Copper-cerium oxide catalysts for the selective oxidation of carbon monoxide in hydrogen-containing mixtures: I. Catalytic activity

Article abstract of DOI:10.1134/S0023158407030135

A series of copper-cerium oxide catalysts were prepared, and their properties toward the reaction of CO oxidation in hydrogen-containing gas mixtures were studied. It was found that the copper-cerium oxide catalysts are stable, active, and selective in this reaction. The conditions under which these catalysts decreased the concentration of CO from 1 to <10^-3 vol % in hydrogen containing water vapor and carbon dioxide were determined.

Full text of DOI:10.1134/S0023158407030135

ISSN 0023-1584, Kinetics and Catalysis, 2007, Vol. 48, No. 3, pp. 439–447. © MAIK “Nauka /Interperiodica” (Russia), 2007.
Original Russian Text © P.V. Snytnikov, A.I. Stadnichenko, G.L. Semin, V.D. Belyaev, A.I. Boronin, V.A. Sobyanin, 2007, published in Kinetika i Kataliz, 2007, Vol. 48, No. 3,
pp. 463–471.
Copper–Cerium Oxide Catalysts for the Selective Oxidation
of Carbon Monoxide in Hydrogen-Containing Mixtures:
I. Catalytic Activity
P. V. Snytnikova,b, A. I. Stadnichenkoa, G. L. Semina, V. D. Belyaeva,
A. I. Boronina,b, and V. A. Sobyanina,b
a Boreskov Institute of Catalysis, Siberian Branch, Russian Academy of Sciences, Novosibirsk, 630090 Russia
b Novosibirsk State University, Novosibirsk, 630090 Russia
E-mail: pvsnyt@catalysis.ru
Received February 26, 2006
Abstract—A series of copper–cerium oxide catalysts were prepared, and their properties toward the reaction
of CO oxidation in hydrogen-containing gas mixtures were studied. It was found that the copper–cerium oxide
catalysts are stable, active, and selective in this reaction. The conditions under which these catalysts decreased
the concentration of CO from 1 to <10–3 vol % in hydrogen containing water vapor and carbon dioxide were
determined.
DOI: 10.1134/S0023158407030135
INTRODUCTION
The catalytic properties of the CuO–CeO2 system
toward the reaction of CO oxidation in the presence of
hydrogen were compared with the properties of other
well-known Pt- and Au-containing catalysts [1, 7]. It
was found that Au/α-Fe2O3 was the most active (low-
temperature) catalyst for both a gas mixture free of H2O
and CO2 and gas mixtures containing water and carbon
dioxide, whereas CuO–CeO2 was more active than
Pt/γ-Al2O3 and superior to all of these catalysts in selec-
tivity.
In general, the above suggests that the CuO–CeO2
system is a very promising catalyst for the fine purifica-
tion of a hydrogen-containing gas to remove CO. Tak-
ing into account the fact that this system is free of noble
metals, it is reasonable to study this system in more
detail.
The development of highly active catalysts for the
fine purification of hydrogen to remove CO with the use
of selective CO oxidation is an important problem of
the preparation of hydrogen for fuel cells with a poly-
mer proton-conducting electrolyte. A hydrogen-con-
taining gas mixture prepared by the steam or air reform-
ing of hydrocarbon fuels followed by the CO water gas
shift reaction usually contains 0.5–2 vol % CO. At the
same time, it is well known that CO in this concentra-
tion is a poison for fuel cell electrodes, and the concen-
tration of CO should be decreased to a level lower than
10–100 ppm. A serious attempt has been made to find
inexpensive catalytic systems containing no noble met-
als for the fine purification of a hydrogen-containing
gas to remove CO. Catalysts based on oxide copper–
cerium systems are of greatest current interest.
This work was devoted to a systematic study of the
oxidation reaction of carbon monoxide in hydrogen-
containing gas mixtures on Cu/CeO2 − x catalysts ori-
ented to the determination of conditions for the fine
purification of hydrogen to remove CO to a level of
10 ppm.
Avgouropoulos et al. [1] were the first to propose
copper–cerium oxide catalysts for the reaction of CO
oxidation in the presence of hydrogen. They studied
CuO–CeO2 catalysts prepared by a coprecipitation
method. They found that the CuO–CeO2 system is an
active and selective catalyst for the oxidation of CO in
the presence of H2 and it can be used for the fine purifi-
cation of a hydrogen-containing gas to remove CO. The
subsequent publications [2–6] have fully confirmed this
conclusion. Note that, along with the CuO–CeO2 sys-
tem, copper catalysts supported on CeO2 doped with
zirconium oxide or samarium oxide were studied [2–4,
6]. These catalysts were found to be less active and
EXPERIMENTAL
Preparation of Cu/CeO2 – x Catalysts
by the Coprecipitation Method
A concentrated aqueous solution of oxalic acid was
rapidly added (for 1 min) to an aqueous solution of
Cu(NO3)2 and Ce(NO3)3 with stirring at room tempera-
ture, and the mixture was stirred for 20 min with a
selective than CuO–CeO2 in the reaction of CO oxida- mechanical stirrer. The resulting precipitate was
tion in the presence of H2. allowed to stand for 30 min without stirring for aging.
