Effect of the nanoparticle size of supported copper-containing catalysts on their activity in the selective oxidation of CO in an excess of H2

Article abstract of DOI:10.1134/S0023158411030049

The catalytic properties of systems prepared by the supporting of CuO onto CeO₂, ZrO₂, and Zr_0.5Ce_0.5O ₂ with particle sizes of 15-25 nm (nitrate pyrolysis (p)) and 5-6 nm (microemulsion method (me)) in the reaction of CO oxidation in an excess of H₂ were studied. In the latter case, the supports had an almost homogeneous surface and a small number of defects. The catalytic activity of (me) and (p) supports was low and almost the same, whereas the catalytic activity of CuO/(CeO₂, ZrO₂, and Zr_0.5Ce _0.5O₂)(me) samples was lower than that of CuO/(CeO ₂ and ZrO₂)(p). The maximum CO conversion (～100% at 125°C) was observed on 5% CuO/CeO₂ (p). The CO and CO₂ adsorption species on (p) and (me) catalysts were studied by TPD. Differences in the compositions of copper-containing centers on the surfaces of (p) and (me) systems were found using TPR. The nature of the active centers of CO oxidation and the effect of support crystallite size on the catalytic activity were considered.

Full text of DOI:10.1134/S0023158411030049

ISSN 0023ꢀ1584, Kinetics and Catalysis, 2011, Vol. 52, No. 3, pp. 434–445. © Pleiades Publishing, Ltd., 2011.
Original Russian Text © A.A. Firsova, T.I. Khomenko, A.N. Il’ichev, V.N. Korchak, 2011, published in Kinetika i Kataliz, 2011, Vol. 52, No. 3, pp. 445–456.
Effect of the Nanoparticle Size of Supported CopperꢀContaining
Catalysts on Their Activity in the Selective Oxidation
of CO in an Excess of H2
A. A. Firsova, T. I. Khomenko, A. N. Il’ichev, and V. N. Korchak
Semenov Institute of Chemical Physics, Russian Academy of Sciences, Moscow, 119991 Russia
eꢀmail: firsova@chph.ras.ru
Received June 24, 2010
Abstract—The catalytic properties of systems prepared by the supporting of CuO onto CeO2, ZrO2, and
Zr0.5Ce0.5O2 with particle sizes of 15–25 nm (nitrate pyrolysis (p)) and 5–6 nm (microemulsion method
(me)) in the reaction of CO oxidation in an excess of H2 were studied. In the latter case, the supports had an
almost homogeneous surface and a small number of defects. The catalytic activity of (me) and (p) supports
was low and almost the same, whereas the catalytic activity of CuO/(CeO2
samples was lower than that of CuO/(CeO2 and ZrO2)(p). The maximum CO conversion (~100% at 125°C
, ZrO2, and Zr0.5Ce0.5O2)(me)
)
was observed on 5% CuO/CeO2 (p). The CO and CO2 adsorption species on (p) and (me) catalysts were studꢀ
ied by TPD. Differences in the compositions of copperꢀcontaining centers on the surfaces of (p) and (me)
systems were found using TPR. The nature of the active centers of CO oxidation and the effect of support
crystallite size on the catalytic activity were considered.
DOI: 10.1134/S0023158411030049
INTRODUCTION
150°С. At the temperature
Т 150°С, the conversion
>
of CO dramatically decreased because of the dissociaꢀ
tive adsorption of hydrogen and the blocking of active
centers by adsorbed water. The temperature range in
which a high CO conversion (90–99.5%) was retained
was very narrow, only 20 K. It was found that, upon the
addition of Fe and Ni oxides, which are inactive in this
reaction, to the CuО/СеО2 system, the conversion of
CO increased at a lower concentration of CuO (2.5%)
[20] and the temperature range of a maximum converꢀ
sion simultaneously broadened to 40–50 K toward
both low and high temperatures.
The purification of commercially produced hydroꢀ
gen to remove an impurity of CO, whose concentraꢀ
tion is 0.5–2 vol % CO, is mandatory for the use of this
hydrogen in new fuel cells, which can replace internal
combustion engines [1, 2]. The concentration of CO
in mixtures with hydrogen should be lower than 10 or
100 ppm in operations with cells containing platinum
electrodes or bimetallic catalysts like PtRu, respecꢀ
tively [3–5]. The lowꢀtemperature catalytic oxidation
of CO in an atmosphere of hydrogen with oxygen or
air is most promising for the removal of CO impurities
from the mixtures. Various supported catalytic systems
The studies of nanosized catalytic systems syntheꢀ
involving Pt, Ru, Pd, and Au [3, 6–9] or Co [10–13] sized based on materials with high surface areas and
and Cu oxides [14–22] were proposed. In the latter uniform crystallite sizes in the range of 3–10 nm [23]
case, with the use of СеО2 and ZrO2 dioxides and and containing supported small metal or oxide partiꢀ
ZrxCe1 – xО2 solid solutions as supports, the CuО/СеО2 cles are currently underway. There are published data
catalysts were found the most effective in this process. on the effect of the size of such supported particles on
On the surface of these catalysts, highly active oxygen, the catalytic activity of the systems [24–26] in dehyꢀ
which is capable of oxidizing CO at 100–160°С, was drogenation and oxidation reactions. It was noted that
formed as a result of the strong interaction of CuO– a decrease in the particle size of a catalyst can lead to
CeO2 oxides. It was found [19] that Cu–O–Ce clusꢀ either an increase or a decrease in the catalytic activity.
ters were formed on the surface of CuО/СеО2, and New methods have been developed for the preparation
CO was adsorbed at these clusters with the formation of nanodispersed oxides; these are the thermal decomꢀ
of СО–Cu+ carbonyl complexes and the subsequent position of an organometallic precursor in a flow reacꢀ
CO oxidation by the oxygen of these clusters. In the tor [27], the sol–gel synthesis [28], and the microꢀ
study of the oxidation of CO in an excess of hydrogen emulsion method [29], which was used to synthesize
[19], it was found that the conversion of CO increased СеО2, ZrO2, and Zr0.5Ce0.5O2 [30]. Il’ichev et al. [30]
with increasing CuO concentration (from 0.5 to 6 wt %) used Xꢀray diffraction (XRD) analysis, temperatureꢀ
and reaction temperature to reach a maximum of programmed desorption (TPD), and EPR spectrosꢀ
99.5% at a 5–6 wt % CuO concentration and 140– copy to find that these oxides have a high specific surꢀ
434

