138
K. Shimura et al. / Applied Catalysis A: General 462–463 (2013) 137–142
Table 1
3 h. ZrO2 was prepared by hydrolysis of zirconium oxynitrate 2-
hydrate in distilled water by gradually adding an aqueous NH4OH
solution (1.0 mol dm−3), filtration of precipitate, washing with dis-
tilled water three times, drying at 100 ◦C, and calcining at 500 ◦C.
SiO2 (Q-10) was supplied from Fuji Silysia Chemical Ltd. C (active
carbon) was purchased from Kishida Chemical Co., Ltd. Ni metal,
NiO and Ni2O3 were purchased from Mitsuwa Chemical. Raney Ni
(B113W, Ni >90%) was supplied from Evonik Industries.
Ni K-edge EXAFS analysis of CeO2-supported Ni (5 wt%) samples.
TH2a/◦C
Shell
Nb
R /A
ꢀd/A
Rf e/%
c
˚
˚
–
O
Ni
O
Ni
O
Ni
Ni
O
Ni
O
Ni
Ni
6.9
7.8
5.8
6.4
2.4
2.0
1.8
0.8
5.3
0.5
5.6
0.4
5.9
6
1.99
2.95
2.08
2.95
2.02
2.49
2.94
1.88
2.45
1.86
2.44
1.84
2.44
2.02
2.94
2.49
0.145
0.106
0.129
0.100
0.100
0.085
0.080
0.040
0.103
0.00
0.107
0.03
0.109
–
1.4
200
250
6.6
4.2
The catalyst precursor (named NiO/CeO2) was prepared by an
impregnation method. A mixture of the support and an aqueous
solution of Ni nitrates were evaporated at 50 ◦C, followed by drying
at 90 ◦C for 12 h, calcination in air at 350 ◦C for 4 h. Before catalytic
experiments, the precursor (NiO/CeO2) was pre-reduced in a pyrex
tube under a flow of H2 (20 cm3 min−1) at a reduction tempera-
ture (TH2) for 0.5 h. The specific surface area of 5 wt% Ni/CeO2 was
300
350
400
NiOf
1.46
0.86
1.1
–
–
–
69 m2 g−1
.
12
12
–
–
Nif
a
Temperature of H2 reduction.
Coordination number.
Bond distance.
Debye-Waller factor.
2.2. Characterization
b
c
d
e
f
XRD patterns of the powdered catalysts were recorded with a
Rigaku MiniFlex II/AP diffractometer with Cu K␣ radiation. Specific
surface area was obtained by measuring N2 adsorption at −196 ◦C
by BELCAT.
Crystallographic data of NiO and Ni metal.
Ni K-edge XAFS measurements were performed in a transmis-
sion mode at the BL01B1 in SPring-8 operated at 8 GeV. A Si(1 1 1)
single crystal was used to obtain a monochromatic X-ray beam. A
pressed pellet of the pre-reduced Ni/CeO2 sample (Ni = 5 wt%) was
and extended X-ray absorption fine structure (EXAFS) shown in
Fig. 1. Table 1 shows the results of curve-fitting analyses of the
EXAFS. The curve-fitting of the EXAFS of unreduced precursor,
˚
named NiO/CeO2, gave a Ni O shell at 1.99 A with Ni Ni coordina-
placed in a quartz in situ cell in a flow of 100% H2 (100 cm3 min−1
)
˚
tion number (N) of 6.9 and a Ni Ni shell at 2.95 A of N = 7.8 (Table 1).
