Boron oxide modified bifunctional Cu/Al2O3 catalysts for the selective hydrogenolysis of glucose to 1,2-propanediol

Article abstract of DOI:10.1016/j.mcat.2019.110514

A series of B₂O₃ modified Cu/Al₂O₃ catalysts were prepared for the hydrogenolysis of glucose. The catalysts were fully characterized by BET, ICP, N₂O adsorptive decomposition, XRD, SEM, TG, H₂-TPR, CO-FTIR, XPS, and NH₃-TPD. The strong interaction between B₂O₃ and CuO could promote the dispersion of copper and inhibit the reduction of CuO, creating a proper mol ratio of Cu^δ+/Cu⁰ for the hydrogenolysis of glucose to oxygen-containing chemicals. Furthermore, the doping of B₂O₃ also introduced more acid sites onto the CuB/Al₂O₃ catalysts, which is favorable for the cleavage of hydroxyl through dehydration. Therefore, the selective hydrogenolysis of glucose to 1,2-propanediol was dependent on the contribution of Cu^δ+, Cu⁰, and acid sites. The catalytic activity and 1,2-propanediol selectivity were improved significantly by doping B₂O₃ into Cu/Al₂O₃. Among the catalysts, 1CuB/Al₂O₃ showed the highest selectivity for 1,2-propanediol, with the value of 49.5% at 96.6% conversion of glucose.

Full text of DOI:10.1016/j.mcat.2019.110514

MolecularCatalysis485(2020)110514
Contents lists available at ScienceDirect
Molecular Catalysis
journal homepage: www.elsevier.com/locate/mcat
Boron oxide modified bifunctional Cu/Al2O3 catalysts for the selective
hydrogenolysis of glucose to 1,2-propanediol
Chengwei Liua, Yaning Shanga, Shen Wanga, Xin Liua, Xianzhou Wangb,c, Jianzhou Guia,d,*,
Chenghua Zhangb,e,**, Yulei Zhub,e, Yongwang Lib,e
a State Key Laboratory of Separation Membranes & Membrane Processes, Tianjin Key laboratory of Green Chemical Technology and Process Engineering, College of
Chemistry and Chemical Engineering, Tiangong University, Tianjin 300387, People’s Republic of China
b State Key Laboratory of Coal Conversion, Institute of Coal Chemistry, Chinese Academy of Sciences, Taiyuan 030001, People’s Republic of China
c University of Chinese Academy of Sciences, Beijing 100049, People’s Republic of China
d College of Science, Engineering and Technology, University of South Africa, Nanotechnology and Water Sustainablility Research Unit, Florida Science Campus 1710,
South Africa
e National Energy Center for Coal to Liquids, Synfuels China Technology Co., Ltd., Huairou District, Beijing 101407, People’s Republic of China
A R T I C L E I N F O
A B S T R A C T
Keywords:
B2O3
A series of B2O3 modified Cu/Al2O3 catalysts were prepared for the hydrogenolysis of glucose. The catalysts were
fully characterized by BET, ICP, N2O adsorptive decomposition, XRD, SEM, TG, H2-TPR, CO-FTIR, XPS, and NH3-
TPD. The strong interaction between B2O3 and CuO could promote the dispersion of copper and inhibit the
reduction of CuO, creating a proper mol ratio of Cuδ+/Cu0 for the hydrogenolysis of glucose to oxygen-con-
taining chemicals. Furthermore, the doping of B2O3 also introduced more acid sites onto the CuB/Al2O3 cata-
lysts, which is favorable for the cleavage of hydroxyl through dehydration. Therefore, the selective hydro-
genolysis of glucose to 1,2-propanediol was dependent on the contribution of Cuδ+, Cu0, and acid sites. The
catalytic activity and 1,2-propanediol selectivity were improved significantly by doping B2O3 into Cu/Al2O3.
Among the catalysts, 1CuB/Al2O3 showed the highest selectivity for 1,2-propanediol, with the value of 49.5% at
96.6% conversion of glucose.
Cu/Al2O3
Glucose
Hydrogenolysis
1,2-Propanediol
1. Introduction
Considering the excessive amount of hydroxyl groups in the glucose
molecule, the selective elimination of surplus oxygen atoms is one of
The exploration and utilization of renewable resources has gained
increasing attention [1,2]. Among the renewable resources, such as
biomass, solar, wind energy, geothermal, etc., biomass, as the only
sustainable carbon resource, has the potential to provide renewable
high value-added chemicals and oxygenated liquid fuels [3]. Cellulose,
which is the main component of lignocellulose in nature, is composed of
glucose monomers through β-1,4-glucosidic bonds [4]. Recently, the
successful decomposition of cellulose by acid and enzyme encourages
us to study the catalytic conversion of glucose [5,6], which could not
only produce chemicals and fuels, but also help us understanding the
complex reaction process of biomass at the molecular level.
the major challenges in biomass conversion. In this respect, hydro-
genolysis, which could selectively break CeC and CeO bonds, has been
regarded as an efficient and economical method [7]. Furthermore,
many value-added chemicals such as 1,2-propanediol (1,2-PDO) and
C1-C6 alkanes can be obtained from the hydrogenolysis of glucose.