439

440
SNYTNIKOV et al.
Thereafter, the supernatant solution was decanted and from entering the capillary tube system of gas inlet and
the precipitate was washed with distilled water, dried in outlet lines with a gas flow. The temperature was mea-
air at 110°C for 4 h, and then calcined at a temperature sured with a Chromel–Alumel thermocouple, which
from 400 to 900°C for 2 h.
was placed in the well at the center of the catalyst bed.
No catalyst pretreatment processes were applied before
the experiments.
Preparation of Cu/CeO2 − x Catalysts
The concentrations of reactants and reaction prod-
ucts at the reactor inlet and outlet were determined
using a Kristall 2000 chromatograph (Russia) equipped
with a thermal-conductivity detector and a flame-ion-
ization detector with a methanator containing an NKM-
4 nickel catalyst. The combination of a methanator and
a flame-ionization detector allowed us to analyze any
gaseous hydrocarbons present in the gas mixture in
combination with CO and CO2 with a high sensitivity.
The test mixture was separated on a column with
molecular sieves NaX (thermal-conductivity detector)
and on a column with Porapak Q (flame-ionization
detector). The sensitivity of the technique allowed us to
measure the concentrations of CO, ëç4, and CO2 to the
by the Impregnation Method
To prepare catalysts by this method, we initially pre-
pared cerium oxide in accordance with the following
procedure: A concentrated solution of oxalic acid was
rapidly added (for 1 min) to an aqueous solution of
Cu(NO3)2 with stirring at room temperature, and the
mixture was stirred for 20 min with a mechanical stir-
rer. The resulting precipitate was allowed to stand for
30 min without stirring for aging. Thereafter, the super-
natant solution was decanted and the precipitate was
washed with distilled water and dried in air at 110°C for
4 h; then, the sample was calcined at 400°C in air for
2 h. The specific surface area of the resulting cerium
oxide was 66 m2/g.
level of ~10–4 vol % and O2 to 10–3 vol %. The occur-
rence of the reaction of carbon monoxide oxidation in
hydrogen-containing mixtures was characterized by the
The cerium oxide powder thus prepared was incipi-
ent wetness impregnated with an aqueous solution of
Cu(NO3)2. Then, the sample was dried in air at 110°C
for 4 h and calcined at a temperature from 400 to 900°C
for 2 h.
conversions of CO (Xëé) and é2 (XO ) and selectivity
2
(S), which were calculated from the equations
The powders prepared by coprecipitation and
impregnation were pressed into pellets, which were
then crushed. A fraction with a catalyst particle size of
0.25–0.50 mm was used in catalytic experiments.
[CO]0 – [CO]
---------------------------------
XCO
=
× 100%;
× 100%;
× 100%,
[CO]0
[O2]0 – [O2]
------------------------------
[O2]0
XO
=
The concentrations of the main components in the
copper–cerium catalysts were determined by induc-
tively coupled plasma atomic emission spectrometry on
an Optima instrument from Perkin-Elmer. The analyti-
cal procedure involved the dissolution of a solid cata-
lyst sample, the dilution of the resulting solution to a
concentration required for spectrometric analysis, and
photometric measurements using a calibration graph
method. The copper content of a catalyst (at %) was
determined from the equation 100NCu/(NCu + NCe),
where NCu and NCe are the numbers of copper and
cerium atoms, respectively, in the test sample.
2
[CO] – [CO]
1
0
-----------------------------------
S =
2 [O2]0 – [O2]
where [ëé]0 and [é2]0 are the concentrations of CO
and O2, respectively, at the reactor inlet; [CO] and [O2]
are the concentrations of CO and O2, respectively, at the
reactor outlet.
RESULTS AND DISCUSSION
The specific surface areas (SBET) and pore volumes
(Vpore) of supports and catalysts were determined from
the full isotherms of low-temperature nitrogen adsorp-
tion at 77 K, which were measured on an ASAP-2400
instrument (Micrometrics, United States).
The procedure used in the kinetic experiments was
described in detail elsewhere [8]. The reaction was per-
formed in a quartz flow reactor at atmospheric pressure.
The reactor was a U-shaped tube 40 cm in length and
Activity and Selectivity of Cu/CeO2 − x Catalysts
Prepared by Impregnation
and Coprecipitation Methods
The reaction of CO oxidation in the presence of H2
was studied on oxide copper–cerium catalysts contain-
ing from 2.5 to 30 at % Cu calcined in air at tempera-
tures from 400 to 900°C.
