The role of SO2 in the reduction of NO by CO on La2O2S

Article abstract of DOI:10.1016/j.jcat.2006.10.025

The decomposition of NO and the reduction of NO by CO on La₂O₂S were studied using temperature-programmed reaction technique coupled with fast mass spectrometry, powder X-ray diffraction and X-ray photoelectron spectroscopy. The role of SO₂ plays in the reduction is explored. It is found that the decomposition of NO is a favorable reaction step for the reduction of NO by CO achieved through a sulfur-assisted path. The oxygen produced in the decomposition is removed by the sulfur in the oxysulfide as SO₂, which in turn is reduced back to sulfur by CO on La₂O₂S. An external supply of sulfur, such as SO₂, in the feed is needed to maintain the population of sulfur in the oxysulfide and thus make the reduction sustainable.

Full text of DOI:10.1016/j.jcat.2006.10.025

Journal of Catalysis 245 (2007) 301–307
www.elsevier.com/locate/jcat
The role of SO2 in the reduction of NO by CO on La2O2S
Ngai T. Lau ∗, Ming Fang, Chak K. Chan
Department of Chemical Engineering, The Hong Kong University of Science and Technology, Clear Water Bay, Hong Kong
Received 7 June 2006; revised 11 October 2006; accepted 23 October 2006
Available online 28 November 2006
Abstract
The decomposition of NO and the reduction of NO by CO on La O S were studied using temperature-programmed reaction technique coupled
2
2
with fast mass spectrometry, powder X-ray diffraction and X-ray photoelectron spectroscopy. The role of SO plays in the reduction is explored.
2
It is found that the decomposition of NO is a favorable reaction step for the reduction of NO by CO achieved through a sulfur-assisted path.
The oxygen produced in the decomposition is removed by the sulfur in the oxysulfide as SO , which in turn is reduced back to sulfur by CO on
2
La O S. An external supply of sulfur, such as SO , in the feed is needed to maintain the population of sulfur in the oxysulfide and thus make the
2
2
2
reduction sustainable.
© 2006 Elsevier Inc. All rights reserved.
Keywords: Lanthanum oxysulfide; Decomposition; Reaction step; Temperature-programmed reaction; XPS
1. Introduction
La2O3 by Huang et al. [7] supports the R2 center decomposi-
tion mechanism. It is interesting to note that when NO is fed
to La2O3 that has been heated in helium at 973 K and cooled
to room temperature, a small amount of N2O is produced. This
indicates that decomposition of NO will readily occur even at
room temperature if anion vacancies are made available:
Lanthanum oxysulfide (La2O2S) has been shown to be
highly active for the simultaneous reduction of SO2 and NO
by CO, but when SO2 is absent, the reduction of NO cannot
be sustained, and the conversion of NO drops significantly with
time [1]. This suggests that the reduction of NO is closely cou-
pled to the reduction of SO2. However, the role that SO2 plays
in the reduction of NO remains unknown, and the reaction steps
involved in the reduction of NO on La2O2S have not been stud-
ied and reported. In contrast, the reduction and decomposition
of NO on La2O3 have been studied extensively [2–8] and can be
used as a reference for the study of the oxysulfide reaction sys-
tem. La2O3, the precursor of La2O2S [9], has a structure close
to that of La2O2S, with one oxide ion replaced by a sulfide ion.
The reduction and decomposition of NO on La2O3 are usu-
ally associated with oxygen anion vacancies. In particular, Win-
ter [2] proposed that the decomposition of NO starts with the
adsorption of NO molecules at R2 centers. An R2 center is ba-
sically a pair of adjacent oxygen anion vacancies; the oxygen
remaining from the NO occupies the center after decomposi-
tion. The study of the adsorption and decomposition of NO on
2NO → N2O + O.
