High-surface area CuO-CeO2 catalysts prepared by a surfactant-templated method for low-temperature CO oxidation

Article abstract of DOI:10.1016/j.jcat.2006.11.021

High-surface area nanosized CuO-CeO₂ catalysts were prepared by a surfactant-templated method and tested for CO oxidation. The catalysts were characterized by XRD, TEM, N₂ sorption, H₂-TPR, and CO-TPR. The surfactant method can be used for preparing CuO-CeO₂ mixed oxides with a crystallite size of about 5 nm. The highest BET surface area of the catalysts was 215 m² g^-1, achieved over a 3.3% CuO content catalyst. XRD results indicated that the absence of a CuO phase with < 12 % CuO content may partially incorporate in the CeO₂ lattice to form Cu_x Ce_{1 - x} O_{2 - δ} solid solution, whereas a higher CuO content causes the formation of bulk CuO. These high-surface area nanosized catalysts were found to be very active for CO oxidation reaction; the lowest T₉₀ was 80 °C, achieved over a 12.0% CuO content catalyst. In addition, the CuO-CeO₂ catalysts also show high catalytic activity for selective oxidation of CO in excess H₂ at relatively low temperature. H₂-TPR results reveal three reduction peaks for these catalysts, which could be attributed to reduction of the highly dispersed CuO, the Cu²⁺ in the CeO₂ lattice, and the bulk CuO. Removal of the finely dispersed CuO in the catalyst by acid treatment resulted in a decline in catalytic activity for CO oxidation, indicating that the finely dispersed CuO species are the active sites for the reaction.

Full text of DOI:10.1016/j.jcat.2006.11.021

Journal of Catalysis 246 (2007) 52–59
www.elsevier.com/locate/jcat
High-surface area CuO–CeO catalysts prepared by a surfactant-templated
2
method for low-temperature CO oxidation
∗
Meng-Fei Luo , Jing-Meng Ma, Ji-Qing Lu, Yu-Peng Song, Yue-Juan Wang
Zhejiang Key Laboratory for Reactive Chemistry on Solid Surfaces, Institute of Physical Chemistry, Zhejiang Normal University, Jinhua 321004, China
Received 10 September 2006; revised 22 November 2006; accepted 22 November 2006
Available online 21 December 2006
Abstract
High-surface area nanosized CuO–CeO catalysts were prepared by a surfactant-templated method and tested for CO oxidation. The catalysts
2
were characterized by XRD, TEM, N sorption, H -TPR, and CO-TPR. The surfactant method can be used for preparing CuO–CeO mixed
2
2
2
2
−1
oxides with a crystallite size of about 5 nm. The highest BET surface area of the catalysts was 215 m g , achieved over a 3.3% CuO content
catalyst. XRD results indicated that the absence of a CuO phase with <12% CuO content may partially incorporate in the CeO lattice to form
2
CuxCe
O
solid solution, whereas a higher CuO content causes the formation of bulk CuO. These high-surface area nanosized catalysts
1
−x 2−δ
◦
were found to be very active for CO oxidation reaction; the lowest T was 80 C, achieved over a 12.0% CuO content catalyst. In addition, the
90
CuO–CeO catalysts also show high catalytic activity for selective oxidation of CO in excess H at relatively low temperature. H -TPR results
2
2
2
2+
reveal three reduction peaks for these catalysts, which could be attributed to reduction of the highly dispersed CuO, the Cu in the CeO lattice,
2
and the bulk CuO. Removal of the finely dispersed CuO in the catalyst by acid treatment resulted in a decline in catalytic activity for CO oxidation,
indicating that the finely dispersed CuO species are the active sites for the reaction.
©
2006 Elsevier Inc. All rights reserved.
Keywords: CO oxidation; CuO–CeO ; Surface area; Catalytic activity
2
1
. Introduction
Although precious metal catalysts have high activity and de-
sirable air and temperature stability, their high cost may limit
their application. Recent reports have shown that the activity
of Ce-related oxides in total oxidation reactions is greatly en-
hanced not only by noble metals, but also by transitional metals
Cerium dioxide powders have potential applications in pol-
ishing powders, coatings for high-temperature optical and ce-
ramic materials, and gas sensors [1]. Because of its low tem-
perature reducibility, oxygen storage, and release properties,
ceria is also widely used as a support for many oxidation cata-
lysts. These catalysts are active in oxidation of CO [2], water–
like copper, and the CeO2 -supported CuO catalyst is compa-
−x
rable or superior to platinum catalysts for the preferential oxida-
tion of CO in excess hydrogen [7–9]. The promoting effect had
been correlated with the synergism of the redox properties of
the system, which is achieved by the formation of copper–ceria
interactions, with both components being significantly more
readily reduced or oxidized than the corresponding independent
components [10].
gas shift reactions [3], elimination of CO and NO contam-
x
inants from automotive exhaust gases, combustion of hydro-
carbons [4], and other reactions. One of the most important
applications of ceria is its use as an additive in the three-way
catalysts for automotive exhaust gas treatment. The interaction
of ceria with precious metals (Pd, Pt, Rh) and its effect on cat-
alytic activity are well documented [5,6].
