WOx modified Cu/Al2O3 as a high-performance catalyst for the hydrogenolysis of glucose to 1,2-propanediol

Article abstract of DOI:10.1016/j.cattod.2015.06.030

Glucose is one of the most important platform molecules of biomass in nature. The selective hydrogenolysis of glucose to 1,2-PDO is still a challenge. However, The hydrogenolysis of fructose has higher activity and selectivity to 1,2-PDO. Therefore, A series of Cu-WO_x/Al₂O₃ catalysts with high activity for glucose isomerization to fructose and fructose hydrogenolysis to 1,2-PDO were designed for glucose hydrogenolysis to 1,2-PDO. The W surface density was controlled as low as 0.8 W/nm². The low W surface density could make the WO_x species present as isolated WO₄ structure, which could only provide more Lewis acid sites. As a result, the isolated WO₄ species could form a complex with glucose and then promote the isomerization of glucose to fructose. The isolated WO₄ species also have coverage, dispersion, and electronic effects on copper sites, resulting more stable copper sites and proper amount of hydrogenation sites on Cu-WO_x/Al₂O₃ surface. The correlations between the ratio of Lewis acid amount to Cu surface area and the selectivity of 1,2-PDO suggest that the hydrogenolysis of glucose to 1,2-PDO follows the bifunctional reaction route which contains the reactions on Lewis acid and metal sites. Furthermore, the highest 1,2-PDO selectivity of 55.4% was obtained on Cu-WO_x(0.8)/Al₂O₃.

Full text of DOI:10.1016/j.cattod.2015.06.030

G Model
CATTOD-9687; No. of Pages12
ARTICLE IN PRESS
Catalysis Today xxx (2015) xxx–xxx
Contents lists available at ScienceDirect
Catalysis Today
journal homepage: www.elsevier.com/locate/cattod
WOx modified Cu/Al2O3 as a high-performance catalyst for the
hydrogenolysis of glucose to 1,2-propanediol
Chengwei Liua,c, Chenghua Zhanga,b,∗, Shunli Haob, Sikai Sunb, Kangkai Liub, Jian Xub,
Yulei Zhua,b, Yongwang Lia,b
a State Key Laboratory of Coal Conversion, Institute of Coal Chemistry, Chinese Academy of Sciences, Taiyuan 030001, PR China
b National Energy Center for Coal to Liquids, Synfuels CHINA Co., Ltd., Huairou District, Beijing, 101400, PR China
c University of Chinese Academy of Sciences, Beijing 100049, PR China
a r t i c l e i n f o
a b s t r a c t
Article history:
Glucose is one of the most important platform molecules of biomass in nature. The selective hydrogenol-
ysis of glucose to 1,2-PDO is still a challenge. However, The hydrogenolysis of fructose has higher activity
and selectivity to 1,2-PDO. Therefore, A series of Cu-WOx/Al2O3 catalysts with high activity for glucose
isomerization to fructose and fructose hydrogenolysis to 1,2-PDO were designed for glucose hydrogenol-
ysis to 1,2-PDO. The W surface density was controlled as low as 0.8 W/nm2. The low W surface density
could make the WOx species present as isolated WO4 structure, which could only provide more Lewis
acid sites. As a result, the isolated WO4 species could form a complex with glucose and then promote
the isomerization of glucose to fructose. The isolated WO4 species also have coverage, dispersion, and
electronic effects on copper sites, resulting more stable copper sites and proper amount of hydrogenation
sites on Cu-WOx/Al2O3 surface. The correlations between the ratio of Lewis acid amount to Cu surface
area and the selectivity of 1,2-PDO suggest that the hydrogenolysis of glucose to 1,2-PDO follows the
bifunctional reaction route which contains the reactions on Lewis acid and metal sites. Furthermore, the
highest 1,2-PDO selectivity of 55.4% was obtained on Cu-WOx(0.8)/Al2O3.
Received 12 May 2015
Received in revised form 27 June 2015
Accepted 29 June 2015
Available online xxx
Keywords:
WOx
Cu-WOx/Al2O3
Glucose
Hydrogenolysis
1,2-Propanediol
© 2015 Elsevier B.V. All rights reserved.
1. Introduction
reviews [2,6–9]. Many value-added chemicals and biofuels such as
5-hydroxymethylfufural (HMF), 2,5-dimethylfuran, levulinic acid,
Increasing interests have been stimulated in developing novel
chemical processes based on renewable biomass materials for pro-
ducing energy and chemicals in view of the depletion of fossil
resources and ongoing climate change [1–4]. Among biomass based
feedstocks, glucose, as the monomer of cellulose, is an attractive
platform chemical which can be efficiently converted into vari-
ous chemicals, fuels, foods, medicines [5]. Furthermore, it is the
most abundant monosaccharide and readily available from the
hydrolysis of cellulose by mineral acids [5]. Therefore, the cat-
alytic transformation of glucose is of great importance for biomass
utilization. On this respect, the chemical routes for monosac-
charide reactions such as oxidation, dehydration, esterification,
hydrogenation, and hydrogenolysis have been reported on many
ꢀ-valerolactone, gluconic acid, pentanoic acid ester, and various
polyols have been studied extensively recently [10]. Especially,
hydrogenolysis, which results in the cleavage of C C and C
O
bonds by hydrogen, can produce valuable platform chemicals such
as propanediol (PDO) and ethylene glycol (EG) that are already inte-
grated in today’s value chain. Furthermore, the hydrogenolysis of
glucose in aqueous solution can avoid the fatal disadvantages in
biomass conversion, such as the feedstock transportation, batch
reaction, coking, organic solution, environment pollution, etc. All
these merits make glucose an ideal feedstock in future large-scale
biorefining. Therefore, the process of glucose hydrogenolysis in
aqueous solution is respected to bear the potential to bridge cur-
rently available technologies and the future biomass-based refinery
concepts [2].
In the earliest stage of the hydrogenolysis reaction of glucose
and its derived-polyols sorbitol, glycerol was considered as the
most desirable product. However, with the surplus of glycerol
from biodiesel production, the importance of polyols-related value-
added chemicals has shifted towards 1,2-propanediol (1,2-PDO)
and ethylene glycol (EG). Furthermore, 1,2-PDO is an industrially
∗
Corresponding author at: State Key Laboratory of Coal Conversion, Insti-
tute of Coal Chemistry, Chinese Academy of Sciences, Taiyuan 030001, PR China.
Tel.: +86 010 69667802.
E-mail addresses: zhangchh@sxicc.ac.cn (C. Zhang), zhuyulei@sxicc.ac.cn
(Y. Zhu).
http://dx.doi.org/10.1016/j.cattod.2015.06.030
0920-5861/© 2015 Elsevier B.V. All rights reserved.
Please cite this article in press as: C. Liu, et al., WOx modified Cu/Al2O3 as a high-performance catalyst for the hydrogenolysis of glucose
to 1,2-propanediol, Catal. Today (2015), http://dx.doi.org/10.1016/j.cattod.2015.06.030

