Full text of DOI:10.1070/MC2001v011n05ABEH001446

, 2001, 11(5), 186–188
New evidence for the electronic nature of the strong metal-support interaction
effect over a Pt/TiO2 hydrogenation catalyst
Aleksandr Yu. Stakheev,*a Yurii M. Shulga,b Natalia A. Gaidai,a Natalia S. Telegina,a Olga P. Tkachenko,a
Leonid M. Kustova and Khabib M. Minacheva
a N. D. Zelinsky Institute of Organic Chemistry, Russian Academy of Sciences, 119991 Moscow, Russian Federation.
Fax: +7 095 135 5328; e-mail: st@ioc.ac.ru
b Institute of Problems of Chemical Physics, Russian Academy of Sciences, 142432 Chernogolovka, Moscow Region,
Russian Federation
10.1070/MC2001v011n05ABEH001446
Analysis of the Pt 4f line asymmetry evidenced that the suppression of the hydrogenation activity of Pt/TiO2 in the SMSI state is
caused by a decrease in the d-electron density at the Fermi level of platinum particles, while the net charge of the metal particles
remains unaltered.
Interest in the effect of strong metal-support interaction (SMSI)
in M/TiO2 systems (M = Pt, Rh, Pd, Ni, etc.) has increased since
1980 because the performance of these catalysts can change dra-
matically depending on the reduction temperature.1 Despite inten-
sive research efforts directed toward the elucidation of the nature
of this phenomenon, its mechanism remains unclear. Several hy-
potheses have been proposed including encapsulation of metal
particles by the support material or alteration of their electronic
properties.2–7
In this study, we used a new method for the evaluation of the
electronic state of supported metal particles. This method is based
on the analysis of the lineshape asymmetry in the XPS spectra of
the metal.8 The importance of this parameter for studying the
electronic state of supported metal clusters stems from the fact that
the asymmetry of an XPS line is a function of the density of
d-electrons at the Fermi level.9 Taking into account that d-elec-
trons are deeply involved in the catalytic conversion over metals,
an analysis of the XPS line asymmetry may provide a new in-
sight into the relation between the electronic structure of sup-
ported metals and their catalytic performance. Therefore, the
aim of this study was to find a relationship between variations
in the catalytic activity and the electronic state of metal particles
upon transition of the Pt/TiO2 catalyst between the SMSI and
non-SMSI states.
In order to minimise experimental uncertainties resulting from
possible phase transitions in the carrier material and from varia-
tions in the dispersion of platinum, a 3 wt% Pt/TiO2 catalyst
supported on high-purity rutile type TiO2 with a narrow particle-
size distribution was used. The catalyst was prepared by impreg-
nation of TiO2 (95 m2 g–1) with an aqueous solution of H2PtCl6
(Aldrich, 99.995%) followed by drying and calcination at 500 °C.
The removal of residual chlorine was monitored by XPS.
After calcination, the catalyst samples were reduced in a
hydrogen flow at 200 and 500 °C, and the particle size of pla-
tinum was estimated by transmission electron microscopy (EM-
125 electron microscope operated at 75 kV) and from X-ray
diffraction line broadening. The X-ray diffraction line broadening
gave a particle size of ~6.0 nm for both samples, which is in a
good agreement with the estimates on the basis of XPS data
(see below). The TEM data also ensured the absence of bigger
platinum particles (>10 nm) in both samples.
X-ray photoelectron spectra were obtained using an XSAM-
800 spectrometer (Kratos) with AlKα 1,2 radiation for spectra
excitation. The binding energies of peaks were corrected with
account of sample charging by referencing to the C 1s peak at
285.0 eV. The Pt/Ti surface atomic ratio was calculated from the
integral intensities of XPS peaks using the Scofield photoioniza-
tion cross-sections for AlKα 1,2 excitation.10 The particle size of
Pt was calculated from the Pt 4f /Ti 2p intensity ratio using the
Kerkof ‘stacking sheets’ model.11 Reductive and oxidative pre-
treatments were performed using a home-made reactor attached
directly to the analytical chamber of the spectrometer.12 The
samples were treated in flowing H2 and O2 and transferred to
the spectrometer without exposure to ambient air.
A Pt foil was cleaned using a standard procedure13 of several
cycles of sputtering with Ar ions (2 kV, 0.25 µ a, t ~ 30 min)
followed by heating in oxygen (pO = 6×10–5 mbar, t = 45 min).
2
Surface contaminations were below the detection limits of XPS.
