Size limit of support particles in an oxide-supported metal catalyst: Nanocomposite Ni/ZrO2 for utilization of natural gas

Article abstract of DOI:10.1021/jp030127l

Oxide-supported metal catalysts are a very important class of industrial catalysts that are closely related with many key technologies in the chemical/petrochemical industries, in environmental protection, in chemical sensors, etc. The size effect of zirconia particles on the catalysis of Ni metal for CO₂ re-forming of methane was studied. When the sizes of zirconia particles becoming < 25 nm, the oxide formed nanocomposite catalysts with size-comparable Ni-metal nanocrystals (10-15 nm). The nanocomposite catalysts showed extremely stable catalysis, which was in strong contrast with the deactivating Ni catalyst supported on bigger zirconia particles (> 25 nm). A lower limit in the nanometer range might exist for the size of oxide support particles in supported metal catalysts.

Full text of DOI:10.1021/jp030127l

J. Phys. Chem. B 2003, 107, 5203-5207
5203
Size Limit of Support Particles in an Oxide-Supported Metal Catalyst: Nanocomposite
Ni/ZrO2 for Utilization of Natural Gas
†
Bo-Qing Xu,* Jun-Mei Wei, Ying-Tao Yu, Yan Li, Jin-Lu Li, and Qi-Ming Zhu
State Key Lab of C1 Chemistry & Technology, Department of Chemistry, Tsinghua UniVersity,
Beijing 100084, China
ReceiVed: January 29, 2003
2
This paper explores the size effect of zirconia particles on the catalysis of Ni metal for CO re-forming of
methane. It is revealed for the first time that when the sizes of zirconia particles become samller than 25 nm,
the oxide forms nanocomposite catalysts with size-comparable Ni-metal nanocrystals (10-15 nm). The
nanocomposite catalysts show extremely stable catalysis, which is in strong contrast with the deactivating Ni
catalyst supported on bigger zirconia particles (>25 nm). This raises the possibility of tailoring the catalytic
behavior of oxide-supported metal catalysts by reducing the particle size of oxide to make high-performance
nanocomposite catalysts.
1
. Introduction
to methane, also a large fraction of CO2. Moreover, the use of
CO2 provides a clean and cheaper oxygen source for the reac-
Oxide-supported metal catalysts are a very important class
tion, which eliminates the costly separation processes required
for partial oxidation of methane. This reaction is also considered
of industrial catalysts that are closely related with many key
technologies in chemical/petrochemical industries, in environ-
mental protection, in chemical sensors, and in the manufacture
6-8
for energy transformation or chemical heat-pipe applications.
Ni-based catalysts possess high initial activity for the reaction
but suffer from a rapid deactivation due to carbon deposition,
1
of fine and special chemicals. Often in this kind of catalyst
system, discrete metal nanocrystals (typically 1-15 nm) are
dispersed on support particles that are one to several orders of
magnitude larger than the supported metal nanoparticles. The
most important function of a support is to provide proper texture/
pore structure and high surface area to disperse and maintain
the active metal catalyst. Structure-sensitive or -insensitive
concepts are developed to describe the observations that catalysis
of metal catalyst is dependent or independent of the size of the
4,5
which hinders the process from industrial applications. In
9-11
earlier work, we demonstrated
that the use of nano-ZrO2
particles (15-20 nm) as “support” for nickel catalyst led to
extremely stable Ni/ZrO2 catalysts that are highly active for the
CO2/CH4 reaction. We have discovered that a change of the
conventional hydrogels of metal hydroxides to their alcogels
followed by nanocrystal formation under controlled atmospheres
at elevated temperature can lead to size control synthesis of
2
3
metallic nanocrystals. When the particle sizes of an oxide
12
13,14
nanocrystals of zirconia and other oxides.
makes it possible to address the particle size effect of ZrO2
support” on the performance of Ni catalyst by reducing the
This discovery
support are reduced to such an extent that they become
comparable to that of the active metal particles, the oxide may
deviate dramatically from its function as a conventional catalyst
support. Such metal/oxide catalyst with size-comparable metal
and oxide nanocrystals may be better called a metal/oxide
“
particle size of ZrO2 down to the extent that it becomes
indistinguishable from particles of the metal catalyst.
4
nanocomposite rather than an “oxide-supported” metal catalyst.
