Selective hydrogenolysis of glycerol to propanediols on supported Cu-containing bimetallic catalysts

Article abstract of DOI:10.1039/c0gc00058b

Supported Cu-containing bimetallic catalysts were prepared and used to convert glycerol to propanediols. The effects of supports, metals, metal loadings, and impregnation sequences were examined. A synergistic effect was observed between Cu and Ag when they were impregnated on γ-Al ₂O₃. Characterizations revealed that the addition of Ag not only resulted in an in situ reduction of CuO, but also improved the dispersion of the Cu species on the support. A CuAg/Al₂O₃ catalyst with optimal amounts of Cu and Ag (Cu/Ag molar ratio 7:3, 2.7 mmol Cu+Ag per gram of γ-Al₂O₃) showed a near 100% selectivity to propanediols with a glycerol conversion of about 27% under mild reaction conditions (200 °C, 1.5 MPa initial H₂ pressure, 10 h, (Cu+Ag)/glycerol molar ratio of 3/100). Compared with a commercial copper chromite catalyst commonly used for this reaction, the CuAg/Al₂O ₃ catalyst had much higher activity and did not need a reduction pretreatment.

Full text of DOI:10.1039/c0gc00058b

View Online / Journal Homepage / Table of Contents for this issue
PAPER
www.rsc.org/greenchem | Green Chemistry
Selective hydrogenolysis of glycerol to propanediols on supported
Cu-containing bimetallic catalysts
a
a
b
a
a,b
Jinxia Zhou, Liyuan Guo, Xinwen Guo, Jingbo Mao and Shuguang Zhang*
Received 28th April 2010, Accepted 20th July 2010
DOI: 10.1039/c0gc00058b
Supported Cu-containing bimetallic catalysts were prepared and used to convert glycerol to
propanediols. The effects of supports, metals, metal loadings, and impregnation sequences were
examined. A synergistic effect was observed between Cu and Ag when they were impregnated on
g-Al . Characterizations revealed that the addition of Ag not only resulted in an in situ
reduction of CuO, but also improved the dispersion of the Cu species on the support. A
CuAg/Al catalyst with optimal amounts of Cu and Ag (Cu/Ag molar ratio 7 : 3, 2.7 mmol
Cu+Ag per gram of g-Al
2
O
3
2
O
3
2
O
3
) showed a near 100% selectivity to propanediols with a glycerol
◦
conversion of about 27% under mild reaction conditions (200 C, 1.5 MPa initial H
2
pressure,
1
0 h, (Cu+Ag)/glycerol molar ratio of 3/100). Compared with a commercial copper chromite
catalyst commonly used for this reaction, the CuAg/Al
did not need a reduction pretreatment.
2
O
3
catalyst had much higher activity and
1
. Introduction
Ru catalyst exhibited a markedly high activity, 90.1% glycerol
◦
conversion at 180 C and 5 MPa H
2
pressure, but that the
As important commodity chemicals, propanediols are widely
used as functional fluids such as de-icing reagents, an-
tifreeze/coolants, and as precursors for the syntheses of un-
saturated polyester resins and pharmaceuticals. Propanediols
are currently produced from propylene via a process involving
propanediol selectivity was only 20.6%. The use of supported
noble metal catalysts with various acidic materials was also
9–12
studied. The combination of a Ru/C and an Amberlyst resin
1
presented a glycerol conversion of 79.3% and a propanediol
◦
selectivity of 74.7% at 120 C, 8 MPa initial H
1
2
pressure, and
selective propylene oxidation to propylene oxide and subsequent
9
0 h. The use of Ru/C with other acidic materials, such as
2,3
hydrolysis. The process is restricted by the supply of propylene
derived from crude oil, and that stimulated a search for
more economical and renewable alternative feeds. Glycerol is a
renewable chemical which is usually produced by steam splitting
or saponification of animal fat or vegetable oil. Now it can
also be obtained as a major by-product from the biodiesel
niobia, 12-tungstophosphoric acid (TPA) supported on zirconia,
the caesium salt of TPA and the caesium salt of TPA supported
on zirconia, was studied by Balaraju et al. They found a linear
11
correlation between conversion and acidity on the catalysts.
12
Alhanash et al. prepared a bifunctional catalyst by loading Ru
onto a heteropoly acid salt Cs2.5
H0.5[PW12O40], which achieved
4
production process. Readily available glycerol has made the
9
6% selectivity to 1,2-propanediol at 21% glycerol conversion
catalytic hydrogenolysis of glycerol to propanediols a highly
viable alternative route. The reactions and processes for glycerol
conversion to valuable compounds have been well covered in
◦
at 150 C and 5 MPa initial H
supported bimetallic noble metal catalysts, Ma et al. discovered
2
pressure in 10 h. As of
13
a promoting effect of Re on the catalytic performances of
5,6
previous reviews. The following discussion intends to provide
more details about catalyst preparation and testing conditions
for glycerol hydrogenolysis.
Ru/Al
2
O
3
, Ru/C, and Ru/ZrO , both on the conversion of
2
glycerol and the selectivity to propanediols.
Transition metal oxide catalysts used for the hydrogenolysis
of glycerol often exhibit good selectivity towards propanediols.
Supported noble metals and transition metal oxides have been
reported in the literature as catalysts for the hydrogenolysis of
glycerol. Noble metal based catalysts are usually more active
than transition metal oxide catalysts, but the selectivity to
3
Chaminand et al. showed that nearly 100% selectivity to 1,2-
propanediol was achieved on a CuO/ZnO catalyst, but the
activity of the catalyst was so low that it took 90 h to reach 19%
3
propanediols is lower. Among several supported metal catalysts
14
glycerol conversion. Wang and Liu obtained a selectivity of
(
metal: Rh, Ru, Pt, Pd; support: active carbon, SiO
2
, Al
2
O
3
),
◦
8
3.6% to 1,2-propanediol at 22.5% glycerol conversion at 200 C
7
Furikado et al. found that Rh/SiO
2
was the most active
and 4.2 MPa initial H
2
pressure, after 12 h on a Cu–ZnO catalyst
and selective one, with a glycerol conversion of 19.6% and a
15
(Cu/glycerol = 3 wt%). Dasari et al. reported that a commercial
◦
propanediol selectivity of 39.8% at 120 C, 8 MPa initial H
2
◦
copper chromite catalyst (pre-reduced at 300 C) converted
8
pressure, and 10 h. Feng et al. found that a TiO
2
supported
glycerol to 1,2-propanediol with a selectivity of 85.0% at 54.8%
◦
conversion at 200 C and 1.4 MPa initial H
2
pressure, after
2
4 h. Considering the good performance of copper chromite
a
College of Environmental and Chemical Engineering, Dalian University,
Dalian, 116622, China. E-mail: dluzhang@126.com;
catalysts under relatively mild reaction conditions as well as the
vulnerability of noble metal catalysts to impurities in glycerol,
the process using copper chromite is considered to be the most
Fax: +86-411-87402449; Tel: +86-411-87403214
b
State Key Laboratory of Fine Chemicals, Dalian University of
4,15
Technology, Dalian, 116012, China
promising one for commercialization. Another catalyst that
This journal is © The Royal Society of Chemistry 2010
Green Chem., 2010, 12, 1835–1843 | 1835

View Online
1
6
may be used at relatively low hydrogen pressure or even without
hydrogen is RANEYꢀR Ni.
