H. Xin et al. / Journal of Catalysis xxx (xxxx) xxx
3
scanning electron microscopy (SEM) was performed using a Hita-
chi S-4800 microscope. The actual Pt loading of the samples were
determined by Thermo Elemental IRIS Intrepid II XSP inductively
coupled plasma-atomic emission spectroscopy (ICP-AES).
9:1). The reaction began when hydrogen (2.0 MPa, 99.999% purity)
was introduced with stirring (1200 rpm) at 363 K. The reaction was
stopped after a proper time and the products were analyzed using
a GC (GC-2014, Shimadzu) equipped with a flame ionization detec-
tor (FID) and a capillary column (DM-WAX, 30 m ꢂ 0.32 mm ꢂ
Temperature-programmed reduction of hydrogen (H
conducted with a Micromeritics AutoChem II Chemisorption Ana-
lyzer. Prior to H -TPR experiments, the samples were pretreated
in flowing Ar at 423 K for 0.5 h and then cooled to RT in flowing
2
-TPR) was
0.25lm). The response factor of each component was calculated
2
using standard samples and was used to calculate the conversion
and selectivity.
ꢀ1
Ar (99.999%, 30 mLꢁmin ). After that, the sample was heated from
It should be pointed out that the absence of internal and
external mass transfer limitation in the reaction was verified by
Weisz-Prater criterion and Mears criterion, respectively. The
absence of heat transfer limitation in the reaction was verified by
a Mears criterion [31,32]. At the maximum reaction rate, the
ꢀ1
RT to 1073 K with a ramp of 10 Kꢁmin under a mixture of 10%
ꢀ1
H
2
-Ar (30 mL ꢁ min ). The rate of H
by a gas chromatograph (GC) with a thermal conductivity detector
TCD).
Temperature-programmed desorption of hydrogen (H
2
consumption was monitored
(
ꢀ
6
2
-TPD)
C
WP = 8.09 ꢂ 10 < 1 and the C
M
= 0.0633 < 0.15, which assured
was also carried out on Micromeritics AutoChem II Chemisorption
Analyzer. The samples were pretreated at 473 K for 2 h under 10%
the absence of mass transfer limitation. Mears criterion for exter-
0
ꢀD
H
r
ðꢀr
b
Þq RE
g
A
nal (interphase) heat transfer gave j
2
j < 0:15, demon-
h
t
T
R
H
2
-Ar atmosphere and then cooled down to 323 K under flowing
He atmosphere. After H adsorption in 10% H -Ar mixture
50 mL ꢁ min ) at 323 K for 1 h, the samples were purged with
He for 2 h to remove the physically adsorbed species. The H des-
orption was then conducted under He atmosphere with a temper-
b
strating no heat transfer limitations (see the Support Information
for the details). Moreover, we also made the Madon-Boudart test
to confirm that conditions of strict kinetic control were maintained
in this case (see Fig. S1 in the Supporting Information for the
details) [33].
For the recycling reactions, the catalyst was recovered by cen-
trifugation, washed with isopropanol for several times to remove
the residual substrate and product after each run and then submit-
ted to the next run with fresh solvent and reactant.
2
2
ꢀ1
(
2
ꢀ1
ature ramp of 10 Kꢁmin from 323 to 1073 K.
3
Temperature programmed desorption of ammonia (NH -TPD)
was also performed on Micromeritics AutoChem II Chemisorption
Analyzer to characterize the acidity of the samples. The samples
ꢀ1
were firstly outgassed in flowing He (50 mLꢁmin ) at 673 K for
1
NH
sisorbed NH
h, and then adsorption of ammonia was performed in a 10%
ꢀ1
3
-He (50 mLꢁmin ) at 323 K for 0.5 h. After removal of the phy-
3. Results and discussion
ꢀ1
3
by flowing He (50 mLꢁmin ), the chemisorbed NH
3
was analyzed with a TCD detector by heating from RT to 873 K at
a ramp rate of 10 Kꢁmin and then maintained at 873 K for 0.5 h
under the flowing He.
3.1. General characterization of the supports and related Pt catalysts
ꢀ1
The Co
sol-gel method using glucose as template. The structure of the
Co Fe1-xAl 4+d SCMOs was first characterized using XRD. As
shown in Fig. 1A, FeAl 4+d and Co0.3Fe0.7Al 4+d composites exhib-
x 2
Fe1-xAl O4+d SCMOs were synthesized via a modified
The FT-IR spectra was measured to confirm the structure of
Co
x
Fe1-xAl
2
O
4+d SCMOs using a Nexus 870 FT-IR spectrometer in
x
2
O
ꢀ
1
ꢀ1
the range of 400–4000 cm at a spectral resolution of 4 cm
2
O
2
O
and 32 scans. The sample powders were mixed with KBr (1 wt%)
and pressed into self-supported disks at RT. The FT-IR spectra of
the samples using CO as probe molecules were measured with
Nicolet iS50 FT-IR spectrometer in transmission-absorption mode.
