Effect of Ag loading on Cu/Al2O3 catalyst in the production of 1,2-propanediol from glycerol

Article abstract of DOI:10.1016/j.apcata.2014.01.015

The vapor-phase hydrogenolysis of glycerol was performed at a gradient temperature and ambient hydrogen pressure over a commercial Cu/Al ₂O₃ catalyst modified with silver. Addition of Ag into Cu/Al₂O₃ catalysts was found to be efficient in inhibiting the decomposition of glycerol to produce ethylene glycol and gave a yield of 1,2-propanediol higher than that of the original Cu/Al₂O₃ without Ag. To minimize the ethylene glycol formation, the suitable loading of Ag was 1 wt.%. Since the Ag loading decreases the hydrogenation ability of the Cu/Al₂O₃ catalyst, 1 wt.% Ag-loaded Cu/Al ₂O₃ catalyst was placed on the upper layer of the fixed catalyst bed and the Cu/Al₂O₃ catalyst was placed on the bottom layer for achieving much higher 1,2-propanediol yield. The effect of temperatures of the top, the interlayer, and the bottom of the catalyst bed on the yield was also examined: a high 1,2-propanediol yield of 98.3%, which is the highest value under ambient H₂ pressure conditions, was achieved at a gradient temperature from 170 to 105 C and a glycerol concentration of 15 wt.%.

Full text of DOI:10.1016/j.apcata.2014.01.015

Applied Catalysis A: General 475 (2014) 63–68
Contents lists available at ScienceDirect
Applied Catalysis A: General
journal homepage: www.elsevier.com/locate/apcata
Effect of Ag loading on Cu/Al2O3 catalyst in the production of
1,2-propanediol from glycerol
Daolai Sun, Yasuhiro Yamada, Satoshi Sato∗
Graduate School of Engineering, Chiba University, Chiba 263-8522, Japan
a r t i c l e i n f o
a b s t r a c t
Article history:
The vapor-phase hydrogenolysis of glycerol was performed at a gradient temperature and ambient hydro-
gen pressure over a commercial Cu/Al2O3 catalyst modified with silver. Addition of Ag into Cu/Al2O3
catalysts was found to be efficient in inhibiting the decomposition of glycerol to produce ethylene glycol
and gave a yield of 1,2-propanediol higher than that of the original Cu/Al2O3 without Ag. To minimize
the ethylene glycol formation, the suitable loading of Ag was 1 wt.%. Since the Ag loading decreases the
hydrogenation ability of the Cu/Al2O3 catalyst, 1 wt.% Ag-loaded Cu/Al2O3 catalyst was placed on the
upper layer of the fixed catalyst bed and the Cu/Al2O3 catalyst was placed on the bottom layer for achiev-
ing much higher 1,2-propanediol yield. The effect of temperatures of the top, the interlayer, and the
bottom of the catalyst bed on the yield was also examined: a high 1,2-propanediol yield of 98.3%, which
is the highest value under ambient H2 pressure conditions, was achieved at a gradient temperature from
170 to 105 ◦C and a glycerol concentration of 15 wt.%.
Received 28 November 2013
Received in revised form
24 December 2013
Accepted 8 January 2014
Available online 15 January 2014
Keywords:
Glycerol
1,2-Propanediol
Cu/Al2O3 catalyst
Silver loading
© 2014 Elsevier B.V. All rights reserved.
1. Introduction
In the vapor phase, copper metal works in the formation of
1,2-PDO from glycerol at H2 atmosphere [13–18]. Zhu et al. also
The application of renewable biomass provides a facile route
to alleviate the shortage of fossil fuels and reduce CO2 emission
[1,2]. Glycerol, as a biomass derivate, is currently produced in
the biodiesel production process which brings a huge amount of
glycerol close to 10 wt.% of the overall biodiesel production [3–5].
Nowadays, a large surplus of crude glycerol is directly inciner-
ated, although the crude glycerol must be a potential raw material
[6]. Many kinds of value-added chemicals [1–5], such as acrolein
[6–9], lactic acid [10,11], 1,2-propanediol (1,2-PDO) [12–19], and
1,3-propanediol (1,3-PDO) [20–23], can be derived from glycerol.
1,2-PDO is a valuable chemical, which is mainly used for produc-
ing polymers [24]. In the liquid-phase processes, hydrogenolysis of
glycerol into 1,2-PDO proceeds over various metal catalysts such as
Rh [25,26], Ru [27,28], Ni [29,30], Pt [31,32], Ag [33], and Cu [12,19].
