The effect of support nature and the temperature of preliminary calcination on the atomic catalytic activity of supported palladium catalysts for the complete oxidation of hydrocarbons

Article abstract of DOI:10.1023/B:KICA.0000032177.05279.4b

It is shown that for palladium catalysts supported on various supports, there is no effect of thermal activation consisting of an increase in the turnover number (TON) of palladium with an increase in the temperature of sample calcination from 500 to 700°

Full text of DOI:10.1023/B:KICA.0000032177.05279.4b

Kinetics and Catalysis, Vol. 45, No. 3, 2004, pp. 406–413. Translated from Kinetika i Kataliz, Vol. 45, No. 3, 2004, pp. 432–439.
Original Russian Text Copyright © 2004 by Chzhu, Tsyrul’nikov, Kryukova, Borbat, Kudrya, Smolikov, Bubnov.
The Effect of Support Nature and the Temperature
of Preliminary Calcination on the Atomic Catalytic
Activity of Supported Palladium Catalysts
for the Complete Oxidation of Hydrocarbons
D. P. Chzhu*, P. G. Tsyrul’nikov*, G. N. Kryukova**, V. F. Borbat***,
E. N. Kudrya*, M. D. Smolikov*, and A. V. Bubnov*
Boreskov Institute of Catalysis (Omsk Branch), Siberian Division, Russian Academy of Sciences, Omsk, 644053 Russia
* Boreskov Institute of Catalysis, Siberian Division, Russian Academy of Sciences, Novosibirsk, 630090 Russia
*
*
*
** Omsk State University, Omsk, 644077 Russia
Received September 20, 2002
Abstract—It is shown that for palladium catalysts supported on various supports, there is no effect of thermal
activation consisting of an increase in the turnover number (TON) of palladium with an increase in the temper-
ature of sample calcination from 500 to 700°C, as was observed in the case of similar supported platinum cat-
alysts; for two palladium–alumina catalysts calcined at 500°C and differing in the concentration of palladium
and dispersity, the TON of low-dispersity palladium is substantially higher than the TON of high-dispersity pal-
ladium; all other conditions being the same, the TON of supported palladium is determined by the support
nature.
INTRODUCTION
does not depend on the size of the metal particles,
whereas Baldwin and Burch [8] reported a change in
Hydrocarbon oxidation processes over supported the TON but without any relation to the particle size.
platinum and palladium catalysts have been discussed
in a huge number of papers. However, as follows from
the published data, there is no common opinion regard-
ing the relationships between the structural and surface
changes of the catalysts in the course of preliminary
treatment and a reaction and changes in the catalyst
activity.
Earlier we established that the effect of a drastic
increase in the TON after catalyst calcination in air at
6
00–700°C, which has been called the effect of thermal
activation, reveals itself over the supported catalysts
Pt/MO (where MO is Al O , CeO , La O , and ZrO )
x
x
2
3
2
2
3
2
in the complete oxidation of n-pentane [9]. One might
expect that this phenomenon is only characteristic of
platinum catalysts. In this connection, the goal of the
study was to search for ways to change the TON of sup-
ported palladium in model reactions of the complete
oxidation of methane and n-pentane after thermal treat-
ment of the samples in air at elevated temperatures, and
finding the relationships between the TON and the size
of the metal particles.
One of the important characteristics of platinum and
palladium catalysts that determines whether a reaction
is structure-sensitive is the turnover number (TON),
which is the activity per one surface atom of a sup-
ported metal.
In several recent publications, the relationships
between the TON, the state of the active component and
the size of the metal particles on the catalyst surface
were studied for hydrocarbon oxidation reactions. Briot
et al. [1, 2] showed that the catalytic combustion of
methane on platinum and palladium catalysts supported
on alumina is structure-sensitive, because the TON
increases with an increase in the size of the particles.
An analogous conclusion was drawn for the catalytic
oxidation of propylene [3] and n-heptane [4] over the
same catalytic systems. Hicks et al. [5, 6] studied the
behavior of supported platinum and palladium in the
reaction of methane oxidation at 300°C. On both met-
EXPERIMENTAL
Catalyst preparation. Support materials Al O ,
2
3
ZrO , and CeO , formed as chips, were prepared by
2
2
extrusion of hydroxides peptized with nitric acid;
hydroxides were preliminarily precipitated with ammo-
nia from the nitrates of the corresponding metals (ana-
lytical purity grade) with further drying at 120°C and
calcination at 500°C for 4 h. To prepare another sup-
port, ZrO was calcined at 750°C for 4 h.