Especially, the hydrogenolysis of saccharides and alditols to 1,2-PDO
has attracted increasing interest recently [8–11]. Furthermore, com-
pared with cellulose, the hydrogenolysis of saccharides can be operated
in a continuous system with high concentration, which makes it a more
efficient and energy saving reaction process [12,13]. Usually, basic
auxiliaries (such as Ca(OH)2 and NaOH) were added to enhance CeC
⁎ Corresponding author at: State Key Laboratory of Separation Membranes & Membrane Processes, College of Chemistry and Chemical Engineering, Tiangong
University, Tianjin 300387, People’s Republic of China.
⁎⁎ Corresponding author at: State Key Laboratory of Coal Conversion, Institute of Coal Chemistry, Chinese Academy of Sciences, Taiyuan 030001, People’s Republic
of China.
E-mail addresses: guijianzhou@tiangong.edu.cn (J. Gui), zhangchh@sxicc.ac.cn (C. Zhang).
https://doi.org/10.1016/j.mcat.2019.110514
Received 28 March 2019; Received in revised form 19 June 2019; Accepted 15 July 2019
2468-8231/©2019PublishedbyElsevierB.V.

C. Liu, et al.
MolecularCatalysis485(2020)110514
bond cleavage [11,14–17], which resulted in a difficult and expensive
separation process. Moreover, the metal-oxide bifunctional catalysts,
which are more convenient and efficient for biomass hydrogenolysis,
have been widely studied. It is reported everywhere that, the metal sites
in the catalyst are mainly responsible for the hydrogenation and hy-
drogenolysis reactions. The acid sites on the metal oxides could catalyze
CeO bond breaking through dehydration reaction [18–23]. Based on
the reaction mechanism of saccharides, it has been suggested that the
metal and oxide components in the bifunctional catalyst could catalyze
the tandem reactions of biomass to value-added chemicals. Further-
more, the selective conversion of glucose is dependent on the reason-
able modulation of the structure and properties of the metals and oxides
[24,25]. For example, SnO2 has been used to modify the Ru/AC cata-
lysts for glucose conversion. The highly dispersed SnO2 could improve
1,2-PDO yield, with the value of 25% on the optimal 5%Ru3%Sn/AC
catalyst [8].
Among the bifunctional catalysts, Cu has often been used as the
active metal in biomass conversion due to its tunable hydrogenolysis
activity [26,27]. γ-Al2O3, which has abundant acid sites, is benefit for
saccharide and polyols hydrogenolysis [28,29]. For example, Vila has
prepared Cu-Al2O3 catalysts for the hydrogenolysis of glycerol [30].
The results indicate that the acid property of Al2O3 has an important
effect on glycerol hydrogenolysis.
Fig. 1. XRD spectra of the CuB/Al2O3 catalysts.
2.2. Catalyst characterization
Boron is suggested to be an efficient promotor to improve the sta-
bility and selectivity of transition metal, which has been widely studied
on the hydrogenation reaction [31–34]. Furthermore, the doping of
B2O3 could not only affect the physicochemical properties of Cu, but
also influence the hydrogenolysis activity. For example, Zhao has pre-
pared the Cu-B2O3/SiO2 bifunctional catalyst, in which B2O3 could
promote the dispersion and stability of Cu when a proper amount of
boron was introduced [31]. The CuB/SiO2 catalyst system had been
prepared and used for the selective hydrogenolysis of glycerol [34]. The
results turned out that the introduction of B2O3 onto the CuB/SiO2
catalyst brought new acid sites. In this text, B2O3 was used to modify
the structure and properties of Cu/Al2O3 catalysts. The role of B2O3 on
the properties and activity of CuB/Al2O3 catalysts for glucose trans-
formation was studied.
The contents of copper and B2O3 were detected on ICP-OES
(Optima2100DV, PerkinEler).
N2 physical adsorption was performed on an ASAP 2420 (micro-
meritics) apparatus. The catalysts were vacuumed for 8 h at 350 °C prior
to the measurements.
The metallic copper surface area was measured with the N2O che-
misorption method reported before [35,36]. Typically, 200 mg sample
was reduced for 2 h in H2 at 250 °C, and then flushed with Ar flow for
1 h. 10 vol.% N2O/N2 flow was used for the oxidation of surface Cu
species at 50 °C for 30 min, following by being flushed for 30 min with
Ar flow. Then the consumption of H2 was monitored by a TCD detector.
XRD spectra were acquired on a D/max-RA apparatus, using Cu Kα
radiation.
The scanning electron microscopy (SEM) images were obtained on
JSM-6360LV (JEOL, Japan).
Thermogravimetric analysis (TG) was operated on Netzsch STA
2500 in air. The samples were heated to 1000 °C with a rate of 10 °C/
min.
2. Experimental
2.1. Catalysts preparation
H2-TPR experiments of the catalysts were performed on
Micromeritics Autochem II 2920. The temperature was increased to
800 °C at a rate of 10 °C/min in 10 vol.% H2/Ar flow.