Huang et al. reported 98% selectivity to 1,2-PDO with a conversion
of ca. 82% over Cu/SiO2 catalyst at 180 ◦C and an H2 pressure of
6.0 MPa, while ethylene glycol (EG) as a by-product increases with
increasing the reaction temperature [12]. In the liquid phase, the
formation of 1-propanol is preferable over Cu/boehmite at 200 ◦C
and an H2 pressure of 4.0 MPa, so that the stepwise hydrogenolysis
of 1,2-PDO decreases the selectivity to 1,2-PDO at high conversion
[19].
reported 98% selectivity to 1,2-PDO with a complete conversion
over boron oxide-loaded Cu/SiO2 catalyst at 200 ◦C at an H2 pres-
sure of 5.0 MPa [18]. It is known that this reaction proceeds through
two steps: the first step is glycerol dehydration to produce hydrox-
yacetone (HA) [34,35], and the second step is HA hydrogenation to
produce 1,2-PDO [15,16]. In our previous research, we performed
the reaction at a gradient temperature, which was efficient for both
endothermic dehydration of glycerol at high temperature and the
exothermic hydrogenation of HA at low temperature: the selectiv-
ity to 1,2-PDO higher than 95% with a complete conversion was
obtained even at ambient pressure [15]. Under an H2 pressure
of 5.0 MPa at 220 ◦C, Ag supported on porous manganese oxide
exhibits the stable glycerol hydrogenolysis activity with the 1,2-
PDO selectivity above 65% at the glycerol conversion of 60% [33].
In the vapor-phase processes, a major by-product, EG, reduces
the 1,2-PDO selectivity [14–18,33], while another by-product, HA,
can be hydrogenated into 1,2-PDO because of the shift of equilib-
rium in the hydrogenation [14,15]. Furthermore, the reduction of
the EG yield is also important from the view of industrial applica-
tion because the separation of EG from 1,2-PDO is difficult due to
the similar properties of 1,2-PDO and EG. Therefore, inhibiting EG
formation is regarded as the key for achieving a complete yield of
1,2-PDO from glycerol. Recently, we have reported the dehydra-
tion of glycerol into HA over Ag/SiO2 catalyst: the selectivity to HA
is 91.1 mol% in H2 flow at 240 ◦C [36], while the selectivity to EG was
only 1.6%, which is much lower than 5.4% over Cu/Al2O3 without Ag
∗
Corresponding author. Tel.: +81 43 290 3376; fax: +81 43 290 3401.
E-mail address: satoshi@faculty.chiba-u.jp (S. Sato).
0926-860X/$ – see front matter © 2014 Elsevier B.V. All rights reserved.
http://dx.doi.org/10.1016/j.apcata.2014.01.015

64
D. Sun et al. / Applied Catalysis A: General 475 (2014) 63–68
at 230 ◦C [15]. It is expected that the modification of Cu/Al2O3 cata-
lyst with silver inhibits the generation of EG and increases 1,2-PDO
yield.
Shimadzu) was used for identification of the products in the efflu-
ent. The conversion and selectivity were calculated as follows [15]:
In this study, a commercial Cu/Al2O3 catalyst was modified
with Ag for the glycerol dehydration–hydrogenation reaction, and
optimum reaction conditions for the 1,2-PDO formation were also
studied at gradient temperatures and ambient hydrogen pressure.
sum of moles of all products
conversion(%) =
selectivity(%) =
× 100
mole of the reactant
moles of carbon in specific product
moles of carbon in all products
× 100
+H2
Cu
-H2O
HO
OH
HO
HO
Cu-Ag
OH
O
OH
2.3. Characterization of catalysts
2. Experimental
2.1. Sample preparation
Temperature-programmed reduction (TPR) measurements
were performed for characterizing metal state of the catalysts in a
mixed flow of H2/N2 (=1/9) at a flow rate of 10 cm3 min−1 at a heat-
ing rate of 5 ◦C min−1 from 25 to 900 ◦C, the details are described
elsewhere [15,37]. The XRD patterns of the samples were recorded
on a D8 ADVANCE (Bruker, Japan) using Cu K␣ radiation.
Glycerol and silver nitrate were purchased from Wako Pure
Chemical Industries Ltd. Cu/Al2O3 (N242) purchased from Nikki
Chemical Co., Ltd. has the CuO content of 55.1% and the surface area
of 118 m2 g−1 [15]. Supported silver catalysts were prepared by an
incipient wetness impregnation method using an aqueous solution
of silver nitrate as the precursor of Ag and N242 as a support. After
the impregnation process, the catalyst was dried at 110 ◦C for 12 h,
and finally calcined at 400 ◦C for 3 h. The percentage of Ag in the
Ag-loaded N242 catalyst was calculated based on the weight of Ag
metal. For comparison, unloaded N242 was also calcined at 400 ◦C
for 3 h prior to the reaction. The sample name is denoted by “N242-
x%Ag”, where x is weight percentage of Ag loaded on N242 catalyst.
For example, 1 wt.% Ag-containing N242 catalyst, hereafter denoted
as N242-1%Ag.
3. Results
3.1. Reaction of 1,2-PDO over different loading of Ag–Cu/Al2O3
catalysts
Prior to the temperature-gradient experiment, the reaction was
performed at a constant temperature of 200 ◦C and short contact
time in order to clarify the effect of Ag loading on the catalytic
performance of N242 catalyst. Table 1 demonstrates catalytic data
at conversion level lower than 100% using a catalyst of 1 g and
an aqueous glycerol solution with the concentration of 30 wt.% as
the reactant. The glycerol conversion decreased with increase in
Ag loading. In addition, the selectivity to 1,2-PDO was decreased
with increase in Ag loading while the selectivity to HA was signif-
icantly increased by Ag loading. The Ag loading on N242 catalyst
clearly indicates the decrease in the catalytic activities of N242 cat-
alyst for both the dehydration of glycerol and the hydrogenation of
HA. It should be noticed that the selectivity to EG was significantly
reduced by Ag loading.