EFFECT OF THE NANOPARTICLE SIZE OF SUPPORTED COPPERꢀCONTAINING
435
face area of ~100 m2/g and consist of uniform particles respectively, and for ZrO2 (p) and (2.5, 5)%CuO/ZrO2 (p),
with a crystallite size of ~5 nm. The use of such oxides 55–50 m2/g, respectively. For the samples prepared by
for preparing supported catalysts for CO oxidation in the (me) method, Ssp was much higher: 100 m2/g in
the presence of hydrogen makes it possible to study the ZrO2 (me) or 80 m2/g in СеО2 (me) and Zr0.5Ce0.5O2 (me).
influence of the particle size of a catalyst on its activity Upon supporting 2.5 and 5 wt % CuO onto the (me)
because the number of supported atoms accessible to systems, the values of S
sp decreased by 5–7 m2/g.
the reaction increases with the dispersity. On the other
hand, the probability of formation of active centers
that include several metal and oxygen atoms arranged
in a certain manner on the surface decreases with
decreasing support particle size, and this can lead to a
decrease in the catalytic activity.
The aim of this work was to comparatively study
copperꢀcontaining catalytic systems in which the
СеО2, ZrO2, and Ce0.5Zr0.5O2 supports differed in crysꢀ
tallite sizes (15–25 and ~5 nm in the case of (p) samꢀ
ples and samples prepared by (me) synthesis, respecꢀ
tively) in the reaction of CO oxidation in an excess of
H2. The effects of the particle sizes of the catalysts on
their structure, bond strength of oxygen with the surꢀ
face, catalytic activity, and CO and СО2 adsorption
species were studied using XRD analysis, temperaꢀ
tureꢀprogrammed reduction (TPR), and TPD.
The phase composition was studied by XRD analyꢀ
sis on a Dronꢀ3M diffractometer over the angle range
of
mator), which was calibrated with the use of the (112)
line of SiO2 powder ( ꢀquartz; interplanar spacing,
2
θ
= 8– 70° (СuK radiation; graphite monochroꢀ
α
α
1.818 Å). All of the calculations and the processing of
spectra were performed using computer programs.
The catalytic properties of the synthesized samples
in the reaction of CO oxidation in an excess of hydroꢀ
gen were studied in a flowꢀtype system over the temꢀ
perature range of 20–400°С. The reaction mixture
containing 98 vol % H2, 1 vol % CO, and 1 vol % О2
was passed through a quartz tube reactor (3 mm in
diameter) with a catalyst (20 mg) at a flow rate of
40 ml/min; the temperature was increased stepwise. A
thermocouple was arranged outside the reactor at a
level of the center of a catalyst bed (whose height was
4–5 mm). In the course of the analysis of reaction
products at each particular temperature, which was
kept for 20–30 min, no increase in the catalyst temꢀ
perature as a result of heat release during exothermic
reactions of CO and Н2 oxidation was detected. The
reaction products were analyzed by chromatography;
they were separated on two columns packed with
molecular sieves NaX (13 Å) and Porapak QS. Kathaꢀ
rometers were used as detectors. The reaction was
characterized by the degree of conversion, the temperꢀ
ature at which a maximum conversion of CO into СО2
was reached (Tmax), and the selectivity for oxygen. This
selectivity was determined as the fraction of oxygen (on a
EXPERIMENTAL
The cerium and zirconium dioxides (СеО2 and
ZrO2) were prepared by two different methods: (1) the
pyrolysis of nitric acid salts (p) and (2) a microemulꢀ
sion method (me). In the former case, СеО2 was
obtained from Се(NO3)3 6H2O on heating in air to
⋅
500оС at a rate of 7 K/min and the subsequent calcinaꢀ
tion at this temperature for 2 h, and ZrO2 was obtained
from zirconium(IV) oxide nitrate (ZrO(NO3)2
by consecutive thermal decomposition in air at 300оС
for 2 h and at 500 for 3 h. In the latter case, СеО2
⋅
6H2O)
°С
,
ZrO2, and Се0.5 Zr0.5O2 were synthesized using a
microemulsion procedure, which was described elseꢀ
where [30]. The emulsions were prepared by mixing
consumed basis) used for the formation of СО2
.
The TPR of the catalysts was performed in a flow of
6%Н2/Ar mixture at a rate of 100 ml/min. Heating
a
the aqueous solutions of the salts Се(NO3)3
ZrOCl2 8H2O or their mixture (with the ratio Zr : Ce =
1 between the cations) and N(CH3)4OH 5H2O with
⋅
6H2O and
over the temperature range of 20–650
formed at a rate of 12 K/min. Before the experiments,
the samples (100 mg) were calcined at 500 in a flow
°С was perꢀ
⋅
⋅
°
С
an organic solution containing hexane, hexanol, and
the surfactant Triton Xꢀ100. Commercial reagents
from Aldrich were used. The reaction was carried out
with stirring the corresponding emulsions for 24 h;
after this, the products were separated by centrifugaꢀ
tion, washed in methanol, dried at room temperature,
then heated in air at a constant heating rate of 2 K/min
of air for 1 h. Water vapor released upon reduction was
absorbed in a trap packed with Anhydrone and
arranged immediately after the reactor. A katharomeꢀ
ter was used as a detector. The amount of absorbed
hydrogen was determined from TPR peak areas to
within ~10%.
The adsorbed CO and СО2 species on copper oxide
supported on (me) and (p) samples were studied using
TPD. A sample (50 mg) was placed in a quartz
ampoule, evacuated to 10–4 Pa, heated to 500°С for
to 500°С, and kept at this temperature for 2 h.
CuO was supported onto the resulting systems by
impregnating them with a copper nitrate solution;
thereafter, they were dried at 150°С and calcined at
500°С for 2 h. The samples contained 2.5 and 5 wt %
CuO.
1 h, oxidized in О2 for 30 min (
20°С, and evacuated once again. After this, the
adsorption of CO ( = 30 Pa) was performed for
Р = 600 Pa), cooled to
Р
The specific surface areas (
measured by the BET method using the lowꢀtemperaꢀ 15 min, and the sample was heated at the constant rate
ture adsorption of argon. The values of sp for CeO2 (p) = 10 K/min in the course of evacuation. The TPD
and (2.5, 5)%CuO/CeO2 (p) were 67 and 65–63 m2/g, spectrum as the dependence of the pressure of desꢀ
Ssp) of the samples were 10 min; then, the gas phase was pumped out for
S
β
KINETICS AND CATALYSIS Vol. 52
No. 3
2011

436
FIRSOVA et al.
Debye–Scherrer formula. The average crystallite size
Conversion of CO, %
(а)
of СеО2 (me), ZrO2 (me), and Zr0.5Се0.5O2 (me) was
3–5 nm; it increased to 5–7 nm after supporting
5 wt % CuO. The size of the particles prepared by
pyrolysis was much greater: 13–15 nm in СеО2 (p) or
20–25 nm in ZrO2 (p).
100
90
80
70
3
60
50
40
30
20
10
2
5
1
4
6
Catalytic Activity of (p) and (me) Samples
in the Reaction of CO Oxidation with Oxygen
in an Excess of Hydrogen
0
50 100 150 200 250 300 350 400
T, °C
Supports. The conversion of CO on СеО2 (p) came
into play at 200°С and reached 30% at 350°С; it was
almost the same as the conversion of CO on
СеО2(me).
Both of the samples of ZrO2 (me) and ZrO2 (p)
were inactive in the range of 20–300°С; the oxidation
of CO on them was observed at temperatures higher
than 300°С; in this case, the conversion was higher
than 3–4%. The mixed oxide Zr0.5Се0.5O2 (me) occuꢀ
pied an intermediate place between the individual
oxides in terms of catalytic activity: the oxidation of
Selectivity for O2, %
100
(b)
4
90
80
2
70
60
50
40
30
20
10
3
1
0
50
100
150
200
250
T, °C
CO came into play at temperatures higher than 200
as on СеО2 (me); however, the conversion of CO was
lower by a factor of about 2, namely, ~17% at 350
°С,
Fig. 1. (a) Conversion of CO into СО and (b) process
selectivity for oxygen on the catalysts (
2
°С.
1) 2.5% CuO/CeO (p),
2
(
(4
2
) 2.5% CuO/CeO (me), ( ) 5% CuO/CeO (p),
3
2
2
2
The conversion of CO on the samples containing
the supported copper oxide CuO depended on both
CuO concentration on the surface and the nature of
the support.
) 5% CuO/CeO (me), (
5
) CeO (p), and (
6
) CeO (me).
2
2
orbed gas (W) on temperature (
Т
) was measured using
CuO/СеО2. As can be seen in Fig. 1a, the converꢀ
a Pirani gage [31] in a temperature range from 20 to
500°С
sion and the temperature range of a maximum converꢀ
sion (Тmax) depend on both the amount of CuO supꢀ
ported on the surfaces of (p) and (me) СеО2 samples
and the СеО2 preparation method. For all of the CuOꢀ
containing samples, the temperature dependence of
CO conversion is an extremal function. In the oxidaꢀ
tion of CO on a sample of 2.5% CuO/СеО2 (p) (Fig. 1a,
.
RESULTS
XRD Analysis of the Samples
According to XRD data, cerium dioxide (СеО2
crystallized in a cubic modification (C) regardless of curve
preparation procedure; this crystal structure was 95% to reach a maximum at
retained upon supporting 2.5 and 5 wt % CuO onto it. as the temperature was further increased. On 2.5%
In the synthesis of ZrO2(p), a monoclinic modificaꢀ CuO/СеО2 (me) (curve ), the conversion of CO
)
1
), the conversion increased from 5 (80°С) to
Тmax = 145°С decreased
2
tion (M) was mainly formed; in addition, lowꢀintenꢀ increased in the range of 100–180°С to reach a maxiꢀ
sity reflections due to a tetragonal modification (T), mum value of 87% at Тmax = 180–200°С and also
whose concentration was no higher than 10%, were decreased at higher temperatures. As the amount of
detected in a diffraction pattern. The phase composiꢀ supported copper oxide (CuO) was increased from 2.5
tion of the catalyst remained unchanged upon supꢀ to 5%, the conversion of CO increased and Тmax
porting 2.5 and 5 wt % CuO. On the contrary, only a decreased on both (p) and (me) samples. On the 5%
metastable T phase was detected in the sample of CuO/СеО2(p) catalyst, the maximum conversion was
ZrO2(me); this is characteristic of crystallites with a ~100% at Тmax = 125°С (curve 3); in this case, the
size of ~5 nm [30]. The formation of 20–25% M phase temperature range of a maximum conversion broadꢀ
was noted after supporting 5% CuO; that is, the imperꢀ ened and shifted to the region of low temperatures.
fection of the sample increased. In the case of The accuracy of the measurement of a minimum
Zr0.5Се0.5O2(me), a solid solution is formed, which residual concentration of CO in a gas mixture was
was, probably, a pseudoisometric phase with a lattice about 0.005%, that is, ~50 ppm. Thus, the residual
parameter intermediate between the lattice parameꢀ concentration of CO was no higher than 50 ppm on
ters of СеО2 (me) and ZrO2 (me). The phase composiꢀ the 5% CuO/СеО2 (p) sample under optimum condiꢀ
tion of the sample after supporting CuO remained tions at a CO conversion of ~100%. The same behavꢀ
unchanged. The crystallite sizes of the synthesized ior was also observed in the 5% CuO/СеО2 (me) samꢀ
samples were determined from XRD data using the ple: the conversion increased to 99% with a simultaꢀ
KINETICS AND CATALYSIS Vol. 52
No. 3
2011