2
The support (0.3 mol) CeO –(0.7 mol) ZrO and
2
2
als, the reaction was also structure-sensitive. Cullis and supports based on modified alumina (Al O –CeO ,
Willat [7] found for methane oxidation that the TON Al O –ZrO , and Al O –La O ) in the form of chips
2
3
2
2
3
2
2
3
2
3
0
023-1584/04/4503-0406 © 2004 MAIK “Nauka /Interperiodica”

THE EFFECT OF SUPPORT NATURE AND THE TEMPERATURE
407
were prepared by the extrusion of hydroxides prelimi- tion) or 300°C (in the case of n-pentane oxidation) at
narily coprecipitated with ammonia from the nitrates of 50% conversion, which corresponded to a concentra-
the corresponding metals (analytical purity grade) and tion of methane of 0.25 vol % or n-pentane of 0.05 vol %.
peptized with nitric acid. They were dried at 120°C and
The TONs of the samples were calculated using the
following formulas:
calcined at 500°C for 4 h. The 0.5–1.0 mm fraction was
used to further support the palladium.
Palladium catalysts were obtained by the impregna-
WTON = W/NPd,
(1)
tion of the same fraction of ZrO and CeO –ZrO with
2
2
2
the solution of palladium nitrate. The samples based on
3
–1 –1
where W
is the turnover number (ncm CH atom s
4
TON
CeO and γ-Al O were prepared by chemisorption
3
–1 –1
2
2
3
or ncm C H atom s ), W is the activity per 1 g of
5 12
from the solution of palladium nitrate at room temper-
ature. For the modified supports Al O –CeO (7.2 wt %
3
–1 –1
3
–1 –1
the catalyst (ncm CH g s or ncm C H g s );
4
5
12
2
3
2
and N is the number of surface palladium atoms per
1 g of the catalyst (atom/g),
Pd
CeO ), Al O –ZrO (4.9 wt % ZrO ), and Al O –La O
2
2
3
2
2
2
3
2
3
(
7.2 wt % La O ), palladium catalysts were obtained by
2 3
chemisorption from the solution of palladium nitrate at
room temperature. The samples were then dried at
NPd = (0.01C D N )/M ,
(2)
Pd Pd
A
Pd
1
20°C for 1 h. Thermal treatment of the samples was
carried out in a muffle furnace by calcination in air at where C is the concentration of palladium in the cat-
Pd
5
00, 600, and 700°C for 2 h.
alyst (wt %), D is the dispersity of supported palla-
Pd
dium, N is Avogadro’s number, and M is the molar
A
Pd
The phase composition of the support and palla-
weight of palladium.
dium catalysts were studied by XRD (on a DRON-3
The experimental error, including the error in deter-
mining the catalytic activity and chemisorption mea-
surements, was ~20%.
instrument using MoK irradiation with a β-filter after
α
the sample before a counter (for Pd/Al O , Pd/CeO , and
2
3
2
Pd/La O ) and CuK irradiation with a β-filter after the
2
3
α
sample before a counter (for Pd/ZrO , Pd/Al O –CeO ,
The dispersity of supported palladium (D ) was
2
2
3
2
Pd
and Pd/Al O –ZrO )). Samples were analyzed by the determined using H –O titration in the all-soldered
2
3
2
2
2
Bregg–Brentano method of powder diffractometry.
glass setup using the known technique [11] (except for
the samples that had CeO in their composition). The
2
Electon microscopic imaging of the samples was
carried out using a transmission electron microscope
JEM-2010 (resolution, 0.14 nm; accelerating voltage,
determination of dispersity included the following
main steps:
(
1) Drying the sample charged into the reactor at a
low pressure and 120°C for 2 h;
2
00 kV). The samples were prepared for study by sup-
porting from an alcohol suspension on an ultrasound
dispersant UZDN-3.
(
2) Sample reduction at 500°C (with a continuous
increase in the temperature from room temperature) in
the reactor with hydrogen circulation and freezing (at
T = –196°C) the reaction products for 1 h and further
vacuuming;
Chemical analysis of catalysts for determining the
palladium concentration was carried out according to
the known technique [10].
The catalytic activity of the samples was deter-
(3) “Blank” titration with oxygen at room tempera-
mined in the kinetics-controlled regime using a flow- ture for 20 min without checking the result with further
circulation setup under gradientless conditions in the vacuuming;
model reactions of methane and n-pentane oxidation.
