Lanthanum oxycarbonate modified Cu/Al2O3 catalysts for selective hydrogenolysis of glucose to propylene glycol: Base site requirements

Article abstract of DOI:10.1039/c7cy01571b

This work reports on the base properties of La₂O₂CO₃-Cu/Al₂O₃ catalysts and their effect on the catalyst's performance in the aqueous hydrogenolysis of glucose to propylene glycol. The catalysts show promising performance in glucose isomerisation, C-C bond cleavage of fructose and hydrogenation of the resulting fragments. The base properties of the catalyst influence the selectivity. More specifically, the basicity of the catalyst facilitates the isomerisation of glucose to fructose and the C-C bond cleavage of fructose to produce lighter products. This is evidenced by the correlation between the acid-base properties of the catalysts and their catalytic performance. In contrast, acid sites promote side reactions such as condensation reactions. Based on this study, a surface reaction mechanism is proposed for the first step of the tandem reactions, the isomerisation of glucose to fructose. The effect of the reaction time and temperature on the product distribution was investigated to get more insight into the reaction pathways. In this regard, the conversion of intermediates was also investigated. Then, a reaction mechanism was proposed based on the findings of this study and previous works.

Full text of DOI:10.1039/c7cy01571b

View Article Online
View Journal
Catalysis
Science &
Technology
Accepted Manuscript
This article can be cited before page numbers have been issued, to do this please use: P. Yazdani, B.
Wang, Y. Du, S. -. Kawi and A. Borgna, Catal. Sci. Technol., 2017, DOI: 10.1039/C7CY01571B.
This is an Accepted Manuscript, which has been through the
Volume 6 Number 1 7 January 2016 Pages 1–308
Royal Society of Chemistry peer review process and has been
accepted for publication.
Catalysis
Science &
Technology
Accepted Manuscripts are published online shortly after
www.rsc.org/catalysis
acceptance, before technical editing, formatting and proof reading.
Using this free service, authors can make their results available
to the community, in citable form, before we publish the edited
article. We will replace this Accepted Manuscript with the edited
and formatted Advance Article as soon as it is available.
You can find more information about Accepted Manuscripts in the
author guidelines.
Please note that technical editing may introduce minor changes
to the text and/or graphics, which may alter content. The journal’s
ISSN 2044-4753
standard Terms & Conditions and the ethical guidelines, outlined
in our author and reviewer resource centre, still apply. In no
event shall the Royal Society of Chemistry be held responsible
PAPER
Qingzhu Zhang et al.
Catalytic mechanism of C–F bond cleavage: insights from QM/MM
analysis of fluoroacetate dehalogenase
for any errors or omissions in this Accepted Manuscript or any
consequences arising from the use of any information it contains.
rsc.li/catalysis

Page 1 of 13
Catalysis Science & Technology
Catalysis Science & Technology
Article
View Article Online
DOI: 10.1039/C7CY01571B
Lanthanum oxycarbonate modified Cu/Al2O3 catalysts for
selective hydrogenolysis of glucose to propylene glycol: base site
requirements
P. Yazdani,a,b B. Wang,a Y. Du,a S. Kawi,b* A. Borgnaa,b*
This work reports on the base property of La2O2CO3-Cu/Al2O3 catalysts and its effect on the catalyst performance for the
aqueous hydrogenolysis of glucose to propylene glycol. The catalysts show promising performance in glucose
isomerisation, C-C bond cleavage of fructose and hydrogenation of the resulting fragments. The base property of the
catalyst influences the selectivity. More specifically, the basicity of the catalyst facilitates the isomerisation of glucose to
fructose and C-C bound cleavage of fructose to produce lighter products. This is evidenced by the correlation between
acid-base properties of the catalysts and the catalytic performance. In contrast, acid sites promote side reactions such as
condensation reactions. Based on this study, a surface reaction mechanism is proposed for the first step of the tandem
reactions; the isomerisation of glucose to fructose. The effect of the reaction time and temperature on the product
distribution was investigated to get more insight into the reaction pathways. In this regards, the conversion of
intermediates was also investigated. Then, a reaction mechanism is proposed based on the findings of this study and
previous works.
selectivity to produce high yield of high value chemicals such
as propylene glycol and ethylene glycol.
These polyols are widely used as building blocks for production
Introduction
The production of chemical and fuel still heavily relies on fossil
resources as feedstock. However, this long lasting landscape is
being challenged by environmental concerns, increasing
energy and chemical demands due to the fast population
growth in the world, and the limited reserves of fossil
resources. Therefore, the utilisation of biomass as a renewable
carbon source to replace fossil resource has gained much
attention. In this context, cellulose, the most abundant
component in the lignocellulosic biomass with a yearly
production of nearly 40 billion1, is considered as a promising
renewable feedstock. The building blocks of cellulose are
glucan chains, which are glucose polymers. Several processes
for conversion of glucose have been reported, including
dehydration, reduction, gasification, dry reforming,
esterification, and hydrogenolysis1. It is important to stress
that glucose hydrogenolysis has received much more attention
compared to the other processes. The high content of oxygen
atoms in the glucose molecule makes it a promising feedstock
for producing oxygen containing chemicals such as polyols via
hydrogenolysis2. Glucose hydrogenolysis involves the cleavage
of the C-O and C-C chemical bonds in the molecule of glucose.
Then, the hydrogenation of the fragments formed during the
chemical bond cleavage can further produce various value-
added chemicals. In addition, the uncontrolled reactivity of
glucose is the main issue lowering down the carbon balance2.
Hence, it is required to design effective catalysts to control the
of polymers such as polyethylene terephthalate, polyester
resins and other industrial chemicals3. It is worth noting that
the use of propylene glycol is growing over ethylene glycol due
to the toxicity of ethylene glycol to humans and animals4.
Considering this fact and the higher price of propylene glycol
compared to ethylene glycol, significant efforts have been
devoted to develop new catalysts for the hydrogenolysis of
cellulose/glucose to produce propylene glycol as major
product. Several authors reported the reaction mechanism for
glucose hydrogenolysis5-8. A simplified reaction pathway is
depicted in Scheme 1 based on the results of this work and
literature. Glucose initially undergoes isomerisation to form
fructose, being this step catalysed by Lewis acid or base sites9.
Subsequently, retro-aldol condensation takes place to convert
fructose to C3 chemicals, including glycerol, pyruvaldehyde,
and lactic acid10-12. Finally, propylene glycol is produced by
hydrogenation of pyruvaldehyde via acetol as an intermediate.
Hence, the conversion of glucose to propylene glycol needs a
bifunctional catalyst, which could catalyse isomerisation, retro-
aldol condensation, dehydration, and hydrogenation reactions.
In addition to the mentioned cascade reactions, there are
other competing reactions including hydrogenation of glucose
and fructose to hexitols, degradation of glucose to ethylene
glycol and erythritol, dehydration of fructose to 5-
hydroxymetylfurfural, and formation of humins and soluble
condensation products13. Therefore, precise control of the
catalyst properties is required to achieve high yield of
propylene glycol.