Varying the preparation routes often leads to changes in the
morphology and dispersion of copper species. Methods in cur-
rent use include co-precipitation [11], deposition–precipitation,
impregnation [12], sol–gel, and urea methods [13]. We have
previously reported [14] the synthesis of CuO/CeO by an
2
*
impregnation method. The catalysts exhibit high catalytic ac-
tivities in CO oxidation. The activity is close to that of the
Corresponding author. Fax: +86 579 2282595.
E-mail address: mengfeiluo@zjnu.cn (M.-F. Luo).
0021-9517/$ – see front matter © 2006 Elsevier Inc. All rights reserved.
doi:10.1016/j.jcat.2006.11.021

M.-F. Luo et al. / Journal of Catalysis 246 (2007) 52–59
53
◦
Au/MnOx catalyst and clearly superior to that of the Pt/SnOx
catalyst reported in the literature [15]. Increasing the surface
area and reducing the particle size to nanoscale will provide
numerous, more reactive sites [16] so as to enhance the cat-
alyst activity. Several routes for synthesizing nanosized ceria
and cerium-based mixed oxides have been investigated. Hi-
rano and Inagaki [17] obtained fine CeO2 from Ce(NO3)3,
sample heated in N2 at 800 C was designated as N8. The N8
◦
sample followed by calcinations in air at 400 C was designated
as N8A4.
2
.2. Characterizations
X-ray diffraction (XRD) patterns were collected on a
PHILIPS PW 3040/60 powder diffractometer using CuKα ra-
◦
Ce(SO4)2, and Ce(NH4)4(SO4)4 solutions at 180 C under au-
togenous pressure. Tok et al. [18] reported the synthesis of
Gd2O3-doped CeO2 solid solution nanoparticles via carbonate
co-precipitation. Skårman et al. [19] obtained nanostructured
CuOx/CeO2 composites by inert gas condensation (IGC). Shen
et al. [20] reported the synthesis of CuO-loaded mesoporous
CeO2 using the ordered mesoporous silica KIT-6 as a hard tem-
plate. The surface area of the catalyst with 20% CuO loading
diation. The working voltage was 40 kV, and the current was
◦
4
0 mA. The intensity data were collected at 25 C in a 2θ range
◦
◦
◦
−1
of 20 –130 with a scan rate of 1.2 min . The microstruc-
tural parameters of samples were determined by the Rietveld
method [22] using MAUD software (material analysis using
◦
diffraction) [23]. CeO2 annealed at 1450 C for 10 h in air was
used as a nonintrinsic broadening sample to measure the instru-
ment function and to extract the microstructure values of the
catalyst.
2
−1
◦
was 88 m g , and a T50 of 116 C was obtained for the CO
oxidation reaction.
In this paper, a series of high-surface area nanosized CuO–
CeO2 catalysts were prepared by a surfactant-templated meth-
od, and their catalytic performance was tested for the CO ox-
idation reaction. The active phase for CO oxidation was also
investigated.
N2 sorption isotherms were obtained at 77 K on an Auto-
sorb-1 apparatus. BET and BJH analysis were used to deter-
mine the surface area and pore size distribution of the sam-
ples. Transmission electron microscopy (TEM) observations
were made using a JEM2010 microscope operated at 200 kV.