The XPS spectra were measured on an ESCALAB 250Xi spectrometer
equipped with Al Kα radiation (1486.6 eV). The calcined sample was in
situ reduced for 2 h in H2 at 250 °C.
Boric oxide (B2O3) modified Cu/Al2O3 (labeled as CuB/Al2O3) cat-
alysts were prepared with a stepwise impregnation method. Firstly,
B2O3 was deposited on γ-Al2O3 with an aqueous solution of boric acid
for 12 h. The precursor was dried at 80 °C for 12 h and then im-
pregnated with Cu(NO3)2 solution. The impregnated material was dried
for 12 h at 80 °C and calcined for 4 h at 350 °C. The samples were de-
noted as yCuB/Al2O3, in which y% refers to the designed weight
loading of Boric oxide (y = 1, 3, 5, 7).
NH3-TPD patterns were obtained on Micromeritics Autochem II
2920. In a typical run, 200 mg catalysts were pretreated for 2 h at
250 °C in H2 flow. After, the catalyst was flushed for 60 min in helium at
Table 1
Textural and physicochemical properties of the CuB/Al2O3 catalysts.
Catalyst
Metal loading (wt/%)a
SBET (m2/g)b
VP (cm3/g)b
DP (nm)b
SCu (m2/gCat.)c
NH3 (mmol/g)d
Cu
B2O3
Cu/Al2O3
5.01
5.02
4.92
5.03
5.00
–
263.5
270.7
279.6
275.4
268.4
0.42
0.42
0.40
0.39
0.37
4.9
4.9
4.8
4.8
4.9
16.3
9.7
9.5
9.2
5.7
73.9
1CuB/Al2O3
3CuB/Al2O3
5CuB/Al2O3
7CuB/Al2O3
0.85
2.51
4.54
6.31
80.5
90.7
101.4
115.5
a
b
c
Metal contents measured by ICP.
N2 physical adsorption and desorption method.
Determined by N2O-TPD.
d
Determined by NH3-TPD.
2

C. Liu, et al.
MolecularCatalysis485(2020)110514
250 °C. Subsequently, the glucose aqueous solution (5 wt%) was in-
troduced to the gas flow with a high-pressure infusion pump. The liquid
products were analyzed off-line on an HPLC with an RID detector. The
total carbon (TC) balance was calculated based on the following
equation:
carbon mol of products and reactant after reaction
TC(%) =
× 100
carbon mol of reactant in feedstock
3. Results and discussion
3.1. Structural properties of the CuB/Al2O3 catalysts
The contents of copper detected by ICP-OES are nearly the same as
designed. The detected contents of B2O3 were close to the designed
values. The surface area of the catalysts increased with the increase of
B2O3 loading from 263.5 m2/g of Cu/Al2O3 to 279.6 m2/g of 3CuB/
Al2O3. Then it decreased with the B2O3 contents further increasing.
However, the surface areas of the catalysts with B2O3 doping were
higher than that of Cu/Al2O3, which suggested that new surface was
formed by doping B2O3. The pore volume decreased slightly with B2O3
contents increasing, indicating that the pore was blocked when a high
amount of B2O3 was doped (Table 1).
Fig. 2. H2-TPR spectra of the CuB/Al2O3 catalysts.
Fig. 1 displays the crystalline structures of the calcined CuB/Al2O3
catalysts. Only the XRD peaks of γ-Al2O3 at 37.4, 46.0, and 67.0° were
observed [37]. However, no signals about B2O3 and CuO were detected,
which suggested that B2O3 and CuO nanoparticles were well dispersed
on the catalyst surface.
3.2. H2-TPR
To study the reducibility of CuO on the CuB/Al2O3 catalysts, H2-TPR
was performed (Fig. 2). The reduction pattern of Cu/Al2O3 showed one
main reduction peak with a shoulder peak. The reduction peak at
203 °C is attributed to CuO nanoparticles with high dispersion. The
reduction peak shifted to a higher temperature with the increase of
B2O3 contents, which indicates the strong interaction between B2O3 and
CuO. The shoulder peak at around 270 °C was attributed to large CuO
nanoparticles [34,38]. However, the shoulder peak disappeared gra-
dually with the increase of B2O3 contents, which illustrates that the
doping of B2O3 could enhance the dispersion of copper. N2O adsorptive
decomposition was carried out to measure copper surface area. The
results turned out that the exposed Cu surface area on CuB/Al2O3 cat-
alysts decreased with the increase of B2O3 contents. Combining the H2-
TPR patterns with the metallic copper areas measured by N2O chemi-
sorption, it was supposed that the B2O3 species had both dispersion and
coverage effects on copper sites [39].
Fig. 3. CO-DRIFT spectra of the reduced CuB/Al2O3 catalysts.
250 °C and cooled to 100 °C. The sample was flushed with ammonia and
then purged with helium for 1 h. The TPD pattern was started to record
in the temperature range of 100–800 °C.
2.3. Catalytic reaction process
A fixed-bed reactor system was used for the hydrogenolysis of glu-
cose. First, 2.0 g catalysts were loaded and reduced for 2 h in H2 at
Fig. 4. Cu 2p XPS (A) and Cu LMM Auger spectra (B) of the reduced and used CuB/Al2O3 catalysts.
