Effects of sulfur on Mo2C and Pt/Mo2C catalysts: Water gas shift reaction

Article abstract of DOI:10.1016/j.jcat.2010.04.004

Molybdenum carbide (Mo₂C) and Pt/Mo₂C catalysts were evaluated for the water gas shift reaction without and with 5 ppm H ₂S. The Mo₂C catalyst was quickly poisoned by sulfur, achieving a rate that was ～10% of that prior to sulfur exposure. The Pt/Mo₂C catalyst was initially more active than the Mo₂C catalyst and deactivated more gradually to a level similar to that for the Mo₂C catalyst. X-ray photoelectron spectroscopy revealed Mo ₂C, MoS₂, and S-Mo on the spent catalysts; the Pt/Mo ₂C catalyst also contained PtS. The results are consistent with Mo₂C sites on the Mo₂C and Pt/Mo₂C catalysts being partially sulfur tolerant, in that these sites could be reactivated on treatment in 15% CH₄/H₂ at 590 °C. High activity sites associated with Pt nanoparticles were irreversibly deactivated. Residual activities for the Mo₂C and Pt/Mo₂C catalysts in the presence of H₂S appeared to be associated primarily with the presence of MoS₂ domains.

Full text of DOI:10.1016/j.jcat.2010.04.004

Journal of Catalysis 272 (2010) 235–245
Contents lists available at ScienceDirect
Journal of Catalysis
journal homepage: www.elsevier.com/locate/jcat
Effects of sulfur on Mo C and Pt/Mo C catalysts: Water gas shift reaction
2
2
Joshua A. Schaidle, Adam C. Lausche, Levi T. Thompson *
University of Michigan, Department of Chemical Engineering, 3020 H.H. Dow Building, 2300 Hayward Street, Ann Arbor, MI 48109-2136, USA
a r t i c l e i n f o
a b s t r a c t
Article history:
Molybdenum carbide (Mo C) and Pt/Mo C catalysts were evaluated for the water gas shift reaction with-
2
2
Received 26 January 2010
Revised 7 April 2010
Accepted 8 April 2010
Available online 21 May 2010
out and with 5 ppm H
2
S. The Mo
2
C catalyst was quickly poisoned by sulfur, achieving a rate that was
ꢀ
10% of that prior to sulfur exposure. The Pt/Mo
alyst and deactivated more gradually to a level similar to that for the Mo
spectroscopy revealed Mo C, MoS , and S–Mo on the spent catalysts; the Pt/Mo
PtS. The results are consistent with Mo C sites on the Mo C and Pt/Mo C catalysts being partially sulfur
tolerant, in that these sites could be reactivated on treatment in 15% CH /H at 590 °C. High activity sites
associated with Pt nanoparticles were irreversibly deactivated. Residual activities for the Mo C and Pt/
Mo C catalysts in the presence of H S appeared to be associated primarily with the presence of MoS
domains.
2
C catalyst was initially more active than the Mo
C catalyst. X-ray photoelectron
C catalyst also contained
2
C cat-
2
2
2
2
2
2
2
Keywords:
4
2
Molybdenum carbide
Water gas shift
Sulfur poisoning
2
2
2
2
2
H S
Platinum
Ó 2010 Elsevier Inc. All rights reserved.
1
. Introduction
Sulfur is a contaminant in fossil fuels including crude oil and
Research described in this paper investigated the tolerance of
Mo
2
C and Pt/Mo
2
C catalysts to sulfur exposure during the WGS
reaction
coal, and often remains in products derived from fossil fuels. Sulfur
can also be present in biomass-derived products. For example,
Robinson et al. [1] reported that, depending on the type of biomass,
the sulfur content varied from ꢀ14–2200 ppm. Sulfur severely and
irreversibly deactivates most catalytic materials, and typically has
to be removed upstream of the reactor. For some reactions, early
transition metal carbides and nitrides have been reported to be
sulfur tolerant, in particular at high temperatures and/or pressures
ꢁ
CO þ H
2
O $ CO
2
þ H
2
D
H ¼ ꢂ41:1 kJ=mol
ð1Þ
This reaction is a critical step in the conversion of hydrocarbon and
alcohol feedstocks into reformate or syngas [15]. Due to the intoler-
ance of most WGS catalysts, sulfur concentrations in the feed typi-
cally have to be reduced to ppb levels [16]. The effects of H
exposure on the WGS activities of Mo C and Pt/Mo C catalysts were
evaluated, and the materials were characterized to define any asso-
ciated changes in catalyst structure and composition. Hydrogen sul-
fide was used as the sulfur species because it is often present in
process streams [17]. Results presented in this paper were deter-
mined at more moderate temperatures and pressures than those
used in most prior investigations [2–4] and provide new insights
2
S
2
2
[
2–4]. For example, DaCosta et al. [3] report that tetralin hydroge-
nation activities for Mo C/Al and WC/Al catalysts were not
significantly impacted by the presence of 200 ppm H S at 300 °C
2
2
O
3
2 3
O
2
and 4 MPa. Carbides and nitrides can be synthesized with high sur-
2
face areas (up to 200 m /g) and are catalytically active for a variety
of reactions including hydrodesulfurization [4], alcohol amination
into the effects of sulfur on the catalytic properties of Mo
2
C-based
[
5], alkane isomerization [6], alkane hydrogenolysis [7], and CO
catalysts.
hydrogenation [8]. Because of their high surface areas, carbides
and nitrides have attracted attention for use as support materials,
especially for electrocatalysts where their high electronic conduc-
tivities can be exploited [9–11]. Recently, a technique has been
developed to support metals like Pt and Ni directly onto carbide
and nitride surfaces [12]. Some of the resulting materials have
been demonstrated to be more active than conventional catalysts
for reactions such as water gas shift (WGS) [13] and methanol
steam reforming [14].
2. Materials and methods
2.1. Catalyst preparation
The Mo
programmed reaction procedure [18]. Approximately 1.3 g of
ammonium paramolybdate (AM, (NH Mo O, 81–83%
24ꢃ4H
MoO , Alpha-Aesar) was loaded into a quartz tube reactor on
top of a quartz wool plug. In order to maintain consistent
catalyst properties, the AM was sieved to 125–250 m prior to
carburization. The AM was reduced in H at 375 mL/min as the
2
C catalyst was synthesized using a temperature-
4
)
6
7
O
2
3
l
*
Corresponding author. Fax: +1 734 763 0459.
E-mail address: ltt@umich.edu (L.T. Thompson).
2
0
021-9517/$ - see front matter Ó 2010 Elsevier Inc. All rights reserved.
doi:10.1016/j.jcat.2010.04.004

2
36
J.A. Schaidle et al. / Journal of Catalysis 272 (2010) 235–245
temperature was increased from room temperature (RT) to
50 °C (heating rate of 278 °C/h), and then held at this temper-
ature for 12 h. The reactant gas was then switched from H to
a 15% CH /H mixture and the temperature was increased from
50 °C to 590 °C at a rate of 160 °C/h. The final temperature
was maintained for 2 h prior to quenching the material to RT.
This procedure has been shown to produce b-Mo C [19]. The
resulting material was passivated using a 1% O /He mixture at
0 mL/min for at least 5 h.