X-ray photon spectroscopy (XPS) experiments were carried out
on a RBD upgraded PHI-5000C ESCA system (Perkin-Elmer)
with MgKα radiation (hν = 1253.6 eV). The pass energy was
fixed at 46.95 eV to ensure sufficient sensitivity. Binding ener-
gies were calibrated by using the containment carbon (C1s =
2
. Experimental
2
.1. Catalyst preparation
The CuO–CeO catalysts with different CuO content (3.3,
2
2
84.6 eV). The data analysis was performed using the RBD
6
.1, 12.0, 28.3, 50.1 mol%) were prepared using a surfactant-
AugerScan 3.21 software (RBD Enterprises).
templated method. For a typical synthesis, cetyltrimethyl am-
monium bromide (CTAB; 6 mmol) was dissolved in 200 mL
of distilled water, followed by the addition of Ce(NO3)3·6H2O
The reduction properties of CuO–CeO2 catalysts were mea-
sured by temperature-programmed reduction (TPR) techniques,
including H2-TPR and CO-TPR. A 10-mg sample was placed
in a quartz reactor connected to a homemade TPR apparatus,
(
9.5 mmol) and Cu(NO3)2·3H2O (5 mmol). The mixture was
stirred for 0.5 h, after which a solution of sodium hydroxide
◦
and the reactor was heated from 100 to 900 C at a rate of
2
(
50 mmol) in 300 mL of distilled water was added and stirred
◦
−1
0 C min . The amount of H2 uptake during the reduction
◦
for 12 h. The final precipitate was aged at 90 C for 3 h, washed
with hot water, dried at 110 C for 6 h, and finally calcined at
4
was measured using a thermal conductivity detector (TCD),
and the amount of CO2 desorbed during the reduction was
monitored by a Balzers Omnistar200 mass spectrometer at
m/e = 44.
◦
◦
00 C for 4 h. Other catalysts with different CuO contents were
prepared in a similar manner. The actual CuO content in the
catalyst was analyzed by atomic absorption spectrophotometry
(
AAS); for example, a CeCu3 catalyst means a CuO–CeO cat-
2
.3. Catalytic activity measurements
alyst with CuO content of 3.3 mol%. It was found the actual
CuO content (3.3 mol%) was lower than the CuO content in the
starting solution (5 mol%).
The catalytic activity of CO oxidation was evaluated in
For comparison, CuO/SiO2 and CuO/γ -Al2O3 catalysts
were prepared by an incipient wetness method. A SiO2 (BET
a fixed-bed reactor (6 mm i.d.) using 500 mg of catalyst (20–
−1
40 mesh). The total flow rate of the feed gas was 80 mL min
corresponding to a space velocity (SV) of 9600 mL g h
,
.
2
−1
2
−1
−1 −1
area, 730 m g ) or γ -Al2O3 (BET area, 212 m g ) support
was immerged with a proper amount of Cu(NO3)2·3H2O so-
The catalysts were directly exposed to reaction gas as the re-
actor temperature was stabilized at the reaction temperature
without any pretreatment. For CO oxidation, a gas mixture con-
sisted of 1% CO and 1% O2 in N2 with a total flow rate of
◦
lution for 4 h. The slurry was heated at 80 C to remove the
◦
water and dried at 100 C for 7 h, followed by calcination at
00 C for 4 h in air. The CuO content in these catalysts was
2 mol%. Pure CuO and CeO2 were prepared by decomposi-
tion of Cu(NO3)2·3H2O and Ce(NO3)3·6H2O at 400 C for 4 h
in air. In addition, a series of Ce–Cu–O catalysts with 10 mol%
CuO content were prepared according to the typical method re-
ported in the literature [21]. The sample directly calcined in air
◦
4
1
−
1
80 mL(NTP) min . The reaction temperature was monitored
by a thermocouple placed in the middle of the catalyst bed. The
CO concentration in the reactor effluent was analyzed using an
Agilent 6850 gas chromatograph equipped with a TCD detector
attached to an HP PLOT column (30 m × 0.32 mm × 12 µm).
For comparison, the reaction was carried out using 50 mg of
◦
◦
at 400 or 800 C was designated as A4 or A8, respectively. The

54
M.-F. Luo et al. / Journal of Catalysis 246 (2007) 52–59
catalyst and a feed gas total flow rate of 100 mL min 1, corre-
−
to 50.1 mol% causes an apparent increase in the intensities of
the peaks.
−
1
−1
sponding to an SV of 120,000 mL g
h .
Catalytic tests for the selective oxidation of CO in excess hy-
drogen were carried out in a fixed-bed reactor (5 mm i.d.) at at-
mospheric pressure. A 50-mg sample of catalyst (60–80 mesh)
was diluted with same amount of inert silicon dioxide. The gas
The results of crystallite size and cell parameter measure-
ments are summarized in Table 1. The table shows that cell pa-
rameter decreases with increasing CuO content up to 6.1 mol%,
but remains almost constant between 12.0 and 50.1 mol%.