3

C. Liu, et al.
MolecularCatalysis485(2020)110514
Fig. 5. (A) NH3-TPD patterns of reduced CuB/Al2O3, (B) The acid amount as a function of B2O3 added.
3.3. CO-DRIFT
3.4. Chemical states of Cu species
CO-DRIFT experiments were performed to study the chemical en-
vironment of Cu in CuB/Al2O3 (Fig. 3). According to the previous re-
ports, the IR bands assigned to the linear adsorption of CO on copper
are often below 2140 cm−1, which are changeable according to the
chemical environment of copper [40]. All the catalysts showed a broad
peak at around 2050–2140 cm−1, indicating the high dispersion of
metallic copper nanoparticles. An obvious blue shift of the bands was
observed with B2O3 contents increasing, which demonstrates that B2O3
has strong electronic effects on Cu.
The Cu 2p XPS spectra were obtained to study the chemical state of
copper (Fig. 4A). The broad peak at 933.7 eV with a satellite peak in
Fig. 4A indicates that CuO species could not be completely reduced,
with partial Cu2+ species existing on the reduced catalyst. The reduc-
tion degree of Cu species was also analyzed by Cu LMM XAES spectra.
The results showed that there were both Cu0 and Cuδ+ (δ ≥ 1) species
existed on the reduced catalyst [35]. Furthermore, compared with the
Cu/Al2O3 catalyst, the ratios of Cuδ+/(Cuδ+ + Cu0) were higher on the
catalysts with B2O3 doping, which confirmed the strong electronic ef-
fect between B2O3 and copper. Furthermore, the highest ratio of Cuδ+
(Cuδ+ + Cu0) was obtained on 1CuB/Al2O3.
/
Fig. 6. (A) Glucose conversion, (B) sorbitol selectivity, (C) glycerol selectivity, and (D) 1,2-PDO selectivity in glucose conversion on the CuB/Al2O3 catalysts.
4

C. Liu, et al.
MolecularCatalysis485(2020)110514
Scheme 1. The main reaction route of the hydrogenolysis of glucose.
3.5. NH3-TPD
3.6. The hydrogenolysis of glucose
The NH3-TPD patterns were recorded to study the acidity of the CuB/
Fig. 6 shows the catalytic performance versus WHSV. As shown in
Fig. 6A, all the CuB/Al2O3 catalysts with B2O3 doped showed higher
glucose conversion than that of Cu/Al2O3. Sorbitol selectivities on CuB/
Al2O3 with B2O3 doping were lower than that on Cu/Al2O3 catalysts.
Furthermore, 1CuB/Al2O3 showed the lowest sorbitol selectivity. The
3CuB/Al2O3, 5CuB/Al2O and 7CuB/Al2O3 catalysts display similar
sorbitol selectivities. The glycerol selectivity in Fig. 6C showed a special
order. With the increase of B2O3 contents, glycerol selectivity decreased
first and then increased. The lowest glycerol selectivity was shown on
1CuB/Al2O3. Furthermore, the catalysts with B2O3 doping showed
higher 1,2-PDO selectivity than Cu/Al2O3. The optimal value of 49.5%
at 96.6% glucose conversion was obtained on 1CuB/Al2O3. The results
illustrated that B2O3 could promote the hydrogenolysis of glucose.
Al2O3 catalysts (Fig. 5A). All the reduced catalysts showed a broad ammonia
desorption peak at 130–500 °C. With B2O3 contents increasing, the amount
of acid sites calculated based on the adsorbed NH3 decreased as follows:
7CuB/Al2O3 (115.5 mmol/g) > 5CuB/Al2O3 (101.4 mmol/g) > 3CuB/
Al2O3
(90.7 mmol/g) > 1CuB/Al2O3
(80.5 mmol/g) > Cu/Al2O3
(73.9 mmol/g). The ratio of the acid amount to B2O3 added has also been
analyzed (Fig. 5B). An almost linear relationship between the amount of the
acid sites and the content of B2O3 was obtained, which indicated that the
new acid sites were mainly provided by the doped B2O3 species.
Table 2
XAES parameters of the Cu 2p region in CuB/Al2O3 catalysts.
Catalyst
K.E. (eV)a
Cu0
XCu (%)b
+
Cuδ+
Cu/Al2O3
917.9
917.9
917.6
917.9
917.9
917.6
914.5
914.7
913.8
914.3
914.0
913.8
49
57
77
56
53
54
1CuB/Al2O3
1CuB/Al2O3-used
3CuB/Al2O3
5CuB/Al2O3
7CuB/Al2O3
a
b
Kinetic energy.
0
+
XCu (%) = Cu +/(Cu + Cu ) × 100
.
+
Fig. 7. 1,2-PDO selectivity and the physicochemical properties on the CuB/
Al2O3 catalysts.