Since the gradient temperature conditions have been proved
to be efficient in achieving high yields of 1,2-PDO even at ambi-
ent pressure in our previous study [15], the additive effect of Ag
into N242 catalyst was examined at the gradient temperatures
of 200–130 ◦C in the following section. Table 1 also shows the
results of glycerol hydrogenolysis over Ag-loaded N242 catalysts
with different Ag contents. Glycerol was completely converted in
all the reactions. 1,2-PDO was the major product while HA, EG, and
methanol were the main by-products with a small amount of 1-
propanol. The Ag-loaded catalysts show the selectivity to EG lower
2.2. Catalytic reaction
The catalytic reaction of glycerol was performed in a fix-bed
down-flow glass reactor with an inner diameter of 17 mm at ambi-
ent H2 pressure: a detail procedure is described elsewhere [15]. A
catalyst (weight, 8.7 g; volume, 7.2 cm3; height, 3.4 cm) was placed
in the catalyst bed. When two types of catalysts such as N242 and
N242-2%Ag catalysts were used, the Ag-loaded N242 catalyst was
placed over the N242 catalyst bed where the total weight of the cat-
alysts was 8.7 g. Prior to the reaction, the catalysts were reduced by
H2 at 250 ◦C for 1 h. After the temperature of the catalysts bed had
been kept at the prescribed temperature, an aqueous solution of
glycerol was fed through the top of the reactor at the liquid feed
rate of 1.32 cm−3 h−1 with an H2 flow of 360 cm−3 min−1. The liq-
uid effluents collected in a dry ice-acetone trap (−78 ◦C) every hour
were analyzed by a FID–GC (GC-2014, Shimadzu) with a 30 m cap-
illary column of TC-WAX (GL-Science, Japan). A GC–MS (QP5050A,
Table 1
Effect of Ag loading over N242 on the catalytic reaction of glycerola.
Selectivity/%b
Conversion/%b
Ag content/%
1,2-PDO
HA
EG
1-Propanol
0c
94.3
93.2
84.1
100
100
100
100
100
100
70.5
69.9
59.9
94.1
93.8
96.2
95.7
95.1
92.7
23.7
25.9
35.6
1.2
1.0
1.2
1.1
1.2
3.2
3.6
2.1
2.8
3.4
2.8
1.6
1.7
2.3
2.7
0.2
0.1
0.1
0.1
0.1
0.1
0.1
0.1
0
1.0c
5.0c
0d
0.5d
1.0d
1.5d
2.0d
5.0d
a
Reaction conditions: feed rate, 1.32 cm3 h−1; H2 flow rate, 360 cm3 min−1; H2/glycerol = 700. An aqueous solution of glycerol at a concentration of 30 wt.% was used as
the reactant.
b
Average activity between 1 and 5 h.
c
Reaction temperature, 200 ◦C; catalyst weight, 1.0 g (0.4 cm).
Gradient temperature, 200 to 130 ◦C.; catalyst weight, 8.7 g (3.4 cm).
d

D. Sun et al. / Applied Catalysis A: General 475 (2014) 63–68
65
100
95
90
85
80
75
25
CuO
a
b
c
N242-5%Ag
20
15
10
5
1,2-PDO
EG
N242-2%Ag
N242-1%Ag
N242
HA
0
20
30
40
50
60
70
1
2
3
4
5
Time on stream/ h
2θ / degree
100
95
90
85
80
75
25
20
15
10
5
Fig. 2. XRD patterns of N242 and Ag-loaded N242 catalysts.
Cu
N242-5%Ag-600 oC
1,2-PDO
EG
N242-5%Ag-250 oC
HA
N242-1%Ag-600 oC
N242-1%Ag-250 oC
N242-250 oC
0
1
2
3
4
5
Time on stream/ h
100
95
90
85
80
75
25
20
15
10
5
20
30
40
50
60
70
2θ / degree
Fig. 3. XRD patterns of N242 and Ag-loaded N242 catalysts reduced at different
1,2-PDO
EG
temperatures.
HA
N242-5%Ag
N242-2%Ag
N242-1%Ag
N242
0
1
2
3
4
5
Time on stream/ h
Fig. 1. Changes in selectivity with time on stream over (a) N242, (b) N242-1%Ag,
and (c) N242-5%Ag. Reaction conditions: gradient temperature, 200 to 130 ◦C; feed
rate, 1.32 cm3 h−1; H2 flow rate, 360 cm3 min−1; H2/glycerol = 700; catalyst weight,
8.7 g (3.4 cm). An aqueous solution of glycerol at a concentration of 30 wt.% was used
as the reactant.