G Model
CATTOD-9687; No. of Pages12
ARTICLE IN PRESS
2
C. Liu et al. / Catalysis Today xxx (2015) xxx–xxx
important chemical and is extensively used in the production of
polyester resins, pharmaceuticals, tobacco humectants, paints, cos-
metics and antifreeze [11]. It is currently produced at an industrial
scale using petroleum based propylene as the carbon feedstock. The
hydrogenolysis of biomass-derived glucose to 1,2-PDO provides a
sustainable and economic route instead of the oxidation of fossil
fuel-derived propylene.
stability [24]. The ratio of C C/C O bond cleavage can be tuned by
modulating the concentration of acid to metal sites. A good balance
between the two functions is very important to obtain a high yield
of target polyols. Therefore, a non-noble Cu-WOx/Al2O3 catalyst
with different ratios of acid to metal sites by changing the den-
sity of W was designed. Cu/Al2O3 with proper activity for C C/C
O
bond cleavage and C O bond hydrogenation showed a high activity
for the hydrogenolysis of fructose to 1,2-PDO, and high dispersed
WOx species have a high selectivity for the isomerization of glucose
to fructose. The synergy effect of the catalytic active site promotes
glucose conversion and 1,2-PDO selectivity. This work focuses on
the effect of W on the properties and catalytic performance of the
Cu-WOx/Al2O3 catalysts.
Cu, Ni, Ru based catalysts together with base additives such as
CaO were often used for the hydrogenolysis of glucose and sorbitol,
and a mixture of lower polyols such as glycerol, EG, and 1,2-PDO
was mainly obtained. For example, Ye [12] reported a Ce-promoted
Ni/Al2O3 catalyst for aqueous-phase hydrogenolysis of sorbitol to
glycols. At 513 K and 7.0 MPa, above 90% of sorbitol conversion and
55–60% of glycol selectivities were obtained. Chen [13] found that
the Ni/MgO catalyst with Ni/MgO ratio of 3:7 exhibited the best
performance for sorbitol hydrogenolysis at 473 K and 4 MPa H2,
with 67.8% conversion and 80.8% total selectivity of EG, 1,2-PDO
and glycerol. Banu [14] studied the conversion of sorbitol on Ni-
NaY and Pt-NaY catalysts. They found that 1,2-PDO was the major
product on Ni-NaY and glycerol was the main product on Pt-NaY.
Meanwhile, the highest 1,2-PDO selectivity of 72% was achieved at
59% sorbitol conversion on Ni(6%)-Pt(1%)/NaY at 493 K and 6 MPa.
A copper-chromium catalyst was more selective for 1,2-PDO. As
reported by Zartman [15], a 60% yield of 1,2-PDO was obtained from
glucose hydrogenolysis in supercritical ethanol. The CuCr catalyst
was also reported to be efficient for the hydrogenolysis of cellulose
to 1,2-PDO, with the yield of 42.6% [16]. Apparently, the study still
encounters many problems such as poor activities and selectivi-
ties, harsh condition, and environmental pollution. Moreover, safe,
efficient and economic catalysts are the key factor for the selective
conversion of glucose to 1,2-PDO in aqueous solution.
2. Experiment
2.1. Catalyst preparation
Cu/Al2O3 and WOx modified Cu/Al2O3 catalysts (denoted as
Cu-WOx/Al2O3) with different WOx weight loadings in the range
of 1–7 wt% were prepared by a stepwise impregnation method.
Firstly, ꢀ-Al2O3 (purchased from Sinopharm Chemical Reagent Co.,
Ltd, China) was impregnated for 12 h with a certain amount of
(NH4)6W7O24·6H2O (Sinopharm Chemical Reagent Co., Ltd, China)
aqueous solution. The impregnated sample was dried at 120 ◦C
overnight. Afterwards, they were calcined at 600 ◦C for 4 h. The
obtained sample was then impregnated for 12 h with an aque-
ous solution of Cu(NO3)2·3H2O (Sinopharm Chemical Reagent Co.,
Ltd, China). After impregnation, the sample was dried at 120 ◦C
overnight, followed by calcined at 350 ◦C for 4 h. The catalysts were
labeled as Cu-WOx(y)/Al2O3, in which y stands for the surface den-
sity of W (atoms/nm2). The Cu/Al2O3 catalyst was prepared using
the stepwise impregnation method mentioned above.
It is well known that the hydrogenolysis of saccharides and poly-
ols can be performed through a bifunctional reaction pathway that
involves three primary reactions, C O bond cleavage on acid sites,
C
C bond cleavage and C O/C C bond hydrogenation on metal
sites [17–19]. Furthermore, retro-aldol condensation is the key
C bond cleavage reaction and occurs on metal sites [17]. How-
2.2. Catalyst characterization
C
ICP optical emission spectroscopy (Optima2100DV, PerkinEler)
was performed to determine the chemical compositions of calcined
catalysts.
BET surface areas were measured on a micromeritics ASAP 2420
instrument. All the samples were degassed at 110 ◦C for 1 h and
350 ◦C for 8 h prior to the measurement.
ever, it is well known that the retro-aldol condensation of glucose
mainly produces C2 and C4 polyols, and the conversion of glucose
to C3 polyols includes two steps: the isomerization of glucose to
fructose, which can be promoted by Lewis acid; the retro-aldol
condensation of fructose to dihydroxyacetone and glyceraldehyde
[20,21]. Therefore, based on the understanding of the reaction path-
way, an efficient bifunctional catalyst with high activity for glucose
isomerization, fructose retro-aldol condensation, dehydration and
hydrogenation is required for the selective conversion of glucose to
1,2-PDO. Among the catalysts, Ni and Ru catalysts were frequently
used for the hydrogenolysis of sugars and sugar alcohols due to their
high activity for C C bond cleavage. However, excessive C C bond
scission often happened, and large amount of unwanted products
such as methane were formed [22]. It is known that copper cata-
lyst has high activity for C O bond cleavage and low activity for
The surface areas of copper were calculated using N2O
chemisorption method performed based on the assuming spherical
shape of the copper metal particles and 1.4 × 1019 copper atoms/m2
[25,26]. Before the measurements, 200 mg sample was reduced by
flowing H2 at 250 ◦C for 2 h, followed by purging with He for 1 h.
After cooled down to 50 ◦C, the sample was flushed with 10 vol.%
N2O/N2 (50 mL/min) flow for 30 min. Then changed the gas flow to
Ar (50 mL/min) and held for 30 min to clean the catalyst surface,
finally the H2-TPR was recorded with the 10 vol% H2/Ar flow.
Powder X ray diffraction (XRD) patterns were recorded on a
D2/max-RA X-ray diffractometer (Bruker, Germany), with Cu-K␣
radiation operated at 30 KV and 10 mA. The 2ꢁ angle was scanned
in the rage of 10–80◦.
H2-TPR experiments were carried out on the Auto Chem II 2920
equipment (Micromeritics, USA) with a TCD detector. The sample
(200 mg) was loaded in a quartz reactor and flushed with a 10 vol.%
H2/Ar flow at 50 ◦C. Then an isopropyl alcohol gel (−88 ◦C) cooled
trap was added to condense the water vapor. The hydrogen con-
sumption was monitored from 50 ◦C to 600 ◦C using a heating rate
of 10 ◦C/min.
C
C bond cleavage. Such properties of copper based catalysts are
suitable for the hydrogenolysis of glucose to glycols [22]. More-
over, the application of cheap metal can improve the economy of
the hydrogenolysis route, which is one of the main challenges of
the transformation process of renewable feedstocks [2]. As for the
acid sites, alumina is a typical industrial support with large amount
of acid sites on the surface. Furthermore, many studies have also
been performed on the development of acid sites on tungsten oxide
supported on alumina catalyst as a function of W surface den-
sity [23,24]. The studies turned out that the acid amount and acid
species were easily tuned by changing the surface density of W.
Furthermore, due to the strong interaction between the oxides, the
WOx phase is molecularly dispersed as a two-dimensional metal
oxide overlayer on a high surface area support oxide with a high
Raman spectra were recorded on a LabRAM HR800 System
equipped with a CCD detector at room temperature. The 532 nm
of the air-cooled frequency-doubled Nd-Yag laser was employed
as the exciting source with a power of 30 MW.
Please cite this article in press as: C. Liu, et al., WOx modified Cu/Al2O3 as a high-performance catalyst for the hydrogenolysis of glucose
to 1,2-propanediol, Catal. Today (2015), http://dx.doi.org/10.1016/j.cattod.2015.06.030