for 0.5 h at each temperature under atmospheric pressure, then the
sample was cooled to 40 ◦C under a flow of He, and then a XAFS
spectrum was obtained. The analysis of EXAFS was performed using
the REX version 2.5 program (RIGAKU). The Fourier transformation
of the k3-weighted EXAFS oscillation from k space to r space was
performed over the range 22–126 nm−1 to obtain a radial distri-
bution function. The inversely Fourier filtered data were analyzed
By comparison with the crystallographic data of NiO, the EXAFS fea-
ture of NiO/CeO2 was assigned to NiO species. On the other hand,
the EXAFS of the Ni/CeO2 catalyst reduced at 400 ◦C consists of
˚
˚
a Ni Ni shell at 2.44 A (N = 5.9) with a weak Ni O shell at 1.84 A
(N = 0.4). By comparison with the crystallographic data of Ni metal,
the Ni Ni shell was assigned to metallic Ni. The Ni Ni coordina-
tion number (5.9) was smaller than that of Ni foil (12), indicating
a small size of Ni metal nanoparticles on Ni/CeO2. The weak con-
tribution of the Ni O shell indicates the presence of NiO species
as minor species. The EXAFS of the Ni/CeO2 catalyst reduced at
a medium temperature (250 ◦C) consists of characteristic features
with a usual curve fitting method in the k range of 22–126 nm−1
.
The parameters for the Ni O and Ni Ni shells have been provided
by the FEFF6.
2.3. Typical procedures of catalytic reactions
˚
due to NiO (a Ni O shell at 2.02 A with N = 2.4 and a Ni Ni shell at
2.94 A with N = 1.8) and a feature due to Ni metal (a Ni Ni shell at
2.49 A with N = 2.0), which indicates the co-presence of NiO and Ni
˚
˚
Commercially available organic compounds were used without
further purification. Typically, 3 or 5 wt% Ni/CeO2 (3 mol% of Ni with
respect to alcohol) was used in catalytic experiments. After the
pre-reduction, we carried out catalytic tests without exposing the
catalyst to air as follows. The mixture of o-xylene (1.5 mL), alcohol
(1 mmol), and n-dodecane (0.5 mmol) was injected to the pre-
reduced catalyst inside a reactor (cylindrical glass tube) through
a septum inlet. Then, the reactor was purged by N2 and set in a
reaction vessel equipped with a condenser. The resulting mixture
was stirred at 80–144 ◦C. Conversion and yields of products were
determined by GC using n-dodecane as an internal standard. The
GC-FID (Shimadzu GC-14B) and GC–MS (Shimadzu GCMS-QP5000)
analyses were carried out with a Rtx-65 capillary column (Shi-
madzu) using nitrogen or helium as the carrier gas. The products
were identified by GC–MS and by comparison with commercially
pure products.
metal species. To estimate the fraction of Ni metal and NiO species
in the catalysts, we carried out pattern fitting analysis of Ni K-edge
XANES. We obtained the fractions of the Ni metal and NiO by the
least-squares fitting of the XANES spectrum of the Ni/CeO2 sample
with a linear combination of XANES spectra of Ni foil and NiO. The
fraction of Ni metal and NiO species in each catalyst was plotted in
Fig. 2A as a function of the reduction temperature. The fraction of
metallic Ni increased with increase in the reduction temperature.
First, we compared the activity of various Ni(5 wt%)-loaded cat-
alysts pre-reduced at 500 ◦C and Ni compounds for self-coupling of
2-octanol (Table 2). Among various Ni catalysts, Ni/CeO2 showed
highest yield of dimer products (ketone 1a and alcohol 2a). In
contrast, Ni-loaded ZrO2, ␥-Al2O3, TiO2, SiO2 and active carbon
catalyzed the dehydrogenation of 2-octanol to give 2-octanone as
the main products (entries 3–7). Yield of the dimer products for
Ni/CeO2 was much higher than those on other Ni-loaded cata-
lysts even under a similar conversion level (38–74%, entries 2–7).
Ni-loaded ZrO2, ␥-Al2O3 or TiO2 catalysts pre-reduced at lower
temperature (250 ◦C) were also tested (not shown), but the yields of
the dimers were below 8%. Cu/CeO2 and Co/CeO2 catalysts (entries
3. Results and discussion
3.1. Structure of Ni/CeO2
The XRD spectrum of 5 wt% Ni/CeO2 showed no peaks due to
metallic Ni and NiO possibly due to high dispersion of Ni species.
Thus, the structure of Ni species on the Ni/CeO2 catalysts was stud-
ied by Ni K-edge X-ray absorption near-edge structures (XANES)