The Mo C-supported Pt catalyst was prepared via wet impreg-
nation of the unpassivated Mo C with a deaerated aqueous solu-
tion containing 1.3 mg/mL of dihydrogen hexachloriplatinate
hexahydrate (H PtCl O, 99.95% metal basis, Alfa Aesar). After
ꢃ6H
decanting the excess solution, the material was loaded into a
quartz reactor and dried in H at 375 mL/min for 3 h at RT. Subse-
was equipped with an in situ XPS reaction chamber. The spectra
were deconvoluted using nonlinear least squares method
3
a
2
employing a combination of Gaussian (80%) and Lorentzian
(20%) distributions and CasaXPS, a commercially available XPS
analysis program. Parameter constraints were imposed during
deconvolution of the Mo, Pt, and S spectra. The Mo 3d spectra
were fit using doublets with a splitting of 3.2 eV between the
3d5/2 and 3d3/2 peaks and an intensity ratio of 3:2. The Pt 4f
spectra were fit using doublets with a splitting of 3.3 eV be-
tween the 4f7/2 and 4f5/2 peaks and an intensity ratio of 4:3.
The S 2p spectra were fit using doublets with a splitting of
1.2 eV between the 2p3/2 and 2p1/2 peaks and an intensity ratio
of 2:1. For all spectra, the peak widths (FWHM) for the doublets
were constrained to be similar. Shirley backgrounds were used
for the Mo 3d, Pt 4f, and S 2p spectra, while a linear background
was used for the C 1s and O 1s spectra. The peak areas were
normalized using the appropriate atomic sensitivity factors. This
allowed comparison of the relative atomic fractions of each spe-
cies on the catalyst surfaces.
4
2
3
2
2
2
2
2
2
6
2
2
quently, the temperature was increased to 110 °C in ꢀ1 h and held
there for 2 h. The temperature was then increased to 450 °C at a
rate of 340 °C/h and held for 4 h. Finally, the material was
quenched to room temperature and passivated in a 1% O
2
/He mix-
ture at 20 mL/min for at least 5 h.
2.3. Reaction rate measurements
2.2. Characterization
The WGS rates were measured using a 4 mm I.D. quartz U-
X-ray diffraction analysis was performed using a Rigaku Mini-
flex Diffractometer with Cu radiation and Ni filter
k = 1.540 Å). The range (10° < 2h < 90°) was scanned at a rate
of 5°/min with a 0.02° step size. The BET surface areas were
determined via N physisorption using a Micromeritics ASAP
010 analyzer. Prior to these measurements, the catalysts were
degassed at 300 °C for 4 h. Pulse chemisorption experiments
were performed using a Micromeritics AutoChem 2910 Chemi-
sorption Analyzer equipped with a thermal conductivity detector
tube in which 20–30 mg of catalyst was supported on a quartz
wool plug. As necessary, the catalysts were diluted with inert,
low surface area (<1 m /g) silica to prevent channeling, avoid
problems with axial dispersion, and minimize temperature gradi-
ents in the bed. Prior to the reaction rate measurements, the cat-
Ka
a
2
(
2
2
alysts were pretreated at 590 °C for 4 h in a mixture of 15% CH
[13]. The effluent gas was passed through an ice-bathed con-
denser to remove most of the H O, and the composition was
4
/
H
2
2
analyzed using a SRI 8610C gas chromatograph (GC) equipped
with a Supelco Carboxen-1000 column and a thermal conductiv-
ity detector. The GC sampled the effluent gas every 30 min. To
ensure that there was no residual sulfur, the reactor system
and a mass spectrometer. Prior to analysis, the Mo
Mo C catalysts were pretreated in 15% CH /H for 4 h at 590 °C
13]. The catalysts were then degassed in He for 1 h. After cool-
ing to RT, the catalysts were repeatedly dosed with 5 mL of 5%
CO/He until saturation was achieved. The Cu/Zn/Al catalyst
was reduced in 5% H /Ar at 200 °C until complete reduction of
2
C and Pt/
2
4
2
[
2
(reactor and lines) was heated to 500 °C in flowing H for 12 h
2
O
3
in between runs. This treatment was sufficient to reproduce
the sulfur-free WGS rate for a commercial Cu–Zn–Al benchmark
catalyst.
2
Cu was achieved. The sample was then degassed in flowing He
for 1 h and cooled to 60 °C. The gas flow was then switched to
The rates were measured under differential conditions (con-
version 610%) at atmospheric pressure over a temperature range
of 200–240 °C. The equilibrium WGS conversion under condi-
tions used in the experiments would be greater than 90%. The
dry reactant simulated the effluent stream from a partial oxida-
N
2
O for 1.5 h. The N
the sample was cooled to RT. The N
by performing a H temperature-programmed reduction and
measuring the H consumption.
2
O flow was then replaced with He, and
2
O uptake was determined
2
2
The catalyst compositions were determined by inductively cou-
pled plasma optical emission spectroscopy (ICP–OES) using a Var-
ian 720-ES analyzer. The materials were dissolved in aqua regia (3
tion reformer and consisted of 13% CO, 56% H
ance N . The dry reactant was passed through a H
maintained at 69.4 ± 0.4 °C resulting in a wet reactant containing
2
, 8% CO
2
, and bal-
2
2
O saturator
parts HCl–1 part HNO
3
) and the emission spectra of dissolved spe-
30% H
2
O. The flow rate of the reactant was 262 mL/min corre-
ꢂ1
cies were compared to those for a series of standard solutions of
known concentrations. The surface morphologies of the fresh and
spent catalysts were characterized using scanning electron micros-
copy (SEM) with a Phillips XL30 FEG SEM operating at an acceler-
ating voltage of 15–25 kV and a nominal resolution of 2–5 nm.
Prior to analysis, the materials were sputter coated with Pd–Au
to mitigate charging effects. Sulfur adsorption/incorporation
experiments were performed using the microbalance on a TA
Instruments Q50 Thermogravimetric Analyzer. These experiments
sponding to a gas hourly space velocity of 125,000 h . Hydrogen
production (CO consumption) rates and conversions were deter-
mined by monitoring the concentration of CO in the product
stream. Features attributable to H
matograms but could not be quantified because He was used
as the carrier gas. No products other than CO
served in the effluent.
For measurements including sulfur, the catalysts were ini-
tially allowed to achieve pseudo steady-state rates at 240 °C in
2
were present in the chro-
2 2
and H were ob-
characterized interactions of the catalyst with H
2
S. The flow rate to
2
the sulfur-free reactant stream. Subsequently, H S was added
catalyst ratio and temperatures were similar to those used during
the reaction rate measurements.
to the reactant stream to produce concentrations ranging up
to 50 ppmv. In order to maintain the same gas hourly space
The fresh and spent catalysts were characterized using X-ray
photoelectron spectroscopy (XPS) to determine the compositions
and oxidation states of species on the surfaces. The XPS experi-
ments were performed using a Kratos Axis Ultra X-ray photo-
velocity, the N
for the addition of H
investigated by treating the spent materials in 15% CH
4 h at 590 °C. After this treatment, the catalysts were again ex-
posed to the sulfur-free feed at 240 °C to determine the recov-
ered rate.
2
balance gas flow rate was adjusted to account
S. The regenerability of the catalysts was
/H for
2
4
2
electron spectrometer with an Al anode (K
a
radiation at
1
486.6 eV) operating at 10 mA and 14 kV. The spectrometer

J.A. Schaidle et al. / Journal of Catalysis 272 (2010) 235–245
237
3
. Results
3.1. Pre-reaction characterization
Diffraction patterns for the Mo
2
2
C and Pt/Mo C catalysts (Fig. 1)
contained peaks for b-Mo C [20] and a-MoC1ꢂx [21]. The relative
2
peak areas suggested similar amounts of each. These phases have
Mo:C ratios near two; therefore, we will refer to this material as
Mo
cating that synthesis achieved complete bulk carburization. There
were no clearly discernable Pt peaks for the Pt/Mo C catalyst
24], indicating that crystallites, if present, were below the detec-
2 2 3
C. No peaks were observed for MoO [22] or MoO [23], indi-
2
[
tion limit of the X-ray diffractometer. A detailed description of
the Pt dispersion is beyond the scope of this work, but will be ex-
plored in a subsequent paper. Results from catalyst characteriza-
tion are given in Table 1. The reduced surface area for the Pt/
Mo
blocking by Pt nanoparticles. Interestingly, the addition of Pt de-
creased the CO uptake when compared to Mo C.