2
. Experimental Section
.1. Preparation and Catalytic Testing of Ni/ZrO2 Cata-
But it is unknown whether there is a lower limit to the particle
size of an oxide support for the so-called oxide-supported metal
catalysts. Here, we investigate the size effect of zirconia
nanocrystals on the catalysis of Ni metal for CO2 re-forming
of methane. We find that nanocomposites of size comparable
to Ni metal (10-15 nm) and zirconia (7-25 nm) show
extremely stable catalysis whereas Ni catalyst supported on
bigger zirconia particles (>25 nm) deactivates very rapidly. This
raises the possibility of tailoring the catalytic behavior of oxide-
supported metal catalysts by reducing the particle size of oxide
to make metal/oxide nanocomposite catalysts.
2
4
,13
lyst. Following our previous sample coding, ZrO2-AS is the
sample prepared by drying under supercritical ethanol (270 °C,
7
.0 MPa) of an ZrO(OH)2 alcogel, which was obtained by
washing a conventional ZrO(OH)2 hydrogel with anhydrous
ethanol. A processing of the alcogel at 270 °C in flowing
nitrogen under normal pressure gave sample ZrO2-AN. And,
the digesting of the same alcogel in anhydrous ethanol under
reflux (78-80 °C) and then processing at 270 °C in flowing
nitrogen produced the ZrO2-AD sample. The later physical
characterizations (XRD, TEM, BET, etc.) of these alcogel-
derived ZrO2 made use of the samples further calcined at 650
CO2 re-forming of methane over heterogeneous catalysts to
produce a syngas with equal moles of CO and H2 is one of the
attractive routes for the utilization of the methane and CO2
°
C for 5 h in flowing air. The ZrO2-CP and -CN samples were
4
-8
prepared respectively by conventional calcination at 650 °C of
the ZrO(OH)2 hydrogel in flowing air and nitrogen.
resources.
Some natural gas reservoirs contain, in addition
Ni/ZrO2 catalysts were obtained by impregnation of aqueous
Ni(NO3) 2 onto the oxide precursors dried at 270 °C with a rotary
evaporator. Details for the catalyst preparation were described
*
Corresponding author. E-mail: bqxu@mail.tsinghua.edu.cn.
Present address: Department of Chemical Engineering, University of
†
California at Berkeley, Berkeley, CA 94720.
1
0.1021/jp030127l CCC: $25.00 © 2003 American Chemical Society
Published on Web 05/10/2003

5
204 J. Phys. Chem. B, Vol. 107, No. 22, 2003
Xu et al.
after
earlier.10,11,13 The loading of nickel was kept between 12% and
1
Earlier data from this lab have revealed that this amount of Ni
TABLE 1: Physicochemical Properties of ZrO
2
Calcination at 650 °C
3% by the weight and was checked out with XRF analysis.
BET surface ZrO phasea XLBA sizeb TEM sizec
2
4
,11
2
loading seems to be an optimum for Ni/ZrO2 catalyst. After
drying at 110 °C, the samples were calcined in air at 650 °C
for 5 h, and then stored for further use. Our preparations and
catalytic tests, to be reported elsewhere,15 show that the behavior
of the alcogel-derived Ni/ZrO2 catalysts in methane re-forming
reaction by CO2 is basically not affected by the precalcination
temperature (270-650 °C) of the ZrO(OH)2 alcogel in nitrogen,
where precalcination means processing at the elevated temper-
atures before the impregnation of the nickel salt.
sample
(m /g)
(%)
(nm)
(nm)
ZrO
2
2
2
2
2
-AD
85
44
43
28
26
M45/T55
M30/T70
M64/T36
M80/T20
M98/T2
M9.9/T10.0
M15.1/T15.4
M20.8/T20.2
M23.8/T20.6
M22.5
7-12
15-20
18-25
30-60
40-200
ZrO
ZrO
ZrO
ZrO
-AS
-AN
-CN
-CP
a
M and T represent monoclinic and tetragonal phases, respectively.