It was found that catalysts containing ~15 at % Cu
8 mm in i.d. with a wall thickness of ~1 mm. A thermo- calcined at 400°C exhibited the highest activity and
couple well 3 mm in o.d. was arranged at the center selectivity toward the oxidation reaction of CO in
along the axis of the reactor. A weighed portion of the hydrogen-containing mixtures. Therefore, we report
catalyst was mixed with powdered quartz (particle size here the results obtained on the catalysts containing
of ~1 mm) and placed in the reactor. The length of the ~15 at % Cu calcined at 400°C. These catalysts, which
catalyst bed was 12 cm. Filters were placed at the reac- were prepared using coprecipitation and impregnation,
tor inlet and outlet to prevent small catalyst particles contained 14.4 and 14.6 at % Cu, respectively; they are
KINETICS AND CATALYSIS Vol. 48 No. 3 2007

COPPER–CERIUM OXIDE CATALYSTS FOR THE SELECTIVE OXIDATION... I
441
Table 1. Characteristics of oxide copper–cerium catalysts
SBET Vpore
subsequently referred to as Cu/CeO2 − x(C) and
Cu/CeO2 − x(I), respectively.
,
,
Catalyst
m2/(g Cat) cm3/(g Cat)
Data given in Table 1 indicate that catalyst charac-
teristics such as the copper content, specific surface
area (SBET), and pore volume (Vpore) were identical in
catalyst samples before and after the reaction of CO
oxidation in the presence of H2; this suggests the stabil-
ity of catalyst structures under reaction conditions.
CeO2
66
70
0.22
0.18
14.4 at % Cu/CeO2 – x(C)
coprecipitation, before reaction
14.4 at % Cu/CeO2 – x(C)
69
65
65
0.19
0.20
0.20
coprecipitation, after reaction
Figure 1 shows the temperature dependence of the
conversions of CO and O2 and selectivity in the reaction
of CO oxidation in the presence of hydrogen. It can be
seen that, in all cases, the conversion of CO (Xëé) ini-
tially increased and than decreased with temperature.
The maximum values of Xëé were as high as 99.9% at
170°C on the Cu/CeO2 − x(I) catalyst, 98% at 205°ë on
the Cu/CeO2 − x(C) catalyst, and only 60% at 330°C on
pure CeO2. The conversion of oxygen increased with
temperature and reached 100% at T ≥ 170°ë on the
14.6 at % Cu/CeO2 – x(I)
impregnation, before reaction
14.6 at % Cu/CeO2 – x(I)
impregnation, after reaction
the composition of a hydrogen-containing mixture on
the course of CO oxidation in the presence of H2. The
experimental results are considered below.
Effect of CO2 and H2O vapor. We performed a
number of experiments with the use of the following
hydrogen-containing mixtures:
Cu/CeO2 − x(I) catalyst,
T
≥
205°ë on the
Cu/CeO2 − x(C) catalyst, and T ≥ 330°ë on pure CeO2.
The 14.6 at % Cu/CeO2 − x catalyst prepared by
impregnation was found to be the most selective cata-
lyst toward CO oxidation in the presence of H2. Almost
100% selectivity was observed on this catalyst at T ≤
140°C. A further increase in the temperature resulted in
a dramatic decrease in selectivity to 33%.
The catalyst prepared by coprecipitation also exhib-
ited high selectivity; however, its value was lower than
that of the Cu/CeO2 − x(I) catalyst. It gradually
decreased from 80 to 60% over the temperature range
130–190°C and then rapidly decreased to 33%. The
selectivity of CeO2 was much lower than that of oxide
copper–cerium catalysts, and it gradually decreased
from ~30 to 20% as the temperature was increased.
Thus, the experimental results suggest that cerium
dioxide is not an inert material toward CO oxidation in
the presence of H2. However, it exhibited much lower
activity and selectivity than those of oxide copper–
cerium systems. Of these systems, the Cu/CeO2 − x(I)
catalyst was found to be much more active and selective
than Cu/CeO2 − x(C). Taking into account that the cop-
per contents of both of the copper–cerium oxide cata-
lysts were approximately equal, the above difference
can be explained by the higher dispersion of copper in
the catalyst prepared by impregnation, as compared
with that in the coprecipitated catalyst. This conclusion
is consistent with published data [1–3].
(1) 1 vol % CO + 1 vol % O2 + 65 vol % H2
+ balance He;
(2) 1 vol % CO + 1 vol % O2 + 65 vol % H2
+ 20 vol % CO2 + balance He;
(3) 1 vol % CO + 1 vol % O2 + 65 vol % H2
+ 10 vol % H2O + balance He;
(4) 1 vol % CO + 1 vol % O2 + 65 vol % H2
+ 20 vol % CO2 + 10 vol % H2O + balance He.
Figure 2 illustrates the experimental data. It can be
seen that, with the use of a gas mixture free of CO2 and
H2O, the conversion of CO increased and then
decreased with temperature. A temperature window in
which the conversion of CO was higher than 99.9% was
observed over the temperature range 150–170°C. The
conversion of O2 increased much more slowly than the
conversion of CO with temperature, and it reached
~50% at 130°C at almost 100% selectivity. By this is
meant that the oxidation of hydrogen did not occur
under these conditions. An increase in the temperature
above 130°C resulted in a rapid growth of oxygen con-
version to 100% and, correspondingly, a decrease in
selectivity to 40% at 180°C.
