Effect of SiO2 support properties on the performance of Cu-SiO2 catalysts for the hydrogenation of levulinic acid to gamma valerolactone using formic acid as a hydrogen source

Article abstract of DOI:10.1039/c7cy00902j

Vapor phase catalytic transfer hydrogenation of aqueous levulinic acid (LA) with formic acid (FA) as a hydrogen source was carried out over copper loaded on to several SiO₂ supports having different physicochemical properties. The SiO₂ supports and Cu-SiO₂ catalysts with different support properties were characterized by XRD, TPR, XPS, NH₃-TPD and UV-vis techniques to evaluate the copper species and its interaction with the support. The SiO₂ support alone showed lower LA conversion and low selectivity to GVL whereas Cu-SiO₂ catalysts were found to show high conversion and selectivity to the desired product with formic acid as the hydrogen source. The difference in the activities and product selectivity among different copper supported silica appears to be a complex function of acidity and the nature of copper species in the Cu-SiO₂ catalysts. No direct correlation was observed between a specific catalyst property such as acidity and pore size, and the performance of the Cu-SiO₂ catalysts. From the characterization, it appears that the species responsible for higher activity and selectivity to GVL are monomeric partially oxidized copper species in a strong interaction with the support in combination with a higher number of acid sites, both of which are present in Cu-SiO₂-Q6. Cu-SiO₂-Q6 showed the highest conversion and selectivity of 66 and 81%, respectively, under optimized conditions.

Full text of DOI:10.1039/c7cy00902j

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: A. Sultana, S.
Lomate and T. Fujitani, Catal. Sci. Technol., 2017, DOI: 10.1039/C7CY00902J.
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 40
Catalysis Science & Technology
View Article Online
DOI: 10.1039/C7CY00902J
Effect of SiO2 support properties on the performance of Cu-SiO2 catalysts
for the hydrogenation of levulinic acid to gamma valerolactone using
formic acid as hydrogen source
Samadhan Lomate, Asima Sultana* and Tadahiro Fujitani
Research Institute for Innovation in Sustainable Chemistry, National Institute of Advanced
Industrial Science and Technology (AIST), 1-1-1 Higashi, Tsukuba, Ibaraki 305-8565, Japan
*asima.sultana@aist.go.jp
Abstract
Vapor phase catalytic transfer hydrogenation of aqueous levulinic acid (LA) with
formic acid (FA) as hydrogen source was carried out over copper loaded on to several SiO2
supports having different physicochemical properties. The SiO2 supports and CuꢀSiO2
catalysts with different support properties were characterized by XRD, TPR, XPS, NH3ꢀTPD
and UVꢀVis techniques to evaluate the copper species and its interaction with the support.
The SiO2 support alone showed lower LA conversion and low selectivity to GVL whereas,
CuꢀSiO2 catalysts were found to show high conversion and selectivity to desired product with
formic acid as the hydrogen source. The difference in the activities and product selectivity
among different copper supported silica appear to be a complex function of acidity and the
nature of copper species in the CuꢀSiO2 catalysts. No direct correlation was observed between
specific catalyst property such as acidity, pore size with the performance of the CuꢀSiO2
catalysts. From the characterization it appears that the species responsible for higher activity
and selectivity to GVL are monomeric partially oxidized copper species in strong interaction
with the support in combination with higher number of acid sites both of which are present in
CuꢀSiO2ꢀQ6. CuꢀSiO2ꢀQ6 showed the highest conversion and selectivity of 66 and 81%
respectively under optimized conditions.

Catalysis Science & Technology
Page 2 of 40
View Article Online
DOI: 10.1039/C7CY00902J
Introduction
Significant advances have been made in recent years in using the heterogeneous
catalysts for converting biomassꢀderived compounds to fuels and chemicals.1ꢀ7 The rapidly
accelerating research on the biomass conversion has led to the identification of novel
platform molecules such as 5ꢀ(hydroxymethyl)ꢀ2ꢀ furaldehyde (5ꢀHMF)8, levulinic acid
(LA)9 and gammaꢀvalerolactone (GVL),10 which could either replace the currently used
fossilꢀbased chemicals or serve as renewable feedstock routes for their production.
Levulinic acid, owing to its low cost of production and presence of reactive functional
groups has been considered as a sustainable liquid and proposed for the production of valueꢀ
added chemicals11. Selective hydrogenation of levulinic acid (LA) to γꢀvalerolactone (GVL)
is one of the target reactions for converting lignocellulose into renewable fuels and chemicals.
Typically, catalytic hydrogenation of LA is performed with molecular hydrogen in the
presence of a precious metal catalyst.12ꢀ17 Production of GVL using a nonꢀprecious metal low
cost catalyst and biomass acid hydrolysis byꢀproduct formic acid (FA) as hydrogen source
would be of great advantage due to the reduced process cost. Recently, several research
groups have demonstrated formic acid as a promising source of molecular hydrogen for
hydrogenation of LA to GVL.18ꢀ26 However, so far either the reaction is carried out in liquid
phase, using high concentration of formic acid, using 1, 4 dioxane as solvent or using high
transition metal loading.12, 27 The dioxanes are not the best solvents to use due to its
carcinogenity.28 In addition, the real life feedstock used as hydrolysis mixture always
contains H2O. Hence, it is important to test the catalytic activity in the presence of water.
Therefore, one of the greatest challenges is to use crude solution obtained directly from
biomass hydrolysis (LA+FA+H2O) for the conversion of LA into GVL. Recently, Upare et
al.27 have reported the LA hydrogenation at atmospheric pressure over nickel promoted
copperꢀsilica catalyst using formic acid as hydrogen feed, however, 1, 4 dioxane was used as