2 2
C catalyst compared to the Mo C catalyst may be due to pore
2
The in situ XPS results tracked effects of the various treatments
on the catalyst surface chemistries without exposure to air. Follow-
ing treatment, the samples were purged with N then cooled to
2
room temperature in the in situ XPS reaction chamber. The pres-
ꢂ8
sure was reduced to <10 Torr, and the samples were transferred
into the analysis chamber. Results from the deconvolution of spec-
tra for the as-synthesized and pretreated (15% CH
/H
4 2
at 590 °C for
2 2
Fig. 1. X-ray diffraction patterns for the (a) Pt/Mo C and (b) Mo C catalysts, and
peak positions for polycrystalline (c) b-Mo
2
C [20], (d) a-MoC1ꢂx [21], and (e) Pt [24]
reference materials.
Table 1
2
BET surface areas, N O uptakes, CO uptakes, site densities, Pt loadings, and TOFs for
Mo
2
C, Pt/Mo
2
C, Cu/Zn/Al
2
O
3
, and Pt/oxide catalysts.
Catalyst
BET
surface
N
2
O
CO
Site
uptake density
Pt
TOF
a
ꢂ1
uptake
mol/
loading (s
(wt.%)
)
area
(l
(lmol/ (sites/
g)
2
2
ꢄ 1018)
(
m /g)
g)
m
c
Mo
Pt/Mo
Cu/Zn/Al
Pt/Al
Pt/ZrO
Pt/TiO
2
C
98
70
60
180
75
74
–
–
192
–
–
–
268
151
–
91
56
9
1.65
1.30
1.93
0.31
0.45
0.07
–
3.7
–
3
3
0.08
b
c
c
2
C
0.80
0.32
2
O
[25]
3
2
O
3
0.03d
d
2
[25]
[25]
0.20
0.82d
2
3
a
Determined from uptake and BET surface area.
b
Determined using ICP–OES. Corresponds to a surface coverage of 12%, assuming
2
1
0 Pt atoms/nm .
c
2 2
Based on rates measured at 240 °C. Extrapolated TOFs for the Mo C, Pt/Mo C,
ꢂ1
ꢂ1
ꢂ1
and Cu/Zn/Al
2
O
3
catalysts at 250 °C are 0.11 s , 1.02 s , and 0.44 s , respectively.
d
Based on rates measured at 250 °C.

2
38
J.A. Schaidle et al. / Journal of Catalysis 272 (2010) 235–245
4
h) catalysts are summarized in Table 2. Peaks that accounted for
less than 10% of the total spectral area typically did not contribute
significantly to the goodness of fit. Surfaces of the catalysts con-
tained varying concentrations of species attributable to Mo
and Pt oxides, and carbon oxides; no chlorine residue (from the
Pt precursor) was observed for the Pt/Mo C catalyst. The C 1s spec-
tra for the Mo C and Pt/Mo C catalysts are shown in Fig. 2. Peaks
2
C, Mo
2
2
2
centered at 284.8 eV corresponded to adventitious carbon and
were used to reference the other binding energies. In addition to
adventitious carbon, surfaces of the Mo
contained carbidic carbon and adsorbed carbon oxides. Peaks at
83.5 ± 0.1 eV were assigned to carbidic carbon in Mo C [26]. The
2 2
C and Pt/Mo C catalysts
2
2
peaks at 286.6 ± 0.6 eV and 288.4 ± 0.4 eV were assigned to species
containing CAO and C@O bonds, respectively [27,28]. These could
be associated with carbonates and/or formates on the catalyst
surface.
2 2
The Mo 3d spectra for the Mo C and Pt/Mo C catalysts con-
tained four doublets (see Fig. 2). Doublets with Mo 3d5/2 peaks at
2
32.3 ± 0.2 eV are characteristic of Mo6 and suggested the pres-
ence of MoO [2,27,29,30]. The peaks at 229.7 ± 0.2 eV were likely
due to Mo in MoO [27,30,31]. The peaks at 228.5 ± 0.1 eV were
assigned to Mo in Mo C, based on comparisons with spectra
reported in the literature [2,26,32,33]. For most of the materials,
the ratio of the normalized area for the 1s peak at
83.5 ± 0.1 eV to the normalized area for the Mo 3d5/2 peak at
28.5 ± 0.1 eV was consistent with the presence of Mo C (see
+
3
4
+
2
2
+
2
Fig. 3. Pt 4f XPS spectra for the (a) as-synthesized and (b) pretreated Pt/Mo
2
C
catalyst. The catalyst was pretreated at 590 °C for 4 h in a mixture of 15% CH /H .
4
2
C
2
2
Table 3
Selected atomic ratios for species on surfaces of the as-synthesized and pretreated
Mo C and Pt/Mo C catalysts.
2
Table 3). The Mo 3d5/2 peak at 228.9 ± 0.1 eV was designated as
Mo (2 < d < 4), perhaps in an oxycarbide [30,34].
d+
2
2
Catalyst
Mo2C
Treatment
Ca/Mob ratio
O/Mo ratio
O /Pt ratio
c
d
The O 1s spectra for the Mo
2
C and Pt/Mo
2
C catalysts are also
shown in Fig. 2. Two different species were present on the Mo
2
C
1.9
0.7
0.5
0.5
2.6
1.5
1.9
1.2
–
–
e
surfaces. Peaks at 530.7 ± 0.2 eV are typically assigned to oxygen
in Mo oxides [26,27,29,30]. Peaks at 532.4 ± 0.2 eV are indicative
2
Mo C
CH
4
/H
/H
2
Pt/Mo
Pt/Mo
2
C
C
0.8
0.4
e
ꢂ
ꢂ
2
CH
4
2
of strongly bound O , OH , H
the Pt/Mo C catalyst contained an additional peak at 531.7 ±
.2 eV corresponding to Pt oxide [35,36].
Two doublets were observed in the Pt 4f XPS spectra (Fig. 3).
2
O and/or O@C [28,30]. Spectra for
a
b
c
C 1s peak corresponding to carbidic carbon (283.5 ± 0.1 eV).
Mo 3d doublet corresponding to Mo (228.5 ± 0.1 eV).
O 1s peak corresponding to PtO (531.7 ± 0.2 eV).
Pt doublet corresponding to PtO (73.0 ± 0.3 eV).
4 2 2
Pretreated in 15% CH /H at 590 °C for 4 h, purged with N , then cooled to room
2
2+
0
d
e
The dominant doublet with Pt 4f7/2 peak at 71.7 ± 0.1 eV was as-
0
signed to Pt [36,37]. The minor component with Pt 4f7/2 peak at
temperature.
+
7
3.0 ± 0.3 eV was assigned to Pt2 , most likely in the form of PtO
[
37].
Pretreatment of the as-synthesized catalysts in 15% CH
90 °C for 4 h caused a significant change in the surface chemistry.