5

C. Liu, et al.
MolecularCatalysis485(2020)110514
Table 3
The hydrogenolysis of glucose on the CuB/Al2O3 catalysts.a
Catalyst
WHSV (h−1
)
Conv. (%)
Selectivity (on a carbon basis, %)
TC (%)
Sorbitol
Erythritol
EG
Glycerol
1,2-PDO
Ethanol
Othersb
Cu/Al2O3
0.36
0.48
0.60
0.36
0.48
0.60
0.36
0.48
0.60
0.36
0.48
0.60
0.36
0.48
0.60
0.36
0.48
0.60
91.6
87.9
87.2
96.6
94.6
92.6
88.7
87.4
84.6
96.6
93.9
92.2
96.6
94.7
93.0
96.6
94.3
92.0
18.0
22.7
38.5
11.0
14.6
17.9
35.7
39.9
44.1
15.9
19.8
22.5
16.7
18.4
20.7
17.8
19.5
21.9
4.8
5.4
2.3
4.3
4.6
5.1
1.0
1.1
0.4
4.5
5.1
5.4
4.8
5.1
5.2
4.5
5.3
5.5
13.0
12.6
10.4
14.3
13.4
13.0
0.9
15.2
16.0
17.6
13.9
14.8
15.8
10.2
11.4
15.5
17.6
17.9
17.9
18.3
17.5
17.7
18.4
17.8
17.6
43.4
38.9
27.8
49.5
47.1
43.5
36.3
33.8
27.1
44.9
41.4
38.3
43.9
41.2
38.1
42.6
39.5
36.5
2.2
0.0
0.0
2.4
2.0
1.7
0.1
0.1
0.1
2.1
1.8
1.6
2.2
2.0
1.5
2.0
1.9
1.6
3.4
4.4
3.4
4.6
3.5
3.0
15.8
12.5
11.3
0.1
0.6
1.2
0.1
1.4
3.3
0.1
1.5
4.1
96.9
96.1
97.0
95.6
96.7
97.2
86.0
89.1
90.4
99.9
99.4
98.9
99.9
98.7
96.9
99.9
98.6
96.2
1CuB/Al2O3
1CuB/Al2O3
3CuB/Al2O3
5CuB/Al2O3
7CuB/Al2O3
c
1.2
1.5
14.9
13.4
13.1
14.0
14.4
13.5
14.6
14.5
12.8
a
b
c
Reaction condition: 5 wt% glucose aqueous solution, 4 MPa H2, 180 °C.
Others: humins.
Reduced at 300 °C.
3.7. Correlation between the properties and performance of CuB/Al2O3
that the high-temperature reduction changed the ratio of Cu/Cuδ+ and
caused the aggregation of Cu species, which resulted in the lower hy-
drogenolysis activity of the catalyst. Considering the tandem reaction
process, it suggested that 1,2-PDO selectivity was determined by both
the acidic and metal sites. An interesting found was that 1,2-PDO se-
lectivity displayed nearly the same trend with the Cuδ+ percent on the
CuB/Al2O3 catalysts. It seems that the proportion of Cuδ+/Cu0 plays
great roles on the selective hydrogenolysis of glucose. The 1CuB/Al2O3
catalyst after reaction was also characterized by XPS and XAES to detect
the ratio of Cu cation to Cu metal (Fig. 4 and Table 2). The Cu 2p XPS
spectra of the used 1CuB/Al2O3 catalyst displayed the similar results to
the reduced 1CuB/Al2O3 catalyst, which indicates the presence of both
Cu0 and Cuδ+ species. Furthermore, the XAES results turned out that
the percentage of Cuδ+ was 77%, which is much higher than that on the
reduced 1CuB/Al2O3 catalyst. We speculate that the small Cu nano-
particles on the 1CuB/Al2O3 catalyst are prone to be oxidized in air.
It has been reported that the metal cations such as Snx+ and Wx+ on
the solid catalysts could act as Lewis acid in the hydrogenolysis of
saccharides, which could catalyze the isomerization between glucose
and fructose [29,42]. Furthermore, many researchers also pointed out
that the Cuδ+ species on the reduced copper-based catalysts could
function as Lewis acid sites [29,43,44]. It can be concluded that the
electronic deficient Cuδ+ species could provide Lewis acid sites, which
B2O3 has been widely used as an additive for various catalysts due to
its structure and electronic effects [33,34]. Based on the N2O-TPD and
H2-TPR results, the close contact between B2O3 and copper not only
promotes the dispersion of copper but also has a coverage effect on
copper species, resulting in the decrease of copper surface area with the
increase of B2O3 contents. The coverage and dispersion effects have
been widely reported in the case of WOx [29], B2O3 [32], MoOx [41],
etc. For example, Zhu et al. have prepared the CuB/SiO2 catalysts. They
suggested that the doping of B2O3 could inhibit the agglomeration of
copper species during the heating treatment and facilitate the formation
of small CuO nanoparticles. As revealed by H2-TPR, CO-DRIFTS, and Cu
LMM XAES results, the close contact between B2O3 and copper resulted
in a strong electron interaction, which inhibited the reduction of CuO
and decreased the ratio of Cu0/Cuδ+ on the reduced catalysts. Ad-
ditionally, the doping of B2O3 introduced more acidity sites onto the
catalysts.
