely recoverable as SO^2" using the titra-
tion; therefore, the results of the adsorption experiments are reported in
Table 3 where all irreversible adsorption has been considered to be due only
to the formation of the sulfate ion.
TABLE 3. REACTIVE CAPACITIES OF MnY ZEOLITE CATALYSTS
Manganese concentration
(Mn (II) /Unit Cell)
0. 0
0.4
1.7
9.0
16
0.0
0.4
1.7
9.0
16
0. 0
0.4
1.7
9.0
16
Pretreatment Irreversible
procedure _
(no. S0~ /Unit Cell)
Fill Iv nphvHt-aforl TPAPF
1. \lMMjf LyCl 1 y UJ. u L.CU H\f\\^Ci
1.9
2.1
1.6
2.2
Partially Hydrated 0
9.9
12.1
19.0
21.2
Fnl Iv HuHr Ai-oH TPnfP
-tvJXijrri^'«^H.clUt;*«l J. XUAV^ij
45.2
50.5
46.7
49.1
adsorption
(no. SO" /Mn(II))
4.7
1.2
0.2
0.2
0
24
7.0
2.1
1.4
113
29
5.1
3.2
10
-------
From these results one may conclude that both the Mn2+ ion and Oz must be
present for the oxidation of S02 to occur. The formation of S(V~ was not ob-
served on any of the zeolite samples in the absence of 02 while only trace
amounts of S042- could be detected on NaY in the absence of Mn^ . These trace
amounts of SOi*2' could be explained by the presence of iron impurities in the
original zeolite.
The reactive capacity of the zeolites was found to be strongly dependent
'
on the degree of dehydration of the MnY catalysts. The amount of S(X ' formed
per unit cell in the fully hydrated samples was more than an order of magnitude
greater than the amount formed in the correspond ihg fully hydrated samples.
For any given sample pretreatment, the number of S0i*2~ ions formed per unit
cell was nearly constant. In contrast, the number of SOi,2' ions produced per
Mn2+ ion was highly variable and reached its maximum value of 113 for the fully
hydrated MnY sample containing 0.4 Mn2+ ions per unit cell. This large turn-
over simply demonstrates that the reaction was catalytic with respect to manga-
nese.
The above observations show that water in the zeolite strongly affects the
reaction. It is possible that water functions as a reactant according to the
following scheme:
S02(g) -> S02(adsorbed)
S02(adsorbed) + ^2 -> S03 (2)
S0 + H20 -> H2SQk
This production of sulfuric acid would be expected to make the catalyst more
acidic. Since the catalytic oxidation of S02 by some solutions of Mn(II) salts
is known to have pH limitations (27), it is possible that the reaction within
the zeolite micropores is likewise pH limited. As an alternative, SOa may
react with lattice oxide ions
S03 + 02-(lattice) -> S042-(adsorbed) (4)
and water may simply provide a transport mechanism for the migration of SOi* "
ions away from the active sites within the zeolite.
The crystallinity of the zeolite during reaction comes under question
since the above mechanisms involve the participation of lattice oxide ions and
the production of sulfuric acid. The x-ray powder diffraction patterns of the
catalysts indeed were modified after the reaction occurred. There was a de-
crease in the intensity of some of the reflections while the background level
remained reasonably constant. It is not known at this time whether (a) the
crystal lattice was partially disrupted as a result of the reaction or (b) tne
changes in the pattern were simply a result of the added electron density in
the interior of the catalyst due to the production of large amounts of sulfate
product. This matter is presently under investigation.
Initial rate experiments were conducted using partially hydrated MnY zeo-
lite to determine the reaction rate dependence, if any, on the concentrations
11
-------
of S02, 02 and Mn2+. Initial rate studies were not conducted on the fully
hydrated and fully dehydrated MnY sample because in the former case the rate
of adsorption of the reactant gases was extremely slow due to occlusion of
the micropores by water molecules, and in the latter case only small amounts
of S042~ ions could be formed, which led to a relatively large experimental
error.
The results of the rate study are presented in Table 4 and in Figure 4.
There is an apparant first order rate dependence which is best described in
Figure 4 by the linear rate dependence on the pressure of S02. Table 4A de-
monstrates that the reaction was independent ot total pressure as was found by
increasing the total pressure in the system with helium. Furthermore, Tables
4B and 40 show that the initial rates of reaction were independent of 02 pres-
sure and temperature respectively.
It is recognized that the observed kinetics in such heterogeneous reac-
tions on porous catalysts may not be representative of the true reaction kine-
tics if there are diffusion processes in action which can influence the rate
of reaction. The fact that the initial rate did not significantly increase as
the temperature was raised from 5°C to 60 C is an indication that diffusion in-
deed may be limiting the overall reaction. Further evidence that the reaction
is diffusion limited may be found in the fact that the reaction rate increased
significantly upon addition of small amounts of Mn2+ to the zeolite, but addi-
tional exchange of Mn + resulted in only modest rate enhancement as shown in Table 4C,
Of the three various types of diffusional limitations, only Knudsen dif-
fusion fits the experimental observations. The Knudsen diffusion coefficient
is independent of total pressure, P, while the rates of bulk diffusion and
diffusion across a static film are inversely proportional to /P". That Knudsen
diffusion is operative in this system is not surprising since the conditions
for Knudsen flow is that X, the mean free path of a gas molecule be much larger
than the micropore radius. The value of X is given approximately by
which is on the order of ID"4 cm in this system. Obviously, this is consider-
ably larger than the micropore radius of 8 x 10~8 cm in the Y zeolite.