A difference in the crystallite size of the directions (d111, d200,
d220) indicates the presence of anisotropy in the phase. In all
samples, the crystallite size is <5 nm.
consisted of 1% CO, 1% O , and 50% H in Ar, with a total
2
2
−
1
flow rate of 100 mL min . The reactor inlet and outlet streams
were measured by an on-line gas chromatograph equipped with
a TCD and a flame ionization detector (FID). H , O , CO, and
Fig. 2 shows the N2 sorption isotherms and the pore size
distributions of the CuO–CeO2 catalysts with different CuO
contents. The N2 sorption isotherms of all samples (Fig. 2) are
of type IV, typical of a mesoporous sample. The BET surface
area and pore size of these samples are listed in Table 1. From
Table 1 and Fig. 2, it can be seen that with increasing CuO
content from 3.3 to 50.1 mol%, the BET surface areas of CuO–
2
2
CO2 were separated using a carbon molecular sieve (TDX-01)
column. CO and CO2 were converted to methane by a metha-
nation reactor and analyzed using an FID. The carbon balance
was close to 100 ± 3%.
3
. Results
2
−1
CeO2 catalysts decrease from 215 to 91 m g , whereas the
pore sizes increase from 6.4 to 17.8 nm.
3
.1. Structural characterization
Fig. 3 shows the TEM images of the sample CeCu6 and
CeCu12. The images do not clearly show the presence of or-
dered mesopores, but reveal a rather foamlike structure result-
ing from closely aggregated metal oxide nanoparticles [24]. The
size of the nanoparticles is around 5 nm, consistent with the
crystallite size obtained by XRD.
Fig. 1 shows the XRD patterns of the CuO–CeO catalysts
with different CuO contents. The distinct fluorite oxide diffrac-
2
tion pattern of CeO is seen in all of the samples. However,
2
XRD peaks due to CuO are not detected in the samples with
low CuO content. At a CuO content of 28.3 mol%, weak dif-
fraction peaks of CuO are seen; further increasing CuO content
3
.2. Catalytic testing
Fig. 4 shows catalytic activities of CO oxidation over the
CuO–CeO catalysts at different SVs. It shows that the activ-
2
Table 1
Cell parameter, crystallite size, BET surface area and pore size of CeCux cata-
lysts
Sample Cell parameter Crystallite size (nm)
nm)
S (BET) Pore size
A
(
2
−1
(nm)
d(111) d(200) d(220) (m g
)
CeCu0
CeCu3
CeCu6
CeCu12 0.5416(2)
CeCu28 0.5416(2)
CeCu50 0.5415(2)
0.5418(1)
0.5416(1)
0.5415(1)
4.58
4.89
3.40
3.21
3.88
5.10
4.41
4.89
3.54
3.41
3.71
4.56
4.54
4.89
3.49
3.27
3.24
4.27
202
215
180
150
116
91
6.6
6.4
7.1
9.6
11.2
17.8
Fig. 1. XRD patterns of CuO–CeO catalysts with different CuO contents.
2
Fig. 2. N sorption isotherms (a) and pore size distributions (b) of CuO–CeO catalysts with different CuO contents.
2
2

M.-F. Luo et al. / Journal of Catalysis 246 (2007) 52–59
55
Fig. 3. TEM images of (a) CeCu6 and (b) CeCu12 catalysts.
Fig. 5. CO oxidation over various CuO containing catalysts.
Fig. 4. CO oxidation over CeCux catalysts. (a) Space velocity = 9600 mL
−
1
−1
−1 −1
h .
g
h
; (b) space velocity = 120,000 mLg
ity is significantly enhanced with increased CuO content in
the catalyst. With increasing CuO content, T90 (the temper-
ature for 90% conversion of CO) of the samples decreases
◦
sharply. The lowest T90 is 80 C, achieved at the 12.0 mol%
CuO content; T90 remains constant with increasing CuO con-
tent up to 50.1 mol%. For the same catalyst, it is seen that CO
Fig. 6. CO oxidation in excess of H over CeCu12 catalyst.
2
−
1
−1
h
conversion is lower at high SV (120,000 mL g
) than at
−1
h ; this is because an increase in SV decreases
−1
9
600 mL g
results are presented in Fig. 6. The CeCu12 catalyst is inactive
for H2 oxidation at reaction temperatures up to 145 C. There-
the contact time with the catalyst.
◦
For comparison, Fig. 5 shows the catalytic performance of
the CuO/SiO2 and CuO/γ -Al2O3 catalyst, as well as the pure
CuO and CeO2, on CO oxidation. It is seen that these catalysts
are less active than the CuO–CeO2 catalysts.
