The role of Sn in Pt-Sn/CeO2 catalysts for the complete oxidation of ethanol

Article abstract of DOI:10.1016/j.molcata.2005.03.018

The structural features and catalytic properties of Pt-Sn/CeO₂ catalysts prepared by modified polyol method were extensively investigated for the complete oxidation of ethanol. CO chemisorption, TPR, DTA and XPS measurements identically indicated that the electronic configuration of Pt by Sn as well as the formation of PtSn alloy were the key factors in determining the nature of the active sites. A strong Pt/Sn atomic ratio dependence of catalytic performances was observed, which was explained in terms of the changes in the surface structure of metal phases and the electronic Pt-Sn interaction.

Full text of DOI:10.1016/j.molcata.2005.03.018

Journal of Molecular Catalysis A: Chemical 235 (2005) 122–129
The role of Sn in Pt–Sn/CeO catalysts for the complete
2
oxidation of ethanol
∗
Xiaolan Tang, Baocai Zhang, Yong Li, Yide Xu, Qin Xin, Wenjie Shen
State Key Laboratory of Catalysis, Dalian Institute of Chemical Physics, Chinese Academy of Sciences,
457 Zhongshan Road, P.O. Box 110, Dalian 116023, PR China
Received 16 November 2004; received in revised form 2 March 2005; accepted 21 March 2005
Abstract
ThestructuralfeaturesandcatalyticpropertiesofPt–Sn/CeO
2
catalystspreparedbymodifiedpolyolmethodwereextensivelyinvestigatedfor
the complete oxidation of ethanol. CO chemisorption, TPR, DTA and XPS measurements identically indicated that the electronic configuration
of Pt by Sn as well as the formation of PtSn alloy were the key factors in determining the nature of the active sites. A strong Pt/Sn atomic ratio
dependence of catalytic performances was observed, which was explained in terms of the changes in the surface structure of metal phases and
the electronic Pt–Sn interaction.
©
2005 Elsevier B.V. All rights reserved.
Keywords: Pt–Sn/CeO2; Pt–Sn alloy; Modified polyol method; Complete oxidation of ethanol
1
. Introduction
that the beneficial effects may be obtained by the addition of
promoters to the supported noble metal catalysts [7–9]. For
example, platinum–tin catalysts presented extremely high
catalytic activities at lower temperatures in the preferential
CO oxidation [7] and the oxidation of methanol [8]. Pt–Sn/C
as anode catalyst also exhibited higher electro-catalytic per-
formance than that of Pt/C catalyst due to the enhancement
of the cleavage of C C bond by the addition of Sn [10].
The role of tin is often attributed to the formation of Pt–Sn
alloy, which had better catalytic properties than Pt alone
[9–12], and to the presence of well-dispersed Pt particles
stabilized by tin [13]. Considering the recently reported
Pt/CeO2 catalysts, which attracted much attention with quite
promising catalytic activities in various reactions [15–17], it
is expectable that Pt–Sn/CeO2 may be potential catalyst for
the complete oxidation of ethanol at lower temperatures.
In the present study, Pt/CeO2 and Pt–Sn/CeO2 catalysts
with the Pt/Sn molar ratio ranging from 3:1 to 1:2 prepared
by modified polyol method, were investigated for the com-
plete oxidation of ethanol. The electronic effect of tin on plat-
inum in Pt–Sn/CeO2 catalysts was also explored by means of
TPR, XPS, CO chemisorption and DTA measurements. The
temperature for complete oxidation of ethanol to CO2 was
Ethanol has been regarded as promising alternative fuel
for vehicles due to its high energy density, less toxicity and
availability in large quantities from biomass [1]. However,
the undesirable partially oxidized products in complete
oxidation of ethanol, such as acetaldehyde, CO and acetic
ester, can lead to severe environment and health problems.
As a result, the development of efficient catalysts, which can
directly converting ethanol to CO2 and H2O at relatively low
temperatures, has been receiving renewed attention [1,2].
Supported noble metal catalysts were found to be highly
active for directly converting ethanol into CO2 with less ac-
etaldehyde formation [1–6]. For instance, 100% conversion
of ethanol was achieved over Ag/La0.6Sr0.4MnO3 catalyst
at 493 K [1], and that was obtained over Pt/Al2O3 even at
4
73 K [2]. Even though, the rapid application of ethanol
in energy aspects still calls for the development of more
efficient catalysts to realize the total oxidation of ethanol at
temperatures as low as possible. Recent studies have shown
∗
Corresponding author. Tel.: +86 411 84379085; fax: +86 411 84694447.