Under conditions of Knudsen diffusion the pseudo-order of the initial
rate of reaction with respect to SOa pressure may be found from the equation:
Observed Order = (n + l)/2 (6)
where n is the true reaction order. Thus, it is postulated that the oxidation
of S02 on MnY catalysts would be a first order process in the absence of the
Knudsen diffusion limitation. This result is reassuring since it agrees with
the findings of many other authors, most notably Cheng e_t al_. (6), who have
studied other heterogeneous catalysts containing maganese.
12
-------
<£>
C
Oi
Ol
n>
o
o
-r ti
cu
ci-
O>
to
o
to
TJ
O)
n-
o>
-a
3
5 10 15
SOa. PARTIAL PRESSURE (torr)
20
-------
TABLE 4. INITIAL RATES OF REACT J.UN .
A. Initial rate as functions
of SC>2 pressure and total
Manganese concentration Reaction temperature
(Mn(II)/Unit Cell) <°c>
20.2
20.2
20.2
20.2
20.2
20.2
20.2
B. Initial rate as function
25
25
25
25
25
25
25
of O2 pressure
Manganese concentration Reaction temperature
(Mn(II)/Unit Cell) (°c)
20.2
20.2
20. 2
20.2
20.2
25
25
25
25
25
Pso2
(Torr)
2.5
5.0
8.0
13
17
17
17
Pso2
(Torr)
10
10
10
10
10
pressure
Po2
(Torr)
20
20
20
20
20
20
20
P°2
(Torr)
40
25
20
10
5
PHe
(Torr)
27.5
25
22
17
13
63
163
PHe
(Torr)
0
15
20
30
35
Initial rate
(mg 504 g catalyst"1 min )
19.3
31.1
37.5
71.6
84.2
83.6
85.2
Initial rate
(mg SO^ g catalyst"1 min"1)
97
103
116
101
96
(rv-mhirmpdl
-------
TABLE 4 (cont.)
C. Initial rate as function of manganese concentration
Manganese concentration Reaction temperature Pso->
/ O (~* \ ^
(Torr)
(Mn (II) /Unit Cell)
(Torr)
Initial
(mg SO^ g catalyst" rain" )
0
0.4
1.7
9
16
23
D. Initial rate as function
Manganese concentration
(Mn (II) /Unit Cell)
16
16
16
16
16
16
16
25
25
25
25
25
25
of temperature
Reaction temperature
60
50
40
30
20
10
5
7.5
7.5
7.5
7.5
7.5
7.5
Pso2
(Torr)
10
10
10
10
10
10
10
24
24
24
24
24
24
(Torr)
40
40
40
40
40
40
40
0
26.2
29.0
35.8
39.1
40.9
Initial rate
(mg SC>4 g catalyst" min"1)
56
55
59
47
47
59
54
-------
SURFACE REACTIONS OF H2S WITH 02
The ESCA spectrum of a MnY zeolite after the introduction of H2S and 02
showed only one sulfur species which is assigned to elemental sulfur by its
lpv?and 2pi/? binding energies of 162.8 and 164.0, respectively. A spectrum
of sulfur howlver of much lower intensity, was obtained without the exposure
of the sample to oxygen. The elemental sulfur remained on the surface after
it had been exposed to the atmosphere for 2 weeks. On NaY a "miTar spectrum
was observed after the reaction, although the concentration of elemental sul-
fur was less than half as great as with the MnY zeolite. The dissociation of
H?S over zeolites to form elemental sulfur has been previously studied, and
the results reported here further demonstrate that no higher oxidation states
of sulfur are formed. On amorphous silica-alumina virtually no sulfur lines
were present in the ESCA spectrum.
The ESCA spectrum of the dehydrated MgO, after it had been treated with
H2S, followed by a 3 min exposure to air during the transfer of the sample to
the spectrometer, indicated the formation of two sulfur species; elemental
sulfur and sulfide ions (Figure 5A). Brief exposure to the atmosphere
showed no evidence of further oxidation; however, when the H2S-treated MgO
was purposely exposed to pure oxygen, the ESCA spectrum indicated an additional
peak (Figure 5b). Deconvolution of the spectrum suggested the sulfur 2p3/2
maxima at 167.0, 162.9 and 160.8 eV, which are in good agreement with the
binding energies of sulfites. elemental sulfur and sulfides, respectively. The
formation of sulfite ions was favored by low coverage of H2S and by long expo-
sure of the sample in air, which is evident in Figure 5c. This sample of MgO
was first exposed to H2S and 02, and then to the atmosphere for 2 weeks. On a
hydrated MgO/Mg(OH)2 surface similar results were obtained; however, the sulfur
bands were much less intense. The fractional monolayer coverage of the sulfite
ions formed on MgO varied from 0.1 to 0.8, depending on the equilibrium pres-
sure of the H2S and the duration of exposure in air.