EFFECT OF THE NANOPARTICLE SIZE OF SUPPORTED COPPERꢀCONTAINING
437
neous decrease in Тmax to 160°С (curve
the residual concentration of CO was no higher than
100 ppm. Figure 1a (curves ) shows the conversion
of CO on (p) and (me) CeO2.
Data on the catalytic activity of samples based on
СеО2 show that the catalysts obtained by supporting
copper oxide on СеО2 (p) were more active than the
samples with CeO2 (me). Note that the extremal temꢀ
perature dependence of CO conversion was also characꢀ
teristic of other supported copper oxide catalysts [20–
4). In this case,
(а)
Conversion of CO, %
100
5, 6
– 1
– 2
– 3
– 4
– 5
– 6
80
60
2
40
4
20
5
3
1
6
0
50 100 150 200 250 300 350 400 450
T, °C
22], and a decrease in the conversion of CO at
Т > Тmax
Selectivity for O2, %
100
(b)
was not related to an irreversible change in the catalyst
because the conversion increased to a maximum value
once again as the reaction temperature was decreased
to Тmax. Furthermore, the stable operation of (p) and
(me) samples at Тmax was retained for 2 h.
Figure 1b shows the selectivity for oxygen of the
process on the samples based on СеО2. For the (me)
– 1
– 2
– 3
– 4
2
80
60
3
40
20
4
1
samples (curves 2, 4), the selectivity in the lowꢀtemꢀ
perature region of 80–150°С was much higher (100–
65%), as compared with 64–45% for the (p) samples
(curves 1, 3). A change in the amount of copper oxide
0
50 100 150 200 250 300
T, °C
supported onto the surface from 2.5 to 5% had almost
no effect on the value of selectivity both in the case of
Fig. 2. (a) Conversion of CO into CO and (b) process selecꢀ
2
tivity for oxygen on the catalysts (1) 2.5% CuO/ZrO (p),
2
CuO/СеО2 (p) (curves
CuO/СеО2 (me) (curves
1, 3) and in the case of
(
2
) 5% CuO/ZrO (p), (3) 2.5% CuO/ZrO (me),
2 2
2
,
4).
(4
) 5% CuO/ZrO (me), (5) ZrO (p), and (6) ZrO (me).
2 2 2
CuO/ZrО2. The catalytic systems in which ZrО2 (p)
and (me) are used as the supports are less active than
the systems with СеО2; however, the temperature
dependences of conversion also exhibited an extremal
character. The activity of CuO/ZrО2 (p) catalysts (Fig. 2a,
rapidly decreased with temperature to be 45–50% in
the temperature range of 160–190 , where a maxiꢀ
mum conversion of CO was observed. On the
2.5% CuO/ZrО2 (p) and (me) samples, the selectivity
was much lower (Fig. 2b, curves 1, 3), and its value
depended on the ZrО2 preparation procedure: (p) or
(me). In the case of the 2.5% CuO/ZrО2(p) sample,
°
С
curves 1, 2) is determined by the copper oxide content
of the sample. Thus, the oxidation of CO on
2.5%CuO/ZrО2 (p) came into play at a temperature
higher than 100°С, and the conversion passed through
a maximum at 190°С. As the concentration of CuO on
the surface of ZrО2 (p) increased by a factor of 2 (from
2.5 to 5%), the maximum conversion also increased by
a factor of 2 from 32% (190°С) to 65% (170°С), and
Тmax decreased by 20 K. A further increase in the
amount of supported CuO to 10% had almost no effect
on the degree of CO conversion and Тmax. This effect
of the concentration of CuO on the catalyst activity
was not observed in CuO/ZrО2 (me) samples. The
the selectivity increased from 8% at 130
190 and then decreased (curve ), whereas an
almost constant selectivity of 42–45% was observed
on the 2.5% CuO/ZrО2 (me) sample at 120–190
(curve ). As the temperature was further increased,
°С to 38% at
°
С
1
°
С
3
the selectivity decreased, as in the case of other cataꢀ
lysts.
CuO/Zr0.5Се0.5O2
dation of CO on the samples of Zr0.5Се0.5O2 (me)
. Figure 3 shows data on the oxiꢀ
(curve 1) and 2.5 and 5% CuO/Zr0.5Се0.5O2 (me)
conversion of CO came into play at 130
maximum value of 29 or 32% at Тmax = 190
2.5% CuO/ZrО2 (me) (Fig. 2a, curve
5% CuO/ZrО2 (me) (Fig. 2a, curve
decreased and was no higher than 5% at 250
(curves and ) shows data on the conversion of CO
°С to reach a
(curves
2
and , respectively). The regularities of CO
3
°
3)
C
on
or
oxidation on CuO/Zr0.5Се0.5O2 (me) are similar to oxiꢀ
dation on CuO/СеO2 (me), although the activity of the
latter is much higher. The surface concentration of
copper oxide influences the conversion of CO: for
4
); then, it rapidly
°С. Figure 2a
5
6
2.5% CuO, a maximum conversion was 36% at Тmax
220 (Fig. 3a, curve ), whereas it was 51% for
5% CuO at max = 180 (Fig. 3a, curve ); that is, the
=
on ZrО2 (p) and (me).
°
С
2
The process selectivity for oxygen (Fig. 2b) on
CuO/ZrО2(p) and (me) samples increased considerꢀ
ably as the concentration of CuO on the surfaces of
ZrО2 (p) and (me) was increased from 2.5 to 5%. As
can be seen in Fig. 2b, a maximum selectivity of 100%
was observed at low temperatures of 100–120°С upon
Т
°
С
3
conversion of CO increased by a factor of 1.5 and Тmax
decreased by 40 K with increasing the amount of supꢀ
ported copper oxide.
For these samples, the selectivity of О2 consumpꢀ
supporting 5% CuO onto both (p) and (me) ZrО2 tion for the oxidation of CO (Fig. 3b) depended on the
(curves ). The process selectivity on these catalysts concentration of CuO on the surface of Zr0.5Се0.5O2
2, 4
KINETICS AND CATALYSIS Vol. 52
No. 3
2011