E-mail address: shen98@dicp.ac.cn (W. Shen).
1
381-1169/$ – see front matter © 2005 Elsevier B.V. All rights reserved.
doi:10.1016/j.molcata.2005.03.018

X. Tang et al. / Journal of Molecular Catalysis A: Chemical 235 (2005) 122–129
123
significantly reduced to 363 K that is about 90–120 K lower
than those of previously reported [1,2].
an accelerating voltage of 12.5 kV. The powder samples were
pressed into thin discs and mounted on a sample rod placed in
a pretreatment chamber, in which the catalysts were reduced
at desired temperatures by flowing hydrogen for 2 h. Then it
was transferred into the analysis chamber where the spectra
of Pt 4d, Sn 3d, Ce 3d and O 1s levels were recorded. Charg-
ing effects were corrected by adjusting the binding energy of
C 1s peak to a position of 284.6 eV.
2
. Experimental
2
.1. Catalyst preparation
The CeO2 support was prepared by precipitation of
DTA experiments were performed on a Shimadzu DT-
Ce(NO3)3·6H2O aqueous solution with the addition of
NH3·H2O solution until the pH value of the mixture was
greater than 9.0 under vigorous stirring at 333 K. The pre-
cipitant was then aged in the mother liquid for 4 h, followed
by filtration and washing with distilled water. The obtained
solid was dried at 373 K for 10 h, and calcined at 723 K for
2
3
0B thermoanalysis instrument in the temperature range
00–1073 K in N2. A temperature ramp of 10 K/min was
employed.
2.3. Activity measurements
4
h in air.
The catalytic activity tests were performed under at-
The metal precursors, chloroplatinic acid and tin(II) chlo-
mospheric pressure in a continuous-flow fixed-bed reactor.
Samples of 200 mg (40–60 mesh) were placed between two
layers of quartz wool inside a quartz tube (i.d. = 6 mm).
Prior to catalytic reactions, the catalysts were reduced with
ride, was dissolved into 60 mL ethylene glycol (EG) and
mixed with 3 g CeO2 powder. The pH value of the mixture
was then adjusted to 12 by the addition of 1 M NaOH solu-
tion, and maintained at 403 K for 1 h to ensure the complete
reduction of Pt. The resulting solid was filtered, washed with
10% H2/He at 473 K for 2 h. The oxidation of ethanol was
−
performed between 298 and 473 K. Liquid ethanol was
injected by a micro-feeder and mixed with O2/He mixture
hot distilled water until no Cl anion could be detected in
filtrate by AgNO3 solution. After drying at 373 K for 10 h,
the solid was calcined at 473 K for 2 h. The Pt loading for all
samples was 3% by weight, and the Pt/Sn molar ratio varied
from 3:1 to 1:2.
(
C2H OH/O2/He = 0.5/3/96.5 vol.%) with a gas hourly space
5
−
1
velocity (GHSV) of 40,000 h . Effluents from the reactor
were analyzed on-line by a HP6890 gas chromatography
equipped with TCD and FID detectors. A Haysep D packing
column was used to detect the CO, CO2, O2 and H2, and
an Innowax capillary column was employed to separate
oxygenates and hydrocarbons.
For comparison, monometallic 3 wt.% Pt/CeO2 and
5
wt.% Sn/CeO2 catalysts were prepared in the same way.
2
.2. Characterization
The specific surface areas (SBET) of the catalysts were cal-
culated from a multipoint Braunauer–Emmett–Teller (BET)
analysisofthenitrogenadsorptionisothermsat77 Krecorded
on a micrometrics ASAP 2000 instrument.
3. Results and discussion
3.1. Surface area and structure
The powder X-ray diffraction (XRD) patterns were
recorded using a Rigaku D/max-RB with Cu K␣ radiation
operated at 40 kV and 100 mA.
The BET surface area of the CeO2 support was measured
to be 103 m /g, and the deposition of platinum and tin
2
The CO chemisorption measurements were carried out
in CHEMBET 3000 adsorption instrument (QuantaChrome
Co., USA). One hundred milligrams samples were reduced in
H2 flow at 473 K for 1 h, and then purged with He at 473 K for
caused a significant decrease in the specific surface area.