The hydrogenolysis of glucose to 1,2-PDO undergoes some tandem
reactions, as shown in Scheme 1 [29]: (1) the isomerization of glucose
to fructose; (2) The CeC bond cleavage of fructose to dihydroxyacetone
and glyceraldehyde through retro aldol condensation reaction, which is
supposed to be conducted on metal sites; (3) The CeO bond cleavage of
dihydroxyacetone and glyceraldehyde through dehydration reaction;
(4) The hydrogenation of the intermediates.
As previously suggested, sorbitol is obtained by the hydrogenation
of glucose, which acts as the competitive reaction with the retro aldol
condensation reaction [29]. Moreover, the selectivity of 1,2-PDO, the
percent of Cuδ+
(
X
Cu
+), the surface area of Cu (SCu), and the amount of
acid sites on different CuB/Al2O3 catalysts were shown in Fig. 7. The
results demonstrated that, with the increasing B2O3 content, the acid
amounts increased while the copper surface area decreased. Accord-
ingly, the optimal 1,2-PDO selectivity was achieved on the 1CuB/Al2O3
catalyst. The hydrogenolysis of glucose has been performed on the
1CuB/Al2O3 catalyst reduced at 300 °C. The results turned out that the
conversion of glucose was lower than that on the 1CuB/Al2O3 catalyst
reduced at 250 °C. Furthermore, the reduction temperature of 1CuB/
Al2O3 also affected the selectivity of products. For example, the se-
lectivity of sorbitol was higher than that on 1CuB/Al2O3 reduced at
250 °C. Correspondingly, the selectivity of 1,2-propanediol decreased
when 1CuB/Al2O3 was reduced at higher temperature. We speculated
Fig. 8. Stability of 1CuB/Al2O3 in glucose hydrogenolysis.
6

C. Liu, et al.
MolecularCatalysis485(2020)110514
Fig. 9. Characterization results of the calcined and used 1CuB/Al2O3 catalyst: (A) XRD; (B) TG; (C) the SEM image of the calcined catalyst; (D) the SEM image of the
used catalyst.
can catalyze the isomerization reaction.
mainly responsible for the hydrogenation of glucose and the inter-
mediates, which can be easily conducted at mild conditions [46]. Fur-
thermore, the retro aldol condensation reaction could also be carried
out on Cu metal sites. In this respect, the presence of Cuδ+ could tune
the activity of different elementary reactions. For example, the Cuδ+
species could act as acid sites, which could enhance the isomerization,
retro aldol condensation and dehydration reactions. Overall, the doping
of B2O3 could tune the properties of metal and acid sites on CuB/Al2O3,
which could enhance 1,2-PDO selectivity (Table 3).
Furthermore, many researchers also found that the retro aldol
condensation reaction could be promoted by Lewis acid sites [21–23].
For example, a bifunctional Ru-WOx catalyst has been used for the
hydrogenolysis of glucose by Liu et al [21]. They pointed out that WOx
could promote the retro aldol condensation reaction. Vilcocq pointed
out that Lewis acid sites could change to L-OH- Brønsted basic sites in
association with water and promote retro aldol condensation reaction
[23]. Therefore, the Cuδ+ species, which act as Lewis acid, are pro-
posed to promote the retro aldol condensation reaction. Moreover, the
acid sites could also catalyze the dehydration of C3 intermediates.
Therefore, it was proposed that the Cuδ+ species could tune the reac-
tion rate of elementary reactions and enhance 1,2-PDO selectivity.
Furthermore, the adjustment of Cu0/Cuδ+ could also modify the hy-
drogenation activity and decrease sorbitol selectivity. Based on the
reaction mechanism, it was proposed that the acid sites provided by
Al2O3 or B2O3 could promote the isomerization of glucose to fructose,
the retro aldol condensation of fructose to unsaturated intermediates,
and the dehydration of the intermediates [45]. Cu metal sites are
3.8. The stability of the 1CuB/Al2O3 catalyst
The stability of the 1CuB/Al2O3 catalyst has been evaluated (Fig. 8).
The results turned out that the catalyst could be operated over 120 h
without obvious deactivation. The 1CuB/Al2O3 catalyst after the long-
term operation has been characterized by XRD, SEM, TG and ICP
(Fig. 9). The XRD pattern of the used catalyst displayed the diffraction
peaks of boehmite at 14.4, 28.1, and 38.3° (JCPDS 05-0190), which
suggested that the AlOOH structure had been formed in the
Table 4
the ICP and TG results of the 1CuB/Al2O3 catalyst.
Catalyst
Metal loading (wt/%)
Weight loss (%)c
Cua
B2O3
Cub
B2O3
a
b
calcined
used
5.02
4.41
0.85
0.68
–
–
10.2
18.6
4.81
0.74
a
b
c
Measured by ICP-OES.
Calibrated with the TG results: Cub (wt/%) = Cua (wt/%)/(100 18.6 + 10.2) × 100; B2O3b (wt/%) = B2O3a (wt/%)/(100 18.6 + 10.2) × 100.
Measured by TG.
