L. Ma, D. He / Catalysis Today 149 (2010) 148–156
149
that the activity and selectivity for the glycerol hydrogenolysis to
12 h and stored in the desiccator. The samples reduced by H
2 2
–N
propanediols with Ru catalysts alone are still not satisfying. The
modification of Ru catalysts for improving their performance in the
hydrogenolysis of glycerol is necessary. It has been reported that
bimetallic catalysts can be superior to monometallic catalysts in
the catalytic activity and selectivity for many reactions [14]. Maris
et al. applied bimetallic PtRu/C and AuRu/C catalysts to the
flow were denoted as Ru/SiO -rT and Ru–Re/SiO -rT, and here ‘‘T’’
2
2
refers to the pre-reduction temperatures (450, 300, 200 8C). The
actual loading of Ru and Re on the catalysts was analyzed by
Inductive Coupling Plasma-Atomic Emission Spectroscopy (ICP-
AES), that is, 3.55% Ru/SiO
SiO
2 2
, 3.23% Ru-3.57% Re/SiO and 4.01% Re/
2
.
aqueous-phase hydrogenolysis of glycerol at 473 K and 40 bar H
2
and found that the PtRu catalyst appeared to be stable under the
aqueous-phase reaction conditions, but it seems that the activity
and selectivity of AuRu/C or PtRu/C in the hydrogenolysis of
glycerol was similar to the monometallic Ru/C [12,13]. Generally,
the synergistic effect of bimetals and the interaction between
metal and support could be influenced by the preparation methods
and pretreatments of catalysts, which in turn affect the catalytic
2.2. Characterization of catalysts
The specific surface areas, cumulative pore volume and average
pore diameter of the catalysts and supports were measured by N
adsorption/desorption with the BET and BJH methods on a
Micromeritics ASAP 2010C analyzer. Before measurement, the
samples were degassed at 200 8C for 2 h. The phase structures of
the catalysts were determined by X-ray Diffraction (XRD) with a
2
2 3
performance of the catalysts [15–22]. For Pt–Re/Al O reforming
catalysts, it has been reported that bimetallic particles of platinum
and rhenium could be formed when the catalyst was dried in air at
temperatures ꢀ500 8C before reduction at 480 8C [15]. On the other
Bruker D8 Advance X-Ray Powder Diffractometer with Cu K
a
(l
= 0.15406 nm). The crystal sizes of metal or metal oxides were
determined by means of the X-ray line broadening method
according to the well-known Scherrer formula. Raman spectra
were recorded at room temperature on a Microscopic Confocal
2 3 2 3
hand, the degree of surface hydroxylation of Al O in Pt–Re/Al O
reforming catalysts essentially affected aggregation or segregation
of Pt and Re during reduction step [16]. Thus, pretreatments at
severe conditions could lead to dehydroxylation of the alumina
surface and result in Pt–Re segregation, while pretreatments at
moderate conditions preserved surface hydroxyl groups, which
favored the transport of ReOx species to Pt during the reduction
step, and resulted in the aggregation of metals [16]. Epron et al.
found that the interactions between platinum and copper on a Pt–
Cu bimetallic catalyst were noticeably affected by reducing and
oxidizing pretreatments, and consequently the catalytic activity
for nitrate reduction in water was changed [17].
In our previous work, we found a remarkable promoting effect
of Re component on the activity of Ru catalysts in the
hydrogenolysis of glycerol to propanediols [23], and investigated
the behaviors of Ru–Re bimetallic catalysts in the hydrogenolysis
of glycerol and confirmed the significant synergistic effect of Ru
and Re on increasing the activity of the catalysts [24]. Recently, we
also found that the preparation parameters could obviously affect
the catalytic performance of Ru–Re bimetallic catalysts and
corresponding Ru monometallic catalysts. In this paper, we focus
+
Raman spectrometer (Renishaw, RM 2000) equipped with Ar
laser. The 633 nm line from an argon ion laser was used as the
excitation source. The laser was operated at a power of 15 mW. The
morphologies and microstructure of the catalysts were character-
ized by high-resolution transmission electron microscopy (HR-
TEM, JEM-2010 of JEOL) equipped with an energy dispersive X-ray
detector (EDX). The accelerating voltage was 120 kV. The samples
were ultrasonically dispersed in ethanol and deposited on a holey
carbon copper grid before measurement. The X-ray photoelectron
spectroscopy (XPS) measurements were performed on a PHI
Quantera Scanning X-ray Microprobe of ULVAC-PHI Inc. The
spectra were referenced with respect to C 1s line at 284.8 eV.