Upon freezing the H2S and 02 above the dehydrated MgO which was at -196°C
a small amount of SO3 was evident in the EPR spectrum with g = 2.0034. Upon
warming to 25°C for 5 min the catalyst turned yellow and the spectrum of SOs
disappeared. A much greater concentration of S20~, which is characterized by
9! = 2.030, g2 = 2.010 and g3 = 2.001 was observed. Moreover, the EPR spectrum
recorded at -196°C indicated the formation of the $3 ion, which is charac-
terized by gi = 2.043 and gii= 2.004. A relatively small peak at g = 2.052
was observed at both 25°C and -196°C, in addition to the spectrum of S20~ and
$3, following the oxidation of H2S by 02. This species dominated the spectrum
at higher coverages of H2S, and the g values (gi = 2.052, g2 = 2.033, g3 =
2.002) correspond well with those assigned to Sn. No EPR signals were detected
for sulfur species on zeolites or on silica-alumina.
ELECTRON PARAMAGNETIC RESONANCE EVIDENCE FOR THE FORMATION OF S20~ ON MgO
Hydrogen sulfide and sulfur dioxide react on many types of surfaces, form-
ing elemental sulfur and water (the Claus reaction). In addition, on MgO a
paramagnetic ion identified as S20~ was also formed. This ion exhibited the
spectrum shown in Figure 6 with the indicated g values. Four hyperfine lines
with a splitting of 47.6G, centered on gxx, were observed when S02 enriched
to 25.5% 33S was used (Figure 7a). When H2S enriched to 25.5% 33S was
16
-------
(0
K
Z
3
o
u
175 167 159
BINDING ENERGY
EV
151
Figure 5. XPS spectrum of sulfur 2p lines, formed on MgO after the adsorp-
tion of (a) H2S, (b) H2S and 02 and (c) H2S and 02, in atmosphere for 2
weeks, at 25°C.
17
-------
g«=2.oio
9 =2.030
9xx=2.001
Figure 6. EPR spectrum of S20 formed by allowing H2S and S02 to react
on magnesium oxide at room temperature, followed by irradiation at 254 nm:
(a) X-band, (b) Q-band.
18
-------
employed, another set of four hyperfine lines with a splitting of 32.6G, also
centered on gxx, was detected as shown in Figure 7b. Both sets of hyperfine
lines of approximately equal intensity were observed as the result of the re-
action of H2S with enriched S02 which contained 59.2% 33S. Hyperfine struc-
ture resulting from the adsorption of S02 enriched to 41.7% in 170 was not ob-
served. No change in the EPR spectrum was detected when D2S was used instead
of H2S.
From the 33S (1=3/2) hyperfine splitting observed it is evident that two
sets of four lines result from two nonequivalent 33S atoms. Although some iso-
tope mixing occurred during the formation of tno paramagnetic ion, the results
suggest that the hyperfine splitting of 48.66 was mainly due to the 33S which
originated from S02; whereas, the splitting of 32.6 was mainly due to the 33S
which originated from H2S. The mixing was more apparent in the spectrum of
the more highly enriched (59.2% 33S) S02.
Morton (25) originally based the assignment of the S20~ ion on a compari-
son of trends in principal g values for the ions S02, S20~, and 83 where the
effect of the larger spin orbit coupling becomes increasingly greater with
more sulfur atoms in the molecule. Since oribitals available for mixing with
the ground state are rather far removed in energy, the effect of local environ-
ment is small, and the spectra are reasonably independent of the particular
matrix or surface. This is evident, for example, when one compares the g
values of S02 on $3 on surfaces and irradiated crystals, as shown in Table 5.
TABLE 5. g-VALUES OF THE ISOELECTRONIC SO", S2(T AND 85
SO^/MgO (A)
(B)
S02 /Zeolites
SQa'/KCl
S2OVMgO
S^'/Sodalite
S20-/Na2S203-5H20
SsYMgO
Sl/Sodalite
Sa/KCl
9i
2.0097
2.0078
2.002
2.0100
2.030
2.029
2.0287
2.043
2.046
2.0499
92
2.0052
2.0033
2.002
2.0071
2.010
2.011
2.0106
2.043
2.036
2.0319
93
2.0028
2.0014
2.009
2.0025
2.001
2.001
2.0035
2.004
2.005
2.0026
ref .
26
26
27
28
this work
29
25
30
29
28
The absence of any oxygen hyperfine splitting is somewhat difficult to
understand since 170 has a rather large nuclear magnetic moment, and even a
spin density as low as 0.05 would result in a separation of approximately 25G
between terminal hyperfine lines. It seems more reasonable that the oxygen in
19
-------
»OG -|
Figure 7 EPR spectrum of S20' formed by allowing (a) H2S to react wit
25.5% 33S-enriched S02, (b) 25.5% 33S.enriched H2S to react with S02 on MgO
20
-------
$20~ is derived from the oxide ions of the lattice, rather than from the S02.
Furthermore,justification for this model comes from the observation that S20"
could be formed by irradiation of CSz or COS adsorbed on MgO. We suggest that
sulfur molecules (S2, S3,...Sn) are deposited on the surface of MgO by the
reaction of H2S with S02 and by irradiation of adsorbed COS or CS2. The sub-
sequent reactions
S2 + 02-(MgO) + S20- + e- (7)
and nv
S3 + e- £v S-3 (8)
yield the observed paramagnetic ions. It is likely that S2 was also formed,
but the EPR spectrum was not observed at -196°C because of a short spin relaxa-
tion time. The partial pressures of H2S and S02 over the MgO may affect the
distribution of molecular weights of the sulfur species, and thereby alter the
concentration of S20". The role of oxide ions is further substantiated by the
absence of S20~ on MgO-Mg(OH)2 where the surface is covered with hydroxide
ions.