438
Conversion of CO, %
100
FIRSOVA et al.
of both of the samples was almost the same at temperꢀ
atures higher than 150°С
(а)
.
80
TemperatureꢀProgrammed Reduction (TPR)
of (p) and (me) Samples with Hydrogen
60
40
20
3
2
The study of the (p) and (me) nanodispersed sysꢀ
tems by TPR with hydrogen allowed us to reveal differꢀ
ences in the reduction of both СеО2, ZrO2, and
Zr0.5Се0.5O2 supports synthesized by different methꢀ
ods and the copper oxide catalysts obtained on their
basis. Figure 4 indicates that the reduction temperaꢀ
1
0
50 100 150 200 250 300 350 400
T, °C
Selectivity for O2, %
100
90
80
70
60
50
40
30
20
10
ture of the СеО2 (me) sample (spectrum
siderably lower than that of СеО2 (p) (spectrum
both cases, the TPR spectra exhibited two peaks: specꢀ
trum , at 270 and 345 and spectrum , at 400 and
526 . Two peaks of the oxidation of hydrogen by oxyꢀ
1
) was conꢀ
2
2). In
(b)
1
С
°С
2
3
°
gen of the catalyst suggest two stages of the reduction
process. Note that the total amounts of hydrogen
absorbed by (p) and (me) samples were almost the
same (see Table 1): 5.3
×
10–4 and 4.9
×
10–4 mol/g,
respectively. Assuming that Н2 is completely oxiꢀ
dized by oxygen from the surface of oxides, the
amount of reacted oxygen per unit surface area is
0
50
100
150
T, °C
200
250
(4–4.3)
×
1018 molecule/m2, which corresponds to
Fig. 3. (a) Conversion of CO into CO and (b) process
2
the interaction of hydrogen with approximately 60%
of surface oxygen ions. According to published data
selectivity for oxygen on the catalysts (
(me), 2.5% CuO/Zr Се
) 5% CuO/Zr Се O (me).
1
) Zr Се O
(me), and
0.5 0.5 2
(2)
O
0.5 0.5
2
[21, 30], at the first stage the reduction of СеО2
,
(3
0.5 0.5
2
hydrogen is oxidized by surface oxygen, and oxidation
at the second stage was accompanied by the formation
of nonstoichiometric surface oxides СеОх. As can be
seen in Table 1, both of the processes effectively
occurred on the СеО2 (p) sample, whereas the conꢀ
sumption of hydrogen on СеО2 (me) at the first stage
was higher than that at the second stage by a factor of
10. This fact suggests that the surface of СеО2 (me) was
more uniform than that of the СеО2 (p) sample.
5% СuO/CeO2 (me)
5% СuO/CeO2 (p)
6
5
The TPR spectra of the CuO/CeO2 (me) and
CuO/CeO2 (p) samples were also strongly different
(Fig. 4, curves 3–6). Spectrum
2.5% CuO/CeO2 (me) sample contained three peaks.
The first of them at 137 C belongs to the reduction of
surface finely dispersed copper oxide particles [32,
33]. The second peak at 173 C is likely due to the
reduction of Cu–O–Ce clusters formed at the CuO–
CeO2 boundary, whereas the third peak at 226
belongs to the reduction of СеО2 (me). As the concenꢀ
tration of CuO was changed from 2.5 to 5% (curves
and , respectively), the number of reduced finely disꢀ
3
of the
2.5% СuO/CeO2 (p)
2.5% СuO/CeO2 (me)
°
4
3
°
CeO2 (me)
CeO2 (p)
2
1
°
C
3
0
50 100 150 200 250 300 350 400 450 500
T, °C
6
persed copper oxide particles increased, and the
amount of the reacted surface oxygen of СеО2 (me) did
not increase. At the same time, the amount of oxygen
in Cu–O–Ce clusters cannot be accurately estimated
as a result of the overlapping of peaks at 137 and
173°С; it likely remains constant. The spectrum of the
Fig. 4. TPR spectra of (
) 2.5% CuO/CeO (me), (4) 2.5% CuO/CeO (p),
2
1) СеО (me), (2) СеО (p),
2 2
(3
2
(5) 5% CuO/CeO (p), and (6) 5% CuO/CeO (me).
2
2
(me) only in the lowꢀtemperature region of 80–
150 ; moreover, the process selectivity on the sample
containing 2.5% CuO was higher (curve ) than that
on the catalyst with 5% CuO (curve ). The selectivity support at 400 and 526°С were absent from this specꢀ
2.5% CuO/СеО2 (p) sample (Fig. 4, curve
tained three peaks at 127, 153, and 340
istic peaks due to the surface reduction of the СеО2 (p)
4
) also conꢀ
°
С
°
С. Characterꢀ
1
2
KINETICS AND CATALYSIS Vol. 52
No. 3
2011

EFFECT OF THE NANOPARTICLE SIZE OF SUPPORTED COPPERꢀCONTAINING
439
Table 1. TPR data for the CuO/(CeO2, ZrO2, and Zr0.5Ce0.5O2) samples
H2 consumption
104, mol/g
×
T
, °C
Sample
Degree of reduction, %
–
max
CeO2 (p)
400
526
2.6
2.7
Σ
5.3*
CeO2 (me)
270
345
4.5
0.4
4.9
–
Σ
0
2.5% CuO/CeO2 (p)
2.5% CuO/CeO2 (me)
127
153
1.7
5.5
100(CuO
100(CeO
Cu )
**
Ce O3 )
2
2
surf
340
1.2
8.4
Σ
Σ
0
137(sh)
173
7.5
100(CuO
90(CeO
Cu )
**
Ce O3 )
2
2
surf
226
316(w)
0
5% CuO/CeO2 (p)
5% CuO/CeO2 (me)
114
150
2.8
6.9
100(CuO
84(CeO
Cu )
**
Ce O3 )
2
2
surf
Σ
9.7
0
138
228
Σ
11.0
100(CuO
Cu )
**
100(CeO
Ce O3 )
2
2
surf
ZrO2 (p)
–
–
–
–
ZrO2 (me)
326
0.3
0
2.5% CuO/ZrO2 (p)
149(1)***
1.7(1) ***
81(CuO
78(CuO
98(CuO
97 (CuO
Cu )
207(2)
311(3)
0.8 (2+3)
2.5
Σ
0
2.5% CuO/ZrO2 (me)
5% CuO/ZrO2 (p)
162(1)
2.4 (1+2)
Cu )
202(sh, 2)
319(3)
0.3 (3)
2.7
Σ
0
142(1)
4.5(1)
Cu )
192(2)
212(3)
1.6 (2+3)
6.1
Σ
0
5% CuO/ZrO2 (me)
150
325
376
173
278
Σ
6.0
Cu )
Zr0.5Ce0.5O2 (me) (me)
4.4
8.6
–
0
2.5% CuO/Zr0.5Ce0.5O2 (me)
Σ
100 (CuO
Cu )
**
100 (CeꢀZrꢀO )surf
0
5% CuO/Zr0.5Ce0.5O2 (me)
147
286
Σ
10.4
100 (CuO
Cu )
**
98 (CeꢀZrꢀO )surf
* TPR up to 650°C.
** Fraction of ceria reduced in the absence of CuO.
*** The peak number in the TPR spectrum is given in parentheses; sh = shoulder, and w = weak.
KINETICS AND CATALYSIS Vol. 52
No. 3
2011