G Model
CATTOD-9687; No. of Pages12
ARTICLE IN PRESS
C. Liu et al. / Catalysis Today xxx (2015) xxx–xxx
3
UV–vis DRS spectra were collected on an Agilent cary 5000
UV–vis spectrometer with BaSO4 as a reference. The calcined cata-
lysts were measured in the 200–800 nm regions.
Cu/Al O
2
3
Cu-WO (0.2)/Al O
x
2
3
3
3
3
Cu-WO (0.4)/Al O
NH3 temperature-programmed desorption (NH3-TPD) mea-
surements were performed on an Auto Chem II 2920 equipment
(Mircromeritics, USA). Prior to the test, 200 mg catalyst was
reduced in a stream of pure H2 at 250 ◦C for 2 h. Subsequently,
the gas flow was changed to helium at 250 ◦C and then purged for
60 min to clean the surface. After cooling to 100 ◦C in the helium
flow, the sample was adsorbed with NH3 until saturation, then
flushed with helium to remove the physical adsorption for 1 h. The
NH3 desorption pattern was recorded from 100 ◦C to 800 ◦C at a
heating rate of 10 ◦C/min.
x
2
Cu-WO (0.6)/Al O
x
2
Cu-WO (0.8)/Al O
x
2
CO-FTIR spectra were collected using an infrared spectrome-
ter (VERTEX70, Bruker, Germany), equipped with KBr optics which
works at the liquid nitrogen environment. To realize the in situ mea-
surement for the adsorption of CO, the infrared cell equipped with
ZnSe windows was connected to a gas-feed system with a set of
stainless steel gas lines. The catalysts were reduced in situ at 250 ◦C
for 2 h in H2 flow (20 mL/min), followed by flushing the sample with
helium for 1 h at 250 ◦C. Then the CO-FTIR spectra were recorded
after cooling to 30 ◦C in helium flow.
The Py-FTIR spectra were obtained on a VERTEX 70 Bruker FT-
IR spectrometer. All the samples were reduced at 250 ◦C in H2 for
2 h and then degassed to a pressure of 10−2 MPa at 350 ◦C for 1 h.
Adsorption of pyridine was carried out after cooling down to 30 ◦C,
The Py-IR spectra were measured at 30 ◦C and 350 ◦C after evacua-
tion for 30 min.
0
10
20
30
Pore size (nm)
Fig. 1. The pore size distribution of calcined Cu-WOx/Al2O3 catalysts.
where m: mass, M: relative molecular weight, N: carbon numbers
in product.
More details about the analysis methods are in Supporting Infor-
mation.
3. Results
3.1. Physicochemical properties of catalysts
X-ray photoelectron spectroscopy (XPS) was performed on a
Thermal XPS ESCALAB 250Xi spectrometer with Al K␣ radiation
(1486.6 eV) and a multichannel detector. The obtained binding
energies were calibrated using the C 1s peak at 284.6 eV as the ref-
erence. The experiment error reported in this study is within 0.1 eV.
For the in situ reduction process, the sample was reduced at 250 ◦C
for 2 h with a ramp of 1 ◦C/min
The results of physicochemical properties of Cu-WOx/Al2O3 cat-
alysts are showed in Table 1. The copper loadings show little
difference for different catalysts. The surface density of W species
(W/nm2) was calculated from Eq. (1) [27,28]. The density of W
increased with the designed amount. The highest W density was
only 0.8 W/nm2, which was much lower than the theoretical ꢀ-
Al2O3 monolayer coverage (7 W/nm2) [29,30]. The specific surface
area and the pore structure of Cu-WOx/Al2O3 catalysts are shown in
Table 1 and Fig. 1. The BET surface area increased obviously with the
increasing W density from 175.6 m2/g of Cu/Al2O3 to 218.7 m2/g
of Cu-WOx(0.8)/Al2O3. Both the pore volume and the pore size
decreased slightly with increasing W density. Fig. 1 showed the
pore size distribution of Cu-WOx/Al2O3 catalysts. All the catalysts
presented broad peaks centered at about 6.3 nm and a narrow peak
centered at about 3.6 nm. The intensity of the peak at about 6.3 nm
decreased while that of the peak at about 3.6 nm increased with
increasing W density, indicating the penetration of WOx into large
pores to form pores of relatively small diameter. The metal dis-
persion is considered to be important parameter which strongly
influences the catalytic performance. As shown in Table 1, the
surface copper area decreased from 9.6 m2/gCat on Cu/Al2O3 to
6.0 m2/gCat on Cu-WOx(0.8)/Al2O3.
2.3. Catalytic reaction process
Hydrogenolysis of glucose was performed in a vertical fixed-bed
reactor (i.d. 12 mm, length 600 mm) with a cold trap. Prior to the
test, 2.0 g catalyst (20–40 mesh) was loaded at the isothermal zone
and in situ reduced at 250 ◦C for 2 h in hydrogen (100 mL/min). After
reduction, a 5 wt% aqueous glucose solution was pumped into the
reactor, and mixed with H2 co-feeding. The liquid products were
collected in a gas–liquid separator immersed in an ice-water trap.
Typical reaction conditions were as follows: 4 MPa H2 (45 mL/min),
5 wt% aqueous glucose solution, aqueous glucose solution flow
rates were set as 24, 19.2, and 14.4 mL/h, which can be shown as
0.6, 0.48, and 0.36 h−1 in WHSV (weight hourly space velocity).
The liquid products collected in the cold trap were analyzed
offline by an HPLC equipped with a RID detector. The tail gas prod-
ucts were analyzed by online gas chromatograph (GC) (models
6890N and 4890D; Agilent). The conversion of glucose and the
selectivity of products were calculated based on the total carbon
according to the following equations:
[WO3 loading (wt%/100)] × 6.02 × 105
ꢂW
=
(1)
BET surface area (m2/g) × 231.84
3.2. Structural properties of the catalysts
carbon mol of reactant after reaction
Conversion (%) = 100 −
XRD patterns of the calcined catalysts are displayed in Fig. 2.
All the catalysts showed the diffraction peaks of ꢀ-Al2O3 (2ꢁ = 37.4,
46.0 and 67.0◦). With the increase of W density, these peak inten-
sities for ꢀ-Al2O3 decreased notably. This phenomenon was due to
the coverage of WOx on the alumina surface. Two diffraction peaks
at 35.6◦ and 38.8◦ ascribed to CuO were detected. The peak inten-
sities declined with increasing W density and even disappeared on
Cu-WOx(0.8)/Al2O3. The results indicated that the doping of WOx
promoted the dispersion of Cu. However, the results were opposite
carbon mol of reactant in feedstock
× 100
(mproduct/Mproduct) × Ncarbon numbers in product
(mglucose in feedback − mglucose in products)/Mglucose × 6
Selectivity (%) =
× 100
Please cite this article in press as: C. Liu, et al., WOx modified Cu/Al2O3 as a high-performance catalyst for the hydrogenolysis of glucose
to 1,2-propanediol, Catal. Today (2015), http://dx.doi.org/10.1016/j.cattod.2015.06.030

G Model
CATTOD-9687; No. of Pages12
ARTICLE IN PRESS
4
C. Liu et al. / Catalysis Today xxx (2015) xxx–xxx
Table 1
The physicochemical properties of the calcined Cu-WOx/Al2O3 catalysts.
Catalyst
Metal loading
(mmol/gCat)a
ꢂw (W/nm2)
SBET (m2/g)b
VP (cm3/g)b
DP (nm)b
SCu (m2/gCat.)c
NH3 (␮mol/g)d
Cu
W
Cu/Al2O3
0.79
0.77
0.78
0.76
0.77
–
–
175.6
178.9
189.6
211.6
218.7
0.45
0.44
0.43
0.41
0.40
7.5
7.2
6.6
5.9
5.7
9.6
7.1
6.9
6.1
6.0
49.4
50.6
56.6
68.2
73.6
Cu-WOx(0.2)/Al2O3
Cu-WOx(0.4)/Al2O3
Cu-WOx(0.6)/Al2O3
Cu-WOx(0.8)/Al2O3
0.05
0.13
0.21
0.30
0.2
0.4
0.6
0.8
a
Metal loading determined using ICP.
BET method.
Calculated from the N2O-TPD results.
Calculated by NH3-TPD.
b
c
d
Table 2
Peak-fitting results of W 4f XPS spectra of catalysts.
a
Sample
Bingding energy for W 4f (eV)
W5+/(W5+ + W6+
)
W6+ (W 4f5/2
)
W6+ (W 4f7/2
)
W5+ (W 4f5/2
)
W5+ (W 4f7/2
35.0
35.1
34.9
34.9
35.7
35.7
)
Cu-WOx(0.2)/Al2O3
Cu-WOx(0.4)/Al2O3
Cu-WOx(0.6)/Al2O3
Cu-WOx(0.8)/Al2O3
WOx(0.6)/Al2O3
40.0
39.5
39.2
39.1
40.0
40.1
37.9
37.4
37.1
37.0
37.9
38.0
37.1
37.2
37.0
37.0
37.8
37.8
0.42
0.44
0.47
0.48
0.50
0.57
G-WOx(0.6)/Al2O3
a
Calculated according to the curve-fitting results of the W 4f XPS spectra of catalysts.
to the N2O chemisorption results, which showed that the Cu sur-
face area decreased with increasing W density. This observation
suggests perimeter of the copper particle was partially decorated
with WOx species, which have both dispersed effect and coverage
effect on Cu [31]. The XRD patterns did not show any information
about WOx species, which indicates the high dispersion of WOx on
the catalyst surface.
CuO particles. The Cu-WOx/Al2O3 catalysts presented a band at
945 cm−1. This band was ascribed to the symmetric W O stretch-
ing vibration mode of distorted isolated WO4 tetrahedra [34–36].
Furthermore, no shift of the surface W O band was observed with
increasing W density, which indicates that no polymerized WOx
species existed in all the Cu-WOx/Al2O3 catalysts [24]. The absence
of Raman bands in the 500–700 cm−1 region reflected the absence
Raman spectroscopy is suitable for the discrimination of
tungsten oxide species such as isolated surface WO4 species, poly-
tunstate species, WO3 crystalline NPS, which is strongly dependent
on the surface W density [31–33]. The Raman spectra of Cu-
WOx/Al2O3 catalysts are shown in Fig. 3. The ꢀ-Al2O3 support has
no Raman active modes [34]. Cu/Al2O3 catalyst presented three
characteristic bands of CuO at 293, 341, 629 cm−1. As the W den-
sity increasing, the intensity of these bands decreased obviously
and almost disappeared when W density increasing to 0.6 W/nm2.
It confirms that the addition of WOx promoted the dispersion of
or negligible proportion of bridge W O W bond in the catalysts
[37], which confirms the isolated WO4 structure of WOx species on
the catalyst surface.
UV–vis DRS is another useful tool to differentiate tungsten oxide
species deposited on Al2O3 surface. As shown in Fig. 4, only a sin-
gle ligand-to-metal charge transfer (LMCT) band at 232 nm was
observed in the UV–vis DRS spectra of the Cu-WOx/Al2O3 catalysts,
which is the unique feature of UV–vis DRS for the isolated WO4
reference compounds [37]. It confirms the result that only isolated
WO4 species are present on the Cu-WOx/Al2O3 catalyst.
CuO
945
e
d
Cu-WOx(0.8)/Al O
2
3
Cu-WOx(0.6)/Al O
2
3
c
b
a
293
341
Cu-WOx(0.4)/Al O
2
3
Cu-WOx(0.2)/Al O
2
3
Cu/Al O
2
3
20
40
60
80
o
2θ ( )
200
400
600
Raman shift (cm
800
1000
-1
)
Fig. 2. XRD patterns of calcined Cu-WOx/Al2O3 catalysts: (a) Cu/Al2O3, (b)
Cu-WOx(0.2)/Al2O3, (c) Cu-WOx(0.4)/Al2O3, (d) Cu-WOx(0.6)/Al2O3, (e) Cu-
WOx(0.8)/Al2O3.
Fig. 3. Raman spectra of the calcined Cu-WOx/Al2O3 catalysts.
Please cite this article in press as: C. Liu, et al., WOx modified Cu/Al2O3 as a high-performance catalyst for the hydrogenolysis of glucose
to 1,2-propanediol, Catal. Today (2015), http://dx.doi.org/10.1016/j.cattod.2015.06.030

