Effect of thermal activation of supported catalysts Pt/MeOx, where MOx = Al2O3, La2O3, and ZrO2, for complete oxidation

Article abstract of DOI:10.1023/A:1016061919189

A sharp increase in the atomic catalytic activity (ACA) of supported platinum catalysts in the model reaction of n-pentane complete oxidation is found on going from the preliminary calcination temperature of 500-600°C to a temperature of 700°C. ACA increases by an order of magnitude for the Pt/γ-Al₂O₃ system, ～3 times for Pt/ZrO₂, and ～1.5 times for Pt/CeO₂. The per-gram activities of all catalysts decrease because of a decrease in the dispersion of supported platinum with an increase in the temperature of preliminary calcination.

Full text of DOI:10.1023/A:1016061919189

Kinetics and Catalysis, Vol. 43, No. 3, 2002, pp. 379–383. Translated from Kinetika i Kataliz, Vol. 43, No. 3, 2002, pp. 410–414.
Original Russian Text Copyright © 2002 by Chzhu, Tsyrul’nikov, Kudrya, Smolikov, Borbat, Bubnov.
Effect of Thermal Activation of Supported Catalysts Pt/MeOx,
Where MOx = Al2O3, CeO2, La2O3, and ZrO2,
for Complete Oxidation
D. P. Chzhu*, P. G. Tsyrul’nikov*, E. N. Kudrya*,
M. D. Smolikov*, V. F. Borbat**, and A. V. Bubnov*
* Boreskov Institute of Catalysis (Omsk Branch), Siberian Division, Russian Academy of Sciences, Omsk, 644053 Russia
** Omsk State University, Omsk, 644077 Russia
Received May 30, 2001
Abstract—A sharp increase in the atomic catalytic activity (ACA) of supported platinum catalysts in the model
reaction of n-pentane complete oxidation is found on going from the preliminary calcination temperature of
500–600°ë to a temperature of 700°ë. ACA increases by an order of magnitude for the Pt/γ-Al2O3 system,
~3 times for Pt/ZrO2, and ~1.5 times for Pt/CeO2. The per-gram activities of all catalysts decrease because of
a decrease in the dispersion of supported platinum with an increase in the temperature of preliminary calcina-
tion.
INTRODUCTION
starting samples are thermally treated in air at tempera-
tures ≥600°C, the surface Pt “oxide” structures become
unstable and decompose, whereas disperse “ionic” Pt,
which was previously stabilized on the support defects
and interacted with the support, is reduced to form
metallic Pt crystallites. Hence, the Ptn+/Pt0 ratio in the
samples decreases; that is, the number of Pt0-contain-
ing sites on which hydrocarbon molecules can adsorb
dissociatively increases [10].
The thermal activation effect, which consists in a
sharp increase in the atomic (specific) catalytic activity
upon calcination at elevated temperatures, has been
found and studied for the MnOx/Al2O3 and MnOx/ZrO2
oxide systems [1–5]. This effect is due to the initial
interaction (at relatively low temperatures) between the
components of the catalytic system and often produces
new phases or solid solutions, which decompose upon
high-temperature treatment or reduction. The above
effect was seen over many catalysts including plati-
num-based ones. It was shown in [6, 7] by the X-ray
radial atom distribution method that in aluminum–plat-
inum catalysts modified with rare-earth metal oxides
[6], mixed compounds are formed in the subsurface
support layer, for example, a solid solution of Pt+4 in
cerium aluminate, which decomposes to evolve into
disperse platinum upon calcination in air at ~720°ë or
reduction with hydrogen at 500°ë. As a result, disperse
platinum is formed and strongly interacts with the sup-
port and is therefore resistant to sintering. The atomic
catalytic activity of this platinum is lowered compared
to platinum supported on unmodified γ-Al2O3. At the
same time, the thermal treatment of platinum supported
on pure γ-Al2O3 oxide at temperatures >750°C signifi-
cantly enhances the specific catalytic activity per 1 m2
of platinum in the complete oxidation of n-butane,
which is due to a change in the Pt+n/Pt0 ratio on the sur-
face of the platinum crystals, as was assumed in [8]. It
was pointed out in [9] that the activity of ionic Pt in
hydrocarbon complete oxidation is substantially lower
than that of metallic platinum. This was confirmed in
[10] by the fact that the activity of Pt/γ-Al2O3 was max-
imal when Pt was only partially oxidized. When the
An increase in the atomic catalytic activity of plati-
num in aluminum–platinum catalysts for complete oxi-
dation with an increase in the calcination temperature
was reported in other papers. It was found in [8, 9, 11]
by XRD and electron microscopy that an increase in
Tcal of aluminum–platinum catalysts from 500 to
1000°ë results in the polymorphic transformation of
the support Al2O3(γ
(δ, θ)
α) and strong Pt
sintering, whereas both the starting and calcined alumi-
num–platinum catalysts are polydispersed. The mor-
phology of the supported platinum crystallites also
changes dramatically with an increase in the calcina-
tion temperature. Thermal treatment in air at 750–
980°C results in the transformation of initially almost
spherical small particles of platinum into large crystals
with a regular shape and clear faceting. An increase in
the size of the supported metal crystallites with an
increase in the temperature of preliminary treatment
leads to a decrease in the donor–acceptor interaction
with the support. This results in the enhancement of the
electron density on the surface atoms of the supported
metal and, hence, the catalytic activity [12, 13]. It was
shown in these papers that the strength of the Pt–éads
bond in the calcined samples does not change with Tcal,
0023-1584/02/4303-0379$27.00 © 2002 MAIK “Nauka /Interperiodica”

380
CHZHU et al.
