Catalytic oxidation of cellobiose over TiO2 supported gold-based bimetallic nanoparticles

Article abstract of DOI:10.1039/c4cy01566e

A series of Au-M (M = Cu, Co, Ru and Pd) bimetallic catalysts were supported on TiO₂via a deposition-precipitation (DP) method, using urea as a precipitating agent. The resulting catalysts were employed in the catalytic oxidation of cellobiose to gluconic acid and the properties of these catalysts were carefully examined using various characterization techniques. Cu-Au/TiO₂ and Ru-Au/TiO₂ catalysts demonstrated excellent catalytic activities in the oxidation of cellobiose to gluconic acid, though with contrasting reaction mechanisms. Complete conversion of cellobiose (100%) with a gluconic acid selectivity of 88.5% at 145 °C within 3 h was observed for reactions performed over Cu-Au/TiO₂; whereas, a conversion of 98.3% with a gluconic acid selectivity of 86. 9% at 145°C within 9 h was observed for reactions performed over Ru-Au/TiO₂. A reaction pathway was proposed based on the distribution of reaction products and kinetic data. It is suggested that cellobiose is converted to cellobionic acid (4-O-beta-d-glucopyranosyl-d-gluconic acid) and then gluconic acid is formed through the cleavage of the β-1,4 glycosidic bond in cellobionic acid over Cu-Au/TiO₂ catalysts. On the other hand, for reactions over the Ru-Au/TiO₂ catalyst, glucose was observed as the reaction intermediate and gluconic acid was formed as a result of glucose oxidation. For reactions over Co-Au/TiO₂ and Pd-Au/TiO₂ catalysts, fructose was observed as the reaction intermediate, along with small amounts of glucose. Co and Pd remarkably promoted the successive retro-aldol condensation reactions of fructose to glycolic acid, instead of the selective oxidation to gluconic acid. This journal is

Full text of DOI:10.1039/c4cy01566e

Catalysis
Science &
Technology
View Article Online
View Journal
PAPER
Catalytic oxidation of cellobiose over TiO2
supported gold-based bimetallic nanoparticles†
Cite this: DOI: 10.1039/c4cy01566e
a
b
a
Prince Nana Amaniampong,ab Xinli Jia, Bo Wang, Samir H. Mushrif,
*
b
a
*
*
Armando Borgna and Yanhui Yang
A series of Au–M (M = Cu, Co, Ru and Pd) bimetallic catalysts were supported on TiO2 via a deposition–pre-
cipitation (DP) method, using urea as a precipitating agent. The resulting catalysts were employed in the
catalytic oxidation of cellobiose to gluconic acid and the properties of these catalysts were carefully exam-
ined using various characterization techniques. Cu–Au/TiO2 and Ru–Au/TiO2 catalysts demonstrated excel-
lent catalytic activities in the oxidation of cellobiose to gluconic acid, though with contrasting reaction
mechanisms. Complete conversion of cellobiose (100%) with a gluconic acid selectivity of 88.5% at 145 °C
within 3 h was observed for reactions performed over Cu–Au/TiO2; whereas, a conversion of 98.3% with a
gluconic acid selectivity of 86. 9% at 145 °C within 9 h was observed for reactions performed over Ru–Au/
TiO2. A reaction pathway was proposed based on the distribution of reaction products and kinetic data. It
is suggested that cellobiose is converted to cellobionic acid IJ4-O-beta-D-glucopyranosyl-D-gluconic
acid) and then gluconic acid is formed through the cleavage of the β-1,4 glycosidic bond in cellobionic
acid over Cu–Au/TiO2 catalysts. On the other hand, for reactions over the Ru–Au/TiO2 catalyst, glucose
was observed as the reaction intermediate and gluconic acid was formed as a result of glucose oxidation.
For reactions over Co–Au/TiO2 and Pd–Au/TiO2 catalysts, fructose was observed as the reaction intermedi-
ate, along with small amounts of glucose. Co and Pd remarkably promoted the successive retro-aldol con-
densation reactions of fructose to glycolic acid, instead of the selective oxidation to gluconic acid.
Received 28th November 2014,
Accepted 29th January 2015
DOI: 10.1039/c4cy01566e
www.rsc.org/catalysis
Homogeneous catalysis has been widely used for the oxida-
tion of cellulose to gluconic acid.7–9 However, it imposes
Introduction
Abundant and renewable lignocellulosic biomass is consid-
ered as a promising alternative to non-renewable petroleum-
based resources. The use of second-generation biomass as a
feedstock to produce fuels and chemicals is sustainable and
is of great interest to help solve energy and environmental
problems in the future.1–5 Cellulose is the largest component
of lignocellulosic biomass, accounting for 40–50% by mass.2,3
Recently, extensive research to study the chemical/biological
transformations of cellulose into high-value added chemicals
such as gluconic acid and its derivatives has been performed
since gluconic acid is an important platform chemical which
is widely used in pharmaceutical applications and in the food
many disadvantages arising from product separation, forma-
tion of by-products and disposal of wastewater.8 The cost of
traditional fermentation processes to produce gluconic acid is
high due to the system-specific requirements, e.g., high tem-
perature, high level of purification and cost of bioreactors.
Hence, the development of novel heterogeneous catalysts is of
great significance to provide efficient and economical alterna-
tives to these primitive techniques.10
Attributed to the cohesive interactions between metal spe-
cies and supports, with either Lewis or Brønsted acid sites,
multifunctional catalysts have shown outstanding catalytic
performance in the conversion of cellulose to gluconic
acid.11–28 However, cellulose is a very complex macromole-
cule and is insoluble in most solvents. Hence, it is not easy
to perform fundamental research using cellulose as a feed-
stock. Cellobiose, which is a D-glucose dimer connected by a
β-1, 4 glycosidic bond, represents the simplest model of cellu-
lose. The studies on the catalytic conversion of cellobiose
may also provide helpful clues to efficiently transform cellu-
lose into valuable chemicals. Wang and co-workers examined
the conversion of cellobiose over gold nanoparticles
supported on various supports.12 They found that the
industry as
a
water-soluble cleanser and additive.6–9
a School of Chemical and Biomedical Engineering, Nanyang Technological
University, 62 Nanyang Drive, 637459, Singapore. E-mail: yhyang@ntu.edu.sg,
SHMushrif@ntu.edu.sg
b Institute of Chemical and Engineering Sciences, A*STAR (Agency for Science,
Technology and Research), 1 Pesek Road, Jurong Island, 627833, Singapore.
E-mail: Armando_Borgna@ices.a-star.edu.sg
† Electronic supplementary information (ESI) available: S1 XRD patterns of
bimetallic catalysts studied in this work, S2 UV-vis spectra of bimetallic catalysts,
and S3 NH3-TPD of Cu–Au/TiO2. See DOI: 10.1039/c4cy01566e
This journal is © The Royal Society of Chemistry 2015
Catal. Sci. Technol.