(1)
On La2O2S, oxygen anion vacancies are probably formed in a
manner comparable to that on the oxide, and also can be pro-
duced via the sulfur pathway, as reported by Lau and Fang [10]:
∗S → S + ∗,
(2)
∗
∗
where S is a sulfide ion in the surface lattice and is an an-
ion vacancy with a negative charge. Thus, it is fair to suggest
that NO is also readily decomposed on La2O2S provided that
these anion vacancies are available. Winter [11] pointed out
that in the decomposition of NO, the desorption of oxygen to
regenerate the vacancies can be the limiting step. Thus, if we
can identify a chemical species in La2O2S that promotes the
removal of oxygen and regenerates the anion vacancies at tem-
peratures typical of NO reduction, then the proposition that the
decomposition of NO is an important step in the reduction of
NO on La2O2S can be substantiated. The goal of this paper is
to identify the species in the oxysulfide that can remove the
*
Corresponding author. Fax: +852 2719 6710.
E-mail address: ntlau@ust.hk (N.T. Lau).
0021-9517/$ – see front matter © 2006 Elsevier Inc. All rights reserved.
doi:10.1016/j.jcat.2006.10.025

302
N.T. Lau et al. / Journal of Catalysis 245 (2007) 301–307
oxygen at the typical reaction temperatures, and by doing so
demonstrate that the decomposition of NO is an important re-
action step for the reduction of NO by CO on La2O2S.
In this study, we used temperature-programmed reaction
coupled with fast mass spectrometry (TPR/MS) to track the
decomposition and the reduction of NO in the presence and
absence of SO2 by CO on La2O2S with reference to those on
La2O3. Changes in the catalyst composition due to the NO re-
actions were monitored with powder X-ray diffraction (XRD)
and X-ray photoelectron spectroscopy (XPS). The TPR/MS
technique temporally and thermally isolates the interactions be-
tween the reactants and the catalyst and allows identification of
the active species in the catalyst.
(C1) 0.04%v NO,
(C2) 0.4%v SO2, 0.8%v CO,
(C3) 0.16%v NO, 0.16%v CO, and
(C4) 0.16%v NO, 0.4%v SO2, 0.96%v CO,
was obtained by diluting the preblended reaction gases with Ar
in the gas-mixing flask before it was fed to the reactor. The com-
position of the effluent gas from the reactor was continuously
monitored with the mass spectrometer (Extrel MS250), which
reported the composition data each second.
The typical TPR procedure started with purging the sample
with feed gas at room temperature (∼25◦C) until the composi-
tion of the effluent from the reactor was steady. The sample was
then heated from room temperature to 700 ◦C at 10 ◦C min−1
and kept at 700 ◦C for 2 h or until the composition of the efflu-
ent gas was steady.
2. Experimental
2.1. Catalyst preparation
2.3. Catalyst composition
La2O2S was synthesized in situ in a quartz microreactor fol-
lowing the sulfidization procedure reported by Ma et al. [9]
from La(OH)3 powder (Yiaolong, China) at 700 ◦C for 3 h.
The sample was kept in flowing Ar or He before use. The
composition of the sample material synthesized was La2O2S,
as confirmed by XRD, and the sample’s specific surface area
was approximately 4 m2 g−1 as obtained with the BET method
(ASAP2000, Micromeritics).
La2O3 was also prepared in situ in a quartz microreactor by
calcining the La(OH)3 powder at 700 ◦C for 3 h in flowing Ar
at ∼90 ml min−1. The composition of the sample was La2O3,
as confirmed by XRD, and the sample’s specific surface area
was approximately 4 m2 g−1, as obtained with BET.
The bulk composition of the catalyst was monitored with
XRD using a Philips model PW1830 powder X-ray diffraction
system equipped with a Cu anode and a graphite monochroma-
tor operated at 2 kW. The samples were spread and pressed onto
glass sample holders with no pretreatment.