0
200
400
600
800
Temperature/ oC
Fig. 4. TPR profiles of N242 and Ag-loaded N242 catalysts.
than that of N242. The selectivity to EG decreased with increasing
the loading of Ag at Ag loading lower than 1 wt.%. The lowest selec-
tivity to EG of 1.6% was obtained over N242-1%Ag catalyst, which
gave the highest 1,2-PDO selectivity of 96.2%.
diffraction patterns of N242 and no peak attributed to Ag and/or
AgO was detected. The loaded Ag is not agglomerated on the N242
catalyst because only peaks of CuO were observed. Fig. 3 shows the
XRD patterns of Ag-loaded N242 catalysts reduced at different tem-
peratures for 1 h. N242 reduced at 250 ◦C shows typical diffraction
peaks of Cu at 2ꢀ = 43.5 and 50.1◦ [JCPDS file 4-0836]. The diffrac-
tion patterns of N242-1% Ag and N242-5%Ag reduced at 250 ◦C are
the same as those of reduced N242. Even at a high reduction tem-
perature of 600 ◦C, the Ag-loaded samples had no diffraction peaks
attributed to Ag metal, while they show higher crystallinity of Cu
than those reduced at 250 ◦C.
Fig. 4 shows TPR profiles of N242 and Ag-loaded N242 catalysts
with different Ag loading. N242 had two reduction peaks at 210 ◦C
and 215 ◦C. The reduction peaks of 0, 1, 2, and 5 wt.% Ag-loaded
N242 catalysts are observed at 210, 207, 199, and 198 ◦C, respec-
tively. The reduction peak shifts from the high temperature to the
low temperature with increasing Ag loading.
Fig. 1 shows the changes in the catalytic activities of N242 and
Ag-loaded N242 catalysts with time on stream. It is obvious that
the decrease in the selectivity to EG and the increase in the selec-
tivity to 1,2-PDO are achieved by 1 wt.% Ag loaded on N242 (Fig. 1a
and b). However, at Ag loading higher than 1 wt.%, the selectiv-
ity to 1,2-PDO decreased and the selectivity to EG increased with
increasing the Ag loading. The selectivity to HA increases to 3.2%
and the selectivity to 1,2-PDO decreases to 92.7% at Ag content of
5 wt.%. The changes in the catalytic activity with time on stream
over N242-5%Ag is shown in Fig. 1c. The increase of the selectivity
to HA is observed after 3 h of the reaction.
Fig. 2 shows the XRD patterns of modified N242 catalysts
with different Ag loading. The unloaded N242 without Ag showed
typical diffraction peaks of CuO at 2ꢀ = 35.5 and 38.7◦ [JCPDS
file 5-0661]. All the Ag-loaded N242 catalysts showed the same

66
D. Sun et al. / Applied Catalysis A: General 475 (2014) 63–68
Table 2
Effect of calcination temperature of N242-1%Aga.
Selectivity/%b
Conversion/%b
Calcined temperature/◦C
1,2-PDO
HA
EG
1-Propanol
400
500
600
100
100
100
96.2
95.3
92.0
1.2
1.0
1.0
1.6
1.7
1.8
0.1
0.2
1.1
a
Reaction conditions: gradient temperature, 200 to 130 ◦C; feed rate, 1.32 cm3 h−1; H2 flow rate, 360 cm3 min−1; H2/glycerol = 700; catalyst weight, 8.7 g (3.4 cm). An
aqueous solution of glycerol at a concentration of 30 wt.% was used as the reactant.
b
Average activity between 1 and 5 h.
3.2. Effect of calcination temperature of N242-1%Ag catalysts
EG selectivity was similar to that of single-layered N242-1%Ag cat-
Table 2 summarizes the catalytic performance of N242-1%Ag
alyst bed (Entry 1). In Entry 2, the EG selectivity was higher than
catalysts calcined at different temperatures. The conversion of 1,2-
that of Entry 1.
PDO is 100% in these reactions. The calcination temperature clearly
The reaction was also performed at a gradient temperature from
affected HA and EG formation. The selectivities to HA and EG
a top temperature of 200 ◦C to a low bottom temperature of 120 ◦C,
increased with increasing the calcination temperature while the
and the selectivity to HA decreased to 0.5% and the selectivity to
selectivity to 1,2-PDO decreased. 1-Propanol and some unidentified
1,2-PDO significantly increased to 96.8% (Entry 4). However, the
products were formed over the catalysts calcined at high tempera-
conversion decreased to 99.7% and the HA selectivity increased to
tures.
1.8% at the gradient temperature from 190 to 130 ◦C (Entry 5).
Fig. 5 shows the XRD patterns of N242-1%Ag calcined at differ-
ent temperatures. The intensity of the diffraction peaks of CuO in
3.4. Reaction using aqueous glycerol solution at a concentration
N242-1%Ag increased with increasing the calcination temperature.
of 15 wt.% over double-layered catalysts
The crystallinity of CuO increases with increasing the calcination
temperature. This tendency indicates that large particles of CuO
When an aqueous glycerol solution at the concentration of
were formed at high temperatures.