ELECTRON PARAMAGNETIC RESONANCE EVIDENCE FOR THE FORMATION OF H2S2 ON MgO
Trapped electrons (S centers) may be produced by irradiating MgO in the
presence of hydrogen with a 254-nm ultraviolet lamp. Upon warming a sample
tube containing frozen H2S and MgO with S centers to -78°C the EPR spectrum
shown in Figure 8a was observed. This spectrum can best be understood in
terms of two forms of H2S2. One form (A) is characterized by g^ (A)=2.015
and gn (A)=2.003 with two nonequivalent protons having aj_H(A)= an H(A)=9.4G
and a H' (A) = an H'(A) = 6.6G. The other form (B) is isotopic and is charac-
terizid by giso(B) = 2-009 with AisoH(B) = 9-4G and aisoH'(B) = 6.6G. The
simulated spectrum using these parameters and a concentration ratio of A/B =
5.7 is given in Figure 8b.
The relative intensities of A and B may be altered by changing the mode
of formation. When the sample was irradiated at -196°C following the adsorption
of H2S on S centers, the color of the sample became yellow and species A domi-
nated the spectrum. The hydrogen hyperfine lines in the parallel direction be-
came more evident, although some of the isotropic spectrum was still observed.
Upon warming the sample to 25°C the color changed to white and no paramagnetic
species was observed.
When H2S enriched to 25.5% in sulfur -33 was adsorbed on MgO containing
S centers, the EPR spectrum revealed additional hyperfine lines as depicted
in Figure 9A. Upon irradiation at 254 nm for 10 min the relative intensity
of species B was greatly reduced; whereas, two sets of quartet lines having
splitting constants 61 and 41 G, centered on g,, of species A became apparent.
These additional hyperfine lines were much more obvious when H2S enriched to
59.2% in sulfur - 33 was used. Using the statistical isotope distributions
for 25.5% 33S the spectrum of Figure 9b was simulated for the g values and hy-
drogen hyperfine splitting indicated previously, and with |aj.s(A)| = jar5
(A)|= 26, |a,,s(A)| = 61G, |a,,sY(A)| = 41G, |aiso s(B)| =136, |aiso s' (Bj =96,
and a concentration ratio of A/B = 5.7.
21
-------
9=2.009
Figure 8. EPR spectra of H2S2" obtained by allowing H2S to react with trapped
electrons on MgO at -78°C: (a) experimental spectrum; (b) simulated spectrum
using a mixing ratio A/B =5.7.
22
-------
111 II '
Figure 9. EPR spectra of H2 S2"(B) obtained by allowing 25.5% 33S-enriched
H2S to react with trapped electrons on MgO at -78°C; (a) experimental spec-
trum; (b) simulated spectrum.
23
-------
The g values obtained for species A are in reasonably good agreement
with those reported by Bennett £t al_. (31) for the spectrum assigned to the
H2S" radical. We believe, however, that the hyperfine structure is derived
from two nonequivalent hydrogen atoms and two nonequivalent sulfur atoms for
species A, and perhaps species B. The situation is not so clear with respect
to species B since there is overlapping of hyperfine lines in the EPR spec-
trum. Better agreement between the experimental and simulated spectra was
obtained using the model H2S2, rather than H2S". It is proposed that the
isotropic spectrum is the result of ion motion on the surface in three dimen-
sions.
The mode of formation of the radicals via the transfer of electrons from
the surface to the adsorbed H2S confirms that the observed species are anions.
When H2S is adsorbed on the S centers of MgO, the H2S~ ion may be formed ini-
tially. This ion may then react with a neutral H2S molecule, forming H2S~2
and H2. Subsequent irradiation undoubtedly leads to the photolysis of HaS
and other reactions which may result in additional H2S2 and elemental sulfur.
The nonequivalency of the two sulfur and two hydrogen atoms may be due to the
effect of the surface environment. The reason for the high mobility ex-
perienced by part of the H2S2 molecules is not understood at this time.
CALCULATIONS OF THE g TENSORS AND SPIN DENSITIES OF SMALL PARAMAGNETIC SULFUR-
CONTAINING ANIONS ON MAGNESIUM OXIDE
Using a CNDO semiempirical calculation principal values of g tensors and
spin densities have been determined for the sulfur-containing molecules SQ~2,
SSO~, S"3, SCO", and CS~2, as well as CQ~2. A deviation A is defined as the
average of the absolute deviation between the principal values of the experi-
mental and the calculated g tensors: A = I(|AX«| + |Ayi| + |AZI|). In the
course of minimizing A, we started with the bond lengtn suggested for the
neutral molecule, and the bond angles were varied between 90 and 150°. By
plotting A versus bond angle a curve with a minimum was generally obtained.
By varying both the bond length and bond angle within reasonable values a true
minimum, which corresponds to the most likely molecular structure of the ion
on the surface of MgO was found. These calculate g tensors that correspond
to the minimum A values are tabulated in Table 6. The calculated g values
were often within experimental error.
The calculations also yielded spin densities which are compared with the
experimental values in Table 7. The spin densities of the sulfur and oxygen
atoms are generally consistent with the experimental finding; however, the
spin'densities in the p orbital of the central sulfur atoms are usually lower
by 7 to 16% than the experimental values. The difference is reasonable since
the estimates based on hyperfine data were obtained by neglecting the d-
orbital contribution. Nevertheless, the theoretical calculations indicate
that the contribution of sulfur d orhitals is only 3 to 11%.