The introduction of 20 vol % CO2 or 10 vol % H2O
resulted in a decrease in the activity of the
Cu/CeO2 − x(I) catalyst in the oxidation reaction of CO
in the presence of H2 and in a shift of the temperature
dependence of CO and O2 conversions and selectivity
to the region of higher temperatures by ~30–70°C, as
compared with the dependence obtained in a mixture
containing no carbon dioxide and water vapor. In the
presence of 20 vol % CO2, a maximum conversion of
CO (99.9%) was reached at 185–190°C. In the presence
of 10 vol % H2O, a maximum conversion of CO
(99.6%) was observed at 185°C.
The subsequent studies of the reaction of CO oxida-
tion in the presence of H2 were performed with the
14.6 at % Cu/CeO2 − x catalyst prepared by impregna-
tion, which is the most active and selective copper–
cerium catalyst.
Effect of the Composition of the Hydrogen-Containing
Mixture on the Course of Reaction
In the case that carbon dioxide and water vapor were
To determine optimum conditions for the operation simultaneously present in a mixture, a maximum con-
of the Cu/CeO2 − x(I) catalyst, we studied the effect of version of CO was equal to 97.7% at T = 195°C. Thus,
KINETICS AND CATALYSIS Vol. 48 No. 3 2007

442
SNYTNIKOV et al.
XCO, %
XCO, %
2
100
(‡)
100
80
60
40
20
0
(‡)
3
80
60
40
20
0
1
1
2
3
4
100
150
200
250
300
(b)
350
70
XO , %
100
130
130
130
160
160
160
190
220
220
220
X , %
O2
2
100
(b)
100
80
60
40
20
0
1
80
60
40
20
0
2
3
70
100
190
(c)
100
150
200
250
300
(c)
350
S, %
S, %
100
100
80
60
40
20
0
80
60
40
20
2
3
1
70
100
190
100
150
200
250
300
350
T, °C
T, °C
Fig. 2. Dependence of the conversions of (a) CO and (b) O
2
and (c) selectivity on the reaction temperature of CO oxidation
in the presence of H on the CuO/CeO (I) catalyst for vari-
Fig. 1. Dependence of the conversions of (a) CO and (b) O
2
2
2 − x
and (c) selectivity on the reaction temperature of CO oxida-
tion in the presence of H on (1) CeO , (2) CuO/CeO (I),
ous gas mixtures: (1) 1 vol % CO, 1 vol % O , 65 vol % H ,
2
2
2
2
2 − x
and the balance He; (2) mixture 1 + 20 vol % CO ; (3) mix-
and (3) CuO/CeO
(C). Gas mixture: 1 vol % CO,
2
2 − x
ture 1 + 10 vol % H O; (4) mixture 1 + 20 vol % CO +
1.5 vol % O , 65 vol % H , 3 vol % H O, 20 vol % CO , and
2
2
2
2
2
–1
–1
10 vol % H O. Space velocity of the gas mixture: 24000 h .
the balance He. Space velocity of the gas mixture: 12000 h .
2
lyst operation in the oxidation of CO in hydrogen-con-
taining mixtures. Nevertheless, all of the hydrogen-
the presence of CO2 and H2O in a hydrogen-containing
mixture resulted in a decrease in activity and in a nar-
rowing of the temperature window of Cu/CeO2 − x cata- containing mixtures used exhibited a temperature
KINETICS AND CATALYSIS Vol. 48 No. 3 2007

COPPER–CERIUM OXIDE CATALYSTS FOR THE SELECTIVE OXIDATION... I
443
region in which almost 100% selectivity was reached:
XO , %
2
T < 130°C in the model mixture (without CO2 and
H2O), T < 140°C in the presence of CO2, T < 140°C in
the presence of H2O, or T < 160°C with the use of a real
mixture (containing CO2 and H2O). A further increase
in the temperature resulted in a rapid decrease of selec-
tivity to 40%. Behavior of this kind, which is character-
istic of copper–cerium oxide systems, was also found in
previous studies [1–3].