30 wt.% was used as the reactant (Entry 5, Table 3), the HA hydro-
genation to 1,2-PDO was not completed and the conversion of
3.3. Hydrogenolysis of glycerol over double-layered catalysts
glycerol decreased at a top temperature of 190 ◦C. Thus, further
study for achieving high 1,2-PDO yield was performed using an
Since the addition of Ag decreased the hydrogenation ability, the
aqueous glycerol solution at a concentration of 15 wt.%. Table 4
selectivity to 1,2-PDO decreased and that to HA increased (5 wt.%
shows the catalytic reaction results. The selectivity to 1,2-PDO was
Ag loading, Table 1). Catalytic reactions, in which N242-1%Ag in the
97.5% with a complete conversion at a gradient temperature from
bottom layer of the catalyst bed was replaced by N242 catalyst with
185 ◦C to 120 ◦C (Entry 1). The selectivity to EG slightly decreased
powerful hydrogenation ability, was performed to shift the equilib-
with decreasing the top temperature, and the selectivity to 1,2-PDO
rium to the 1,2-PDO side in the hydrogenation. The effect of reaction
was 98.0% at a gradient temperature from 170 to 120 ◦C (Entry 2).
temperatures in the top, the interlayer, and the bottom position of
The selectivity to HA was only 0.2% and the selectivity to 1,2-PDO
the catalyst bet on the selectivity to 1,2-PDO was also examined
increased to 98.3% in the reaction performed at a gradient temper-
(Table 3). In the single-layered N242-1%Ag catalyst at the same gra-
ature from 170 ◦C to a low bottom temperature of 105 ◦C (Entry
dient temperature (Entry 1 of Table 3), the selectivity to 1,2-PDO
3). Fig. 6 shows the changes of the catalytic activity with time on
and EG were 96.2 and 1.6%, respectively, at a gradient tempera-
stream: the catalytic activity was stable during the initial 5 h we
ture from 200 to 130 ◦C. In the double-layered catalyst bed, where
tested. For comparison, the reaction at a gradient temperature from
the total weight of N242-1%Ag and N242 was 8.7 g, the interlayer
170 to 120 ◦C over the unloaded N242 catalyst was performed: the
temperature was controlled by the charged amounts of N242-1%Ag
selectivity to 1,2-PDO was 96.0% (Entry 4), which was lower than
and N242. The selectivity to 1,2-PDO and EG were 95.5 and 2.1%,
98.3% over double-layered catalysts (Entry 3).
respectively, at a gradient temperature from 200 to 130 ◦C with an
interlayer temperature of 185 ◦C (Entry 2). The selectivity to the by-
product EG decreased with decreasing the interlayer temperature:
when the interlayer temperature decreased to 170 ◦C (Entry 3), the
100
95
90
10
5
CuO
1,2-PDO
EG
HA
N242-1%Ag-600 oC
N242-1%Ag-500 oC
N242-1%Ag-400 oC
0
1
2
3
4
5
Time on stream/ h
N242
Fig. 6. Changes in selectivity with time on stream over double-layered catalysts at a
gradient temperature from 170 ◦C to 105 ◦C (Table 4, Entry 3). Reaction conditions:
feed rate, 1.32 cm3 h−1; H2 flow rate, 360 cm3 min−1; H2/glycerol = 1400; catalyst
weight, 8.7 g (3.4 cm). An aqueous solution of glycerol at a concentration of 15 wt.%
was used as the reactant.
20
30
40
50
60
70
2θ / degree
Fig. 5. XRD patterns of N242 and N242-1%Ag calcined at different temperatures.

D. Sun et al. / Applied Catalysis A: General 475 (2014) 63–68
67
Table 3
Hydrogenolysis of glycerol over double-layered catalystsa.
Entry
Temperature/◦C
Conversion/%b
Selectivity/%b
1,2-PDO
Catalyst (g)
Top
Bottom
HA
EG
1
2
A(8.7)
A(1.9)
B(6.8)
A(3.7)
B(5.0)
A(3.3)
B(5.4)
A(3.6)
B(5.1)
200
200
185
200
170
200
170
190
165
130
185
130
170
130
170
120
165
130
100
100
96.2
95.5
1.2
1.0
1.6
2.1
3
4
5
100
100
99.7
96.1
96.8
96.1
1.0
0.5
1.8
1.7
1.6
1.5
a
Reaction conditions: feed rate, 1.32 cm3 h−1; H2 flow rate, 360 cm3 min−1; H2/glycerol = 700; catalyst weight, 8.7 g (3.4 cm). An aqueous solution of glycerol at a concen-
tration of 30 wt.% was used as the reactant. Catalyst A and B are N242-1%Ag and N242, respectively.
b
Average activity between 1 and 5 h.
Table 4
Hydrogenolysis of glycerol at a concentration of 15 wt.% over double-layered catalystsa.
Conversion/%b
Entry
Temperature/◦C
Bottom
Selectivity/%b
Catalyst
Top
1,2-PDO
97.5
HA
0.5
EG
1.6
1
2
3
A(3.5 g)
B(5.2 g)
A(3.5 g)
B(5.2 g)
A(3.5 g)
B(5.2 g)
B(8.7 g)
185
164
170
150
170
144
170
164
120
150
120
144
105
105
100
100
100
100
98.0
98.3
96.0
0.5
0.2
0.2
1.4
1.4
3.0
4
a
Reaction conditions: feed rate, 1.32 cm3 h−1; H2 flow rate, 360 cm3 min−1; H2/glycerol = 1400; catalyst weight, 8.7 g (3.4 cm). An aqueous solution of glycerol at a
concentration of 15 wt.% was used as the reactant. Catalyst A and B are N242-1%Ag and N242, respectively.
b
Average activity between 1 and 5 h.