(4) Hydrogen chemisorption at 150°C for 30 min;
The catalyst loading was ~1.5 g. In the case of CH oxi-
4
and
dation, the initial gas mixture consisted of methane
(
0.5 vol %) and air (99.5 vol %). In the case of n-pentane
(5) Titration of adsorbed hydrogen by a certain vol-
oxidation, the initial mixture consisted of n-pentane ume of oxygen.
(
0.1 vol %) and air (99.9 vol %). The rate of the gaseous
The dispersity of catalyst supported on CeO or on
2
mixture flow through the reactor with a sample was var-
ied from 10 to 60 l/h. The circulation rate was ~800 l/h.
Analysis of the reaction mixture before and after the
reactor was carried out using an LKhM-8 MD chro-
matograph with a flame-ionization detector and a stain-
less steel 3 mm × 2 m column packed with Polisorb-1.
The reaction was controlled by changes in the concen-
tration of unreacted methane (or n-pentane) at the reac-
tor inlet. No other peaks except for the peak of methane
modified Al O was determined by CO adsorption at
2
3
room temperature on the surface of samples preliminar-
ily prepared in the same setups where H –O titration
2
2
was carried out. The method for determining the disper-
sity included the following steps:
(
1) Drying the samples charged into the reactor at a
–
2
low pressure (10 Torr) and 120°C for 2 h;
(2) Sample reduction at 350°C for 2 h (with a con-
(
or n-pentane) were observed. The TON of the samples tinuous temperature increase from room temperature)
was compared at 500°C (in the case of methane oxida- in the reactor with hydrogen circulation and freezing
KINETICS AND CATALYSIS Vol. 45 No. 3 2004

4
08
CHZHU et al.
increases. These data suggest that a portion of the zir-
conia is initially in an amorphous state and crystallizes
to form the monoclinic modification with an increase in
the temperature of calcination.
In the Pd/ZrO sample (T
ZrO , 750°C) calcined
2
calcin
2
at 500 and 700°C, the above three crystalline phases are
also present, but zirconia is basically monoclinic
because it was preliminarily calcined at 750°C.
3
. The Pd/CeO sample after calcination in air at 500
2
and 700°C consists of two phases: metallic palladium
crystals and the CeO phase.
2
4
. The Pd/Al O –CeO sample calcined at 500 and
2 3 2
1
0 nm
7
00°C consists of three phases: metallic palladium
crystals, the γ-Al O phase, and the CeO with a coher-
2
3
2
ent scattering region (CSR) d ~ 35–40 Å.
Fig. 1. Microscopic image of the freshly prepared sample
%Pd/Al O –CeO , which was preliminarily calcined in
2
2
3
2
air at 500°C (2500000 magnification).
5. The Pd/Al O –ZrO sample after calcination at
2
3
2
5
00°C contains the phases of γ-Al O and PdO. After
2
3
calcination in air at 700°C, the cubic ZrO phase (CSR
2
(
at T = –196°C) the reaction products and further vacu- d ~ 80 Å) and palladium crystals also appear.
–2
uming to 10 Torr;
3) “Blank” titration with oxygen at room tempera-
ture (without documenting the result) for 20 min and
further vacuuming);
Electron microscopic images of modified palla-
dium–aluminum samples 2% Pd/Al O –CeO are
(
2
3
2
shown in Figs. 1–3. The initial freshly prepared sample
calcined at 500°C in air but not studied in the oxidation
reaction (Fig. 1) is characterized by an average size of
supported palladium particles of ~10 nm, although
some particles are elongated. The high-resolution
image (Fig. 1) shows that the particles have cuttings
and their internal structure is well ordered. Figure 2
shows the same sample calcined in air at 500°C and
(
4) Hydrogen chemisorption at 150°C (for platinum
catalysts) or at room temperature (for palladium cata-
lysts) for 30 min;
–
2
(
5) H desorption from the catalyst surface (10 Torr)
2
with an increase in temperature from 150 to 350°C for
1
–1.5 h;
(
6) Cooling to room temperature at a low pres- tested in methane oxidation. Upon the reaction the size
sure; and
of the supported particles insignificantly increases (up
to 15 nm). These are still isolated particles with well
cutting (Fig. 2a). It is likely that alumina particles inter-
act with ceria particles, which reveals itself in the form
of a characteristic regular contrast on high-resolution
images of the support (Fig. 2b). The interaction of alu-
mina and ceria and the formation of aluminum phases
were noted in [12]. After calcination in air at 600°C,
particles are sintered to form a size of 40 nm, but there
are some particles with a size of 10 nm (Fig. 3b, where
palladium crystallites are shown as dark spots). Sinter-
ing was disordered with the formation of noncoherent
boundaries between particles, where the structure of the
material is strongly distorted (Fig. 3a). Isolated parti-
cles with a size of 10 nm have a well-ordered internal
structure and are characterized by regular cutting
(
7) CO chemisorption on the sample prepared.