The specific surface areas of Pt/CeO2, Pt–Sn/CeO2 (Pt/Sn
molar ratio = 1:1) and Pt–Sn/CeO2 (Pt/Sn molar ratio = 1:2)
2
catalysts were measured to be 92, 93 and 90 m /g, respec-
3
0 min to remove the adsorbed species. The CO chemisorp-
tively. Fig. 1 shows the XRD patterns of CeO2, Pt/CeO2 and
Pt–Sn/CeO2 samples. As shown, the distinct fluorite oxide
diffraction patterns of CeO2 were observed in all samples,
and the average crystallite size of ceria was estimated to
be about 9.3 nm using the Scherrer equation. However, no
diffraction peaks assigned to crystallines of platinum or tin
species can be detected, even in samples after reduced at
773 K, suggesting the high dispersion of platinum and tin
species on the surface of CeO2.
Table 1 lists the CO uptakes of the catalysts. The amount
of CO adsorbed on Pt/CeO2 catalyst was measured to be
101 ␮mol/gcat. The uptakes of CO adsorption on the bimetal-
lic Pt–Sn/CeO2 catalysts were lower than that of Pt/CeO2
catalyst, ranging from 89 to 62 ␮mol/gcat. This is consistent
tion was conducted after cooling down the samples to room
temperature.
The temperature-programmed reduction (TPR) measure-
ments were conducted in a conventional setup equipped with
a TCD detector. Samples of 100 mg Pt/CeO2 or Pt–Sn/CeO2
were loaded and pretreated in He flow at 473 K for 1 h to
remove the adsorbed carbonates and hydrates. After cooling
down to ambient temperature and introducing the reduction
agent of 5% H2/Ar, the temperature was then programmed to
rise at 10 K/min.
X-ray Photoelectron Spectroscopy measurement was per-
formed with an ESCALAB MK-II spectrometer (VG Scien-
tific Ltd., UK) using Al K␣ radiation (1486.6 eV) operated at

1
24
X. Tang et al. / Journal of Molecular Catalysis A: Chemical 235 (2005) 122–129
Fig. 1. XRDpatternsofCeO2, Pt/CeO2 andPt–Sn/CeO2 catalysts:(a)CeO2;
b) Pt/CeO2; (c) Pt–Sn/CeO2 (Pt/Sn = 1:1); (d) Pt–Sn/CeO2 (Pt/Sn = 1:2); (e)
Fig. 2. TPR profiles of Sn/CeO2 and Pt–Sn/CeO2 catalysts.
(
Pt–Sn/CeO2 (Pt/Sn = 1:1) reduced at 773 K; (f) Pt–Sn/CeO2 (Pt/Sn = 1:2)
reduced at 773 K.
molar ratio of Pt/Sn was 3:1, respectively. The former can be
assigned to the reduction of platinum oxide and the reduction
of SnO2 to SnOx in intimate contact with platinum, and
the later might be associated with the reduction of SnOx to
with the results of Verbeek and Sachtler [17], who observed
the drastic decrease in the capacity of CO adsorption over
Pt–Sn systems when the Sn/Pt atomic ratio increased from
0
Sn . When the Pt/Sn ratio arrived at 1:1, the reduction peaks
0
to 2.0. However, it should be noted that the measured CO
shifted to 390 and 490 K corresponding to the reductions
of platinum and tin oxides, respectively. Considering the
disappearance of reduction peak at 576 and 680 K, which
appeared in Sn/CeO2, it is reasonable to say that the presence
of platinum induced the reduction of SnOx by hydrogen
spillover, and thus promoted the reduction of tin oxide.
On the other hand, comparing with Pt/CeO2 sample, the
reduction peak of Pt oxide in Pt–Sn/CeO2 shifted to higher
temperature, indicating that the reduction of platinum was
depressed by the addition of tin. With respect to the case of
Pt/Sn = 1:2, two overlapped reduction peaks were observed
at 365 and 476 K, but the intensity of low temperature
reduction peak decreased accompanying by an increase in
the intensity of the high temperature peak. Thus, it can be
concluded the reduction of tin in Pt–Sn bimetallic catalysts
depends on the Pt/Sn molar ratio, and the reduction of
platinum and tin species was preceded in a synergistic way.
uptakes are relative values because CO can be adsorbed on
the ceria surface, especially in the presence of metals [14]. In
a comparative way, the results of CO adsorption still qualita-
tively indicated the change of accessibility of Pt to CO with
the increase of tin content.
3
.2. Temperature-programmed reduction
In our previous study, the Pt/CeO2 was found to display a
single reduction peak of PtO at 244 K and reduction peak of
surface CeO2 at 670 K [16]. The TPR profiles of Sn/CeO2
and Pt–Sn/CeO2 catalysts are presented in Fig. 2. Sn/CeO2
presented three reduction peaks at 576, 680 and 778 K. The
two overlapped reduction peaks at 576 and 680 K can be
ascribed to the reduction of SnO2 to SnOx (1 < x < 2) and
removal of surface oxygen in CeO2, and the peak at 778 K
can be assigned to the surface reduction of SnOx to Sn [19].