View Article Online
Paper
Catalysis Science & Technology
Brønsted acid sites in the carbon nanotube (CNT) had a pro-
nounced effect on the enhancement of the hydrolysis of cello-
biose to glucose, reporting a cellobiose conversion of 91%,
while the presence of Au nanoparticles promoted the oxida-
tion of glucose to gluconic acid with a selectivity of 60% at a
reaction time of 3 h. This group further investigated the con-
version of cellobiose over CNT-supported palladium, plati-
num, rhodium, copper, and silver catalysts and the best
results were obtained with platinum/CNT (cellobiose conver-
sion of 53% and gluconic acid selectivity of 23%).12 These
results revealed that the nature of the support and the type
of metal catalyst played important roles in the cellobiose con-
version and the selectivity towards gluconic acid.
Compared to pure metals, employing bimetallic catalysts
is believed to be a promising option for biomass feedstock
upgrading, as the interaction between the two metals can
modify the surface and electronic properties of the catalysts,
thereby enhancing significantly their catalytic activity and sta-
bility, even in the presence of undesired biomass by-prod-
ucts. Bimetallic Au-containing mixed metal catalysts have
been developed and it has been shown that bimetallic cata-
lysts exhibit higher catalytic activities than their monometal-
lic analogues.18 The effect of Cu, Co, Ru, Pt and Ni addition
to a Au catalyst was studied for several reactions, including
2,3-dihydrofuran synthesis,18 partial oxidation of methanol,19
water-gas shift29 and naphthalene hydrogenation.30
In this study, we prepared and characterized bimetallic
catalysts, such as Co–Au, Cu–Au, Pd–Au and Ru–Au,
supported on urea treated TiO2 for the oxidation of cellobiose
under aqueous conditions. In tune with the advantages of
the catalyst preparation methods described by Zanella
et al.,31 we employed the deposition–precipitation method
which has been proven to be an efficient method for prepar-
ing Au-based bimetallic catalysts32–34 to obtain metallic nano-
particles with a small size of only a few nanometers. The
catalytic activities of the bimetallic catalysts were then
compared with those of their corresponding monometallic
analogues and the influence of the second metal on the con-
version of cellobiose and selectivity towards gluconic acid
was studied. The aim of this work is to gain better insights
into the reaction mechanisms of cellobiose conversion to
gluconic acid and its derivatives over gold-based bimetallic
catalysts by identifying the key intermediate compounds and
also to attain better selectivity towards gluconic acid. To
better understand the structure–function relationship of the
catalysts, the catalyst nanoparticles were studied by X-ray
diffraction (XRD), X-ray photoelectron spectroscopy,
transmission electron microscopy (TEM), ammonia tempera-
ture programmed desorption (NH3-TPD), UV-vis spectroscopy,
X-ray fluorescence (XRF) and N2 physisorption. A reaction
pathway for cellobiose conversion to gluconic acid was also
proposed, as shown in Scheme 1.
Scheme 1 Proposed reaction pathways of cellobiose conversion to gluconic acid and its derivatives.
Catal. Sci. Technol.
This journal is © The Royal Society of Chemistry 2015

View Article Online
Catalysis Science & Technology
Paper
Table 1 Structural properties of the TiO2 support, actual metal loading and particle sizes of Au–M/TiO2 (M = Cu, Co, Ru and Pd)
BET surface area
Pore volume
Average pore size
(nm)
M contenta
(wt%)
Au contenta
(wt%)
Metal particle sizeb
(nm)
Sample
(m2 g−1
)
(cm3 g−1
)
TiO2
53.5
54.8
52.6
53.7
53.6
52.3
52.6
53.8
52.8
0.3
0.5
0.4
0.5
0.5
0.5
0.6
0.6
0.5
21.4
33.7
32.0
34.7
32.5
34.1
36.4
40.9
36.2
—
—
—
—
—
—
TiO2 (urea treated)
0.5Cu–0.5Au/TiO2
0.5Co–0.5Au/TiO2
0.5Ru–0.5Au/TiO2
0.5Pd–0.5Au/TiO2
1Ru/TiO2
0.450
0.440
0.350
0.482
0.970
0.860
—
0.480
0.480
0.488
0.501
—
—
1.010
5.6
3.0
4.6
3.6
1.5
1.6
2.1
1Cu/TiO2
1Au/TiO2
a
Actual metal contents obtained from XRF analysis. b Metal particle size measured by TEM.
Results and discussion
such there is no significant loss of metal precursors during
the catalyst preparation due to the effective precipitation
action of the urea utilized.
Catalyst characterization
It is well known that the surface properties of a catalyst play
an important role in its activity. High surface area and large
pore volume favour mass transport during the reaction.
Hence the textural properties of all as-prepared bimetallic
samples and their monometallic analogues were also
investigated.
As seen in Table 1, the pore volume of the titania support
treated with urea increases slightly as compared to that of
the commercial TiO2 support without urea treatment. An ear-
lier report35,36 showed that urea was a good pore forming
agent and as such helped in the formation of mesoporosity
in the titania framework. The current study also confirms
that adding urea increases the pore diameter to 34 nm, com-
pared to 21 nm of the standard sample without urea. It is
worth stating that the aim of this work is not to thoroughly
investigate the effect of urea on the pore structure of the
catalysts. X-ray fluorescence (XRF) analysis performed on
the as-synthesized catalysts evidences that the actual metal
loadings are close to the theoretical values (Table 1) and as
The XRD characteristic peaks corresponding to palladium,
gold, copper, ruthenium and cobalt are not detected for
Pd–Au, Cu–Au, Ru–Au and Co–Au, respectively, because of
the low amounts of metals loaded, high metal dispersion and
probably, the very high intensity of the TiO2 peaks (ESI† Fig. S1).
The particle dispersion and size distribution of both
monometallic and bimetallic catalysts on the TiO2 support
are further investigated by TEM observations, as illustrated
in Fig. 1. The metallic particles are finely dispersed on the
support surfaces.
The monometallic Ru, Cu and Au catalysts exhibit smaller
particle sizes compared to their bimetallic analogues, with an
average particle size of 2.1, 1.6 and 1.5 nm for Au/TiO2, Cu/
TiO2 and Ru/TiO2, respectively, obtained from the TEM
analysis.
For brevity, the TEM images of the monometals are not
shown. Few agglomerations of the metal particles are
observed for the bimetallic catalysts with average particle
sizes of 3.0, 3.6, 4.6 and 5.6 nm for Co–Au/TiO2, Pd–Au/TiO2,
Fig. 1 Transmission electron microscopy (TEM) images of as-prepared catalysts (a) Ru–Au/TiO2, (b) Cu–Au/TiO2, (c) Pd–Au/TiO2 and (d) Co–Au/TiO2.
This journal is © The Royal Society of Chemistry 2015
Catal. Sci. Technol.

View Article Online
Paper
Catalysis Science & Technology
Ru–Au/TiO2 and Cu–Au/TiO2, respectively, as shown in
Fig. 1a–d. The results reveal that the deposition–precipitation
method using urea is indeed viable for the synthesis of metal
nanoparticles with smaller size compared with other prepara-
tion methods reported in the literature.37 Chimentão et al.37
employed an impregnation method to prepare Au–Cu nano-
particles on TiO2 and only managed to achieve a wider parti-
cle size distribution of around 5–10 nm in the case of Au–Cu/
TiO2 and 8–15 nm for Au/TiO2 with large agglomerations up
to 100–250 nm.
XPS spectra were recorded for the prepared catalysts and
the high-resolution XPS spectra were curve fitted to evaluate
the chemical states of the different species present on the
surface of the TiO2 support. The individual XPS spectrum
and the corresponding binding energies (BE) of the different
chemical species are shown in Fig. 2. The BE of the Au 4f7/2
line displayed in Fig. 2 was observed at around 83.0–84.6 eV
for all the catalysts, indicating the presence of both reduced
Au nanoparticles (metallic Au0) and cationic nanoparticles
(Auδ+ and Au3+).
Fig. 2 X-ray photoemission spectra of as-prepared bimetallic samples.
Catal. Sci. Technol.
This journal is © The Royal Society of Chemistry 2015