Catalysis Science & Technology
Page 2 of 13
Catalysis Science & Technology
Article
View Article Online
DOI: 10.1039/C7CY01571B
Scheme 1. Proposed reaction pathways for glucose hydrogenolysis towards propylene glycol and by-products adapted from6, 7, 14
;
HG, hydrogenation; H, hydrolysis; DH, dehydration; RAC, retro-aldol condensation; HDO, hydrodeoxygenation; AC, aldol
condensation.
Based on these considerations, the development of homogenous base promoter and a heterogeneous supported
heterogeneous catalyst with a suitable balance between metal catalyst. Thus, the hydrogenolysis of cellulose over
°
acid/base properties and selective hydrogenation sites has Cu/CrOx+Ca(OH)2 gave 42.6% of propylene glycol at 245 C and
received considerable attention recently15. For instance, 60 bar of H219. Cellulose hydrogenolysis over Ni/Ac+La(OH)3
several more selective catalysts towards propylene glycol than produced 27.3% and 25.1% propylene glycol and ethylene
11
°
ethylene glycol have been reported: Ni-SnOx/Al2O3
,
glycol at 245 C and 50 bar, respectively12. More recently, dual
Ni/Ac+SnO16, Ru/Ac+ZnO17, CuO/ZnO/Al2O3 , and Ni/ZnO18. association of Pt/BaZrO3 with CeCl3∙7H2O resulted in 19.2%
8
Selectivities towards propylene glycol between 18.4% and 34% and 21.7% propylene glycol and ethylene glycol, respectively,
°
were obtained. Those studies have shown that switching the when the hydrogenolysis of cellulose was carried out at 230 C
10
reaction pathway towards the production of propylene glycol and 25 bar of H2
.
can be achieved using appropriate base promoters. Another Cu/C and Pd/C were reported to be more selective to
strategy was to perform the hydrogenolysis of propylene glycol compared with other transition metals
glucose/cellulose over dual system consisting of
a
supported on carbon20. It is worth noting that the ultimate aim

Page 3 of 13
Catalysis Science & Technology
Catalysis Science & Technology
ARTICLE
of this study is to produce propylene glycol as the main aged at 60 °C for 16 hours under vigorous stirring.
View Article Online
product by glucose hydrogenolysis in a sustainable manner. Subsequently, warm deionised water (70D°OC)I:w10a.1s03a9d/dCe7CdYt0o15t7h1eB
Hence, copper, a non-precious and cheaper metal, was chosen solution until the pH of the solution decreased to 7. The
over palladium. In addition, heterogeneous base promoted precipitated solids were obtained by filtration and dried
catalysts are a better choice than homogeneous catalysts in overnight at 100 °C in a vacuum oven. The dried catalysts were
view of alkali separation or neutralization, and corrosion then calcined at 400 °C for 5 hours using a heating ramp of 1 °C
issues21. Based on these concerns and inspired by previous /min in following air. The activation of the catalysts was
reports, Cu-La2O2CO3/Al2O3 catalysts were developed for performed at 400 °C for 2 hours with a ramping rate of 2 °C
glucose hydrogenolysis, allowing the production of propylene /min in hydrogen flow. Finally, the reduced catalyst was
glycol from glucose via a more sustainable route. Lanthanum passivated using a mixture of 1%O2 in N2 for 30 min before
oxide has been used as a promoter in various catalytic exposure to ambient air. The prepared catalysts were
reactions22-24. The basicity of lanthanum oxide makes it a designated by the theoretical Cu/La molar ratio.
suitable additive in hydrogenolysis of glucose to propylene
glycol25, 26. However, dissolution of lanthanum ions in the Catalytic activity
reaction medium promotes glucose degradation towards
The hydrogenolysis of glucose was carried out in a 300 ml
ethylene glycol12. To overcome this issue, the metal active
stirred Parr-reactor with teflon liner under 34 bar of H2 gas.
phases are usually supported on the carriers such as TiO2,
High pressure of H2 gas was introduced in the parr reactor
carbon, and Al2O3 to improve the hydrothermal stability.
before heating up the reactor to the reaction temperature.
Moreover, it was reported that mixed oxides had better
The heating time to 200 °C was 35 min. Time zero was taken as
stability compared to the corresponding pure oxides in hot
the time at which temperature reached to 200 °C. The
°
liquid water (>200 C)27, 28. In addition, some other studies
resulting solution obtained after reaction was analysed with
showed that the thermal stability of aluminium oxide was
HPLC. 0.5 mM aqueous solution of sulfuric acid was used as
increased after modification with rare earth metal oxides such
mobile phase in the HPLC analysis, using a flow rate of 0.5
as lanthanum oxide22, 28. Hence, in this study, both copper and
ml/min. The temperature of the aminex HPX-87H column was
lanthanum were deposited on alumina support by co-
kept at 60 °C.
precipitation method, where copper provides the
hydrogenation site.
Catalyst characterisation
Cu-La2O2CO3/Al2O3 catalysts with different Cu/La ratios were
The powder X-ray diffraction (XRD) patterns were recorded on
investigated in this work. This approach allows for the
a Bruker AXS D8 diffractometer using CuKα radiation operating
at 40 kV and 40 mA. The diffraction patterns were recorded in
the 2 range of 20-80°. The Brunauer Emmett Teller (BET)-
correlation of acid/base properties of the catalysts with
product distribution during glucose hydrogenolysis to
understand the role of the different active sites. In addition,
hydrogenolysis of intermediates such as fructose,
pyruvaldehyde, and acetol over the developed catalyst was
investigated to gain more insights on the reaction mechanism.
surface area was determined by N2 physical adsorption at
liquid N2 temperature using a Micromeritics ASAP 2420
system. The pore volume of the catalysts was calculated by
BJH method. Prior to BET analysis, the catalysts were
pretreated at 200 °C for 2 hours under vacuum. Temperature-
programmed reduction (TPR) studies were performed on a
Thermo Scientific TPDRO 1100 instrument equipped with a
thermal conductivity detector (TCD). Prior to TPR studies, 40
mg of the catalyst powder was outgassed for 1 hour under Ar
flow at 100 °C using a ramping rate of 5 °C/min. Then TPR
measurements were performed under 30 ml/min flow of
10%H2/90%Ar from room temperature to 600 °C at a ramping
rate of 5 °C/min. Temperature programmed desorption (TPD)
of CO2 was used to determine the total basicity and base
strength of the prepared catalysts. Prior to analysis, 100 mg of
the catalyst was first flashed with N2 for 5 min at room
temperature followed by reduction in 30 ml/min flow of 5% H2
at 400 °C for 60 min. After cooling down to 50 °C, the catalyst
was saturated with CO2 (30 ml/min) for 30 min. Physically
adsorbed CO2 was removed using flowing N2 gas for 30 min.
Then, CO2 TPD was carried out heating up from 50 °C to 800 °C
under 30ml/min of flowing He and the desorbed CO2 was
quantified using a TCD. The CO2-TPD profiles were analysed
using multiple Gaussian functions to carry out the fittings. The
area of the deconvoluted peaks was calculated using OriginPro
Experimental
Catalyst preparation
Cu-La2O2CO3/Al2O3 catalysts with different Cu/La ratios along
with a Cu/La2O2CO3 sample were prepared by the co-
precipitation method described by Xin Jin et al.6. Nitrate
precursors of copper, lanthanum, and aluminium were added
to deionised water. The amounts of the copper and aluminium
precursors used to synthesize 30%Cu/Al2O3 were 1.6 and 11.6
gram in 200 ml of water, respectively. Catalysts with Cu/La
molar ratios of 9, 6, 3 and 1.5 were prepared by adding 0.33,
0.5, 1.0, 2.0 gram of the lanthanum precursor. The amount of
copper and lanthanum precursors used to synthesize 30%Cu/
La2O2CO3 were 4.7 and 8.7 gram, respectively. Another
solution was prepared by dissolving 3.0 of 25.4 gram of sodium
hydroxide and sodium carbonate in 300 ml of deionised water.
The prepared solutions were added drop wise to the solution
of the nitrate precursors, which was kept at 60 °C in a silicon
oil bath under vigorous stirring. The pH of the solution was
kept above 10 during the precipitation. The solution was then