Average crystal size obtained with the Scherrer equation by using
b
the ( 1h 11) diffraction (2θ ) 28.5°) for monoclinic and the (111)
c
diffraction (2θ ) 30.4°) for tetragonal crystals, respectively. Particle
The catalytic reaction was conducted at 757 °C in a vertical
size measured by TEM.
fixed-bed U-shape quartz reactor (i.d. 10 mm) under atmospheric
pressure, as described previously.4
,11,13
The reaction temperature
that the XLBA sizes of tetragonal crystals are very close to the
sizes of the monoclinic ones in the alcogel-derived samples,
which seems at variance with conventional opinion that the
tetragonal crystal should be significantly smaller than the
monoclinic at the temperature used, but the discussion of the
crystal sizes of the two phases is beyond the scope of this work.
The agreement in particle sizes from XLBA and TEM measure-
ments for the ZrO2-AD (XLBA, 10 nm; TEM, 7-12 nm), -AS
XLBA, 15-16 nm; TEM, 15-20 nm), and -AN (XLBA, 20-
1 nm; TEM, 18-25 nm) samples indicates that the alcogel-
derived ZrO2-AD, -AS, and -AN samples are composed of
discrete nanocrystals with sizes less than 25 nm.
was measured by a thermal couple in a quartz-well inserted into
the catalyst bed. Unless otherwise specified, the flow rate of
the feed, a 1:1 (molar) mixture of methane and CO2 without
4
using any inert diluent gas, was GHSV ) 2.4 × 10 to 3.8 ×
5
10 mL/(h‚g of cat). Before the reaction, the catalyst was reduced
in situ with a flow of 10% H2/N2 at 700 °C for 3 h.
.2. Sample Characterizations. The crystal structure of ZrO2
2
and Ni/ZrO2 samples were characterized with powder X-ray
diffraction (XRD) on a Bruker D8 Advance X-ray diffractometer
using the Cu K2R source at 40 kV and 40 mA. The percentage
of monoclinic phase (M%) of the oxide solids was measured
(
2
16
according to the equation of Srinivasan et al.:
The ZrO2-CP and ZrO2-CN samples contain ZrO2 nanocystals
averaged respectively to 20-24 and 22.5 nm by XLBA;
however, TEM data show that the nanocrystals form larger hard
aggregates of 30-60 nm in ZrO2-CN and 40-200 nm in ZrO2-
M% ) IM( 1h 11)/(1.6IM( 1h 11) + IT(111)
)
where IM( 1h 11) and IT(111) are the diffraction intensities of the
monoclinic ( 1h 11) and tetragonal (111) planes, respectively. The
average size of ZrO2 crystals was measured with the X-ray
broadening analysis (XLBA) by using the well-known Scherrer
equation:17
2
CP samples. The surface area data (ZrO2-AD, 85 m /g; ZrO2-
2
2
2
AS, 44 m /g; ZrO2-AN, 43 m /g; ZrO2-CN, 28 m /g; ZrO2-CP,
2
2
6 m /g) are closely related to the sizes of zirconia particles,
also confirming that the conventionally prepared samples (ZrO2-
CP and -CN) were present as hard aggregates of sintered
d ) 0.089λ/(B(2θ) cos θ)
nanocrystals.
The physicochemical properties of the reduced Ni/ZrO2
catalysts are presented in Table 2. The XRD measurements
where B(2θ) is the width of the XRD pattern line at half peak
height in radians, λ is the wavelength of the X-ray, θ is the
angle between the incident and diffracted beams in degrees, and
d is the crystal size of the powder sample in nanometers.
Surface areas of the samples were measured with nitrogen
adsorption at -196 °C on a Micromeritics ASAP 2010C
instrument. The samples were dehydrated with flowing dry
nitrogen at 200 °C for 5 h before the adsorption measurement.
TEM measurements of some samples were performed on a
Hitachi H-800 electron microscope to check the crystal sizes
of Ni and ZrO2.