View Article Online
Catalysis Science & Technology
Paper
However, in the case of Co–Au, Cu–Au, Pd–Au and Ru–Au,
the significant decrease in the BE value for the Au 4f7/2 line
(83.2, 83.1, 83.1 and 83.5 for Co, Cu, Pd and Ru containing
catalysts, respectively) suggests that electrons are transferred
from M (i.e., Co, Cu, Pd and Ru) to Au atoms and these inter-
actions of Au with Co, Cu, Pd and Ru lead to a change in the
electronic structure of Au.38
Copper in bimetallic Cu–Au/TiO2 exhibits peaks due to Cu
2p3/2 transitions at 932.2 eV and 933.0 eV, both of which cor-
respond to the presence of metallic Cu on the surface of
TiO2.
support. The peaks at high and low temperatures can be
attributed to the desorption of NH3 from the medium and
weak acid sites, respectively. The high temperature desorp-
tion of ammonia is ascribed to the desorption of coordi-
nated NH3 bound to Lewis acid sites.27 The peak at 261 °C
is attributed to the desorption of ammonia on the weak
Brønsted acid sites of the TiO2 lattice. The Lewis acid sites
contribute to the primary acidity responsible for the cellobi-
ose and cellobionic hydrolysis, and the successive retro-
aldol reaction of fructose to glycolic acid (a reaction by-
product of cellobiose oxidation) in the oxidation of cellobi-
ose to gluconic acid (Scheme 1). The total acidity estimated
from the analysis of NH3-TPD is 354.54 μmol g−1 which is
comparable to the acid strength of TiO2 estimated by
Yang et al. (362 μmol g−1, with a BET surface area of
79 m2 g−1).42
In addition, the peak at 934.7 eV is indicative of CuO after
calcination.
Palladium in bimetallic Pd–Au also shows characteristic
peaks at around 335.9 and 341.3 eV, indicating the formation
of metallic Pd and PdO on the surface of TiO2. In the case of
Co–Au/TiO2, cobalt species display peaks at 779.7 and
781.4 eV due to the Co 2p3/2 transition, indicating the pres-
ence of two forms of cobalt species on TiO2. These, in turn,
might correspond to Co3O4 and Co2O3 species, implying that
the cobalt present in the catalyst is not transformed into its
metallic form at a calcination temperature of 300 °C. The
results obtained are in agreement with most literature
reports.36,39,40
The UV-vis spectra of the bimetallic samples are shown in
the ESI† (Fig. S2). The absorbance in the visible region is sig-
nificantly pronounced for the Cu–Au/TiO2 catalyst. The broad
absorption peak at around 570 nm can be ascribed to the sur-
face plasmon resonance of Au nanoparticles. This plasmon
band is sensitive to the environment and can be shifted
depending on the stabilizer or the substrate. Because of the
coupling between the metal nanoparticles and the TiO2 sup-
port, the plasmon bands in almost all the bimetallic samples
are red-shifted.41
Oxidation of cellobiose
Cellobiose oxidation typically involves the hydrolysis of cello-
biose to glucose as the first step. Glucose possesses 3 differ-
ent types of carbons: aldehyde, primary alcohol and second-
ary alcohol. All three positions can be oxidized leading to
different products. The oxidation of the aldehyde produces a
carboxylic acid (e.g., gluconic acid), whereas the oxidation of
the primary and secondary alcohol affords glucuronic acid,
and keto-glucose and keto-acids, respectively. An appropri-
ately designed catalyst may greatly influence this well-studied
reaction pathway. According to the HPLC results in this
study, which will be discussed later in Tables 2 and 3, the
observed oxidized products are mainly gluconic acid and
glycolic acid, although there are other organic acids of lower-
carbon containing compounds formed as a result of the
decomposition or fragmentation of glucose and/or gluconic
acid.
NH3-TPD characterization was also conducted to survey
the acid strength of the Cu–Au/TiO2 catalyst sample. Here,
Cu–Au/TiO2 was chosen for the analysis because it was
found to be the most active catalyst towards the selective
oxidation of cellobiose to gluconic acid. Fig. S3 in the ESI†
represents the strength of the acid sites on the TiO2
Activity and product selectivity of various monometallic and
bimetallic catalysts
The effect of various monometallic and bimetallic catalysts
on the conversion of cellobiose was studied under the
Table 2 Oxidative conversion of cellobiose by O2 in H2O over Ru–Au/TiO2 catalysta
Cellobiose
conversion
[%]
Selectivity [%]
Glucose
Reaction
time [h]
Gluconic acid
Ethylene glycol
Glycolic acid
Erythritol
1
2
3
4
5
7
9
11
12
75.9
95.5
97.4
98.2
98.6
98.5
98.3
98.5
100
69.4
33.5
31.8
23.8
20.0
13.8
5.2
30.6
60.1
64.2
68.3
75.0
77.4
86.9
87.8
86.7
0.0
0.0
0.0
0.0
0.0
3.0
2.3
1.6
2.6
0.0
4.3
3.0
2.6
3.2
4.4
4.0
3.7
3.8
0.0
2.1
1.0
0.9
1.7
1.4
1.6
1.7
2.3
2.5
4.6
a
Reaction conditions: cellobiose: 0.6 mmol, catalyst (Ru–Au/TiO2): 0.100 g, H2O: 20 mL, pressure (O2): 1 MPa, temperature: 145 °C, stirring
speed: 1200 rpm.
This journal is © The Royal Society of Chemistry 2015
Catal. Sci. Technol.

View Article Online
Paper
Catalysis Science & Technology
Table 3 Oxidative conversion of cellobiose by O2 in H2O over Cu–Au/TiO2 catalysta
Cellobiose
conversion
[%]
Selectivity [%]
Glucose
Reaction
time [h]
Gluconic acid
Ethylene glycol
Glycolic acid
Cellobionic acid
1
2
3
4
5
7
9
11
12
96.7
98.6
100
100
100
100
100
100
100
6.0
9.6
11.5
0.9
0.0
0.0
0.0
0.0
0.0
66.1
70.4
88.5
86.8
81.4
68.6
58.8
43.0
41.3
0.0
0.0
0.0
4.3
7.6
11.3
17.7
27.5
28.8
0.0
0.0
0.0
27.9
20.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
8.9
11.0
15.0
22.0
26.0
25.6
a
Reaction conditions: cellobiose: 0.6 mmol, catalyst (Cu–Au/TiO2): 0.100 g, H2O: 20 mL, pressure (O2): 1 MPa, temperature: 145 °C, stirring
speed: 1200 rpm.
reaction conditions described at the end of the article. The
activities of the monometallic catalysts with respect to cello-
biose conversion and selectivity towards gluconic acid are
shown in Fig. 3. Catalytic activities are essentially similar over
various catalysts, with a cellobiose conversion of about 98%.
However, the product distribution and the selectivity towards
gluconic acid are different. Au/TiO2 exhibits excellent selectiv-
ity towards gluconic acid (75%), followed by Cu/TiO2 (63%).
A significant amount of formic acid (~25%) is formed over
Cu/TiO2, suggesting the superior hydrogenolysis abilities of
the Cu metal as compared to Au, where C–C bonds are
cleaved to yield lower carbon containing carboxylic acid com-
pounds. On the other hand, Ru/TiO2 shows poor selectivity
towards gluconic acid, suggesting that Ru nanoparticles may
not efficiently catalyze glucose oxidation to gluconic acid.
Moreover, Ru catalysts have been reported to be efficient cat-
alysts for the hydrolysis of β-1,4-glycosidic bonds.43 Pd/TiO2
and Co/TiO2 catalyze cellobiose to gluconic acid with selectiv-
ities of ~25% and ~23%, respectively, along with significant
amounts of glucose and fructose. The type of metal greatly
influences the product distribution and particularly the
selectivity towards the formation of gluconic acid. The activ-
ity of the monometallic catalysts followed the order: Au/TiO2
> Cu/TiO2 > Pd/TiO2 ≈ Co/TiO2 > Ru/TiO2.
It has been reported in several studies27,44 that the cata-
lytic oxidation of cellobiose in the presence of molecular oxy-
gen over Au nanoparticles supported on oxide supports gives
rise to some conversion of cellobiose and modest selectivity
towards gluconic acid. However, complete conversion of cel-
lobiose and good selectivity towards the formation of
gluconic acid over Au/TiO2 and Au-containing bimetallic cata-
lysts supported on TiO2 have not yet been achieved and
remain unclear. The highest conversion of cellobiose
reported till date is about 96% with a gluconic acid selectivity
of 63%, which was obtained by Wang et al.27 using Au/TiO2
with molecular oxygen at a pressure of 0.5 MPa and 145 °C.
Additionally, An et al. achieved a gluconic acid selectivity of
57% over Au/TiO2 using molecular oxygen as the oxidizing
agent at 145 °C.44 As reported in our previous work, a
gluconic acid selectivity of 86.0% within 3 h over Au/TiO2
reduced under hydrogen flow at 700 °C was achieved.35 It
was clarified that the generation of oxygen vacancies at such
a high reduction temperature significantly contributed to the
enhanced catalytic activity of the Au/TiO2 catalyst. In the
present study, the complete conversion of cellobiose and a
gluconic acid selectivity of 88.5% are demonstrated over Cu–
Au/TiO2 prepared by the deposition–precipitation method
using urea, in the presence of molecular oxygen at 145 °C in
3 h. The addition of a second metal to Au nanoparticles
increases the activity and selectivity to gluconic acid by 15%
compared to those of monometallic Au/TiO2 under the same
reaction conditions.
Reaction mechanism and reaction pathway investigations of
individual bimetallic catalysts
To understand the reaction pathways for the conversion of
cellobiose over the as-prepared bimetallic catalysts, detailed
analysis of the reaction products at different times was
performed and the results are shown in Tables 2–4. The
selectivity patterns for gluconic acid and the distribution of
products differ for all the catalysts used in this study. A
Fig.
catalysts.
3 Catalytic activity and product distribution of monometallic
Catal. Sci. Technol.
This journal is © The Royal Society of Chemistry 2015