100
80
60
40
20
0
1
2
3
4
To reveal the reason for the occurrence of a temper-
ature region in which almost 100% selectivity was
reached on the Cu/CeO2 − x(I) catalyst, we performed
experiments on the separate oxidation of CO and the
oxidation of H2 both in the absence of CO2 and H2O and
in the presence of 20 vol % CO2 and 10 vol % H2O. For
this purpose, we used the following mixtures:
70
100
130
160
190
220
T, °C
(1) 1 vol % CO + 0.5 vol % O2 + balance He;
Fig. 3. The temperature dependence of the conversion of O in
2
(2) 1 vol % CO + 0.5 vol % O2 + 20 vol % CO2 +
10 vol % H2O + balance He in the reaction of CO oxi-
dation and
the course of the oxidation of (1, 3) CO and (2, 4) H (1, 2) in
2
the absence of H O and CO or (3, 4) in the presence of
2
2
10 vol % H O and 20 vol % CO on the CuO/CeO
2
2
2 − x
(1) 65 H2 + 0.5 O2 + balance He;
(I) catalyst. Gas mixtures in the reaction of CO oxidation:
(1) 1 vol % CO, 0.5 vol % O , and the balance He; (3) mix-
2
(2) 65 H2 + 0.5 O2 + 20 vol % CO2 + 10 vol % H2O
+ balance He in the reaction of H2 oxidation.
ture 1 + 20 vol % CO + 10 vol % H O. Gas mixtures in the
2
2
reaction of H oxidation: (2) 65 vol % H , 0.5 vol % O , and
2
2
2
To compare the activity of a copper–cerium oxide
catalyst in the reactions of CO and H2 oxidation, the
courses of both of the reactions were monitored by
measuring the conversion of O2.
the balance He; (4) mixture 2 + 20 vol % CO + 10 vol %
2
–1
H O. The space velocity of the gas mixture was 24000 h
2
in all of the experiments.
In Fig. 3, it can be seen that only the oxidation of CO
occurred on the copper–cerium oxide catalyst in the
absence of H2O and CO2 from the mixture to T ~100°C,
and the oxidation of H2 was insignificant (it came into
play only at a higher temperature).
Effect of O2 and CO. One of the main parameters
affecting the efficiency of CO removal from hydrogen-
containing mixtures is the initial concentration of oxy-
gen. To choose optimum conditions, the minimum
amount of oxygen required for decreasing the concen-
tration of CO to 10 ppm should be known.
In the presence of H2O and CO2 in the reaction mix-
ture, the activity of the copper–cerium oxide catalyst in
the oxidation of both CO and H2 decreased. Neverthe-
less, the oxidation of CO primarily occurred to
T ~150°C, whereas the oxidation of H2 was observed in
the region of higher temperatures.
Figure 4 shows the dependence of the outlet concen-
tration of CO, the conversion of O2, and the selectivity
on the reaction temperature of CO oxidation in the pres-
ence of H2 at various initial concentrations of oxygen
(0.5–1.5 vol %) on the Cu/CeO2 − x(I) catalyst. It can be
seen that an increase in the concentration of oxygen had
no effect on the final concentration of CO at tempera-
tures lower than 150°C. In this case, the values of selec-
tivity for all of the three concentrations of oxygen were
close to each other and insignificantly decreased from
100 to 90%.
By this is meant that the activity of the
Cu/CeO2 − x(I) catalyst in CO oxidation was much
higher than the activity in H2 oxidation both in the
absence and in the presence of CO2 and H2O in the
reaction mixture. It is likely that, because of this, there
is a temperature region in which almost 100% selectiv-
ity was observed in the reaction of CO oxidation in the
presence of H2. The selectivity considerably decreased
at the instant the oxidation of H2 came into play.
From Fig. 4, it follows that, with the use of a stoichi-
It is most likely that the catalyst activity in the pres- ometric mixture containing 1 vol % CO and 0.5 vol %
ence of H2O and CO2 decreased because of the block- O2, a minimum outlet concentration of CO (300 ppm)
ing of catalyst active sites by adsorbed CO2 and H2O at 100% O2 conversion was observed at 170°C. An
molecules. This hypothesis is based on the following increase in the concentration of oxygen to 1 vol %
experimental data: No products were formed when a allowed us to decrease the final concentration of CO to
mixture of 20 vol % CO2 and 80 vol % H2 was supplied 10 ppm at 180°C. However, in this case, the conversion
to the catalyst. By this is meant that the reverse CO of oxygen rapidly reached a value of 100% to result in
water gas shift reaction does not occur under the reac- a decrease in selectivity to 50%. At an initial oxygen
tion conditions of CO oxidation in the presence of H2.
concentration of 1.5 vol %, a minimum concentration
KINETICS AND CATALYSIS Vol. 48 No. 3 2007

444
SNYTNIKOV et al.
regardless of the initial concentration of oxygen, and
[CO]outlet, ppm
104
even an insignificant increase in the temperature
resulted in a rapid increase in the concentration of CO
at the reactor outlet. For example, even at 210°C and an
O2/CO ratio of 0.5, 1.0, or 1.5, the outlet concentration
of CO was equal to 2000, 1500, or 900 ppm, respec-
tively. Thus, the temperature window in which the con-
centration of CO at the reactor outlet was lower than
10 ppm was narrow and equal to 10–15°C for the oxide
copper–cerium catalyst.