Quantitative TPR measurement was done on a homemade
temperature-programmed analysis system with a temperature
ramp of 15 K/min with procedures similar to those reported in
refs 13 and 18. The TPR measurement was followed by an H2-
TPD to measure the chemisorption of hydrogen. The ratio of 2
(Figure 1) showed no evidence for any possible compound
formation between nickel and zirconia in the samples. The
loaded nickel ions were found completely reducible, as indicated
by the reducibility data in Table 2, in the temperature-
programmed reduction (TPR) with hydrogen. It is important to
note that the particle sizes of Ni metal estimated from the Ni-
dispersion data for the Ni/ZrO2-AD, -AS, and -AN samples were
comparable to the sizes of zirconia “support” in the same
samples as shown in the last two columns of Table 2. This
information was further confirmed by TEM micrograms of the
reduced Ni/ZrO2 catalysts in Figure 2. Energy dispersive analysis
of X-rays (EDAX) showed that the majority of the dark particles
in the TEM micrograms were images of the Ni-metal crystals
whereas particles in lower contrasts were most likely crystals
of zirconia. A further selected area EDAX probe detected no
evidence of any significant contamination between the metal
and oxide components. Thus, these catalysts (Ni/ZrO2-AD, -AS,
and -AN) can better be called nanocomposites of size-
comparable Ni-metal and zirconia nanocrystals. High-resolution
TEM measurements coupled with EDAX and electron diffrac-
tion analysis confirmed the nanocomposite nature by showing
that the nanosize (10-15 nm) Ni-metal crystals were surrounded
by zirconia nanocystals of 15-25 nm, as shown in Figure 3
with Ni/ZrO2-AS as an example. On the other hand, the metal
particles in Ni/ZrO2-CP and -CN samples, which used big
particles of zirconia (>30 nm) as the support, were much smaller
×
(area of the TPD peak)/(area of the TPR peak) gives the
dispersion (D) of the reduced nickel metal, whereas the inverse
of the dispersion data, 1/D, provides an estimation of the crystal
size of the nickel metal catalyst.1
3,17
3. Results and Discussion
After calcination in air at 650 °C, the alcogel derived ZrO2-
AD, -AS, and -AN samples exist as mixtures of the monoclinic
M) and tetragonal (T) phases. The phase composition and
(
XLBA sizes of the monoclinic and tetragonal crystals together
with the TEM particle sizes are given in Table 1. It is apparent

Nanocomposite Ni/ZrO2 for Utilization of Natural Gas
J. Phys. Chem. B, Vol. 107, No. 22, 2003 5205
TABLE 2: Physicochemical Properties of Ni/ZrO Catalysts After the Reduction at 700 °C
2
Ni loada
(wt %)
BET surface
Ni redb
(Ni %)
ZrO
(%)
phasec
Ni dispd
(%)
Ni sizee
(nm)
ZrO
2
sizef
(nm)
2
2
sample
(m /g)
Ni/ZrO
2
2
2
2
2
-AD
12.6
12.5
12.1
12.7
12.9
62
35
38
29
18
98.6
109.1
95.2
93.5
98.5
M51/T49
M45/T55
M41/T59
M35/T65
M5/T95
11.1
9.0
9.4
5.6
5.8
9.1
11.1
10.6
17.2
17.8
7-12
15-20
18-25
30-60
40-200
Ni/ZrO
Ni/ZrO
Ni/ZrO
Ni/ZrO
-AS
-AN
-CN
-CP
a
Measured from X-ray fluoroscence analysis. b Reducibility of nickel measured from temperature-programmed reduction (TPR) with 5% H
/Ar,
2
c
d
as described in ref 13. M and T represent monoclinic and tetragonal phases, respectively. Dispersion of nickel obtained from temperature-
programmed desorption (TPD) of hydrogen after the TPR measurement. Particle size of Ni metal (d) calculated with the assumption: d ) 1/dispersion.
Particle size of zirconia measured by TEM.
e
f
2
Figure 3. High-resolution TEM microgram of the reduced Ni/ZrO -
AS catalyst, showing that the Ni-metal nanocrystal is surrounded by
size-comparable crystals of zirconia. Energy dispersive analysis of
X-rays focused on individual particles showed little contamination
between Ni-metal and zirconia nanocrystals.
Figure 1. X-ray diffraction patterns of Ni/ZrO
at 700 °C.
2
catalysts after reduction
Figure 4. Catalytic activity by methane conversion and stability of
Ni/ZrO
AN, -CN, and -CP are 7-10, 15-20, 20-25, 30-60, and 40-200
nm, respectively.
2 2
catalysts at 757 °C. Note: the particle sizes of ZrO -AD, -AS,
-
reaction time-on-stream over the different Ni/ZrO2 catalysts;
similar curves were obtained for the CO2-conversion data. The
selectivity of syngas in the products was higher than 97% with
a CO/H2 molar ratio of ca. 1.2, and this ratio did not change
during the long lasting reaction. Quantitative analysis of product
4
,10,11
H2O confirms, as in our earlier reports,
that the deviation
from the theoretical product ratio (CO/H2 ) 1.0) is due to an
occurrence of the unavoidable reverse water-gas shift reaction.