The surface composition of the catalyst was measured by
XPS (Perkin–Elmer Surface Science Analysis System model
PHI 5600) equipped with a monochromatic AlKα X-ray source
according to the procedure described by Lau and Fang [10]. The
powder samples were pressed into cup-shaped sample mounts
for direct insertion into the instrument. Low-energy flooding
electrons were used to neutralize the charges built up on the
samples, and the binding energy (BE) scale was so adjusted
to make the adventitious carbon peak at 284.5 eV. Multiplex
scans of carbon C1s, nitrogen N1s, oxygen O1s, sulfur S2p,
and lanthanum La3d were acquired at constant pass energy of
23.5 eV.
2.2. TPR/MS
TPR/MS measurements were obtained using the experimen-
tal setup shown in Fig. 1. A 0.1-g sample of La2O2S was used
in each measurement. The feed gas, at 100 ml min−1, 25 ◦C,
and 1 atm, with typical composition
3. Results
3.1. Change in catalyst composition
The XPS C1s, O1s, S2p, and La3d spectra of La2O2S used
in the decomposition and reduction of NO in the presence and
absence of SO2 are shown in Fig. 2. The as-sulfidized sample is
also presented as a reference. The typical O1s and S2p peaks are
listed in Table 1. The as-sulfidized sample XPS spectra are com-
parable to those reported by Lau and Fang [10]. Although most
of the samples were subjected to the reactions involving NO,
no detectable amount of nitrogen was found on their surfaces.
The C1s spectrum contained two peaks: a higher-intensity peak
at 284.5 eV, attributable to adventitious carbon, and a lower-
intensity peak at ∼289 eV, associated with C–O species, such
as carbonate. S2p at BE ∼160.1 eV can be assigned to the sul-
fide in the lattice or the sulfur in the surface not bonded to
oxygen. The broader group of S2p at ∼166.4–168.5 eV can
be attributed to S–O species, such as SO2 adspecies, sulfite,
and sulfate. O1s also contained two peaks, one at lower BE
Fig. 1. Flow schematic of experiment setup.

N.T. Lau et al. / Journal of Catalysis 245 (2007) 301–307
303
Fig. 2. XPS spectra (a) C1s, (b) O1s, (c) S2p and (d) La3d of La O S—effect of NO. Curves in the figures from top to bottom, (A) no treatment (as-sulfidized),
5/2
2 2
(B) TPR reaction with NO, SO and CO, (C) TPR reaction with NO and CO, and (D) TPR reaction with NO. All the spectra have been rescaled to the same vertical
2
span.
Table 1
Effect of NO reactions on XPS O1s and S2p peaks of La O S
XRD revealed no significant changes in the bulk composi-
tion of the La2O2S catalyst after the NO reactions, although
XPS revealed subtle changes in the surface composition. Ta-
ble 2 shows the changes in the carbon, oxygen, and sulfur
species in the surface of the catalyst due to the NO reactions.
The decomposition of NO significantly decreased the concen-
tration of sulfur in the surface relative to lanthanum (S/La), es-
pecially the sulfur not associated with the oxygen (SONLY/La),
while increasing the concentrations of oxygen (O/La) and S–O
species in the surface. Similar changes were also observed on
La2O2S subjected to the reduction of NO by CO. The lattice
oxygen O1s was almost completely diminished, and the O1s at
higher BE shifted to 530.9 eV, indicating a dominant contribu-
tion from the carbonate [12]. The reduction of NO in the pres-
2
2
Reactions
Binding energy (eV)
S2p
O1s
528.9
As-sulfidized
531.6
531.5
530.9
531.5
160.0
160.2
160.2
160.1
166.4–168.0
168.5
168.3
NO decomposition
NO reduction by CO
NO reduction by CO in
528.9
–
528.9
166.4–167.7
the presence of SO
2
The oxysulfide was produced by sulfidizing lanthanum hydroxide powder with
a stoichiometric mixture of SO and CO.
2
(∼528.9 eV) due to lattice oxygen or oxygen not related to C–O
or S–O species, and the other at higher BE comprising contri-
butions from the C–O and S–O species.

304
N.T. Lau et al. / Journal of Catalysis 245 (2007) 301–307
Table 2
and
Effect of NO reactions on the surface concentration of carbon, oxygen and sul-
2NO → N2O + O.