The total BET surface area of the samples was
measured using a Sorpty-1750 instrument from single-
point nitrogen adsorption at P = 135 Torr and T = 77 K.
RESULTS AND DISCUSSION
The XRD analysis of the samples showed that
1
. The Pd/γ-Al O samples (with palladium concen-
2
3
trations of 1 and 2 wt %) after preliminary calcination
at 500 and 700°C are a two-phase system consisting of
palladium metal crystals and the phase of γ-Al O .
2
3
However, for the sample Pd/γ-Al O (1 wt % Pd) cal-
2
3
cined at 600°C, a new PdO phase appears in addition to
the two phases mentioned above, although the corre-
sponding peak is poorly resolved.
(
Fig. 3b).
2
. The Pd/ZrO sample (T
ZrO , 500°C) consists
Thus, as follows from the electron microscopic data,
2
calcin
2
of three crystalline phases: metallic palladium, ZrO
with an increase in the calcination temperature of the
2
with monoclinic and tetragonal structures. In the sam- Pd/Al O –CeO samples to 600°C, the formation of
2
3
2
ple calcined at 500°C, the monoclinic structure domi- noncoherent boundaries between sintered palladium
nates over the tetragonal one. With an increase in the crystallites takes place. This phenomenon can be asso-
calcination temperature to 700°C, the intensity of the ciated with the fact that in the course of sintering the
peak of tetragonal phase in the XRD pattern remains CeO fragments migrate together with the palladium
2
the same, whereas that of the monoclinic phase crystallites with which they interact.
KINETICS AND CATALYSIS Vol. 45 No. 3 2004

THE EFFECT OF SUPPORT NATURE AND THE TEMPERATURE
409
1
0 nm
10 nm
(
‡)
(‡)
1
0 nm
100 nm
(
b)
(b)
Fig. 2. Microscopic image of (a) the sample 2% Pd/Al O –
Fig. 3. Microscopic image of the sample 2% Pd/Al O –
2 3
2
3
CeO , tested in the complete oxidation of methane
CeO calcined at 600°C and tested in the reaction of the
2
2
(
2500000 magnification) and (b) the support.
complete oxidation of methane ((a) 2500000 magnifica-
tion, (b) 200000 magnification).Arrows show the regions of
the formation of noncoherent boundaries of the particles,
where the structure of the material is strongly distorted.
The total BET specific surface area for the supports
2
was (m /g): γ-Al O , 201; ZrO , 39; CeO , 52.
2
3
2
2
of the sample somewhat decreases with an increase in
Tcalcin in air from 500 to 700°C. Palladium catalysts
based on CeO and CeO –ZrO have about the same
Catalytic Activity of the Samples
2
2
2
The specific catalytic activity of the samples catalytic activity in methane oxidation; their activity is
based on 1 g of the catalyst) for the process of the an order of magnitude lower than for the catalysts sup-
(
complete oxidation of methane. The characteristics of ported on Al O and ZrO . With an increase in the tem-
2
3
2
the 2% Pd catalysts supported on γ-Al O , ZrO (T
perature of preliminary calcination from 500 to 700°C,
2
3
2
calcin
5
00 and 750°C), and CeO , as well as CeO (0.3 mol)– the catalytic activity of these samples nearly halves.
2
2
ZrO (0.7 mol) and calcined at two different tempera-
2
For 1% Pd/γ-Al O (Table 1), the catalytic activity
2
3
tures in air (500 and 700°C), are shown in Table 1. As
follows from these data, the most active systems are
supported palladium–aluminum catalysts, whose cata-
lytic activity changes insignificantly with an increase in
in the complete oxidation of methane decreases by a
factor of ~1.5 with an increase in the calcination tem-
perature from 500 to 700°C and becomes much lower
(
~2–3 times lower) than the activity of the correspond-
the temperature of calcination in air. The Pd/ZrO sam-
2
ing 2% sample.
ples are also rather active in this reaction, and their
activity is close to that of Pd/Al O . Note that an
Thus, in methane oxidation, the activity per 1 g
2
3
increase in the calcination temperature of the support decreases with an increase in the calcination tempera-
ZrO from 500 to 750°C leads to an increase in the cat- ture from 500 to 700°C for all supported palladium cat-
2
alytic activity by a factor of ~1.5. As in the case of alu- alysts. However, the apparent decrease in the activity
minum–palladium catalysts, the catalytic activity of 1 g with an increase in the calcination temperature for var-
KINETICS AND CATALYSIS Vol. 45 No. 3 2004

4
10
CHZHU et al.