Page 3 of 40
Catalysis Science & Technology
View Article Online
DOI: 10.1039/C7CY00902J
solvent. Varkula et al.16 reported high LA conversion of 95% over Ni/Al2O3, but 30% high
metal loading was used and also catalyst showed fast deactivated in the presence of water and
after 10 h reaction time showed only 30% conversion.
Recently, vapor phase hydrogenation of LA was investigated over Ni supported on
Al2O3, MgO and hydrotalcite using feedstock solution consisting of formic acid and levulinic
acid and found Ni/Al2O3 to be most efficient catalyst.18 Kumar et al.15 have studied the
influence of the nickel supported SiO2, ZrO2 and Al2O3 catalysts on the GVL yield at ambient
pressure and found that Ni/SiO2 exhibited better activity compared to Ni/ZrO2 and Ni/Al2O3
catalysts. In a preliminary investigation, on copper supported on Al2O3, SiO2, TiO2, ZSMꢀ5
and SiO2ꢀAl2O3 catalyst we have found that CuꢀSiO2 is the best catalyst for conversion of LA
to GVL. Luo et al.29 also reported on the influence of the support by comparing the acidity of
HꢀZSMꢀ5, HꢀBeta, TiO2 and ZrO2 for the hydrogenation of levulinic acid in liquid phase and
using different solvents. They achieved almost 100% conversion of LA and 97.5% GVL
yield but at high reaction temperature of 200 °C and 40 bar high H2 pressure.
On the basis of the available literature, we conclude that a large number of supports
have been tested for the hydrogenation of LA to GVL. However, a proper comparison in
terms of the specific property of an individual support is difficult as most studies are focused
on different supports and only a limited number of systematic studies are available in a series
of single support varying in their physicochemical properties. Furthermore, there are only a
scarce number of reports on vapor phase hydrogenation of levulinic acid using the more
realistic amount of formic acid and cost effective low metal loaded catalyst in the presence of
water. Herein, we show that the CuꢀSiO2 catalyst system can be used to form GVL from
reaction mixture consisting of LA, FA, and H2O. Our previous work on copper supported on
different supports revealed that under the action of acidꢀbase nature of support the electronic
and oxidationꢀreduction properties of the supported metal, its degree of dispersion, metal

Catalysis Science & Technology
Page 4 of 40
View Article Online
DOI: 10.1039/C7CY00902J
support interaction and a number of other characteristics are considerably changed. In the
present investigation, we report on our findings aimed at identifying the properties of a series
SiO2 support varying in their physicochemical properties that influence the performance of
CuꢀSiO2 catalyst for the conversion of levulinic acid to GVL.
Experimental
Materials and catalyst preparation
Five types of commercial SiO2 with average pore diameters of 2.8 nm (SiO2ꢀQ3), 2.9
nm (SiO2ꢀQ6), 9.6 nm (SiO2ꢀQ10), 11.4 nm (SiO2ꢀQ15), and 22.9 nm (SiO2ꢀQ30) were
purchased from Fuji Silysia Chemical Company Ltd. and used as catalysts. Table 1
summarizes the main characteristics of the catalyst samples. Cu(NO3)2.3H2O and formic acid
(FA) were obtained from WAKO, and levulinic acid (LA) from Alfa Aesar. All the catalysts
were prepared by the incipient wet impregnation method. In the typical procedure, an
aqueous solution containing required amount of Cu(NO3)2.3H2O salt was added to the
support in a beaker. The mixture was then placed at room temperature for 2 h and finally the
excess water was removed by heating at 85 °C using a hot plate with constant stirring. The
prepared 6wt%CuOꢀSiO2 catalysts were dried for 12 h at 110 °C and then calcined at 600 °C
for 3 h in static air with a temperature ramp rate of 2 °C minꢀ1.
Catalyst characterization
The wideꢀangle XRD patterns are collected on a Bruker D8 Advance Xꢀray
diffractometer using nickelꢀfiltered CuKα radiation (λ= 0.15406 nm) with a scanning angle
(2θ) range of 10~70°, a scanning speed of 2° minꢀ1, and a voltage and current of 40 kV and
40 mA, respectively. The full width at half maximum (FWHM) of Cu (111) reflection is
measured for calculating crystallite sizes using the Scherrer equation.

Page 5 of 40
Catalysis Science & Technology
View Article Online
DOI: 10.1039/C7CY00902J
The particle size and distribution were determined by using a highꢀresolution
transmission electron microscope (HRTEM; FEI Tecnai Osiris) operated at 200 kV.
The BET surface area of the catalysts was determined by N2 adsorptionꢀdesorption at
liquid N2 temperature using micrometrics ASAP 2020 instrument using 0.5 g of sample. The
sample was initially heated to 100 °C with a heating rate of 5 °C/min and then evacuated at
this temperature for 1 h and then at 350 °C for 3 h. Pore volume and average pore size were
calculated by the BJH method.
Temperature programmed reduction (H2ꢀTPR) of the catalysts was performed using
Belcatꢀbasic instrument. In this measurement, 0.1 g of the sample was placed in Uꢀshaped
quartz tube and the mixture of 10% H2/Ar (50 ml minꢀ1) was introduced to the sample tube at
room temperature until the baseline was stable. Then TPR measurement was performed in the
stream of 10% H2/Ar (50 ml minꢀ1) from room temperature to 715 °C with heating rate of
10 °C minꢀ1. Then the effluent gas was analyzed using Thermal Conductivity Detector (TCD).
UVꢀVis spectra of the catalysts were recorded by using a UVꢀVis (UVꢀ2600)
spectrometer in a diffuse reflectance mode between 200 and 800 nm range at a step of 0.5 nm
with a band width of 2 nm. BaSO4 was used as a reference sample to measure the baseline
spectrum.
Xꢀray photoelectron spectroscopy (XPS) was used to study the surface chemical composition
and the oxidation state of the species present in the reduced catalysts. Prior to the analysis, the
fresh catalysts were reduced in H2 at 500°C for 1 h (identical reduction conditions to those applied
in the activity tests). XPS analyses were performed using an ESCALAB 220i spectrometer
equipped with Mgꢀ Kα Xꢀray source (1253.6 ev). The XPS data were corrected with respect to
the adventitious carbon C 1s peak at 284.5 eV.

Catalysis Science & Technology
Page 6 of 40
View Article Online
DOI: 10.1039/C7CY00902J
The acidity of the samples was measured by ammonia temperature programmed
desorption (NH3ꢀTPD) technique. A quadrupole mass spectrometer (BEL JAPAN) was used
to detect desorbed NH3. Ca. 100 mg of the sample was out gassed at 500 °C for 1 h with the
temperature ramp rate of 10 °C minꢀ1 in He flow, then the sample was cooled down to 100 °C
and the reactor was flushed with He for 30 minutes followed by ammonia adsorption at
100 °C for 1 h. Subsequently, the sample was flushed with He for 30 min at 100 °C to
remove physically adsorbed ammonia. Then desorption of NH3 was carried out from 100 °C
to 700 °C using a heating ramp rate of 10 °C minꢀ1 The effluent NH3 gas was analyzed using
.
a thermal conductivity detector (TCD).
Pyridine adsorption was studied using selfꢀsupporting sample disk of about 15 mg cm2 in
a high vacuum system using infrared (FTIR) technique to determine the type of acid sites on
the samples. Prior to pyridine adsorption, the samples were heated to 500 °C at 10 °C minꢀ1 in
vacuum and then cooled down to room temperature. After this pretreatment, they were
exposed to saturated pyridine vapour for 10 min and infrared spectra were recorded after out
gassing at 150, 250, 350 and 450 °C using a Nicolet 6700 FTꢀIR spectrometer at a resolution
of 4 cm−1.
Catalytic activity
The catalytic hydrogenation of LA into GVL was conducted in a fixed bed reactor with
12 mm inner diameter and 364 mm length at atmospheric pressure. The required amount of
catalyst (0.5 g, unless specified) was sandwiched between two quartz wool beds in the middle
section of the reactor and a thermocouple was placed at the center of the catalyst bed to
measure the actual temperature of catalyst bed during the reaction. Then the reactor was
placed in an electric furnace consisting of three heating zones lower, middle and upper
equipped with three temperature controllers. Prior to the reaction, the catalysts were reduced
at 500 °C for 1 h in H2 at a flow rate of 50 ml minꢀ1. In the catalytic run, a mixture containing