440
FIRSOVA et al.
5.0% СuO/ZrO2 (p)
5
5.0% СuO/ZrO2 (me)
2.5% СuO/ZrO2 (p)
4
3
2.5% СuO/ZrO2 (me)
ZrO2 (me)
2
1
0
100
200
300
400
500
T, °C
Fig. 5. TPR spectra of (
1) ZrO (me), (2) 2.5% CuO/ZrO (me), (3) 2.5% CuO/ZrO (p), (4) 5% CuO/ZrO (me), and (5)
2 2 2 2
5% CuO/ZrO (p).
2
trum. At the same time, the temperatures of peak maxꢀ in the two regions of 140–150 and 190–210°С.
imums in the spectrum depends on the amount of According to Firsova et al. [22], the lowꢀtemperature
CuO on the surface: a change in the concentration of peak belongs to the reduction of copper cations that
CuO from 2.5 to 5% (Fig. 4, curve
the temperature of the first maximum from 127 to form Cu–O–Zr clusters, and the second group of
114 or from 153 to 150 for the second maximum. peaks at ~200 belongs to the reduction of the surꢀ
Furthermore, a peak at 340 disappeared. As in the face particles of a CuO phase. The concentration of
5) led to a shift in strongly interact with the surface defects of ZrO2 to
°
С
°
С
°С
°
С
case of the CuO/СеО2 (me) samples, the first and the clusters on the surface increased with the copper conꢀ
second peaks were due to the reduction of finely disꢀ tent of the sample.
persed copper oxide particles and Cu–O–Ce clusters,
respectively. A comparison of spectra (Fig. 4)
Analogous behaviors were observed in the (2.5–
5)% CuO/ZrO2 (me) samples (Fig. 5, spectra and ).
3–6
2
4
shows that the number of Cu–O–Ce clusters in the
(2.5–5%) CuO/СеО2 (p) samples was greater than
that in the corresponding (me) samples. It should be
noted that the amount of hydrogen absorbed by both
CuO/СеО2 (p) and (me) samples during the reduction
was considerably greater than that required for the
complete reduction of supported copper from the state
Cu2+ to Cu0 (see Table 1), and it was also sufficient for
100% reduction of the surface ions Ce4+ that are
reduced in the absence of CuO to Ce3+ (see Table 1).
In this case, the reduction of clusters occurred at a
higher temperature of 150–160°С and the reduction
of the particles of a CuO phase occurred at ~200°С.
Note that the total amount of hydrogen absorbed
by the (p) and (me) samples was approximately the
same (see Table 1).
Figure 6 shows the TPR spectra of the Zr0.5Се0.5O2
(me) (curve
samples (curves
spectrum exhibits one peak at
gously to the spectrum of СеО2 (me) (Fig. 4, curve
with a peak at max = 270 ; the amount of absorbed
1
) and 2.5 and 5% CuO/Zr0.5Се0.5O2 (me)
and , respectively). As is evident,
max = 376 analoꢀ
2
3
1
T
°С
Figure 5 shows the TPR spectra of the (p) and (me)
samples of ZrO2 and CuO/ZrO2. Unlike ZrO2 (p),
2)
Т
°С
which was not reduced in hydrogen to 650
spectrum of ZrO2 (me) exhibited a small narrow peak of
the absorption of hydrogen at 326 (Fig. 5, curve ).
°С, the TPR
hydrogen was almost the same as that observed during
the reduction of СеО2 (me). On this basis, we can
assume that the surface of the solid solution
Zr0.5Се0.5O2 (me) was enriched in СеО2, and an
increase in the temperature of its reduction by ~100 K,
as compared with that of СеО2, was due to the strong
interaction of Ce and Zr cations, which strengthened
°С
1
The amount of oxygen removed from the sample corꢀ
responding to this peak was no higher than 3% on a
surface oxygen ion basis; it was most likely due to
defects formed upon the synthesis of the sample [30].
In the (2.5–5)% CuO/ZrO2 (p) samples (Fig. 5, the bond of surface oxygen [30]. In the TPR spectra of
spectra and ), the absorption of hydrogen occurred the 2.5 and 5% CuO/Zr0.5Се0.5O2 (me) samples, two
3
5
KINETICS AND CATALYSIS Vol. 52
No. 3
2011

EFFECT OF THE NANOPARTICLE SIZE OF SUPPORTED COPPERꢀCONTAINING
441
5.0% СuO/Zr0.5Ce0.5O2
3
2.5% СuO/Zr0.5Ce0.5O2
2
Zr0.5Ce0.5O2
1
0
100
200
300
400
500
T, °C
Fig. 6. TPR spectra of (1) Zr Ce O (me), (2) 2.5% CuO/Zr Ce O (me), and (3) 5% CuO/Zr Ce O (me).
0.5 0.5 2 0.5 0.5 2 0.5 0.5 2
peaks were observed. One of these peaks at 173оС for (curve
2.5% CuO (curve ) or 147°С for 5% CuO (curve samples are substantially lower; moreover, for the latꢀ
was due to the reduction of surface copper oxide, and ter sample, it is even lower than that of 5% CuO/СеО2
3) and 5% CuO/Zr0.5Се0.5O2 (me) (curve 4)
2
3)
the other at 286 or 278°С for 2.5 and 5% CuO, respecꢀ (me) by a factor of 2 (see curves
tively, was due to the reduction of Zr0.5Се0.5O2 (me). In
2, 4).
Using massꢀspectrometric analysis, we found that
the products of desorption are CO (28 amu) and СО2
(44 amu) and their TPD spectra overlap. To separate
spectra, we performed analogous TPD experiments
and recorded the spectra of CO desorption without
СО2. For this purpose, in the process of desorption,
СО2 was frozen in a trap cooled with liquid nitrogen,
which was arranged between the reactor and a
manometer. The desorption spectrum of СО2 was
obtained by the subtraction of the spectrum of CO
from the overall spectrum. The results of this analysis
are given in Fig. 8 for the 5% CuO/СеО2 (p) sample.
this case, a decrease in the support reduction temperꢀ
ature under the action of supported copper oxide was
also observed.
Study of the CO and CO2 Adsorption Species
on CuO/(CeO2, ZrO2, and Zr0.5Ce0.5O2)(p)
and (me) Samples
TPD spectra were obtained for the oxidized supꢀ
ports and copperꢀcontaining samples after the adsorpꢀ
tion of CO on them (
time, 10 min). Desorption from the (p) and (me) CeO2
oxides was observed only at > 400 C. The number of
desorbed molecules per unit surface area was almost
the same for both of the sample (approximately 2.0
1016 m–2). The only product of desorption was СО2
Р = 30 Pa; Т = 20°С; adsorption
Curve
(CO + СО2). Spectrum
tion, which occurred in the range of 20–200
maximum rate of desorption at Тmax = 110
trum , which is the difference of spectra
relates to the desorption of СО2, which occurs in the
range of 50–400 with max = 150 . In the range of
200–400 , it is likely that the wing of the СО2 desꢀ
orption spectrum consists of two additional peaks with
Тmax at 240 and 300 . Similar peaks of CO and CO2
desorption at Тmax = 100–110 and ~150 , respecꢀ
1
in Fig. 8 shows the total TPD spectrum
corresponds to CO desorpꢀ
with a
. Specꢀ
and
T
°
2
°
С
×
°
С
,
3
1
2,
which, probably, resulted from the decomposition of
surface carbonates formed upon the adsorption of CO.
This conclusion was supported by the absence of any
peaks from the TPD spectra of the oxidized samples
without the adsorption of CO. At the same time, the
temperature range of the observed desorption of СО2
coincides with the region of the decomposition of carꢀ
bonates on СеО2 [34].
°
С
Т
°С
°
С
°
С
°
С
tively, were also observed for the other copper oxide
catalysts containing the (me) supports: СеО2, ZrO2,
Figure 7 shows the TPD spectra of the oxidized and Zr0.5Се0.5O2. However, as can be seen in Figs. 7
CuO/(СеО2, ZrO2, and Zr0.5Се0.5O2) samples after and 8, the desorption of СО2 on them occurs only to
CO adsorption on them. The intensity of the spectrum 300°С. Table 2 summarizes the total amounts of desꢀ
depends on the nature of the support and the method orbed molecules N(СО) and N(CO2) per unit surface
of its preparation. Thus, the intensity of the spectrum of area of the samples. It turned out that CO and СО2
desorption from the 5% CuO/СеО2 (p) sample (Fig. 7, were desorbed in approximately the same amounts
curve
5% CuO/СеО2 (me) sample (curve
desorption products from the 5% CuO/ZrO2 (me) and depended on the nature of the support.
1
) is 1.5 times higher than that of the from the surfaces of each of the (me) samples,
2). The amount of although the total desorption (see Fig. 7) was different
KINETICS AND CATALYSIS Vol. 52
No. 3
2011