Table 1. Characteristics of supported palladium catalysts for the complete oxidation of CH calcined at various temperatures
4
Pd concen-
tration,
wt %
22
3
S ,
m /g
W
(CH ) × 10 ,
Tcalcin
,
sp
2
W(CH ) × 10 ,
4
AKA
3
4
No.
Sample
DPd
3
–1 –1
°C
ncm g s
–1 –1
ncm (atom Pd) s
1
2
3
4
5
6
7
8
9
Pd/Al O
500
600
700
500
700
244
198
183
203
168
–
0.24
0.16
0.13
0.11
0.09
0.10
0.06
0.06
0.05
0.29
0.13
0.18
0.05
69
51
51
56
2
3
″
″
1
48
65
Pd/Al O
158
140
100
71
132
140
91
2
3
2
2
2
2
2
″
Pd/ZrO (T
ZrO = 500°C) 500
2
calcin
2
″
700
–
104
218
225
3.6
4
Pd/ZrO (T
ZrO = 750°C) 500
44
148
128
12
2
calcin
2
″
700
500
700
44
1
1
1
1
0 Pd/CeO2
49
1
″
35
6
2 Pd/CeO (0.3)/ZrO (0.7)
500
700
49
18
9
2
2
3
″
17
4
7
ious samples differs. This is due to the fact that the the active metal state as follows from [6]. For the spe-
activity is proportional to the dispersity of the sup- cies formed after sample calcination in air under the
ported metal, and a decrease in the dispersity for all reaction conditions, some time is required for their
studied samples that are thermally treated under the transition into the active state. A similar phenomenon
same conditions differs. Therefore, a decrease in the of the increase in the reaction rate of methane oxidation
activity also differs.
per 1 g of the catalyst in the course of the reaction has
been observed for the systems Pd/Al O , Pd/ZrO , and
2
3
2
The dynamics of the activity of platinum and palla-
dium supported on alumina were studied in methane
combustion at 200–600°C [1, 2]. The authors of those
papers established that the Pd/Al O sample treated at
Pd/SiO –Al O calcined in air at 500–850°C [13]. To
2
2
3
observe a noticeable increase in the reaction rate of
methane oxidation with time, several tens of hours are
needed. In our case the experimental times were much
shorter.
2
3
6
00°C in a methane–oxygen–nitrogen mixture for 14 h
is much more active (based on 1 g of the catalyst) than
the same sample calcined at 500°C in a flow of argon
and reduced by hydrogen. They also observed a
decrease in the dispersity of the catalyst “aged” under
the conditions of the reaction medium by a factor of ~2
compared to the sample reduced in hydrogen. Note that
samples aged in the absence of methane (i.e., in an oxy-
gen–nitrogen mixture) at 600°C did not result in an
increase in the catalytic activity. Analogous results
were obtained for the platinum–alumina sample. These
observations led the authors of [1, 2] to conclude that
the presence of both reactants (methane and oxygen) is
a necessary condition for an increase in the catalyst
activity. In our case, all samples were calcined in air
and the activity per 1 g of the catalyst naturally
decreases with a decrease in the dispersity. This differ-
ence can be explained by the fact that an increase in the
Specific catalytic activity of the samples (per 1 g
of the catalyst) for the process of complete n-pen-
tane oxidation. Modification of alumina-supported
palladium catalysts (~1 wt % Pd) by La O and CeO
2
3
2
leads to some decrease in the activity compared to
Pd/Al O , although the dependences of activity on the
2
3
temperature of preliminary calcination are similar. In
the case of alumina modification by zirconia, the activ-
ity per 1 g of the catalyst remains practically the same
as for the nonmodified sample after preliminary calci-
nation at various temperatures. The catalytic activity of
Pd/Al O passes through a slight maximum after pre-
2
3
liminary calcination at 600°C, which is also character-
istic to some degree of the Pd/Al O –La O and
2
3
2
3
Pd/Al O –ZrO systems (Table 2).