The percentage of Mo in the form of Mo C increased from 15 to 20%
/H
4 2
at
tion and caused the formation of a small amount of a species with
Mo 3d5/2 peaks at 231.0 ± 0.1 eV (see Fig. 2). Peaks with similar
binding energies have been assigned to Mo [29–31]. Pretreat-
ment also increased the relative amount of carbidic carbon at the
surface and decreased the amount of oxygen and carbon oxides
5
5
+
2
in the as-synthesized catalysts to more than 50% in the pretreated
catalysts. Pretreatment completely reduced the Mo6 concentra-
+
Fig. 2. C 1s, Mo 3d, and O 1s XPS spectra for the (a) as-synthesized Mo
Mo C catalyst. The catalysts were pretreated at 590 °C for 4 h in a mixture of 15% CH
2
2
C catalyst, (b) as-synthesized Pt/Mo
/H
2 2
C catalyst, (c) pretreated Mo C catalyst, and (d) pretreated Pt/
4
2
.

J.A. Schaidle et al. / Journal of Catalysis 272 (2010) 235–245
239
(
Fig. 2). Approximately 70% of Mo on the as-synthesized Mo
face was in the form of MoO and MoO , which is in good agree-
ment with an O/Mo ratio of 2.6 (Table 3). The O/Mo ratio for the
as-synthesized Pt/Mo C catalyst was 1.9. Pretreatment at 590 °C
in 15% CH /H resulted in a substantial reduction in the O/Mo ratio
to 1.5 for Mo C and 1.2 for Pt/Mo C. While pretreatment reduced
the overall oxygen to molybdenum ratio for both catalysts, the
presence of Pt on Pt/Mo C seemed to facilitate the reduction of
Mo oxides to a greater extent than Mo C alone. There was only a
2
C sur-
(TOS). To probe the nature of this deactivation, the activity decay
was fit to models of the form [38]:
3
2
da
dt
2
m
ꢂ
¼ k
d
aðtÞ
ð2Þ
4
2
2
2
where a(t) is the ratio of the rate at time t to the initial rate, k
d
is the
specific decay constant, and t is time on stream (TOS). The best-fit
for the Mo C and Pt/Mo C catalysts was obtained using the recipro-
2
2
2
2
2
+
cal power form (Tables 4 and 5). This form is consistent with
deactivation by carbon deposition [39,40].
slight reduction in the percentage of Pt after pretreatment with
5% CH /H at 590 °C.
1
4
2
Rates for the Mo
exposure to sulfur. After the introduction of 5 ppm H
gen production rate for the Mo
C catalyst decreased by ꢀ90% with-
in 10 min. After ꢀ32 h on stream, the catalyst regained some of its
activity, reaching a H
production rate that was ꢀ25% of its sulfur-
free reactant steady-state rate. These temporal trends were repro-
ducible. When H S was removed from the reactant, the rate for the
C catalyst quickly decreased to zero. Upon treating the spent
C catalyst with 15% CH /H at 590 °C for 4 h, 25–30% of its ini-
2 2
C and Pt/Mo C catalysts also decreased on
2
S, the hydro-
3
.2. Reaction rates
2
The WGS rates for the Pt/Mo
2
C catalyst were almost an order of
C catalyst and 2–3 times
catalyst in the absence of
S (Fig. 4). The apparent activation energies for the Mo C, Pt/
Mo C, and Cu/Zn/Al catalysts were 59, 42, and 47 kJ/mol,
respectively. These results are consistent with those reported pre-
viously [12,13,18]. Turnover frequencies for the Pt/Mo C catalyst
determined using CO uptakes were higher than those for the Cu/
Zn/Al catalyst determined using the N O update, and similar
2
magnitude higher than those for the Mo
higher than those for the Cu/Zn/Al
2
2 3
O
2
H
2
2
Mo
Mo
2
2
2 3
O
2
4
2
tial rate was recovered.
General temporal trends for the Pt/Mo
to those for the Mo C catalyst, although the deactivation rate for
the Pt/Mo C catalyst was much slower and occurred over a per-
iod of several hours. The Pt/Mo C catalyst did not regain any of
its lost activity after removal of H S from the reactant, and treat-
ment in 15% CH /H at 590 °C for 4 h resulted in a very slight
reactivation of the catalyst to a rate similar to that for the
Mo C catalyst.
To better understand the nature of interactions between sulfur
and the Pt/Mo C catalyst, the H S concentration in the reactant
was varied. The deactivation rates were strong functions of the sul-
fur concentration. The ratio of the rate following exposure to H S to
the rate just before exposure to H S, a (t), is plotted in Fig. 6 as a
function of TOS with 5, 25, and 50 ppm H S in the reactant. The
activity decay was again fit to models of the form [38]:
2
2
C catalyst were similar
2
2
O
3
2
2
to those reported for the most active Pt-based catalysts (see
Table 1).
2
2
The Mo
0–15 h of exposure to sulfur-free reformate. Fig. 5 shows the rates
for Mo C and Pt/Mo C at 240 °C as a function of time on stream
2 2
C and Pt/Mo C catalysts deactivated during the first
4
2
1
2
2
2
2
2
2
2
s
2
da
dt
s
n
H
m
ꢂ
¼ kd;s
C
2S;0a
s
ðtÞ
ð3Þ
where kd,s is the specific decay constant due to sulfur exposure, t is
the TOS after sulfur introduction, and CH2S,0 is the concentration of
H S in the reactant (assumed to be constant). The best-fit was
2
achieved for m = 1 (exponential form) and n = 0.51 ± 0.05, although
the hyperbolic form was also a good fit for the data. The fit param-
eters for each of the models are included in Table 6. The exponential
form is typical of deactivation caused by poisoning [41], while the
hyperbolic form often indicates deactivation by sintering [38]. The
Fig. 4. Hydrogen production rates during WGS for the Mo
at 200–240 °C. The H production rates for a commercial Cu/Zn/Al
also illustrated. The reformate feed consisted of 9% CO, 30% H O, 6% CO
2
C and Pt/Mo
catalyst are
2
, 39% H ,
2
C catalysts
2
half-order dependence on concentration also suggests that H S dis-
sociated on the Pt/Mo C catalyst surface [41].
2
2
2 3
O
2
2
2
and 16% N .
Table 4
Table 5
Results from nonlinear regression of sulfur-free activity data for the Mo
four empirical decay rate laws.
2
C catalyst to
Results from nonlinear regression of sulfur-free activity data for the Pt/Mo
to four empirical decay rate laws.
2
C catalyst
Type
Linear
Exponential Hyperbolic Reciprocal
Type
Linear
Exponential
Hyperbolic Reciprocal
power
power
Differential
form
ꢂ
da ¼ kd
ꢂ
da ¼ kda
ꢂ
da ¼ kda
2
da
1=5
0
am
Differential
form
ꢂ
da
¼ k
d
ꢂ
da
¼ k
d
a
ꢂ
da
dt
¼ k
d
a2
ꢂ
da
¼ k A am
1=5
0
ꢂ
dt ¼ k
d
A
dt
dt
dt
d
dt
dt
dt
Integral form
ꢂ1
a ¼ 1 ꢂ k
0.06 ± 0.01 0.18 ± 0.05
d
t
a ¼ 1
1
þk
a ¼ A tꢂk
d
Integral form a ¼ 1 ꢂ k
t
a ¼ eꢂk t
d
a ¼
1
a ¼ A
0
t
ꢂk
d
a ¼ eꢂk
d
t
d
t
0
1þk
d
t
d
kd (hꢂ1
k
d
(h
)
0.41 ± 0.08 0.20 ± 0.02
)
0.045 ± 0.006 0.066 ± 0.009 0.10 ± 0.05 0.136 ± 0.007
A
0
2
–
–
–
0.46 ± 0.03
0.996
A
0
2
–
–
–
0.75 ± 0.02
0.999
0.371
0.732
0.893
R
0.952
0.970
0.982
R
adj
adj
a
F
a
573
926
1570
60,300
F
18.3
84.7
260
4140
a
a
Calculated by dividing the mean square model by the mean square error. P
value for all models was <0.0001.