The spectra were analysed by a curve fitting procedure using
the Doniach–Sunjic function14
G(1 – a)
~
I(e) =
cos[0.5π a + q(e)],
(1)
(2)
(e2 + g2)(1 – a)/2
q(e) = (1 + a)tan–1(e/g),
where e is the kinetic energy of the photoelectrons, G is the
gamma-function, g is the lifetime width of the core hole created
as a result of photoemission and a is the line asymmetry param-
eter.
The function was convoluted with a Gaussian curve for taking
into account an experimental broadening (instrumental resolution,
sample inhomogeneity, etc.).
Kinetic investigations of toluene hydrogenation activity under
stationary conditions were carried out in a gradientless flow-cir-
16.0
14.0
Treatment in
O2 at 200 °C
12.0
Table 1 Variations of the Pt 4f7/2/Ti 2p3/2 atomic ratio, Pt particle size, and
binding energies (BE) of Pt 4f7/2 and Ti 2p3/2 lines with the reduction tem-
perature.
10.0
8.0
6.0
4.0
Pt 4f7/2/Ti 2p3/2 Particle size Pt 4f7/2 BE/ Ti 2p3/2 BE/
Pre-treatment
intensity ratio
of Pt/nma
eV
eV
H2, 200 °C
H2, 500 °C
0.015
0.013
5.7
6.5
6.5
70.9
70.8
70.8
458.9
458.9
458.8
2.0
0.0
O2, 200 °C + 0.013
H2, 200 °C
200
300
400
500
200
H2, 300 °C
H2, 400 °C
H2, 500 °C
0.014
0.013
0.013
6.1
6.5
6.5
70.7
70.8
70.7
458.9
458.9
458.8
Reduction temperature/°C
Figure 1 Effect of the reduction temperature on the reaction rate of toluene
hydrogenation at 130 °C over Pt/TiO2.
aCalculated from XPS data using the Kerkof model.5
– 186 –

, 2001, 11(5), 186–188
0.20
0.18
Pt-foil
0.16
0.14
0.12
0.10
0.08
0.06
500 °C, H2
Treatment in
O2 at 200 °C
0.04
200
500
200
300
400
500
Reduction temperature/°C
Figure 3 Dependence of the Pt 4f7/2 line asymmetry (a) on the reduction
temperature.
200 °C, H2
analysis demonstrates that the narrowing resulted from a sig-
nificant decrease in the asymmetry parameter a (Figure 2).
The dependences of the catalytic activity (Figure 1) or the
Pt 4f7/2 line asymmetry (Figure 3) on the reduction temperature
reveals a good correlation between these parameters. An increase
in the reduction temperature results in a simultaneous decrease in
the catalytic activity and in the a value. On the other hand, the
oxidation of the deactivated catalyst followed by the reduction
at 200 °C completely restores the activity and the a value.
The above data allowed us to interpret the observed effect.
Two main mechanisms were proposed for the explanation of
changes in the catalytic performance of the catalyst in the SMSI
state. One mechanism implies the decoration or encapsulation of
metal particles with the support material species. The other mech-
anism assumes the alteration of the electronic properties of the
metal particles due to their interaction with defects on the sup-
port surface created in the course of high-temperature treatment.
However, the Pt 4f/Ti 2p atomic ratio remained almost un-
changed upon the treatments (Table 1). Therefore, we can pre-
sume that the decoration or encapsulation of the metal particles
with the support moieties seems not to be the governing factor
of the catalytic performance. The constancy of the Pt 4f/Ti 2p
intensity ratio also implies the absence of any pronounced sin-
tering of metal particles (Table 1).
Pt-foil
81
78
75
72
69
Binding energy/eV
Figure 2 XPS spectra and the results of the curve-fitting analysis of the
Pt 4f line in Pt/TiO2 after reduction at 200 and 500 °C and in Pt foil. Dotted
lines correspond to the Pt 4f line with a = 0 and are shown for comparison.
On the other hand, a decrease in the catalytic activity is con-
sistent with the changes of the electronic state of the metal
particles deduced from the variations of the Pt 4f line asym-
metry. The decrease in the line asymmetry is presumably caused
by a decrease in the local density of a d-state at the Fermi level.
In turn, the structure of the d-band is the crucial factor deter-
mining the reactivity of metal surface.15,16 This allows us to
suggest that the strong metal–support interaction can result in
the lowering of the density of d-electrons on the Fermi level of
the metal particle, thus decreasing the overall activity of the
catalyst.
culation system at atmospheric pressure with the on-line gas-
chromatographic analysis of the reaction products. The reaction
rates were measured at low conversions at 130 °C and PC H
/
7
8
PH = 0.004.