Page 7 of 40
Catalysis Science & Technology
View Article Online
DOI: 10.1039/C7CY00902J
28% LA in aqueous mixture of LA and FA with a required molar ratio was fed to the reactor
through HPLC pump with nitrogen flow rate 50 ml minꢀ1. The liquid products condensed in
an ice cold trap are collected with an interval of 1 h. The collected products were analyzed by
a gas chromatography equipped with a DB WAX (60 M) capillary column and FID detector.
Results and discussion
The XRD spectra of the copper supported on SiO2 catalysts are shown in Fig. 1. The
XRD patterns exhibited diffraction peaks corresponding to amorphous SiO2 and crystalline
CuO.30 All catalysts, except CuꢀSiO2ꢀQ3, showed diffraction peaks at 2θ ≈35.48 and 38.70
corresponding to CuO. From the XRD pattern of CuꢀSiO2ꢀQ3 no detectable diffraction peaks
corresponding to CuO were detected. This led us to assume that the CuO species to be
present in the highly dispersed form. Further such high dispersion can be rationalized due to
the high surface area of the support SiO2ꢀQ3 which can lead to higher dispersion of copper
species. On the other hand SiO2ꢀQ6 with similar surface area as that of SiO2ꢀQ3 showed large
CuO crystallite size phase in XRD indicating that the support physicoꢀchemical properties in
addition to surface area are playing a significant role. Other small peaks characteristic of CuO
are also seen around 32.3, 48.7, 53.4 and 58.1. The formation of bulk CuO on silica, even at
Cu loadings smaller than 2%, was reported previously and it indicates the weak interaction
between the support and the Cu oxide.31 The relatively more intense diffraction peaks at 2θ
≈35.48 and 38.70 in CuꢀSiO2ꢀQ15, and CuꢀSiO2ꢀQ30 compared to CuꢀSiO2ꢀQ6 and CuꢀSiO2ꢀ
Q10 indicates the presence of bulk CuO in some CuꢀSiO2 samples. The diffraction peaks
corresponding to metallic copper at 2θ ≈43.3° and 50.5° and Cu2O at 36.5° were not
observed on any of the oxidized catalysts.32 The XRD profiles of reduced CuꢀSiO2 showed
only the diffraction pattern corresponding to Cuo while, there were no detectable CuO
diffraction patterns. It is interesting to point out that the CuꢀSiO2ꢀQ3 sample in oxidized form

Catalysis Science & Technology
Page 8 of 40
View Article Online
DOI: 10.1039/C7CY00902J
did not show the CuO crystallites however upon reduction the Cuo crystallites are visible
through XRD supporting the high dispersion of copper species in CuꢀSiO2ꢀQ3.
The crystallite sizes of the unreduced and reduced catalyst catalysts were calculated by
the Xꢀray line broadening method from the major diffraction peak of CuO at 2θ of 35.48°
(002) and metallic Cu0 at 2θ of 43.5 (111) using the Scherrer equation. The particle size of
unreduced and reduced (values in parenthesis for reduced catalysts) showed relatively large
CuO and Cu crystallites in CuꢀSiO2ꢀQ6 catalyst (Table.1). From the XRD results, there was
no specific assignable relationship between support surface area, pore volume, and particle
size. For the reduced sample, particle sizes trend for different catalyst are similar as
unreduced, but larger in size. This indicates that a growth of the supported particles occurred
during the reduction treatment at 500oC.
TEM of reduced catalysts was conducted to obtain insight into the crystallite size of
copper species. As shown in Fig. 2, the CuꢀSiO2ꢀQ3 catalyst consists of relatively uniformly
distributed spherical particles of about ~5 nm. On the other hand the CuꢀSiO2ꢀQ6 had
somewhat nonꢀuniformly distributed and larger (about 70nm), compared to CuꢀSiO2ꢀQ3,
crystallite size copper species. As expected the CuꢀSiO2ꢀQ30 catalyst also showed
significantly larger copper species among the catalysts characterized with TEM. The TEM
observations are consistent with the findings from TPR and XRD.
The surface area, pore diameter pore and volume of the CuꢀSiO2 catalysts and SiO2
supports are summarized in Table 1. The pore size distributions of the supports are shown in
Fig. 3. As can be seen in Fig. 3 and Table 1, in the case of SiO2 supports with decrease in the
surface area, the total pore volume and pore diameter increased from 0.08 to 1.3 cm3 gꢀ1 and
2.8 to 22.9 nm. It was found that the surface area, pore volume and pore diameter was
decreased in CuꢀSiO2 catalysts. The decrease in the surface area of SiO2ꢀQ3 and SiO2ꢀQ6
supports is more obvious, decreased from 786 and 720 m2 g−1 to 466 and 590 m2 g−1,

Page 9 of 40
Catalysis Science & Technology
View Article Online
DOI: 10.1039/C7CY00902J
respectively compared to other supports. The conspicuous reduction in surface area of SiO2ꢀ
Q3 and SiO2ꢀQ6 supports, on Cu impregnation suggests copper crystallites are blocking or
present in the pores of the supports. The CuꢀSiO2ꢀQ6 and CuꢀSiO2ꢀQ3 with high surface area
together with narrow pores in 20–50 Å range resulted into the most active catalysts. On the
other hand, the less active CuꢀSiO2ꢀQ10, CuꢀSiO2ꢀQ15, and CuꢀSiO2ꢀQ30 showed a broad
range of pores sizes. AlꢀShaal et al.33 reported that the high surface area of Ru supported
catalyst is beneficial for the catalytic hydrogenation of LA to GVL. Furthermore, Tsubaki et
al.34 reported that small Cu clusters are preferentially formed in the small or narrow pores of
SiO2ꢀQ3 compared to SiO2ꢀQ50 with bimodal small and large pores. The narrow pores could
render confinement of Cu particles to the pore walls, leading to easy accessibility of the
active sites for the reaction.35 The smaller Cu clusters are expected to be more active than
large clusters in the hydrogenation of levulinic acid to GVL in agreement with previous
observation by Zhang et al.36 who showed high activity of small Ru clusters for the
hydrogenation of levulinic acid to GVL.
The H2ꢀTPR profiles of CuO and different copper supported SiO2 catalysts are shown
in Fig. 4 and the total hydrogen consumption corresponding to the reduction of copper oxide
species are summarized in Table 1. In general two reduction peaks of copper oxide are
reported, the low temperature reduction peak is attributed to CuO/Cu2+ to Cu+ and a high
temperature peak to Cu+ to Cu0.37 It is also reported that the surface copper oxide species on
the support is more easily reduced than the bulk CuO.38 In the present study, TPR profiles of
all the CuꢀSiO2 catalyst showed only a single broad prominent reduction profile at lower
temperature region which indicates the presence of monoꢀdistribution of copper species on all
the supports. Similar observation of highly dispersed copper species with one reduction peak
was observed previously.39 TPR revealed lower onset reduction temperatures of CuO species
in CuꢀSiO2ꢀQ3 compared with other catalysts. As observed from XRD the CuO is well