Catalysis Science & Technology
Page 4 of 13
ARTICLE
Catalysis Science & Technology
2016 software. NH3-TPD of was performed to analyse the Table 1. Textural properties of the Cu-La2O2CO3/AVl2ieOw3AcrtaictleaOlynslintes
acidity of the catalysts. Pretreatment and reduction with different Cu/La ratios along with Cu/DLaO2IO: 120C.1O033.9/C7CY01571B
procedures were similar to those used for CO2-TPD. The
reduced catalysts were cooled down to 100 °C and saturated Cu/La2O2CO3-
with NH3 (50ml/min) for 60 min. Then, the sample was treated
under 50 ml/min flow of He for 60 min. The acidity of the
CuO
/La2O3
(wt%)a
SBET
dpore
D
Cu SA
Cu PS
(nm)e
Al2O3
(m2/g)
(nm)b
(%)c (m2/g)d
Cu/Al2O3
Cu/La=9.0
22.7/-
22.4/8.
94.0
90.0
18.6
24.3
8.3
11.5
53.8
75.1
12.5
8.9
sample was determined by heating the catalyst from 100 °C to
800 °C under 50ml/min flow of He and the desorbed NH3 was
detected with a TCD. Copper surface area and dispersion of
the catalysts were determined by the dissociative N2O
chemisorption as reported elsewhere29. The catalysts were
pretreated and reduced as previously described for CO2-TPD
method. Then, oxidation of the copper surface was performed
by introducing 30 ml/min of N2O at 80 °C. N2O was assumed to
be dissociated, resulting in the oxidation of two copper atoms.
Subsequently, the catalyst was purged with 30 ml/min of N2 to
remove the residual N2O gas. Then, Cu2O was reduced by using
30 ml/min flow of 5% H2 in Ar. X-ray absorption spectra (XAS)
of Cu-K edge and La-LIII edge were recorded at the XAFCA
beam line at the Singapore Synchrotron Light Source30. The
samples were firstly mixed with boron nitride to prepare the
samples. The samples were reduced under similar conditions
to those used for the catalyst activation described in the
catalyst preparation section. TEM images were taken on a
JEM-2100F transmission electron microscope and SEM images
of the catalysts were recorded on a JEOL FEG SEM (SM 6700F).
Cu/La=6.0 17.9/11 97.7
28.7
14.5
93.8
7.2
Cu/La=3.0 17.6/18 99.2
Cu/La=1.5 15.4/28 106.9
21.4
20.2
14.8
15.3
96.3
99.6
7.0
6.7
Cu/La2O2CO3 23.6/60 34.8
a XRF analysis
25.9
9.8
63.3
10.6
b Pore Diameter (dpore) from BJH method
c Cu Dispersion (D) calculated from N2O chemisorption
d Cu Surface Area (SA) calculated from N2O chemisorption
e Cu Particle Size (PS) calculated from N2O chemisorption
XRD Patterns
Figure 1 compares the XRD patterns of the calcined catalysts.
The diffraction pattern corresponding to the monoclinic
structure of La2O2CO3 (JCPDS 00-048-1113) is observed in all
XRD patterns of La-containing catalysts. Moreover, the
diffraction peaks appearing at 35.5°, 38.6°, and 66.2° can be
indexed to CuO phase (JCPDS 04-012-7238). It is clearly shown
that the intensity of the CuO diffraction peaks is decreased
with the increase of lanthanum loading and eventually
vanished for the catalyst with Cu/La ratio= 1.5. This suggested
that the addition of the lanthanum enhance the dispersion of
the copper particles on the catalyst surface, being consistent
with N2O chemisorption results.
Results and discussions
Textural properties
The physical properties of the catalysts were characterised by
N2O chemisorption and N2 physisorption methods, and the
results are presented in Table 1. As shown in the Table, the
surface area and dispersion of the copper particles for
Cu/Al2O3 catalyst are 53.8 m2/g and 8.3%, respectively,
indicating a copper particle size of 12.5 nm. Interestingly, the
surface area and dispersion of the copper particles increase
with the addition of lanthanum to the Cu/Al2O3 catalyst to 99.6
m2/g and 15.3%, respectively. In the case of Cu/La2O2CO3, they
are 63.3 m2/g and 9.8%, respectively. Hence, the metal
dispersion in all Cu-La2O2CO3/Al2O3 catalysts is higher than that
in the Cu/Al2O3 catalyst. As a result, smaller copper particles
are produced on the surface of the Cu-La2O2CO3/Al2O3 and
Cu/La2O2CO3 compared to that of the Cu/Al2O3. The
mentioned trend is consistent with previous studies on the
effect of the lanthanum loading on the crystal growth of nickel
particles supported on aluminium oxide22. The surface areas of
the catalysts with different lanthanum loadings are quite
similar, while the area of Cu/La2O2CO3 catalyst is only 34.8
m2/g. This value is much lower than the surface areas of the
other catalysts. Thus, it is reasonable to expect that alumina
enhances the metal dispersion by providing an increased
surface area. Moreover, the addition of lanthanum could
further enhance the Cu dispersion.
Figure 1. XRD spectra of the calcined catalysts.
.