Although Ni/ZrO2-CP and -CN catalysts exhibited high activities
that conversions of the reactants were close to their equilibrium
values (86%-90%) under the conditions used, rapid deactivation
of these two catalysts was apparent in less than 50 h. Rapid
deactivation of conventionally prepared Ni/ZrO2 catalyst was
2
Figure 2. TEM micrograms of the reduced Ni/ZrO catalysts. Electron
diffraction pattern analysis (not shown here) reveals that the samples
are mixtures of metallic nickel and zirconia crystals.
than the support particle; these two samples could be regarded
as conventional zirconia-supported nickel catalysts.
The catalytic data of the methane re-forming reaction at 757
4
19
°
C with GHSV of the reactants at 2.4 × 10 mL/(h‚g of cat)
also reported in earlier literature. In contrast, the nanocom-
posite Ni/ZrO2-AS, -AN, and -AD catalysts showed very stable
are given in Figure 4 by plotting CH4 conversion against the

5206 J. Phys. Chem. B, Vol. 107, No. 22, 2003
Xu et al.
Figure 5. Catalytic activity by methane conversion and stability of
Ni/ZrO -AS catalysts at 757 °C at different space velocities of the
reactants: GHSV ) 24 000 (A), 48 000 (B), 72 000 (C), 96 000 (D),
92 000 (E), 240 000 (F), 288 000 (G), 336 000 (H), and 384 000 mL/
h‚g of cat.) (I).
2
Figure 6. Effect of Ni loading on the catalytic activity and stability
of Ni/ZrO -AS catalyst at 757 °C. Reaction conditions are the same as
2
for Figure 4.
1
(
catalytic activities for the re-forming reaction in 100-250 h.
Previously, we reported4
,10,11
that the catalysis of a 27%Ni/ZrO2-
AS (denoted as 27%Ni/ZrO2-B in the original publication)
catalyst is stable for more than 600 h.
The Ni/ZrO2-AD catalyst was further tested for 500 h in the
present study, and we detected no catalyst deactivation. Another
catalytic test up to 1100 h with the Ni/ZrO2-AN catalyst also
2
0
showed no deactivation during the entire reaction period. To
demonstrate the catalyst stability at methane and CO2 conver-
sions far below their thermodynamic equilibrium values (86%-
90%), the present Ni/ZrO2-AS catalyst was tested with stepwise
increasing of the space velocity by 15-fold up to GHSV ) 3.8
5
×
10 mL/(h‚g of cat), and the results are given in Figure 5.
The observed catalytic methane and CO2 (not shown in the
figure) conversions were stable every time after changing the
reaction GHSV. And, surprisingly, the conversion at every
specific GHSV was kept constant after switching between many
different GHSVs during 550 h reaction time-on-stream. Other
tests on the deactivating Ni/ZrO2-CP and -CN catalysts with
2
Figure 7. X-ray diffraction patterns of reduced Ni/ZrO -AS catalysts
with different nickel loadings.
on the preparation of Ni/ZrO2 also indicates that nanozirconia
crystals in stable Ni/ZrO2-AN catalysts can be mainly either
tetragonal or monoclinic, providing that the sizes of the crystals
5
GHSV ) 2.4 × 10 mL/(h‚g of cat) showed that they deactivate
15
are not bigger than 25 nm. To our knowledge, this is the first
much more rapidly than they were reacted in Figure 4 with the
example to show a size limit of nano-oxide particles for a metal
catalyst. We believe that the oxide particles in these stable Ni/
ZrO2 catalysts can no longer be named as a support for the
crystals of Ni metal because these particles are not capable of
holding/supporting a number of the metal crystals (refer to
Figures 2 and 3).
1
0-fold lower GHSV. These results further demonstrated the
extremely stable catalysis of the nanocomposite Ni/ZrO2-AD,
AS, and -AN catalysts. TEM measurements of the used stable
-
catalysts showed that the particle sizes of both the Ni-metal
and zirconia crystals were virtually maintained when the flow
of CO2 was used to cool the catalyst.