Temperature-programmed reduction (TPR) measurements
were carried out on a dynamic-flow gas sorption instrument
(Quantachrome, CHEMBET 3000 TPR/TPD). The catalyst samples
(about 100 mg) were treated in Ar at 350 8C for 0.5 h before TPR
was performed. All TPR measurements were carried out in a flow of
2
4.97% H /Ar (20 ml/min) at a heating rate of 15 8C/min. A cold trap
(liquid nitrogen + iso propanol) was placed before the TCD to
remove water produced during TPR measurements. The hydrogen
consumption was calibrated using TPR of copper oxide (CuO) at the
same conditions.
on the influence of pretreatment of Ru–Re/SiO
2 2
and Ru/SiO
precursors on catalytic performance in glycerol hydrogenolysis to
propanediols.
The dispersion of Ru metal on supports was measured by CO
chemisorption method. CO chemisorption was operated on
dynamic-flow gas sorption instrument (Quantachrome, CHEMBET
3000 TPR/TPD). Before CO chemisorption was performed, catalyst
samples (about 200 mg) were treated in a quartz reactor in a flow
2
. Experimental
2.1. Catalyst preparation
SiO
support, which was calcined at 400 8C for 4 h before used.
Re (CO)10 was purchased from Stream Company and used as
received. RuCl O and HReO , which were purchased from
ꢁ4H
Institute of ShenYang Youse Jinshu and Alfa Aesar Company,
respectively, were used for preparing Ru/SiO2, Re/SiO and Ru–Re/
SiO bimetallic catalysts.
Supported monometallic catalysts (Ru/SiO
bimetallic Ru–Re/SiO catalyst were prepared by impregnation
method. The powder of SiO was impregnated with RuCl O or
ꢁ4H
HReO aqueous solution or the mixture aqueous solution of
RuCl O and HReO . After impregnation and solvent removal
ꢁ4H
2
was obtained from Tianjin Chemical Institute and used as a
2
of H (20 ml/min) at 400 8C for 2 h and then purged in He flow. The
temperature of samples was decreased to 40 8C in He flow. The CO
chemisorption was performed by pulse injection of pure CO gas at
40 8C. The stoichiometry of CO to Ru was 1 for CO chemisorption
[25].
2
3
2
4
2
2
2.3. Hydrogenolysis reaction of glycerol
2
2
and Re/SiO ) and
2
The hydrogenolysis of glycerol (ultra pure, Alfa Aesar Company)
was carried out in a stainless steel autoclave of 100 ml with a
magnetic stirrer. Detailed procedure was described in our previous
paper [23]. The standard reaction conditions were 160 8C, 8 MPa
hydrogen pressure and 8 h reaction time, using 10 ml 40 wt%
glycerol aqueous solution and 0.15 g supported catalyst in every
run. After the reaction, the liquid and the solid catalyst in the
mixture were separated by centrifugation and filtration.
2
3
2
4
3
2
4
by evaporation, the precursors were dried at 110 8C for 12 h, and
calcined at 350 8C in air for 5 h. The calcined samples were denoted
as Ru/SiO
calcined samples were further reduced in H
= 3/5, 80 ml/min) at 450 8C (or 300, 200 8C) for 4 h, and finally
passivated in a flowing of CO –N gas (CO /N = 2/5, 70 ml/min) for
2
-c350, Re/SiO
2
-c350 and Ru–Re/SiO
2
-c350. Some
2
–N
2
flow (H /
2
N
2
The products in liquid phase were analyzed qualitatively by
GC–MS (GCMS-QP2010, SHIMADZU Corporation) and analyzed
2
2
2
2
Ma, Lan