Page 5 of 13
Catalysis Science & Technology
Catalysis Science & Technology
Article
View Article Online
DOI: 10.1039/C7CY01571B
TEM and SEM images
TEM images of the reduced Cu-La2O2CO3/Al2O3 with Cu/La
ratio at 6 is presented in Figure 2. These images demonstrate
that copper particles are spherical in shape, having a wide
range of copper particle sizes on the catalyst surface. The
average copper particle size is estimated using Nano Measure
software over 200 particles. An average particle size of 8.3 nm
with a standard deviation of 3.9 nm was obtained. The
calculated mean particle size from TEM images is in good
agreement with the particle size determined by N2O
chemisorption (Table 1).
SEM images of the calcined catalysts are presented in Figures
S1. SEM images reveals that the catalysts prepared in this
study are composed of agglomerates with different shapes and
sizes in the range of 10-50 µm. SEM-EDX pattern were also
measured and presented in the Figure S2. EDX analysis
confirmed that both Cu and La atoms are present on the
surface of the calcined catalysts. The concentration of the
copper particles on the surface of the catalysts decreases from
31.2wt% to 10.6wt% by increasing the lanthanum loading from
0 to a Cu/La ratio= 1.5, while the concentration of lanthanum
increases from 0.0 wt% to 29.2 wt%. The concentration of the
Cu and La atoms on the surface of the Cu/ La2O2CO3 catalyst
are 15.2 wt% and 39.5 wt%, respectively. Hence, SEM images
clearly show that upon addition of lanthanum, the
concentration of the copper atoms on the surface decreases
while the lanthanum concentration increases.
Figure 3. CO2-TPD profile of the catalysts.
Acid-Base properties
CO2-TPD profiles of the synthesized catalysts are depicted in
Figure 3. It can be seen that there are different desorption
peaks in the range of 50-650 °C. These desorption peaks
correspond to base sites with different strengths. Table 2
shows the relative concentration of base sites with different
strengths, estimated by integrating the area under the
deconvoluted peaks. Di Cosimo et al. assigned the desorption
peaks of CO2 at increasing temperature to bicarbonates,
unidenate carbonates and bidenate carbonates, adsorbed on
hydroxyl groups, metal-oxygen pairs, and oxygen anions,
respectively31. It is worth noting that all samples possess a
similar amount of hydroxyl groups on the surface of the
catalysts, except for the Cu/ La2O2CO3 catalyst. The
concentration of hydroxyl groups on the surface of the Cu/
La2O2CO3 catalyst is 0.08 µmol/m2, while for other catalysts it
is between 0.56-0.62 µmol/m2 (Table 2). Meanwhile, the
concentration of the moderate and strong base sites clearly
increases as the loading of the lanthanum increases, reaching
the highest value for the Cu/ La2O2CO3 catalyst. In contrast, the
NH3-TPD profiles of the catalysts only possess one peak at 200
°C. The concentration of these weak acid sites, calculated from
the integration of the NH3-TPD profile, is presented in Table 2.
Clearly, the concentration of acid sites decreases as a result of
the increasing lanthanum loading. Moreover, Cu/ La2O2CO3
catalyst does not possess acid sites. In summary, the acid/base
properties of the catalysts can be effectively modified by
changing the Cu/La/Al ratio.
Figure 2. TEM images and particle size distribution of Cu-
La2O2CO3/Al2O3 sample with a Cu/La ratio=6.

Catalysis Science & Technology
Page 6 of 13
Catalysis Science & Technology
Article
View Article Online
DOI: 10.1039/C7CY01571B
Table 2. Concentration of base and acid sites on the catalyst surface
Concentration of basic sites,
Cu-La-Al
Mixed
oxides
µmol/m2
Concentration of
acid sites,
µmol/ m2
0.45
Weak
Moderate Strong
Cu/Al2O3
Cu/La=9.0
Cu/La=6.0
Cu/La=3.0
Cu/La=1.5
Cu/La2O2CO3
0.56
0.62
0.61
0.60
0.61
0.08
0.22
0.33
0.44
0.52
0.53
0.4
-
0.23
0.25
0.32
0.37
0.52
0.16
0.12
0.05
0.05
-
TPR profiles
TPR profiles of the catalysts are depicted in Figure 4. All
catalysts display a broad reduction peak in the temperature
range of 200-330 °C. Two reduction peaks are observed at low
temperature for Cu/Al2O3, which can be assigned to the two
step reduction of the bulk CuO, whereas the peak observed at
higher temperatures is ascribed to well-dispersed CuO
particles strongly interacting with the support32. As shown in
Figure 4, copper oxide reduces at higher temperature when it
is supported on alumina as compared to La2O2CO3-supported
copper oxide. This can be ascribed to the stronger interaction
of the copper oxide species with alumina. Interestingly,
promoting the Cu/Al2O3 with relatively small amounts of
lanthanum oxycarbonate (Cu/La=9, 6 and 3), shifts the
reduction of copper oxide to higher temperature due to the
increasing CuO dispersion. Thus, stronger interaction between
CuO species and alumina clearly exist because of the higher
dispersion of CuO species in presence of lanthanum
oxycarbonate. However, the reduction temperature of the
CuO- La2O2CO3/Al2O3 sample with Cu/La=1.5 is shifted to lower
temperatures as a results of the week interaction between
copper and lanthanum oxycarbonate. Therefore, it is clear that
the reduction of copper oxide species is strongly affected by
addition of the lanthanum and it depends on the lanthanum
loading. Indeed, some contradictory observations of the effect
of lanthanum oxide on the reducibility of metals were
reported in the literature as well. The delay of Ni reduction by
lanthanum oxide was reported by Znak et al.33 and Liu et al.24,
Figure 4. TPR profiles of the catalysts.
Figure 5. Copper 2p core level XPS spectra of the calcined
catalysts, a) Cu/Al2O3, b) Cu/La=9, c) Cu/La=6, d) Cu/La=3, e)
Cu/La=1.5, f) Cu/ La2O2CO3.
while others reported that lanthanum oxide facilitates the
34,
reduction of nickel and copper22,
35. Herein, our results
indicate that the reducibility of the CuO- La2O2CO3/Al2O3
catalysts depends on the loading of lanthanum oxycarbonate.
XPS and XAS Characterisations
Copper 2p core level XPS spectra of the calcined catalysts are
shown in Figure 5. These samples exhibit the Cu 2p3/2 and Cu
2p1/2 peaks at the binding energy of 933.9 eV and 954.2 eV,
respectively. A sharp satellite peak is observed in the XPS
spectra of the catalysts, indicating the presence of the CuO on
the surface of the samples.