(‡)
103
102
1
2
3
101
Usually, after the low-temperature CO water gas
shift reaction, the hydrogen-containing mixture con-
tained from 0.5 to 2 vol % CO depending on reaction
conditions. Because of this, the effect of the initial con-
centration of CO on the occurrence of CO oxidation in
the presence of H2 should be studied. Figure 5 shows
the dependence of the conversion of CO and O2 and the
selectivity on the reaction temperature of CO oxidation
in hydrogen-containing mixtures on the Cu/CeO2 − x(I)
catalyst at CO concentrations varied from 0.5 to 2.0 vol %
and a constant O2/CO ratio of 0.5.
100
110
130
150
170
190
210
(b)
XO , %
2
100
80
60
40
20
0
It can be seen that the temperature dependence of
Xëé, XO , and S was almost the same for all of the three
2
gas mixtures. A maximum CO conversion of ~90% was
reached at 180°C in gas mixtures containing 2.0, 1.0,
and 0.5 vol % CO. A further increase in the temperature
caused a rapid decrease in the conversion of CO.
An analogous dependence was also observed at
other O2/CO ratios. Table 2 summarizes the experimen-
tal results. It can be seen that a maximum conversion of
CO was practically the same at various initial concen-
trations of CO and a constant O2/CO ratio. The conver-
sion of CO was increased only by an increase in the
O2/CO ratio.
110 130 150 170 190 210 230
S, %
(c)
100
80
60
40
20
0
Optimum conditions for the fine purification of
hydrogen-containing mixtures to remove CO. Not
only the data considered above but also data on the
effect of the flow rate of a reaction mixture on the
course of CO oxidation in the presence of O2 are
required for optimizing conditions for the fine purifica-
tion of a hydrogen-containing gas to remove CO. The
experiments were performed with the following mix-
ture: 1 vol % CO, O2/CO = 0.5, 65 vol % H2, 10 vol %
H2O, 20 vol % CO2, and the balance He. As can be seen
in Fig. 6, an increase in the flow rate shifted the temper-
ature dependence of the final concentration of CO and
the conversion of O2 to the region of higher tempera-
tures. In this case, the minimum concentration of CO
increased. Thus, at a space velocity of 6000 h–1, a min-
imum concentration of CO was equal to 300 ppm at
170°C, whereas it was equal to 1000, 2300, or
2700 ppm at a space velocity of 24000, 42000, or
60000 h–1 and a temperature of 180, 190, or 195°C,
respectively. Note that the conversion of O2 reached
100% at the specified temperatures. A further increase
in the temperature resulted in a dramatic increase in the
110 130 150 170 190 210 230
T, °C
Fig. 4. Dependence of (a) the outlet concentration of CO,
(b) the conversion of O , and (c) selectivity on the reaction
2
temperature of CO oxidation in the presence of H on the
2
CuO/CeO
(I) catalyst at O concentrations of (1) 0.5,
2 − x
2
(2) 1.0, and (3) 1.5 vol % for a gas mixture of 1 vol % CO,
65 vol % H , 10 vol % H O, 20 vol % CO , and the balance
2
2
2
–1
He. Space velocity of the gas mixture: 6000 h .
of CO (2 ppm) was reached at 185°C at 100% conver-
sion of oxygen and 33% selectivity.
Note that a minimum outlet concentration of CO outlet concentration of CO at the complete conversion
was observed over a very narrow temperature range of oxygen. Note that, even at 220°C, the final concen-
KINETICS AND CATALYSIS Vol. 48 No. 3 2007

COPPER–CERIUM OXIDE CATALYSTS FOR THE SELECTIVE OXIDATION... I
445
X
CO, %
[CO]outlet, ppm
104
(‡)
100
80
60
40
20
0
(‡)
103
102
1
2
3
4
1
2
3
101
100
130
150
170
170
170
190
190
190
210
230
230
230
110 130 150 170 190 210 230
XO , %
XO , %
2
2
(b)
100
80
60
40
20
0
(b)
100
80
60
40
20
0
130
150
210
(c)
110 130 150 170 190 210 230
S, %
S, %
100
100
(c)
80
60
40
20
0
80
60
40
20
0
130
150
210
T, °C
110 130 150 170 190 210 230
T, °C
Fig. 6. Dependence of (a) the outlet concentration of CO,
Fig. 5. Dependence of the conversions of (a) CO and (b) O
2
(b) the conversion of O , and (c) selectivity on the reaction
2
and (c) selectivity on the reaction temperature of CO oxida-
tion in the presence of H on the CuO/CeO (I) catalyst at
temperature of CO oxidation in the presence of H on the
2
2
2 − x
CuO/CeO
(I) catalyst at space velocities of (1) 6000,
CO concentrations of (1) 0.5, (2) 1.0, and (3) 2.0 vol % in a
gas mixture also containing 65 vol % H , 10 vol % H O,
2 − x
–1
2
2
(2) 24000, (3) 42000, and (4) 60000 h . Gas mixture:
1 vol % CO, 0.5 vol % O , 65 vol % H , 10 vol % H O,
20 vol % CO , and the balance He at O /CO = 0.5. Space
2
2
2
2
2
–1
20 vol % CO , and the balance He.
velocity of the gas mixture: 24000 h .