The catalytic performance of the CeCu12 catalyst for the se-
lective oxidation of CO in excess H2 was also investigated; the
fore, 100% O2 selectivity can be obtained over the CeCu12 cat-
◦
alyst at reaction temperatures below 145 C, due to the lack of
competitive oxidation of H2. However, with a further increase
in reaction temperature, H2 oxidation reaction emerges, and the
selectivity to O2 decreases.

56
M.-F. Luo et al. / Journal of Catalysis 246 (2007) 52–59
The N2 sorption experiments show that all of the samples are
porous. However, the lack of low-angle XRD peaks indicates
they do not have regularly ordered mesoporous pores. This is
also consistent with the TEM observations (Fig. 3). In addition,
with increasing CuO content of the catalysts, formation of the
bulk CuO results in decreased BET surface area and, conse-
quently, increased pore size.
The CuO–CeO2 catalysts reported in this work have slightly
higher catalytic activities than other reported Cu-modified or
-
doped cerium oxide materials for the oxidation of CO to
CO2 at relatively low temperatures [19,20,25]. Tang et al. [25]
reported the synthesis of a CuO–CeO2 (10 wt% Cu) cata-
lyst by a co-precipitation method. The T90 for the CO oxida-
◦
tion reaction of this catalyst was about 90 C with an SV of
Fig. 7. H -TPR profiles of CuO–CeO catalysts with different CuO contents.
2
2
−1 −1
1
0,000 ml g
h
. In the present work, the T90 of the CO reac-
◦
tion over the CeCu12 catalyst is about 80 C with almost the
same SV (Fig. 4a). The decline in T90 of the catalysts with
different CuO contents (Fig. 4a) implies that the CuO species
should be responsible for the activity. In the H2-TPR results
for the catalysts, all profiles show peak α and peak β. We pre-
viously reported two reduction peaks of H2-TPR [14], with
the peak α ascribed to finely dispersed CuO and the peak β
assigned to the larger particles of the bulk CuO. The same
results also have been reported by Kundakovic and Flytzani-
Stephanopoulos [31]. However, in H2-TPR of the present work
3
.3. TPR
TPR has been extensively used to characterize the reducibil-
ity of CuO–CeO2 catalysts. H2-TPR profiles of the CuO–CeO2
catalysts and pure CeO2 are shown in Fig. 7. The pure CeO2 has
◦
two reduction peaks at about 450 and 900 C, ascribed to the re-
duction of surface and bulk oxygen of CeO2, respectively [25].
The profiles of all of the CuO–CeO2 samples show two reduc-
◦
tion peaks (α and β) in the range of 150–280 C. As is well
known, the reduction profile of pure CuO is characterized by
(
Fig. 7), the peaks α and β coexist in all TPR profiles even
at 1.5 mol% CuO content, and the area of peak β always larger
than that of peak α. In the meantime, the XRD results show that
part of the Cu ions replaces Ce to form a CuxCe1−xO2−δ
solid solution at low CuO content. Therefore, it is more likely
that peak α is due to the reduction of finely dispersed CuO and
peak β is due to the reduction of Cu ions in CuxCe1−xO2−δ
◦
a single peak at about 380 C [14]. The H2-TPR peak temper-
ature of CuO–CeO2 catalysts is much lower than that of pure
CuO. With increasing CuO content from 3.3 to 12.0 mol%, the
intensities of the peak α and β increase, and the peak positions
shift to lower temperatures. With increasing CuO content up
to 28.3 mol%, the peaks α and β coalesce and overlap, and
2+
4+
2+
◦
a new peak, called γ , appears at 280 C. With a further in-
solid solution. In addition, with further increases in CuO con-
tent up to 28.3 mol%, the bulk CuO becomes present in the
sample (Fig. 1), and a new peak γ is found by H2-TPR, sug-
gesting that this peak γ is due to the reduction of bulk CuO.
It is commonly believed that the finely dispersed CuO is
the active phase for CO oxidation [14]. Therefore, the ac-
tivity should decrease if the active phase is removed. The
finely dispersed CuO particles can be easily dissolved in nitric
acid [14]. In this work, the CeCuO12 catalyst was immersed
crease in CuO content to 50.1 mol%, the intensity of peak γ
increases and the position shifts to higher temperature, but no
visible changes occur in peaks α and β.