Cu, S represents the amount of total metals (mmol) loaded on
a support (per gram), and Cu:X is the molar ratio of Cu to the
17
In spite of many research efforts, this important reaction has
not been successfully commercialized due to several common
drawbacks with the catalysts used. The main problem of
supported noble metal catalysts is the low selectivity towards
propanediols. The Cu-based catalysts exhibited superior perfor-
mances in terms of the selectivity towards propanediols, but
their activities are usually low. The Cu on these catalysts is
initially in an oxidic state. In order to achieve a decent activity,
the catalysts often need to be reduced in situ under reaction
other metal. For example, CuAg/Al
with 2.7 mmol Cu plus Ag, a Cu/Ag molar ratio 7 : 3, supported
on 1 g of Al
2
3
O (2.7)(7 : 3) is a catalyst
2
3
O .
A copper chromite catalyst (Engelhard Corporation, US) was
tested for comparison purposes.
2.2 Catalyst characterization
X-Ray Diffraction (XRD) patterns of the catalysts were
recorded at room temperature on an X-ray diffractometer
14
conditions with very high H
2
pressure, or to be pre-reduced
to generate Cu species which are catalytically active under mild
(
D/max-2400) with a graphite monochromator attachment,
utilizing Ni-filtered Cu-Ka radiation (40 kV, 100 mA) with a
1
8
reaction conditions.
◦
scanning speed (2q) of 1 per min.
1
9
In previous work, we reported that a selectivity to propane-
diols of about 97% with a glycerol conversion of about 50% was
Nitrogen adsorption experiments for pore size distribution,
pore volume, and surface area measurements were conducted
on an AUTOSORB-1 instrument (Quantachrome). All samples
achieved on a Cu/Al
catalysts, however, this Cu/Al
in H . By employing another metal besides Cu to prepare
2
O
3
catalyst. Like most transition metal
2
O
3
catalyst had to be pre-reduced
◦
were calcined at 350 C under vacuum before the measurements.
2
Temperature Programmed Reduction (TPR) studies of the
supported bimetallic catalysts, the present study aimed at
eliminating the pre-reduction step while maintaining or further
improving the activity and selectivity. Various characterization
techniques were applied to investigate the in situ generation of
active sites under reaction conditions.
catalysts were carried out in a 10% H
flow rate of 50 ml min with a temperature ramp of 10 C min .
Before TPR tests the catalysts were pretreated in argon at 300 C
2
/Ar gas mixture at a
-
1
◦
-1
◦
for 2 h. Hydrogen consumption was monitored using a thermal
conductivity detector (TCD).
X-Ray Photoelectron Spectroscopy (XPS) measurements
were performed on an ESCALAB-250 spectrometer with a
monochromatic X-ray source (Al Ka line of 1486.6 eV energy
and 150 W) to characterize surface species on the catalysts. XPS
elemental spectra were acquired in a fixed analyzer transmission
mode with steps of 0.1 eV energy at a pass energy of 50 eV.
All spectra were aligned based on the adventitious C 1 s peak
at 284.6 eV. The spectra were fitted with an XPSPEAK 41
program using a Shirley background and Gaussian line shapes
for the deconvolution into individual states and for obtaining
peak areas.
2
. Experimental
2
.1 Catalyst preparation
Our Cu-based catalysts were synthesized using an incipient
wetness impregnation method similar to those described in
the literature
and/or another metal nitrate (AgNO
Cr (NO ·3H O, Ni(NO ·3H O, or Co(NO
ports used in this study include g-Al (Shandong Filiale of
China Aluminium Co., Ltd., China), HY zeolite (Wenzhou
Huahua Co., Ltd., China), HZSM-5 zeolite (Nankai University
Catalyst Co., Ltd., China), and Hb zeolite (PQ Zeolites B.V.).
Some physical properties of these supports are summarized in
Table 1.
2
0,21
with aqueous solutions of Cu(NO
Zn(NO
·H O). The sup-
3
)
2
·3H
2
O
O,
3
,
3
)
2
·6H
2
2
3
)
3
2
3
)
2
2
3
)
3
2
2
O
3
2.3 Catalytic activity measurement
The hydrogenolysis of glycerol was carried out in a high
throughput batch reactor system consisting of ten independent
stainless steel autoclaves (165 ml each) with mechanical stir-
ring. An aqueous solution of glycerol (50 wt% concentration)
prepared with pure glycerol (>99%, China National Medicines
Corporation Ltd., China) and deionized water was used as feed.
In a typical run, 65 g of the glycerol solution and a specified
quantity of the catalyst were loaded into the reactor. The molar
ratio of the active metal(s) on the catalyst to the glycerol (M:G)
◦
After impregnation, the catalysts were dried at 110 C for
◦
12 h and calcined in air for 3 h at 400 C, except in the
study of calcination temperature. Unless specifically stated,
these catalysts were tested or characterized directly after the
calcination without a pre-treatment in H
2
. In a few cases, the
stream
),
). All catalysts were in powder
◦
catalysts were pre-reduced at 300 C for 3 h in a H
2
-
1
(
30 ml min ), and then the names were postfixed with (H
2
for example, Cu/Al (2.0)(H
2
O
3
2
was 3 : 100. The reactor was purged five times with H
Dalian F.T.Z Gredit Chemical Technology Development Co.,
Ltd.) and pressurized with H to 1.5 MPa at room temperature.
2
(99.99%;
form with particle size less than 0.32 mm in diameter. The
catalysts prepared were designated as CuX/Support(S)(Cu:X),
in which X represents any other metal that was loaded besides
2
With stirring at 400 RPM, the mixture of the glycerol and
◦
Table 1 Physical properties of supports
the catalyst was heated to 200 C and maintained for 10 h.
The stirring speed was selected to eliminate the influence of
external mass transfer and to avoid creating splash inside the
reactor which would make sampling and temperature control
very difficult. Hydrogen was fed on demand so as to keep the
total reaction pressure at 3.6 MPa during the 10 h period.
After the reaction, the gas phase products were collected in a
gas bag and the liquid phase products were separated from the
SiO
(mol/mol)
2
/Al
2
O
3
BET surface
area/m g
Pore volume/
mL g
2
-1
-1
Support
g- Al
2
O
3
313
718
356
731
0.37
0.42
0.31
0.87
HY
5.6
48
27
HZSM-5
Hb
1
836 | Green Chem., 2010, 12, 1835–1843
This journal is © The Royal Society of Chemistry 2010

View Online
catalyst by filtration. These products were analyzed using a gas
chromatograph (GC HP5890) equipped with a flame ionization
detector. The GC column used was a PEG2W capillary column
the formation of propanediols with little or no selectivity to-
wards ethylene glycol and other degradation by-products.