Catalysis Science & Technology
Page 10 of 40
View Article Online
DOI: 10.1039/C7CY00902J
dispersed in CuꢀSiO2ꢀQ3 catalyst and the wellꢀdispersed copper oxide species reduces more
easily than the relatively large copper oxide species by H2. For the most active CuꢀSiO2ꢀQ6
catalyst the reduction of copper species occurred at higher temperatures indicating strong
interaction with the support. Based on the activity data such catalyst would also contribute to
its superior activity and stability under reaction conditions. Furthermore, it is known from the
literature that TPR peaks at low temperatures indicate smaller CuO particles or oligomeric
clusters with a relatively weak interaction with the support. The reduction signal at higher
temperature is indicative of the formation of relatively stronger interaction with the support
and larger CuO particles.40 From TPR results, it can be concluded that CuO species in strong
interaction with SiO2 facilitates the hydrogenation reaction.
The oxidation state of copper and the potential copper species in the catalysts are
further characterized by UVꢀVIS spectroscopy (Fig. 5). The absorption bands are observed
between 230ꢀ300, 300ꢀ500 and 600ꢀ800 nm respectively which are typical of copper
containing catalysts.41 The band between 230ꢀ300 nm has been assigned to the charge transfer
between mononuclear Cu2+ and oxygen and, between 300ꢀ500 nm to the presence of [Cu–O–
Cu]nꢀtype (oligomeric species), which are different from the bulk CuO. The band between
600ꢀ800 nm corresponds to the d–d transition of Cu2+ ions in an octahedral environment,
which can be attributed to bulk CuO. Besides this, in the present investigation, the calcined
catalysts showed different colors for example, CuꢀSiO2ꢀQ3 (greenish blue), CuꢀSiO2ꢀQ6
(grey), CuꢀSiO2ꢀQ15 (dark grey), CuꢀSiO2ꢀQ30 (bluish grey), indicating the potential
presence of different copper species.
The UVꢀVis spectrum of CuꢀSiO2ꢀQ3 and CuꢀSiO2ꢀQ6 catalysts exhibited absorption
bands around 250 nm and a broad band between 400ꢀ600 nm range, with significantly higher
relative intensities indicating the presence of monomeric Cu2+ and oligomeric [Cu–O–Cu]n+
copper clusters respectively, being in agreement with those previously reported and discussed

Page 11 of 40
Catalysis Science & Technology
View Article Online
DOI: 10.1039/C7CY00902J
in the literature for Cu+ and Cu2+ species.40 UVꢀVis in conjunction with XRD of CuꢀSiO2ꢀQ6,
which showed larger crystallites of CuO compared to CuꢀSiO2ꢀQ3 indicates that in the former
catalyst significant proportion of copper species are present as monomeric species and a
portion of copper species are present as large CuO crystallites. In CuꢀSiO2ꢀQ10, CuꢀSiO2ꢀ
Q15 and CuꢀSiO2ꢀQ30 relatively weak absorption bands were observed between 230 ꢀ250 nm
compared to CuꢀSiO2ꢀQ3 and CuꢀSiO2ꢀQ6. However they showed a broad band between 600ꢀ
800 nm corresponding to the d–d transition of Cu2+ ions in an octahedral environment, which
can be attributed to bulk CuO. The relative intensity of the band in 400ꢀ600 range is higher in
CuꢀSiO2ꢀQ3 compared to CuꢀSiO2ꢀQ6 indicating the presence of higher amount of
oligomeric copper in the former catalyst. The appearance of two different bands, with
different intensities in CuꢀSiO2ꢀQ3 and CuꢀSiO2ꢀQ6 catalysts with high surface area (Table 1),
suggests the presence of wellꢀdispersed copper species with different copper coordination and
the environment in these catalysts.
The nature of the copper species and its interaction with SiO2 in most active catalysts
were further investigated by XPS characterization. Fig. 6 shows the comparison of highꢀ
resolution XPS spectra of Cu in the 2p region for different copper catalyst. In general the
XPS spectrum of the catalysts shows typical Cu 2p3/2 and Cu 2p1/2 peaks with the measured
binding energy of around 933 and 953 eV, respectively, as well as their concomitant shakeꢀup
lines around 943 and 963 eV. The binding energy of the Cu 2p3/2 peak at 933 ~934 eV and
the characteristic shakeup feature at a binding energy of around 944 eV are indicative of Cu2+
species, while lower binding energy (931ꢀ932 eV) and the absence of shakeup peaks is
characteristic of Cu1+. The XPS spectra of the reduced CuꢀSiO2ꢀQ6 and CuꢀSiO2ꢀQ3 samples
exhibited a lack of a Cu2p peak typically observed between 933ꢀ944 eV, indicating the
absence of the Cu2+ species as expected42. The binding energies of Cu 2p1/2 at around 951.6
eV and 952.8 eV and Cu 2p3/2 at 931.5 eV and 932.8 eV, respectively were observed. The