Page 7 of 13
Catalysis Science & Technology
Catalysis Science & Technology
Article
View Article Online
DOI: 10.1039/C7CY01571B
It is worth noting that the chemical state of copper oxide is not
affected by the presence of lanthanum oxycarbonate, as no
change in the binding energy of Copper 2p core level XPS
spectrum is observed.
X-ray absorption spectroscopy (XAS) was employed to
measure the local structure and electronic states of copper
and lanthanum species in the synthesized catalysts. The
Fourier Transforms (FT) of Cu K-edge EXAFS spectra of the
calcined catalysts along with CuO as reference are compared in
Figure 6a. These results confirm that copper species on the
surface of the calcined catalysts exist only as CuO, which is
consistent with XRD diffraction patterns and XPS data. Hence,
the electronic state of the copper species is not significantly
influenced by lanthanum oxycarbonate and alumina support
after the calcination. The FTs of the EXAFS spectra for the
reduced catalysts are compared with metallic copper in Figure
6b. This Figure shows that CuO is reduced to the metallic
copper in all samples after reduction at 400 °C for 2 hour,
being Cu the active site for hydrogenation. The curve fitting of
Cu K-edge EXAFS of the catalysts reduced at 400 °C shows the
existence of Cu-Cu distances at 2.50-2.51 Ǻ and coordination
numbers (CN) of ca. 10
 1 (Table S1). The fitting parameters
of the catalysts are summarised in Table S1. These results
indicate that there is no evidence of copper-lanthanum
interaction. In addition, it is worth noting that the coordination
numbers obtained by the fitting procedures indicate the
presence of relatively large Cu particles with particle sizes > 4
nm36. Figure 7 shows the La LIII XANES spectra of La2O2CO3 and
the reduced catalysts. The XANES spectra of the catalysts are
similar to that of the bulk lanthanum oxide. However, the
white line intensity for the synthesized catalysts is slightly
higher when it is compared with bulk lanthanum oxide. This
feature suggests the modification of the electronic structure of
the lanthanum cations, resulting from the interaction between
lanthanum oxycarbonate and the support37.
Figure 6. Cu K-edge EXAFS spectra of a) CuO/ La2O2CO3-Al2O3
with different ratio of Cu/La, and CuO/ La2O2CO3, b) Cu/
La2O2CO3-Al2O3 with different ratio of Cu/La, and Cu/
La2O2CO3.
Catalytic Results
The performance in glucose hydrogenolysis of the Cu-
La2O2CO3/Al2O3 catalysts with different Cu/La ratios along with
the Cu/ La2O2CO3 catalyst is compared in Table 3. 5-hydroxyl
methyl furfural is the main product in the blank experiment,
yielding 25.2% after 10 min reaction at 200 °C and 34 bar of H2
(Table 3, entry 10). The conversion of glucose to 5-
hydroxylmethylfurfural was reported by several authors2, 18
under similar reaction conditions applied in this study. As it
was reported, glucose easily undergoes transformation under
our reaction conditions. Thus, the formation of 5-
hydroxymethylfurfural and soluble condensation products is
dominant in absence of catalyst18. On the contrary, 5-
hydroxymethylfurfural selectivity is dramatically decreased in
the presence of a catalyst, being glucose fully converted within
10 minutes at 200 °C and 34 bar of H2 (Table 3, entry 1-6).
Figure 7. XANES spectra of the reduced Cu-La2O2CO3/Al2O3
with different Cu/La ratios and Cu/ La2O2CO3 at the La LIII-edge.