On all these catalysts, the majority of the propanediols
formed was 1,2-propanediol (1,2-PD) and the amount of 1,3-
propanediol (1,3-PD) was very small. However, the activities
3,14,15,19
(
30 m ¥ 0.32 mm ¥ 0.5 mm) manufactured by Dalian Institute
of Chemical Physics, Chinese Academy of Sciences. Solutions
of n-butanol with known amounts of internal standards were
prepared and used for quantification of various glycerol-derived
compounds in the products.
of these catalysts differed significantly. CuAg/Al
active, with a glycerol conversion of 27%, 3–6 times that of
the CuZn/Al and CuCr/Al catalysts. Other bimetallic
catalysts tested, CuNi/Al and CuCo/Al (results not
shown here), also showed poor activities.
Cu and Ag, separately supported on Al
and Ag/Al ), did not generate a comparable result to
CuAg/Al
reduction in hydrogen at an optimized temperature is nec-
essary for Cu/Al catalysts to generate good activity and
selectivity. Interestingly, the activity of the Ag/Al catalyst
2
O
3
was most
2
O
3
2
O
3
The conversion of glycerol, the selectivity, and the yield of
propanediols were used to evaluate the performance of each cat-
alyst. They were defined by the following equations. The amount
of glycerol converted was calculated from the total amount of
carbon based species formed in the product. Only trace amount
of product was detected in the gas phase. The carbon on a used
catalyst was analyzed with thermogravimetry and C/H elemen-
tal analyses (results not shown here). Overall carbon balance in
the product was >98%. Products besides propanediols, such as
ethylene glycol and 1-propanol, were also identified and listed
as “Others” in the tables below. Repeated runs showed that data
variation was in the range of ±5% (relative value).
2
O
3
2
O
3
2
O
3
(Cu/Al
2
O
3
2
O
3
19
2
3
O . In a previous paper, we showed that pre-
2
O
3
2
O
3
is quite impressive considering the relatively low Ag loading. Its
catalytic performance was better than those of the supported
7
–13
noble metal catalysts reported
selectivity. This type of Ag/Al
in terms of activity and
catalysts will be investi-
2
O
3
gated further in a separate paper. The physical mixture of
Cu/Al (1.9) and Ag/Al (0.8) generated a glycerol con-
version of 20%, much higher than the sum of the conver-
sions on Cu/Al (1.9) and Ag/Al (0.8) and comparable
to that on CuAg/Al (2.7)(7 : 3). It should be noted that
the amount of alumina in the physical mixture doubles that
in CuAg/Al (2.7)(7 : 3). Although it has no activity in the
amount of glycerol converted (mole)
Conversion (%) =
×100
2
O
3
2
O
3
total amount of glycerol in the feed (mole)
2
O
3
2
O
3
amount of glycerol converted to a product (mole)
amount of glycerol converted (mole)
2
O
3
Selectivity (%) =
×100
2
O
3
Yield (%) = [Conversion (%) ¥ Selectivity (%)]/100
absence of metals (as shown later in Fig. 6), alumina is not
solely an inert support in the cases of alumina supported metal
catalysts. It probably contributes to the glycerol conversion by
promoting the formation of reaction intermediates on its acid
sites, and also facilitating the hydrogen spillover discussed below.
3
. Results and discussion
3
.1 Investigation of catalyst composition
14,19,22
It has been proposed
that coordinatively unsaturated
Active metal selection. A series of g-Al
2
O
3
supported
copper species on the surface of Cu-based catalysts are the
active sites for the hydrogenolysis of glycerol. That may be the
reason that most of the Cu-based catalysts, such as Cu–ZnO,
copper chromite, and Cu/Al O , usually need to undergo pre-
bimetallic catalysts were prepared and tested without pre-
reduction treatment. One of the two metals was Cu, and
the other one was selected from Ag, Zn, Cr, Ni, or Co.
2
3
Two supported monometallic catalysts, Cu/Al
Ag/Al (0.8), were also prepared. These two catalysts and
their physical mixture were tested for comparison with the
CuAg/Al (2.7)(7 : 3) catalyst. The reaction results are shown
in Table 2.
The Cu-containing catalysts presented pretty high selectivity
to propanediols. This is in agreement with the literature, which
shows that copper-based catalysts are usually selective towards
2
O
3
(1.9) and
reduction in order to possess activity. Cu–ZnO oxide catalysts
showed some activity in the hydrogenolysis reaction without
2
O
3
14
a high temperature pre-reduction, but a high H2 pressure
(≥8 MPa) seems indispensable for the reaction. Under the
reaction conditions employed here, however, we believe that
such a reduction of CuO occurred significantly only when Ag
co-existed on the catalysts presented in Table 2. For Al O
2
O
3
2
3
supported metal catalysts reported in the literature, spillover
2
3
of activated hydrogen from Ag to MnO
2
and CoO and from
Table 2 Hydrogenolysis of glycerol on Al
2
O
3
supported bimetallic
24
Pd to CuO has been proposed to explain the reduction of the
metal oxides. In our case, we believe that under the mild reaction
conditions hydrogen was activated on the metallic Ag in the
catalysts
Selectivity (%)
Conv. (%) 1,2-PD 1,3-PD Others
CuAg/Al
2
3
O (2.7)(7 : 3) catalyst, and then spillover of hydrogen
Catalyst
M : G
occurred to the CuO to reduce it in situ into Cu sites that are
active for the reaction. This was confirmed by our TPR (Fig. 1)
and XPS (Fig. 2) results.
CuAg/Al
2
O
3
(2.7)(7 : 3) 3 : 100 27
96
94
92
83
94
94
0
0
2
0
0
0
4
6
6
17
6
CuZn/Al
CuCr/Al
2
O
3
(2.7)(7 : 3) 3 : 100
(2.7)(7 : 3) 3 : 100
6
1
2
O
3
The TPR pattern of Cu/Al
2
O displays a peak centered
3
Cu/Al
Ag/Al
Cu/Al
2
2
2
3
3
3
(1.9)
(0.8)
(1.9)
2.1 : 100 2
0.9 : 100 8
3 : 100 20
◦
at about 250 C, which corresponds to the reduction of the
◦
◦
6
CuO starting at about 200 C and finishing at about 260 C.
The pattern of the Ag/Al sample is similar to that of the
a
+
Ag/Al
3
(0.8)
2
O
3
a
Cu/Al
2
O
3
catalyst except that the reduction of the silver oxide
1
: 1 (wt/wt) physical mixture.