although the size and shape of the Pt crystallites change number of surface platinum atoms in 1 g of the catalyst
essentially due to sintering.
(atom/g).
The NPt value was calculated according to the for-
mula
The effect of the atomic catalytic activity enhance-
ment with an increase in the preliminary calcination
temperature prompted us to study the thermal activa-
tion of the Pt/MOx system, where MOx = Al2O3, GeO2,
La2O3, and ZrO2.
NPt = (0.01CPtDPtNA)/MPt,
where CPt is the platinum concentration in the catalyst
(wt %), DPt is the dispersion of supported platinum, NA
is the Avogadro number, and MPt is the platinum atomic
weight.
EXPERIMENTAL
Supported platinum dispersion (D) was estimated
by ç2–é2 titration in an all-sealed glass setup accord-
ing to a known procedure [15] (except for ëÂé2 sam-
ples) including the following main stages:
(1) Drying the sample loaded into the reactor in a
vacuum at 120°ë for 2 h;
(2) Reduction of the sample at 500°ë with a gradual
increase in temperature in the reactor with hydrogen
circulation and freezing out the reaction products at T =
–196°ë for 1 h;
(3) Blank titration with oxygen at room temperature
for 20 min;
Catalyst preparation. Supports Al2O3, La2O3,
CeO2, and ZrO2 in the form of rods were prepared by
the extrusion of nitric acid-peptized hydroxides prelim-
inarily precipitated with ammonia from the corre-
sponding metals (analytical grade) followed by drying
at T = 120°ë and calcination at T = 500°ë for 4 h
(600°C for La2O3). A fraction of 0.5–1.0 mm was cho-
sen for subsequent platinum deposition.
Catalysts were prepared by the impregnation of the
above fraction of the ZrO2 and La2O3 supports with an
acetone solution of platinum carbonyl followed by ace-
tone evaporation in an argon flow and drying the sam-
ples at 100°ë for 1 h; in the case of the CeO2 and Al2O3
supports, the catalysts were prepared by platinum car-
bonyl chemisorption. Thermal treatment was carried
out in a muffle oven by calcination in air at 500, 600,
and 700°ë for 2 h.
(4) Hydrogen chemisorption at 150°ë for 30 min;
and
(5) Titration of adsorbed hydrogen with a specified
oxygen volume.
The specific surface area of the samples was mea-
sured on a Sorpty-1750 instrument using the BET
method from nitrogen adsorption at p = 135 torr and
T = 77 K.
The phase composition of the supports and cata-
lysts was studied by X-ray phase analysis (a DRON-3
instrument, ëÓKα irradiation with a β filter behind the
a sample and in front of a counter). Diffraction patterns
were obtained by powder diffractometry according to
the Bregg–Brentano method.
RESULTS AND DISCUSSION
X-Ray phase analysis showed that
(1) The Pt/γ-Al2O3 sample is a two-phase system
consisting of metallic platinum crystals and the γ-Al2O3
phase.
(2) The Pt/ZrO2 sample consists of three crystalline
phases: metallic platinum crystals, monoclinic ZrO2,
and tetragonal-lattice ZrO2. The monoclinic structure
prevails in all three samples calcined at 500, 600, and
700°ë, and the relative amount of the tetragonal phase
decreases with an increase in the calcination tempera-
ture, whereas that of the monoclinic phase increases.