For the Pt–Sn/CeO2 samples, their TPR profiles remarkably
depended on the loading of tin, suggesting the presence
of interaction between the two metals. The two separated
reduction peaks appeared at 450 K and 595 K when the
3.3. X-ray photoelectron spectroscopy
Fig. 3 compares the X-ray photoelectron spectra of Pt 4f
in Pt–Sn/CeO2 catalysts (Pt/Sn = 1:1 and 1:2) experienced
reduction with hydrogen. The binding energies and the
relative concentrations of platinum and tin species are
summarized in Table 2. The XPS spectra of Pt 4f were
resolved into three sets of spin-orbit doublets, which can
Table 1
CO chemisorption on Pt/CeO2 and Pt–Sn/CeO2 catalysts
Sample
CO uptake (␮mol/gcat)
0
2+
4+
Pt/CeO2
101
89
81
be assigned to Pt , Pt and Pt , respectively. For the
Pt–Sn/CeO2 (Pt/Sn = 3:1)
Pt–Sn/CeO2 (Pt/Sn = 1:1)
Pt–Sn/CeO2 (Pt/Sn = 1:2)
as-prepared Pt–Sn/CeO (Pt/Sn = 1:1), Pt species presented
2
2+
4+
0
as Pt (46%), Pt (15%) and Pt (38%). After reduction
with hydrogen at RT and 473 K, the percentages of Pt
62
0

X. Tang et al. / Journal of Molecular Catalysis A: Chemical 235 (2005) 122–129
125
Fig. 3. (A) XPS spectra of Pt 4f for Pt–Sn/CeO2 (Pt/Sn = 1:1): (a) as-prepared sample, and reduced with hydrogen at (b) RT, (c) 473 K, (d) 773 K; (B) XPS
spectra of Pt 4f for Pt–Sn/CeO2 (Pt/Sn = 1:2): (a) as-prepared sample, and reduced with hydrogen at (b) RT, (c) 473 K, (d) 773 K.
greatly increased to 53% and 100%. The majority of the
Pt species over the as-prepared Pt–Sn/CeO2 (Pt/Sn = 1:2)
also was oxidized platinum species, and the percentages
the proportion of reduced Pt species sharply increased to
81%, and the Pt species were completely reduced to Pt
0
after reduction at 473 K. Compared with the monometallic
of Pt2 and Pt were 49% and 24%. Once reduced at RT,
+
4+
Pt/CeO2 catalyst in which 94% of Pt was presented even
0
Table 2
Reduction temperature dependence of component chemical states
Catalyst
Pt/CeO2
Reduction temperature (K)
Platinum species
Binding energy of Pt 4f7/2 (eV)
Tin species
Binding energy of Sn 3d5/2 (eV)
0
a
Fresh
Pt (54)
71.5
73.1
74.6
71.6
73.3
71.6
71.5
–
–
–
–
–
–
–
–
–
–
–
–
–
–
2
+
Pt (31)
4
+
Pt (15)
0
RT
Pt (94)
2
+
Pt (6)
0
4
7
73
73
Pt (100)
0
Pt (100)
Pt–Sn/CeO2 (Pt/Sn = 1:1)
Fresh
0
0
b
Pt (39)
71.2
73.0
74.3
71.1
72.8
71.1
Sn (22)
485.6
486.8
–
485.7
486.9
485.9
486.8
485.8
487.1
2
+
2+
Sn , Sn4+ (78)
Pt (46)
4
+
Pt (15)
–
0
0
RT
Pt (53)
Sn (28)
2
+
Sn , Sn4+ (72)
2+
Pt (47)
0
0
4
7
73
73
Pt (100)
Sn (49)
Sn , Sn4+ (51)
2+
0
0
Pt (100)
71.2
Sn (68)
Sn , Sn4+ (32)
2+
Pt–Sn/CeO2 (Pt/Sn = 1:2)
Fresh
0
0
Pt (27)
71.2
72.7
73.6
71.2
72.8
71.3
Sn (20)
485.6
487.0
–
485.9
486.9
485.7
486.7
485.7
486.8
2
+
2+
Sn , Sn4+ (80)
Pt (49)
4
+
Pt (23)
–
0
0
RT
Pt (81)
Sn (24)
2
+
Sn , Sn4+ (76)
2+
Pt (19)
0
0
4
7
73
73
Pt (100)
Sn (56)
Sn , Sn4+ (44)
2+
0
0
Pt (100)
71.2
Sn (71)
Sn , Sn4+ (29)
2
+
a
The numbers within the parentheses is referred to the content of Pt species.