View Article Online
Catalysis Science & Technology
Paper
plausible reaction mechanism is proposed and shown in
Scheme 1. Based on the distribution of the reaction products,
as shown in Table 2, for reactions performed over the Ru–Au/
TiO2 catalyst, the pathway for the conversion of cellobiose to
gluconic acid is suggested to be via glucose IJC6H12O6) as the
reaction intermediate. It is proposed that (i) water molecules
adsorb onto the acid site of the TiO2 support, (ii) cellobiose
then diffuses into the internal pores of TiO2, (iii) cellobiose
undergoes hydrolysis in the acid site with the adsorbed water
and the hydrolysis products diffuse out of the pores, and (iv)
glucose (the main reaction intermediate) is then oxidized to
gluconic acid in the presence of molecular oxygen with Au–Ru
nanoparticles as active sites. The gluconic acid IJC6H12O7) pro-
duced from glucose oxidation can further undergo decomposi-
tion or C–C, C–O and C–H bond cleavages to yield ethylene gly-
col, glycolic acid and erythritol, as observed in Table 2.
Table 3 illustrates the detailed distribution of the reaction
products obtained over the Cu–Au/TiO2 catalyst. It is found
that cellobionic acid is formed in significant amounts (~28%)
together with trace amounts of glucose (6%) in the initial
reaction stage. The selectivity for cellobionic acid decreases
with prolonged reaction time. The selectivity for glucose
increases sharply from 6% to 11.5% at 3 h and finally
decreases significantly when the reaction time reaches 4 h.
The selectivity for gluconic acid, the target product, increases
significantly, indicating that gluconic acid is formed from
the consecutive conversion of cellobionic acid. To ascertain
these phenomena, cellobiose oxidation reactions are
performed over Cu–Au/TiO2 for 15 min reaction time.
Cellobionic acid and glucose selectivities of 98.2% and 1.8%,
reactions performed over Co–Au/TiO2 and Pd–Au/TiO2
catalysts, as shown in Tables 4 and 5, respectively, similar
product distributions are observed. The slight difference
is that trace amounts of xylitol and formic acid are observed
at prolonged reaction times for reactions performed over
Pd–Au/TiO2. The presence of significant amounts of fructose
suggests that Co–Au/TiO2 and Pd–Au/TiO2 catalysts promote
the isomerisation of glucose to fructose and hence, fructose
is seen as the reaction intermediate, a phenomenon which is
different from what is observed for reactions performed over
Ru–Au/TiO2 and Cu–Au/TiO2. The presence of both glyceral-
dehyde and glycolic acid suggests that fructose which is the
reaction intermediate is converted to glyceraldehyde by a
retro-aldol reaction. The triose fragments are subsequently
converted to glycolic acid via the retro-aldol reaction of
glyceraldehyde to glycolaldehyde and further oxidation of
the latter to glycolic acid. These observations allow us to
suggest that the reactions performed over Co–Au/TiO2 and
Pd–Au/TiO2 follow the same reaction pathway, as evidenced
by the presence of similar reaction products, although Pd–
Au/TiO2 reveals a better oxidizing ability because glucaric
acid which is a typical over-oxidation product of gluconic acid
is observed. It is noteworthy that the formation of ethylene
glycol over Co–Au/TiO2 and Pd–Au/TiO2 is anticipated to
occur via the decomposition of gluconic acid at an elevated
reaction time. Based on these results, a plausible reaction
mechanism is proposed and shown in Scheme 1. More
intriguingly and interestingly, it is observed that different cat-
alysts with different reaction times exhibit enormously differ-
ent catalytic behaviours in either enhancing or suppressing
the over-oxidation and decomposition of gluconic acid to
other by-products, as well as the successive retro-aldol frag-
mentation reaction of fructose to produce glycolic acid.
respectively, and
a cellobiose conversion of 42.1% are
obtained. These observations strongly suggest that the first
step in cellobiose conversion over the Cu–Au/TiO2 catalyst is
the formation of cellobionic acid via oxidation of on glucose
unit in the cellobiose structure. Glucose can be formed as a
minor product through the hydrolysis of cellobionic acid cat-
alyzed by the acid sites. The small amount of glucose
observed even at prolonged reaction time suggests the ability
of glucose to rapidly oxidize to gluconic acid and other degra-
dation products such as glycolic acid and ethylene glycol. For
Correlation between metal–metal interactions and the
catalytic performance
Reactions performed over Ru–Au/TiO2 show an increase in
the selectivity for gluconic acid with time from 30.6% to
87.8% at 1 and 11 h, respectively. The XPS analysis shows the
Table 4 Oxidative conversion of cellobiose by O2 in H2O over Co–Au/TiO2 catalysta
Cellobiose
conversion
[%]
Selectivity [%]
Glucose
Reaction
time [h]
Gluconic acid
Ethylene glycol
Glycolic acid
Glyceraldehyde
Fructose
1
2
3
4
5
7
9
11
12
91.2
96.3
96.8
96.9
97.0
97.0
97.0
97.1
97.2
29.0
21.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
22.1
42.7
33.6
32.1
30.4
27.1
25.4
25.5
24.4
0.0
0.0
4.7
3.2
8.6
4.9
10.4
7.3
7.6
0.0
2.3
0.0
0.0
10.6
9.1
16.5
17.0
21.4
23.5
26.0
48.9
32.2
28.6
30.7
9.1
6.2
5.8
7.0
6.5
18.1
19.5
30.6
36.1
30.4
30.2
28.6
a
Reaction conditions: cellobiose: 0.6 mmol, catalyst (Co–Au/TiO2): 0.100 g, H2O: 20 mL, pressure (O2): 1 MPa, temperature: 145 °C, stirring
speed: 1200 rpm.
This journal is © The Royal Society of Chemistry 2015
Catal. Sci. Technol.