Calculated by dividing the mean square model by the mean square error. P
value for all models was <0.0002.

2
40
J.A. Schaidle et al. / Journal of Catalysis 272 (2010) 235–245
Fig. 5. Hydrogen production rates for the Mo
free reformate after treatment of the catalyst at 590 °C for 4 h in a mixture of 15% CH
2
C and Pt/Mo
2
C catalysts with (A) sulfur-free reformate, (B) reformate with 5 ppm H
2
S, (C) sulfur-free reformate, and (D) sulfur-
O, 6% CO , 39% H , and 16% N
4
/H . The reformate contained 9% CO, 30% H
2
2
2
2
2
.
2
pretreated material, the C 1s spectrum (Fig. 7) for the Mo C cata-
lyst presented peaks associated with C-O and C@O, in addition to
a prominent carbidic carbon peak. The resulting Mo spectra for this
catalyst were similar to that for the pretreated catalyst; however,
the oxygen spectra were different. In addition to peaks at
5
30.7 ± 0.2 and 532.4 ± 0.2 eV (also observed for the pretreated
materials), a peak at 533.7 ± 0.2 eV was observed (Table 7). This
peak is believed to correspond to oxygen in adsorbed carbon oxi-
des [24]. Under reaction conditions, the O/Mo ratio for the Mo C
2
catalyst was 5.7, compared to a O/Mo ratio of 1.5 for the pretreated
material. This increase in the O/Mo ratio was due to an increase in
ꢂ
ꢂ
the density of adsorbed O , OH , and/or H
2
O, and not to an in-
crease in Mo oxides.
For the Pt/Mo
bled those after pretreatment, except for an additional C 1s peak
that was attributed to C@O. As observed for the Mo C catalyst,
2
C catalyst, the C and Mo spectra (Fig. 7) resem-
2
the oxygen spectra included an additional peak at 533.7 ± 0.2 eV
that was attributed to adsorbed carbon oxides. The O/Mo ratio
Fig. 6. Activity, a
the introduction of 5, 25, and 50 ppm H
s
(t), for the Pt/Mo
2
C catalyst as a function of time on stream after
S to the reformate.
for the Pt/Mo
2
C catalyst increased to 4.2 under reaction conditions
2
from a value of 1.2 after pretreatment. This increase was primarily
due to increases in the concentrations of O , OH , and/or H
ꢂ
ꢂ
2
O.
Temporal changes in the catalyst weight during exposure to H
2
S
3
.3. In situ characterization
provided insight regarding the nature of interactions between the
catalysts and sulfur. Prior to these measurements, the catalysts
were pretreated at 590 °C for 4 h in the CH /H mixture, then de-
gassed in He at 240 °C for 1 h. Changes in weight were assumed
to be due to the adsorption onto and/or incorporation of sulfur into
Prior to collection of the in situ XPS spectra, the materials were
pretreated in 15% CH /H at 590 °C for 4 h, exposed to a reformate
containing 9% CO, 30% H O, 6% CO , 39% H in N at 240 °C for 4 h,
4
2
4
2
2
2
2
2
purged with N
2
, then cooled to room temperature in the XPS reac-
the catalysts. During the first 10 min at 240 °C in a 5 ppm H S/He
mixture, there was a rapid weight gain (Fig. 8). Subsequently, the
rate of weight gain decreased, but remained somewhat steady until
2
ꢂ8
tion chamber. After reducing the pressure to <10 Torr, the mate-
rial was transferred into the analysis chamber. Compared to the
Table 6
2
Results from nonlinear regression of activity data for the Pt/Mo C catalyst to four empirical decay rate laws. The WGS rates were measured using reformate containing 5, 25, and
5
2
0 ppm H S.
Type
Linear
Exponential
Hyperbolic
Reciprocal power
n
da
s
dt
n
da
s
dt
n
2
da
s
dt
n
1=5
0
m
s
Differential form
Integral form
ꢂ
da
dt
s
¼ kd;s
C
ꢂ
¼ k
C
as
ꢂ
¼ kd;sC S;0a
H
2
S;0
d;s
H
2
S;0
H
2
s
ꢂ
¼ kd;s
C
ꢂkd;s
H
2
S;0A
a
n
as ¼ eꢂk
C
n
t
1
C
H S;0
2
n
as ¼ 1 ꢂ kd;s
C
S;0t
d;s
H
2
S;0
as ¼ 1þkd;s
n
H
2
C
H2S;0
t
as ¼ A0t
n
0.38 ± 0.06
0.12 ± 0.02
–
0.51 ± 0.05
0.25 ± 0.03
–
0.6 ± 0.1
0.5 ± 0.1
–
0.1 ± 0.3
0.6 ± 0.5
0.31 ± 0.04
0.737
ꢂn ꢂ1
kd,s (ppm
h )
A
0
2
adj
R
0.781
0.986
0.965
a
F
39.7
777
304
18.4
a
Calculated by dividing the mean square model by the mean square error. P value for all models was <0.0001.

J.A. Schaidle et al. / Journal of Catalysis 272 (2010) 235–245
241
Fig. 7. C 1s, Mo 3d, and O 1s XPS spectra for the (a) Mo
2
C and (b) Pt/Mo
2
C catalysts following pretreatment in 15% CH
4
/H
2
at 590 °C for 4 h and exposure to reformate
containing 9% CO, 30% H O, 6% CO , 39% H , and 16% N at 240 °C in the in situ XPS reaction chamber.
2 2 2 2
Table 7
2 2 4 2
Binding energies from in situ spectra for species on surfaces of the Mo C and Pt/Mo C catalysts following pretreatment in 15% CH /H at 590 °C for 4 h, exposure to reformate
containing 9% CO, 30% H O, 6% CO , 39% H in N at 240 °C for 4 h, purge with N
2
2
2
2
2
, then cooling to room temperature.
Catalyst
C 1s (eV)a,b
Mo
Mo 3d5/2 (eV)a
Mo2+
Pt 4f7/2 (eV)a
O 1s (eV)a
MoO
Mod+
Mo4+
Pt0
Pt2+
ꢂ
ꢂ
OAC
2
C
CAO
x
2
O , OH , H O, O@C
Mo
Pt/Mo
2
C
283.1 (18)
283.2 (11)
286.5 (14)
286.5 (16)
228.3 (55)
228.2 (69)
228.7 (23)
228.7 (16)
229.7 (20)
229.9 (15)
–
–
530.7 (12)
530.5 (14)
532.3 (68)
532.3 (40)
533.7 (20)
533.9 (36)
2
C
71.6 (83)
73.3 (17)
a
The number in parentheses represents the atomic percentage.
Balance of atomic percentages for C 1s is adventitious carbon.
b
reaching a weight gain equivalent to ꢀ2 monolayers (ML) of sulfur,
and H
ꢂ32 kJ/mol at 240 °C.
2 f
S is also thermodynamically favorable with a DG of
2
based on a site density of 10 sites/nm and 1 sulfur atom/site. After
the adsorption or incorporation of ꢀ2 ML of sulfur, the rate of
weight gain slowed then leveled off. The overall rate of sulfur up-
3.4. Post-reaction ex situ characterization
2 2
take was slightly higher for the Pt/Mo C catalyst than for the Mo C
catalyst, and shapes of the curves were different suggesting that
the presence of Pt caused a change in the mechanism for sulfur
incorporation.