2
tration of CO was equal to ~4000 ppm at all of the flow marizes minimum outlet concentrations of CO and the
rates.
reaction temperature depending on the rate of a gas
flow and the initial O2/CO concentration ratio. Based
An analogous dependence was also obtained at
other initial O2/CO concentration ratios. Table 3 sum- on these data, the following conclusions can be made:
KINETICS AND CATALYSIS Vol. 48 No. 3 2007

446
SNYTNIKOV et al.
Table 2. Maximum conversion of CO (%) at various O2/CO
ratios and initial CO concentrations in the oxidation of CO on
the Cu/CeO2 – x(I) catalyst in the presence of H2
flow space velocity should be decreased. For example,
the maximum space velocity was equal to 6000 h–1 at
O2/CO = 1.
Initial CO concentration, vol %
Comparison between the catalytic properties of
Cu/CeO2 - x systems. The known published data on the
oxidation of CO in the presence of H2 on oxide copper–
cerium catalysts were obtained during the period when
this study was performed. Therefore, it seemed reason-
able to compare the above results with published data
[1, 3, 5]. In the works cited, the reaction of CO oxida-
tion in hydrogen-containing mixtures was studied on
copper–cerium oxide catalysts prepared using various
methods in both a model mixture with no CO2 and H2O
and a mixture containing CO2 and H2O. It was found
that catalysts containing 15–20 at % Cu exhibited the
highest activity. We also found that the 14.6 at %
Cu/CeO2 − x catalyst prepared by impregnation exhib-
ited the highest activity and selectivity in CO oxidation
in the presence of H2.
O2/CO ratio
í, °C
2.0
1.0
0.5
0.5
1.0
1.5
180
190
195
87.0
95.0
97.5
90.0
95.0
97.7
91.0
94.0
96.0
Note: The initial mixture contained 0.5–2 vol % CO, 0.25–3 vol %
O , 65 vol % H , 10 vol % H O, 20 vol % CO , and the bal-
2
2
2
2
–1
ance He. The space velocity of the gas mixture was 24000 h .
• an increase in the flow rate resulted in an increase
in the minimum concentration of CO at the outlet and
in an increase in the temperature at which the minimum
concentration of carbon monoxide was reached;
• an increase in the concentration of O2 resulted in a
decrease in the minimum concentration of CO at the
outlet and in an increase in the reaction temperature at
which this concentration was reached. This relation was
true at all gas flow rates. For example, at a space veloc-
ity of 6000 h–1, an increase in the initial O2/CO concen-
tration ratio by a factor of 3 decreased the outlet con-
centration from 300 to 2 ppm, whereas the addition of
oxygen decreased the outlet concentration of CO from
2700 to 2000 ppm at a space velocity of 60000 h–1.
Avgouropoulos et al. [1] demonstrated that the
14.3 at % Cu/CeO2 catalyst prepared by coprecipitation
decreased the concentration of CO in a hydrogen-con-
taining gas to 10 ppm in a gas mixture containing
1 vol % CO, 1.25 vol % O2, 50 vol % H2, and the bal-
ance He at T = 130–150°C and a space velocity of
10000 h–1. In our experiments (see Fig. 2), the
Cu/CeO2 − x(I) catalyst provided the above degree of
CO removal from a hydrogen-containing gas under
similar experimental conditions (inlet concentrations of
1 vol % CO, 1 vol % O2, 65 vol % H2, and the balance
He; T = 150–170°C; and space velocity of 24000 h–1).
Avgouropoulos et al. [1] also studied the effect of
water and carbon dioxide on the course of CO oxidation
in hydrogen-containing mixtures. These experiments
were performed at a space velocity of 10000 h–1. They
found that the presence of 15 vol % CO2 and 10 vol %
Figure 7 shows the dependence of the outlet concen-
tration of CO on the space velocity of the gas flow and
the initial O2/CO concentration ratio; this dependence
was obtained at the optimum temperature (170–215°C),
which was determined previously (see Figs. 4–6). The
graph was subdivided into four regions with respect to
the outlet concentrations of CO: 1000–10000, 100–
1000, 10–100, and 1–10 ppm.