◦
is in a much lower temperature region, from 50 to 120 C. This
This journal is © The Royal Society of Chemistry 2010
Green Chem., 2010, 12, 1835–1843 | 1837

View Online
Apparently, the Ag species in the catalyst system promote the
reduction of CuO, and this promotion is more effective when
Ag and Cu are in close proximity. On the other hand, the TPR
patterns of CuZn/Al
2
O
3
(2.7)(7 : 3) and CuCr/Al
2
O (2.7)(7 : 3)
3
◦
showed major reduction peaks at about 210 C and broad peaks
◦
above 250 C. This means that the reduction of the CuO on
these two Cu-based catalysts requires more severe conditions
◦
(
temperature higher than 200 C and/or higher H
2
pressure).
The CuO on these catalysts was not (or not fully) reduced under
our reaction conditions. In other words, Zn, Cr, Ni and Co are
not as effective as Ag in terms of facilitating the reduction of
CuO on the Al
2
O support. Other researchers indicated that the
3
18
reduction temperature, needed to make Cu–ZnO and copper
25
chromite active for the hydrogenolysis of glycerol, should be
◦
above 200 C, and that these catalysts show poor activities when
used without a reduction pre-treatment.
XPS spectra of as-prepared, used, and hydrogen reduced
CuAg/Al
of the Cu/Al
2
O
3
catalysts are shown in Fig. 2. The XPS spectra
and Ag/Al catalysts are presented for
2
O
3
2
O
3
comparison. The analyses showed the oxidation states of those
near-surface species. The spin–orbit split peaks at 952.5 eV
2 3
Fig. 1 TPR patterns of Cu and other metal impregnated Al O :
(
2p1/2) and 932.3 eV (2p3/2) (Fig. 2A) are attributed to Cu(0)
(
(
a) Ag/Al
d) CuZn/Al
2 3
(1.9)+Ag/Al O
2
O
3
(2.0), (b) Cu/Al
(2.7)(7 : 3), (e) CuCr/Al
(0.8).
2
O
3
(2.0), (c) CuAg/Al
2 3
O (2.7)(7 : 3),
26,27
and Cu(I).
Differentiation of these two oxidation states is
2
O
3
2 3
O
(2.7)(7 : 3), and (f)
not possible solely based on the 2p signals because the binding
Cu/Al
2
O
3
26
energy difference is only 0.1 eV. Fully oxidized CuO species
are characterized by both a second set of spin–orbit split peaks,
shifted 1.3 ± 0.2 eV to higher binding energy (at 953.8 eV
(
2p1/2) and 933.5 eV (2p3/2)), and the appearance of characteristic
26,27
satellite peaks around 963 and 943 eV.
XPS spectra in Fig.
2
B displays the Ag 3d states of the catalysts. The spin–orbit split
26,28,29
Ag peaks are around 374 eV (3d3/2) and 368 eV (3d5/2).
Compared with the metallic Ag peaks, the peaks of Ag species
in the 1+ and 2+ (an average of 1+ and 3+) oxidation state are
26
shifted to lower binding energies by 0.3 and 0.7 eV, respectively.
Sometimes ranges of binding energies, 368.0–368.3, 367.6–367.8,
and 367.3–367.4 eV were reported for metallic Ag, Ag O, and
2
28
AgO, respectively.
Gaussian fitting of the Cu 2p binding energy region shows
that the surface of Cu/Al (1.9) contains only CuO. The
2
O
3
XPS spectrum of as-prepared CuAg/Al
calcined at 400 C in air also shows CuO as the only Cu
species. After one reaction cycle at 200 C in the presence of
2
O (2.7)(7 : 3) which was
3
◦
◦
Fig. 2 XPS spectra in the binding energy range of (A) Cu 2p and
glycerol and hydrogen, the satellite peaks (963 and 943 eV)
from the catalyst diminished significantly and the Cu 2p
signals were slightly shifted to lower binding energy (952.5
and 932.3 eV), indicating that CuO was mostly reduced and
that there was only a small amount of CuO left (7% based on
(
(
B) Ag 3d: (a) Cu/Al
2
O
3
(1.9), (b) as-prepared CuAg/Al
2
O
3
(2.7)(7 : 3),
(2.7)(7 : 3)
◦
c) CuAg/Al
2
O
3
(2.7)(7 : 3) used at 200 C, (d) CuAg/Al
2
O
3
◦
reduced in H
2
at 200 C, and (e) Ag/Al
2
O
3
(2.0).
means that any silver oxide on the Ag/Al
reduced to metallic silver under our reaction conditions. The
2
O
3
catalyst can be
deconvolution and peak area integration). The XPS spectrum
◦
of Cu species on CuAg/Al
with H
confirms that during reaction the bimetallic catalyst underwent
2
O
3
(2.7)(7 : 3) reduced at 200
C
TPR pattern of the CuAg/Al
major peak around 150 C, which could be assigned to the
reduction of the CuO to metallic copper. The reduction of oxidic
2
O
3
catalyst contains a broad
2
is almost identical to that of the used catalyst. This
◦
an in situ reduction, which is very similar to that in H
200 C.
2
at
◦
◦
species on this catalyst is complete at about 200 C. Therefore,
we conclude that a similar reduction has happened under our
The Ag 3d signals indicate that the Ag in the bimetallic
catalysts is in the metallic state, while the Ag/Al (2.0)
catalyst contains both Ag(0) and Ag(I). The amount of Ag(I)
is about 50% of the total Ag in Ag/Al (2.0) (according to
◦
reaction conditions (200 C, 3.6 MPa H
2
). It is clear that the Ag
2
O
3
on the CuAg/Al
2
3
O catalyst lowered the reduction temperature
of the copper oxide. Interestingly, the TPR pattern of the
2
O
3
physical mixture of Cu/Al
two distinct CuO reduction peaks at around 175 and 215 C.
2
O
3
(1.9) and Ag/Al
2
O
3
(0.8) showed
the XPS peak area). The observation of Ag(I) concurs with
our TPR characterization (Fig. 1). The formation of metallic
◦
1
838 | Green Chem., 2010, 12, 1835–1843
This journal is © The Royal Society of Chemistry 2010

View Online
Ag from the decomposition of Ag
2
O or AgO in air has been
Table 3 Effects of different supports on catalytic performance
investigated by Waterhouse et al. using XRD, FT-IR and Raman
spectroscopies. In the heating range of 100–200 C, they
30
◦
Selectivity (%)
observed the transformation of AgO (polycrystalline powder
Catalyst
Conv. (%) 1,2-PD 1,3-PD Others
◦
of silver(I,III) oxide) to Ag
and completely decomposed to Ag and O
2
O, which was stable up to 350 C
◦
CuAg/Al
CuAg/HY(2.7)(7 : 3)
CuAg/Hb(2.7)(7 : 3)
CuAg/HZSM-5(2.7)(7 : 3) nd
2
O
3
(2.7)(7 : 3)
27
2
2
96
80
89
—
0
2
1
—
4
2
at 400 C. For
18
10
—
Ag species on support, the states of Ag species may depend
on the loading of the Ag and the nature of the support.
a
31
Bethke and Kung found that Ag/Al
wt% Ag contained small Ag O particles and/or isolated Ag(I)
atoms after calcination at 700 C. Park and Boyer claimed
that Ag existed as well-dispersed Ag O on alumina support
when the loading was 2 wt%, while metallic Ag appeared at
wt% Ag loading. This coincides with our observation of both
metallic silver and silver oxide on Ag/Al (2.0). However, at
present we do not have solid data to rationalize the absence
of silver oxide on the CuAg/Al (2.7)(7 : 3) catalyst which
contains about 7 wt% Ag. A hypothesis is that the Cu species
in CuAg/Al (2.7)(7 : 3) facilitates the decomposition of Ag
by weakening the interaction between Ag O and the support or
by consuming the oxygen in Ag O through formation of CuO.