2
The dependence of the rate of toluene hydrogenation on the
reduction temperature is depicted in Figure 1. The reaction rate
abruptly decreases with increasing reduction temperature above
300–400 °C. However, the oxidation of the catalyst reduced at
500 °C in O2 at 200 °C followed by the reduction at 200 °C
completely restores the activity.
It is noteworthy that the Pt 4f7/2 line asymmetry in the SMSI
catalyst (a = 0.06) becomes close to that for Au (a = 0.03 for
Au 4f7/2). The d-electron density of states for gold is low, which
is in a good agreement with the low catalytic activity of Au in
the reaction requiring the activation of H–H, C–H or C–C bonds.
Thus, the results of this study indicate that the strong metal–
support interaction induced by the high-temperature reduction
presumably decreases the density of d-electrons at the Fermi
level thus suppressing the activity of metal clusters. Another
important finding is that the interaction between a metal particle
and the support may affect the distribution of d-electrons in the
metal cluster and change its catalytic performance, while the net
charge of the particle remains nearly constant.
It was found that the oxidation–reduction cycles could be
repeated five times without appreciable changes in the reaction
rates. Clearly, the catalyst demonstrates the behaviour typical of
reversible transition between non-SMSI and SMSI states.
The XPS spectra of the Pt 4f line (Figure 2) indicated that the
binding energy of the Pt 4f line in the catalyst reduced at 200 °C
is higher by ~0.4 eV as compared to the position of the Pt 4f
line of platinum foil, which is typical of supported metals. The
curve fitting analysis of the line shape reveals a very similar
value of the Pt 4f line asymmetry a for platinum foil and Pt/TiO2
reduced at 200 °C.
The reduction at 500 °C does not affect the position of the
Pt 4f line (Figure 2, Table 1). However, it leads to the pro-
nounced narrowing of the line (Figure 2). Note that, after reduc-
tion at 500 °C, the Pt 4f line becomes narrower than that in Pt
foil, which is unexpected for supported metal. The curve fitting
This work was supported by the Russian Foundation for Basic
Research (grant no. 99-03-32222).
– 187 –

, 2001, 11(5), 186–188
11 F. P. J. M. Kerkof and J. A. Moulijn, J. Phys. Chem., 1979, 83, 1612.
12 W. Grunert, A. Yu. Stakheev, E. S. Shpiro, K. Anders, R. Feldhaus, Kh. M.
Minachev and W. Haupt, Reaktor-Schubstange System zur Hermetishen
Uberfuhrung von Festkorperproben in Vakuumgerate, Patent no. 289596
BRD 02.05.91, 1–11.
13 Y. M. Sun, D. Sloan, D. J. Alberas, M. Kovar and J. M. White, Surf. Sci.,
1994, 319, 34.
14 S. Doniach and M. Sunjic, J. Phys. C, 1970, 3, 285.
15 A. Ruban, B. Hammer, P. Stoltze, H. L. Skriver and J. K. Nørskov, J.
Mol. Catal. A, 1997, 115, 421.
References
1
S. J. Tauster, S. C. Fung and R. L. Garten, J. Am. Chem. Soc., 1978,
100, 170.
2
3
B. C. Bruce and P. N. Ross, J. Phys. Chem., 1986, 90, 6811.
D. E. Resasco, R. J. Fenoglio, M. P. Suarez and J. O. Cechini, J. Phys.
Chem., 1986, 90, 4330.
4
T. H. Fleisch, A. T. Bell, J. R. Regalbuto, R. T. Thomson, G. S. Lane,
E. E. Wolf and R. F. Hicks, Stud. Surf. Sci. Catal., 1988, 38, 791.
S. Tang, G. X. Xiong and H. G. Wang, J. Catal., 1988, 111, 136.
G. L. Haller and D. E. Resasco, Adv. Catal., 1989, 36, 173.
A. Dandekar and M. A. Vannice, J. Catal., 1999, 183, 344.
S. Hüfner, G. K. Wertheim and J. H. Wernick, Solid State Commun.,
1975, 17, 417.
5
6
7
8
16 B. Hammer and J. K. Nørskov, Surf. Sci., 1995, 343, 211.
9
D. Briggs and J. C. Riviere, in Practical Surface Analysis by Auger and
X-ray Photoelectron Spectroscopy, eds. D. Briggs and M. P. Seach,
John Wiley, New York, 1983.
10 J. H. Scofield, J. Electron Spectrosc., 1976, 9, 29.
Received: 1st March 2001; Com. 01/1772
– 188 –