7

C. Liu, et al.
MolecularCatalysis485(2020)110514
hydrothermal condition. The SEM images of the calcined and spent
catalysts displayed that the catalyst was broken down into smaller
particles after the long-term operation, which could be due to the effect
of the phase transformation of γ-Al2O3. Considering the stability of the
catalyst, it suggested that the phase transformation of γ-Al2O3 had little
effect on glucose hydrogenolysis. Furthermore, Wu et al. have com-
pared the activity of Cu/boehmite and Cu/γ-Al2O3 on the hydro-
genolysis of glycerol to 1,2-PDO [47]. The results turned out that Cu/
boehmite showed higher glycerol conversion and 1,2-PDO selectivity,
which was ascribed to the small size of Cu metal clusters stabilized by
boehmite and the enhanced dehydration of the intermediate by the
Lewis acid sites of boehmite. The mass change of the 1CuB/Al2O3 cat-
alyst before and after reaction has been quantified by TG analysis in air
atmosphere (Table 4). The results turned out that the weight loss of the
used catalyst was larger than the calcined catalyst when they were
heated to 1000 °C in air, which was supposed to be caused by the de-
hydration of boehmite and the removal of coke on the used catalyst.
Therefore, considering the mass change caused by the phase transfor-
mation and coke on the surface of 1CuB/Al2O3, the contents of Cu and
B2O3 measured by ICP-OES were calibrated with the TG results. The
values were a little lower than that on the calcined catalyst, which
indicated the stability of Cu and B2O3 on the catalyst surface.
[2] I. Jiménez-Morales, M. Moreno-Recio, J. Santamaría-González, P. Maireles-Torres,
A. Jiménez-López, Appl. Catal. B: Environ. 164 (2015) 70–76.
[3] Y. Huang, Y. Fu, Green Chem. 15 (2013) 1095–1111.
[4] W. Zhu, H. Yang, J. Chen, C. Chen, L. Guo, H. Gan, X. Zhao, Z. Hou, Green Chem. 16
(2014) 1534–1542.
[5] J. Wang, J. Xi, Y. Wang, Green Chem. 17 (2015) 737–751.
[6] Y. Zhang, X. Liu, X. Liu, G. Xu, C. Wang, Q. Lu, L. Ma, Green Chem. 21 (2019)
5647–5656.
[7] A. Yepez, A. Pineda, A. Garcia, A.A. Romero, R. Luque, Phys. Chem. Chem. Phys. 15
(2013) 12165–12172.
[8] J. Pang, M. Zheng, X. Li, Y. Jiang, Y. Zhao, A. Wang, J. Wang, X. Wang, T. Zhang,
Appl. Catal. B: Environ. 239 (2018) 300–308.
[9] J. Wang, J. Xi, Y. Wang, Appl. Catal. B: Environ. 220 (2018) 251–263.
[10] M. Rivière, N. Perret, D. Delcroix, M. Cabiac, C. Pinel, M. Besson, ACS Sustain.
Chem. Eng. 6 (2018) 4076–4085.
[11] Z. Xiao, S. Jin, G. Sha, C.T. Williams, C. Liang, Ind. Eng. Chem. Res. 53 (2014)
8735–8743.
[12] S. Esposito, B. Silvestri, V. Russo, B. Bonelli, M. Manzoli, F.A. Deorsola, A. Vergara,
A. Aronne, M.D. Serio, ACS Catal. 9 (2019) 3426–3436.
[13] M.J. Climent, A. Corma, S. Iborra, Green Chem. 13 (2011) 520–540.
[14] C.L. Lautenschlä ger, M. Bockmühl, G.Ehrhart, W. Kroβ, German Patent 541,362,
(1932).
[15] K.L. Lautenschlä ger, M. Bockmühl, G. Ehrhart, W. Krohs, U.S. Patent 1,915,431,
(1933).
[16] A.W. Larcher, (to Du Pont Co.) U.S. Patent 1,963,997, (1934).
[17] H.S. Rothrock, (to Du Pont Co.), U.S. Patent 2,004,135, (1935).
[18] P. Yazdani, B. Wang, Y. Du, S. Kawi, A. Borgna, Catal. Sci. Technol. 7 (2017)
4680–4690.
[19] X. Jin, J. Shen, W. Yan, M. Zhao, P.S. Thapa, B. Subramaniam, R.V. Chaudhari, ACS
Catal. 5 (2015) 6545–6558.
[20] E. Girard, D. Delcroix, A. Cabiac, Catal. Sci. Technol. 6 (2016) 5534–5542.
[21] Y. Liu, Y. Liu, Y. Zhang, Appl. Catal. B: Environ. 242 (2019) 100–108.
[22] F. Chambon, F. Rataboul, C. Pinel, A. Cabiac, E. Guillon, N. Essayem,
ChemSusChem 6 (2013) 500–507.