The numbers within the parentheses is referred to the content of Sn species.
b

1
26
X. Tang et al. / Journal of Molecular Catalysis A: Chemical 235 (2005) 122–129
Fig. 4. (A) XPS spectra of Sn 3d for Pt–Sn/CeO2 (Pt/Sn = 1:1): (a) as-prepared sample, and reduced with hydrogen at (b) RT, (c) 473 K, (d) 773 K; (B) XPS
spectra of Sn 3d for Pt–Sn/CeO2 (Pt/Sn = 1:2): (a) as-prepared sample, and reduced with hydrogen at (b) RT, (c) 473 K, (d) 773 K.
0
reduced ceria) on platinum surface and the simultaneously
migration of tin to the surface which is often observed in
CeO2-supported metal catalysts may be another reason
after reduction at RT, whereas only 50 – 80% of Pt was
obtained in the Pt–Sn/CeO2 catalysts after reduction at RT,
suggesting that tin suppress the reduction of Pt species. On
the other hand, the binding energy of Pt 4f7/2 in Pt–Sn/CeO2
catalysts slightly shifted towards lower values (∼0.3 eV),
evidencing a possible charge transfer from Sn to Pt.
[
20]. With regard to the Pt–Sn/CeO2 (Pt/Sn = 1:2), the Pt/Sn
molar ratio also increased initially, and then significantly
decreased with the reduction temperature until it is close
to the bulk Pt/Sn ratio. This behavior can be understood by
considering Pt sintering and the migration of Sn from bulk
to surface at reductive atmosphere. Consequently, exposed
Pt surface atoms decreased with increasing tin loading and
reduction temperature, suggesting some Pt particles were
covered with Sn domains, especially in tin-rich sample.
Recent studies concerning Pt–Sn bimetallic system has
been focused on the oxidation state of tin and the possibility
of alloy formation [21,22]. In our study, XPS measurements
XPS spectra of Sn 3d are shown in Fig. 4, which were
deconvoluted into two components: one at lower BE value
0
(
485.3 – 485.8 eV) corresponds to Sn and the other at higher
BE (468.6 – 487.0 eV) represents oxidized tin species. As
shown in Table 2, Pt–Sn/CeO2 catalysts showed that the
0
percentage of Sn gradually increased with the reduction
temperature. Reduction at 473 K, the percentages of Sn0
in Pt–Sn/CeO2 (Pt/Sn = 1:1) and Pt–Sn/CeO2 (Pt/Sn = 1:2)
increased to 49% and 56%, whereas the corresponding
values further increased to 68% and 71% after reduction
at 773 K, respectively. This finding proved that about 70%
of surface tin oxides were reduced simultaneously with
platinum at temperatures below 773 K. The increase in the
metallic Sn concentration also evidenced the possibility of
PtSn alloys formation. Together with the downshift of the
binding energy of Pt 4f7/2, as shown in Fig. 4, the binding
energy of Sn 3d5/2 in Pt–Sn/CeO2 catalysts shifted towards
higher value, indicating possible electronic modification of
tin through the Pt–Sn interaction.
Table 3
Reduction temperature dependence of Pt and Sn surface compositions
Catalyst
Reduction temperature (K)
Atomic ratio
Pt/Ce
Pt:Sn
Ce/O
Pt/CeO2
Fresh
RT
0.74
0.43
0.25
0.58
–
–
–
–
0.99
2.28
3.27
2.49
4
7
73
73
The relative surface compositions of Pt, Sn, Ce and
O elements also changed after the samples were reduced
with hydrogen, as shown in Table 3. For the as-prepared
Pt–Sn/CeO2 (Pt/Sn = 1:1), the Pt/Ce and Pt/Sn ratios were
Pt-Sn/CeO2 (Pt/Sn = 1:1)
Fresh
RT
473
0.43
0.93
0.75
0.52
0.99
1.32
1.07
0.94
1.11
0.86
1.14
1.23
773
0
.43 and 0.99, respectively. Reduction at RT led to obvious
increase of Pt/Ce and Pt/Sn ratios. When the reduction
temperature increased from RT to 773 K, the atomic ratios
of Pt/Ce and Pt/Sn decreased, indicating the highly possible
sintering of Pt particle, especially at higher reduction
temperatures. Meanwhile, the coverage of ceria (or partially
Pt–Sn/CeO (Pt/Sn = 1:2)
2
Fresh
RT
0.50
0.61
0.52
0.35
0.83
1.26
0.71
0.51
1.10
1.03
1.68
1.79
4
7
73
73

X. Tang et al. / Journal of Molecular Catalysis A: Chemical 235 (2005) 122–129
127
clearly indicated the presence of metallic tin in both the
as-prepared and the reduced Pt–Sn/CeO2 catalysts, and the
proportion of reduced tin increased with tin loading and
reduction temperature. After reduction at 773 K, the pro-
on SiO2, forming so-called “sites isolation platinum crystals”
[24]. The proportion of reduced tin species is close to 50%,
higher than those usually reported for Pt–Sn and Rh–Sn cat-
alysts prepared by conventional impregnation methods [25].