(1)
fur in La O S
2
2
Reactions
C/La O/La S/La
S
/La S–O/La S–O/S
ONLY
Instead of oxygen, SO2 was produced; its formation was closely
associated with the decomposition of NO. As NO decompo-
sition diminished with time, SO2 formation also slowed. SO2
started to form at ∼320, ∼40◦C below the temperature at which
significant amounts of N2 and N2O are formed. Furthermore,
the SO2 formation curve consisted of two temperature peaks.
Based on the formation of COS by the reaction between CO
and sulfur in La2O2S [10], the lower peak can be attributed to
the reaction with labile sulfur in the surface, whereas the higher
peak may be associated with the reaction with the lattice sulfur.
Note that a significant amount of NO was desorbed immedi-
ately before the start of decomposition.
As-sulfidized
1.3
1.9
1.5
1.2
3.2
4.0
4.1
3.2
0.65 0.46
0.55 0.32
0.40 0.24
0.52 0.40
0.20
0.23
0.16
0.12
0.30
0.41
0.40
0.23
NO decomposition
NO reduction by CO
NO reduction by CO in
the presence of SO
2
C/La, O/La and S/La are the relative atomic concentration of carbon, oxygen
and sulfur to lanthanum obtained with the XPS measurements. S is the
sulfur species in the surface that is not bonded to oxygen such as sulfide. S–O/S
is the relative atomic concentration of surface S–O species to the total sulfur
species S.
ONLY
3.3. Reduction of NO: TPR/MS
Fig. 4 shows the thermal TPR/MS profile for the reduction
of NO in the presence of SO2. The conversion, C, is computed
from the mass intensities using the following equation:
I0 − I
C =
,
(4)
I0
where I and I0 are the net intensities of the selected mass frag-
ment of the reactant species in the effluent and feed streams,
respectively. For example, C_CO, the CO conversion, is com-
puted from the intensities of fragment with mass-to-charge ratio
m/z = 28 in effluent and feed. This intensity usually measures
the CO concentration when CO is involved in the reaction.
However, the formation of N2 in the reduction of NO appar-
ently lowers the conversion of CO due to the interference of
N2. The mass fragment also includes the contribution from CO2
and N2O, but their contribution is usually negligible relative
to CO and N2. Similarly, C_NO and C_SO2, the conversions
of NO and SO2, are computed from intensities m/z = 30 and
64, respectively. C_NO also contains the fragment contribu-
tion from N2O, but this usually is very small. C_SO2 is usu-
ally not affected by other species except S2 vapor. S2 vapor
is not volatile and is usually deposited on the cool surface of
the reactor and transfer line; only a negligible fraction can en-
ter into the mass spectrometer. The CO2 intensity (m/z = 44)
measures the concentration of the products of CO2 but also
may include contributions from N2O. When there is significant
conversion of CO, CO2 will be the dominating species; how-
ever, when CO conversion is low, N2O can be the dominant
species.
Fig. 3. Decomposition of NO on La O S. The nominal feed composition was
2
2
0.04%v NO. The relative concentrations of the species were obtained by cor-
recting the intensity of m/z = 28, 30, 44 and 64 for zeroes and fragmentation
effect, and then adjusting for the sensitivity of the compounds. The dashed line
◦
marks the beginning of the 700 C reaction temperature zone. The dotted line
marks the maximum NO desorption point.
ence of SO2 reduced the concentrations of the sulfur and S–O
species in the surface but did not change the surface oxygen
concentrations. Thus, the decomposition and reduction of NO
in the absence of SO2 consumed surface sulfur and produced
S–O species, whereas the reduction by CO kept the concen-
tration of S–O species low. In the presence of SO2, the sulfur
consumed during NO reduction likely was replenished by the
sulfur produced during SO2 reduction.