As can be seen in Fig. 7, the maximum gas-flow H2O in a model mixture resulted in a shift of the tem-
space velocity at which the removal of CO from the gas perature dependence of CO and O2 conversions by 15–
mixture below 10 ppm was reached was equal to 70°C to the region of higher temperatures, as compared
~14000 h–1 at an O2/CO ratio of no lower than 1.5. with the dependence observed in the model mixture. A
Either an increase in the velocity or a decrease in the maximum CO conversion of >99% at 40% selectivity
initial concentration of O2 resulted in an increase in the was reached at 190°C. An analogous decrease in the
final concentration of CO. Thus, to remove CO at a activity of a copper–cerium oxide catalyst was also
level lower than 10 ppm and at a lower O2/CO ratio, the observed in our experiments on the Cu/CeO2 − x(I) cata-
Table 3. Minimum outlet concentrations of CO and reaction temperature depending on space velocity and initial O2/CO con-
centration ratio in the course of CO oxidation on the Cu/CeO2 – x(I) catalyst in the presence of H2
Inlet O2/CO concentration ratio
Space
0.5
1.0
1.5
[CO]outlet, ppm
velocity, h–1
T, °C
[CO]outlet, ppm
T, °C
[CO]outlet, ppm
T, °C
6000
24000
42000
60000
170
180
190
195
300
1000
2300
2700
180
190
205
210
10
500
1200
2100
185
195
210
215
2
230
580
2000
Note: The initial mixture contained 1 vol % CO, 65 vol % H , 10 vol % H O, 20 vol % CO , and the balance He; O /CO = 0.5–1.5.
2
2
2
2
KINETICS AND CATALYSIS Vol. 48 No. 3 2007

COPPER–CERIUM OXIDE CATALYSTS FOR THE SELECTIVE OXIDATION... I
447
Space velocity, h–1
60000
1000–10 000 ppm
42000
100–1000 ppm
24000
10–100 ppm
1–10 ppm
6000
0.5 0.6 0.7 0.8 0.9 1.0 1.1 1.2 1.3 1.4 1.5
Initial O2/CO concentration ratio
Fig. 7. Dependence of the outlet concentration of CO on the initial O /CO concentration ratio and the space velocity of a gas flow
2
in the reaction of CO oxidation in the presence of H on the CuO/CeO
(I) catalyst. Gas mixture: 1 vol % CO, 0.5–1.5 vol % O ,
2
2 − x
2
65 vol % H , 10 vol % H O, 20 vol % CO , and the balance He.
2
2
2
lyst. A CO conversion of >99.9% and 38% selectivity work of the Federal Special Scientific and Technical
were reached at 180°C in a mixture containing 1 vol % Program “Research and Development in Priority Areas
CO, 1.3 vol % O2, 65 vol % H2, 20 vol % CO2, 10 vol % of Science and Technology” (2006-RI-19.0/001/0.15).
H2O, and the balance He at a space velocity of 10000 h–1.
Kim and Cha [5] found that the most active copper–
cerium oxide catalyst containing 20 at % Cu, which
REFERENCES
was prepared by the coprecipitation method, decreased
the concentration of CO below 10 ppm in a mixture
containing 1 vol % CO, 1 vol % O2, 50 vol % H2,
13.5 vol % CO2, 20 vol % H2O, and the balance He at a
space velocity of 15000 h–1 over the temperature range
167–176°C. In our experiments, the Cu/CeO2 − x(I) cat-
alyst decreased the concentration of CO below 10 ppm
in a mixture containing 1 vol % CO, 1.5 vol % O2,
65 vol % H2, 20 vol % CO2, 10 vol % H2O, and the bal-
ance He at a space velocity of 14000 h–1 only over a
narrow temperature window of 185–195°C.
The above comparison suggests that the copper–
cerium oxide catalysts studied in this work and
described in the literature are similar in terms of both
activity and selectivity in the reaction of CO oxidation
in hydrogen-containing mixtures.
1. Avgouropoulos, G., Ioannides, T., Matralis, H.K.,
Batista, J., and Hocevar, S., Catal. Lett., 2001, vol. 73,
p. 33.
2. Ratnasamy, P., Srinivas, D., Satyanarayana, C.V.V.,
Manikandan, P., Senthil Kumaran, R.S., Sachin, M., and
Shetti, V.N., J. Catal., 2004, vol. 221, p. 455.
3. Avgouropoulos, G. and Ioannides, T., Appl. Catal., A,
2003, vol. 244, p. 155.
4. Wang, J.B., Lin, S.-C., and Huang, T.-J., Appl. Catal., A,
2002, vol. 232, p. 107.
5. Kim, D.H. and Cha, J.E., Catal. Lett., 2003, vol. 86,
nos. 1–3, p. 107.
6. Manzoli, M., Monte, R.D., Boccuzzi, F., Coluccia, S.,
and Kaspar, J., Appl. Catal., B, 2005, vol. 61, p. 192.
7. Avgouropoulos, G., Ionnides, T., Papadopolou, C.,
Batista, J., Hocevar, S., and Matralis, H.K., Catal. Today,
2002, vol. 75, p. 157.
ACKNOWLEDGMENTS
8. Snytnikov, P.V., Sobyanin, V.A., Belyaev, V.D., Tsyrulni-
kov, P.G., Shitova, N.B., and Shlyapin, D.A., Appl.
Catal., A, 2003, vol. 239, nos. 1–2, p. 149.
This work was supported in part by the Federal
Agency for Science and Innovations within the frame-
KINETICS AND CATALYSIS Vol. 48 No. 3 2007