View Article Online
Paper
Catalysis Science & Technology
Table 5 Oxidative conversion of cellobiose by O2 in H2O over Pd–Au/TiO2 catalysta
Cellobiose
conversion
[%]
Selectivity [%]
Reaction
time [h]
Xylitol
Gluconic acid
Ethylene glycol
Glycolic acid
Glyceraldehyde
Formic acid
Sorbitol
Fructose
1
2
3
4
5
11
91.2
96.3
96.8
96.9
97.0
97.0
0.0
0.0
0.0
0.0
0.0
2.3
17.3
29.6
35.6
36.5
41.8
5.1
0.0
0.0
0.0
5.6
8.6
0.0
10.4
19.3
24.3
25.2
34.3
0.0
2.6
3.2
8.0
3.0
2.5
0.0
0.0
0.0
0.0
5.0
0.0
0.0
3.9
4.2
5.5
9.9
61.2
33.4
12.7
0.0
0.0
0.0
21.3
11.7
a
Reaction conditions: cellobiose: 0.6 mmol, catalyst (Pd–Au/TiO2): 0.100 g, H2O: 20 mL, pressure (O2): 1 MPa, temperature: 145 °C, stirring
speed: 1200 rpm.
presence of Au0 and Ru0, and the contact between Au and Ru
atoms favours the transfer of electrons from Ru0 towards Au0
due to the difference in the electronegativity, resulting in a
positive partial charge of Ru atoms and a negative partial
charge of Au atoms.45
explain why the reactions performed over Cu–Au/TiO2 pro-
ceed via cellobionic acid to gluconic acid. The formation of
an Au–CuO composite structure after calcination promotes
the oxidation of one glucose unit in the cellobiose structure
by the lattice oxygen in the CuO framework of the Au–CuO
composite structure. The consumption of the lattice oxygen
in the CuO framework of the Au–CuO composite structure on
the titania support creates oxygen vacancies on the copper
surface. These oxygen vacancies serve as potential adsorption
and activation sites for molecular oxygen. When the pro-
duced cellobionic acid is hydrolysed to gluconic acid and glu-
cose (via the Lewis acid sites located on the titania support),
the activated and highly reactive oxygen (O*) on the surface
of Cu–Au may initiate a nucleophilic attack on glucose to pro-
duce gluconic acid. The enhanced activity of Cu–Au/TiO2 in
the selective conversion of cellobiose to gluconic acid sug-
gests that the Au sites are accessible for the oxidation reac-
tion and interaction with CuO that can activate molecular
oxygen. According to other studies, CuO may be located on
titania and at the interface between the gold particles and
the titania support or as patches on the gold particles. Based
on EPR, XRD and XANES results, Liu et al.46 postulated that
for oxidation reactions involving Cu–Au bimetallic catalysts
supported on metal oxides, the catalytic structure was com-
posed of gold particles decorated by a layer or patches of
CuO, which maximized the perimeter between gold and
CuOx. The reactant adsorbed on gold would react with oxygen
activated on the neighbouring CuOx. In this way, the catalytic
activity of Cu–Au/TiO2 was greatly enhanced in comparison
with that of the monometallic gold and copper. This is con-
sistent with the fact that the plasmon band of Cu–Au/TiO2
evolves slightly differently from that of the other bimetallic
catalysts used in this study.
It has been postulated that for a molecule of an α,β-unsat-
urated aldehyde, the CO group could be adsorbed on the
Ru metal surface45 due to the partial negative charge (−δ) of
the O atom. The partially oxidized second metal species
strongly binds and activates the CO group and paves the
way for the selective oxidation of the CO bonds in the
aldehyde group of glucose by Au nanoparticles to yield
gluconic acid. Interestingly, it has also been shown that when
a second metal is added to Au, the oxidized state of the sec-
ond metal (Ru0 to RuO2) is able to strongly interact with the
support and can also function as an anchor to prevent the
aggregation of gold nanoparticles.46,47 The synergistic effect
of the Ru–Au/TiO2 catalyst contributed to the enhanced activ-
ity in selective conversion of cellobiose to gluconic acid,
where partially oxidized Ru (RuO2) interacts strongly with the
aldehyde functional group of glucose, as well as with the sup-
port material to afford strong metal–support interactions,
leading to highly active Au atoms for the selective oxidization
of the bound-aldehyde functional group of glucose. Consider-
ing the overall energy input and cost effectiveness of design-
ing a green approach for the production of high value-added
chemicals from renewable sources, employing Ru–Au/TiO2 as
a catalyst for the oxidative conversion of cellobiose into
gluconic acid may not be ideal since it requires an extended
reaction time to obtain high selectivity for the product which
translates into more energy input.
Reactions over Cu–Au/TiO2 show the presence of signifi-
cant amounts of glycolic acid, ethylene glycol and erythritol
at extended reaction times, although it exhibits the highest
activity among all the bimetallic catalysts investigated in this
study. The high activity of the Cu–Au/TiO2 catalyst in the con-
version of cellobiose to gluconic acid may also be correlated
with the charge transfer between the Au and Cu metal spe-
cies, which has been reported to increase the availability of
reactive oxygen in most oxidation reactions.48 According to
XPS analysis in this study, Cu is not fully reduced as
evidenced by the presence of a CuO peak; this may therefore
It has been reported that the type of metal–metal interac-
tion and particle morphology greatly influence the reaction
route.49–51 It has also been well documented that, when Pd–
Au/TiO2 is used in oxidation reactions, e.g., in the selective
oxidation of alcohols and biomass-derived compounds, sur-
face enrichment of Pd might take place, forming an Au-rich
core and a Pd-rich shell. In this case, the catalytic perfor-
mance of the Pd–Au bimetallic phase is actually determined
by the chemical composition of the Pd-rich shell, while Au
Catal. Sci. Technol.
This journal is © The Royal Society of Chemistry 2015

View Article Online
Catalysis Science & Technology
Paper
behaves more like a promoter to prevent over-oxidation and
poisoning of the Pd metal by intermediates or products. The
superior catalytic performance of Pd–Au/TiO2 in promoting
the cleavage of the C–C bond by the successive retro-aldol
condensation reaction of fructose to glyceraldehyde and
glycolaldehyde and subsequent oxidation of the latter to
glycolic acid, instead of the selective oxidation of glucose to
gluconic acid, is attributed to the fact that Pd containing cat-
alysts not only catalyze glucose isomerisation to fructose but
also largely decrease the energy barrier for the retro-aldol
fragmentation of fructose to trioses, i.e., dihydroxyacetone
and glyceraldehyde, as revealed by theoretical studies.51,52
Furthermore, Au can disperse/isolate Pd sites, preventing oxy-
gen poisoning of Pd in liquid-phase oxidation.53 Au, with
higher electronegativity, withdraws electrons from Pd atoms,
leading to an enhanced interaction of Pd atoms with the sub-
strate and thereby making the Pd atoms available to catalyze
the isomerisation of glucose to fructose (where Pd acts as a
Lewis acid catalyst),52 evidenced by the large amount of fruc-
tose present in the product mixture, and the subsequent low-
ering of the energy barrier to enhance retro-aldol reactions
yielding glycolic acid.
Reactions over Co–Au/TiO2 follow the same mechanism as
that observed with Pd–Au/TiO2 because product analysis
reveals substantial amounts of fructose and glyceraldehyde
and also a significant amount of glycolic acid. It is important
to state that the synergistic effect between gold and cobalt in
catalyzing glucose isomerisation, successive retro-aldol frag-
mentation and further oxidation reaction to glyceraldehyde
and glycolic acid is unclear and further studies to unravel the
mechanistic insight are under way in our laboratory.
observed with glucose oxidation over Cu–Au/TiO2 indicating
the slow rate of glucose isomerisation to fructose, which is
also evidenced by the presence of only a trace amount of fruc-
tose in the products. On the other hand, glycolic acid is pro-
duced in a significant amount while ethylene glycol is pres-
ent only in trace amounts, suggesting the further
decomposition or fragmentation of gluconic acid to glycolic
acid and ethylene glycol. As the main product of glucose oxi-
dation, gluconic acid is present in a large amount with a
selectivity of 68% as compared to all the other by-products.
The reaction with gluconic acid as the starting compound
shows a large amount of glycolic acid and ethylene glycol,
confirming that glycolic acid can be produced directly from
the decomposition of gluconic acid as well as small amounts
of ethylene glycol. Glucaric acid is also observed, which is a
typical over-oxidation product of gluconic acid. Fructose oxi-
dation shows only a small amount of gluconic acid with a
selectivity of 10.4%, while glycolic acid is formed in a compa-
rably large amount with a selectivity of 37.7%, confirming
the reaction pathway where fructose undergoes a retro-aldol
reaction catalyzed by the Lewis acid sites of the TiO2 support
to form glyceraldehyde and glycolaldehyde and finally oxida-
tion of the latter to yield glycolic acid.
A series of control experiments were also conducted to val-
idate the presence of other C1–C4 compounds formed in the
reaction pathway. Gluconic acid, glucose and fructose were
employed as substrates in the presence of molecular O2 at
145 °C within 3 h over the Cu–Au/TiO2 catalyst as shown in
Table 6 (entries 1–3). Reactions were also performed in the
absence of a catalyst (Table 6, entries 4 and 5). The presence
of a large amount of C1–C4 compounds (formic acid, xylitol,
oxalic acid, glycolic acid and ethylene glycol) indicates that
these compounds can be formed from the direct decomposi-
tion or fragmentation of glucose and fructose in the catalytic
oxidation of cellobiose.54 The absence of gluconic acid con-
firms that the bimetallic nanoparticles (Cu–Au) are responsi-
ble for providing the active sites for the formation of gluconic
acid.
To gain more insights into the reaction pathway for the
oxidation of cellobiose over the Cu–Au/TiO2 catalyst in water,
a number of catalytic oxidation experiments were performed
using glucose, gluconic acid and fructose as the starting sub-
strates, as shown in Table 6. In glucose conversion, the reac-
tion pathway may involve isomerisation of glucose to fruc-
tose. The distribution of reaction products as discussed
earlier shows that the rate of glucose isomerisation is highly
dependent on the intrinsic nature of the bimetallic catalyst
used in the reaction and the rate of glucose isomerisation
controls the selectivity to glyceraldehyde (which is the main
precursor to form glycolic acid). No glyceraldehyde is
Influence of acidity or basicity on product distribution
The effect of different solvents on the distribution of reaction
products and particularly on the selectivity towards gluconic
Table 6 Oxidative conversion of different biomass substrates by O2 in H2O over Cu–Au/TiO2 catalysta
Selectivity [%]
Conversion
[%]
Glycolic Ethylene Gluconic
Formic
Glyceraldehyde acid
Glucaric Oxalic
Substrate
Glucosea
Fructose acid
glycol
acid
Glucose Xylitol acid
acid
Others
91.3
5.7
0.0
10.0
51.4
—
20.5
33.9
37.7
5.5
1.6
25.3
10.8
6.4
68.4
—
10.4
0.0
0.0
0.0
6.6
0.0
0.0
0.0
0.0
21.6
28.6
15.0
—
0.0
2.0
—
43.3
1.9
0.0
0.0
4.2
6.1
0.0
11.8
0.0
0.0
0.0
—
1.9
29.0
—
—
—
Gluconic acida 59.0
0.0
0.0
3.9
6.8
Fructosea
Glucoseb
Fructoseb
90.0
88.9
95.2
16.0
2.2
0.0
a
Reaction conditions: cellobiose: 0.6 mmol, catalyst (Cu–Au/TiO2): 0.100 g, H2O: 20 mL, pressure (O2): 1 MPa, temperature: 145 °C, stirring
b
speed: 1200 rpm. Glucose and fructose: no catalyst, H2O: 20 mL, pressure (O2): 1 MPa, temperature: 145 °C, stirring speed: 1200 rpm.
This journal is © The Royal Society of Chemistry 2015
Catal. Sci. Technol.