Catalysis Science & Technology
Page 12 of 40
View Article Online
DOI: 10.1039/C7CY00902J
Cu2p3/2 peaks in CuꢀSiO2ꢀQ6 appeared at slightly lower binding energy at 931.5 and 951.6
eV (Fig. 6). The Cu 2p3/2 peak shift to slightly lower binding energy is an indication of the
partial reduction of the Cu species. The strong interaction with support in CuꢀSiO2ꢀQ6,
leading to more difficult to reduce species, could lead to formation of partially charged Cu
species. Monte et al43 reported presence of Cuδ+ atoms having a binding energy of 933.25 ±
0.35 eV, i.e. situated in between the Cu+ and Cu2+ binding energy values. Relatively strong
interaction of copper with the support is further evident from TPR and UVꢀVis
characterization.
With different textural properties of the support materials used they are expected to
have different acidꢀbase characteristics and influence the nature of active sites as well as
impacts the performance of the catalysts. The acidꢀbase properties of the supports and
catalysts are characterized by the NH3ꢀTPD technique from which the total acidic sites and
strength of acid sites were evaluated (Fig. 7). The acid sites responsible for NH3 desorption
below 200 °C are classified as weak and between 200ꢀ400 °C as medium and above 400 °C
as strong acid sites. SiO2 supports did not show significant NH3 storage as expected, on the
other hand, the presence of copper introduced the acid sites in the catalyst. The Cu+ can act as
Lewis acid site.44 The total amount and strength of acid sites in CuꢀSiO2 catalysts follows the
trend CuꢀSiO2ꢀQ3 ˃ CuꢀSiO2ꢀQ6 ˃ CuꢀSiO2ꢀQ10> CuꢀSiO2ꢀQ15 ˃ CuꢀSiO2ꢀQ30. Among the
CuꢀSiO2 catalyst, the CuꢀSiO2ꢀQ3 and CuꢀSiO2ꢀQ6 catalysts showed distinct features with a
peak appearing between 200ꢀ400 °C and the relative intensity of this peak is higher in Cuꢀ
SiO2ꢀQ6 catalyst than in CuꢀSiO2ꢀQ3 indicating presence of a large number of medium acid
strength sites in the CuꢀSiO2ꢀQ6 catalyst. The NH3ꢀTPD profiles of the reduced CuꢀSiO2
catalyst showed significantly lower NH3 storage than the oxidized catalysts. However the
NH3 storage on CuꢀSiO2ꢀQ3 and CuꢀSiO2ꢀQ6 still showed slightly higher NH3 storage than
the respective supports alone.

Page 13 of 40
Catalysis Science & Technology
View Article Online
DOI: 10.1039/C7CY00902J
The infrared spectra of pyridine adsorbed CuꢀSiO2ꢀQ3 and CuꢀSiO2ꢀQ6 samples at 150,
250 and 350 °C are shown in Fig. 8. The bands around 1540ꢀ1548 and 1445ꢀ1460 cmꢀ1 are
characteristic of Bronsted (PyH+) and Lewis (LꢀPy) acid sites, respectively.45 Both the
catalysts showed bands at 1450 and 1610 cmꢀ1 assigned to Lewisꢀbound pyridine (LꢀPy) and
a band at 1490 cmꢀ1 attributed to pyridine adsorbed on both Lewis and Bronsted acid sites (Bꢀ
Py and LꢀPy). The intensity of 1490 cmꢀ1 band is much higher in CuꢀSiO2ꢀQ3 and the
intensity increased with increase in evacuation temperature indicating that it possessed strong
Bronsted acid sites compared to CuꢀSiO2ꢀQ6 and other copper supported catalysts. The
higher lewis acidity in CuꢀSiO2ꢀQ6 could be associated with low coordination sites on the
metal particle. In addition, CuꢀSiO2ꢀQ3 catalyst showed a band at 1570 cmꢀ1 assigned to
hydrogen bonded pyridine.
The catalytic performances of different CuꢀSiO2 catalysts in the conversion of LA to
GVL are presented in Fig. 9A. All CuꢀSiO2 catalysts showed activity in reducing LA to GVL.
Among the catalysts the LA conversion over CuꢀSiO2ꢀQ6 and CuꢀSiO2ꢀQ3 was significantly
higher than other catalysts. The LA conversion and GVL selectivity reached up to 55 and
80% over the CuꢀSiO2ꢀQ6 catalyst at 250 °C, with others following in the decreasing order of
CuꢀSiO2ꢀQ6 > CuꢀSiO2ꢀQ3 > CuꢀSiO2ꢀQ10 > CuꢀSiO2ꢀQ15 >CuꢀSiO2ꢀQ30. Typically,
hydrogenation of LA gives GVL as a major product and AL and pentanoic acid (PA) as
minor products. Only the CuꢀSiO2ꢀQ6 showed significantly different selectivity trend
compared to rest of the catalyst with different conversion levels. Upare et al.27 obtained LA
conversion of 66 and 83% and GVL selectivity of 40 and 81% over the 10%CuꢀSiO2 and
80%CuꢀSiO2 using formic acid as a hydrogen source (LA: FA=1: 1). The copper loadings
they used are unusually high and moreover, they carried out the reaction at a low space
velocity (0.512 hꢀ1). In addition, the reaction mixture consisted of 10% diluted LA in 1, 4
dioxane and at a relatively higher reaction temperature of 265 °C. The route followed in this

Catalysis Science & Technology
Page 14 of 40
View Article Online
DOI: 10.1039/C7CY00902J
article is expected to have many advantages over some of the previously reported routes for
converting LA to GVL. We have obtained high conversion rates of LA to GVL using formic
acid as a hydrogen source and aqueous LA.
It is reported that the LA conversion to GVL proceeds through two consecutive steps,
dehydration of LA to angelica lactone over acid sites and followed by hydrogenation to GVL
on copper metal (Scheme .1).16, 46 Therefore the textural and chemical properties of support
as well as nature of copper species, are expected to impact the conversion and selectivity
behaviors. The SiO2 support and CuꢀSiO2 catalyst with high surface area and higher acidity
are expected to provide a large number of active sites for dehydration reactions, which is
obvious from the higher conversions observed on CuꢀSiO2ꢀQ3 and CuꢀSiO2ꢀQ6 catalysts
which showed LA conversions of 48 and 55%.
Fig. 9B compares the LA hydrogenation activity of different supports alone as catalysts.
In general, the SiO2 catalysts showed lower conversion of LA and relatively higher selectivity
to AL compared to CuꢀSiO2 catalysts, and SiO2ꢀQ6 gave the highest AL selectivity of 81%
among all the different SiO2 catalysts. SiO2ꢀQ3 having higher surface area gave a slightly
higher LA conversion of 19% compared to SiO2ꢀQ6 and Q10. When supports alone are used
for the conversion of LA, no hydrogenation is expected and therefore the main product is AL,
which only proceeds through acid catalyzed reaction. However all the supports showed
somewhat similar conversion and selectivity to products indicating no strong correlation of
performance to the support acidity. The results obtained in this study are in close agreement
with the report by Sato et al,47 where they reported lactonization of levulinic acid to angelica
lactone over the same SiO2 supports used in this study. From the correlation of acidity and
activity, Fig. 10, it appears that though the acidity is important in determining the catalytic
activity, the differences in total acidity are not proportionally reflected for determining
conversion and selectivity to various products.