G Model
CATTOD-9687; No. of Pages12
ARTICLE IN PRESS
C. Liu et al. / Catalysis Today xxx (2015) xxx–xxx
5
232
Cu-WOx(0.8)/Al2O3
Cu-WOx(0.6)/Al2O3
Cu-WOx(0.4)/Al2O3
Cu-WOx(0.2)/Al2O3
200
300
400
500
600
700
800
Wavelength (nm)
Fig. 4. UV–vis DRS spectra of the calcined Cu-WOx/Al2O3 catalysts.
3.3. Chemical state of the Cu-WOx/Al2O3 catalysts
XPS analysis of binding energies and intensities of the surface
species provides information on chemical state and relative quanti-
ties of the surface species. Fig. 5A shows the W 4f XPS spectra of the
different calcined Cu-WOx/Al2O3 catalysts. All the catalysts show
two broad XPS peaks in the range of 32–42 eV. Deconvolutions
were performed to distinguish WOx species in different chemical
states from the position of the W 4f level by a curve-fitting pro-
cedure. The deconvoluted results shows two doublets with the W
4f7/2 B.E. values at 35.0 and 37.1 eV, which are ascribed to W5+ and
W6+ species [38,39]. Table 2 also gives the quantitative results of
the molar ratio of W5+/(W5+ + W6+). The results displays that the
surface W5+ species contents increased with increasing WOx den-
sity, which indicates that electronic interactions exist between Cu
and WOx with an intimate contact [30]. The photoelectron peak
at 932.9 eV with a strong satellite peak at ca. 943 eV in Cu 2p3/2
region (Fig. 5B) suggests that the predominant species is CuO on
the calcined Cu-WOx/Al2O3 surface [40]. Furthermore, the Cu 2p3/2
XPS spectra and Cu LMM XAES spectra of the reduced Cu/Al2O3
and Cu-WOx(0.8)/Al2O3 catalysts were also recorded to confirm
the electronic interaction between Cu and WOx species (Fig. 5C
and D). The photoelectron peak at 932.3 eV without the presence
of satellite peak indicates that Cu2+ was reduced to Cu0 and/or Cu+,
which could be further differentiated through XAES spectra [41].
The Cu LMM XAES spectra show that the relative amount of Cu+ to
(Cu+ + Cu0) increased with the addition of WOx species. The results
confirm that strong electronic interaction is present between Cu
and WOx species.
Cu-WOx(0.8)/Al2O3
Cu+/(Cu++Cu0) = 65.1%
D
3.4. Reducibility and surface acidic properties
Cu/Al2O3
Cu+/(Cu++Cu0) = 54.6%
In order to reveal the reduction performance of these catalysts,
the H2-TPR profiles were measured. As displayed in Fig. 6, the
Cu/Al2O3 catalyst showed a strong reduction peak at around 226 ◦C
and a weak shoulder peak at 276 ◦C. The peak at lower tempera-
ture was assigned to the reduction of higher dispersed CuO, while
the peak at higher temperature was ascribed to the reduction of
large CuO particles with slightly interaction with the support, as
reported previously [11,42]. Furthermore, the reduction of the bulk
CuO particles was also performed, which displayed an even higher
reduction temperature at about 344 ◦C. With the increase of W
density, the intensity of the peak at higher temperature decreased
gradually, indicating the decrease of the amount of large CuO par-
ticles. The result was consistent with the XRD and Raman results.
The low temperature peak at about 226 ◦C shifted to a higher
910
912
914
916
918
920
922
924
Kinetic Energy (eV)
Fig. 5. (A) W 4f XPS spectra and (B) Cu 2p XPS spectra of the calcined Cu-WOx/Al2O3
catalysts, (C) Cu 2p XPS spectra and (D) Cu LMM XAES spectra of the reduced
Cu/Al2O3 and Cu-WOx(0.8)/Al2O3 catalysts.
Please cite this article in press as: C. Liu, et al., WOx modified Cu/Al2O3 as a high-performance catalyst for the hydrogenolysis of glucose
to 1,2-propanediol, Catal. Today (2015), http://dx.doi.org/10.1016/j.cattod.2015.06.030

G Model
CATTOD-9687; No. of Pages12
ARTICLE IN PRESS
6
C. Liu et al. / Catalysis Today xxx (2015) xxx–xxx
A
Cu-WO (0.8)/Al O
2 3
x
Cu-WOx(0.8)/Al2O3
Cu-WOx(0.6)/Al2O3
Cu-WOx(0.4)/Al2O3
Cu-WO (0.6)/Al O
x
2
3
3
3
Cu-WO (0.4)/Al O
x
2
Cu-WOx(0.2)/Al2O3
Cu/Al2O3
Cu-WO (0.2)/Al O
x
2
CuO
Cu/Al O
2
3
100
200
300
400
500
600
100
200
300
400
500
600
700
800
Temperature (oC)
Temperature (oC)
B
Fig. 6. H2-TPR profiles of the Cu-WOx/Al2O3 catalysts.
30 oC
350 oC
1447
temperature when W density increased to 0.6 and 0.8 W/nm2. It
was reported that the highly interacting Cu2+ species were more
difficult to be reduced than the weakly interacted CuO particles
[42]. Therefore, the shift of the reduction temperature could be
due to the strong electronic interactions between CuO and WOx.
These results were correlated well with the XRD and XPS results.
No reduction peak of WOx was detected below 600 ◦C. Horsley
reported that the highly dispersed WOx species were not easily
reduced, and the reduction temperature was often above 900 ◦C
according to the literature [33,39].
1610
1450
1498
1489
e
e
d
c
d
c
It is well known that the catalyst acidity plays an important role
in determining the product distribution in polyols hydrogenolysis
[18,19]. Thus, NH3-TPD performances were conducted, as shown
in Fig. 7A. The desorbed NH3 was monitored with a MS detector. In
order to eliminate the effect of water (Fig. S1), iron current at m/z 16
was used to calculate the desorbed NH3 amount. Cu/Al2O3 showed
a broad peak from 130 to 400 ◦C which centered at 190 ◦C, revealing
that the surface acid strength has a wide distribution on the cata-
lyst [43]. The main ammonia desorption peak before 350 ◦C was
attributed to weak acid sites while the small ammonia desorption
peak in the range of 350–400 ◦C was attributed to stronger acid sites
[44]. The results showed that there were mainly weak acid sites
on the Cu-WOx/Al2O3 catalysts. The ammonia desorption intensity
improved gradually with increasing surface W density. The con-
centration of the total acid amount per gram of catalyst gives the
following increasing order: Cu/Al2O3 < Cu-WOx(0.2)/Al2O3 < Cu-
WOx(0.4)/Al2O3 < Cu-WOx(0.6)/Al2O3 < Cu-WOx(0.8)/Al2O3.
The Py-FTIR spectra of reduced catalysts evacuated at 30 ◦C and
350 ◦C were recorded to measure the acidity of Cu-WOx/Al2O3 cat-
alysts, as shown in Fig. 7B. The bands at 1447 cm−1, 1489 cm−1
and 1610 cm−1 were indicative of coordination of Py molecules to
Lewis acid sites and consequent formation of LPy species. Moreover,
the decrease of the intensity and the increase in the frequency of
the maximum after evacuation at subsequent higher temperatures
indicate that there are both weak and strong Lewis acid sites on the
surface of the samples [45,46]. No band at 1540 cm−1 was detected
in the spectra of the Cu-WOx/Al2O3 catalysts, indicating the absence
of Brønsted acid sites on the catalyst surface. The generated Lewis
acid sites can be ascribed to the coordinatively unsaturated Wx+
cations [34]. It is reported that the surface W density plays a key
role in the formation of Brønsted acid sites which requires the
delocalization of the negative charge. However, the isolated WOx
species cannot delocalize the negative charge, while it occurs over
the extended network of polytunstate species [30]. Herein, it can
b
a
b
a
1400 1450 1500 1550 1600 1650 1700 1400 1450 1500 1550 1600 1650 1700
Wavenumber (cm-1)
Fig. 7. (A) NH3-TPD and (B) Py-FTIR patterns of the reduced Cu-WOx/Al2O3
catalysts: (a) Cu/Al2O3, (b) Cu-WOx(0.2)/Al2O3, (c) Cu-WOx(0.4)/Al2O3, (d) Cu-
WOx(0.6)/Al2O3, (e) Cu-WOx(0.8)/Al2O3.
be the explanation for the absence of Brønsted acid sites at low W
surface density.
3.5. The hydrogenolysis of glucose on Cu-WOx/Al2O3 catalysts
The hydrogenolysis of glucose was performed in a continuous
fixed-bed reactor at 180 ◦C and 4 MPa, as shown in Fig. 8 and Table 3.
In the conversion of glucose, only polyols were detected in the liq-
uid products. Moreover, sorbitol, glycerol and 1,2-PDO were the
main products, small amount of other polyols such as erythritol,
EG, 1,2-butanediol (1,2-BDO) and ethanol were observed, while
trace amount of isopropanol, propanol and 2-butanol were also
detected. It was interesting to note that all the polyols were C2,
C3, C4 and C6 products, no product with one or five carbon num-
bers was detected. Moreover, no product was detected in gas phase.
It is inferred that the C C bond cleavage occurs mainly at C2 C4 or
C3 C3 bond.
The doping of WOx on the Cu-WOx/Al2O3 catalysts influenced
not only the activity of glucose hydrogenolysis but also the selec-
tivity of products. The conversion of glucose as a function of
WHSV is illustrated in Fig. 8A. The activity of glucose conver-
sion increased first and then decreased slightly with increasing
W density. Both the selectivities of sorbitol and glycerol showed
Please cite this article in press as: C. Liu, et al., WOx modified Cu/Al2O3 as a high-performance catalyst for the hydrogenolysis of glucose
to 1,2-propanediol, Catal. Today (2015), http://dx.doi.org/10.1016/j.cattod.2015.06.030