4
. Discussion
In this work, a series of CuO–CeO catalysts with CuO
2
content of 3.3–50.1 mol% were prepared using a surfactant-
2
−1
templated method. The surface area was high, 91–215 m g
Table 1), higher than that for other reported Ce–Cu–O ma-
in concentrated nitric acid (15 ml of 50% HNO /g catalyst)
for 2 h, filtered, then washed with plenty of distilled water to
3
(
terials [19,20,25]. XRD (Fig. 1) showed no CuO diffraction
peaks at lower CuO contents. This may be attributed to fine
remove the residual HNO or other impurities, and dried at
120 C. This catalyst was designated as CeCu12-H. This cat-
3
◦
dispersion of CuO on the surface of CeO [26], solid solution
alyst’s CuO content was 8.1 mol% by AAS analysis. After
the acid treatment, its BET surface area increased from 150
2
formation [27,28], or a combination of these two phenomena.
The presence of CuO diffraction with high CuO content should
be ascribed to bulk CuO. The cell parameter decreases with in-
creasing CuO content up to 6.1 mol%, because the ionic radius
2
−1
to 172 m g , suggesting that removing the finely dispersed
CuO species produced an increase in surface area. However,
the cell parameter and crystallite size were 0.5416(2) nm and
3.16 nm, respectively, with no visible changes seen compared
with the untreated sample (Table 1). CO oxidation activities of
the CeCuO12 catalyst and the CeCu12-H catalyst are shown in
Fig. 8. The catalytic activity evidently drops after the acid treat-
ment.
2
+
4+
of Cu (0.072 nm) is smaller than that of Ce (0.097 nm).
When part of the CuO incorporates into the CeO2 lattice and
2+
4+
Cu replaces Ce , a reduction in the cell parameter of ceria
occurs. Therefore, changes in the cell parameter of the catalysts
(Table 1) indicate that the missing CuO phase in XRD at low
CuO content may have partly formed a Cu Ce
solution [29,30].
O
solid
H -TPR profiles of the CeCuO12 catalyst and CeCu12-H
catalyst are shown in Fig. 9a. The figure shows that the peak α
x
1−x 2−δ
2

M.-F. Luo et al. / Journal of Catalysis 246 (2007) 52–59
57
Table 2
Comparison of Cu content before and after nitric acid treatment
Sample
Cu/(Cu + Ce) (mol%)
Nominal
XPS
CeCu12
CeCu12-H
12.0
8.1
23.7
8.7
generated by dissociative adsorption on the metallic Cu surface
can spill over onto the surface of the lattice Cu2 , thereby de-
creasing its reduction temperature. After the finely dispersed
CuO is removed by acid treatment, the reduction temperature
of peak β evidently increases. However, in CO-TPR, the car-
bon monoxide spillover effect does not exist; thus, the position
of peak β remains constant before and after acid treatment.
+
Fig. 8. CO oxidation over CeCu12 and CeCu12-H at a space velocity of
◦
−
1
−1
.
A weak peak is present at about 300 C in Fig. 9b. It is evi-
120,000 mL g
h
dent that the temperatures of the reduction peaks of CO-TPR
are lower than those of H2-TPR. Therefore, we believe that the
◦
reduction peak of CO-TPR seen at about 300 C is probably
due to reduction of some of the CeO2 surface oxygen.
To further clarify the surface changes caused by the acid
treatment, XPS analysis was conducted. Table 2 summarizes
the surface concentrations of Cu before and after the acid
treatment. Clearly, that for the untreated sample, the surface
Cu concentration (23.7%) is much higher than the nominal
one (12.0%), due to the aggregation of the CuO species on the
surface. In contrast, after acid treatment, the surface Cu con-
centration (8.7%) is close to the nominal one (8.1%), indicating
complete removal of the surface CuO species. These results are
in good agreement with the H -TPR and the CO-TPR results,
2
strongly suggesting that the surface CuO species are responsi-
ble for CO oxidation, and that the decreased activity is due to
the removal of surface CuO species.