Page 15 of 40
Catalysis Science & Technology
View Article Online
DOI: 10.1039/C7CY00902J
The difference in the activities and product selectivity among different copper
supported silica therefore appear to be a complex function of acidity and redox properties of
the CuꢀSiO2 catalysts. For instance, CuꢀSiO2ꢀQ3 having the highest total acidity among the
different supports used showed lower activity than CuꢀSiO2ꢀQ6. Furthermore, the NH3ꢀTPD
and TPR data suggest that the presence of medium strength acid sites in CuꢀSiO2ꢀQ6 and
presence of a relatively higher number of mononuclear Cu2+ species could be responsible for
its higher activity. As observed in TPR of CuꢀSiO2ꢀQ3, the lower reduction temperature
indication of weak metal support interaction results in higher mobility of the Cu2+ ions during
the thermal treatment which ultimately leads to partial clustering of the copper oxides and
forming oligomeric copper oxides (CuꢀOꢀCu)n+.40 The CuꢀSiO2ꢀQ6 catalyst with relatively
higher number of monomeric Cu2+ species showed higher activity compared to CuꢀSiO2ꢀQ3
with higher number of oligomeric copper species.
Based on the acidity and UVꢀVis characterization, it is concluded that the presence of
higher number of mononuclear Cu2+ and Lewis acid sites with medium acid strength are
important for the conversion of LA to GVL. In addition to conversion given in Fig. 9, the
activity defined as number moles of LA converted gꢀ1 of catalyst hꢀ1, are provided in table 1.
Based on the activity and copper species crystallite size data (Table 1), it appears that the
conversion and selectivity to be less sensitive to crystallite size indicating the conversion of
LA and selectivity to GVL are not structure sensitive reactions. On the other hand, the Cuꢀ
SiO2ꢀQ3 catalyst with a higher number of strong Bronsted acid sites showed higher
selectivity of AL and decreased GVL selectivity. Thus, based on FTIR results it can be
speculated that Lewis acid sites are more important than Bronsted acid sites in the conversion
of LA to GVL. This result is in accordance with the earlier report by Venugopal et al.17 who
demonstrated that the presence of Lewis acid sites influences the conversion of LA to AL and

Catalysis Science & Technology
Page 16 of 40
View Article Online
DOI: 10.1039/C7CY00902J
its subsequent conversion of AL to GVL over Ni sites and presence of Bronsted acid sites
have led to hydrogenolysis of GVL to VA and hydrocarbons over Ni supported catalysts.
The SiO2ꢀQ6 used in CuꢀSiO2ꢀQ6 yielded optimal conversion and desired selectivity
trends therefore in the rest of the work to optimize the catalyst and reaction conditions and to
further increase the LA conversion and GVL selectivity this support was utilized.
Effect of reaction conditions
Various reaction conditions were varied to influence the activity and product selectivity
to gain insights into the reaction path way as well as to optimize the performance. Fig. 11
demonstrates the LA conversion and GVL yield of the CuꢀSiO2ꢀQ6 catalyst at different
reaction temperatures ranging from 230 to 290 °C. The conversion of LA increased from
53% at 230 °C to 68% at 290 oC. However, the GVL selectivity remained almost same in the
temperature range of 230 to 270 °C. Further, increase in temperature to 290 °C the selectivity
of GVL decreased to 75%. The decrease in GVL selectivity at a higher temperature > 270 °C
could be due to the consecutive conversion of GVL by ring opening leading to the formation
of other products such as pentanoic acid due to overꢀhydrogenation at higher reaction
temperature. Hence, it was concluded that 270 °C is the optimum temperature for the LA
hydrogenation reaction over copper supported silica catalyst.
The presence of strong acid sites is reported to be favor dehydration reaction leading to
the formation of angelicalactone.16 Furthermore, both the dehydration and hydrogenation can
occur in parallel during the hydrogenation of LA to GVL, where the angelica lactone formed
gets further hydrogenated to GVL. It is expected that the reaction first proceeds by
hydrogenation of LA to 4ꢀhydroxypentanoic acid over metal sites exposed on the surface
followed by dehydration to GVL on support acidic sites16. However, as angelica lactone is
also obtained we believe that when hydrogenation is coupled with dehydration over

Page 17 of 40
Catalysis Science & Technology
View Article Online
DOI: 10.1039/C7CY00902J
bifunctional acid/metal catalyst it leads to high GVL selectivity. The presence of acids sites
with stronger strength in CuꢀSiO2ꢀQ6 combined with elevated reaction temperature could
favor the ring opening of GVL to pentanoic acid. Based on the product distribution obtained
in the present investigation it can be concluded that the hydrogenation is predominant than
dehydration, particularly higher reaction temperatures favors hydrogenation.
A general treatment after the catalyst preparation is calcination (heating the sample in
air, in order to clean up and stabilize the catalyst) and/or reduction (heating the sample in a
reducing atmosphere, usually containing hydrogen, in order to activate the catalytic metal) of
the catalyst. Since the reduction temperature is expected to strongly affect the catalytic
activity, the effect of reduction temperature on the hydrogenation of levulinic acid was
investigated. As expected with increase in reduction temperature from 250 to 500 oC, the LA
conversion and GVL selectivity was found to increase from 29 to 55% and 71 to 80%
respectively (Fig. 12).
The hydrogenation of LA using FA as a hydrogen source was reported to be strongly
depended on the concentration of FA in the reaction mixture.20, 29 The LA: FA ratio was
varied from 1:1 to 1:3 and results are presented in Fig. 13. As seen in the figure, the LA
conversion and GVL selectivity strongly depend on the LA: FA ratio. At otherwise similar
reaction conditions, both the LA conversion and GVL selectivity increased as the molar ratio
of FA increased. The conversion of LA and GVL selectivity with the molar ratio of LA: FA
was 1: 1 was 58% and 64%, and when the ratio of LA: FA increased to 1: 3 the LA
conversion and GVL selectivity increased to 73% and 84%. The low catalytic activity at low
LA: FA ratio is due to the insufficient H2 formation by decomposition of formic acid.
Furthermore, at the low LA: FA ratio (1:1) the lower selectivity to AL indicates that the
reaction mostly proceed first via the conversion of LA to AL with the subsequent
hydrogenation to GVL. If the reaction proceeds via dehydration as a first step the AL