2
3
2
The results obtained suggest that modifying the pal-
size of the palladium particles in the atmosphere of the ladium–alumina catalysts by ZrO does not lead to a
2
reaction medium is accompanied by the formation of substantial change in their activity toward the complete
KINETICS AND CATALYSIS Vol. 45 No. 3 2004

THE EFFECT OF SUPPORT NATURE AND THE TEMPERATURE
411
Table 2. Characteristics of supported palladium–aluminum catalysts for the complete oxidation of n-C H calcined at var-
5
12
ious temperatures
Pd concen-
3
22
w × 10 (C H ), W × 10 (C5H12),
2
5
12
No.
Sample
Tcalcin, °C tration, S , m /g Modifier, wt % DPd
sp
3
–1 –1
ncm (atom Pd) s–1
3
–1
ncm g
s
wt %
1
2
3
4
5
6
7
8
9
Pd/Al O
500
600
700
500
600
700
500
600
700
750
500
600
700
1.0
1.0
1.0
1.16
1.16
1.16
1.1
1.1
1.1
1.1
1.1
1.1
1.1
244
198
183
201
180
165
201
210
185
176
181
186
157
0.24
0.16
0.13
0.27
0.23
0.21
0.28
0.24
0.22
0.21
0.24
0.21
0.19
2.9
4.5
3.5
4.0
4.5
3.5
2.1
1.9
2.0
1.6
2.5
3.0
2.5
2.1
4.9
4.7
2.3
3.0
2.5
1.2
1.3
1.4
1.2
1.7
2.3
2.1
2
3
″
–
″
Pd/Al O –ZrO
2
2
3
″
4.9
″
Pd/Al O –CeO
2
2
3
″
7
.2
″
″
1
1
1
1
0
1 Pd/Al O –La O
3
2
3
2
2
3
″
7.2
″
oxidation of n-pentane. The addition of CeO and methane, we observed the fulfillment of Boreskov’s
2
La O to the Pd/Al O system leads to some decrease in rule of constant specific catalytic (TON in our case).
2
3
2
3
the activity of the catalyst in this reaction compared to
the nonmodified sample. Note that modification by the
The TON of the samples in the reaction of com-
plete oxidation of n-pentane. Table 2 shows the
dependences of TON in n-pentane oxidation over palla-
dium–alumina catalysts modified by ZrO , CeO , and
cited rare-earth element oxides and ZrO leads to the
2
stabilization of the values of the specific catalytic activ-
ity of the samples calcined in air at various tempera-
tures.
2
2
La O , on the temperature of their preliminary calcina-
2
3
tion in air. It can be seen from the preliminary data that
the addition of rare-earth element oxides and ZrO to
2
Pd/Al O significantly stabilizes the dispersity of the
2
3
Turnover Number
supported palladium; its value insignificantly decreases
upon preliminary calcination of the catalysts at 700°C.
For the nonmodified sample, the dispersity almost
halves. In the case of palladium–alumina catalysts
modified by ZrO , CeO , and La O , the TON slightly
The TON of the samples in the complete oxida-
tion of methane. The dependence of the TON of the
nonmodified supported palladium samples (1 and 2 wt
2
2
2
3
%
Pd) on the temperature of their preliminary calcina-
changes in the whole calcination temperature range
00–700°C. For the nonmodified 1% palladium–alumi-
num catalyst, the TON increases 2.5 times when T
tion (500 and 700°C) is described in Table 1. In the case
5
of the systems Pd/Al O , Pd/ZrO (T
(ZrO ) =
2
2
3
2
calcin
calcin
5
00°C), and Pd/ZrO (T
(ZrO ) = 750°C) contain-
2
calcin
2
is increased from 500 to 600°C but then remains the
ing 2% Pd, an increase in the calcination temperature
from 500 to 700°C leads to an insignificant increase in
the TON despite a substantial decrease in the dispersity.
same when T is increased from 600 to 700°C.
calcin
Analogous results were obtained in [14]: the stabili-
zation of the TON values for platinum–aluminum cata-
lysts modified by the same rare-earth element oxides
For the Pd/CeO system, the dispersity almost halves
2
with an increase in the calcination temperature, but the
TON increases insignificantly as well. In the case of the
system Pd/CeO (0.3 mol)–ZrO (0.7 mol), a small
and ZrO after the thermal treatment of samples in air.