G Model
CATTOD-9687; No. of Pages12
ARTICLE IN PRESS
C. Liu et al. / Catalysis Today xxx (2015) xxx–xxx
7
A
C
95
15
10
90
Cu/Al O
Cu/Al O
2
3
2
3
Cu-WO (0.2)/Al O
Cu-WO (0.2)/Al O
x
2
3
3
3
3
x
2
3
3
3
3
5
0
Cu-WO (0.4)/Al O
Cu-WO (0.4)/Al O
x
2
x
2
Cu-WO (0.6)/Al O
Cu-WO (0.6)/Al O
x
2
x
2
Cu-WO (0.8)/Al O
Cu-WO (0.8)/Al O
x
2
x
2
85
0.60
0.55
0.50
0.45
0.40
0.35
0.60
0.55
0.50
0.45
0.40
0.35
WHSV (h-1)
WHSV (h-1)
55
50
45
40
35
30
25
20
D
Cu/Al O
2 3
B
Cu-WO (0.2)/Al O
30
25
20
15
x
2
3
3
3
3
Cu-WO (0.4)/Al O
x
2
Cu-WO (0.6)/Al O
x
2
Cu-WO (0.8)/Al O
x
2
Cu/Al O
2
3
Cu-WO (0.2)/Al O
x
2
3
3
3
3
Cu-WO (0.4)/Al O
x
2
Cu-WO (0.6)/Al O
x
2
Cu-WO (0.8)/Al O
x
2
10
0.60
0.55
0.50
WHSV (h-1)
0.45
0.40
0.35
0.60
0.55
0.50
0.45
0.40
0.35
WHSV (h-1)
Fig. 8. (A) Glucose conversion, (B) sorbitol selectivity, (C) glycerol selectivity and (D) 1,2-PDO selectivity in glucose hydrogenolysis on the Cu-WOx/Al2O3 catalysts.
an obvious decrease with the increase of W density (Fig. 8B
and C). However, as shown in Fig. 8D, the selectivity of 1,2-PDO
showed a completely opposite changing trend, which increased
with increasing W density. The highest 1,2-PDO selectivity of 55.4%
was obtained at WHSV = 0.36 h−1 on Cu-WOx(0.8)/Al2O3 with glu-
cose conversion of 95.0%, which is one of the best performances
reported in this reaction. This information suggested that the dop-
ing of WOx promoted the conversion of glucose and the selectivity
of 1,2-PDO.
3.6. Correlation between the properties and performance of the
Cu-WOx/Al2O3 catalysts
Tomishige’s work has shown that transition metals such as Pt,
Rh, and Ir have strong interactions with MOx (W, Mo, Re and so
on), and the combination of metal sites and metal oxide species is
efficient to the hydrogenolysis of C-O bond [47–53]. In this work,
the N2O-TPD, XRD, Raman, H2-TPR, and XPS results turn out that
WOx species have coverage, dispersion, and electronic effects on
Table 3
The product distribution of glucose hydrogenolysis on Cu-WOx/Al2O3 catalysts.a
Catalyst
WHSV (h−1
)
Conv. (%)
Selectivity (%)
Sorbitol
Erythritol
EG
Glycerol
1,2-PDO
1,2-BDO
Ethanol
Othersb
Cu/Al2O3
0.60
0.48
0.36
86.3
90.8
93.9
31.9
22.6
16.3
5.1
4.7
4.3
6.6
7.2
8.1
18.1
16.8
15.9
29.2
35.9
43.0
2.7
3.8
4.6
1.7
2.1
2.4
4.7
6.9
5.4
Cu-WOx(0.2)/Al2O3
Cu-WOx(0.4)/Al2O3
Cu-WOx(0.6)/Al2O3
0.60
0.48
0.36
90.1
92.2
93.9
26.7
19.3
13.8
5.3
4.9
4.2
7.2
7.5
7.4
15.8
14.2
11.9
37.5
42.3
44.0
3.1
4.3
5.0
2.1
2.1
2.3
2.3
5.4
11.4
0.60
0.48
0.36
93.4
93.7
94.9
21.6
16.5
12.9
5.5
5.1
4.7
8.1
8.1
8.2
15.0
13.1
11.7
42.5
46.9
50.4
4.4
5.4
6.2
1.9
2.6
3.5
1.0
2.3
2.4
0.60
0.48
0.36
94.6
94.5
95.2
14.4
11.1
9.0
5.5
5.0
4.3
8.5
8.1
8.0
12.1
10.1
9.0
50.2
52.3
54.5
6.0
7.0
8.0
2.2
4.0
3.7
1.1
2.4
3.5
Cu-WOx(0.8)/Al2O3
0.60
0.48
0.36
92.7
93.3
95.0
11.8
9.8
8.3
5.5
5.0
4.4
8.0
7.8
8.0
9.9
8.8
8.1
51.1
52.1
55.4
7.0
7.5
8.5
2.6
4.6
3.8
4.1
4.4
3.5
a
Reaction conditions: 4 MPa H2, 180 ◦C, 5 wt% glucose aqueous solution.
Others: isopropyl alcohol, propanol, 2-butanol, humins, etc.
b
Please cite this article in press as: C. Liu, et al., WOx modified Cu/Al2O3 as a high-performance catalyst for the hydrogenolysis of glucose
to 1,2-propanediol, Catal. Today (2015), http://dx.doi.org/10.1016/j.cattod.2015.06.030