Catalysis Science & Technology
Page 8 of 13
Article
Catalysis Science & Technology
View Article Online
DOI: 10.1039/C7CY01571B
Over Cu/Al2O3, the main product is sorbitol (34.1% yield) and ethylene glycol production. So, an experiment was designed to
the total selectivity towards C3 chemicals, including glycerol, determine if, in our case, the reaction proceeds over
lactic acid, acetol, and propylene glycol, is only 24.2%. This lanthanum ions dissolved in the reaction media. In order to
result shows that the concentration of base sites over Cu/ verify this, Cu- La2O2CO3/Al2O3 with a Cu/La ratio= 6 was
Al2O3 was not enough to produce propylene glycol as the main treated in water at 200 °C and 34 bar of H2 for 10 min. Then,
product over sorbitol. Zheng et al.15 also reported that alumina the reaction media was separated from the solid catalyst using
is not effective to promote retro-aldol condensation of a centrifuge. Then, glucose was added to the solution and used
glucose, in line with our catalytic results. In addition, minor in a blank experiment (entry 11, Table 3). In this experiment,
products such as ethylene glycol (2.4%), fructose (2.3%), acetol yield is 6.1% and the carbon balance is only 46.4%,
arabitol (5.0%), erythritol (1.8%), 1,2-butanediol (2.1%), and which is much lower than the one obtained in the blank
2,5-hexanediol (0.5%) are also detected. The mass balance of experiment reported in entry 10 of Table 3. Hence, this result
the reaction products analysed by HPLC is 73.0% in this case. clearly suggests that glucose hydrogenolysis to acetol in this
Low carbon balances have also been reported in several study is catalysed by
a heterogeneous catalyst, Cu-
studies on glucose hydrogenolysis to propylene glycol in batch La2O2CO3/Al2O3, other than the homogeneous lanthanum ions.
reactors11, 12, 18, 19. The carbon loss is ascribed to unidentified In addition, retro-aldol condensation of fructose is favoured
and/or undetected products by HPLC analysis, most likely under our reaction conditions, resulting in the production of
soluble condensation products and humins. These chemicals propylene glycol as the main product.
are produced via the reaction between glucose with HMF Glucose hydrogenolysis over commercial La2O3 was also
and/or HMF with 2,5-dioxo-6-hydroxy-hexanal (hydrated examined to understand the role of La2O2CO3 and Cu
HMF)13. In addition, the formation of gas products is not the individually. These catalytic results are presented in entry 9,
main reason to explain the low carbon balance, as the Table 3. It shows that the main product of glucose
difference between total organic carbon (TOC) of the reaction hydrogenolysis over La2O3 is lactic acid with yield of 42.0%.
products and feedstock is lower than 5%, as determined by Acetol was produced as well with a yield of 11.5%. This result
TOC analysis.
suggests that lanthanum oxycarbonate is responsible for the
Notably, upon addition of lanthanum oxycarbonate, the total glucose isomerisation to fructose and retro-aldol condensation
selectivity towards C3 chemicals increases with the concurrent of fructose to C3 chemicals (Scheme 1). Hence, it can be
decreasing of the sorbitol yield. This trend indicates that postulated that lanthanum oxycarbonate, formed on the
lanthanum oxycarbonate facilitates glucose conversion to surface of the synthesized catalyst, plays a similar role during
propylene glycol. However, selectivity towards lactic acid glucose hydrogenolysis. As shown by CO2-TPD, NH3-TPD, and
increases to 15.3% by increasing the Cu/La ratio from 9.0 to XRD, lanthanum oxycarbonate was formed on the surface of
1.5. Moreover, over the catalyst without alumina (Cu/ the catalyst, changing the catalyst acid/base properties
La2O2CO3), the total yield of (acetol + propylene glycol) is 40.1 dramatically. This change explains the enhancement of the
%. Clearly, aluminium oxide plays the role of support for selectivity towards propylene glycol by controlling the rate of
copper and lanthanum oxycarbonate, as evidenced by the competitive reactions. Cu/Al2O3 contains the lowest
comparing the catalytic results of Cu/Al2O3 and Cu/ La2O2CO3. basicity and highest concentration of acid sites, being this
Hence, a proper balance of base sites and hydrogenation sites catalyst less active towards C-C bond cleavage and more active
is required to facilitate glucose hydrogenolysis to acetol and for hydrogenation. Then, higher concentration of moderate
minimise the yields of by-products such as sorbitol and lactic and strong base sites, along with lower concentration of acid
acid. In addition a proper ratio of acid sites and base sites sites lead to an increase in the selectivity towards acetol and
(Cu/La=6) is required to maximised the carbon balance, as propylene glycol. Hence, base sites play a vital role in
evidenced by carbon balance results presented in Table 3 producing C3 chemicals from glucose hydrogenolysis. However,
along with the characterisation results Therefore, the reaction it is difficult to distinguish whether the required active sites
time of glucose hydrogenolysis over these two catalysts (Cu/La are moderate or strong base sites due to the co-existence of
ratio of 6 and Cu/ La2O2CO3) was extended to 3 h to convert both base sites. Therefore, it can only be concluded that the
acetol into propylene glycol and the catalytic results are shown presence of a suitable amount of moderate and strong base
in entry 7-8, Table 3. Propylene glycol yields over these sites are required for efficiently promote this reaction. Cu/
catalysts reach 29% and 32%, being comparable to the highest La2O2CO3 contains negligible amounts of weak base sites while
yields reported in the literature5, 17-19. It is worth noting that the concentration of weak base sites in Cu-La-Al catalysts is
propylene glycol yield exceeds the yield of ethylene glycol over similar to the one on Cu/Al2O3 catalysts, thus suggesting that
these catalysts at 200 °C and 34 bar of H2. These reaction weak base sites are not effective to promote propylene glycol
results are in contrast with those obtained during cellulose production.
°
hydrogenolysis over Ni/C+LaOH at 245 C and 50 bar of H2,
where ethylene glycol was reported as the main product by
Sun et al12. They have shown that lanthanum ions in solution,
acting as
a homogenous catalyst, along with a Ni/C
heterogeneous catalyst directed the reaction pathway towards

Page 9 of 13
Catalysis Science & Technology
Article
Catalysis Science & Technology
View Article Online
DOI: 10.1039/C7CY01571B
Table 3. Catalytic results of glucose conversion over Cu-La2O2CO3/Al2O3 catalysts.
Selectivity (%)
C3 chemicals
Ethylene HMF Othersb Carbon
Lactic
acid
Propylene
glycol
Entry
Catalysts
Conversion Sorbitol Glycerol
Acetol
glycol
balance
1
2
Cu/Al2O3
Cu/La=9.0
95.5
98.8
100
100
100
100
100
100
99.8
63.1
95.5
34.1
9.6
3.9
2.0
2.3
0.8
3.3
1.3
-
13.4
9.8
9.6
7.3
5.5
2.2
8.1
4.1
-
2.4
4.9
5.7
18.8
36.9
37.8
39.3
37.9
-
2.7
3.5
3.1
1.1
1.0
2.2
29.0
32.0
1.4
-
2.4
3.9
6.9
5.0
5.1
3.3
6.9
7.1
-
0.6
11.7
10.8
13.6
12.7
13.0
8.5
69.8
60.6
82.1
79.1
81.5
69.7
70.6
68.6
69.3
41.6
44.4
-
3
Cu/La=6.0
8.1
-
4
Cu/La=3.0
13.2
15.3
14.8
8.1
-
5
Cu/La=1.5
-
6
Cu/La2O2CO3 La2O3
Cu/La=6.0a
-
7
-
-
15.2
16.1
14.4
18.8
18.4
a
8
Cu/La2O2CO3La2O3
8.0
-
9
La2O3
42.1
5.7
11.5
0.8
6.4
-
10
11
Blank 1c
Blank 2d
-
-
-
40.0
15.0
-
-
6.6
-
-
a
Reaction condition: 0.3 wt% glucose in 100 ml water, 0.1 g catalyst, 200 °C, 34 bar of H2, reaction time: 10 min. Reaction time: 3 hours.
bOthers: ethanol, 1,2-butandiol, erythritol, methanol, 2,5-hexandiol, 2.5-hexanedion, levulinic acid, arabitol, fructose, xylitol. cBlank 1,
without catalyst. dBlank 2, without catalyst and using water previously contacted with a Cu-La2O2CO3/Al2O3 catalyst for 10 min at 200 C and
34 bar of H2 as described above.
The mentioned correlation between catalytic results and
acid/base properties of the catalysts are further considered to
understand the mechanism of the glucose isomerisation to
fructose. It is known that this reaction follows two different
mechanisms: proton transfer and hydride shift38, 39. Proton
transfer involves base catalysed deprotonation of the carbonyl
group in the glucose molecule, known as Lobry-de Bruyn-van
Ekenstein (LdB−AvE) mechanism. The hydride shift involves the
polarization of the carbonyl group of glucose and it is catalysed
by Lewis acid sites9, known as Meerwein-Ponndorf-Verley
mechanism. The role of each pathway in the isomerisation of
glucose to fructose has been widely debated. The correlation
between the catalytic results and the acid-base properties
(Figure 8) suggests that glucose isomerisation follows the
Figure 8. Correlation between acid/base properties of the
proton-transfer mechanism, because Cu/La2O2CO3 without any
catalysts and the yield of acetol
+ propylene glycol.
acid site converts glucose to propylene glycol, resulting in a
yield of 32%. Hence, glucose isomerisation to fructose is
believed to follow the LdB−AvE mechanism38. Based on this
mechanism, base sites of the catalysts deprotonate α-carbonyl
carbon of the acyclic glucose to form enolate intermediates.
The open-ring form of fructose is then produced by
transferring a proton from C2 to C1 and O2 to O1 of the enolate
intermediates, as shown in Figure 9. Then the open chain form
of fructose is converted to C3 chemicals via retro-aldol
condensation40.