High catalytic activity of the CeCu12 catalyst for selective
oxidation of CO in excess H2 at lower temperature is under-
standable. Comparing Figs. 9a and 9b clearly shows that the
◦
reduction temperature (120 C) of the finely dispersed CuO is
◦
lower in CO than in H2 (200 C), indicating that the reaction of
CO is easier than that of H2. These findings may relate to the
CO selectivity of the CuO–CeO2 catalyst in excess H2. At the
same SV, the activity for CO oxidation in excess H2 (Fig. 6) is
slightly lower than that in CO atmosphere alone (Fig. 4b). This
Fig. 9. H -TPR profiles (a) and CO-TPR profiles (b) of CeCu12 and CeCu12-H
2
samples.
may be due to the competitive adsorption of H and CO on the
2
active sites.
of CeCu12 disappears after the acid treatment, indicating that
the finely dispersed CuO species are the active sites for CO
In addition, to prove the influence of the surface area on
the catalytic activity, the CO oxidation activities of Cu–Ce–O
catalysts with similar CuO contents but different surface ar-
eas were tested. The results are shown in Fig. 10, and the BET
surface areas, calculated average crystallite sizes, and cell pa-
rameters are given in Table 3. The CuO content is 10 mol% in
the N8A4, A4, and A8 samples. Fig. 4 shows that the CuCe6
and the CuCe12 samples have similar catalytic activity; thus,
if a CuCe10 sample does exist, then its catalytic activity should
be close to that of CuCe12. It can be seen that the larger the sur-
face area, the greater the catalytic activity. The CuCe12 catalyst
oxidation, and also shows that peak β is represented by Cu2
+
incorporating into the CeO2 lattice. However, in Fig. 9a, the
position of the peak β shifts to higher temperature after acid
treatment. To explain this shift in peak β, CO-TPR was car-
ried out on the CeCu12-H and CeCu12 samples; the profiles
are also presented in Fig. 9b. After acid treatment, the peak α of
CeCu12 disappears, but the peak β remains consistent, in con-
trast to the H2-TPR findings. The difference between H2-TPR
and CO-TPR can be explained by the following mechanism.
In H2-TPR, due to hydrogen spillover [32], when the finely dis-
persed CuO of CeCu12 sample is reduced, the atomic hydrogen
2
−1
has the largest surface area (150 m g ) and greatest catalytic
activity. This is because increasing the surface area of the cat-

58
M.-F. Luo et al. / Journal of Catalysis 246 (2007) 52–59
CeO2 catalysts. Higher CO oxidation activities of the catalysts
were obtained compared with samples with low surface areas.
H2-TPR suggests the presence of three different CuO species:
highly dispersed CuO, Cu2 in the CeO2 lattice, and bulk CuO
species. The activity of the catalyst for CO oxidation obviously
declined after the removal of highly dispersed CuO by acid
treatment, indicating that the highly dispersed CuO species are
the active phase for CO oxidation.
+
Acknowledgments
This work was supported by the Natural Science Foundation
of China (grant 20473075) and the Zhejiang Provincial Na-
ture Science Foundation of China (grant Z404383). The authors
thank Professor Guang-Lie Lv (Zhejiang University, China) for
calculating the microstructural parameters and Dr. Zhi-Gang
Liu (Zhejiang University, China) for carrying out the selective
oxidation of CO in excess H2.
Fig. 10. CO oxidation over Cu–Ce–O catalysts with similar CuO content but
different surface areas.
Table 3
Cell parameter, crystallite size and BET area of different Ce–Cu–O catalysts
References
Sample
Cell parameter
nm)
Crystallite size (nm)
S (BET)
A
(
2
−1
d(111)
d(200)
d(220)
(m g )
[1] N. Izu, W. Shin, N. Murayama, S. Kanzaki, Sens. Actuators B 87 (2002)
CeCu12
N8A4
A4
0.5416(2)
0.5418(2)
0.5419(1)
0.5418(1)
3.21
6.95
8.08
6.36
3.41
6.87
8.10
6.36
3.27
6.89
8.09
6.36
150
131
60
95.
[2] A. Martínez-Arias, M. Fernández-García, O. Gálvez, J.M. Coronado, J.A.
Anderson, J.C. Conesa, J. Soria, G. Munuera, J. Catal. 195 (2000) 207.
[3] X. Wang, J.A. Rodriguez, J.C. Hanson, D. Gamarra, A. Martínez-Arias,
M. Fernández-García, J. Phys. Chem. B 110 (2006) 428.
A8
10
[4] P. Bera, K.C. Patil, V. Jayaram, G.N. Subbanna, M.S. Hegde, J. Catal. 196
(2000) 293.
[5] F. Mariño, C. Descorme, D. Duprez, Appl. Catal. B 54 (2004) 59.
[6] C.N. Costa, S.Y. Christou, G. Georgiou, A.M. Efstathiou, J. Catal. 219
(
2003) 259.
7] P.G. Harrison, I.K. Ball, W. Azalee, W. Daniell, D. Goldfarb, Chem. Mater.
2 (2000) 3715.
8] S. Ho cˇ evar, U.O. Krasovec, B. Orel, A.S. Arico, H. Kim, Appl. Catal. B 28
2000) 113.