G Model
CATTOD-9687; No. of Pages12
ARTICLE IN PRESS
8
C. Liu et al. / Catalysis Today xxx (2015) xxx–xxx
Cu metal sites and have an electronic effect on it. The variation
Cu
A
e∗
Cu
of the properties of the Cu-WOx/Al2O3 catalysts could lead to dif-
ferent catalytic performances. As reported by Huber [18,19], the
reaction of polyols hydrogenolysis was regulated by the synergis-
tic catalysis of metal and acid sites. A large amount of literature had
reported the linear correlation between the ratio of metal surface
area/acid amount and its behaviour on polyols hydrogenolysis such
as glycerol, dimethyl oxalate hydrogenation, 5-methylfurfuryl alco-
hol hydrogenolysis, and carbon dioxide hydrogenation to methanol
[11,55–57]. For example, a linear relationship between the yield
of 1,2-PDO from glycerol and copper metal surface area or the
Lewis acid amount was widely reported [11,58]. Moreover, surface
copper and Lewis acid sites are important factors in the structure-
performance relationship for polyols hydrogenolysis considering
their activity on the elemental reactions involving C C/C O bond
cleavage. The highest 1,2-PDO selectivity on Cu-WOx(0.8)/Al2O3
was suggested to be related with the proper Lewis acid/metal
surface molar ratio. A similar relationship between the product
selectivity and the ratio of Lewis acid to copper site amount was
proposed and shown in Fig. 10A. When WHSV = 0.6 h−1, sorbitol
and glycerol selectivities decreased linearly with increasing the
molar ratio of Lewis acid to metal sites. Meanwhile, the 1,2-PDO
selectivity increased linearly with the increase of the molar ratio
of Lewis acid to metal sites. Actually, the correlation between the
product selectivities and the molar ratio of Lewis acid to metal sites
in WHSV = 0.48 and 0.36 h−1 also displayed the same linear rela-
tionship, as shown in Fig. 10B and Fig. 10C. The correlations strongly
suggest that metal and Lewis acid sites are the active sites for glu-
cose hydrogenolysis in H2 atmosphere, and the proper molar ratio
of Lewis acid to metal sites was crucial for the efficient hydrogenol-
ysis of glucose to 1,2-PDO. The correlations also indicate that the
increase of molar ratio of Lewis acid to metal sites promotes the
cleavage of C C/C O bonds. In order to identify the stability and
the real acidity of the catalysts in the hydrothermal environment,
a special Py-FTIR experiment was designed (Fig. S2 and Table S1
of Supporting Information). The samples were reduced at 250 ◦C in
H2 for 2 h and then degassed to a pressure of 10−2 MPa at 350 ◦C
for 1 h. After cooling down to 30 ◦C, vapor was adsorbed for 30 min.
After evacuation for 30 min, Pyridine was adsorbed for 30 min. The
Py-IR spectra were measured at 30 ◦C and 180 ◦C after evacuation
for 30 min. The results turn out that very small amount of Brønsted
acid sites were generated when the catalysts were exposed to the
vapor environment. The ratios of Brønsted to Lewis acid were only
0.091 and 0.018 on Cu/Al2O3 and Cu-WOx(0.8)/Al2O3, respectively
after evacuation at 30 ◦C. The Brønsted acid sites on the surface
of the two catalysts nearly disappeared after evacuation at 180 ◦C.
The results confirm that Lewis acid sites on Cu-WOx/Al2O3 are the
catalytic active sites in the hydrothermal condition. The promoting
mechanism of WOx species on glucose hydrogenolysis to 1,2-PDO
will be attempted to discuss below.
e
d
c
b
a∗
a
30
40
50
60
70
80
2θ (o)
B
2093.3
2088.5
Cu-WOx(0.8)/Al2O3
Cu/Al2O3
2000
2040
2080
Wavenumber (cm-1)
2120
2160
2200
Fig. 9. (A) The XRD patterns of the spent (a, b, c, d, e) and reduced
(a*, e*) Cu-WOx/Al2O3 catalysts and (B) CO-FTIR of the reduced Cu/Al2O3
and Cu-WOx(0.8)/Al2O3 catalysts. (a) Cu/Al2O3, (b) Cu-WOx(0.2)/Al2O3, (c) Cu-
WOx(0.4)/Al2O3, (d) Cu-WOx(0.6)/Al2O3, (e) Cu-WOx(0.8)/Al2O3, (a*) Cu/Al2O3, (e*)
Cu-WOx(0.8)/Al2O3. a* and e* were reduced at 250 ◦C for 2 h with a ramp rate of
1
◦C/min, and protected with liquid paraffin.
copper sites. The XRD patterns of the reduced and used catalysts
were also obtained to confirm the strong interaction between Cu
sites and WOx species (Fig. 9A). The results show that the diffrac-
tion peaks at 43.3 and 50.5◦ are present on the used catalysts
which can be ascribed to Cu0. Furthermore, the intensity of the two
peaks decreased with the increasing WOx surface density. The XRD
pattern of the reduced Cu/Al2O3 catalyst also shows the Cu0 diffrac-
tion peaks. However, the diffraction peaks of Cu0 disappeared in
the XRD pattern of Cu-WOx(0.8)/Al2O3 catalyst. These XRD results
confirm that WOx species have dispersion effect on Cu particles.
The CO-FTIR spectra of Cu/Al2O3 and Cu-WOx(0.8)/Al2O3 catalysts
were also recorded (Fig. 9B). CO is used as a probe molecular to
study the electronic environment of the supported metal. Prior to
the measurements, the catalysts were reduced at 250 ◦C for 2 h.
Generally, the band in the range of 2025–2125 cm−1 arises from
linear bound CO molecules on different Cu sites with a prevailing
metal character [54]. The frequency of CO adsorbed on Cu sites over
Cu-WOx(0.8)/Al2O3 catalyst surface shows an obvious blue shift
compared with that over Cu/Al2O3, This may be due to the decrease
of electron density in Cu0 nucleus which is consistent with the XPS
results. It suggests that WOx species have an intimate contact with
3.7. Promoting mechanism of WOx on glucose hydrogenolysis to
1,2-PDO
The decrease of the selectivities of sorbitol and glycerol and
the simultaneous increase of 1,2-PDO selectivity with increasing
WHSV indicate that the hydrogenolysis of glucose to 1,2-PDO may
be a cascade reaction with sorbitol and glycerol as the interme-
diates. In order to clear the reaction route of glucose conversion
to 1,2-PDO on Cu-WOx/Al2O3, the hydrogenolysis of the interme-
diates was performed on Cu-WOx(0.6)/Al2O3 at WHSV = 0.6 h−1
(Fig. S3 in Supporting Information). However, Cu-WOx(0.6)/Al2O3
shows no activity for the hydrogenolysis of the saturated polyols
such as sorbitol, erythritol, glycerol, 1,2-PDO, and EG. It indicates
that the saturated polyols are the terminal products and cannot
undergo the hydrogenolysis reaction. So 1,2-PDO did not come from
Please cite this article in press as: C. Liu, et al., WOx modified Cu/Al2O3 as a high-performance catalyst for the hydrogenolysis of glucose
to 1,2-propanediol, Catal. Today (2015), http://dx.doi.org/10.1016/j.cattod.2015.06.030

G Model
CATTOD-9687; No. of Pages12
ARTICLE IN PRESS
C. Liu et al. / Catalysis Today xxx (2015) xxx–xxx
9
120
WHSV = 0.6
50
Conversion
1,2-PDO selectivity
100
80
60
40
20
0
40
30
20
10
Sorbitol
Glycerol
1,2-PDO
5
6
7
8
9
10
11
12
Molar ratio of acid to metal sites (μmol /m2 Cu)
Cu/Al O
Cu-WOx(0.6)/Al2O3 Cu/Al2O3 Cu-WOx(0.6)/Al2O3
2
3
Glucose
Fructose
Fig. 11. The hydrogenolysis of glucose and fructose on Cu/Al2O3 and Cu-
WOx(0.6)/Al2O3.
WHSV = 0.48
50
40
30
20
10
glucose at a lower temperature of 100 ◦C was performed (Table S2
of Supporting Information). The results showed that only fructose
and sorbitol were produced. Furthermore, the yields of fructose
and sorbitol increased with the addition of WOx species. It sug-
gests that WOx species can promote the activity of Cu metal sites
for C O bond hydrogenation. The C O bond hydrogenation activity
is related with the interaction between Cu sites and WOx species.
As shown in Table 1, the Cu surface areas decreased with increasing
W surface density, indicating the decrease of hydrogenation sites.
Therefore, we suppose that the increased hydrogenation activity of
Cu may be dependent on the electron interaction between Cu sites
and WOx species as proved by XPS and CO-FTIR results. The isomer-
ization reaction was also promoted by WOx species. Therefore, the
decrease of sorbitol and glycerol selectivities could be due to the
promoting effect of WOx on glucose isomerization to fructose and
fructose conversion which is competitive with glucose conversion
to sorbitol and glycerol. As mentioned above, the hydrogenolysis of
glucose to 1,2-PDO mainly includes two steps: (1) the isomerization
of glucose to fructose; (2) the hydrogenolysis of fructose to 1,2-PDO.
In order to confirm the reaction route of glucose hydrogenolysis to
1,2-PDO on Cu-WOx/Al2O3 and reveal the effect of WOx on the dif-
ferent reaction steps of glucose hydrogenolysis, the hydrogenolysis
of fructose was conducted on Cu/Al2O3 and Cu-WOx(0.6)/Al2O3 at
WHSV = 0.6 h−1 in Fig. 11. Compared with glucose hydrogenolysis,
Cu/Al2O3 shows extremely high conversion and 1,2-PDO selectiv-
ity in fructose hydrogenolysis with the value of 94.0% and 57.3%,
respectively, which is much higher than that of glucose conversion.
It suggests that fructose has high activity and selectivity for 1,2-PDO
and the isomerisation of glucose to fructose could be the rate-
controlling step in glucose hydrogenolysis to 1,2-PDO on Cu/Al2O3.
Furthermore, Cu/Al2O3 and Cu-WOx(0.6)/Al2O3 show nearly the
same fructose conversion (94.0 and 94.9%) and 1,2-PDO selectivity
(57.3 and 59.5%). It indicates that the addition of WOx mainly pro-
motes the isomerization of glucose to fructose and has little effect
on fructose hydrogenolysis.
Sorbitol
Glycerol
1,2-PDO
5
6
7
8
9
10
11
12
Molar ratio of acid to metal sites (μmol /m2 Cu)
60
50
40
30
20
10
WHSV = 0.36
Sorbitol
Glycerol
1,2-PDO
5
6
7
8
9
10
11
12
13
Molar ratio of acid to metal sites (μmol /m2 Cu)
Fig. 10. 1,2-PDO selectivity as a function of the molar ratio of Lewis acid to copper
surface area: Reaction condition: 180 ◦C, 4 MPa, 5 wt% glucose solution, WHSV = 0.6,
As generally accepted that the isomerization reaction can be
promoted by Lewis acid in the catalyst by polarizing the car-
bonyl group in the ketone [59–61]. Davis et al. [59] showed that
Sn incorporated in the framework of zeolite Beta, as Lewis acid,
performed the isomerization reaction following an intramolecu-
lar hydride shift mechanism between the carbonyl-containing C-1
and hydroxyl-bearing C-2 of glucose by a way of a 5-member
complex. As discussed above, the isolated WO4, as W5+ or W6+
0.48, and 0.36 h−1
.
the hydrogenolysis of saturated polyols such as sorbitol, erythri-
tol, and glycerol on Cu-WOx/Al2O3. Therefore, sorbitol and glycerol
could not be the intermediates for the hydrogenolysis of glucose
to 1,2-PDO. In order to examine the activities of Cu/Al2O3 and
Cu-WOx/Al2O3 for C O bond hydrogenation, the hydrogenation of
Please cite this article in press as: C. Liu, et al., WOx modified Cu/Al2O3 as a high-performance catalyst for the hydrogenolysis of glucose
to 1,2-propanediol, Catal. Today (2015), http://dx.doi.org/10.1016/j.cattod.2015.06.030