2
O
3
samples with 2 or
a
Not detected.
6
2
◦
32
was observed for the hydrogenolysis of glycerol (Table 3).
Among these supports, the Al stood out unambiguously.
The more acidic supports, HZSM-5, HY, and Hb, did not
generate catalysts with high activity at the same loading of Cu
and Ag. TPR characterizations presented in Fig. 4 show that
the reduction of the metals on these supports was delayed to
higher temperature when compared with the reduction on Al
For example, the reduction of the HZSM-5 supported CuAg
barely started until the temperature reached 250 C and finished
at about 400 C. No glycerol conversion could be detected
2
2
O
3
8
2
O
3
2
O
3
2
3
O .
2
O
3
2
O
◦
2
◦
2
on CuAg/HZSM-5(2.7)(7 : 3). We believe that a strong metal
support interaction due to the strong acidity and high surface
area of the supports (high metal dispersion) may hinder the
reduction of copper oxide. It is known that transition metals
Besides the ease of reduction, another contribution of Ag to
the improved performance of the Cu-based supported catalyst
could come from its capability to increase the dispersion of Cu
species, which correlates with the amount of active Cu species
available to the reactants. A comparison of the XRD patterns
33
on zeolites, when existing as framework metal cation or metal
3
4
silicate, are hard to reduce due to their strong coordination to
the oxygen atoms of the zeolite framework, especially on acidic
of Cu/Al
Although the CuAg/Al
loading as the Cu/Al
display CuO peaks, as observed in the Cu/Al
2
O
3
(1.9) and CuAg/Al
(2.7)(7 : 3) catalyst has the same Cu
(1.9) catalyst, its XRD pattern does not
(1.9) catalyst.
2
O
3
(2.7)(7 : 3) is made in Fig. 3.
2
O
3
3
3,35
zeolites.
2
O
3
2
O
3
It seems that the addition of Ag helped the formation of well-
dispersed Cu crystallites which were too small to be detected
by the XRD analysis. Other metals (Zn, Cr) seem to be able
to increase the Cu dispersion as well. No signals of Ag species
were observed in the XRD spectrum of CuAg/Al
2
3
O (2.7)(7 : 3).
Although the formation of Cu–Ag alloy cannot be completely
ruled out, the chance should be slim considering the low
calcination temperature and the oxidative environment.
Fig. 3 XRD patterns of Cu and other metal impregnated Al
Cu/Al (1.9), (b) CuAg/Al (2.7)(7 : 3), (c) CuZn/Al (2.7)(7 : 3),
and (d) CuCr/Al (2.7)(7 : 3).
2 3
O : (a)
Fig.
4
2 3
TPR patterns of (a) CuAg/Al O (2.7)(7 : 3), (b)
2
O
3
2
O
3
2 3
O
CuAg/HY(2.7)(7 : 3), (c) CuAg/Hb(2.7)(7 : 3), and (d) CuAg/HZSM-
2
O
3
5
(2.7)(7 : 3).
Since the combination of Cu with Ag presented the best re-
action results, the following studies were conducted on catalysts
based on these two metals.
The zeolites used in this study may also catalyze the de-
hydration of glycerol. Using FT-IR, Yada and Ootawa ob-
36
served the selective formation of acrolein on H-MFI zeolite.
Effects of different supports. Strong effect of supports on the
catalytic performance of supported CuAg bimetallic catalysts
However, the large amount of water in our reaction system
should limit the extent of acrolein formation. In addition, the
This journal is © The Royal Society of Chemistry 2010
Green Chem., 2010, 12, 1835–1843 | 1839

View Online
◦
Table 4 Effect of Cu/Ag ratio of CuAg/Al
2
O
3
of Ag O (2q = 32.8, 38.2 ) were observed in the XRD pattern
2
of CuAg/Al
2
O
3
(2.7)(0 : 100) (i.e. Ag/Al
2
O
3
(2.7)). In addition,
Selectivity (%)
◦
XRD signals of metallic Ag (2q = 38.2, 44.3 ) besides Ag
2
O
Catalyst
Cu/Al
CuAg/Al
CuAg/Al
CuAg/Al
CuAg/Al
CuAg/Al
CuAg/Al
CuAg/Al
Conv. (%) 1,2-PD 1,3-PD Others
was also detected on Ag/Al O (2.7), consistent with the XPS
2
3
observation in Fig. 2.
Reaction results on CuAg/Al
metal loadings (0.7 to 3.3 mmol g ) but a constant ratio of
Cu/Ag (7 : 3) are displayed in Fig. 6. The glycerol conversion
rose with increasing amount of supported active metals and
reached a maximum (27%) at 2.7 mmol metals per gram
2
O
3
(2.7)
3
84
99
97
96
96
95
94
93
93
0
0
0
0
0
0
0
0
0
16
1
2
O catalysts with different total
3
2
2
2
2
2
2
2
O
O
O
O
O
O
O
3
3
3
3
3
3
3
(2.7)(95 : 5)
21
-
1
(2.7)(85 : 15) 25
(2.7)(70 : 30) 27
(2.7)(50 : 50) 24
(2.7)(30 : 70) 25
(2.7)(15 : 85) 23
3
4
4
5
6
7
7
(2.7)(5 : 95)
23
21
of Al
2
3
O , after which the conversion decreased with further
Ag/Al
2
O
3
(2.7)
increase of the metal loading. We believe that a serious pore
blockage occurred at high metal loading, as indicated by the
rapid loss of the surface area of the catalysts. The surface
strong acidity of the zeolites may hinder the conversion of
acrolein, as demonstrated by Volckmar et al. in their study
of acrolein hydrogenation on silica/alumina supported silver
catalysts. Even if being hydrogenated, the product of acrolein
hydrogenation would be propionaldehyde, allylalcohol or n-
propanol rather than propanediols. All these factors contribute
to the low conversion of glycerol and low propanediol selectivity
on our zeolite-supported catalysts in Table 3.
2
-1
area of CuAg/Al
CuAg/Al
(2.7)(7 : 3) it was about 200 m g-1. Therefore,
2
O (3.3)(7 : 3) was about 160 m g and of
3
2
2
O
3
37
the total amount of accessible active metal sites could have
actually decreased although the total metal loading increased.
The selectivity toward 1,2-PD (no 1,3-PD) was almost constant
irrespective of the metal loading in the range studied.
