Correlation between the Pt2+/Pt4+ ratio and the catalytic activity for the CO oxidation of Ba12[BaxPt3-x]Pt6O27 (0 ≤ x ≤ 3)

Article abstract of DOI:10.1016/S0025-5408(99)00219-6

A variety of perovskite-type mixed oxides corresponding to the solid solution Ba₁₂[Ba_xPt_3-x]Pt₆O₂₇ (0 ≤ x ≤ 3) was studied by X-ray photoelectron spectroscopy (XPS). Pt was found to be in two valence states, Pt²⁺ and Pt⁴⁺, and the quantity of Pt²⁺ decreases when x increases. The intrinsic activity, for catalytic CO oxidation, was found to be dependent on x. At low temperature (below 170°C), the highest activity is obtained for Ba₁₂Pt₉O₂₇ (x = 0) and the activity decreased with increasing x. The apparent activation energies increased, from 80 to 110 kJ·mol^-1, when x increased; for a sample of metallic platinum catalyst studied in the same conditions, a value of 120 kJ·mol^-1 was found. The partial reaction order/O₂ (in the range of 0.64 to 0.90) increased with x while the partial order/CO (in the range of -0.34 to -0.70) decreased when x increased. Moreover, for the oxide containing only Pt⁴⁺ (Ba₁₅Pt₆O₂₇) the reaction orders were similar to those determined for the metallic platinum. Hence, there is undoubtedly a correlation between the Pt²⁺/Pt⁴⁺ ratio in these oxides and the catalytic activity for CO oxidation. This can be explained by assuming that CO chemisorption on Pt²⁺ is weaker than on Pt⁴⁺, which implies a less inhibitor effect of CO at low temperature.

Full text of DOI:10.1016/S0025-5408(99)00219-6

Materials Research Bulletin, Vol. 34, Nos. 12/13, pp. 2101–2108, 1999
Copyright © 2000 Elsevier Science Ltd
Printed in the USA. All rights reserved
0025-5408/99/$–see front matter
PII S0025-5408(99)00219-6
CORRELATION BETWEEN THE Pt2؉/Pt4؉ RATIO AND THE CATALYTIC
ACTIVITY FOR THE CO OXIDATION OF Ba12[BaxPt3؊x]Pt6O27 (0 < x < 3)
F. Grasset1*, P. Alphonse3, C. Labruge`re2, J. Darriet1, and A. Rousset3
1Institut de Chimie de la Matie`re Condense´e de Bordeaux (ICMCB)-CNRS, Avenue du
Docteur A. Schweitzer, 33608 Pessac Cedex, France
2DMSE-CECAMA/ICMCB, Avenue du Docteur A. Schweitzer,
33608 Pessac Cedex, France
3Laboratoire de Chimie des Mate´riaux Inorganiques, Universite´ Paul Sabatier, 118, route
de Narbonne, 31062 Toulouse Cedex, France
(Refereed)
(Received January 12, 1999; Accepted February 1, 1999)
ABSTRACT
A variety of perovskite-type mixed oxides corresponding to the solid solution
Ba12[BaxPt3Ϫx]Pt6O27 (0 Յ x Յ 3) was studied by X-ray photoelectron
spectroscopy (XPS). Pt was found to be in two valence states, Pt2ϩ and Pt4ϩ
,
and the quantity of Pt2ϩ decreases when x increases. The intrinsic activity, for
catalytic CO oxidation, was found to be dependent on x. At low temperature
(below 170°C), the highest activity is obtained for Ba12Pt9O27 (x ϭ 0) and the
activity decreased with increasing x. The apparent activation energies in-
creased, from 80 to 110 kJ⅐molϪ1, when x increased; for a sample of metallic
platinum catalyst studied in the same conditions, a value of 120 kJ⅐molϪ1 was
found. The partial reaction order/O2 (in the range of 0.64 to 0.90) increased
with x while the partial order/CO (in the range of Ϫ0.34 to Ϫ0.70) decreased
when x increased. Moreover, for the oxide containing only Pt4ϩ (Ba15Pt6O27)
the reaction orders were similar to those determined for the metallic platinum.
Hence, there is undoubtedly a correlation between the Pt2ϩ/Pt4ϩ ratio in these
oxides and the catalytic activity for CO oxidation. This can be explained by
assuming that CO chemisorption on Pt2ϩ is weaker than on Pt4ϩ, which
implies a less inhibitor effect of CO at low temperature. © 2000 Elsevier Science Ltd
*To whom correspondence should be addressed.
2101