G Model
CATTOD-9687; No. of Pages12
ARTICLE IN PRESS
10
C. Liu et al. / Catalysis Today xxx (2015) xxx–xxx
Scheme 1. Promoting mechanism of the isolated WO4 in hexavalent on glucose isomerization to fructose.
species, provides more Lewis acid sites. Therefore, the mechanism
of the isomerization of glucose to fructose on Wx+ may be the
same as that on Snx+ considering their same tetrahedral struc-
ture and Lewis acid role in the reaction step. The XPS spectra of
WOx/Al2O3 with/without glucoses adsorption were obtained to
explore the exact promoting mechanism, as shown in Fig. 12. Base
on the deconvolution of W 4f region of the XPS spectra in the range
of 32–42 eV, both the W5+ and W6+ species existed on the cata-
lyst surface. However, the content of W5+ species on WOx/Al2O3
with glucoses adsorption was higher than that without glucoses
adsorption as shown in Table 2. It seems that WOx was ‘reduced’
by glucose. Therefore, we suppose that glucose may adsorb on the
W6+ species and a Glucose-W6+complex is formed. Moreover, the
adsorption of glucose on W6+ species could promote the isomeriza-
tion of glucose to fructose. Therefore, the promoting mechanism of
W6+ species on glucose isomerization was proposed in Scheme 1.
hydrogenolysis on Cu-WOx/Al2O3 was proposed in Scheme 2 based
on the hydrogenolysis of glucose and the intermediates. Partial
reaction procedures were similar with those illustrated in super-
critical water or subcritical water under hydrogen atmosphere
[20,21,62]. The hydrogenation of glucose to sorbitol is a compet-
itive reaction with the isomerization of glucose to fructose and
the hydrogenolysis of glucose to lower polyols. It is catalyzed by
the hydrogenation of copper. As identified in many reports, the
isomerization between glucose and fructose could occur at super-
critical/subcritical water even with no catalysts [20]. Furthermore,
the isomerization can occur by intramolecular hydride shift in the
present of Lewis acid [59,60]. The new Lewis acid produced by WOx
can promote the isomerization of glucose to fructose. As reported
previously, the hydrogenation of fructose mainly produced man-
nitol. However, it could not be differentiated with the sorbitol in
our experiments. The retro-aldol reaction of glucose to erythrose
and glycolaldehyde and the retro-aldol reaction of fructose to dihy-
droxyacetone and glyceraldehydes intermediates were performed
on the surface copper sites in H2 atmosphere. The hydrogena-
tion of these unsaturated intermediates to lower polyols such as
erythritol, EG, glycerol was also carried out on the copper sites.
The dehydration followed by hydrogenation of these intermediates
was speculated to be conducted on the acid sites and copper sites,
respectively. Polyols such as 1,2-PDO, 1,2-BDO, and erythritol were
obtained. It was confirmed that 1,2-BDO and EG did not come from
erythritol, which could be from the hydrogenolysis of erythrose and
glycolaldehyde, respectively. Furthermore, 1,2-PDO was only from
the hydrogenolysis of the dihydroxyacetone and glyceraldehyde,
which is different from Kanie’s report.
3.8. The possible reaction route of glucose hydrogenolysis on
Cu-WOx/Al2O3
The reaction route of glucose hydrogenolysis has been rarely
studied. Kanie [62] had studied the hydrogenolysis of glucose on Pt
nanoparticles and proposed a reaction route. Based on the model
reactions of fructose and glycerol, they suggested that 1,2-PDO
was obtained from the hydrogenolysis of glycerol and the unsat-
urated intermediates that were from the retro-aldol condensation
of fructose. In this work, the main reaction pathway of glucose
Regarding the synergy of Cu metal sites and WOx species on 1,2-
PDO selectivity, it is thought that the interface between Cu metal
surface and WOx species could play an important role on C-O bond
hydrogenolysis. Tomishige’s group studied the hydrogenolysis of
ethers and polyols over Rh(Ir)-ReOx/SiO2 catalysts and proposed
mechanisms of the hydrogenolysis of C-O bond over the metal and
metal oxide interface [63–65]. Take the hydrogenolysis of glycerol
to 1,2-PDO over Ir-ReOx/SiO2 for example, the interface between
Ir metal surface and ReOx clusters is the catalytically active site.
Glycerol is adsorbed on the ReOx clusters at the terminal oxygen
to form alkoxide species. H2 is adsorbed on Ir metal sites and dis-
sociated into a proton and a hydride. The hydride species attacks
the 3-position of the alkoxide to break the C O bond. Considering
the synergy of Cu and WOx species on glucose hydrogenolysis, a
similar reaction mechanism was proposed and shown in Scheme 3.
Tomishige’s group also found out that high valent Re and W was
efficient to the removal of two C O bonds [66,67]. In our work,
1,2-BDO selectivity was also promoted by WOx species. Therefore,
a reasonable mechanism on erythrose hydrogenolysis to 1,2-BDO
was proposed and shown in Scheme 4.
G-WO (0.6)/Al O
2 3
x
WO (0.6)/Al O
2 3
x
32
34
36
38
40
42
44
46
Binding Energy (eV)
Fig. 12. W 4f XPS spectra of the WOx(0.6)/Al2O3 catalyst with/without glucose
adsorption.
Please cite this article in press as: C. Liu, et al., WOx modified Cu/Al2O3 as a high-performance catalyst for the hydrogenolysis of glucose
to 1,2-propanediol, Catal. Today (2015), http://dx.doi.org/10.1016/j.cattod.2015.06.030

G Model
CATTOD-9687; No. of Pages12
ARTICLE IN PRESS
C. Liu et al. / Catalysis Today xxx (2015) xxx–xxx
11
Scheme 2. The possible reaction route of glucose hydrogenolysis on Cu-WOx/Al2O3.
Scheme 3. Proposed mechanism of dihydroxyacetone and glyceraldehyde hydrogenolysis to 1,2-PDO over Cu-WOx/Al2O3.
Scheme 4. Proposed mechanism of erythrose hydrogenolysis to 1,2-BDO over Cu-WOx/Al2O3.
4. Conclusion
with the increase of the molar ratio of acid to metal sites. The
highest 1,2-PDO selectivity was obtained on Cu-WOx(0.8)/Al2O3
at WHSV = 0.36 h−1, with the value of 55.4%. The adsorption of
glucose on isolated WO4 in hexavalent promotes the isomer-
ization of glucose to fructose which has higher selectivity to
1,2-PDO. The promoting effect of W6+ on glucose isomerization
to 1,2-PDO was proposed. A proper reaction route of glucose
hydrogenolysis on Cu-WOx/Al2O3 in H2 atmosphere was pro-
posed. The mechanisms of the C O bond hydrogenolysis were
discussed.
The hydrogenolysis of glucose was conducted on a series of Cu-
WOx/Al2O3 catalysts. WOx species are present on Cu-WOx/Al2O3
as isolated WO4. The isolated WO4 species provide more Lewis
acid on the catalysts and have shielding, dispersion, electronic
effect on Cu sites. A strong correlation between catalyst proper-
ties and performance was observed. The selectivity of 1,2-PDO
increased linearly with the increase of molar ratio of acid to
metal sites, while the selectivity of sorbitol and glycerol decreased
Please cite this article in press as: C. Liu, et al., WOx modified Cu/Al2O3 as a high-performance catalyst for the hydrogenolysis of glucose
to 1,2-propanediol, Catal. Today (2015), http://dx.doi.org/10.1016/j.cattod.2015.06.030