Catalysis Science & Technology
Page 10 of 13
Catalysis Science & Technology
Article
View Article Online
DOI: 10.1039/C7CY01571B
Figure 9. Proposed reaction mechanism adapted from previous
reports9, 38, 39 for glucose isomerisation to fructose over the
synthesized catalysts.
Figure 10. Product distribution of glucose hydrogenolysis over Cu-
La2O2CO3/Al2O3/Al2O3 with a Cu/La ratio=6 during heating up of the
reactor, reaction condition: 0.3wt% glucose, 0.1 g catalyst, 100 ml
water, 34 bar of H2, others: arabitol, xylitol, erythritol,
pyruvaldehyde, methanol, HMF, 2,5-hexanediol, levulinic acid.
Effect of temperature and reaction time
The effect of reaction time and temperature on the product
distribution over Cu-La2O2CO3/Al2O3 with a Cu/La ratio= 6 was
investigated to get more insight into the reaction pathway. The
product distribution of glucose hydrogenolysis during the heating
process is shown in Figure 10 and the product distribution of
increasing reaction times at 200 °C is depicted in Figure 11.Figure 10
shows that 76.4% of glucose is converted to various products during
the heating process from room temperature to 140 °C. At this
conversion, fructose is the main product with a selectivity of 34.3%
while the selectivity towards C3 chemicals is 21.6%. However,
fructose is completely converted completely to C3 chemicals and
other products by further increasing the temperature to 200 °C,
evidenced by an increase of selectivity towards C3 chemicals to
66.2%, indicating that the retro-aldol condensation of fructose to C3
chemicals is effectively promoted. Thus, the rate-limiting reaction
step of these tandem reactions towards propylene glycol would be
the acetol hydrogenation step. As shown in Figure 11, the
hydrogenation of acetol proceeds when the reaction temperature
reaches 200 °C. After 3 h, the hydrogenation of the acetol at 200 °C
resulted in a 30% yield of the propylene glycol, remaining fairly
constant afterwards. Degradation products of propylene glycol such
as 1-propoanol and 2-propanol are not detected even after 6 hour
reaction. This suggests that propylene glycol is stable under our
reaction conditions. Glycerol and lactic acid are produced with
yields of 11.7% and 12.2%, respectively, during the heating process,
remaining almost constant afterwards. Sugar alcohols such as
sorbitol, mannitol, arabitol, xylitol, erythritol are also produced
during heating process with a total selectivity of 15%.
Figure 11. Product distribution-time profile of glucose
hydrogenolysis over Cu- La2O2CO3/Al2O3/Al2O3 with a Cu/La ratio=6,
reaction condition: 0.3wt% glucose, 0.1 g catalyst, 100 ml water, 34
bar of H2, others: arabitol, xylitol, erythritol, pyruvaldehyde,
methanol, HMF, 2,5-hexanediol, levulinic acid.

Page 11 of 13
Catalysis Science & Technology
Article
Catalysis Science & Technology
View Article Online
DOI: 10.1039/C7CY01571B
Table 4. Hydrogenolysis of the reaction intermediates over Cu-La2O2CO3/Al2O3/Al2O3 with a Cu/La ratio= 6
Entry
Feed
Conversion%
Selectivity%
Glycerol
Lactic
acid
Acetol
Propylene
Ethylene
glycol
Others
glycol
1
2
3
Fructose
Acetola
100
72.9
100
6.8
0
6.2
36.7
-
-
76.6
-
5.1
0
12.2
2.0
-
10.5
26.1
Pyruvaldehyde
-
29.2
-
Reaction condition: 0.3 wt% feed solution, 100ml water, 0.1 g Cu- La2O2CO3/Al2O3Al2O3 with Cu/La=6, 200 °C, 34
bar of H2 and reaction time: 10 min. Others: Ethanol, 1,2-Butandiol, Erythritol, Methanol, 2,5-hexandiol, 2.5-
hexanedion HMF, Levulinic acid, Arabitol, Fructose, Xylitol. a) Reaction time: 2 hours.
mechanism is proposed based on the parametric studies. It is
These chemicals are not further converted by increasing the worth noting that propylene glycol is produced as the main
reaction time to 6 h. One of the competitive reactions for the product over sorbitol and ethylene glycol.
isomerisation of glucose to fructose under our reaction
conditions is the retro-aldol condensation of glucose to
ethylene glycol and erythritol. The yield towards erythritol and
Conflicts of interest
ethylene glycol reaches to 1.0% and 6.9%, respectively,
implying that retro-aldol condensation of glucose is minimal.
In summary, fructose, pyruvaldehyde, and acetol were
identified as reaction intermediates. In order to obtain more
insights in the reaction mechanism, hydrogenolysis of these
intermediates over the Cu- La2O2CO3/Al2O3/Al2O3 catalyst with
a Cu/La ratio= 6 under the same reaction conditions were
carried out and are presented in Table 4. This table shows that
product distribution of fructose hydrogenolysis is similar to
that of the glucose. Acetol is produced with yield of 36.7%
after 10 min at 200 °C. Moreover, acetol is selectively
converted to propylene glycol (Table 4, entry 2), resulting in
72.9% of acetol converted with selectivity of 76.6% after 2
hours reaction. On the other hand, pyruvaldehyde is fully
converted after 10 min. Acetol and lactic acid with 29.2% and
There are no conflicts to declare.
Acknowledgements
The authors gratefully thank The Agency for Science,
Technology and Research (A*STAR) of Singapore, The National
University of Singapore, and NEA (ETRP 1501 103) for their
generous financial supports. P.Yazdani is also grateful to
A*STAR for the Singapore International Graduate Award
(SINGA) scholarship.
References
1.
S. Van de Vyver, J. Geboers, P. A. Jacobs and B. F. Sels,
ChemCatChem, 2011, 3, 82-94.
26.1% yield, respectively, are produced as
a result of
2.
M. Schlaf and Z. Zhang, Reaction pathways and
mechanisms in thermocatalytic biomass conversion I:
Cellulose structure, depolymerization and conversion by
heterogeneous catalysts, Springer 2015.
pyruvaldehyde hydrogenolysis (Table 4, entry 3). It is
important to note that a poor mass balance was obtained for
this reaction and unidentified peaks in the HPLC were
observed within the retention times of glucose hydrogenolysis
products. The poor mass balance of pyruvaldehyde
hydrogenolysis shows that pyruvaldehyde is also involved in
condensation reactions13.
3.
4.
A. Marinas, P. Bruijnincx, J. Ftouni, F. J. Urbano and C.
Pinel, Catalysis today, 2015, 239, 31-37.
C. Rode, A. Ghalwadkar, R. Mane, A. Hengne, S. Jadkar
and N. Biradar, Organic Process Research & Development,
2010, 14, 1385-1392.
5.
6.
E. Girard, D. Delcroix and A. Cabiac, Catalysis Science &
Technology, 2016, 6, 5534-5542.
X. Jin, J. Shen, W. Yan, M. Zhao, P. S. Thapa, B.
Subramaniam and R. V. Chaudhari, ACS Catalysis, 2015, 5,
6545-6558.
B. O. de Beeck, M. Dusselier, J. Geboers, J. Holsbeek, E.
Morré, S. Oswald, L. Giebeler and B. F. Sels, Energy &
Environmental Science, 2015, 8, 230-240.
K. Tajvidi, P. J. Hausoul and R. Palkovits, ChemSusChem,
2014, 7, 1311-1317.
R. Nagorski and J. P. Richard, Journal of the American
Chemical Society, 2001, 123, 794-802.
E. Girard, D. Delcroix and A. Cabiac, Catalysis Science &
Technology, 2016.
Conclusion
In this study, the development of a bifunctional catalyst for
hydrogenolysis of glucose to propylene glycol is reported. The
effect of the relative concentration of base sites and
hydrogenation sites on the surface of the catalyst is discussed.
Catalytic results and catalyst characterisations validate that
the selectivity towards acetol, the precursor of propylene
glycol, in the hydrogenolysis of glucose is promoted by
moderate and strong base sites on the catalyst. In contrast,
the acid sites on the catalysts surface catalyse the reaction
towards condensation products, consequently decreasing the
overall yield of the desired product. Moreover, the reaction
7.
8.
9.
10.