0
portion of Sn in Pt–Sn/CeO2 (Pt/Sn = 1:1) and Pt–Sn/CeO2
(
Pt/Sn = 1:2) increased to 68% and 71%, respectively, to-
3
.4. Activity test
gether with the fully reduced Pt. Thus, the alloying between
tin and platinum is quite favorable under these conditions.
Additionally, the DTA measurements of the Pt-Sn/CeO2
catalysts (Fig. 5) confirmed the occurrence of PtSn alloy
with endothermal peaks emerged at 489 K for Pt–Sn/CeO2
The conversion of ethanol over Pt/CeO2 and Pt–Sn/CeO2
catalysts as a function of temperature is presented in Fig. 6.
Sn/CeO2 exhibited very low activity, no apparent ethanol
conversion can be observed even at 400 K, whereas the
corresponding conversion of ethanol reached 100% over
Pt/CeO2 even at 393 K. For Pt–Sn/CeO2 catalysts, the
conversion of ethanol increased initially with the addition
of Sn, and approached a maximum at Pt/Sn atomic ratio of
(Pt/Sn = 1:1) and 493 K for Pt–Sn/CeO2 (Pt/Sn = 1:2). Sim-
ilar conclusions were reached by Schubert, who confirmed
that a significant fraction of the Sn in a supported PtSn cat-
alyst involved in PtSn alloy formation after reduction in H2,
while the remaining fraction was left as unreduced tin ox-
ides [7]. XPS and M o¨ ssbauer spectroscopy also observed that
1
:1. The complete conversion temperature of ethanol over
Pt–Sn/CeO2 (Pt/Sn = 1:1) decreased to at temperature as
low as 363 K. However, the conversion of ethanol decreased
when the Pt/Sn atomic ratio was 1:2, even lower than that of
monometallic Pt/CeO2 catalyst. For clarity, the calculated
TOF was roughly compared. It is obvious that the TOF
0
from 30% to 70% of Sn presented in the form of PtSn al-
loy in Pt–Sn/Al2O3 and Pt–Sn/SiO2 [22,23]. These results
suggested the presence of a strong and effective interaction
between platinum and tin in the case of bimetallic PtSn cata-
lysts. According to above discussion, the presence of metallic
tin could be explained by the formation of Pt–Sn alloy.
Ontheotherhand, thedownwardshiftofbindingenergyof
platinum together with the upward shift of binding energy of
tin evidenced the presence of a possible charge transfer from
SntoPt. Moreover, thelowercapacityofCOchemisorptionin
the bimetallic catalysts comparing to that of the monometallic
one can also be related to modifications in the structure of the
metallic phase that hinders the availability of surface Pt atoms
to CO.
Several controversial views concerning the exact nature
of Pt–Sn interaction were proposed, probably due to differ-
ent preparation techniques and treatment conditions [11–24].
The way that Sn modifies the catalytic behavior of Pt can
vary from bulk alloying or surface decoration to electronic
interaction. Humblot et al. observed that there exist many
platinum crystals surrounded mostly by tin atoms supported
depended on the Pt/Sn ratio. For instance, the TOF value
−2 −1
−2 −1
was 4.81 × 10
s
s
for the Pt–Sn/CeO2 (Pt/Sn = 1:1)
for the Pt–Sn/CeO2 (Pt/Sn = 3:1) at
and 4.19 × 10
3
53 K, while for Pt/CeO2 and Pt–Sn/CeO2 (Pt/Sn = 1:2),
−2
the corresponding values decreased to 2.01 × 10
and
−2 −1
2
.42 × 10
s
, respectively. The same activity of the
Pt–Sn/CeO2 (Pt/Sn = 1:1) at 353 K was attained only at
3
73 K by the Pt/CeO2 and Pt–Sn/CeO2 (Pt/Sn = 1:2).
Acetaldehyde was the main by-product in the process
of ethanol oxidation, and the amounts of ethyl acetate
and methane were relatively small. Acetaldehyde and
ethyl acetate were observed at low conversion levels of
ethanol, while trace amounts of methane only appeared
at relatively high conversions of ethanol. Figs. 7 and 8
show the temperature dependences of acetaldehyde and
CO2 selectivities over the catalysts investigated. The
CO2 selectivity increased gradually with the reaction
Fig. 5. DTA curves of Pt–Sn/CeO2 catalysts.