Chemical analysis of the catalysts for platinum was
performed according to a known procedure [14].
The catalytic activity of the samples was tested in
n-pentane complete oxidation in a flow-circulation
setup under gradientless conditions in the kinetic
regime at 300°ë. The catalyst loading was ~1.5 g. The
starting gas mixture contained n-pentane (0.1 vol %)
and air (99.9 vol %). The gas flow rate through the reac-
tor with a sample was varied in the range 10–60 l/h to
obtain 50% conversion of the reactant. The circulation
rate was ~800 l/h. The reaction mixture was analyzed
before and after the reactor on an LKhM-8 MD chro-
matograph (ionization-flame detector, 3 mm × 2 m
stainless steel column packed with Polisorb-1). The
reaction was monitored by the change in concentration
of unreacted n-pentane at the reactor outlet. No peaks,
except for that of n-pentane, were found.
(3) The Pt/CeO2 sample consists of two phases:
metallic platinum and CeO2.
(4) The Pt/La2O3 sample is a two-phase system con-
sisting of metallic platinum crystals and La(OH)3
formed during slow cooling of the sample in air. The
relative concentration of La(OH)3 in the support
declines with an increase in the calcination tempera-
ture, which is accompanied by the appearance of a new
phase, lanthanum carbonate La2O3. Nevertheless, we
preserved the designation Pt/La2CO3 for this sample.
The atomic catalytic activity (ACA) of the samples
was calculated using the following formulas:
WÄCÄ = Wsp/NPt,
where WÄCÄ is the atomic catalytic activity
The total specific surface areas of the supports deter-
(cm3 atom–1 s–1); Wsp is the specific catalytic activity mined by the BET method were (m2/g): γ-Al2O3, 201;
calculated per 1 g of the catalyst (cm3 g–1 s–1); NPt is the ZrO2, 39; CeO2, 52; and La2O3, 24.
KINETICS AND CATALYSIS Vol. 43 No. 3 2002

EFFECT OF THERMAL ACTIVATION
381
14
12
10
8
35
0.2
0.5
0.4
0.3
0.2
4
30
25
20
15
10
3
1
2
3
0.1
4
2
6
0.1
0
4
2
1
5
0
0
800
400
500
600
T, °ë
700
800
400
500
600
T, °ë
700
Fig. 2. The atomic catalytic activity of supported platinum
catalysts in n-pentane complete oxidation at 300°C as a
function of the temperature of preliminary calcination for
Fig. 1. The activity of 1 g of the catalyst (1 wt % Pt) in n-
pentane complete oxidation at 300°C as a function of the
temperature of preliminary calcination over the samples:
the samples: (1) Pt/γ-Al O , (2) Pt/ZrO , (3) Pt/CeO , and
2
3
2
2
(1) Pt/γ-Al O , (2) Pt/ZrO , (3) Pt/CeO , and (4) Pt/La O .
2
3
2
2
2 3
(4) Pt/La O .
2
3
Figure 1 shows that the per-gram activity of sup-
ported platinum in the Pt/CeO2 catalyst drops with an
increase in the calcination temperature. The above
activity of the Pt/ZrO2 catalyst also drops and is maxi-
mal in its absolute value among all the samples calcined
at 500°C. The activity of the Pt/γ-Al2O3 sample after
calcinations at 500–600°ë remains practically constant
within the accuracy of the experiment and decreases
after calcination at Tcal = 700°ë. Pt/La2O3 has the low-
est activity, and its variation shown in Fig. 1 is within
the experimental error.
determine the atomic catalytic activity because of the
abnormally high é2 consumption. This fact is due to
the oxidation of the surface ëÂé2 layer preliminarily
reduced with hydrogen. To estimate the platinum dis-
persion in Pt/CeO2 catalysts, we developed a procedure
based on CO adsorption at room temperature on the
surface of preliminarily treated samples. The pretreat-
ment of samples included drying in a vacuum, reduc-
tion at 350°ë for 2 h in the reactor with hydrogen cir-
culation, freezing out the reaction products at T =
−196°ë, and blank titration with oxygen at room tem-
perature followed by hydrogen chemisorption at 150°ë
for 30 min. Then, ç2 is desorbed from the catalyst sur-
face by evacuation at 350°ë for 1–1.5 h and cooled to
room temperature in a vacuum. The study was carried
out in the same setup that was used for ç2–é2 titration.