In the presence of SO2, the reduction of NO by CO began
readily on La2O2S at ∼440◦C, and high and sustained con-
versions of NO, CO, and SO2 were attained at >500◦C. The
reduction was almost impossible on La2O3; only very low con-
versions of NO, SO2, and CO were detectable at >600◦C. This
is consistent with the findings of Lau et al. [13] that anhydrous
La2O3 could not be sulfidized to La2O2S and is inactive for the
reduction of SO2 by CO.
3.2. Decomposition of NO: TPR/MS results
In Fig. 3, the temporal TPR/MS profile of the decomposition
of NO on La2O2S shows that N2 and N2O were produced as
NO was consumed. This indicates the following decomposition
reactions:
When SO2 was absent (Fig. 5), the reduction of NO on
La2O2S could not be sustained. Although a significant amount
2NO → N2 + 2O
(3)

N.T. Lau et al. / Journal of Catalysis 245 (2007) 301–307
305
Fig. 4. Catalytic reduction of NO by CO on (a) La O S and (b) La O in
2
2
2 3
the presence of SO (NO/SO = 0.4). The C_CO, C_NO, and C_SO are the
2
2
2
conversion curves for CO, NO, and SO obtained with the mass intensities
2
m/z = 28, 30 and 64, respectively, and use the left scale. The scale of the CO
2
(m/z = 44) intensity curve is on the right.
of NO was reduced with the formation of SO2, CO2, and pos-
sibly N2O in the first TPR program at >320◦C, a negligible
amount of NO and CO reacted in the second TPR, with no
SO2 formed. The conversion of CO in the first TPR was sig-
nificantly lower than that of NO, probably due to the interfer-
ence of N2 to C_CO and the fact that NO decomposition was
the dominant reaction. Almost no reduction was observed on
La2O3.
In summary, the reduction of NO by CO did not occur on
La2O3 irrespective of the presence or absence of SO2; however,
on La2O2S, the reduction was sustained with high conversion
of NO, CO, and SO2 in the presence of SO2. In the absence of
SO2, the NO reaction also could not be sustained, and it even-
tually died out when the formation of SO2 ceased.
Fig. 5. Catalytic reduction of NO by CO on (a–b) La O S and (c) La O in the
2
2
2 3
absence of SO ; (a) is the result of the first temperature program. The C_CO
2
and C_NO are the CO and NO conversion curves obtained with the mass in-
tensities m/z = 28 and 30, respectively, and use the left scale. The scale of
the CO (m/z = 44) and SO (m/z = 64) intensity curves is on the right. The
2
2
dotted lines mark the two CO bumps in (a).
2

306
N.T. Lau et al. / Journal of Catalysis 245 (2007) 301–307
4. Discussion
that nitrogen was held more tightly by the oxysulfide, probably
at the anion vacancies (the adsorption sites). This suggests that
the NO is adsorbed on La2O2S with the N-end at the vacancies
differing from Winter’s decomposition scheme with the O-end
at the vacancies [2]. The DRIFTS study on the decomposition
of NO on La2O3 by Huang et al. [7] also reported NO ad-
species with the N-end at the oxysulfide. In addition, nitrogen,
being less electronegative than oxygen, can be slightly more
positively charged, making the N-ends more readily approach
the negatively charged anion vacancies. There is also another
possibility that the decomposition of NO on La2O2S can be ac-
complished with two adjacent anion vacancies,
The TPR/MS and XPS results both show that the decompo-
sition and reduction of NO are closely associated with the sulfur
in the catalyst and the oxidation of the sulfur to SO2. This was
also observed in the reduction of NO on supported sulfides on
γ -Al2O3 [14], CoSX–TiO2 [15], and Ti–Sn solid solution cat-
alysts [16]. The sulfur produced in the reduction of SO2 helps
remove the oxygen from the decomposition of NO as SO2, that
is, the sulfur-assisted reduction of NO. It is possible that the
reduction of NO on La2O2S also follows a similar pathway:
S + 2O → SO2.
(5)
NO + 2∗ → ∗N + ∗O,
(6)
Removing the oxygen helps regenerate the anion vacancies for
the next NO decomposition cycle.
where ∗ represents an anion vacancy. However, further research
is needed to clarify this issue.