Optimization of metal ratio and loading. CuAg/Al
2
3
O cat-
alysts which have the same total metal loading (2.7 mmol
of Cu and Ag) but various Cu to Ag molar ratios were
prepared and tested (Table 4). In general, the bimetallic catalysts
were more active than the monometallic ones. For example,
Cu/Al
replacement of 5% of the Cu by Ag, the glycerol conversion
increased sharply to 21% on CuAg/Al (2.7)(95 : 5). With
2
O
3
(2.7) only presented a glycerol conversion of 3%. With
2
O
3
increased activities on these bimetallic supported catalysts, the
selectivity to 1,2-PD increased to above 95% and little 1,3-
PD was observed. The maximal conversion of about 27% was
2 3
Fig. 6 Effect of total metal loading of CuAg/Al O on the formation
of propanediols from glycerol.
achieved on CuAg/Al
XRD patterns of CuAg/Al
to Ag ratios are presented in Fig. 5. The XRD pattern of
CuAg/Al (2.7)(100 : 0), i.e. Cu/Al (2.7), clearly shows the
2
O
3
(2.7)(70 : 30).
Through these studies, an optimized catalyst composition was
established, i.e. 2.7 mmol of Cu and Ag (molar ratio 7 : 3) per
2
O
3
catalysts with various Cu
gram of Al
2
3
O support.
2
O
3
2
O
3
existence of CuO crystallites on the catalyst surface. The signals
of CuO were strongly attenuated once a small amount of Ag
was added to the catalyst (Cu : Ag = 95 : 5). It is likely that
CuO was well dispersed into small crystallites even at this low
content of Ag. This supports the previous observation and
confirms that the activity increase could partially come from
good dispersion of CuO due to the Ag addition. Weak signals
3
.2 Study of catalyst preparation procedure
Metal impregnation sequence. Since CuAg/Al O (2.7)(7 : 3)
2
3
consists of two catalytically active components, i.e. Cu and Ag,
the effect of the metals impregnation sequence on the formation
of propanediols from glycerol was investigated (Table 5). The
catalyst prepared by simultaneously impregnating Cu and Ag
salts onto the support showed the highest activity, about 7–9%
higher than those on the other two catalysts prepared by loading
the two metals sequentially, no matter which one was first.
TPR characterizations (Fig. 7) showed that all three catalysts
◦
have a maximal hydrogen consumption at about 150 C.
Besides this peak, the two catalysts prepared by sequential
impregnations, especially Cu/Ag/Al
2
3
O (2.7)(7 : 3) in which Ag
was impregnated first, also showed a broad peak centered
◦
around 190 C, an indication that the promotion effect of Ag for
the reduction of CuO on these two catalysts is not as effective
as on the CuAg/Al
2
3
O (2.7)(7 : 3) catalyst. Instead of uniformly
mixing well with Cu in the case of (c) simultaneous, Ag particles
may sit on top of the CuO in the case of (a) Cu first, or be
buried by the CuO in the case of (b) Ag first. Depending on
the distance from the Ag, the CuO could be reduced at different
2 3
Fig. 5 XRD patterns of CuAg/Al O (2.7) catalysts with various
Cu/Ag molar ratios: (a) 100 : 0, (b) 95 : 5, (c) 70 : 30, (d) 30 : 70, (e) 5 : 95,
and (f) 0 : 100.
1
840 | Green Chem., 2010, 12, 1835–1843
This journal is © The Royal Society of Chemistry 2010

View Online
Table 5 Effect of Cu and Ag impregnation sequence
Selectivity (%)
1,2-PD
b
Catalyst
Impregnation sequence
Conv. (%)
1,3-PD
Others
Surface ratio of Cu/Ag
a
Ag/Cu/Al
Cu/Ag/Al
2
O
O
3
3
(2.7)(7 : 3)a
Cu first
Ag first
Simultaneous
20
18
27
93
93
96
0
0
0
7
7
4
52/48
51/49
65/35
2
(2.7)(7 : 3)
CuAg/Al
2
O
3
(2.7)(7 : 3)
a
◦
b
An extra calcination at 400 C between the two sequential impregnations. Based on XPS analysis.
on this catalyst. On the two catalysts made by sequential impreg-
nation, the Cu/Ag ratio is 52/48 for Ag/Cu/Al (2.7)(7 : 3)
and 51/49 for Cu/Ag/Al (2.7)(7 : 3), much lower than for
2
O
3
2
O
3
the simultaneously impregnated catalyst, which probably has
the highest Cu dispersion.
It is believed that the low dispersion of the CuO and poor
interaction between Cu and Ag resulted in the lower activities
of the sequentially impregnated catalysts. To help the reduction
and dispersion of the CuO, it seems vital for Ag and Cu to
be mixed well in the impregnation solution and to contact the
support at the same time so that they can be in a close vicinity.
Calcination temperature. The results in Fig. 9 illustrate a rise
and fall in glycerol conversion when the calcination temperature
◦
rose from 200 to 500 C. Maximal yield of 1,2-PD was achieved
◦
at 400 C due to the highest glycerol conversion and the highest
◦
propanediol selectivity. Hence, calcining the catalyst at 400 C
for 3 h was the optimal treatment condition for this catalyst for
the hydrogenolysis reaction.
2 3
Fig. 7 TPR patterns of Cu and Ag impregnated Al O with different
impregnation sequences: (a) Cu first, (b) Ag first, and (c) simultaneous.
temperatures. The Cu close to the Ag would be reduced first
and those relatively far away would have a delayed reduction.
This could be why the TPR patterns of (a) and (b) in Fig. 7
showed two peaks. XRD signals of CuO were detected for the
Cu/Ag/Al
2
O
3
(2.7)(7 : 3) catalyst only (Fig. 8). This means that
this catalyst has some CuO crystallites with relatively large sizes.
2 3
Fig. 9 Effect of calcination temperature of CuAg/Al O (2.7)(7 : 3).
Fig. 10 shows the XRD patterns of CuAg/Al
catalysts dried at 110 C, calcined at 200, 400, and 500 C. Weak
2
O
3
(2.7)(7 : 3)
◦
◦
◦
◦
signals assigned to Cu(NO
3
)
2
crystallites (2q = 13 and 26 ) were
◦
observed on the sample dried at 110 C and the one calcined at
◦
2
00 C. This means that the Cu(NO
3
)
2
salt did not decompose
crystallites (2q = 24,
32, 36 and 55 ) were detected on these two samples, probably due
◦
completely at 200 C or lower. No AgNO
3
◦
Fig. 8 XRD patterns of Cu and Ag impregnated Al
2
O
3
with different
◦
impregnation sequences: (a) Cu first, (b) Ag first, and (c) simultaneous.
to its high dispersion or easy decomposition at 200 C or lower.