The atomic catalytic activity of the Pt/CeO2 samples
calcined at 700°ë was ~1.5 times higher than the activ-
ities of samples calcined at 500 and 600°ë. The exper-
imental error including the error in the determination of
the catalytic activity and chemisorption measurements
was ~20%.
Thus, Wsp of all catalysts decreases with an increase
in the preliminary calcination temperature due to a
decrease in the platinum dispersion (see table).
As can be seen from the table and Fig. 2, the atomic
catalytic activity of the Pt/γ-Al2O3 samples calcined at
700°ë is higher than that of the samples calcined at 500
and 600°ë by approximately an order of magnitude. A
similar dependence is seen for the supported Pt/ZrO2
catalyst. However, the sample calcined at 700°ë was
~3 times more active than that calcined at 500°ë. For
Pt/La2O3, no clear dependence follows from the above
data because of the poor activity and possible experi-
mental error. However, it can be qualitatively con-
cluded that the atomic catalytic activity of platinum
decreases upon the calcination of this sample at 600–
700°C.
To compare the catalytic properties of samples, two
main assumptions were used in the calculation of the
atomic catalytic activity:
(1) All the surface atoms are active (common sug-
inum dispersion in the Pt/CeO2 catalyst, we failed to gestion);
When ç2–é2 titration was used to measure the plat-
KINETICS AND CATALYSIS Vol. 43 No. 3 2002

382
CHZHU et al.
Characteristics of the catalysts calcined at various temperatures
WACA × 1022,
cm3 (atom Pt)–1 s–1
W × 103, cm3 g–1 s–1
Sample
T
cal, °C CPt, wt % Ssp, m2/g W × 103, cm3 g–1 s–1
D
per 1% Pt
Pt/Al2O3
Pt/Al2O3
Pt/Al2O3
Pt/ZrO2
Pt/ZrO2
Pt/ZrO2
Pt/CeO2
Pt/CeO2
Pt/CeO2
Pt/La2O3
Pt/La2O3
Pt/La2O3
500
600
700
500
600
700
500
600
700
500
600
700
0.97
0.97
0.97
0.84
1.1
237
185
180
57
4.2
4.5
3.8
11.4
8.4
7.0
6.7
3.0
2.3
0.2
0.2
0.1
4.3
4.6
3.9
13.6
7.6
5.0
11.2
5.0
3.8
0.2
0.2
0.1
0.49
0.47
0.04
0.94
0.39
0.14
0.76
0.33
0.17
0.14
0.37
0.29
2.8
3.2
33
4.7
16
6.4
1.4
15
12
0.6
48
4.8
0.6
45
4.9
0.6
32
7.2
1.0
24
0.46
0.17
0.11
1.0
22
1.0
28
(2) The number of the surface Pt atoms change Similarly to platinum supported on γ-Al2O3 modified
insignificantly on going from an oxidative to a reduc- with the above oxides, one can suggest that the thermal
tive medium [8, 9, 16].
activation effect is due to the acceleration of the hydro-
gen activation stage.
As can be seen from the data above, the thermal acti-
vation effect was found for the Pt/Al2O3, Pt/ZrO2, and
Pt/CeO2 catalysts. Note that the platinum dispersion in
the Pt/La2O3 sample passes through a maximum with
an increase in the calcination temperature, whereas the
platinum dispersion in other systems decreases. This
behavior of Pt/La2O3 during calcination was not studied
in detail due to the instability of the support composi-
tion in air.
ACKNOWLEDGMENTS
This work was supported by the Russian Foundation
for Basic Research (project no. 99-03-32135).
REFERENCES
1. Tsyrul’nikov, P.G., Shitova, N.B., Kudrya, E.N., and
Savel’eva, G.G., in Zakonomernosti glubokogo okisle-
niya veshchestv na tverdykh katalizatorakh (Complete
Oxidation of Substances on Solid Catalysts), Novosi-
birsk, 2000, p. 282.
2. Tsyrul’nikov, P.G., Sal’nikov, V.S., Drozdov, V.A.,
Stuken, S.A., Bubnov, A.V., Grigorov, E.I., Kalinkin, A.V.,
and Zaikovskii, V.I., Kinet. Katal., 1991, vol. 32, no. 2,
p. 439.