We further investigated the effect of nickel loading in the
4
,10,11
It is well-known that ZrO2 crystallizes in two different
phases: tetragonal and monoclinic. The phase composition of
the zirconia support in the reduced catalysts is characterized
by XRD (Figure 1 and Table 2). Although the transformation
of the tetragonal to monoclinic phase was inhibited in the
support of the deactivating Ni/ZrO2-CP catalyst, the very similar
phase composition of the support between the stable alcogel
derived Ni/ZrO2-AN and the deactivating Ni/ZrO2-CN reveals
that the difference in phase composition of the support seems
not a cause of the difference in catalyst stability in Figure 4.
It is shown that the there are no significant differences in
sizes of tetragonal and monoclinic nanocrystals in the present
support oxides (Table 1). The catalytic performances of Ni
catalysts on the ZrO2-AD (7-10 nm), -AS (15-20 nm), -AN
nanocomposite Ni/ZrO -AS catalyst.
It appears that the
2
catalytic stability seems not affected in the range of 5-27%
nickel loading (Figure 6). The sizes of Ni catalysts measured
by hydrogen adsorption and TEM in these samples are in the
2
range between 5.5 nm in 5% Ni/ZrO and 16.5 nm in 27% Ni/
ZrO . Again, the crystal phase composition of the support oxide,
2
which contains 45%, 55%, and 65% tetragonal phase in the 5%,
12%, and 27%Ni/ZrO catalysts, respectively (Figure 7), did
2
not affect significantly the catalyst stability. Reduction of the
1
3
particle sizes of MgO and Al O has also led to stable Ni/MgO
2
3
1
4
and Ni/γ-Al2O3 catalysts, which implies that the metal/oxide
nanocomposite concept could be applicable to other metal/oxide
catalysts.
The question why nanocomposite Ni/ZrO2 catalysts become
so stable remains to be answered in future research. It is
generally agreed that the long-term activity of the catalyst in
the re-forming reaction is dependent upon the balance between
the rate of CH4 decomposition and the rate of CO2 dissociation.
CH4 decomposition to give hydrogen occurs easily on the Ni-
metal catalyst; the carbon atoms left from this reaction remain
on the metal surface. Oxidative removal of the carbon atoms to
produce CO requires a transfer of surface oxygen atoms from
(
4
“
18-25 nm), -CN (30-60 nm), and -CP (40-200 nm) in Figure
thus can be explained with a particle size effect of the zirconia
support”. For stable Ni catalyst, it seems crucial to keep the
ZrO2 particles as discrete crystals with sizes less than 25 nm.
This is thus an up-limit of zirconia particle for obtaining the
stable Ni catalysts in the high-temperature CO2/CH4 reaction.
Our very recent investigation into the effect of processing
temperature of ZrO(OH)2 alcogel, before the loading of nickel,

Nanocomposite Ni/ZrO2 for Utilization of Natural Gas
J. Phys. Chem. B, Vol. 107, No. 22, 2003 5207
either a reduction of the oxide or from CO2 activation at the
(2) Che, M.; Clause, O.; Marcilly, Ch. Supported catalysts. Part A:
General principles, methods and reaction engineering. Handbook of
heterogeneous catalysis; Wiley-VCH: New York, 1997; Vol. 1, p 191.
metal/oxide boundary or perimeter.1
0,19,21-24
It is believed that
the oxygen transfer ability was significantly enhanced by
reducing the particle size of zirconia to less than 25 nm or by
the formation of metal/oxide nanocomposites. This is not a
surprise because a much higher percentage of metal/oxide
boundaries or perimeters exists in the nanocomposite system.
The present data raise a general issue that a lower limit in
the nanometer range may exist for the size of oxide support
particles in supported metal catalysts. When sizes of the support
particles become further smaller, the metal/oxide catalyst forms
nanocomposites that could show distinctive catalytic perfor-
mances. This may have important implication on the design and
preparation of advanced metal catalysts for many key technolo-
gies.
(
3) Boudart, M.; Djegas-Maria Dassous, G. Kinetics of heterogeneous
catalytic reactions; Princeton University Press: Princeton, NJ, 1984.
4) Wei, J. M.; Xu, B. Q.; Sun, K. Q.; Li, J. L.; Zhu, Q. M. Chem. J.