4. Discussion
the interaction between Cu and Ag. Our previous work elucidates
that EG is formed by C C bond cleavage of glycerol over Cu metal
at high temperatures [15,16]. Thus, we propose that the loading
of Ag in N242 alters the chemical and electronic properties of the
surface, which reduces the cracking ability of Cu in N242 and thus
inhibits the EG formation. However, it is difficult to explain why the
EG formation increases at high Ag loading. We speculate that high
Ag loading results in agglomeration of Ag into large Ag particles,
which could be ineffective for inhibiting the EG formation because
of small interaction between Cu and Ag. Although we observed
the samples using TEM, there is no significant difference in the
morphologies observed in TEM images among the Ag-loaded and
unloaded N242 catalysts (TEM images not shown). This is proba-
bly caused by the large amount of Cu in the catalyst compared to
1 wt.% Ag: Ag can be solved or absorbed in Cu particles because
Ag metal has the same crystal structure of face-centered cubic as
Cu.
The high Ag loading leads an obvious increase of the HA forma-
tion (N242-5%Ag, Table 1 and Fig. 1c). Since Ag has hydrogenation
ability lower than that of Cu [36], it is reasonable that high Ag load-
ing decreases the hydrogenation ability of the catalyst, especially,
under ambient H2 pressure conditions. The hydrogenation of HA,
as the intermediate of 1,2-PDO, cannot proceed sufficiently when
the Ag loading is high, that results in the increase in the selectivity
to HA.
4.1. Effect of the loading of Ag on the catalytic activity of Cu/Al2O3
Bimetallic catalysts have been found to be effective for many
kinds of reactions such as CO oxidation over TiO2 supported Au–Cu
bimetallic catalysts [38]. Cu–Ag bimetallic catalyst was found to
be effective for the epoxidation of propylene [39] and ethylene
[40–42]. The synergistic effect between Ag and Cu was found to be
beneficial to produce more active sites where electrophilic oxygen
species can be absorbed and increase the propylene oxide selectiv-
ity [39]. DFT calculations also proved that Cu in Cu–Ag bimetallic
surfaces altered the chemical and electronic properties of the sur-
face, which affected the ethylene oxide formation [41]. Cu–Ag
bimetallic catalysts were also applied for the glycerol hydrogenol-
ysis in liquid phase [43]. A high 1,2-PDO selectivity of 99% with
a low conversion of 21% was obtained over alumina supported
Cu–Ag with a mole ratio of 95:5 at 200 ◦C under an H2 pressure of
1.5 MPa. The maximum conversion of glycerol was 27% at a Cu/Ag
mole ratio of 7:3, whereas the selectivity to 1,2-PDO decreased to
96%. It was found that the addition of Ag to the Cu-based cata-
lyst facilitates the reduction of the Cu species that generates low
valence Cu species in situ under mild reaction conditions [43].In
this study, the comparison of catalytic activity at a constant tem-
perature clarifies the effect of Ag in the N242 catalyst (Table 1).
The role of Ag loaded on N242 is to reduce the ability of Cu for
the decomposition of glycerol into EG with sacrificing the ability of
Cu for both the dehydration of glycerol and the hydrogenation of
HA into 1,2-PDO. Since no diffraction peak attributed to Ag in the
samples was observed in XRD patters of Figs. 2 and 3, Ag seems
to be highly dispersed on N242. The TPR results that the reduc-
tion peak shifts from the high temperature to low temperature
side with increasing Ag loading (Fig. 4) indicate the existence of
Calcination temperature also affects the catalytic property of
Ag-loaded N242 catalyst (Table 2). Large CuO particles were gener-
ated at high calcination temperatures, which agrees with the results
reported by Gu et al. [44]. Large Cu particles formed from the reduc-
tion of CuO would not be efficient for 1,2-PDO formation because
of low specific surface area of Cu metal. Therefore, we judge that
the calcination temperature of 400 ◦C is suitable for the 1,2-PDO
formation.

68
D. Sun et al. / Applied Catalysis A: General 475 (2014) 63–68
4.2. Hydrogenolysis of glycerol over double-layered catalysts
ability. Because the selectivity to cracking products is low at low top
temperatures and the hydrogenation of HA prefers the low bottom
temperatures, the highest 1,2-PDO yield of 98.3%, was obtained at
ambient hydrogen pressure and a low gradient temperature from
170 ◦C to 105 ◦C using aqueous glycerol solution at the concentra-
tion of 15 wt.%.
The selectivity to EG is 2.1% in the double-layered catalysts at a
gradient temperature from 200 to 130 ◦C with an interlayer tem-
perature of 185 ◦C (Entry 2 in Table 3), which is a little higher than
1.6% over N242-1%Ag in the single-layered catalyst at the same gra-
dient temperature (Entry 1 in Table 3). The interlayer temperature
could be still high for the inhibition of the EG formation through the
decomposition of glycerol. The selectivity to EG decreased to 1.7%
at an interlayer temperature of 170 ◦C (Entry 3 in Table 3). This indi-
cates that EG would be generated at high temperatures, and that the
unloaded N242 decomposes much glycerol to EG at temperatures
between 170 and 185 ◦C. The selectivity to HA decreased to 0.5% at
a gradient temperature from 200 ◦C to a low bottom temperature of
120 ◦C (Entry 4 in Table 3), which indicates the exothermic hydro-
genation of HA to 1,2-PDO prefers low reaction temperatures and
agrees with our previous reports [15,16]. The conversion decreased
to 99.7% and the HA selectivity increased to 1.8% at the top tem-
perature of 190 ◦C (Entry 5 in Table 3). This means that the top
temperature must be higher than 190 ◦C for keeping the complete
glycerol conversion to HA when an aqueous glycerol solution with
the concentration of 30 wt.% was used as the reactant.