24
-------
TABLE 6. A COMPARISON OF THE EXPERIMENTAL AND CNDO/2-CALCULATED g-TENSORS
MOLECULE
S02 (A)
S02(B)
sso~
S3
CO2
SCO
CS2
BOND
LENGTH
S-0 1.560
S-0 1.500
S-0 1.560
S-S 2.100
S-S 2.100
C-0 1.060
C-0 1.159
C-S 1.653
C-S 1.850
BOND
ANGLE
o
120
110
120
110
121
113
130
EXPERIMENTAL
*g
2.
2.
2.
2.
2.
1 .
1.
iso
0059
0042
014
030
0007
9977
9902
gxx
2.0097
2.0078
2.030
2.043
2.0032
2.0049
1.9999
9yy
2.0052
2.0033
2.010
2.043
1.9975
1.986
1.963
gzz
2.0028
2.0014
2.003
2.004
2.0014
2.0022
2.0078
CNDO/2
giso
2.0058
2.0043
2.0151
2.0304
2.0010
1.9973
1.9885
gxx
2.0097
2.0075
2.0287
2.0452
2.0032
2.0049
1.9991
9yy
2.0057
2.0033
2.0144
2.0439
1.9975
1.9858
1.9653
gzz
2.0021
2.0021
2.0021
2.0022
2.0023
2.0012
2.0012
PO
en
isotropic
-------
ro
TABLE 7. A COMPARISON OF SPIN DENSITIES DETERMINED FROM CNDO/2 CALCULATIONS
AND EXPERIMENTAL HYPERFINE DATA
\™
C s
C p
°Pzl
° Pz2
S° sd
S° pd
z
S Pzl
5 Pz2
d-orbital
contribution
SO-(A) S02(B) SSO Sg C02 SCO
ExptlaCNDO Exptl CNDO Exptl CNDO Exptl CNDO Exptl CNDO Exptl CNDO
0.18 0.18 NAb 0.11
_ 0.45 0.34 0.20
0.14 0.15 0.12 0.12 0.07C 0.08 - 0.20 0.24 0.19
0.14 0.150.12 0.12 - - 0.20 0.24 (OPy0.05)
0.01 0.00 0.01 0.00 0.02 0.00 0.01 0.00 - 0.00
0.75 0.64 0.71 0.64 0.54 0.45 0.53 0.37
0.37 0.43 0.27 0.30 - 0.26
_ 0.27 0.30 - (S p 0.16)
0.07 0.11 0.05 0.03 0.00 0.03
CS2
Exptl CNDO
NAb 0 . 08
0.34
0.00
0.24
0.24
0.08
Calculated from experimental hyperfine data
3Not available
=Spin density estimated by difference from unity
Central sulfur atom
-------
SECTION 4
DISCUSSION
The results presented here confirm that a variety of reactions involving
sulfur-containing molecules can occur on solid surfaces. Nitrogen dioxide is
an active molecule for the oxidation of sulfur dioxide to adsorbed sulfuric
acid on hydrated silica and silica-alumina at 25°C. Although direct proof
for NOHSOU as an intermediate is lacking, the experimental evidence strongly
favors this interpretation. Nitrosonium bisulfate is an important inter-
mediate in the gas phase oxidation of S02 to H2SOit with N02 and water at high
pressures and temperatures; e.g., the lead chamber process incorporates this
reaction. In the chamber process the reaction,
2NOHS04 + 2H20 + S02 •*• 3H2S04 + 2ND (9)
is postulated, and a similar reaction apparently takes place on the surface
of silica and silica-alumina. Similar reactions may occur among air pollu-
tants on the surfaces of particulates suspended in the atmosphere. It re-
mains to be demonstrated, however, that the rate of reaction will be signifi-
cant at the lower concentrations of N02 and S02 that are found under most
atmospheric conditions.
In contrast to nitrogen dioxide, it is not possible to oxidize sulfur
dioxide using molecular oxygen on the same silica or silica-alumina surfaces
at ambient temperatures. The latter reaction did occur on a crystalline alu-
minosilicate, known as a zeolite, which contained manganese ions. The reac-
tion was catalytic in that up to 113 sulfate ions were formed per manganese
ion. The maximum amount of sulfate ion was about 48 per unit cell or 7 per
large cavity. This maximum is probably not due to geometric constraints;
more likely, it results from the availability of lattice oxygen ions for sul-
fate formation, assuming that reaction 4 is correct. Since the reaction was
diffusion controlled, the inherent rate of reaction is much greater than was
observed in these experiments. Although the concentration of exchange sites
is more limited, in principle similar reactions should occur on amorphous
silica-alumina containing manganese, or perhaps other transition metal ions
as impurities.
The surface oxidation of hydrogen sulfide also occurs at moderate temper-
atures, but the products are elemental sulfur and sulfite ions on magnesium
oxide In addition, small concentractions of the paramagnetic ions SB, S20~,
S02, SOs, and S~ were also observed on the surface of MgO during the various
stages of the oxidation reaction. The oxidation of H2S by 02 at 25°C was
different on the three materials: MgO, zeolites and amorphous silica-alumina.