Taking SO2 formation as an indicator for the decomposi-
tion of NO, the NO decomposition ignition temperature (Fig. 3)
will be ∼320◦C. This is approximately the same temperature at
which a significant amount of NO was reacted in the reduction
of NO on La2O2S in the absence of SO2. This temperature was
also ∼100◦C lower than the temperature at which significant
NO reduction was observed in the presence of SO2. Thus, the
decomposition of NO is a favored reaction between NO and
La2O2S.
As sulfur is consumed to remove the oxygen and regener-
ate the anion vacancies, the available sulfur in the oxysulfide
is eventually used up if not replenished. Presumably, the SO2
produced in the decomposition of NO can be reduced to replen-
ish the oxysulfide. But insufficient sulfur is available for this,
because a portion of the sulfur species is lost to the effluent
stream. An additional supply of sulfur, such as the SO2 in the
feed, is needed. This is why the reduction of NO can be sus-
tained only in the presence of SO2 in the feed.
Furthermore, N2O is the preferential product for the decom-
position of NO at low temperatures when the reaction started
(see Fig. 3). As more NO is consumed at higher temperatures,
N2 becomes the dominant product. This shift in N2 selectivity
may be associated with the population of N resulting from NO
decomposition. When plenty of N is produced at sufficiently
high temperature, N2 is readily and preferentially formed by the
combination of N. At low temperatures, where N is relatively
scarce, there is greater likelihood of N reacting with another
NO to form N2O before N2 formation can occur.
Therefore, it is likely that the reduction of NO on La2O2S
follows these reaction steps:
Step 1: Adsorption of NO at the adsorption sites.
Step 2: Decomposition of NO to N and O.
Step 3: Removal of O by the oxidation of sulfur in La2O2S to
SO2.
The reduction of NO by COS is probably one of the com-
peting reaction paths for the reduction of NO over La2O2S,
especially when COS is present in the feed. Ma et al. [1] re-
ported that the reduction of NO by COS can be significant at
temperatures as low as 150 ◦C over La2O2S. However, if COS
is absent from the feed and must be produced by the reaction
between the sulfur in the oxysulfide and CO [10], then the COS
reaction path probably will not be as effective as the reaction
path via NO decomposition. This is because the NO decom-
position path is a more direct reaction path involving the sulfur,
whereas the COS path is an indirect path in which COS must be
formed at temperatures above 340 ◦C [10] and there is serious
competition for COS from the reduction of SO2 [18].
Fig. 3 indicates that SO2 was produced before the NO de-
composition reaction. Some of the SO2 could be the residual
adspecies from the sulfidization process when we synthesized
the oxysulfide, whereas the remainder could be formed by the
oxidation of sulfur by the oxygen produced in the decomposi-
tion of the adsorbed NO on the oxysulfide. A significant amount
of NO was desorbed from the oxysulfide when SO2 started to
form, as shown in Fig. 3. The desorption created vacant adsorp-
tion sites. This suggests that additional vacant sites promoted
the decomposition of NO. Furthermore, nitrogen products (N2
and N2O) were formed well after SO2 was produced, indicating
Step 4: Desorption of nitrogen as N2.
Step 5: Reduction of SO2 by CO to sulfur.
The adsorption sites are the anion vacancies in the surface of
La2O2S. Step 5 maintains the sulfur population in the oxysul-
fide and guarantees that sufficient sulfur is available to remove
O in Step 3.
The role of SO2 in NO reduction is schematically illustrated
in Fig. 6. This diagram also explains how NO reduction is
linked to the reduction of SO2 on La2O2S. The lower reaction
cycle shows the COS-intermediate mechanism for the reduction
of SO2 by CO on La2O2S [10,17,18]. The COS intermediate is
formed when CO reacts with sulfur in the oxysulfide. The inter-
mediate then reacts with SO2 on the La2O2S to produce sulfur
and CO2. The NO reaction cycle is represented by the upper cy-
cle. Sulfur in the oxysulfide fuels the decomposition of NO to
N2 by forming SO2, which is in turn reduced by CO to sulfur
in the SO2 reduction cycle. Thus, the S–SO2 pair facilitates the
NO reaction—an indirect reduction path of NO by CO.