2 2
Bulk crystalline structures for the Mo C and Pt/Mo C catalysts
did not change on exposure to reformate without or with 5 ppm
Hydrogen sulfide readily dissociates on Mo
ture [42], and the conversion of Mo C and H S to MoS
dynamically favorable with a
of ꢂ264 kJ/mol at 240 °C [43].
Furthermore, it has been reported that H S dissociates on Pt at
temperatures as low as 100 °C [44]. The formation of PtS from Pt
2
C at room tempera-
2
2
2
is thermo-
DG
f
2
Fig. 9. X-ray diffraction patterns for the (a) Mo
b) Pt/Mo C catalyst after WGS without H S, (c) Mo
ppm H S, and (d) Pt/Mo C catalyst after WGS with 5 ppm H
polycrystalline (e) SiO [45], (f) MoS [46], (g) b-Mo C [20], and (h)
was used as a catalyst diluent
2
C catalyst after WGS without H
C catalyst after WGS with
S. Peak positions for
-MoC1ꢂx [21]
2
S,
(
5
2
2
2
2
2
2
Fig. 8. Weight gain during exposure of the Mo
2
C and Pt/Mo
2
C catalysts to 5 ppm
2
2
2
a
H
1
2
S in He at 240 °C. The dashed lines indicate the weight gains corresponding to
ML and 2 ML of sulfur coverage assuming a material with 98 m /g.
reference materials are also illustrated. The SiO
2
2
during the reaction rate measurements.

2
42
J.A. Schaidle et al. / Journal of Catalysis 272 (2010) 235–245
H
2
S at 240 °C, as evident from the XRD patterns (Fig. 9). The shar-
per peaks were due to the SiO that was used as a diluent during
the reaction rate measurements. The absence of a peak at
crystallites, if present, were below
ferred in air to a desiccator, and stored under vacuum until being
2
transferred to the spectrometer and collection of the ex situ spec-
tra. A comparison of the in situ (Fig. 7) and ex situ (Fig. 11) Mo spec-
tra for materials exposed to the reformate suggests that this brief
2h ꢀ 14° indicated that MoS
2
the detection limit of the X-ray diffractometer. Micrographs of
the fresh and spent catalysts (see for example Fig. 10) were very
similar, indicating that there were no significant changes in the
surface morphology. Although thermodynamically favorable,
exposure to air caused partial oxidation of Mo
MoO . As shown in Fig. 9, however, the catalysts did not undergo
bulk oxidation to MoO or MoO upon exposure to air. Neverthe-
2
C at the surface to
3
2
3
less, an examination of major changes in the ex situ spectra pro-
2
Mo C and Pt are reported to be moderately resistant to bulk sulfi-
vided insight regarding the significant effects of sulfur on the
dation, possibly because the diffusion of sulfur into the sub-surface
is slow due to its size [2,43,47,48]. Surface areas for the spent cat-
alysts both in sulfur-free and in sulfur-containing reactants were
2 2
Mo C and Pt/Mo C catalysts. Results from deconvolution of the
ex situ spectra are summarized in Table 8 for catalysts subjected
to the following treatments (also see Fig. 5):
ꢀ
5–15% lower than those for the fresh catalysts. For example, sur-
face areas for the spent Mo C catalysts were 86 ± 4 (without sulfur)
and 93 ± 4 m /g (with sulfur) compared to a surface area of
2
(A) sulfur-free reformate at 240 °C for 16 h;
(B) reformate with 5 ppm H S at 240 °C for 22 h;
2
(C) sulfur-free reformate at 240 °C for 5 h;
2
2
9
8 ± 5 m /g for the as-synthesized catalyst. These results confirm
2
that the deactivation caused by exposure to H S was not due to
surface area loss or sintering.
(D) sulfur-free reformate after treatment of the catalyst at
4 2
590 °C in CH /H for 4 h.
Due to the difficulty of ensuring the complete removal of sulfur,
experiments involving sulfur could not be carried out in the XPS
reaction system. Instead, catalysts exposed to reformate in the cat-
alytic reactor were cooled to room temperature, quickly trans-
With the exception of the S 2p spectra, spectra for the Mo
prior to and after exposure to sulfur were similar. Exposure to refor-
mate with 5 ppm H S resulted in a doublet with a S 2p3/2 peak at
2
C catalyst
2
2 2
Fig. 10. Scanning electron micrographs of the (a) as-synthesized and (b) spent (WGS with 5 ppm H S at 240 °C) Mo C catalysts. Images were collected at 15 kV accelerating
voltage, 3.0 spot size, and 18,000 magnification.
Fig. 11. Mo 3d XPS spectra for the Mo
2 2 2 2 2
C and Pt/Mo C catalysts (a) after WGS without H S, (b) after WGS with 5 ppm H S, and (c) after WGS with 5 ppm H S, treatment in 15%
CH /H at 590 °C for 4 h, and WGS without H
4
2
2
S.

J.A. Schaidle et al. / Journal of Catalysis 272 (2010) 235–245
243
1
63.1 ± 0.2 eV (Fig. 12). This binding energy is consistent with the
2ꢂ
presence of S2 in MoS
Treatment of the sulfur-deactivated Mo
at 590 °C for 4 h caused an increase in the relative amounts of
Mo that we attributed to Mo C and MoS [2,32] (Fig. 11) and emer-
3
or SH groups [2,49].
2
C catalyst in 15% CH
4
/
H
2
2
2
gence of a small doublet with Mo 3d5/2 peak at 227.5 ± 0.1 eV. This
0
doublet is consistent with the presence of Mo [50]. Three doublets
were resolved in the S 2p spectra (Fig. 12). The doublet with S 2p3/2
peak at 161.8 ± 0.1 eV has been assigned to atomic sulfur strongly
adsorbed to Mo [51,52]. The peak at 162.4 ± 0.1 eV corresponds to
2ꢂ
S
2
species, likely in the form of MoS [2,29,48]. The peak at
1
60.5 ± 0.2 eV has been tentatively attributed to a Mo sulfidocar-
bide [47,53,54] based on the position.
Exposure of the Pt/Mo C catalyst to sulfur significantly affected
the associated XPS spectra. Spectra following exposure to sulfur-
free reformate contained doublet with Pt 4f7/2 peak at
0.1 ± 0.4 eV (Fig. 13), in addition to peaks attributable to Pt and
2
a
0
7
PtO. This binding energy does not match those of any previously
0
reported Pt species, but the shift relative to Pt is consistent with
the presence of anionic Pt. Exposure to sulfur eliminated this
highly reduced Pt species. The relative amount of Mo
following exposure to reformate containing 5 ppm H S. Three dou-
blets were resolved in the S 2p spectra for the Pt/Mo C catalyst
after exposure to reformate with 5 ppm H S at 240 °C (Fig. 12).
2
C increased
2
2
2
The doublet with S 2p3/2 peak at 161.8 ± 0.1 eV corresponds to sul-
fur strongly adsorbed to Mo. The doublet with S 2p3/2 peak at
2
ꢂ
1
62.4 ± 0.1 eV corresponds to S species, possibly in the form of
MoS or PtS [2,29,48,55,56]. The doublet with S 2p3/2 peak at
68.8 ± 0.1 eV is consistent with the presence of sulfate species
2
1
6
+
(
S
) [32,57,58]. Oxygen in the sulfate might have been introduced
from the sub-surface or exposure to air. Treatment of the sulfur-
deactivated Pt/Mo C catalyst in 15% CH /H at 590 °C for 4 h
caused reemergence of peaks that we attributed to anionic Pt
and Mo sulfidocarbide, a slight decrease in the amount of Mo
and slight increase in the amount of MoS
2
4
2
2
C
2
.