4. Conclusion
B2O3 could improve the dispersion of Cu nanoparticles and create
new acid sites on CuB/Al2O3. However, the close contact between B2O3
and Cu also decreases the Cu surface area due to the overlap effect. The
decorated B2O3 on the Cu surface has strong electronic effect on Cu,
which tunes the proportion of Cuδ+/Cu0. The doping of B2O3 could
enhance 1,2-PDO selectivity, with the optimal value of 49.5% on 1CuB/
Al2O3. Cu0, Cuδ+ and B2O3 have synergistic effect on the hydro-
genolysis of glucose, which could promote the selectivity of C3 polyols.
Especially, the Cuδ+ species, which is supposed to act as Lewis acid,
could catalyze the elementary reactions in the transformation of glu-
cose to 1,2-PDO.
[23] L. Vilcocq, R. Koerin, A. Cabiac, C. Especel, S. Lacombe, D. Duprez, Appl. Catal. B:
Environ. 148–149 (2014) 499–508.
[24] Y.T. Kim, J.A. Dumesic, G.W. Huber, J. Catal. 304 (2013) 72–85.
[25] N. Li, G.W. Huber, J. Catal. 270 (2010) 48–59.
[26] K. Tajvidi, P.J. Hausoul, R. Palkovits, ChemSusChem 7 (2014) 1311–1317.
[27] Z. Gao, C. Li, G. Fan, L. Yang, F. Li, Appl. Catal. B: Environ. 226 (2018) 523–533.
[28] S. Zhu, Y. Zhu, S. Hao, H. Zheng, T. Mo, Y. Li, Green Chem. 14 (2012) 2607–2616.
[29] C. Liu, C. Zhang, S. Hao, S. Sun, K. Liu, J. Xu, Y. Zhu, Y. Li, Catal. Today 261 (2016)
116–127.
[30] F. Vila, M.L. Granados, M. Ojeda, J.L.G. Fierro, R. Mariscal, Catal. Today 187
(2012) 122–128.
[31] S. Zhao, H. Yue, Y. Zhao, B. Wang, Y. Geng, J. Lv, S. Wang, J. Gong, X. Ma, J. Catal.
297 (2013) 142–150.
[32] Z. He, H. Lin, P. He, Y. Yuan, J. Catal. 277 (2011) 54–63.
[33] K.F. Tan, J. Chang, A. Borgna, M. Saeys, J. Catal. 280 (2011) 50–59.
[34] S. Zhu, X. Gao, Y. Zhu, Y. Zhu, H. Zheng, Y. Li, J. Catal. 303 (2013) 70–79.
[35] B. Zhang, Y. Zhu, G. Ding, H. Zheng, Y. Li, Appl. Catal. A: Gen. 443–444 (2012)
191–201.
Acknowledgements
This work was financially supported by the Major State Basic
Research Development Program of China (973 Program) (No.
2012CB215305), the Natural Science Foundation of Tianjin
(No.18JCQNJC06600) and the Science & Technology Development
Fund of Tianjin Education Commission for Higher Education (No.
2018KJ203). The authors are grateful for the financial support from the
National Natural Science Foundation of China (No. 21576211),
Program for Tianjin Innovative Research Team in Universities (No.
TD13-5031), and Tianjin 131 Research Team of Innovative Talents.
This work was also supported by Synfuels China Co., Ltd.
[36] C.J.G. Van Der Grift, A.F.H. Wielers, B.P.J. Jogh, J. Van Beunum, M. De Boer,
M. Versluijs-Helder, W. Geus, J. Catal. 131 (1991) 178–189.
[37] X. Chen, G. Clet, K. Thomas, M. Houalla, J. Catal. 273 (2010) 236–244.
[38] Z. Huang, F. Cui, H. Kang, J. Chen, C. Xia, Appl. Catal. A: Gen. 366 (2009) 288–298.
[39] Y. Hong, D. Lee, H. Eom, K. Lee, Appl. Catal. B: Environ. 150–151 (2014) 438–445.
[40] F. Arena, G. Italiano, K. Barbera, S. Bordiga, G. Bonura, L. Spadaro, F. Frusteri,
Appl. Catal. A: Gen. 350 (2008) 16–23.
[41] L. Chen, Y. Zhu, H. Zheng, C. Zhang, Y. Li, Appl. Catal. A: Gen. 411–412 (2012)
95–104.
[42] Y. Román-Leshkov, M. Moliner, J.A. Labinger, M.E. Davis, Angew. Chem. Int. Ed. 49
(2010) 8954–8957.
[43] A. Yin, X. Guo, W.L. Dai, K. Fan, J. Phys. Chem. C 113 (2009) 11003–11013.
[44] B.Z. Sun, W.K. Chen, X. Wang, C.H. Lu, Appl. Surf. Sci. 253 (2007) 7501–7505.
[45] C. Liu, C. Zhang, S. Sun, K. Liu, S. Hao, J. Xu, Y. Zhu, Y. Li, ACS Catal. 5 (2015)
4612–4623.
References
[46] H. Li, W. Wang, J. Deng, J. Catal. 191 (2000) 257–260.
[47] Z. Wu, Y. Mao, M. Song, X. Yin, M. Zhang, Catal. Commun. 32 (2013) 52–57.
[1] A.M. Ruppert, K. Weinberg, R. Palkovits, Angew. Chem. Int. Ed. 51 (2012)
2564–2601.
8