Catalysis Science & Technology
Page 18 of 40
View Article Online
DOI: 10.1039/C7CY00902J
selectivity will be high under lower FA concentration, on the other hand, if reaction proceeds
via hydroxypentanoic acid the AL selectivity was not expected to vary significantly. Angelica
lactone is detected in all experiments and its selectivity decreased with increase in LA: FA
ratio and GVL selectivity increased. Although hydroxypentanoic acid is not detected due to
its instability, its formation and participation in the formation of GVL cannot be excluded.
The effect of WHSV was studied by changing catalyst amount in the reactor and
keeping other reactions conditions same. The effect of WHSV on catalytic performance is
shown in Fig. 14. As can be seen and expected, with an increase in WHSV from 1.1ꢁh−1 to
4.5ꢁh−1, the LA conversion decreased from 63 to 42%. With the decrease in WHSV, the
contact time of feed on the catalyst increased, and thus the increase in conversion was
expectable. The selectivity to GVL increased on increasing the WHSV from 81 to 85% and
the selectivity to pentanoic acid decreased. Pentanoic acid (PA) can be obtained by ringꢀ
opening of GVL to give pentenoic acid (PEA) followed by further hydrogenation, Higher
contact time of the formed GVL with the metal site may favor the ring open and further
hydrogenation of GVL to pentanoic acid. The stability of the 6%CuꢀSiO2ꢀQ6 catalyst during
the hydrogenation of LA was investigated over a period of 10 h timeꢀonꢀstream (TOS) at
250 °C and the results are shown in Fig.15. During TOS up to 10 h, LA conversion and
selectivity of GVL did not change appreciably. The selectivity of AL and other products also
did not increase or decrease during TOS of 10 h investigated. The long term stability
performance of the catalyst is important particularly when the reaction is carried out in the
presence of acid and water, as the corrosive nature of the acid may lead to metal leaching.
The 6%CuꢀSiO2ꢀQ6 catalyst showed a consistent conversion of LA (around 54%) and GVL
selectivity of 81% during the whole period tested indicating that there is no metal leaching at
least in the time period of 10 h tested.
Conclusion

Page 19 of 40
Catalysis Science & Technology
View Article Online
DOI: 10.1039/C7CY00902J
Copper catalysts supported on SiO2 are found to be highly active for the vaporꢀ
phase hydrogenation of levulinic acid to valerolactone using formic acid as the hydrogen
source. The SiO2 supports alone showed lower LA conversion and higher selectivity to
intermediate product AL, compared to CuꢀSiO2 catalysts and no strong correlation of
performance to the support acidity was detected. The difference in the activities and product
selectivity among different copper supported silica appear to be a complex function of acidity
and the nature of copper species in the CuꢀSiO2 catalysts. From the XRD, TPR, UVꢀVis, XPS
and NH3ꢀTPD it was observed that the species responsible for higher activity and selectivity
to GVL are monomeric partially oxidized copper species in strong interaction with the
support in combination with higher number of acid sites both of which are present in Cuꢀ
SiO2ꢀQ6. Among CuꢀSiO2 catalysts, CuꢀSiO2ꢀQ6 showed the highest conversion and
selectivity of 66 and 81% respectively under optimized conditions. From activity and
characterization results it can be rationalized that the strong acid sites may lead to
hydrogenolysis of GVL to pentanoic acid and copper species are responsible for the
selectivity to valerolactone.
Acknowledgements
The author’s gratefully acknowledges Prof. Masaaki Haneda from Nagoya Institute of
Technology for the use of the facilities and help with the FTꢀIR measurements.
References
1. Y. C. Lin and G. W. Huber, Energy Environ. Sci., 2009, 2, 68.
2. G. W. Huber, S. Iborra, and A. Corma, Chem. Rev., 2006, 106, 4044.
3. A. Corma, S. Iborra, and A. Velty, Chem. Rev., 2007, 107, 2411.
4. A. M. Ruppert, K. Weinberg, and R. Palkovits, Angew. Chem. Int. Ed., 2012, 51, 2564.
5. R. Rinaldi and F. Schuth, Chem. Sus. Chem., 2009, 2, 1096.

Catalysis Science & Technology
Page 20 of 40
View Article Online
DOI: 10.1039/C7CY00902J
6. Y. Wang, S. De and N. Yan, Chem. Commun., 2016, 52, 6210.
7. J. Zhang and N. Yan, Green Chem., 2016, 18, 5050.
8. C. O. Tuck, E. Perez, I. T. Horváth, R. A. Sheldon, M. Poliakoff, Science., 2012, 337, 695.
9. K. Yan, C. Jarvis, J. Gu, Y. Yan, Renew. Sustainable Energy Rev., 2015, 51, 986.
10. D. M. Alonso, S. G. Wettstein and J. A. Dumesic, Green Chem., 2013, 15, 584.
11. F. D. Pileidis and M.M. Titirici, ChemSusChem., 2016, 9, 562.
12. A. Piskun, J. G. M. Winkelman, Z. Tang and H. J. Heeres, Catalysts, 2016, 6, 131.
13. B. Putrakumar, N. Nagaraju, V. P. Kumar, K. V.R. Chary, Catal. Today, 2015, 250, 209.
14. M. Sudhakar, V. Vijay Kumar, G. Naresh, M. L. Kantam, S.K. Bhargava, A. Venugopal,
Appl. Catal. B., 2016, 180, 113.
15. V. V. Kumar, G. Naresh, M. Sudhakar, C. Anjaneyulu, S. K. Bhargava, J. Tardio, V. K.
Reddy, A. H. Padmasri and A. Venugopal, RSC Adv., 2016, 6, 9872.
16. V. Mohan, C. Raghavendra, C. V. Pramod, B. D. Raju and K. S. R. Rao, RSC Adv., 2014,
4, 9660.
17. A. M. Ruppert, M. Jędrzejczyk, O. S. Płatek, N. Keller, A. S. Dumon, C. Michel, P.
Sautet and J. Grams, Green Chem., 2016, 18, 2014.
18. M. Varkolu, V. Velpula, D. R. Burria and S. R. R. Kamaraju, New. J. Chem., 2016, 40,
3261.
19. L. Deng, Y. Zhao, J. Li, Y. Fu, B. Liao, and Q. X. Guo, ChemSusChem., 2010, 3, 1172.
20. G. Metzker and A. C. B. Burtoloso, Chem. Commun., 2015, 51, 14199.
21. X. L. Du, L. He, S. Zhao, Y. M. Liu, Y. Cao, H. Y. He, and K. N. Fan, Angew. Chem. Int.
Ed., 2011, 50, 7815.