4
. Discussion
Sulfur tolerance can be manifested in two ways. The most
attractive form of tolerance is when the catalyst maintains its rate
during exposure to sulfur. Another form is when the catalytic activ-
ity can be restored via modest treatment of the spent material. Re-
sults described in this paper indicate that the Mo
2
C and Pt/Mo
2
C
catalysts possessed some degree of sulfur tolerance. The Mo
2
C cat-
alyst was significantly deactivated during exposure to sulfur but
maintained a modest rate, and some of the initial performance
could be restored via treatment in CH
catalyst behaved in a similar manner with the exception that the
extent of reactivation was much lower than that for the Mo C cat-
4 2 2
/H mixtures. The Pt/Mo C
2
alyst. This discussion section will explore changes in the surface
chemistry that correlate with deactivation of these catalysts.
2 2
The Mo C and Pt/Mo C catalysts deactivated during the first
1
0–15 h of exposure to sulfur-free reformate (Fig. 5). Rate decay
equations suggested that the deactivation was caused by carbon
deposition. This result is in agreement with XPS findings. Compar-
ison of XPS spectra from pretreated and in-situ WGS exposed sam-
ples revealed an increase in carbon oxide groups (CAO and C@O) as
well as surface oxygen. Formation of carbonate or formate groups
may have caused blockage of active sites as has been reported for
Au/CeO
the number of active Mo
be the subject of a future paper.
2
Deactivation of the Mo C catalyst by H S appeared to occur in
2
catalysts [59]. Formation of surface oxides could reduce
2
C sites. This deactivation mechanism will
2
three stages. The reaction rate decreased significantly during the
first 10 min of exposure to sulfur. In a similar time frame, the sulfur

2
44
J.A. Schaidle et al. / Journal of Catalysis 272 (2010) 235–245
2 2 2 2 4 2
Fig. 12. S 2p XPS spectra for the Mo C and Pt/Mo C catalysts (a) after WGS with 5 ppm H S for 22 h and (b) after WGS with 5 ppm H S, treatment in 15% CH /H at 590 °C for
4
h, and WGS without H S.
2
ciated with MoS
2
nanoparticles produced via sulfidation of Mo
2
C,
2
and the regenerability of highly active Mo C-based sites poisoned
during initial exposure to sulfur.
2 2
In comparison to the Mo C catalyst, the Pt/Mo C catalyst exhib-
ited a much slower deactivation rate, suggesting that most of the
activity was associated with the presence of Pt. The results are con-
sistent with two types of sites on the Pt/Mo
Pt nanoparticles or at the Pt nanoparticle–Mo
and sites on the Mo C support. The high activity Pt-based sites
2
C catalyst: sites on the
2
C support interface,
2
were severely and irreversibly deactivated, we believe, as a conse-
quence of the conversion of Pt into inactive PtS. Sites associated
with the Mo
similar to that observed for the Mo
Finally, we note that surface oxygen may have played a role in
the interactions of sulfur with the Mo C and Pt/Mo C catalysts.
2
C support could be partially reactivated in a manner
2
C catalyst.
2
2
Oxygen has been reported to facilitate the sulfidation of early tran-
sition metal carbides and nitrides [29]. The O/Mo ratios for the
Mo
substantial amounts of MoS
The increased O/Mo ratio after exposure to WGS reactants was pri-
2
C and Pt/Mo
2
C catalysts were very high and each contained
2
after WGS in the presence of H S.
2
ꢂ
ꢂ
marily associated with the presence of strongly adsorbed O , OH ,
O@C, and/or H O. This oxygen may have facilitated the formation
of MoS . Future experiments will explore this behavior.
Fig. 13. Pt 4f XPS spectra for the Pt/Mo
after WGS with 5 ppm H S, and (c) after WGS with 5 ppm H
CH /H at 590 °C for 4 h, and WGS without H S.
2
C catalyst (a) after WGS without H
2
S, (b)
2
2
2
S, treatment in 15%
2
4
2
2
5
. Conclusions
adsorption and/or incorporation rate increased dramatically. We
believe these responses were interrelated and that highly active
The effects of H
sitions of Mo C and Pt/Mo
These catalysts were severely deactivated on exposure to H
could be partially regenerated. Characterization of the spent cata-
lysts suggested that deactivation of the Mo C catalyst was primar-
ily due to the adsorption of sulfur on Mo C sites and the formation
of surface MoS . The MoS sites were active in the presence of sul-
fur. Deactivation of the Pt/Mo C catalyst appeared to be primarily
due to the irreversible sulfidation of Pt nanoparticles. Other fea-
tures for this catalyst were similar to those for the Mo C catalyst.
Under reaction conditions, the Mo C and Pt/Mo C surfaces also
2
S on the WGS activities, structures, and compo-
C catalysts have been investigated.
S but
2
2
2
Mo C sites were quickly poisoned during this stage. Subsequently,
2
the WGS rate decreased and the sulfur content increased more
gradually. During the final stage, the activity increased slightly.
This increase in activity appeared to correspond with the adsorp-
2
2
tion and/or incorporation of ꢀ2 ML of sulfur on the Mo
surface. When sulfur was removed from the reactant, the activity
decreased suggesting that it was associated with MoS nanoparti-
cles. Molybdenum sulfide is known to be active for the WGS in the
presence of H S, and in fact, requires sulfur in the feed to maintain
its activity [60]. A fraction of the Mo C sites could be restored via
treatment in 15% CH /H at 590 °C for 4 h; however, based on
the XPS results, MoS sites persisted. Therefore, we have concluded
that the sulfur tolerance exhibited by the Mo C catalyst was asso-
2
C catalyst
2
2
2
2
2
2
2
2
2
possessed high concentrations of oxygen, which may have facili-
tated formation of the Mo sulfide. Although Pt improved the activ-
4
2
2
2
ity of the Mo C catalyst, it also altered the interaction of sulfur with
2
the catalyst surface, resulting in an increased susceptibility to

J.A. Schaidle et al. / Journal of Catalysis 272 (2010) 235–245
245
sulfur poisoning. These results suggest that the Mo
catalysts were partially tolerant to sulfur during WGS.
2
C and Pt/Mo
2
C
[26] T.P. St. Clair, S.T. Oyama, D.F. Cox, S. Otani, Y. Ishizawa, R.-L. Lo, K.-I. Fukui, Y.
Iwasawa, Surf. Sci. 426 (1999) 187.
[
27] P. Delporte, F. Meunier, C. Pham-Huu, P. Vennegues, M.J. Ledoux, J. Guille,
Catal. Today 23 (1995) 251.
[
[
[
[
[
28] M. Buo, J.M. Martin, T. Le Mogne, L. Vovelle, Appl. Surf. Sci. 47 (1991) 149.
29] Z.B. Zhaobin Wei, P. Grange, B. Delmon, Appl. Surf. Sci. 135 (1998) 107.
30] J.-G. Choi, L.T. Thompson, Appl. Surf. Sci. 93 (1996) 143.
31] A. Katrib, P. Leflaive, L. Hilaire, G. Maire, Catal. Lett. 38 (1996) 95.
32] A.S. Mamede, J.-M. Giraudon, A. Lofberg, L. Leclercq, G. Leclercq, Appl. Catal. A
Acknowledgments
This work was supported by funding from the Department of
Energy, the National Science Foundation, and the Hydrogen Energy
Technology Laboratory at the University of Michigan.