View Article Online
Paper
Catalysis Science & Technology
Fig. 4 Effect of acidity or basicity on cellobiose oxidation. Reaction
conditions: cellobiose: 0.6 mmol, catalyst (Cu–Au/TiO2): 0.100 g, H2O:
20 mL, pressure (O2): 1 MPa, temperature: 145 °C, stirring speed: 1200 rpm.
acid was also investigated and the results are shown in
Fig. 4. Reactions performed in DI water were benchmarked
against those performed in NaOH (pH of 9) and H2SO4
(pH of 5) aqueous solutions. A complete conversion is
observed for the reactions performed in NaOH solution at
2 h and 145 °C, and the reaction products are composed of a
mixture of C2–C6 organic compounds. The alkaline environ-
ment of the reaction medium does not favour high selectivity
for gluconic acid possibly due to the fact that the degradation
of gluconic acid to other lower carbon containing organic
acids like glycolic acid is slightly more pronounced under
basic conditions. This is evidenced by the high selectivity for
glycolic acid formed in the presence of alkaline medium.
Whereas, only traces of glucose and glycolic acid are observed
for the reaction performed in H2SO4 together with a fructose
selectivity of 35.9%. Therefore, it appears that the bimetallic
Cu–Au/TiO2 can catalyze the oxidation of cellobiose into
gluconic acid more efficiently in neutral water, based on the
results shown in Fig. 4. The promotional effect of H2O in
aiding molecular oxygen adsorption and activation on the
TiO2 support in several oxidation reactions has been well
investigated and documented by several theoretical and DFT
calculations.55–57 This important phenomenon can explain
why reactions performed in water medium results in a higher
selectivity for gluconic acid.
Fig.
5
(a) Recycling of Cu–Au/TiO2 for oxidative conversion of
cellobiose. Reaction conditions: cellobiose: 0.6 mmol, catalyst (Cu–Au/
TiO2): 0.100 g, H2O: 20 mL, pressure (O2): 1 MPa, temperature: 145 °C,
stirring speed: 1200 rpm. (b) Representative TEM image of the spent
Au–Cu/TiO2 after post-treatment at 400 °C.
at 400 °C for 1 h. The purple colour of the catalyst is
restored, and the selectivity for gluconic acid reverts to 76.4%
after the 5th cycle and remains almost unchanged after the
6th cycle (75%). It is worth noting that the post-treatment of
the spent catalyst at 400 °C may generate CuOx that segre-
gates on the surface of the bimetallic nanoparticles. This
phenomenon accounts for the enhanced activity and the
increase in the selectivity for gluconic acid after the 5th cycle.
TEM analysis (Fig. 5b) of the spent catalysts confirmed the
segregation of amorphous CuOx on the surface of the bime-
tallic nanoparticles. These CuOx species served as activation
sites for molecular oxygen and water reduction, which are
critical for oxidation reactions.
In summary, reactions performed over the Ru–Au/TiO2
bimetallic catalyst exhibit lower catalytic activity and selectiv-
ity towards gluconic acid in an initial reaction time of 1 to 4 h,
compared to that of Cu–Au/TiO2. Nonetheless, at extended
reaction times (up to 12 h), Ru–Au/TiO2 is able to catalyze the
oxidation of cellobiose to gluconic acid with a maximum
selectivity of 87% at 9 h with no significant change in the
Catalyst reusability
Reusability is also a key factor in heterogeneous catalysis.
The results indicate that the selectivity for gluconic acid grad-
ually decreased to 70% after the 4th cycle (Fig. 5a). It is
noticed that the colour of the catalyst changed from purple
to deep brown, indicating an accumulation of possible
humins or carboxylic acid by-products on the surface of the
catalyst over time, possibly causing deactivation. Therefore,
post-treatment of the catalyst was performed by calcination
Catal. Sci. Technol.
This journal is © The Royal Society of Chemistry 2015