The samples were pressed into self-supported wafers and put into
the IR cell. Prior to CO adsorption, the samples were in situ pre-
ited too complicated diffraction patterns to assign simply (see
Fig. S2 in the Supporting Information). With increase of Co amount,
Co0.5Fe0.5Al O4+d and Co0.7Fe0.3Al O4+d displayed six representative
2 2
sharp diffraction patterns at 2h of 31.0°, 36.5°, 44.4°, 55.1°, 58.8°
and 64.6°, respectively, which match well with the PDF card of
2 4
FeAl O . Nonetheless, highly dispersed CoO simple oxides may also
treated under H
2
atmosphere at 473 K for 2 h and then cooled
exist although they cannot be detected due to high and uniform
down to 308 K for CO adsorption. The IR spectra of the chemi-
sorbed CO was recorded after physically adsorbed CO was evacu-
ated. All of the FT-IR spectra were collected using 32 scans at a
dispersion (see Fig. S3 in the Supporting Information). If x value
further increased to 1, CoAl
patterns at 2h of 31.2°, 36.7°, 44.7°, 55.5°, 59.2°, and 65.0°, which
can be ascribed to CoAl (2 2 0), (3 1 1), (4 0 0), (4 2 2), (5 1 1)
and (4 4 0) planes, respectively (see Fig. S4 in the Supporting
Information). When compared with those of Co0.5Fe0.5Al
Co0.7Fe0.3Al 4+d, there is a little shift to higher angles for the
diffraction angle of CoAl . In addition, besides the Co
Al 4+d composition presented in the catalyst, we cannot exclude
the existence of simple oxides such as CoO, FeO, Fe because
twice the amount of Co + Fe was adopted for the proposed Co
2 4
O also showed six sharp diffraction
ꢀ1
resolution of 4 cm
.
2 4
O
The X-ray photoelectron spectra (XPS) of the samples were
measured with a Thermo Fisher Scientific ESCALAB 250Xi spec-
2
O
4+d and
Fe1-x
trometer with Al K
a
radiation (1486.6 eV) as incident beam with
2
O
a monochromator. The samples were in situ pre-treated in flowing
hydrogen at 473 K for 2 h in a reactor attachment of the XPS spec-
trometer for comparison. The binding energy (BE) was calibrated
using C-C binding energy at 284.6 eV in order to compare the BE
with data in the literature. The spectra shown in the figures have
been corrected by subtraction of a Shirley background. Spectral fit-
ting and peak integration were done using the XPS PEAK software.
2
O
4
x
2
O
2 3
O
x
x
Fe1-x
Fe1-x
Al
Al
2
O
O
4+d composition. In order to verify the structure of Co
2
4+d SCMOs, FT-IR spectroscopy was adopted (Fig. 1B). For
, the spinel structure can be identified by two sharp bands
CoAl
at 685 and 559 cm , which can be attributed to Al-O vibrations
related to tetrahedral AlO and octahedral AlO groups in CoAl
34]. With decreasing x value while increasing the Fe amount in
Co Fe1-xAl 4+d SCMOs, those two IR bands red-shifted obviously,
which might be caused by impure phase. That is, doping of Fe in
Co Fe1-xAl 4+d SCMOs changed the crystal structure clearly. This
2 4
O
ꢀ1
2.4. Catalytic test
4
6
2 4
O
[
The catalysts were tested for the liquid-phase selective hydro-
x
2
O
genation of CAL in a 100 mL autoclave. For a typical test, 25 mg
of Pt catalyst was pretreated in a specially designed quartz tube
x
2
O
ꢀ
1
under H
2
(99.999% purity, 30 mLꢁmin ) at 473 K for 2 h before
is in good agreement with the XRD results.
use. Then, the pretreated catalyst was immediately transferred into
the autoclave without exposure to air and mixed with CAL and sol-
vent (containing isopropanol and water with a volume ratio of
The relevant Pt catalysts were also characterized using XRD. As
x 2
also displayed in Fig. 1A, the Pt/Co Fe1-xAl O4+d catalysts displayed
the similar diffraction patterns to their supports, demonstrating
Please cite this article as: H. Xin, Y. Xue, W. Zhang et al., Co
x 2