227 (2002) 73.
[33] M.J. Ledoux, C.P. Huu, J. Guille, H. Dunlop, J. Catal. 134 (1992) 383.
[
34] P. Perez-Romo, C. Potvin, J.-M. Manoli, M.M. Chehimi, G. Djega-Mariadassou, J.
Catal. 208 (2002) 187.
References
[35] J.L.G. Fierro, J.M. Palacios, F. Tomas, Surf. Interface Anal. 13 (1988) 25.
[
[
36] J.P. Contour, G. Mouvier, M. Hoogewys, C. Leclere, J. Catal. 48 (1977) 217.
37] G.M. Bancroft, I. Adams, L.L. Coatsworth, C.D. Bennewitz, J.D. Brown, W.D.
Westwood, Anal. Chem. 47 (1975) 586.
[
1] J.M. Robinson, S.R. Barrett, K. Nhoy, R.K. Pandey, J. Phillips, O.M. Ramirez, R.I.
Rodriguez, Energy Fuels 23 (2009) 2235.
[
38] H.S. Fogler, Elements of Chemical Reaction Engineering, third ed., Prentice Hall,
New Jersey, 1999. p. 643.
[
[
2] P.K. Cheekatamarla, W.J. Thomson, Appl. Catal. A 287 (2005) 176.
3] P. Da Costa, J.-L. Lemberton, C. Potvin, J.-M. Manoli, G. Perot, M. Breysse, D.
Djega-Mariadassou, Catal. Today 65 (2001) 195.
[
[
[
39] A. Voorhies Jr., Ind. Eng. Chem. 37 (1945) 318.
40] C.G. Rudershausen, C.C. Watson, Chem. Eng. Sci. 3 (1954) 110.
41] C.H. Bartholomew, R.J. Farrauto, Fundamentals of Industrial Catalytic
Processes, second ed., John Wiley and Sons, Berlin, 2006.
[
4] P.A. Aegerter, W.W.C. Quigley, G.J. Simpson, D.D. Ziegler, J.W. Logan, K.R.
McCrea, S. Glazier, M.E. Bussell, J. Catal. 164 (1996) 109.
[
[
5] M.K. Neylon, S.K. Bej, C.A. Bennett, L.T. Thompson, Appl. Catal. A 232 (2002) 13.
6] F.H. Ribeiro, R.A.D. Betta, M. Boudart, J. Baumgartner, E. Iglesia, J. Catal. 130
[
[
[
[
42] J.A. Rodriguez, J. Dvorak, T. Jirsak, J. Phys. Chem. B 104 (2000) 11515.
43] E. Furimsky, Appl. Catal. A 240 (2003) 1.
44] M.V. Mathieu, M. Primet, Appl. Catal. 9 (1984) 361.
45] V.I. Pakhomov, A.V. Goryunov, P.V. Pakhomov, N.T. Chibiskova, Zh. Neorg.
Khim. 38 (1993) 44.
(
1991) 86.
[
[
[
7] J.S. Lee, S. Locatelli, S.T. Oyama, M. Boudart, J. Catal. 125 (1990) 157.
8] G.S. Ranhotra, A.T. Bell, J.A. Reimer, J. Catal. 108 (1987) 40.
9] M.B. Zellner, J.G. Chen, Catal. Today 99 (2005) 299.
[
46] V. Petkov, S.J.L. Billinge, P. Larson, S.D. Mahanti, T. Vogt, K.K. Rangan, M.G.
Kanatzidis, Phys. Rev. B: Condens. Matter Mater. Phys. 65 (2002) 1.
47] E.J. Markel, J.W. Van Zee, J. Catal. 126 (1990) 643.
48] Z. Paal, P. Tetenyi, M. Muhler, U. Wild, J.-M. Manoli, C. Potvin, J. Chem. Soc.,
Faraday Trans. 94 (1998) 459.
[
[
10] R. Venkataraman, H.R. Kunz, J.M. Fenton, J. Electrochem. Soc. 150 (2003) 278.
11] H. Zhong, H. Zhang, G. Liu, Y. Liang, J. Hu, B. Yi, Electrochem. Commun. 8 (2006)
[
[
7
07.
[
[
[
[
[
12] L.T. Thompson, S.K. Bej, J.J. Patt, C.H. Kim, US Patent 6897178 (2005).
13] T.E. King, Ph.D. Thesis, University of Michigan, 2007.
14] W. Setthapun, S. Bej, L.T. Thompson, Top. Catal. 49 (2008) 73.
15] Q. Ming, T. Healey, L. Allen, P. Irving, Catal. Today 77 (2002) 51.
16] Satterfield, Heterogeneous Catalysis in Industrial Practice, Second ed.,
McGraw-Hill, 1991. p. 445.
[
[
[
[
49] Th. Weber, J.C. Muijsers, J.W. Niemantsverdriet, J. Phys. Chem. 99 (1995) 9194.
50] J.E. De Vries, H.C. Yao, R.J. Baird, H.S. Gandhi, J. Catal. 84 (1983) 8.
51] J.T. Roberts, C.M. Friend, Surf. Sci. 202 (1988) 405.
52] J.A. Rodriguez, S.Y. Li, J. Hrbek, H.H. Huang, G.-Q. Xu, J. Phys. Chem. 100 (1996)
14476.
[
[
[
[
[
[
17] Y.-M. Wu, Oil Gas J. 92 (1994) 38.
18] J. Patt, Ph.D. Thesis, University of Michigan, 2003.
19] J. Patt, D.J. Moon, C. Phillips, L. Thompson, Catal. Lett. 65 (2000) 193.
20] S. Nagakura, M. Kikuchi, S. Oketani, Acta Crystallogr. 21 (1966) 1009.
21] J.C. Schuster, H. Nowotny, Monatsh. Chem. 110 (1979) 321.
22] M. Ghedira, C. Do-Dinh, M. Marezio, J. Mercier, J. Solid State Chem. 59 (1985)
[
[
[
[
[
53] P. Liu, J.A. Rodriguez, J.T. Muckerman, J. Mol. Catal. A: Chem. 239 (2005) 116.
54] S.P. Kelty, G. Berhault, R.R. Chianelli, Appl. Catal. A 322 (2007) 9.
55] M. Muhler, Z. Paal, Surf. Sci. Spectra 4 (1997) 125.
56] J. Dembowski, L. Marosi, M. Essig, Surf. Sci. Spectra 2 (1993) 104.
57] M.-D. Appay, J.-M. Manoli, C. Potvin, M. Muhler, U. Wild, O. Pozdnyakova, Z.
Paal, J. Catal. 222 (2004) 419.
1
59.
[
58] B.J. Lindberg, K. Hamrin, G. Johansson, U. Gelius, A. Fahlman, C. Nordling, K.
Siegbahn, Phys. Scripta 1 (1970) 286.
[
[
23] J.D. Hanawalt, H.W. Rinn, L.K. Frevel, Anal. Chem. 10 (1938) 457.
24] C.A. Vanderborgh, Y.K. Vohra, H. Xia, A.L. Ruoff, Phys. Rev. B: Condens. Matter
Mater. Phys. 41 (1990) 7338.
[
[
59] C.H. Kim, L.T. Thompson, J. Catal. 230 (2005) 66.
60] P. Hou, D. Meeker, H. Wise, J. Catal. 80 (1983) 280.
[
25] H. Iida, A. Igarashi, Appl. Catal. A 298 (2006) 152–160.