(
Chin. UniV. 2002, 23, 2106. Wei, J. M.; Xu, B. Q.; Li, J. L.; Wang, Y. Q.;
Zhu, Q.-M. Chem. J. Chin. UniV. 2002, 23, 92.
(5) Ashcroft, A. T.; Cheetham, A. K.; Green, M. L. H.; Vernon, P. D.
F. Nature 1991, 352, 225-228.
(
6) Bradford, M. C. J.; Vannice, M. A. Catal. ReV.sSci. Eng. 1991,
4
1, 1.
(
7) Seshan, K.; Lercher, J. A. In Carbon Dioxide: enVironmental
Issues; Paul, J., Pradier, C., Eds.; The Royal Society of Chemistry:
Cambridge, U.K., 1994; p 16.
(
8) Woerner, A.; Tamme, R. Catal. Today 1998, 46, 165-174.
(9) Levy, M.; et al. Solar Energy 1992, 48, 395.
(10) Wei, J. M.; Xu, B. Q.; Li, J. L.; Cheng, Z. X.; Zhu, Q. M. Appl.
Catal. A 2000, 196, L167.
(
11) Wei, J. M.; Xu, B. Q.; Li, J. L.; Cheng, Z. X.; Zhu, Q. M. Stud.
Surf. Sci. Catal. 2000, 130, 3687.
12) Xu, B. Q.; Wei, J. M.; Yin, S. F.; Zhang, X.; Zhu, Q. M. Chin.
Pat. CN:1260324A, 2000. (ZL 00100554.5, 2003).
13) Xu, B. Q.; Wei, J. M.; Wang, H. Y.; Sun, K. Q.; Zhu, Q. M.
Catal. Today 2001, 68, 217.
14) Wei, J. M.; Xu, B. Q.; Li, J. L.; Cheng, Z. X.; Zhu, Q. M. Abstr.
Acknowledgment. We are indebted to Ms. X. H. Chen
(
(School of Material Science, Tsinghua University) for help with
(
high-resolution TEM. We also thank Prof. Z. L. Wang (Depart-
ment of Material Science and Technology, Georgia Institute of
Technology, Atlanta, GA) for his help in analyzing some
samples with TEM, EDX, and electron diffraction in his lab.
Financial support of this work was provided by NSF of China
(
Pap. Am. Chem. Soc. 2001, 221, 50.
(15) Li, Y.; Ye, Q.; Wei, J. M.; Xu, B. Q. Manuscript in preparation.
(
16) Srinivasan, R.; Harris, M. B.; Davis, B. H. J. Mater. Sci. 1998, 3,
87.
17) Innes, W. B. In Experimental Methods in Catalytic Research;
7
(grants: 20125035 and 20073023).
(
Anderson R. B., Ed.; Academic Press: New York, 1968; pp 44-46.
(18) Xu, B.-Q.; Sachtler, W. M. H. J. Catal. 1997, 167, 224.
(19) Lercher, J. A.; Bitter, J. H.; Hally, W.; Niessen, W.; Seshan, K.
Stud. Surf. Sci. Catal. 1996, 101, 463.
Note Added after ASAP Posting. This article was released
to the Web on 5/10/2003. Since then the last word in the
paragraph before the Acknowledgment Section has been changed
from techniques to technologies. The correct version was posted
on 5/20/2003.
(20) Xu, B. Q.; Wei, J. M.; Cheng, Z. X.; Li, J. L.; Zhu, Q. M. Chin.
Pat. CN1280882A, 2001. (ZL 00124421.3, 2003).
(21) Bitter, J. H.; Seshan, K.; Lercher, J. A. J. Catal. 1998, 176, 93.
(
22) Seshan, K.; Bitter, J. H.; Lercher, J. A. Stud. Surf. Sci. Catal. 1998,
13, 187.
23) Stagg, S. M.; Romeo, E.; Padro, C.; Resasco, D. E. J. Catal. 1998,
178, 137.
(24) Sharma, S.; Hilaire, S.; Vohs, J. M.; Gorte, R. J.; Jen, H. W. J.
Catal. 2000, 190, 199.
1
References and Notes
(
(
1) Ertl, G.; Knoezinger, H.; Weitkamp, J., Eds. Catalytic process.
Part B. Handbook of heterogeneous catalysis; Wiley-VCH: New York,
997; Vols. 4 and 5.
1