2102
F. GRASSET et al.
Vol. 34, Nos. 12/13
FIG. 1
Representation of the basic structure of Ba12[BaxPt3Ϫx]Pt6O27.
KEYWORDS: A. oxides, B chemical synthesis, C photoelectron spectros-
copy, D catalytic properties
INTRODUCTION
Complex oxides with the perovskite structure and overall stoichiometry ABO3 have received
considerable attention because these compounds exhibit many interesting properties [1],
including ferromagnetism, ferroelectricity, pyro- and piezoelectricity, superconductivity,
large thermal conductivity, fluorescence, and catalytic activity. Perovskite-type oxides have
been examined for application in exhaust gas clean up [2–6], because they are efficient
catalysts for oxidation of CO [7–20] and hydrocarbons [14,17–19,21–26] as well as for NOx
reduction [9,17,27–32].
Whereas the oxidation of CO by O2 over group VIII metals has been studied extensively
during the past decades, relatively little is known about catalytic CO oxidation by perovskite-
containing platinum. To our knowledge, only Johnson et al. [33] has reported that catalytic
activity on BaPtO3 is about twice that of metallic Pt on Al2O3. More recently, Labalme et al.
[34] showed the increase of the activity of Pt/alumina combustion catalysts by barium
addition.
In a previous paper [35], we reported on a structural reinvestigation of the Ba–Pt–O system
for 4/3 Յ y Յ 5/2 (y ϭ Ba/Pt). The existence of the solid solution Ba12[BaxPt3Ϫx]Pt6O27 with
0 Յ x Յ 3 was confirmed. It was shown that the structure of these phases can be
schematically considered as being built up from the periodic stacking along the c direction
of one close-packed [Ba3O9] layer, similar to layers of the hexagonal perovskite and three
[Ba3AЈO9] layers with AЈ ϭ Ba, Pt. The stacking of these mixed layers is hexagonal type,
and columns parallel to c are formed. The columns consist of octahedra and trigonal prisms,
sharing faces, which are occupied by, respectively, Pt4ϩ cations and AЈ cations (AЈ ϭ Ba, Pt)
(Fig. 1).
The purpose of this study was to find a relationship between the Pt2ϩ/Pt4ϩ ratio and
catalytic activity for CO oxidation for Ba12[BaxPt3Ϫx]Pt6O27 with x ϭ 0, 1, 2, 3. The surface

Vol. 34, Nos. 12/13
BARIUM PLATINUM OXIDE
2103
chemical states of platinum in these oxides were determined by X-ray photoelectron spec-
troscopy (XPS). The microstructure was studied by scanning electron microscopy (SEM) and
by specific surface area measurement (BET method).
EXPERIMENTAL
Ba12Pt9O27 (x ϭ 0), Ba13Pt8O27 (x ϭ 1), Ba14Pt7O27 (x ϭ 2), and Ba15Pt6O27 (x ϭ 3) were
prepared by direct solid-state reaction, under flowing oxygen, of stoichiometric mixtures of
analytical grade BaCO3 and metallic Pt. The solid reactants were ground and heat-treated first
for 24 h at 950°C and then for a week at 1150°C.
The samples were examined by X-ray diffraction on a Philips X’Pert MPD diffractometer
using Cu K␣ radiation and by transmission electron microscopy (TEM) [35]. Scanning
electron microscopy (SEM) was carried out with a JEOL JSM 840 operating at an acceler-
ating voltage of 20 kV. The specific surface area of the samples was measured by nitrogen
adsorption (BET method) using a Micromeritics Flowsorb 2300.
Pressed powder pellets of the compounds Ba12Pt9O27 (x ϭ 0), Ba13Pt8O27 (x ϭ 1) and
Ba15Pt6O27 (x ϭ 3) were analyzed by X-ray photoelectron spectroscopy (XPS), using a VG
Scientific 220i-XL Escalab. The nonmonochromatized Mg X-ray source was used for
excitation. The spectrometer was operated in a constant pass energy mode of Epass ϭ 20 eV
(except for survey, where Epass ϭ 150 eV). A flood gun source was applied to the samples
during analysis to compensate for sample charging. Binding energies were referenced to the
C1s energy at 284.6 eV. The Pt4f core electron levels were recorded by a concentric
hemispherical analyzer. Experimental peaks were fitted using a Gaussian–Lorentzian mixture
peak-fitting software provided by VG Scientific. All binding energy (EB) experimental values
reported in the following sections are Ϯ0.2 eV.
Catalytic tests were performed in a differential, tubular, fixed bed flow reactor, at atmo-
spheric pressure, in the temperature range 120–220°C, with a residence time V/F Ϸ 0.01 s
(catalyst mass ϭ 0.010 g, volumetric flow rate ϭ 1.33 cm3sϪ1). Except for the order
determination, the flowing gas composition was CO–O2–N2 (1%–2%–97%). Reactants and
products were analyzed by gas chromatography (thermal conductivity detector, carrier gas
He) using two columns: 13ϫ molecular sieve for N2, O2, and CO and porapak Q for CO2.
Preliminary tests showed significant variations of the conversion with specific surface area,
hence, for comparison of catalysts, intrinsic activity Ai was used (expressed in mol⅐sϪ1⅐mϪ2).
TABLE 1
Variation of Specific Surface Area with x in
Ba12[BaxPt3Ϫx]Pt6O27
x
Oxide
Specific surface area (m2gϪ1
)
0
1
2
3
Ba12Pt9O27
Ba13Pt8O27
Ba14Pt7O27
Ba15Pt6O27
1.2
0.9
0.5
0.8