442
FIRSOVA et al.
arb. units
W
, arb. units
W,
1
0.009
0.008
0.007
0.006
0.005
0.004
0.003
0.002
0.001
0.009
0.008
0.007
0.006
0.005
0.004
0.003
0.002
0.001
1
2
2
3
3
4
0
100
200
300
400
500
600
0
100
200
300
400
500
600
T,
°
C
T, °C
Fig. 7. TPD spectra of the oxidized samples after the
adsorption of CO on them: ( ) 5% CuO/СеO (p),
1
Fig. 8. TPD spectra of the 5% CuO/СеO (p) sample after
2
2
(
2
) 5% CuO/СеО (me), (
3
) 5% CuO/ZrO (me), and
the adsorption of CO on it: (
1) СО + СО , (2) СО, and
2
2
2
(
4) 5% CuO/Zr Ce O (me).
Т
= 20
°
C;
τ
= 10 min.
(
3)
СО = 20 C.
. Т
2
°
0.5 0.5
2
Table 2 also summarizes data on CO2 desorption dized 5% CuO/СеO2 (me) sample after the TPD of
N20(СО2)), which occurred at 20 in the course of CO. In both cases, after the adsorption of СО2 (
CO adsorption ( = 30 Pa; = 20 ; and = 10 min) 30–60 Pa; = 20 ; and СО2 = 10 min), the desorpꢀ
before the onset of TPD. In this case, the amount of tion of СО2 from the samples was observed only in the
adsorbed CO was determined from a pressure change region of 20–200 . It did not exceed 15% of the
(
°
С
Р =
Р
Т
°
С
τ
Т
°С
°
С
upon its adsorption, and the amount of СО2 was deterꢀ amount of СО2 obtained upon the TPD of CO as a
mined by measuring the amount of frozen gas in the result of its oxidation by oxygen of the catalyst surface
trap. The largest amount of СО2 (4.3
obtained from the 5% CuO/СеО2 (me) sample. It was observed temperature dependence of the desorption of
comparable with the amount of СО2 desorbed from СО2 in the region of 50–400 , which is given in Fig. 8,
the catalyst upon TPD. In other samples, the amounts is caused by the activation energy of the oxidation
N20(СО2) were smaller than N(СО2) reaction of adsorbed CO by oxygen surface (Еr) rather
than by the activation energy of the desorption of СО2
×
1017 m–2) was N(СО2) = 4.4
×
1017 m–2. This fact suggests that the
°С
.
.
Note that the adsorption of CO is activated. As the
temperature of CO adsorption on the 5% CuO/СеО2 (me)
The activation energy of the desorption and oxidaꢀ
sample was increased to 100°С, the amount of CO and tion of CO can be estimated using the expression Ed
СО2 upon desorption increased by a factor of about 1.4. 25RTmax, where
= 8.3143 J K–1 mol–1 is the gas conꢀ
The same effect was observed with a change in the presꢀ stant, and Tmax is the temperature ( , K) that correꢀ
sure of CO in the course of adsorption from 30 to 1.7 sponds to the maximum rate of desorption [35]. Thus,
103 Pa. A change in the temperature of CO adsorption for the desorption of CO at Tmax = 383 K, the activation
from 20 to 100 on the 5% CuO/Zr0.5Се0.5O2 (me) energy of desorption is Ed = 79 kJ/mol, the activation
sample also led to an increase in the amounts of CO energy of the oxidation of CO for the desorption peak of
upon TPD from 1.2 max(СO2) = 423 K is r = 88 kJ/mol, and Er
1017 to 5.6 1017 m–2 and of СО2 СО2 at
from 1.5 max(СO2) = 623 K.
1017 to 2.8 1017 m–2. It is likely that the 130 kJ/mol for
activated nature of CO adsorption and its low value at
20 on 5% CuO/Zr0.5Се0.5O2 (me), as compared with
=
R
Т
×
°С
×
×
T
E
=
×
×
T
In the process of TPD, approximately a half of
adsorbed CO is moved away upon desorption in the
°С
other samples, was due to the strong interaction of Zr and
Ce cations in the mixed oxide Zr0.5Се0.5O2 (me) [30].
region of 20–200
dized by surface oxygen to СО2, whose desorption is
. It should be noted that, on the
°С, and the remaining part is oxiꢀ
For explaining the mechanism of СО2 formation observed at 50–400°С
on TPD, we studied CO2 desorption from the preoxiꢀ 5% CuO/СеO2 (me) sample, CO is oxidized to CO2
Table 2. Amounts of CO and CO2 molecules desorbed in TPD after the adsorption of CO on the test samples
Sample
5% CuO/CeO (p)
N(CO)
×
10–17, m–2
N(CO2)
×
10–17, m–2
N20(CO2)
10–17, m–2
×
7.2
10.1
4.4
4.1
1.5
0.34
4.3
0.4
0.6
2
5% CuO/CeO (me)
5.4
3.7
1.2
2
5% CuO/ZrO (me)
2
5% CuO/Zr Ce O (me)
0.5 0.5
2
KINETICS AND CATALYSIS Vol. 52
No. 3
2011