The XRD patterns of the CuAg/Al
2
O (2.7)(7 : 3) catalysts
3
◦
The relative amounts of Cu and Ag on the surfaces of the
three catalysts were quantified with XPS analyses (Table 5).
calcined at 300, 400, and 500 C, as well as their surface areas
2
-1
(about 200 m g ), were very similar. Neither the crystal phase of
The surface Cu/Ag ratio on CuAg/Al
2
O
3
(2.7)(7 : 3) was 65/35,
Cu(NO
3
)
2
nor that of CuO was detected on these three samples.
◦
slightly lower than the bulk ratio of Cu/Ag (7/3) in the catalyst.
Considering that XPS is a semi-quantitative technique, we
believe that both metals were almost homogeneously dispersed
However, the samples calcined at 300 and 500 C were only
about 65% as active as the one calcined at 400 C. These three
catalysts were chosen for further studies with TPR and XPS.
◦
This journal is © The Royal Society of Chemistry 2010
Green Chem., 2010, 12, 1835–1843 | 1841

View Online
Fig. 10 XRD patterns of CuAg/Al
2
O
3
(2.7)(7 : 3) calcined in air at (a)
◦
◦
◦
◦
◦
1
10 C, (b) 200 C, (c) 300 C, (d) 400 C, and (e) 500 C.
As shown by the TPR results in Fig. 11, the maximal hydrogen
◦
◦
consumption on the 400 C calcined sample appears at 150 C.
On the other two samples that had similar catalytic activities,
Fig. 12 XPS patterns in the binding energy range of (A) Cu 2p and (B)
◦
Ag 3d from CuAg/Al O (2.7)(7 : 3) samples calcined in air at (a) 300 C,
2
3
◦
◦
◦
◦
the corresponding peaks are at about 160 C. The order of the
(b) 400 C, and (c) 500 C, and then all reduced in H at 200 C.
2
ease of reduction of these three catalysts was also confirmed
should not have completely decomposed into CuO. This was
confirmed by our study with thermal gravimetric analysis (result
by our XPS studies (Fig. 12). Before the spectra were taken,
◦
the three samples were pretreated in H
2
at 200 C for 3 h to
not shown here), which showed that most of the weight loss of
simulate the reducing environment during reaction. The XPS
characterization revealed the difference of the relative amounts
of low (0 or 1+) and high (2+) valence Cu species in the three
catalysts (Fig. 12A). After the reduction, the majority of Cu
◦
the 110 C pre-dried Cu/Al
2
O catalyst only occurred when the
3
◦
-1
◦
sample was heated in air at 10 C min to 400 C. Cu(OH)
and/or Cu(NO from the incomplete decomposition may be
2
)
3 2
◦
more difficult to reduce than CuO. For the sample calcined
species (93%) in the catalyst calcined at 400 C is Cu(0) or Cu(I),
◦
◦
at 500 C, some copper aluminate may have been generated
while the ones calcined at 300 or 500 C contain a much larger
39
on the catalyst surface. Bolt et al. studied the formation
of nickel, cobalt, copper, and iron aluminates from a- and
g-alumina-supported oxides using Rutherford backscattering
spectrometry, XRD, scanning electron microscopy, and UV-vis
portion of Cu(II) (25–28%) besides Cu(0) and Cu(I).
◦
diffuse reflectance spectroscopy. They found that even at 500 C
small CuAl
formed on the surface phase of each g-alumina grain. Copper
in the CuAl on Al is difficult to reduce. Plyasova et
2
O
4
crystallites, which were invisible to XRD, were
2
O
4
2
O
3
40
al. showed that the reduction of bulk copper aluminate was
significant only in the temperature range of 300 to 400 C and did
not reach 100% even at 700 C. The Ag was present as metallic
◦
◦
phase on all three samples after the H treatment (Fig. 12B).
2
It is obvious that the calcination temperature has a strong
influence on the interaction between Cu and Ag species and
on the interaction between the metals and the support. For the
binary oxide catalysts containing copper and silver, a proper
calcination temperature might yield a suitable environment for
easy hydrogen spillover from the Ag atoms to the Cu and/or a
less electron drawing effect from the support as well so that the
CuO could be easily reduced and stay in lower valences to act as
catalytic active sites.
Comparison with a commercial catalyst. We tested a com-
mercial copper chromite catalyst and compared its performance
with our best bimetallic catalyst under the same reaction
conditions (Fig. 13). The copper chromite, which with pre-
reduction usually shows the best performance among Cu-based
2 3
Fig. 11 TPR patterns of CuAg/Al O (2.7)(7 : 3) calcined in air at (a)
◦
◦
◦
3
00 C, (b) 400 C, and (c) 500 C.
Using X-ray absorption near-edge structure and extended
X-ray absorption fine structure, Wei and co-workers studied
4,15
Cu(NO
3
)
2
loaded aluminium oxide and showed that the prevail-
catalysts,
only had a glycerol conversion of 4% after 10 h
◦
ing Cu species (85%) on the sample calcined at 500 C for 1 h
was CuO, and that the remaining part of the Cu was present
as Cu(OH)
when used directly after calcination in air (copper chromite
(air)). Its activity increased to 13% after a reduction in H
2
38
◦
2
(15%) and as a negligible amount of Cu(NO
3
)
2
.
at 300 C (copper chromite (H
2
)), but is still much lower
◦
Therefore, the Cu(NO
3
)
2
on our 300 C, 3 h calcined sample
than the one of CuAg/Al
2
O
3
(2.7)(7 : 3), which was 27%. We
1
842 | Green Chem., 2010, 12, 1835–1843
This journal is © The Royal Society of Chemistry 2010

View Online
References
1
2
3
4
5
6
7
8
9
R. D. Cortright, M. Sanchez-Castillo and J. A. Dumesic, Appl. Catal.,
B, 2002, 39, 353–359.
G. Luo, S. Yan, M. Qiao, J. Zhuang and K. Fan, Appl. Catal., A,
2
004, 275, 95–102.
J. Chaminand, L. Djakovitch, P. Gallezot, P. Marion, C. Pinel and
C. Rosier, Green Chem., 2004, 6, 359–361.
M. Pagliaro, R. Ciriminna, H. Kimura, M. Rossi and C. D. Pina,
Angew. Chem., Int. Ed., 2007, 46, 4434–4440.
C. H. Zhou, J. N. Beltramini, Y. X. Fan and G. Q. Lu, Chem. Soc.
Rev., 2008, 37, 527–549.
A. Behr, J. Eilting, K. Irawadi, J. Leschinski and F. Lindner, Green
Chem., 2008, 10, 13–30.
I. Furikado, T. Miyazawa, S. Koso, A. Shimao, K. Kunimori and K.
Tomishige, Green Chem., 2007, 9, 582–588.
2 3
Fig. 13 Glycerol conversions on CuAg/Al O (2.7)(7 : 3) and copper
chromite at different reaction time.
J. Feng, H. Fu, J. Wang, R. Li and H. Chen, Catal. Commun., 2008,
9
, 1458–1464.
attribute the low activity of the copper chromite mainly to its
relatively low Cu dispersion. The propanediol selectivity of the
copper chromite catalyst is about 80%, not as good as that of
T. Miyazawa, S. Koso, K. Kunimori and K. Tomishige, Appl. Catal.,
A, 2007, 318, 244–251.