View Article Online
Catalysis Science & Technology
Paper
selectivity from 9 to 12 h. This interesting observation sug-
gests that Ru serves as a non-active oxophilic promoter
protecting gold from over-oxidation. This helps in limiting
the formation of by-products, and glucose (being the main
reaction intermediate in the Ru–TiO2 catalyzed reaction) is
then selectively converted to gluconic acid. The use of Co–Au/
TiO2 and Pd–Au/TiO2 bimetallic catalysts does not favour the
formation of gluconic acid, although both catalysts show
enhanced activities. Product analysis shows the presence of
significant amounts of C4, C3 and C2 organic acids, indicat-
ing the superior ability of Co–Au/TiO2 and Au–Pd/TiO2 in
performing C–O, C–C and C–H bond cleavages.
operating at an accelerating voltage of 200 keV. For the
TEM analysis, the samples were dispersed by ultrasonication
in ethyl alcohol for 30 min and a drop of the supernatant
liquid was placed onto a holey carbon film supported on a
copper grid.
XRF (with a Bruker AXS S4 Explorer) analysis was also
performed on the samples to confirm the metal loading
present.
The UV-vis spectra of the synthesized catalysts were
recorded in the scan range of 210–900 nm, using a UV-visible
spectrophotometer (Shimadzu model UV-2450; Shimadzu,
Kyoto, Japan) equipped with an integrating sphere, with
BaSO4 as the reference.
Temperature programmed desorption of NH3 (NH3-TPD)
was performed on a Micromeritics AutoChem 2920 appara-
tus. 200 mg of the catalyst was placed into a quartz U-tube,
heated for 2 h at 600 °C in Ar, and then kept at 100 °C for
NH3 adsorption. When saturated adsorption was achieved,
the system was swept with He for 3 h. Then the temperature
was programmed to increase to 600 °C under a heating rate
of 10 °C min−1. The desorbed NH3 was analyzed by using a
TCD detector.
Catalyst preparation by deposition–precipitation with urea
In a typical experiment, 1.0 g of TiO2 was dispersed in 50 mL
of 4.2 M urea aqueous solution under vigorous stirring.
Appropriate amounts of gold, ruthenium and copper precur-
sors were added to the suspension to yield 1 wt% gold, ruthe-
nium and copper, respectively, in the monometallic catalysts.
For Au-containing bimetallic samples, an Au concentration of
0.5 wt% and Cu, Co, Ru, Pd and Ni at 0.5 wt% each were
mixed to obtain 1 wt% total metal loading on the catalyst
support. The temperature of each suspension was increased
to 80 °C, and each suspension was stirred for 6 h to enable
slow decomposition of urea to ammonia, which then resulted
in an increased pH and consequently slow precipitation of
the metallic particles. The solids were finally recovered by fil-
tration and washed thoroughly with deionised water (DI), in
order to ensure complete removal of chlorine residues. There-
after, the solids were dried overnight in an oven at 90 °C and
finally calcined at 300 °C in air. The calcined samples were
then stored in desiccators to avoid any adverse effects of light
or atmospheric contaminants.
Reaction setup and product analysis
Oxidation of cellobiose was carried out in a 50 mL Teflon-
lined capacity batch reactor (PARR instrument) under reac-
tion conditions of 145 °C, 20 mL of DI H2O at 2 h, unless
otherwise stated. Cellobiose (typically 0.600 mmol) and the
catalyst (0.100 g) were added into the reactor pre-charged
with DI water. The system was charged with pure O2 at a con-
trolled pressure, after removing air by pressurizing and de-
pressurizing several times with pure O2. The reaction was
performed under different reaction conditions with steady
stirring (1200 rpm). After the reaction, the products were ana-
lyzed by a liquid chromatograph equipped with a RID-6A
refractive index detector and a Hi-Plex H (300 × 6.5 mm)
column with 0.01 M H2SO4 acid solution as the mobile phase
(flow rate of 1 mL min−1).
Catalyst characterization
Powder X-ray diffraction (XRD) patterns were recorded on a
Bruker AXS D8 diffractometer under ambient conditions
using Cu Kα radiation (λ = 0.15406 nm) from a Cu X-ray tube
operated at 40 kV and 40 mA. The diffractograms were
recorded in the 2θ range of 10–90°, in steps of 0.02° with a
count time of 20 s at each point. Prior to the test, the sam-
ples were dried at 80 °C overnight.
Cellobiose conversion is defined as:
Cg,0  Cg
Xg 
(1)
Cg,0
Textural properties were determined by nitrogen
physisorption using a Micromeritics TriStar apparatus. The
specific area was calculated using the Brunauer–Emmett–
Teller (BET) equation and the pore size distribution was ana-
lyzed with the Barrett–Joyner–Halenda (BJH) method.
The samples were also analyzed by X-ray photoemission
spectroscopy (XPS). XPS spectra were collected with a Thermo
Escalab 250 spectrometer. Spectra were recorded using an Al
anode (Al Kα = 1486.6 eV) with a 20 eV pass energy, 0.1 eV
energy step and 0.1 s dwelling time. Energy corrections were
performed using C1s (284.6 eV) as a reference.
where Cg is the concentration of cellobiose after a certain
reaction time and Cg,0 is the initial cellobiose concentration.
Product selectivity for a compound ‘p’ is defined as:
Cp
Sg 
(2)
Cg,0  Cg
where Cp is the concentration of a product after a certain
reaction time.
Conclusions
The morphology and particle size were determined by
transmission electron microscopy (TEM) using a JEOL JEM-2100F,
Cellobiose can be converted to gluconic acid with 88.5%
selectivity and 100% conversion if the reaction is catalyzed by
This journal is © The Royal Society of Chemistry 2015
Catal. Sci. Technol.