EFFECT OF THE NANOPARTICLE SIZE OF SUPPORTED COPPERꢀCONTAINING
443
Table 3. Ratios between the amounts of adsorbed CO and CuO particles on the samples
Sample
N(CO)
×
10–17, m–2
N(CuO)
×
10–18, m–2
N(CO)/ N(CuO) × 100%
5% CuO/CeO2 (p)
17.5
14.0
5.4
4.5
4.6
6.0
4.1
39
30
9
5% CuO/CeO2 (me)
5% CuO/Zr0.5Ce0.5O2 (me)
5% CuO/ZrO2 (me)
7.9
19
during adsorption even at room temperature (20°С), ions is replaced by O2––Ce4+ or O2––Zr4+, and the
and the amount of СО2 is comparable with that formed second copper cation remains in the effective state
in the TPD process (see Table 2).
Cu+, which is necessary for the adsorption of CO. At
the same time, the Cu+ cations can be formed from
Cu2+ by the interaction with negatively charged
anionic vacancies, which appear in the course of CO
oxidation by surface oxygen [39].
The systems containing 5% CuO can be arranged
in the following order of decreasing catalytic activity
(the maximum conversions and the corresponding
values of Тmax are given in parentheses):
The total number of adsorbed CO molecules on the
surface of 5% CuO/СеO2 (p) is greater than that on
(me) samples (see Table 3). If we consider the uniform
surface coverage of the support with copper oxide, the
number of CuO molecules per unit surface area is
almost the same on (p) and (me) samples; however, the
relationships between the amounts of adsorbed CO and
CuO particles on the investigated samples (Table 3) are
essentially different: it is maximal for 5% CuO/СеO2 (p)
or 5% CuO/СеO2 (me) among (me) samples.
CuO/CeO2 (p) (~100%, 125°С)
(99%, 160°С) CuO/ZrO2 (p) (65%, 170°С)
CuO/ZrO2 (me) (32%, 190°С).
>
CuO/CeO2 (me)
>
>
DISCUSSION
Note that copper oxide catalysts in which СеО2 was
used as a support are more active regardless of support
preparation procedure (and hence support particle
size). This is likely due to both the lower strength of the
bond of oxygen in the centers Cu2+–O2––Ce4+, as
compared with Cu2+–O2––Zr4+, and the greater
number of active centers on the surface of CeO2. The
above data indicate that the catalytic activity of the
supported copper oxide systems also changes dependꢀ
The catalytic activity of CuO/(СеO2, ZrO2) in the
reaction of CO oxidation in the presence of hydrogen
depends on both the amount of supported copper
oxide and the particle size (preparation procedure) of
СеO2 and ZrO2. A considerable decrease in the reacꢀ
tion temperature on these samples, as compared with
temperatures observed in the oxidation of CO on the
supports and the copper oxide CuO, and also an
increase in the conversion of CO are consistent with
previously obtained data [19, 20, 26, 36, 37]. In these
publications, it is assumed that an increase in the
activity of the supported copper oxide catalysts is due
to the formation of the new active centers, which are
formed and stabilized on the surface as a result of the
strong interaction of CuO clusters with the support (a
synergetic effect).
ing on the particle size of the support, СеО2 or ZrO2
.
In the case of СеО2, the catalytic activity of (me)
samples is lower than that of (p) samples at the same
concentration of CuO on the surface. It was found by
XRD analysis that СеО2 crystallized in the K modifiꢀ
cation with constant unit cell parameters regardless of
preparation procedure and particle size. This means
that the catalytic properties changed not due to a
According to TPR data, the following two types of change in the support structure. According to TPR
centers were detected on the surface of the supported data (Fig. 4, Table 1), a change in the concentration of
copper oxide catalysts: the clusters of finely dispersed CuO on the surface of СеО2 (p) from 2.5 to 5% leads to
copper oxide and the centers formed at the interface of an increase in the number of Cu–O–Ce clusters and a
the CuO clusters and the support. These latter contain decrease in the temperature of their reduction with
the cations of copper and the support Cu2+–O2––Ce4+ hydrogen, which confirms the occurrence of a synerꢀ
or Cu2+–O2––Zr4+; in accordance with the above, getic effect for these samples. According to TPR data,
they can be the centers of the lowꢀtemperature oxidaꢀ the 5% CuO/СеO2 (me) sample contained a much
tion of CO. It is well known [8–12, 38–40] that strong greater amount of the finely dispersed particles of surꢀ
carbonyl complexes with only Cu+ cations are formed face copper oxide in comparison with the
upon the adsorption of CO. Therefore, copper cations 2.5% CuO/CeO2 (me) sample, and the number of
in the active center on the oxidized catalyst surface will Cu–O–Се (me) clusters remained almost unchanged.
occur in this state. According to Krylov [35], this takes Moreover, a considerable fraction of the free СеO2
place in the structure of the cluster Cu2+–O2––Cu2+
,
(me) surface was retained in both of the samples.
where the effective charge of copper is Cu+. Upon the Probably, a smaller number of Cu–O–Ce clusters was
formation of boundary clusters, one of the copper catꢀ formed on the uniform surface of СеO2 (me), which
KINETICS AND CATALYSIS Vol. 52
No. 3
2011

444
FIRSOVA et al.
consists of small nanoparticles (~5 nm), upon the of Zr and Ce ions, which is consistent with published data
interaction with supported copper oxide. These clusters [30]. Taking into account this fact, we can assume that the
are active centers for the reaction of CO oxidation [20], copper oxide on the surface of Zr0.5Се0.5O2 (me) forms
and these clusters are smaller than those on a more defecꢀ centers for the oxidation of CO {(Zr)Се4+–О2––Cu2+},
tive surface of СеO2 (p) with a crystallite size of ~15 nm. in which the surface ions of Ce interact with the ions
This may be consistent with TPD data, which show of Zr from a nearꢀsurface layer. It is likely that this
that the number of CO molecules adsorbed on interaction leads to a strengthening of the bond of
CuO/СеO2 (p) per unit surface area (at almost the cluster oxygen and hence to an increase in Тmax and a
same numbers of supported CuO particles) is much decrease in the conversion of CO.
greater than that on CuO/СеO2 (me). Furthermore,
according to TPR data, the reduction of copper oxide
(me) samples occurs at a higher temperature than that
of (p) samples (see Fig. 4 and Table 1). This can be due
to an increase in the binding energy of cluster oxygen
with the surface and a weakening of the synergetic
effect in the case of (me) catalysts. All of these results
explain the smaller activity of the CuO/СеO2 (me)
sample with particle sizes of 5–6 nm, as compared
with CuO/СеO2 (p), which has a particle size of 15 nm.
The selectivity of CO oxidation (see Fig. 1b) on (me)
samples is higher than that on (p) samples and almost
independent of the amount of supported copper oxide.
This may serve as an additional support to the greater
uniformity of the surface of CuO/СеO2 (me) catalysts,
as compared with (p) samples.
The dependence of the conversion of CO upon the
concentration of CuO on the surface of ZrO2 was
detected only on (p) samples: as the amount of supꢀ
ported CuO was changed from 2.5 to 5%, the converꢀ
sion of CO increased from 32 to 66%, whereas it
remained unchanged on the (me) catalyst (see Fig. 2).
According to Il’ichev et al. [30], the concentration of
coordinatively unsaturated Zr4+ cations and acid–
base pairs (Zr4+–О2–) was two times higher than that
on ZrO2 (me). Assuming the formation of Cu–O–Zr
active centers on such defects, we can explain the
higher activity of the 5% CuO/ZrO2 (p) sample in
comparison with 5% CuO/ZrO2 (me). According to
TPR data, raising the CuO content of the (me) samꢀ
ples from 2.5 to 5% results primarily in an increase in
the surface concentration of fineꢀparticle CuO, as in
the case of the CuO/CeO2 (me) surface.
Thus, the copper oxide catalysts prepared based on
the synthesized (me) supports were found less active
than CuO/(СеО2,ZrO2)(p) in the oxidation of minor
CO impurities in an excess of H2
.
ACKNOWLEDGMENTS
We are grateful to D.P. Shashkin for performing the
XRD analyses.
This work was supported by the Russian Foundaꢀ
tion for Basic Research (project no. 07ꢀ03ꢀ01074).
REFERENCES
1. Avgoropoulos, G., Ioannides, T., Papadopoulou, Ch.,
Baatista, J., Hocevar, S., and Matralis, H., Catal.
Today, 2002, vol. 75, p. 157.
2. Avgoropoulos, G. and Ioannides, T., Appl. Catal., A
2003, vol. 244, p. 155.
,
3. Kahlich, M.J., Gasteiger, H.A., and Behm, R.J.,
J. Catal., 1997, vol. 171, p. 93.
4. Lee, S.J., Mukerjee, S., Ticinelli, E.A., and VcBreen, J.,
Electrochim. Acta, 1999, vol. 44, p. 3283.
5. Avgoropoulos, G., Ioannides, T., Matralis, H.,
Baatista, J., and Hocevar, S., Catal. Lett., 2001, vol. 73,
p. 33.
6. Igarashi, H., Uchida, H., Suzuki, M., and Watanabe, M.,
Appl. Catal., A, 1997, vol. 159, p. 159.
7. Ito, S.ꢀI., Fujimori, T., Nagashima, K., Yuzaki, K., and
Kumimori, K., Catal. Today, 2000, vol. 57, p. 247.
8. Cheng, W.H., React. Kinet. Catal. Lett., 1996, vol. 58,
p. 329.
As noted previously [22], the number of active centers
on the surface of CuO/ZrO2 (p) (clusters like Cu–O–Zr)
is substantially smaller than that of Cu–O–Ce in the
case of CuO/СеO2 (p); this fact explains its low activꢀ
ity. As the particle size of ZrO2 was decreased from 20–
25 nm in CuO/ZrO2 (p) to 5–6 nm in the case of
CuO/ZrO2 (me) with a simultaneous increase in the
surface uniformity of the support, the number of Cu–
O–Zr clusters also decreased. Upon supporting higher
concentrations of CuO, only the fraction of finely disꢀ
persed surface copper oxide increased.
In the study of CuO/Zr0.5Се0.5O2 (me), we found
that Zr0.5Се0.5O2 (me) is a solid solution, whose surꢀ
face is enriched in СеО2 (according to TPR data), and 15. Liu, W. and FlytzaniꢀStefanopoulos, M., J. Catal.,
the binding energy of surface oxygen in Zr0.5Се0.5O2 (me)
is greater than that in СеО2 (me) at the same size of
crystallites. These data suggest the strong interaction
9. Torres, R.M., Sachez, A., Ueda, K., Tanaka, K., and
Haruta, M., J. Catal., 1997, vol. 168, p. 125.
10. Liotta, L.F., Di Cario, G., Pantaleo, G., and
Deganello, G., Catal. Commun., 2005, vol. 6, p. 329.
11. Kang, M., Song, M.W., and Lee, C.H., Appl. Catal., A
2003, vol. 251, no. 1, p. 143.
,
12. Tang, C.ꢀW., Kuo, C.ꢀC., Kuo, M.ꢀC., Wang, C.ꢀB..,
and Chien, S.ꢀH., Appl. Catal., A, 2006, vol. 309, p. 37.
13. Natile, M.M. and Glisenti, A., Chem. Mater., 2005,
vol. 17, no. 14, p. 3403.
14. Liu, W. and FlytzaniꢀStefanopoulos, M., J. Catal.
1995, vol. 153, p. 304.
,
1995, vol. 153, p. 317.
16. Sedmak, G., Hacevak, S., and Levec, J., J. Catal.
,
2003, vol. 213, p. 135.
KINETICS AND CATALYSIS Vol. 52
No. 3
2011