27
-------
The oxidation was carried out to the greatest extent (sulfite ions) on the
basic MgO surface; whereas, it proceeded to form elemental sulfur on zeolites,
and was negligibly slow on the acidic silica-alumina surface. The Mn2 ions
in the zeolite enhanced the formation of elemental sulfur. There was no evi-
dence for the formation of sulfate ions on any of these surfaces; thus, it
appears unlikely that natural sources of H2S in the environment would con-
tribute significantly to the formation of sulfate ions.
In the process of
netic S20" ions were
f the reaction between H2S and SO. on MgO paramag-
produced, in addition to elemental sulfur and water.
The 33S hyperfine structure observed in these experiments confirms the previous
identification of the ion by EPR spectroscopy. The spin densities of S20~
and the isoelectronic SO; and S; ions are compared in Table 8. The variations
TABLE 8. SPIN DENSITY OF THE ISGELECTRONIC SO;?, S2Q- and $3 ON MgO
SO^A) S02(B) S20- Sj
S3p (total) 0.75 0.71 0.91 1.08
S3p (Sulfur 1) 0.75 0.71 0.54 0.53
S3p (Sulfur 2) - - 0-37 0.27
S3p (Sulfur 3) - °-27
02p (Oxygen 1) 0.14 0.12 0.07
02p (Oxygen 2) 0.14 0.12
ref- 26 26 this 30
work
in spin density for SO;, S20", and S3 are in agreement with a simple argument
based on electron repulsion. The unpaired electron which occupies a highly
antibonding 2b", orbital is repelled to the less electronegative sulfur atom
by the concentration of electron charge in bonding orbitals near the more
electronegative oxygen atom. It is expected', therefore, that the central sul-
fur atom would increase in spin density as one moves up the series from S02 to
S3. By a similar argument one would predict that the replacement of an oxy-
gen atom by sulfur in SOj would result in a greater charge density but a
smaller spin density on the remaining oxygen atom.
At considerable lower temperatures trapped electrons on the surface of
MgO react with H,S forming a paramagnetic ion which has been identified as
H,S; It has been suggested that the ion was H2S~; however, the sulfur -33 hy-
perfine data provides convincing evidence that the molecule contains two, non-
equivalent sulfur atoms. The spin density of the unpaired electron was pre-
dominantly localized in the 3p orbitals of the two nonequivalent sulfur atoms
of the H2$2 radical. Since the sulfur atom has low-lying 3d onbtals, the
28
-------
energy levels for the 3p and 3d orbitals may be close; thus, in principle,
either a 03^ bonding orbital or a o3p* antibonding orbital may accomodate the
unpaired electron. Consideration for the occupancy of the unpaired electron
in the two possible molecular oribtals has been given by Akasaka et al. (32)
in their study of the radicals R-CH2- S-S-CH2-R obtained by y irradiation of
L-cystine dihydrochloride. Using their experimental g tensor they found an
energy splitting between the ground and the first excited state which was much
too large when an occupancy of the a3^ orbital was assumed. Therefore, the
unpaired electron probably occupies the 03^* antibonding orbital.
The CNDO calculations have further confirmed that the d orbital contri-
bution to the bonding in sulfur-containing molecules is relatively small.
These calculations also provide one of the first attempts to determine the
structure of adsorbed molecules by comparing calculated and experimental mag-
netic parameters. The approach is unique in that we attempt to minimize the
differences between calculated and experimental parameters, in contrast to
the usual method of estimating geometries whereby the total energy is mini-
mized with respect to bond lengths and bond angles. The latter method suffers
from the fact that the calculated minimum energies may not be directly com-
pared with experimental values. Comparing the CNDO/2 - calculated g tensors
with those reported by Chuvylkin and Zhidomorov (33) on the CO^ ion we found
that our calculated values (2.0032, 1.9975, 2.0023) are much closer to the
experimental values (2.0032, 1.9975, 2.0014) than theirs (2.0085, 1.9957,
2.0024). Moreover, the CNDO/2 calculation gives 121° as the best possible
bond angle for the CO;? ion, which is very close to the 124° predicted from
the sp hybridization as calculated from the hyperfine structure (34,35). The
calculated bond angles from the EPR spectroscopic study were 115±5° for the
502 ion, according to Dinse and Mb'bius (36), and the CNDO/2 calculation indi-
cates 120 and 110° for S02 (A) and (B), respectively. More support for the
accuracy of the CNDO/2 method comes from comparison with the scattered wave-
SCF-Xa method, by which a bond length and bond angle of the 85 ion were calcu-
lated (37) to be 2.05-2.10A and 110±5°, respectively. The CNDO/2 method to-
gether with experimental g values indicates a best fit at 2.1A and 110°. The
bond lengths of the sulfur-containing anions are long by 0.1 to 0.3A, compared
to the neutral molecules. This may be due to the repulsion which arises from
the extra electron.
29
-------
SECTION 5
LIST OF PUBLICATIONS
1. "Electron Paramagnetic Resonance Evidence for the Formation of S20~ on
Magnesium Oxide", M. J. Lin and J. H. Lunsford, J. Phys. Chem., 8£, 635-
639 (1976).
2 "An Electron Paramagnetic Resonance Study of the H2S£ Radical on Magnesium
Oxide", M. J. Lin and J. H. Lunsford, J. Phys. Chem., 80_, 2015-2018 (1976).