2104
F. GRASSET et al.
Vol. 34, Nos. 12/13
FIG. 2
SEM micrographs of Ba12Pt9O27 (a) and Ba15Pt6O27 (b).
RESULTS AND DISCUSSION
Specific Surface Area and SEM. The specific surface areas of the samples are reported in
Table 1. They are close to 1 m2gϪ1, except for Ba14Pt7O27, whose surface area is only 0.5
m2gϪ1. Figure 2 (a and b) shows the typical microstructure of Ba12Pt9O27 (x ϭ 0) and
Ba15Pt6O27 (x ϭ 3) obtained by scanning electron microscopy. The estimated average
particle size is about 1 ␮m.
X-ray Photoelectron Spectroscopy. The theoretical value (anticipated from the bulk
concentration) and experimental values for the Pt2ϩ/Pt4ϩ ratio and binding energies (EB) of
Pt4f peaks are summarized in Table 2. The Pt4f peaks were found to be symmetrical. The
binding energies, the FWHM (full-width at half-maximum) and ␦EB fit results for x ϭ 0, 1,
and 3, were in agreement with results of previous XPS studies on Pt–O bonding [36,37].
TABLE 2
Experimental and Theoretical Pt2ϩ/Pt4ϩ Ratio and X-ray Photoelectron Core Level
Binding Energies for Pt 4f Electron
Experimental
Theoretical
Pt 4f5/2
(eV)
Pt 4f7/2
(eV)
Compound
Ba12Pt9O27
Pt2ϩ/Pt4ϩ ratio
Pt2ϩ/Pt4ϩ ratio
Source
0.78
0.50
78.1
76.4
74.5
78.2
76.5
78.0
76.3
74.5
74.7
73.0
71.2
74.8
73.2
74.6
72.9
71.2
This work
Ba13Pt8O27
Ba15Pt6O27
Metallic Pt
0.43
0.14
—
0.33
0.00
—
This work
This work
[37]

Vol. 34, Nos. 12/13
BARIUM PLATINUM OXIDE
2105
FIG. 3
XPS Pt 4f spectra of Ba12Pt9O27 (a), Ba13Pt8O27 (b), and Ba15Pt6O27 (c).
Three X-ray photoelectron spectra for the platinum 4f regions are given in Figure 3 for the
samples with composition x ϭ 0, 1, 3. The spectrum of Ba12Pt9O27 (x ϭ 0) is a quintuplet,
which can be decomposed into three doublets. The binding energies of the doublets corre-
spond to Pt0, Pt2ϩ, and Pt4ϩ (Table 2). The presence of Pt0 could be attributed to the possible
reduction, under ultra-high vacuum, of the Pt2ϩ valence state, which is basically unstable.
Anyway, no metallic platinum was detected by X-ray diffraction [35]. The spectra of
Ba13Pt8O27 (x ϭ 1) and Ba14Pt7O27 (x ϭ 3) are triplets, which can be decomposed into two
doublets. However, for x ϭ 3, the small shoulder at low binding energy could perhaps be
attributed to sample charging. As it was expected, the experimental Pt2ϩ/Pt4ϩ ratio decreased
with x, but the values were found to be systematically smaller than the predicted bulk values.
We can assume that these differences, such as the presence of Pt0, could be attributed to
surface reduction of the samples under ultra-high vacuum. Moreover, in order to explain the
observation of difference between experimental and theoretical ratios, it should be borne in