EFFECT OF THE NANOPARTICLE SIZE OF SUPPORTED COPPERꢀCONTAINING
445
17. Ratnasamy, P., Srinavas, D., Satyanarayana, C.V.V., 28. Caracoche, M., Rivas, P.C., and Cervera, M.M.,
Manikandak, P., Kumaran, R.S.S., Sachin, M., and
J. Mater. Chem., 2003, vol. 18, p. 208.
Shetti, V.N., J. Catal., 2004, vol. 221, p. 455.
18. Marban, J. and Fuertes, A.B., Appl. Catal., B, 2005,
29. MartinezꢀArias, A., FernandezꢀGarcia, M., Ballesꢀ
teros, V., Salamanca, L.N., Conesa, J.C., Otero, C.,
and Soria, J., Langmuir, 1999, vol. 15, p. 4796.
vol. 57, no. 1, p. 34.
19. Il’ichev, A.N., Firsova, A.A., and Korchak, V.N., Kinet.
Katal., 2006, vol. 47, no. 4, p. 602 [Kinet. Catal. (Engl.
Transl.), vol. 47, no. 4, p. 585].
20. Firsova, A.A., Il’ichev, A.N., Khomenko, T.I., Gorobꢀ
inskii, L.V., Maksimov, Yu.V., Suzdalev, I.P., and Korꢀ
chak, V.N., Kinet. Katal., 2007, vol. 48, no. 2, p. 298
30. Il’ichev, A.N., Shashkin, D.P., Khomenko, T.I., Fattaꢀ
khova, Z.T., and Korchak, V.N., Kinet. Katal., 2010,
vol. 51, no. 5, p. 769 [Kinet. Catal. (Engl. Transl.),
vol. 51, no. 5, p. 743].
31. Tret’yakov, I.I., Shub, B.R., and Sklyarov, A.V., Zh. Fiz.
Khim., 1970, vol. 44, p. 2112.
[Kinet. Catal. (Engl. Transl.), vol. 48, no. 2, p. 282].
32. Luo, M.ꢀF., Ma, J.ꢀQ., Song, Y.ꢀP., and Wang, Y.ꢀJ.,
21. Firsova, A.A., Khomenko, T.I., Il’ichev, A.N., and
J. Catal., 2007, vol. 246, p. 52.
Korchak, V.N., Kinet. Katal., 2008, vol. 49, no. 5, p. 713
33. Conesa, J.C., Proc. EuropaCat IX, Salamanca, Spain,
[Kinet. Catal. (Engl. Transl.), vol. 49, no. 5, p. 682].
2009, poster 1–17.
22. Firsova, A.A., Khomenko, T.I., Sil’chenkova, O.N.,
and Korchak, V.N., Kinet. Katal., 2010, vol. 51, no. 2, 34. Aneggi, E., Llorca, J., Boaro, M., and Trovarelli, F.,
p. 317 [Kinet. Catal. (Engl. Transl.), vol. 51, no. 2,
J. Catal., 2005, vol. 234, p. 88.
p. 299].
35. Krylov, O.V., Geterogennyi kataliz (Heterogonous
23. Iglesia, E., Proc. EuropaCat IX, Salamanca, Spain,
Catalysis), Moscow: Akademkniga, 2004.
2009.
36. MartinezꢀArias, A., Mungria, A.B., Munuera, G., and
24. Ishikawa, A., Neurock, M., and Iglesia, E., J. Am.
Chem. Soc., 2007, vol. 129, p. 13201.
25. Gololobov, A.M., Bekk, I.E., Bragina, G.O., Zaikꢀ
ovskii, V.I., Ayupov, A.B., Telegina, N.S., Bukhtiyarov, V.I.,
and Stakheev, A.Yu., Kinet. Katal., 2009, vol. 50, no. 6,
p. 864 [Kinet. Catal. (Engl. Transl.), vol. 50, no. 6,
p. 830].
Gamarra, D., Appl. Catal., B, 2006, vol. 65, p. 207.
37. LendzionꢀBielun, Z., Bettahar, M.M., and
Montererdi, S., Proc. EuropaCat IX, Salamanca, Spain,
2009, poster 3ꢀ13.
38. MartinezꢀArias, A., FernandezꢀGarcia, M., Soria, J.,
and Conesa, J.C., J. Catal., 1999, vol. 182, p. 367.
39. MartinezꢀArias, A., FernandezꢀGarcia, M., Galvez, O.,
Coronado, J.M., Anderson, J.A., Conesa, J.C., Soria, J.,
and Munuera, G., J. Catal., 2000, vol. 195, p. 207.
26. GomezꢀCortes, A., Marquez, Y., ArenasꢀAlatorre, J.,
and Diaz, G., Catal. Today, 2008, vols. 133–135,
p. 743.
27. Benfer, S. and Knozinger, E., J. Mater. Chem., 1999, 40. Hadjiivanov, K. and Knözenger, H., J. Catal., 2000,
vol. 9, p. 1203.
vol. 191, p. 480.
KINETICS AND CATALYSIS Vol. 52
No. 3
2011