3. "Surface Reactions of S02 with N02 on Hydrated Silica Gel", S. M. Davis
and J. H. Lunsford, J. Environ. Sci. Health, All. 735-741 (1976).
4. "Low-Temperature Surface Oxidation of Hydrogen Sulfide by Oxygen", M. J.
Lin and J. H. Lunsford, J. Environ. Sci. Health, A12. 127-136 (1977).
5. "CNDO Calculations of the g Tensors and Spin Densities of Small Paramag-
netic Sulfur-Containing Anions on Magnesium Oxide", M. J. Lin and J. H.
Lunsford, J. Magnetic Res., 29_, 151-157 (1978).
6. "Surface Reactions of S02 with N02 on Hydrated Silica and Silica-Alumina",
S. M. Davis and J. H. Lunsford, J. Colloid Interface Sci., 65_, 352-364
(1978).
7. "The Oxidation of S02 with 02 on MnY Zeolites", J. R. Pearce and J. H.
Lunsford, J. Colloid Interface Sci., in press.
30
-------
REFERENCES
1. Maurin, P. G., and J. Jonakin. Removing sulfur oxides from stacks.
Chem. Eng. (New York), 77(9): 173-80, 1969.
2. Shy, C. M., and J. F. Finklea. Air pollution affects community health.
Environ. Sci. Technol., 7:204-208, 1973.
3. Prince, R., and F. F. Rose. Sulfur in air and soil. Water, Air and Soil
Pollut., 1:286-302, 1972.
4. Johnstone, H. F. Metallic ions as catalysts for the removal of sulfur
dioxide from boiler furnace gases. Ind. Eng. Chem., 23:559-561, 1931.
5. Johnstone, H. F. and D. R. Coughanowr. Absorption of sulfur dioxide from
air. Ind. Eng. Chem., 50:1169-1172, 1958.
6. Cheng. R. T., M. Corn and J. 0. Frohliger. Contribution to the reaction
kinetics of water soluble aerosols and S02 in air at ppm concentrations.
Atmospheric Environment, 5:987-1008, 1971.
7. Renzetti, N. A., and G. J. Doyle. Photochemical aerosol formation in sul-
fur dioxide-hydrocarbon systems. Int. J. Air Pollut., 2:327-345, 1960.
8. Jaffe, S., and F. S. Klein. Photolysis of N02 in the presence of S02 at
3660A. Trans. Faraday Soc., 62:2150-7, 1966.
9. Bourbigot Y., J. Bricard, G. Madelaine, and D. Vigla. Identification of
aerosols produced by photolysis in the presence of sulfur dioxide. C. R.
Acad. Sci., 276:547-50, 1973.
10. Canosa, C., W. G. Filby, and R. D. Penzhorn. On the gas phase products
from the thermal and photochemical nitrous oxide-sulfur oxide (SOX) reac-
tion. J. Environ. Sci. Health, Part A, All(l):93-8, 1976.
11. Cook. W. G. and R. A. Ross, Heterogeneous oxidation of hydrogen sulfide at
concentrations below 1000 ppm in nitrogen/air mixtures over supported
metal oxide catalysts. Atmos. Environ., 7:145-51, 1973.
12. Steijns, M., and P. Mars. Role of sulfur trapped in micropores in the
catalytic partial oxidation of hydrogen sulfide with oxygen. J. Catal.,
35:11-17, 1974.
31
-------
13 ct .
hydrogen sulfide oxidation catalyzed by sulfur. J.
1976.
14 Menon, P G., R. Sreeramamurthy, and P. S. Murti. Experimental study
of temperature and rate profiles in a rapidly fouling catalyst bed.
Chem. Eng. Sci., 27:641-4, 1972.
15 Steiins M , P. Koooman, B. Nieuwenhuijse- and P. Mars. The mechanism of
t e c Ulyt c oxidation of hydrogen sulfide. Ill An electron spin reso-
nance study of the sulfur catalyzed oxidation of hydrogen sulfide. J.
Catal., 42:96-106, 1976.
16 Belitskus, D. L. Oxidation of molten aluminum-magnesium alloy in air,
air-sulfur dioxide, and air-hydrogen sulfide atmospheres. Oxid. Metals,
3:313-17, 1971.
17 Fischer F , L. Goens, and H. Kraus. Reactions of hydrogen sulfide in
sorption from oxygen-rich gases on activated carbon. Chem. Ingr. Tech.,
36:963-8, 1964.
18 Pearce, J. R., and J. H. Lunsford. The oxidation of S02 with 02 on MnY
Zeolites. J. Colloid Interface Sci., in press.
19. Davis, S.M., and J. H. Lunsford. Surface reactions of SO, with N02 on
hydrated silica gel. J. Environ. Sci. Health, All.735-741, 19/b.
20 Davis S M. and J. H. Lunsford. Surface reactions of S02 with N02 on
hydrated silica and silica-alumina. J. Colloid Interface Sci., 65:352-
364, 1978.
21. Lin, M. J., and J. H. Lunsford Low-temperature surface oxidation of hy-
drogen sulfide by oxygen. J. Environ. Sci. Health, A12.127-136, I9//.
22 Lin M J , and J. H. Lunsford. An electron paramagnetic resonance study
of the H^1 radical on magnesium oxide. J. Phys. Chem. 80:2015-3018,
1976.
23 Lin, M. J., and J. H. Lunsford. Electron paramagnetic resonance evi-
dence for the formation of S2Q- on magnesium oxide. J. Phys. Chem., 80:
635-639, 1976.