1
1
0 T. Miyazawa, S. Koso, K. Kunimori and K. Tomishige, Appl. Catal.,
A, 2007, 329, 30–35.
CuAg/Al
2
3
O (2.7)(7 : 3) either, which was above 95% consistently
1 M. Balaraju, V. Rekha, P. S. Sai Prasad, B. L. A. Prabhavathi
Devi, R. B. N. Prasad and N. Lingaiah, Appl. Catal., A, 2009, 354,
82–87.
through out the reaction. The reaction rate on the CuAg/Al
2
O
3
showed little change during the first 10 h but seemed to decrease
gradually after that. Detailed study about deactivation is under
way.
1
2 A. Alhanash, E. F. Kozhevnikova and I. V. Kozhevnikov, Catal. Lett.,
2
008, 120, 307–311.
1
1
1
3 L. Ma, D. He and Z. Li, Catal. Commun., 2008, 9, 2489–2495.
4 S. Wang and H. C. Liu, Catal. Lett., 2007, 117, 62–67.
5 M. A. Dasari, P. P. Kiatsimkul, W. R. Sutterlin and G. J. Suppes,
Appl. Catal., A, 2005, 281, 225–231.
4
. Conclusions
1
6 A. Perosa and P. Tundo, Ind. Eng. Chem. Res., 2005, 44, 8535–8537.
A g-Al
2
O
3
supported CuAg bimetallic catalyst can be used di-
17 A. Yin, X. Guo, W. Dai and K. Fan, Green Chem., 2009, 11, 1514–
1
516.
rectly without a reduction pretreatment for glycerol hydrogenol-
ysis to propanediols; it showed good performance under mild re-
action conditions. The TPR and XPS characterizations revealed
that the ease of the reduction of CuO on the supported bimetallic
catalysts correlated well with the catalytic activity. The formation
1
1
2
2
8 M. Balaraju, V. Rekha, P. S. Sai Prasad, R. B. N. Prasad and N.
Lingaiah, Catal. Lett., 2008, 126, 119–124.
9 L. Guo, J. Zhou, J. Mao, X. Guo and S. Zhang, Appl. Catal., A, 2009,
367, 93–98.
0 R. T. Yang, Adsorbents: Fundamentals and Applications, Wiley, New
York, 2003, p. 194.
0
+
of low valence Cu species (Cu or Cu ) is the key for high
activity. The addition of Ag to the Cu-based catalyst facilitated
the reduction of the Cu species that generates low valence Cu
species in situ under mild reaction conditions. XRD and XPS
results showed that Ag could also improve the dispersion of the
Cu species. The effects of Ag were not only dependent on the
impregnation order of Cu and Ag, but also on the calcination
temperature after the impregnation. When optimal amounts of
Cu and Ag (Cu/Ag molar ratio 7 : 3, 2.7 mmol Cu+Ag per
1 J. Zhou, Y. Zhang, X. Guo, A. Zhang and X. Fei, Ind. Eng. Chem.
Res., 2006, 45, 6236–6242.
22 D. G. Lahr and B. H. Shanks, J. Catal., 2005, 232, 386–394.
2
2
3 M. Luo, X. Yuan and X. Zheng, Appl. Catal., A, 1998, 175, 121–129.
4 T. Inui, H. Hara, T. Takeguchi and J. Kim, Catal. Today, 1997, 36,
5–32.
25 X. Huang, L. Ma and M. S. Wainwright, Appl. Catal., A, 2004, 257,
35–243.
2
2
2
2
6 R. Morrish and A. J. Muscat, Chem. Mater., 2009, 21, 3865–3870.
7 J. Batista, A. Pintar, D. Mandrino, M. Jenko and V. Martin, Appl.
Catal., A, 2001, 206, 113–124.
gram of g-Al
2
O
3
) were loaded simultaneously on the support
28 F. Xiao, H. Liu and Y. Lee, Bull. Korean Chem. Soc., 2008, 29(12),
◦
2368–2372.
and calcined at 400 C, a catalytic selectivity to propanediols
2
9 J. F. Weaver and G. B. J. Hoflund, J. Phys. Chem., 1994, 98, 8519–
of nearly 100% with a glycerol conversion of about 27% was
8
524.
◦
achieved at 200 C (1.5 MPa initial H
2
pressure, 10 h, M : G =
30 G. I. N. Waterhouse, G. A. Bowmaker and J. B. Metson, Phys. Chem.
Chem. Phys., 2001, 3, 3838–3845.
3
: 100) due to successfully suppressing the scission of the C–C
3
3
3
1 K. A. Bethke and H. H. Kung, J. Catal., 1997, 172, 93–102.
2 P. W. Park and C. L. Boyer, Appl. Catal., B, 2005, 59, 27–34.
3 R. Bul a´ nek, B. Wichterlov a´ , Z. Sobal’ık and J. Tich y´ , Appl. Catal.,
B, 2001, 31, 13–25.
4 S. Jong and S. Cheng, Appl. Catal., A, 1995, 126, 51–66.
5 A. Jones and B. D. McNicol, Temperature Programmed Reduction
for Solid Materials Characterization, in Chemical Industries, Vol. 24,
Marcel Dekker, New York, 1986, p. 30.
6 E. Yoda and A. Ootawa, Appl. Catal., A, 2009, 360, 66–70.
7 C. E. Volckmar, M. Brona, U. Bentrup, A. Martin and P. Claus, J.
Catal., 2009, 261, 1–8.
8 Y. Wei, H. Wang, Y. Yang and J. Lee, J. Phys.: Condens. Matter, 2004,
16, S3485–S3490.
9 P. H. Bolt, F. H. P. M. Habraken and J. W. Geus, J. Solid State Chem.,
bonds in glycerol, which often occurs on supported precious
metal catalysts. Compared with a commonly used commercial
copper chromite catalyst, the CuAg/Al
2
3
O catalyst had much
3
3
higher activity and did not need a reduction pretreatment.
Acknowledgements
3
3
The authors thank the Educational Department of Liaoning
Province, China for its financial support of this work (Grant#
L2010036). The authors also thank the State Key Laboratory
of Fine Chemicals at Dalian University of Technology for sup-
porting the catalyst characterization work (Grant# KF0703).
We would also like to acknowledge Prof. R. Prins (ETH) for
helpful discussion.
3
3
4
1
998, 135, 59–69.
0 L. M. Plyasova, T. M. Yur’eva, I. Y. Molina, T. A. Kriger, A. M.
Balagurov, L. P. Davydova, V. I. Zaikovskii, G. N. Kustova, V. V.
Malakhov and L. S. Dovlitova, Kinet. Catal., 2000, 41, 429–436.
This journal is © The Royal Society of Chemistry 2010
Green Chem., 2010, 12, 1835–1843 | 1843