5. Conclusion
This study of the decomposition of NO and the reduction
of NO by CO in the presence and absence of SO2 on La2O2S

N.T. Lau et al. / Journal of Catalysis 245 (2007) 301–307
307
HKUST6001/01P). The authors thank Professor J.X. Ma of
Tongji University, Shanghai, China, for his helpful advice and
comments.
References
[1] J.X. Ma, M. Fang, N.T. Lau, Catal. Lett. 62 (1999) 127.
[2] E.R.S. Winter, J. Catal. 22 (1971) 158.
[3] X. Zhang, A.B. Walters, M.A. Vannice, Appl. Catal. B 4 (1994) 237.
[4] X.X. Zhang, A.B. Walters, M.A. Vannice, NO reduction by CH over rare
4
earth oxides, in: G. Centi, C. Cristiani, P. Forzatti, S. Perathoner (Eds.),
Environmental Catalysis—For a Better World and Life: Proc. 1st World
Conference, Pisa, 1–5 May 1995, SCI Publisher, Rome, 1995, p. 33.
[5] M.A. Vannice, A.B. Walters, X. Zhang, J. Catal. 159 (1996) 119.
[6] S.J. Huang, A.B. Walters, M.A. Vannice, J. Catal. 173 (1998) 229.
[7] S.J. Huang, A.B. Walters, M.A. Vannice, J. Catal. 192 (2000) 29.
[8] T.J. Toops, A.B. Walters, M.A. Vannice, Appl. Catal. B 38 (2002) 183.
[9] J.X. Ma, M. Fang, N.T. Lau, J. Catal. 163 (1996) 271.
[10] N.T. Lau, M. Fang, J. Catal. 179 (1998) 343.
Fig. 6. Coupling between the reduction of NO and the reduction of SO on
2
La O S—an illustration.
2
2
found that the decomposition reaction is the favored reaction
step in the reduction of NO by CO and that the reduction pro-
ceeds via the sulfur-assisted path. The oxygen produced in the
decomposition is removed by sulfur in the oxysulfide as SO2,
which is then reduced back to sulfur by CO on La2O2S. This
step helps regenerate the anion vacancies for NO decomposi-
tion. The S–SO2 pair facilitates the indirect reduction of NO by
CO to nitrogen through the SO2 reduction cycle. An external
supply of sulfur, such as SO2 in the feed, must be available to
sustain the NO reduction process.
[11] E.R.S. Winter, J. Catal. 34 (1974) 440.
[12] J.F. Moulder, W.F. Stickle, P.E. Sobol, K.D. Bomben, in: J. Chastain (Ed.),
Handbook of X-ray Photoelectron Spectroscopy—A Reference Book of
Standard Spectra for Identification and Interpretation of XPS Data, Phys-
ical Electronic Division, Perkin–Elmer Corporation, Eden Prairie, MN,
1992.
[13] N.T. Lau, M. Fang, J.X. Ma, Appl. Catal. B 26 (2000) 81.
[14] S. Zhuang, M. Yamazaki, K. Omata, Y. Takahashi, M. Yamada, Appl.
Catal. B 31 (2001) 133.
[15] Z. Zhang, J. Ma, X. Yang, J. Mol. Catal. A Chem. 195 (2003) 189.
[16] Z. Zhang, J. Ma, X. Yang, Chem. Eng. J. 95 (2003) 15.
[17] J.X. Ma, M. Fang, N.T. Lau, Appl. Catal. A 150 (1997) 253.
[18] N.T. Lau, M. Fang, C.K. Chan, J. Mol. Catal. A Chem. 203 (2003)
221.
Acknowledgments
This work was supported by the Research Grant Council of
the Hong Kong Special Administration Region, China (Project