To summarize, this work showed that the effect of a
sharp increase in the atomic catalytic activity of sup-
ported platinum catalysts takes place in the following
systems (in order of decreasing effect): Pt/γ-Al2O3,
Pt/ZrO2, and Pt/CeO2. The per-gram activity of all the
samples decreases regularly with an increase in the cal-
cination temperature.
As was mentioned above, the strength of the Pt–Oads
bond in the calcined Pt/γ-Al2O3 and Pt–Ce(La,Zr)/γ-
Al2O3 samples does not change with an increase in Tcal
despite a significant increase in the size and shape of
the Pt crystallites upon sintering [12, 13]. This means
that the strength of the oxygen bond in these samples
does not determine a change in the platinum specific
catalytic activity per 1 m2 in hydrocarbon complete oxi-
dation upon the calcination of the catalysts at elevated
temperatures. Hence, an increase in SCA Pt or atomic
catalytic activity (ACA Pt) of the calcined samples, that
is, the appearance of the thermal activation effect, is
due to the acceleration of the hydrocarbon activation
stage, that is, a change in the number and/or nature of
the chemisorption and activation sites.
3. Kochubei, D.I., Kriventsov, V.V., Kustova, G.N., Ode-
gova, G.V., Tsyrul’nikov, P.G., and Kudrya, E.N., Kinet.
Katal., 1998, vol. 39, no. 2, p. 294.
4. Tsybulya, S.V., Kryukova, G.N., Vlasov, A.A., Boldy-
reva, N.N., Kovalenko, O.N., and Tsyrul’nikov, P.G.,
React. Kinet. Catal. Lett., 1998, vol. 64, no. 1, p. 113.
5. Tsyrul’nikov, P.G., in Kataliz i katalizatory. Fundamen-
tal’nye issledovaniya (Catalysis and Catalysts: Basic
Studies), Novosibirsk: Inst. of Catal., 1998, p. 39.
6. Drozdov, V.A., Tsyrul’nikov, P.G., Popovskii, V.V.,
Davydov, A.A., and Pestryakov, A.N., Kinet. Katal.,
1988, vol. 29, no. 2, p. 484.
7. Moroz, E.M., Ushakov, V.A., Dzhunusov, A.K.,
Drozdov, V.A., and Tsyrul’nikov, P.G., React. Kinet.
Catal. Lett., 1990, vol. 41, no. 1, p. 109.
We have no data on the bond strength for chemi-
sorbed oxygen or hydrocarbon with a surface of plati-
num supported on pure oxides ZrO2, CeO2, and La2O3.
8. Drozdov, V.A., Kochubei, D.I., Tsyrul’nikov, P.G., and
Popovskii, V.V., Kinet. Katal., 1989, vol. 30, no. 4,
p. 879.
KINETICS AND CATALYSIS Vol. 43 No. 3 2002

EFFECT OF THERMAL ACTIVATION
383
9. Drozdov, V.A., Cand. Sci. (Chem.) Dissertation, Novosi- 14. OST 38.01267.1–82, Alumina–Platinum Catalysts: A
birsk: Inst of Catal., 1987.
Method for Platinum Determination; Ginzburg, S.I. and
Gladelshevskaya, S.A. Rukovodstvo po khimicheskomu
analizu platinovykh metallov i zolota (Manual on Chem-
ical Analysis of Platinum Group Meytals and Gold),
Moscow: Nauka, 1965, p. 159.
10. Burch, R., Crittle, D.J., and Hayes, M.J., Catal. Today,
1999, vol. 47, p. 229.
11. Drozdov, V.A., Tsyrul’nikov, P.G., Popovskii, V.V., et al.,
React. Kinet. Catal. Lett., 1985, vol. 27, no. 2, p. 425.
15. Karnaukhov, A.P., Sibirskie chteniya po katalizu (Sibe-
rian Lectures on Catalysis), Novosibirsk, 1979, part 1,
p. 136.
12. Muller, C.A., Maciewski, M., Koeppel, R.A., and
Baiker, A., Catal. Today, 1999, vol. 47, p. 245.
13. Golodets, G.I., React. Kinet. Catal. Lett., 1981, vol. 17, 16. Luk’yanova, Z.V., Shekhobalova, V.I., and Boronin, V.S.,
nos. 1–2, p. 97.
Zh. Fiz. Khim., 1979, vol. 53, no. 2, p. 410.
KINETICS AND CATALYSIS Vol. 43 No. 3 2002