2
Note a trend toward a decrease in the TON in the
2
2
decrease in the TON with an increase in the calcination series: the nonmodified palladium–alumina catalyst,
temperature is observed. In general, for the catalysts modified by zirconia, and modified by rare-earth ele-
mentioned above and the conditions of their prelimi- ment oxides. As in [14], this can be explained by the
nary treatment in the reaction of complete oxidation of strengthened interaction between a modifier and the
KINETICS AND CATALYSIS Vol. 45 No. 3 2004

4
12
CHZHU et al.
particles of supported platinum, which, on the one defects in palladium crystallites, etc. [16]. We com-
hand, leads to the stabilization of the high dispersity pared palladium–aluminum catalysts calcined at the
and, on the other hand, to a decrease in the TON same temperature, but with different dispersities,
because of the enhanced deficiency of electron density
on the surface palladium atoms.
obtained by varying the concentration of palladium in
the catalysts.
The same but more pronounced trend is seen for pal-
ladium catalysts supported on pure modifying oxides in
the reaction of the complete oxidation of methane
It was found (Table 1) that for 1 and 2% palladium-
alumina catalysts calcined at 500°ë, an increase in the
concentration of palladium from 1 to 2 wt % leads to a
an increase in the average size of particles by a factor of
(Table 1), with the exception of palladium supported on
monoclinic ZrO for which the TON is the highest.
2
2
.2, and the TON decreases by a factor of 2.6. Proceed-
Thus, for methane oxidation, the TON decreases in the
ing from these data, we conclude that an increase in the
size of the supported palladium particles leads to an
increase in the TON. The explanation for such an effect
of particles size on the TON can be a hypothesis of a
decrease in the acceptor effect of the support on the
crystallites of supported palladium with an increase in
their size. On the other hand, the smaller the crystal-
lites, the more pronounced the acceptor effect of sup-
series: (Pd/ZrO , T
= 750°C
(Pd/Al O )
(Pd/CeO ) (2 wt % Pd).
2
2
calcin
2 3
(
Pd/ZrO , T
= 500°C)
2
calcin
For the 1 wt % palladium–alumina catalyst modified
by rare-earth element oxides and ZrO , the TON in the
2
complete oxidation of n-pentane decreases in the
series: (Pd/Al O )
(Pd/Al O –ZrO )
(Pd/Al O –CeO ).
2 3 2
2
3
2 3 2
(
Pd/Al O –La O )
2
3
2
3
A similar dependence of the TON on the nature of the port leading to the deficiency in the electron density on
modifier for platinum–alumina catalyst in the reaction of Pd(Pt) atoms, that is, to an increase in the oxidation
complete oxidation of n-butane was reported in [14].
state. At the same time, according to [17], the activation
As mentioned above, an increase in the TON in the of hydrocarbon molecules is determined by the concen-
reaction of the complete oxidation of hydrocarbons tration of the electron density on the active sites to be
with an increase in the metal particles size was noted by transferred to the antibonding orbital of the C–H bond.
many researchers. This was observed not only in the
An analysis of the data in Table 1 shows that, the
case of the aging of platinum-alumina and palladium–
concentration of palladium being the same and the dis-
persities being close, the TON is largely determined by
the nature of the supporting oxide; the stronger the
alumina samples in the reaction mixture at high temper-
atures [1, 2] but also upon sample calcination in air at
5
00–(850°C) 900°C (for platinum and palladium sup-
ported on Al O , ZrO , and SiO –Al O ) [6, 13]. For metal–support interaction the lower the TON.
2
3
2
2
2
3
the systems Pd/Al O and Pd/ZrO , an increase in the
2
3
2
Considerable changes are observed after calcination
at 600°C (Fig. 3a). As mentioned above, noncoherent
boundaries are formed by sintered species, which point
to the strong distortion in the particle structure, whereas
isolated particles are well ordered and cut. It is probable
that due to the strong interaction, the fragments of the
modifying oxide move together with the palladium par-
ticles interacting with them at an elevated temperature,
and do not allow the palladium particles to sinter. This
explains the existence of noncoherent boundaries
TON was also noted with an increase in the reaction
time [5].