Catalysis Science & Technology
Page 12 of 13
ARTICLE
11.
Catalysis Science & Technology
T. Deng and H. Liu, Journal of Molecular Catalysis A: 39.
Chemical, 2014, 388, 66-73.
Y. Román‐Leshkov, . oliner, J. A. Labinger and . E.
View Article Online
Davis, Angewandte Chemie International Edition, 2010,
DOI: 10.1039/C7CY01571B
12.
R. Sun, T. Wang, M. Zheng, W. Deng, J. Pang, A. Wang, X.
49, 8954-8957.
Wang and T. Zhang, ACS Catalysis, 2015, 5, 874-883.
H. Rasmussen, H. R. Sørensen and A. S. Meyer,
Carbohydrate research, 2014, 385, 45-57.
40.
K. Wang, M. C. Hawley and T. D. Furney, Industrial &
engineering chemistry research, 1995, 34, 3766-3770.
13.
14.
M. Orazov and M. E. Davis, Proceedings of the National
Academy of Sciences, 2015, 112, 11777-11782.
M. Zheng, J. Pang, R. Sun, A. Wang and T. Zhang, ACS
Catalysis, 2017, 7, 1939-1954.
15.
16.
R. Sun, M. Zheng, J. Pang, X. Liu, J. Wang, X. Pan, A. Wang,
X. Wang and T. Zhang, ACS Catalysis, 2015, 6, 191-201.
Y. Hirano, K. Sagata and Y. Kita, Applied Catalysis A:
General, 2015, 502, 1-7.
17.
18.
X. Wang, L. Meng, F. Wu, Y. Jiang, L. Wang and X. Mu,
Green Chemistry, 2012, 14, 758-765.
19.
Z. Xiao, S. Jin, M. Pang and C. Liang, Green Chemistry,
2013, 15, 891-895.
20.
C. Liu, C. Zhang, K. Liu, Y. Wang, G. Fan, S. Sun, J. Xu, Y.
Zhu and Y. Li, Biomass and Bioenergy, 2015, 72, 189-199.
M. Xian, Sustainable Production of Bulk Chemicals:
Integration of Bio‐,Chemo‐ Resources and Processes,
Springer Dordrecht, Netherlands, 2015.
21.
22.
23.
24.
25.
26.
27.
28.
H. L. Rotgerink, R. Paalman, J. Van Ommen and J. Ross,
Applied Catalysis, 1988, 45, 257-280.
R. Kam, C. Selomulya, R. Amal and J. Scott, Journal of
Catalysis, 2010, 273, 73-81.
H. Liu, H. Wang, J. Shen, Y. Sun and Z. Liu, Catalysis Today,
2008, 131, 444-449.
J. Zhang and D. He, Journal of colloid and interface
science, 2014, 419, 31-38.
S. Sato, R. Takahashi, M. Kobune and H. Gotoh, Applied
Catalysis A: General, 2009, 356, 57-63.
H. Xiong, H. N. Pham and A. K. Datye, Green Chemistry,
2014, 16, 4627-4643.
F. Héroguel, B. Rozmysłowicz and J. S. Luterbacher,
CHIMIA International Journal for Chemistry, 2015, 69,
582-591.
29.
30.
J. Evans, M. Wainwright, A. Bridgewater and D. Young,
Applied Catalysis, 1983, 7, 75-83.
Y. Du, Y. Zhu, S. Xi, P. Yang, H. O. Moser, M. B. Breese and
A. Borgna, Journal of synchrotron radiation, 2015, 22, 839-
843.
31.
J. i osimo, Apestegu , . a, . inés and E. glesia,
Journal of Catalysis, 2000, 190, 261-275.
32.
33.
V. D. Dasireddy and B. Likozar, Fuel, 2017, 196, 325-335.
L. Znak, K. Stołecki and J. Zielioski, Catalysis today, 2005,
101, 65-71.
34.
35.
36.
37.
S. Natesakhawat, O. Oktar and U. S. Ozkan, Journal of
Molecular Catalysis A: Chemical, 2005, 241, 133-146.
R. Shi, F. Wang, Y. Li, X. Huang and W. Shen, Green
Chemistry, 2010, 12, 108-113.
A. Borgna, F. Le Normand, T. Garetto, C. Apesteguia and
B. Moraweck, Catalysis letters, 1992, 13, 175-188.
M. Capitan, M. Centeno, P. Malet, I. Carrizosa, J.
Odriozola, A. Marquez and J. Fernandez Sanz, The Journal
of Physical Chemistry, 1995, 99, 4655-4660.
C. Liu, J. M. Carraher, J. L. Swedberg, C. R. Herndon, C. N.
Fleitman and J.-P. Tessonnier, ACS Catalysis, 2014, 4,
4295-4298.
38.

Page 13 of 13
Catalysis Science & Technology
View Article Online
DOI: 10.1039/C7CY01571B