24 Lin M J and J. H. Lunsford. CNDO calculations of the g tensors and
spin densities of small oaramagnetic sulfur-containing anions on mag-
nesium oxide. J. Magnetic Res., 29:151-157, 1978.
25 Morton, J. R. Identification of some sulfur-containing radicals traoped
in single crystals. J. Phys. Chem., 71:89-92, 1967.
32
-------
26. Schoonheydt, R. A., and J. H. Lunsford. Electron paramagnetic resonance
of SO! on magnesium oxide. J. Phys. Chem., 76:323-8, 1972.
27. Ono, Y., H. Tokunaga, and T. Keii. Electron spin resonance study of sul-
fur dioxide (1-) radicals on synthetic zeolites. J. Phys. Chem. 79:
752-6, 1975.
28. Schneider, J., B. Dischler, and A. Raeuber. Electron spin resonance of
sulfur and selenium radicals in alkali halides. Phys. Status Solidi,
13:141-57, 1966.
29. Mclaughlin, S. D., and D. J. Marshall. Paramagnetic resonance of sulfur
radicals in synthetic sodalites. J. Phys. Chem., 74:1359-63, 1970.
30. Lunsford, J. H., and D. P. Johnson. Electron paramagnetic resonance
study of triatomic sulfur ion formed on magnesium oxide. J. Chem. Phys.,
58:2079-83, 1973.
31. Bennett, J.E., B. Mile, and A. Thomas. Preparation and properties of the
H2S negative ion. Chem. Commun. 7:182-4, 1966.
32. Akasaka, K., S. Ohnishi, T. Suita, and I. Nitta. Electron spin resonance
(E.S.R.) of a single crystal of L-cystine dihydrochloride irradiated at
low temperature. J. Chem. Phys., 40:3110-16, 1964. Akasaka, K., S.
Kominami, and H. Hatano. Photoinduced hole-electron recombination in a
y-irradiated single crystal of L-cystine dihydrochloride. J. Phys. Chem.,
75:3746-50. 1971.
33. Chuvylkin, N. D., and G. M. Zhidomirov. Calculations of the anisotropic
hyperfine coupling constants in free radicals. J. Magn. Resonance. 11:
367-72, 1973.
34. Meriaudeau, P., J. C. Vedrine, Y. Ben Taarit, and C. Naccache. Electron
paramagnetic resonance studies of carbon dioxide anion radicals adsorbed
on magnesium oxide. Identification and structure of the species using
carbon-13 and oxygen-17 labeling. J. Chem. Soc. Faraday Trans. II,
71:736-48, 1975.
35. Lunsford, J. H. and J. P. Jayne. Formation of C02 radical ions when C02
is adsorbed on irradiated magnesium oxide. J. Phys..Chem., 69:2182-4,
1965.
36. Dinse, K. P., and K. Moebius. E. P. R. studies of electrolytically gener-
ated S02 . Z. Naturforsch, 23:695-702, 1968.
37. Cotton, F. A., J. B. Harmon, and R. M. Hedges. Calculation of the ground
state electronic structures and electronic spectra of di- and trisulfide
radical anions by the scattered wave SCF-Xa method. J. Am. Chem. Soc.,
98:1417-24, 1976.
33
-------
TECHNICAL REPORT DATA .
(Please read Instructions on the reverse before completing}
" r~^
1 REPORT NO.
EPA-600/7-79-088
\A. TITLE ANDSUBTITLE
SURFACE REACTIONS OF OXIDES OF SULFUR
7. AUTHOR(S)
J.H. Lunsford _
9. PERFORMING ORGANIZATION NAME AND ADDRESS
8. PERFORMING ORGANIZATION REPORT NO.
Department of Chemistry
Texas A&M University
College Station, Texas
77843
12. SPONSORING AGENCY NAME AND ADDRESS
Environmental Sciences Research Laboratory - RTF, NC
Office of Research and Development
U.S. Environmental Protection Agency
Research Triangle ParkT North Carolina 27711
15. SUPPLEMENTARY NOTES
. RECIPIENT'S ACCESSION-NO.
. REPORT DATE
April 1979
,. PERFORMING ORGANIZATION CODE
10. PROGRAM ELEMENT NO.
1NE625 (EA 21) FY77
11. CONTRACT/GRANT NO.
801136
13. TYPE OF REPORT AND PERIOD COVERED
Final 12m to 5/78
14. SPONSORING AGENCY CODE
EPA/600/09
^ereas '""go": °"rie° of 'products^ observed including elemental sulfur,
' ions and a number of paramagnetic ions such as S3, S20
5
function of bond angles and bond lengths.
17.
a.
KEY WORDS AND DOCUMENT ANALYSIS
DESCRIPTORS
b.lDENTIFIERS/OPEN ENDED TERMS
*Air pollution
*Sulfur oxides
*Chemical reactions
*Aerosols
*Surface chemistry
*Ion exchange resins
18. DISTRIBUTION STATEMENT
RELEASE TO PUBLIC
EPA Form 2220-1 (9-73)
19. SECURITY CLASS (This Report)
UNCLASSIFIED
20 SECURITY CLASS (This page/
UNCLASSIFIED
COSATI Field/Group
13B
07B
07D
21. NO. OF PAGES
42
22. PRICE
34
------- |