Fig. 6. C2H5OH conversion on Pt/CeO2 and Pt–Sn/CeO2 catalysts.

1
28
X. Tang et al. / Journal of Molecular Catalysis A: Chemical 235 (2005) 122–129
of 333 – 353 K, while it only was observed between 353
and 373 K over the Pt–Sn/CeO2 (Pt/Sn = 1:2) and Pt/CeO2.
Considering the formation of methane must involve the
leakage of C C bond, so above discrepancy might be caused
by the difference of the ability in activating C C bond. Thus,
it can be concluded that the presence of appropriate Sn can
facilitate the leakage of C C bond to form methane and CO,
and then these products are further oxidized to final oxidation
products. So, the ability in activating C C bond also is an
important factor in improving the catalytic performances of
ethanol oxidation. Clearly, the addition of tin can promotes
the catalytic performances of Pt/CeO2 for ethanol oxidation.
The Pt–Sn/CeO2 (Pt/Sn = 1:1) gave the highest catalytic
performances with the complete oxidation of ethanol to CO2
and H2O at 363 K, while higher temperatures were needed
for the Pt–Sn/CeO2 (Pt/Sn = 1:2) and Pt/CeO2 catalysts.
Combining the reaction data of the complete oxidation of
ethanol with the structural features of the Pt–Sn/CeO2 cata-
lysts, itisreasonabletosaythattheenhancedcatalyticactivity
is associated with the modification of electronic configuration
of Pt and the formation of PtSn alloy stem from the effective
interaction between Pt and Sn. The charge transfer from Sn
to Pt would contribute greatly to polarize the C O and C C
bonds, and therefore favor the reaction of activated ethoxy
molecules with activated oxygen species. In addition, the in-
teraction between Pt and Sn provides different sites in close
proximity of each other, Pt particles facilitate the dissociative
adsorption of ethanol and Sn sites intimately adjacent to the
Pt particles would provide an alternative site for O2 activa-
tion. Consequently, this effective interaction could improve
catalytic activities for ethanol oxidation. On the other hand,
the separate SnOx phase would cover active Pt sites and im-
pede ethanol chemisorption, so higher Sn loading would lead
to the decrease of the catalytic performances.
Fig. 7. Acetaldehyde selectivity over Pt/CeO2 and Pt–Sn/CeO2 catalysts.
temperature, whereas acetaldehyde selectivity presented
different tendency. For Pt–Sn/CeO2 (Pt/Sn = 1:1, 3:1) and
Pt/CeO2 catalysts, acetaldehyde selectivity decreased with
the temperature. Whereas, for Pt–Sn/CeO2 (Pt/Sn = 1:2),
acetaldehyde selectivity initially increased and then dropped
with reaction temperature after passing a maximum at 343 K.
The relatively low acetaldehyde selectivity below 343 K was
due to the formation of ethyl acetate, suggesting that addition
of tin increases the selectivity of ethyl acetate at low conver-
sion levels of ethanol. Over Pt–Sn/CeO2 (Pt/Sn = 1:1) and
Pt–Sn/CeO2 (Pt/Sn = 3:1) catalysts, acetaldehyde selectivity
reached to zero and CO2 became the unique reaction product
at 363 K. Comparatively, CO2 became the unique product
above 393 K over Pt/CeO2 and Pt–Sn/CeO2 (Pt/Sn = 1:2)
catalysts. Even though, this temperature range (363 – 393 K)
is still much lower than the temperatures required for the
complete oxidation of ethanol to CO2 over previous reported
precious metal catalysts [1–3]. Additionally, it is needed
to point out that methane appeared at different temperature
domains for different samples. For Pt–Sn/CeO2 (Pt/Sn = 1:1,
4. Conclusions
3
:1) samples, methane occurred at the temperature range
The Pt–Sn/CeO2 catalysts showed considerably high cat-
alytic performances for the complete oxidation of ethanol to
CO2 at 363 K. The modification of electronic configuration of
Pt and the formation of PtSn alloys determined the nature of
the active sites contributing to the increase of catalytic perfor-
mances for the complete oxidation of ethanol. The separate
SnO2 phase at high tin loading would cover active Pt sites
and impede ethanol chemisorption, resulting in a significant
increase in the temperature required for complete oxidation
of ethanol to CO2.