G Model
CATTOD-9687; No. of Pages12
ARTICLE IN PRESS
12
C. Liu et al. / Catalysis Today xxx (2015) xxx–xxx
Acknowledgement
[32] S.S. Chan, I.E. Wachs, L.L. Murrell, N.C. Dispenziere, J. Catal. 92 (1985) 1–10.
[33] J.A. Horsley, I.E. Wachs, J.M. Brown, G.H. Via, F.D. Hardcastle, J. Phys. Chem. 91
(1987) 4014–4020.
[34] X. Wu, L. Zhang, D. Weng, S. Liu, Z. Si, J. Fan, J. Hazard. Mater. 225–226 (2012)
146–154.
[35] R. Moraesa, K. Thomasa, S. Thomasa, S.V. Donkb, G. Grasso, J.P. Gilsona, M.
Houallaa, J. Catal. 286 (2012) 62–77.
This work was financially supported by the Major State Basic
Research Development Program of China (973 Program) (No.
2012CB215305).
[36] N. Liu, S. Ding, Y. Cui, N. Xue, L. Peng, X. Guo, W. Ding, Chem. Eng. Res. Des. 91
(2013) 573–580.
[37] E.I. Ross-Medgaarden, I.E. Wachs, J. Phys. Chem. C 111 (2007) 15089–15099.
[38] M. Occhiuzzi, D. Cordischi, D. Gazzoli, M. Valigi, P.C. Heydorn, Appl. Catal. A:
Gen. 269 (2004) 169–177.
[39] X. Yang, R. Gao, W. Dai, K. Fan, J. Phys. Chem. C 112 (2008) 3819–3826.
[40] A.J. McCue, C.J. McRitchie, A.M. Shepherd, J.A. Anderson, J. Catal. 319 (2014)
127–135.
Appendix A. Supplementary data
Supplementary data associated with this article can be found,
in the online version, at http://dx.doi.org/10.1016/j.cattod.2015.06.
030
[41] G. Corro, S. Cebada, U. Pal, J.L.G. Fierro, J. Alvarado, Appl. Catal. B: Environ. 165
(2015) 555–565.
[42] Z. Huang, F. Cui, H. Kang, J. Chen, C. Xia, Appl. Catal. A: Gen. 366 (2009)
288–298.
[43] A.H. Karima, S. Triwahyonoa, A.A. Jalilb, H. Hattori, Appl. Catal. A: Gen.
433–434 (2012) 49–57.
[44] L. Vilcocq, R. Koerin, A. Cabiac, C. Especel, S. Lacombe, D. Duprez, Appl. Catal.
B: Environ. 148–149 (2014) 499–508.
[45] M.I. ZaKi, M.A. Hasan, F.A. Al-Sagheer, L. Pasupulety, Colloid. Surf. A 190
(2001) 261–274.
[46] P. Concepción, M.T. Navarro, T. Blasco, J.M. López Nieto, B. Panzacchi, F. Rey,
Catal. Today 96 (2004) 179–186.
[47] K. Chen, S. Koso, T. Kubota, Y. Nakagawa, K. Tomishige, ChemCatChem 2
(2010) 547–555.
[48] Y. Amada, S. Koso, Y. Nakagawa, K. Tomishige, ChemSusChem 3 (2010)
728–736.
[49] Y. Amada, Y. Shinmi, S. Koso, T. Kubota, Y. Nakagawa, K. Tomishige, Appl.
Catal. B: Environ. 105 (2011) 117–127.
[50] S. Kosoa, H. Watanabe, K. Okumura, Y. Nakagawa, K. Tomishige, Appl. Catal. B:
Environ. 111–112 (2012) 27–37.
[51] Y. Amada, H. Watanabe, M. Tamura, Y. Nakagawa, K. Okumura, K. Tomishige, J.
Phys. Chem. C 116 (2012) 23503–23514.
[52] T. Ebashi, Y. Ishida, Y. Nakagawa, S. Ito, T. Kubota, K. Tomishige, J. Phys. Chem.
C 114 (2010) 6518–6526.
[53] S. Koso, H. Watanabe, K. Okumura, Y. Nakagawa, K. Tomishige, J. Phys. Chem.
C 116 (2012) 3079–3090.
[54] F. Arena, G. Italiano, K. Barbera, S. Bordiga, G. Bonura, L. Spadaro, F. Frusteri,
Appl. Catal. A: Gen. 350 (2008) 16–23.
[55] S. Natesakhawat, J.W. Lekse, J.P. Baltrus, P.R. Ohodnicki, B.H. Howard, X. Deng,
C. Matranga, ACS Catal. 2 (2012) 1667–1676.
References
[1] M.J. Climent, A. Corma, S. Iborra, Green Chem. 16 (2014) 516–547.
[2] A.M. Ruppert, K. Weinberg, R. Palkovits, Angew. Chem. Int. Ed. 51 (2012)
2564–2601.
[3] K. Fabicˇovicová, O. Malter, M. Lucas, P. Claus, Green Chem. 16 (2014)
3580–3588.
[4] Y. Liao, Q. Liu, T. Wang, J. Long, L. Ma, Q. Zhang, Green Chem. 16 (2014)
3305–3312.
[5] W. Zhu, H. Yang, J. Chen, C. Chen, L. Guo, H. Gan, X. Zhao, Z. Hou, Green Chem.
16 (2014) 1534–1542.
[6] J.N. Chheda, G.W. Huber, J.A. Dumesic, Angew. Chem. Int. Ed. 46 (2007)
7164–7183.
[7] G.W. Huber, S. Iborra, A. Corma, Chem. Rev. 106 (2006) 4044–4098.
[8] A. Corma, S. Iborra, A. Velty, Chem. Rev. 107 (2007) 2411–2502.
[9] M.E. Zakrzewska, E. Bogel-Łukasik, R. Bogel-Łukasik, Chem. Rev. 111 (2011)
397–417.
[10] J. Song, H. Fan, J. Ma, B. Han, Green Chem. 15 (2013) 2619–2635.
[11] S. Zhu, X. Gao, Y. Zhu, Y. Zhu, H. Zheng, Y. Li, J. Catal. 303 (2013) 70–79.
[12] L. Ye, X. Duan, H. Lin, Y. Yuan, Catal. Today 183 (2012) 65–71.
[13] X. Chen, X. Wang, S. Yao, X. Mu, Catal. Commun. 39 (2013) 86–89.
[14] M. Banu, S. Sivasanker, T.M. Sankaranarayanan, P. Venuvanalingam, Catal.
Commun. 12 (2011) 673–677.
[15] W.H. Zartman, H. Adkins, J. Am. Chem. Soc. 55 (1933) 4559–4563.
[16] Z. Xiao, S. Jin, M. Pang, C. Liang, Green Chem. 15 (2013) 891–895.
[17] N. Li, G.W. Huber, J. Catal. 270 (2010) 48–59.
[18] Y.T. Kim, J.A. Dumesic, G.W. Huber, J. Catal. 304 (2013) 72–85.
[19] G.W. Huber, J.W. Shabaker, S.T. Evans, J.A. Dumesic, Appl. Catal. B: Environ. 62
(2006) 226–235.
[56] A. Yin, X. Guo, W.L. Dai, K. Fan, J. Phy. Chem. C 113 (2009) 11003–11013.
[57] K.L. Deutsch, B.H. Shanks, J. Catal. 285 (2012) 235–241.
[58] A. Bienholz, H. Hofmann, P. Claus, Appl. Catal. A: Gen. 391 (2010) 153–157.
[59] Y. Román-Leshkov, M. Moliner, J.A. Labinger, M.E. Davis, Angew. Chem. Int.
Ed. 49 (2010) 8954–8957.
[60] V. Choudhary, S.H. Mushrif, C. Ho, A. Anderko, V. Nikolakis, N.S. Marinkovic,
A.I. Frenkel, S.I. Sandler, D.G. Vlachos, J. Am. Chem. Soc. 135 (2013)
3997–4006.
[61] W.R. Gunther, Y. Wang, Y. Ji, V.K. michaelis, S.T. Hunt, R.G. Griffin, Y.
Román-Leshkov, Nat. Commun. 3 (2012) 1109.
[62] Y. Kanie, K. Akiyama, M. Iwamoto, Catal. Today 178 (2011) 58–63.
[63] S. Koso, Y. Nakagawa, K. Tomishige, J. Catal. 280 (2011) 221–229.
[64] K. Chen, K. Mori, H. Watanabe, Y. Nakagawa, K. Tomishige, J. Catal. 294 (2012)
171–183.
[65] Y. Amada, H. Watanabe, Y. Hirai, Y. Kajikawa, Y. Nakagawa, K. Tomishige,
ChemSusChem 5 (2012) 1991–1999.
[66] N. Ota, M. Tamura, Y. Nakagawa, K. Okumura, K. Tomishige, Angew. Chem. Int.
Ed. 54 (2015) 1897–1900.
[67] Y. Amada, N. Ota, M. Tamura, Y. Nakagawa, K. Tomishige, ChemSusChem 7
(2014) 2185–2192.
[20] M. Holm, S. Saravanamurugan, E. Taarning, Science 328 (2010) 602–605.
[21] M. Sasaki, K. Goto, K. Tajima, T. Adschiri, K. Arai, Green Chem. 4 (2002)
285–287.
[22] Z. Huang, J. Chen, Y. Jia, H. Liu, C. Xia, H. Liu, Appl. Catal. B: Environ. 147 (2014)
377–386.
[23] X. Chen, G. Clet, K. Thomas, M. Houalla, J. Catal. 273 (2010) 236–244.
[24] T. Kim, A. Burrows, C.J. Kiely, I.E. Wachs, J. Catal. 246 (2007) 370–381.
[25] B. Zhang, Y. Zhu, G. Ding, H. Zheng, Y. Li, Appl. Catal. A: Gen. 443–444 (2012)
191–201.
[26] 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.
[27] W.L. Chu, T. Echizen, Y. Kamiy, T. Okuhara, Appl. Catal. A: Gen. 259 (2004)
199–205.
[28] S. Chen, W. Wang, Y. Zhang, Y. Wei, W. Fang, Y. Yang, J. Mol. Catal. A: Chem.
365 (2012) 60–65.
[29] D.G. Barton, S.L. Soled, G.D. Meitzner, G.A. Fuentes, E. Iglesia, J. Catal. 181
(1999) 57–72.
[30] S. García-Fernández, I. Gandarias, J. Requies, M.B. Güemez, S. Bennici, A.
Auroux, P.L. Arias, J. Catal. 323 (2015) 65–75.
[31] Y. Hong, D. Lee, H. Eom, K. Lee, Appl. Catal. B: Environ. 150–151 (2014)
438–445.
Please cite this article in press as: C. Liu, et al., WOx modified Cu/Al2O3 as a high-performance catalyst for the hydrogenolysis of glucose
to 1,2-propanediol, Catal. Today (2015), http://dx.doi.org/10.1016/j.cattod.2015.06.030