Page 21 of 40
Catalysis Science & Technology
View Article Online
DOI: 10.1039/C7CY00902J
22. P. A. Son, S. Nishimura and K. Ebitani, RSC Adv., 2014, 4, 10525.
23. A. M. Hengne, A. V. Malawadkar, N. S. Biradar and C. V. Rode, RSC Adv., 2014, 4,
9730.
24. D. Kopetzki and M. Antonietti, Green Chem., 2010, 12, 656.
25. V. Fabos, L. T. Mika and I. T. Horvath, Organometallics, 2014, 33, 181.
26. L. Deng, J. Li, D. M. Lai, Y. Fu, and Q. X. Guo, Angew. Chem. Int. Ed., 2009, 48, 6529.
27. P. P. Upare, M. G. Jeonga, Y. K. Hwanga, D. H. Kim, Y. D. Kima, D. W. Hwanga, U. H.
Lee, J. S. Chang, Appl. Catal. A., 2015, 491, 127.
28. US department of health and human resources, 14th report on carcinogens, 2016, 1ꢀ3.
29. W. Luo, U. Deka, A. M. Beale, E. R.H.V. Eck, P. C. A. Bruijnincx, B. M. Weckhuysen.,
J. Catal., 2013, 301, 175.
30. X. Dong, X. Ma, H. Xu and Q. Ge., Catal. Sci. Technol., 2016, 6, 4151.
31. G. Aguila, F. Gracia, J. Cortes, P. Araya, Appl. Catal. B., 2008, 77, 325.
32. A. Patel, P. Shukla, T. Rufford, S. Wang, J. Chen, V. Rudolph, Z. Zhu, Appl. Catal. A.,
2011, 409, 55.
33. M. G. AlꢀShaal, W. R. H. Wright and R. Palkovits., Green Chem., 2012, 14, 1260.
34. Y. Zhang and N. Tsubaki, Stud. Surf. Sci. Catal., 2006, 172, 233.
35. G. Prieto, J. D. Meeldijk, K. P. de Jong and P. E. de Jongh, J. Catal., 2013, 303, 31.
36. Tie Yu, Jun Wang, Meiqing Shen, Wei Li, Catal. Sci. Technol., 2013, 3, 3234.
37. L. Kundakovic, M. F. Stephanopoulos, Appl. Catal. A., 1998, 171, 13.
38. Y. Yao, Z. Wang, S. Zhao, D. Wang, Z. Wu, M. Zhang, Catal. Today, 2014, 234, 245.
39. N. Scotti, M. Dangate, A. Gervasini, C. Evangelisti, N. Ravasio, F. Zaccheria, ACS Catal.
2014, 4, 2818.

Catalysis Science & Technology
Page 22 of 40
View Article Online
DOI: 10.1039/C7CY00902J
40. Q. Xin, A. Glisenti, C. Philippopoulos, E. Poulakis, M. Mertens, J. L. Nyalosaso, V.
Meynen and P. Cool, Catalysts, 2016, 6, 164.
41. C. Angelici, F. Meirer, A. M. J. V. Eerden, H. L. Schaink, A. Goryachev, J. P. Hofmann,
E. J. M. Hensen, B. M. Weckhuysen and P. C. A. Bruijnincx, ACS Catal., 2015, 5, 6005.
42. X. Ma, Z. Yang, X. Liu, X. Tan and Q. Ge, RSC Adv., 2015, 5, 37581.
43. A. Sultana, T. Nanba, M. Haneda, M. Sasaki, H. Hamada, Appl. Catal., B, 2010, 101, 61.
44. M. Monte, G. Munuera, D. Costa, J. C. Conesa and A. M. Arias, Phys. Chem. Chem.
Phys., 2015, 17, 29995.
45. T. Ding, H. Tian, J. Liu, W. Wu, J. Yu, Chin. J. Catal, 2016, 37, 484.
46. F. Zaccheria, N. Scotti, M. Marelli, R. Psaro and N. Ravasio, Dalton Trans., 2013, 42,
1319.
47. D. Sun, Y. Takahashi, Y. Yamada, S. Sato, Appl. Catal. A, 526 (2016) 62.

Page 23 of 40
Catalysis Science & Technology
CuO
SiO2
Cu0
Cu-SiO2-Q30
Cu-SiO2-Q15
Cu-SiO2-Q10
Cu-SiO2-Q6
Cu-SiO2-Q3
10
20
30
40
50
60
Degrees 2θ
Fig.1. XRD pattern of Cu supported on different SiO2 supports (thick line
– calcined samples and dotted line - calcined and reduced)

Catalysis Science & Technology
Page 24 of 40
6%Cu-SiO2-Q6 (Reduced)
6%Cu-SiO2-Q3 (Reduced)
6%Cu-SiO2-Q30 (Reduced)
Fig.2. TEM images of the reduced catalysts.

Page 25 of 40
Catalysis Science & Technology
5
4
3
2
1
A
(1) SiO2-Q3
(2) SiO2-Q6
(3) SiO2-Q10
(4) SiO2-Q15
(5) SiO2-Q30
4
2
1
5
3
(1) Cu-SiO2-Q3
(2) Cu-SiO2-Q6
(3) Cu-SiO2-Q10
(4) Cu-SiO2-Q15
(5) Cu-SiO2-Q30
5
4
3
2
1
0
B
4
2
1
5
3
10
110
210
310
410
510
610
Pore diameter (Å)
Fig.3. Pore size distribution in (A) SiO2 and (B) Cu/SiO2 catalysts.

Catalysis Science & Technology
Page 26 of 40
CuO
Cu-SiO2-Q10
Cu-SiO2-Q6
Cu-SiO2-Q30
Cu-SiO2-Q15
Cu-SiO2-Q3
150
200
250
300
350
400
Temperature (oC)
Fig. 4. H2-TPR profiles of Cu/SiO2 catalysts

Page 27 of 40
Catalysis Science & Technology
Monomeric Cu2+
239
Oligomeric clusters
Cu-SiO2-Q3
235
Cu-SiO2-Q6
Cu-SiO2-Q10
Cu-SiO2-Q15
Cu-SiO2-Q30
200
300
400
500
600
700
800
Wavelength (nm)
Fig. 5. Diffuse reflectance UV–Vis spectra of different Cu-SiO2 catalysts

Catalysis Science & Technology
Page 28 of 40
Cu2p
Cu 2p 3/2
931.5
Cu 2p 1/2
Reduced Cu-SiO2-Q6
932.8
Reduced Cu-SiO2-Q3
975
965
955
945
935
925
Binding energy (eV)
Fig. 6 XPS spectra of Cu 2p band of Cu-SiO2.

Page 29 of 40
Catalysis Science & Technology
Q6
Q3
Q10
Q15
Q30
x 3
Cu-SiO2 – calcined and reduced
Cu-SiO2 -calcined
x 3
Support
100 150 200 250 300 350 400 450 500
Temperature (oC)
Fig. 7. NH3-TPD profiles of different SiO2 and Cu/SiO2 catalysts