2106
F. GRASSET et al.
Vol. 34, Nos. 12/13
FIG. 4
Variation with temperature of intrinsic catalytic activity (␮mol⅐sϪ1⅐mϪ2) for CO oxidation in
Ba12[BaxPt3Ϫx]Pt6O27 oxides.
mind that XPS is essentially a surface technique. It is generally accepted that only electrons
from the top 20–50 Å of the sample can escape.
Catalytic Oxidation of Carbon Monoxide. The variation of the intrinsic catalytic activity
as a function of temperature is shown in Figure 4. Below 170°C, Ba12Pt9O27 (x ϭ 0) has the
highest activity; with increasing Ba2ϩ substitution, the activity decreased. Above 170°C, the
best catalyst was Ba13Pt8O27 (x ϭ 1).
The apparent activation energies, calculated from Arrhenius plots of ln(activity) against
1/T, are given in Table 3. For comparative purposes, we have determined the activation
energy of metallic platinum powder (40 ␮m sifted) in the same conditions. It can be seen that
the activation energy increased, from 80 to 110 kJ⅐molϪ1, when x increased from 0 to 3; the
maximum value, 120 kJ⅐molϪ1, was obtained for metallic platinum.
The apparent reaction orders m and n, corresponding to the equation
Ai ϭ kPmO PnCO
2
were determined at 180°C (Table 4). With increasing Ba2ϩ substitution, n increased while m
decreased. Moreover, the values found for the oxide containing only Pt4ϩ (Ba15Pt6O27) were
the same as those determined for the metallic platinum.
TABLE 3
Apparent Activation Energies
Catalyst
Activation energy (kJ/mol)
Ba12Pt9O27
Ba13Pt8O27
Ba14Pt7O27
Ba15Pt6O27
Metallic Pt
81
93
101
110
120

Vol. 34, Nos. 12/13
BARIUM PLATINUM OXIDE
2107
TABLE 4
Partial and Global Reaction Orders
Algebraic sum of
Catalyst
Order/CO
Order/O2
partial orders
Global order
Ba12Pt9O27
Ba13Pt8O27
Ba15Pt6O27
Metallic Pt
Ϫ0.34
Ϫ0.54
Ϫ0.71
Ϫ0.70
0.64
0.76
0.90
0.90
0.30
0.22
0.19
0.20
0.32
0.24
0.18
It can be observed that, in all cases, the total reaction order is in agreement with the algebraic
sum of the partial orders. Thus, CO2 produced does not take place in the reaction mechanism or,
in other words, we can assume the immediate desorption of CO2 after its formation.
The generally accepted scheme of CO oxidation on the surface of platinum group metals
involves the following steps [38]:
O2 3 2Oads
CO 3 COads
(1)
(2)
(3)
(4)
COads ϩ Oads 3 CO2
CO ϩ Oads 3 CO2
If the dissociative chemisorption of oxygen is necessary, CO chemisorption reaction 2 is
optional, since CO2 can be formed either by reaction 3 according to a Langmuir–Hinshel-
wood mechanism or by reaction 4 via an Eley–Rideal mechanism. Because of the strong
adsorption of CO on these metals, high CO coverage inhibits the chemisorption of O2
slowing down the catalytic reaction. Hence, the reaction kinetics is mainly governed by the
CO coverage of the surface, which is linked to both the temperature and the CO partial
pressure. This explains why a negative partial order with respect to CO is generally found
with platinum metal catalysts.
On the other hand, in perovskite-type oxide catalysts, partial order with respect to CO is
generally found to be in the range 0.5–1.0 [11,15]. But this study shows that the
Ba12[BaxPt3Ϫx]Pt6O27 oxides have a catalytic behavior similar to metallic platinum and the
less Pt2ϩ they contain, the closer their kinetic parameters are to those of metallic platinum.
These results could be explained by assuming that CO chemisorption on Pt2ϩ is weaker than
on Pt4ϩ. Hence, the higher the Pt2ϩ content in the oxide, the less the inhibitor effect of CO
at low temperature.
These results are potentially interesting since, in present automotive catalytic converters
(based on noble metal catalysts), the conversion does not begin below 200°C because of the
inhibitor effect of CO.
REFERENCES
1. E.J. Baran, Catal. Today 8, 133 (1990).
2. W.F. Libby, Science 171, 499 (1971).