References
[
[
[
1] W. Wang, H.B. Zhang, G.D. Lin, Z.T. Xiong, Appl. Catal. B 24
(2000) 219.
2] A. Marques da Silva, G. Corro, P. Marecot, J. Barbier, Stud. Surf.
Sci. Catal. 116 (1998) 93.
3] R.D. Gonzalez, M. Nagaia, Appl. Catal. 18 (1985) 57.
Fig. 8. CO2 selectivity over Pt/CeO2 and Pt–Sn/CeO2 catalysts.

X. Tang et al. / Journal of Molecular Catalysis A: Chemical 235 (2005) 122–129
129
[
[
4] A. Yee, S.J. Morrison, H. Idriss, Catal. Today 63 (2000) 327.
5] P.-Y. Sheng, G.A. Bowmaker, H. Idriss, Appl. Catal. A 261 (2004)
[14] P. Bera, A. Gayen, M.S. Hedge, N.P. Lalla, L. Spadaro, F. Frusteri,
F. Arena, J. Phys. Chem. B 107 (2003) 6122.
1
71.
[15] S. Bernal, J.J. Calvino, M.A. Cauqui, J.M. Gatica, C. Larese, J.A.
P e´ rez Omil, J.M. Pintado, Catal. Today 50 (1999) 175.
[16] X.L. Tang, B.C. Zhang, Y. Li, Y.D. Xu, Q. Xin, W.J. Shen, Catal.
Lett. 97 (2004) 163.
[17] H. Verbeek, W.M.H. Sachtler, J. Catal. 42 (1976) 257.
[18] V. Perrichon, L. Retailleau, P. Bazin, M. Daturi, J.C. Lavalley, Appl.
Catal. A 260 (2004) 1.
[
6] C. Lamy, A. Lima, V. LeRhun, F. Delime, C. Coutanceau, J.M.
L e´ ger, J. Power Sources 105 (2002) 283.
7] M.M. Schubert, M.J. Kahlich, G. Feldmeyer, M.S. Hackenberg,
H.A. Gasteigern, R.J. Behm, Phys. Chem. Chem. Phys. 3 (2001)
[
1
123.
8] P. Chantarvitoon, S. Chavadej, J. Schwank, Chem. Eng. J. 97 (2004)
61.
[
[
1
[19] M. del, C. Aguirre, P. Reyes, M. Oportus, I. Meli a´ n-Cabrera, J.L.G.
Fierro, Appl. Catal. A 233 (2002) 183.
9] A. Erhan Aksoylu, N. Madalena, A. Freitas, J.L. Figueiredo, Catal.
Today 62 (2000) 337.
10] W.J. Zhou, Z.H. Zhou, S.Q. Song, W.Z. Li, G.Q. Sun, P. Tsiakaras,
[20] A. Trovarelli, Catal. Rev. Sci. Eng. 38 (1996) 439.
[21] O.A. Barias, A. Holmen, E.A. Blekkan, Catal. Today 24 (1995) 361.
[22] Y.X. Li, K.J. Klabunde, B.H. Davis, J. Catal. 128 (1991) 1.
[23] Y.X. Li, J.M. Stencel, B.H. Davis, Appl Catal. 64 (1990) 71.
[24] F. Humblot, B. Didillon, F. LePeltier, J.P. Candy, J. Corker, O.
Clause, F. Bayard, M. Basset, J. Am. Chem. Soc. 120 (1998) 137.
[25] C. Kappenstein, M. Gu e´ rin, K. L a´ z a´ r, K. Matusek, Z. Pa a´ l, J. Chem.
Soc., Faraday Trans. 94 (1998) 2463.
[
[
[
[
Q. Xin, Appl. Catal. B 46 (2003) 273.
11] S.R. de Miguel, M.C. Rom a´ n-Mart ´ı nez, D. Cazorla-Amor o´ s, E.L.
Jablonski, O.A. Scelza, Catal. Today 66 (2001) 289.
12] S.A. Bocanegra, A. Guerrero-Ruiz, S.R. de Miguel, O.A. Scelza,
Appl. Catal. A 277 (2004) 11.
13] R.D. Cortright, J.A. Dumesic, J. Catal. 157 (1995) 576.