It is quite possible that an increase in the TON (and
the specific catalytic activity) could be due to the
removal of admixtures from the catalysts during their
calcination in air or in the course of catalyst aging in the
reaction medium. Thus, it was shown in [15] that, when
the palladium catalyst supported on the lanthanum
oxide is prepared from H PdCl , the catalyst activity
2
4
increases for about 10 h in the reaction of n-butane oxi-
dation until it becomes constant due to the removal of between sintered particles. The existence of such struc-
the chloride ions that stabilize the palladium in the oxi- tural barriers stabilizes the dispersity of supported pal-
dized electron-deficient inactive state.
ladium and its activity.
The addition of a modifier stabilizes the dispersity
of palladium in the studied range of calcination temper-
atures (500–700°C), and this fact suggests that there is
an effect of dispersity on the TON of supported palla-
dium. On the other hand, a change in the dispersity with
an increase in the calcination temperature for the non-
modified palladium–alumina catalyst creates uncer-
tainties in the interpretation of the relationship between
the TON and dispersity because at elevated tempera-
tures, the processes of particle sintering and their
Thus, in this work we showed that, for palladium
catalysts supported on various oxides, the effect of ther-
mal activation (an increase in the TON of palladium
with an increase in the calcination temperature of the
samples from 500 to 700°C) is not observed in contrast
to supported platinum catalysts [9]. A comparison of
two palladium-alumina catalysts calcined at 500°C,
differing in the concentration of palladium and its dis-
persity shows that the TON of palladium is much higher
restructuring may occur: for instance, a decrease in the for palladium with a lower dispersity. All other condi-
fraction of a cluster and monatomic states of palladium tions being the same, the TON of supported palladium
on the support surface, a change in the structure of is determined by the support nature.
KINETICS AND CATALYSIS Vol. 45 No. 3 2004

THE EFFECT OF SUPPORT NATURE AND THE TEMPERATURE
413
ACKNOWLEDGMENTS
9. Chzhu, D.P., Tsyrul’nikov, P.G., Kudrya, E.N., Smo-
likov, M.D., Borbat, V.F., and Bubnov, A.V., Kinet.
Katal., 2002, vol. 43, no. 3, p. 410.
This work was supported by the Russian Foundation
for Basic Research (grant no. 99-03-32135a).
1
0. Ginsburg, S.I., Gladyshevskaya, K.A., Ezerskaya, N.A.,
et al., Rukovodstvo po khimicheskomu analizu plati-
novykh metallov i zolota (Handbook on Chemical Anal-
ysis of Platinum Metals and Gold), Moscow: Nauka,
1965, p. 159.
REFERENCES
1
2
3
4
5
6
7
8
. Briot, P., Auroux, A., Jones, D., and Primer, M., Appl.
Catal., 1990, vol. 59, p. 141.
. Briot, P. and Primet, M., Appl. Catal., 1991, vol. 68,
1
1. Karnaukhov, A.P., Sibirskie chteniya po katalizu (Sibe-
rian Lectures on Catalysis), Novosibirsk, 1979, part 1,
p. 136.
p. 301.
. Carballo, L.M. and Wolf, E.E., J. Catal., 1978, vol. 53, 12. Drozdov, V.A., Tsyrulnikov, P.G., Popovskii, V.V., et al.,
p. 366.
React. Kinet. Catal. Lett., 1985, vol. 27, no. 2, p. 425.
. Völter, J., Lietz, G., Spindler, H., and Lieske, L., 13. Ribeiro, F.H., Chow, M., and Dalla Betta, R.A., J. Catal.,
J. Catal., 1987, vol. 104, p. 375.
1994, vol. 146, p. 537.
. Hicks, R., Qi, H., Young, M.L., and Lee, R.G., J. Catal., 14. Drozdov, V.A., Cand. Sci. (Chem.) Dissertation, Novosi-
990, vol. 122, p. 280.
birsk: Institute of Catalysis, 1987.
. Hicks, R., Qi, H., Young, M.L., and Lee, R.G., J. Catal., 15. Starostina, T.G., Cand. Sci. (Chem.) Dissertation,
990, vol. 122, p. 295.
Novosibirsk: Institute of Catalysis, 1987.
. Cullis, C.F. and Willat, B.M., J. Catal., 1983, vol. 83, 16. Chen, J.J. and Ruckenstein, E., J. Phys. Chem., 1981,
p. 267.
vol. 85, p. 1606.
1
1
. Baldwin, T.R. and Burch, R., Catal. Lett., 1990, vol. 6, 17. Golodets, G.I., React. Kinet. Catal. Lett., 1981, vol. 17,
p. 131.
nos. 1–2, p. 97.
KINETICS AND CATALYSIS Vol. 45 No. 3 2004