References
[1] G.W. Huber, S. Iborra, A. Corma, Chem. Rev. 106 (2006) 4044–4098.
[2] A. Corma, S. Iborra, A. Velty, Chem. Rev. 107 (2007) 2411–2502.
[3] M. Pagliaro, R. Ciriminna, H. Kimura, M. Rossi, C. Della Pina, Angew. Chem. Int.
Edit. 46 (2007) 4434–4440.
[4] Y. Nakagawa, K. Tomishige, Catal. Sci. Technol. 1 (2011) 179–190.
[5] C.H. Zhou, H. Zhao, D.S. Tong, L.M. Wu, W.H. Yu, Catal. Rev. Sci. Eng. 55 (2013)
369–453.
[6] A. Ulgen, W.F. Hoelderich, Appl. Catal. A: Gen. 400 (2011) 34–38.
[7] E. Tsukuda, S. Sato, R. Takahashi, T. Sodesawa, Catal. Commun. 8 (2007)
1349–1353.
[8] M.H. Haider, N.F. Dummer, D. Zhang, P. Miedziak, T.E. Davies, S.H. Taylor, D.J.
Willock, D.W. Knight, D. Chadwick, G.J. Hutchings, J. Catal. 286 (2012) 206–213.
[9] K. Omata, S. Izumi, T. Murayama, W. Ueda, Catal. Today 201 (2013) 7–11.
[10] D. Ray, B. Subramaniam, R.V. Chaudari, D. Roy, ACS Catal. 1 (2011) 548–551.
[11] R.K.P. Purushothaman, J. van Haveren, D.S. van Es, I. Melián-Cabrera, J.D.
Meeldijk, H.J. Heeres, Appl. Catal. B: Environ. 147 (2014) 92–100.
[12] Z. Huang, F. Cui, H. Kang, J. Chen, X. Zhang, C. Xia, Chem. Mater. 20 (2008)
5090–5099.
The concentration of aqueous glycerol solution is also an impor-
tant parameter in the glycerol hydrogenolysis into 1,2-PDO. High
concentration of glycerol always leads to low glycerol conversion
and low 1,2-PDO selectivity in the liquid-phase reaction [24,45,46].
The decrease in the glycerol conversion and the increase in the
selectivity to 1,2-PDO (Entry 4 in Table 3) indicate that the active
sites are not enough at the low gradient temperatures from 190
to 130 ◦C when the concentration of glycerol is 30 wt.%. At a low
glycerol concentration of 15 wt.%, the low top temperature is effec-
tive in inhibiting the cracking products such as EG and methanol,
and the low bottom temperature is effective for the HA hydrogena-
tion to 1,2-PDO, the highest 1,2-PDO yield of 98.3% was obtained
over double-layered catalysts at a gradient temperature from 170
to 105 ◦C (Entry 3 in Table 4). On the other hand, the yield of 1,2-
PDO is at most 96% over the unloaded N242 at the same reaction
conditions (Entry 4 in Table 4), which demonstrates the superior-
ity of the Ag-loaded Cu/Al2O3 catalyst on the inhibition of the EG
formation.
[13] C.W. Chiu, A. Tekeei, W.R. Sutterlin, J.M. Ronco, G.J. Suppes, AIChE J. 54 (2008)
2456–2463.
[14] C.W. Chiu, A. Tekeei, J.M. Ronco, M.L. Banks, G.J. Suppes, Ind. Eng. Chem. Res.
47 (2008) 6878–6884.
[15] M. Akiyama, S. Sato, R. Takahashi, K. Inui, M. Yokota, Appl. Catal. A: Gen. 371
(2009) 60–66.
[16] S. Sato, M. Akiyama, K. Inui, M. Yokota, Chem. Lett. 38 (2009) 560–561.
[17] A. Bienholz, H. Hofmann, P. Claus, Appl. Catal. A: Gen. 391 (2011) 153–157.
[18] S. Zhu, X. Gao, Y. Zhu, Y. Zhu, H. Zheng, Y. Li, J. Catal. 303 (2013) 70–79.
[19] Z. Wu, Y. Mao, M. Song, X. Yin, M. Zhang, Catal. Commun. 32 (2013) 52–57.
[20] L. Gong, Y. Lu, Y. Ding, R. Lin, J. Li, W. Dong, T. Wang, W. Chen, Appl. Catal. A:
Gen. 390 (2010) 119–126.
[21] Y. Nakagawa, X. Ning, Y. Amada, K. Tomishige, Appl. Catal. A: Gen. 433–434
(2012) 128–134.
[22] S. Zhu, X. Gao, Y. Zhu, Y. Zhu, X. Xiang, C. Hu, Y. Li, Appl. Catal. B: Environ.