View Article Online
Paper
Catalysis Science & Technology
Cu–Au/TiO2 at 145 °C and an O2 pressure of 1.0 MPa within a
reaction time of 3 h. The M–Au/TiO2 bimetallic catalysts, with
M = Ru, Cu, Co and Pd, used in this study interestingly
showed different catalytic behaviours in converting cellobiose
to gluconic acid. The conversion of cellobiose to gluconic
acid, catalyzed by Cu–Au/TiO2, involves (i) the transformation
of cellobiose into cellobionic acid via the oxidation of one
glucose unit in the cellobiose structure, (ii) the hydrolysis of
the β-1,4 glycosidic bond in cellobionic acid to gluconic acid
and glucose, and (iii) glucose oxidation to gluconic acid. At
extended reaction times, gluconic acid is converted to smaller
carbon-containing carboxylic acids. Reactions over Ru–Au/
TiO2 revealed that glucose was the reaction intermediate,
which is oxidized to gluconic acid. However, kinetics of the
conversion of cellobiose to gluconic acid over Ru–Au/TiO2 is
slower than that over Cu–Au/TiO2. Reactions performed over
Co–Au/TiO2 and Pd–Au/TiO2 revealed fructose as the reaction
intermediate, promoting the successive retro-aldol condensa-
tion reactions of fructose to glycolic acid. The recycled Cu–
Au/TiO2 also showed appreciable amounts of gluconic acid
suggesting the superior stability of the reused catalyst even
after four cycles. The use of neutral water as the reaction sol-
vent proved to be much more efficient than using acidic or
basic reaction medium. Lastly, the enhanced activity, selectiv-
ity and stability of the Au–Cu/TiO2 bimetallic catalyst are due
to the metal–metal interactions caused by the redox proper-
ties of the Cu metal species.
9 H. Zhao, J. E. Holladay, H. Brown and Z. C. Zhang, Science,
2007, 316, 1597–1600.
10 A. Mirescu and U. Prüße, Appl. Catal., B, 2007, 70, 644–652.
11 S. Biella, L. Prati and M. Rossi, J. Catal., 2002, 206, 242–247.
12 X. Tan, W. Deng, M. Liu, Q. Zhang and Y. Wang, Chem.
Commun., 2009, 7179–7181.
13 B. Kamm, Angew. Chem., Int. Ed., 2007, 46, 5056–5058.
14 H. Wang, L. Zhu, S. Peng, F. Peng, H. Yu and J. Yang,
Renewable Energy, 2012, 37, 192–196.
15 V. R. Choudhary, D. K. Dumbre and S. K. Bhargava, Ind. Eng.
Chem. Res., 2009, 48, 9471–9478.
16 C. Della Pina, E. Falletta and M. Rossi, J. Catal., 2008, 260,
384–386.
17 T.-C. Ou, F.-W. Chang and L. S. Roselin, J. Mol. Catal. A:
Chem., 2008, 293, 8–16.
18 L. Leite, V. Stonkus, K. Edolfa, L. Ilieva, D. Andreeva,
L. Plyasova, J. W. Sobczak, S. Ionescu and G. Munteanu,
J. Mol. Catal. A: Chem., 2004, 215, 95–101.
19 F.-W. Chang, L. S. Roselin and T.-C. Ou, Appl. Catal., A,
2008, 334, 147–155.
20 E. V. Murzina, A. V. Tokarev, K. Kordás, H. Karhu,
J.-P. Mikkola and D. Y. Murzin, Catal. Today, 2008, 131, 385–392.
21 J. Kuusisto, A. V. Tokarev, E. V. Murzina, M. U. Roslund,
J.-P. Mikkola, D. Y. Murzin and T. Salmi, Catal. Today,
2007, 121, 92–99.
22 W.-C. Li, M. Comotti and F. Schüth, J. Catal., 2006, 237,
190–196.
23 F.-W. Chang, T.-C. Ou, L. S. Roselin, W.-S. Chen, S.-C. Lai
and H.-M. Wu, J. Mol. Catal. A: Chem., 2009, 313, 55–64.
24 H.-Y. Lin and Y.-W. Chen, Ind. Eng. Chem. Res., 2005, 44,
4569–4576.
25 P. Serna, P. Concepción and A. Corma, J. Catal., 2009, 265, 19–25.
26 M. Conte, A. F. Carley, G. Attard, A. A. Herzing, C. J. Kiely
and G. J. Hutchings, J. Catal., 2008, 257, 190–198.
27 D. An, A. Ye, W. Deng, Q. Zhang and Y. Wang, Chem. – Eur.
J., 2012, 18, 2938–2947.
Acknowledgements
This project is funded by the National Research Foundation
(NRF), Prime Minister's Office, Singapore under its Campus for
Research Excellence and Technological Enterprise (CREATE)
program. The authors also gratefully acknowledge support from
the Singapore Agency for Science, Technology and Research
(A*STAR) for releasing grants for this research.
28 J. Zhang, X. Liu, M. N. Hedhili, Y. Zhu and Y. Han,
ChemCatChem, 2011, 3, 1294–1298.
29 A. Venugopal, J. Aluha and M. Scurrell, Catal. Lett., 2003, 90,
1–6.
30 B. Pawelec, A. M. Venezia, V. La Parola, S. Thomas and
J. L. G. Fierro, Appl. Catal., A, 2005, 283, 165–175.
31 R. Zanella, L. Delannoy and C. Louis, Appl. Catal., A,
2005, 291, 62–72.
32 S. Ajaikumar, M. Golets, W. Larsson, A. Shchukarev,
K. Kordas, A. R. Leino and J. P. Mikkola, Microporous
Mesoporous Mater., 2013, 173, 99–111.
Notes and references
1 J. N. Chheda, G. W. Huber and J. A. Dumesic, Angew. Chem.,
Int. Ed., 2007, 46, 7164–7183.
2 A. Corma, S. Iborra and A. Velty, Chem. Rev., 2007, 107,
2411–2502.
3 D. A. Simonetti and J. A. Dumesic, Catal. Rev.: Sci. Eng.,
2009, 51, 441–484.
4 S. Van de Vyver, J. Geboers, P. A. Jacobs and B. F. Sels,
ChemCatChem, 2011, 3, 82–94.
5 J. A. Geboers, S. Van de Vyver, R. Ooms, B. Op de Beeck,
P. A. Jacobs and B. F. Sels, Catal. Sci. Technol., 2011, 1,
714–726.
33 S. Ajaikumar, J. Ahlkvist, W. Larsson, A. Shchukarev,
A. R. Leino, K. Kordas and J. P. Mikkola, Appl. Catal., A,
2011, 392, 11–18.
6 B. Girisuta, L. P. B. M. Janssen and H. J. Heeres, Chem. Eng.
Res. Des., 2006, 84, 339–349.
34 A. Sandoval, C. Louis and R. Zanella, Appl. Catal., B,
2013, 140–141, 363–377.
7 S. Govindaswamy and L. M. Vane, Bioresour. Technol.,
2010, 101, 1277–1284.
35 P. N. Amaniampong, K. Li, X. Jia, B. Wang, A. Borgna and
Y. Yang, ChemCatChem, 2014, 6, 2105–2114.
8 S. Naik, V. V. Goud, P. K. Rout and A. K. Dalai, Renewable
Sustainable Energy Rev., 2010, 14, 578–597.
36 Z. Jiang, W. Zhang, L. Jin, X. Yang, F. Xu, J. Zhu and
W. Huang, J. Phys. Chem. C, 2007, 111, 12434–12439.
Catal. Sci. Technol.
This journal is © The Royal Society of Chemistry 2015

View Article Online
Catalysis Science & Technology
Paper
37 R. J. Chimentão, F. Medina, J. L. G. Fierro, J. Llorca,
J. E. Sueiras, Y. Cesteros and P. Salagre, J. Mol. Catal. A:
Chem., 2007, 274, 159–168.
38 W. Huang, Z. Zuo, P. Han, Z. Li and T. Zhao, J. Electron
Spectrosc. Relat. Phenom., 2009, 173, 88–95.
39 R. C. King, J. F. Moulder and J. Chastain, Handbook of
X-ray photoelectron spectroscopy: a reference book of standard
spectra for identification and interpretation of XPS data,
Physical Electronics, Eden Prairie, Minn, 1992.
40 J.-Y. Zheng, J.-B. Pang, K.-Y. Qiu and Y. Wei, Microporous
Mesoporous Mater., 2001, 49, 189–195.
47 X. Liu, A. Wang, X. Yang, T. Zhang, C.-Y. Mou, D.-S. Su and
J. Li, Chem. Mater., 2008, 21, 410–418.
48 L. Chmielarz, R. Dziembaj, T. Grzybek, J. Klinik, T. Łojewski,
D. Olszewska and A. Węgrzyn, Catal. Lett., 2000, 70, 51–56.
49 A. Wang, X. Y. Liu, C.-Y. Mou and T. Zhang, J. Catal.,
2013, 308, 258–271.
50 Y. Chen, H. Wang, C.-J. Liu, Z. Zeng, H. Zhang, C. Zhou,
X. Jia and Y. Yang, J. Catal., 2012, 289, 105–117.
51 W. Deng, Q. Zhang and Y. Wang, Catal. Today, 2014, 234,
31–41.
52 W. Dedsuksophon, K. Faungnawakij, V. Champreda and
N. Laosiripojana, Bioresour. Technol., 2011, 102, 2040–2046.
53 A. N. Kholodovich, P. A. Simonov, R. I. Kvon, A. V. Bukhtiyarov,
V. I. Zaikovskii, Y. A. Chesalov and A. V. Romanenko,
Kinet. Catal., 2008, 49, 886–892.
41 X. Li, T. Fan, H. Zhou, B. Zhu, J. Ding and D. Zhang,
Microporous Mesoporous Mater., 2008, 116, 478–484.
42 M. Yang, Y. Men, S. Li and G. Chen, Appl. Catal., A,
2012, 433–434, 26–34.
43 T. Komanoya, H. Kobayashi, K. Hara, W.-J. Chun and
A. Fukuoka, J. Energy Chem., 2013, 22, 290–295.
44 D. An, A. Ye, W. Deng, Q. Zhang and Y. Wang, 10th European
Congress on Catalysis, EUROPACAT-X, Glasgow, Scotland, 2011.
45 L. N. Protasova, E. V. Rebrov, T. S. Glazneva, A. Berenguer-Murcia,
Z. R. Ismagilov and J. C. Schouten, J. Catal., 2010, 271,
161–169.
54 L. Xin, Z. Zhang, J. Qi, D. Chadderdon and W. Li, Appl.
Catal., B, 2012, 125, 85–94.
55 L. Liu, B. McAllister, H. Ye and P. Hu, J. Am. Chem. Soc.,
2006, 128, 4017–4022.
56 A. Bongiorno and U. Landman, Phys. Rev. Lett., 2005, 95,
106102.
57 O. Pozdnyakova-Tellinger, D. Teschner, J. Kröhnert,
F. C. Jentoft, A. Knop-Gericke, R. Schlögl and A. Wootsch,
J. Phys. Chem. C, 2007, 111, 5426–5431.
46 X. Liu, A. Wang, L. Li, T. Zhang, C.-Y. Mou and J.-F. Lee,
J. Catal., 2011, 278, 288–296.
This journal is © The Royal Society of Chemistry 2015
Catal. Sci. Technol.