2108
F. GRASSET et al.
Vol. 34, Nos. 12/13
3. R.J.H. Voorhoeve, J.P. Remeika, P.E. Freeland, and B.T. Matthias, Science 177, 353 (1972).
4. Y.-F. Yu Yao, J. Catal. 36, 266 (1975).
5. R.J.H. Voorhoeve, D.W. Johnson, Jr., J.P. Remeika, and P.K. Gallagher, Science 195, 827 (1977).
6. T. Teraoka, H. Nii, S. Kagawa, K. Jansson, and M. Nygren, J. Mater. Chem. 6, 97 (1996).
7. G. Parravano, J. Am. Chem. Soc. 75, 1497 (1953).
8. O. Parkash, P. Ganguly, G.R. Rao, C.N.R. Rao, D.S. Rajoria and V.G. Bhide, Mater. Res. Bull.
9, 1173 (1974).
9. R.J.H. Voorhoeve, J.P. Remeika, and L.E. Trimble, Ann. N.Y. Acad. Sci. 272, 3 (1976).
10. J.M.D. Tascon and L.G. Tejuca., React. Kinet. Catal. Lett. 15 (2), 185 (1980).
11. N. Gunasekaran, A. Meenakshi Sundaram, and V. Srinivasan, Proc. Cat. Symp./Nat. Catal. Symp.
5, 108 (1980).
12. T. Shimizu, Chem. Lett., 1 (1980).
13. B. Viswanathan and S. George, Indian J. Chem. 22A, 1026 (1983).
14. T. Nakamura, M. Misono, and Y. Yoneda, J. Catal. 83, 151 (1983).
15. D.Y. Rao and D.K. Chakrabarty, Indian J. Chem. 23A, 375 (1984).
16. D.K. Chakrabarty and D.Y. Rao, React. Kinet. Catal. Lett. 33 (1), 131 (1987).
17. L.G. Tejuca, J. Less-Common Met. 146, 261 (1989).
18. C.B. Alcock and J.J. Carberry, Solid State Ionics 50, 197 (1992).
19. P. Salomonsson, T. Griffin, and B. Kasemo, Appl. Catal. A 104, 175 (1993).
20. A.M. Duprat, P. Alphonse, C. Sarda, A. Rousset, and B. Gillot, Mater. Chem. Phys. 37, 76 (1994).
21. L.A. Pedersen and W.F. Libby, Science 176, 1355 (1972).
22. G. Kremenic, J.M.L. Nieto, J.M.D. Tascon, and L.G. Tejuca, J. Chem. Soc., Faraday Trans. 81,
939 (1985).
23. M. Misono and T. Nitadori, in Adsorption and Catalysis on Oxide Surfaces, ed. M. Che and G.C.
Bond, pp. 409–419, Elsevier Science B.V., Amsterdam (1985).
24. T. Nitadori, T. Ichiki, and M. Misono, Bull. Chem. Soc. Jpn. 61, 621 (1988).
25. J.G. McCarty and H. Wise, Catal. Today 8, 231 (1990).
26. T. Seiyama, Catal. Rev. Sci. Eng. 34 (4), 281 (1992).
27. R.J.H. Voorhoeve, J.P. Remeida, and D.W. Johnson, Science 180, 62 (1973).
28. G. Mai, R. Siepmann and F. Kummer, U.S. Patent 3,900,428, 1975.
29. J.P. Remeida and R.J.H. Voorhoeve, U.S. Patent 3,884,837, 1975.
30. R.J.H. Voorhoeve, J.P. Remeika, L.E. Trimble, A.S. Cooper, F.J. Disalvo, and P.K. Gallagher, J.
Solid State Chem. 14, 395 (1975).
31. A.V. Salker, D.K. Chakrabarty, and H.V. Keer, Indian J. Chem., Sect. A, 458 (1989).
32. R. Bontchev, K. Cheshkova, D. Mehandjiev, and J. Darriet, React. Kinet. Catal. Lett. 63 (1), 121
(1998).
33. D.W. Johnson, P.K. Gallagher, G.K. Wertheim, and E.M. Vogel, J. Catal. 48, 87 (1977).
34. V. Labalme, E. Garbowski, N. Guilhaume, and M. Primet, Appl. Catal. A 138, 93 (1996).
35. F. Grasset, F. Weill, and J. Darriet, J. Solid State Chem. 140, 194 (1998).
36. D. Briggs and M.P. Seah, Practical Surface Analysis: Auger and X-ray Photoelectron Spectro-
copy, Vol. 1, Wiley, New York (1990).
37. D. Cahen, J.A. Ibers, and J.B. Wagner, Jr., Inorg. Chem. 13 (6), 1377 (1974).
38. H.P. Bonzel and R. Ku, J. Vac. Sci. Technol. 9, 663 (1972).