140–141 (2013) 60–67.
[23] R. Arundhathi, T. Mizugaki, T. Mitsudome, K. Jitsukawa, K. Kaneda, Chem-
SusChem 6 (2013) 1345–1347.
[24] M. Balaraju, V. Rekha, P.S. Sai Prasad, B.L.A. Prabhavathi Devi, R.B.N. Prasad, N.
Lingaiah, Appl. Catal. A: Gen. 354 (2009) 82–87.
[25] J. Chaminand, L.A. Djakovitch, P. Gallezot, P. Marion, C. Pinel, C. Rosier, Green
Chem. 6 (2004) 359–361.
[26] I. Furikado, T. Miyazawa, S. Koso, A. Shimao, K. Kunimori, K. Tomishige, Green
Chem. 9 (2007) 582–588.
[27] D.G. Lahr, B.H. Shanks, J. Catal. 232 (2005) 386–394.
[28] T. Miyazawa, Y. Kusunoki, K. Kunimori, K. Tomishige, J. Catal. 240 (2006)
213–221.
[29] A. Perosa, P. Tundo, Ind. Eng. Chem. Res. 44 (2005) 8535–8537.
[30] L. Huang, Y. Zhu, H. Zheng, Y. Li, Z. Zeng, J. Chem. Technol. Biotechnol. 83 (2008)
1670–1675.
[31] E.P. Maris, R.J. Davis, J. Catal. 249 (2007) 328–337.
[32] Z. Yuan, P. Wu, J. Gao, X. Lu, Z. Hou, X. Zheng, Catal. Lett. 130 (2009) 261–265.
[33] G.D. Yadav, P.A. Chandan, D.P. Tekale, Ind. Eng. Chem. Res. 51 (2012) 1549–1562.
[34] C.W. Chiu, M.A. Dasari, G.J. Suppes, W.R. Sutterlin, AIChE J. 52 (2006)
3543–3548.
[35] S. Sato, M. Akiyama, R. Takahashi, T. Hara, K. Inui, M. Yokota, Appl. Catal. A: Gen.
347 (2008) 186–191.
[36] S. Sato, D. Sakai, F. Sato, Y. Yamada, Chem. Lett. 41 (2012) 965–966.
[37] T. Nakayama, N. Ichikuni, S. Sato, F. Nozaki, Appl. Catal. A: Gen. 158 (1997)
185–199.
[38] A. Sandoval, C. Louis, R. Zanella, Appl. Catal. B: Environ. 140–141 (2013)
363–377.
[39] X. Zheng, Q. Zhang, Y. Guo, W. Zhan, Y. Guo, Y. Wang, G. Lu, J. Mol. Catal. A:
Chem. 357 (2012) 106–111.
[40] J.T. Jankowiak, M.A. Barteau, J. Catal. 236 (2005) 366–378.
[41] J.T. Jankowiak, M.A. Barteau, J. Catal. 236 (2005) 379–386.
[42] J.C. Dellamorte, J. Lauterbach, M.A. Barteau, Catal. Today 120 (2007) 182–185.
[43] J. Zhou, L. Guo, X. Guo, J. Mao, S. Zhang, Green Chem. 12 (2010) 1835–1843.
[44] C. Gu, G. Li, Y. Hu, S. Qing, X. Hou, Z. Gao, J. Fuel Chem. Technol. 40 (2012)
1328–1335.
[45] C.V. Rode, A.A. Ghalwadkar, R.B. Mane, A.M. Hengne, S.T. Jadkar, N.S. Biradar,
Org. Process Res. Dev. 14 (2010) 1385–1392.
[46] X. Guo, A. Yin, X. Guo, X. Guo, W. Dai, K. Fan, Chin. J. Chem. 29 (2011) 1563–1566.
5. Conclusions
Ag-loaded Cu/Al2O3 catalysts were studied for the glycerol
hydrogenolysis into 1,2-PDO. It was found that the addition of Ag
onto a commercially available Cu/Al2O3 was efficient in inhibi-
ting the EG formation via the decomposition of glycerol and thus
increased the 1,2-PDO selectivity. The suitable Ag loading onto
Cu/Al2O3 was 1 wt.%. The calcination temperature of the catalyst
also affected the reaction: 400 ◦C was the favorable calcination tem-
perature for the 1,2-PDO formation. The role of Ag is reduction of
the ability of Cu/Al2O3 for the decomposition of glycerol into EG at
the expense of the catalytic ability of Cu for both the dehydration
of glycerol and the hydrogenation of HA into 1,2-PDO.
Since the addition of Ag decreases the hydrogenation ability,
the catalytic reaction test with double-layered catalysts, 1% Ag-
containing Cu/Al2O3 loaded on the upper layer of catalyst bed and
Cu/Al2O3 loaded on the bottom, was performed. The effect of the
top, the interlayer, and the bottom temperatures of the catalyst bed
was also examined. Cu/Al2O3 with 1 wt.% Ag was found to mainly
inhibit the EG formation in the first step of glycerol dehydration to
HA over the upper-layer catalyst, while the bottom-layer catalyst
could be replaced by Cu/Al2O3 catalyst with high hydrogenation