9] P.-O. Larsson, A. Andersson, Appl. Catal. B 24 (2000) 175.
[
[
[
1
(
[
10] A. Martínez-Arias, M. Fernández-García, J. Soria, J.C. Conesa, J. Catal.
82 (1999) 367.
1
[11] C.R. Jung, J. Han, S.W. Nam, T.-H. Lim, S.-A. Hong, H.-I. Lee, Catal.
Today 93–95 (2004) 183.
[
[
12] P. Bera, S.T. Aruna, K.C. Patil, M.S. Hegde, J. Catal. 186 (1999) 36.
13] M. Jobbágy, M. Fernando, B. Schönbrod, G. Baronetti, M. Laborde,
Chem. Mater. 18 (2006) 1945.
[
[
[
14] M.F. Luo, Y.J. Zhong, X.X. Yuan, X.M. Zheng, Appl. Catal. A 162 (1997)
Fig. 11. H -TPR profiles of CuCe12, N8A4, A4 and A8 catalysts.
2
121.
15] S.D. Gardner, G.B. Hoflund, B.T. Upchurch, D.R. Schryer, E.J. Kielin,
J. Schryer, J. Catal. 129 (1991) 114.
16] L. Yin, Y. Wang, G. Pang, Y. Koltypin, A. Gedanken, J. Colloid Interface
Sci. 246 (2002) 78.
alyst aids the formation of the finely dispersed CuO. Fig. 11
shows H2-TPR profiles of the CuCe12 and N8A4, A4, and A8
catalysts. These profiles demonstrate that the peak α shifts to-
ward lower temperature with increasing surface area. This may
be due to the CuO dispersion state becoming finer and the CuO
crystallite becoming smaller [30] with increasing surface area
of the catalyst. Therefore, there are more exposed active sites on
the high-surface area catalysts than on the low-surface area cat-
alysts, which could explain the difference in the CO oxidation
activity.
[
[
17] M. Hirano, M. Inagaki, J. Mater. Chem. 10 (2000) 473.
18] A.I.Y. Tok, L.H. Luo, F.Y.C. Boey, Mater. Sci. Eng. A 383 (2004) 229.
[19] B. Skårman, D. Grandjean, R.E. Benfield, A. Hinz, A. Andersson, L.R.
Wallenberg, J. Catal. 211 (2002) 119.
[20] W. Shen, X. Dong, Y. Zhu, H. Chen, J. Shi, Microporous Mesoporous
Mater. 85 (2005) 157.
[
21] G.Q. Xie, M.F. Luo, M. He, P. Fang, J.M. Ma, Y.F. Ying, Z.L. Yan,
J. Nanopart. Res. (2006) 3, online.
[22] R.A. Young, The Rietveld Method, Oxford Univ. Press, Oxford, UK, 1993.
[
[
[
23] L. Lutterotti, S. Gialanella, Acta Mater. 46 (1998) 101.
24] A.K. Sinha, K. Suzuki, J. Phys. Chem. B 109 (2005) 1708.
25] X. Tang, B. Zhang, Y. Li, Y. Xu, Q. Xin, W. Shen, Catal. Today 93–95
5
. Conclusion
(
2004) 191.
The surfactant-templated method presented in this work
[
26] S.-M. Zhang, W.-P. Huang, X.-H. Qiu, B.-Q. Li, X.-C. Zheng, S.-H. Wu,
has allowed us to prepare high-surface area nanosized CuO–
Catal. Lett. 80 (2002) 41.

M.-F. Luo et al. / Journal of Catalysis 246 (2007) 52–59
59
[
[
[
27] C. Lamonier, A. Ponchel, A. D’Huysser, L. Jalowiecki-Duhamel, Catal.
[30] X. Jiang, G. Lu, R. Zhou, J. Mao, Y. Chen, X. Zheng, Appl. Surf. Sci. 173
Today 50 (1999) 247.
28] P. Bera, K.R. Priolkar, P.R. Sarode, M.S. Hegde, S. Emura, R. Kumashiro,
N.P. Lalla, Chem. Mater. 14 (2002) 3591.
(2001) 208.
[31] L. Kundakovic, M. Flytzani-Stephanopoulos, Appl. Catal. A 171 (1998)
13.
29] Y. Li, Q. Fu, M. Flytzani-Stephanopoulos, Appl. Catal. B 27 (2000) 179.
[32] R. Kramer, M. Andre, J. Catal. 58 (1979) 287.