AuPt Alloy on TiO2: A Selective and Durable Catalyst for l-Sorbose Oxidation to 2-Keto-Gulonic Acid

Article abstract of DOI:10.1002/cssc.201501202

Pt nanoparticles were prepared by a sol immobilization route, deposited on supports with different acid/base properties (MgO, activated carbon, TiO₂, Al₂O₃, H-Mordenite), and tested in the selective oxidation of sorbose to 2-keto-gulonic acid (2-KGUA), an important precursor for vitamin C. In general, as the basicity of the support increased, a higher catalytic activity occurred. However, in most cases, a strong deactivation was observed. The best selectivity to 2-KGUA was observed with acidic supports (TiO₂ and H-Mordenite) that were able to minimize the formation of C₁/C₂ products. We also demonstrated that, by alloying Pt to Au, it is possible to enhance significantly the selectivity of Pt-based catalysts. Moreover, the AuPt catalyst, unlike monometallic Pt, showed good stability in recycling because of the prevention of metal leaching during the reaction.

Full text of DOI:10.1002/cssc.201501202

DOI: 10.1002/cssc.201501202
Full Papers
AuPt Alloy on TiO2: A Selective and Durable Catalyst for
l-Sorbose Oxidation to 2-Keto-Gulonic Acid
Carine E. Chan-Thaw,[a] Lidia E. Chinchilla,[c] Sebastian Campisi,[a] Gianluigi A. Botton,[c]
Laura Prati,[a] Nikolaos Dimitratos,*[b] and Alberto Villa*[a]
Pt nanoparticles were prepared by a sol immobilization route,
deposited on supports with different acid/base properties
(MgO, activated carbon, TiO2, Al2O3, H-Mordenite), and tested
in the selective oxidation of sorbose to 2-keto-gulonic acid (2-
KGUA), an important precursor for vitamin C. In general, as the
basicity of the support increased, a higher catalytic activity oc-
curred. However, in most cases, a strong deactivation was ob-
served. The best selectivity to 2-KGUA was observed with
acidic supports (TiO2 and H-Mordenite) that were able to mini-
mize the formation of C1/C2 products. We also demonstrated
that, by alloying Pt to Au, it is possible to enhance significantly
the selectivity of Pt-based catalysts. Moreover, the AuPt cata-
lyst, unlike monometallic Pt, showed good stability in recycling
because of the prevention of metal leaching during the reac-
tion.
Introduction
Carbohydrates represent suitable and sustainable platform
molecules as replacements for fossil-derived resources in the
synthesis of fine chemicals.[1–5] The selective oxidation of carbo-
hydrates could be challenging for the production of several
products.[1–5] The liquid-phase oxidation of glucose to gluconic
acid with noble metals (Pd, Pt, and Au) and molecular oxygen
as the oxidant has been the subject of several investiga-
tions.[2,4,6] By contrast, only a few studies have been reported
with regard to the catalytic oxidation of l-sorbose to 2-keto-
gulonic acid (2-KGUA), an intermediate in the production of l-
ascorbic acid (vitamin C).[7–13] Presently, the transformation of l-
sorbose to 2-KGUA consists of an expensive and time-consum-
ing three-step process that involves the protection of sorbose
with acetone, oxidation with NaOCl or KMnO4, and hydrolytic
deprotection to give 2-KGUA.[14] Efforts have been made to re-
place this methodology with a more convenient catalytic route
consisting of a single-step process in the presence of noble-
metal-based catalysts and molecular oxygen as the oxidant in
aqueous solution.[7–13] Platinum-based catalysts were more
active and more selective for 2-KGUA than palladium-based
ones.[7–13] Moreover, a variety of products deriving from parallel
or consecutive reactions has been reported to occur, as shown
in Scheme 1. In particular, the oxidation of the OH group on
[a] Dr. C. E. Chan-Thaw, S. Campisi, Prof. L. Prati, Dr. A. Villa
Dipartimento di Chimica
Università degli Studi di Milano
via Golgi 19, 20133 Milano (Italy)
alberto.villa @unimi.it
[b] Dr. N. Dimitratos
Cardiff Catalysis Institute, School of Chemistry
Cardiff University
Main Building, Park Place, Cardiff, CF10 3AT (UK)
E-mail: DimitratosN@cardiff.ac.uk
Scheme 1. Reaction scheme for l-sorbose oxidation.
[c] Dr. L. E. Chinchilla, Prof. G. A. Botton
Canadian Centre of Electron Microscopy and Department of Materials Sci-
ence and Engineering
the C6 atom instead of the one on the C1 atom leads to the
formation of 5-keto-gluconic acid (5-KGLA) instead of 2-KGUA
(Scheme 1), with both reactions showing a similar rate. More-
over, products deriving from CÀC cleavage, such as glycolic,
McMaster University
1280 Main Street West, Hamilton, Ontario L8S 4M1 (Canada)
Supporting Information for this article is available on the WWW under
http://dx.doi.org/10.1002/cssc.201501202.
ChemSusChem 2015, 8, 4189 – 4194
4189
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Papers
oxalic, and formic acids, are formed, particularly if high-temper-
ature and alkaline conditions are used.[7–13]
mean diameter of 4.4 nm was measured (Table S1 in the Sup-
porting Information). The catalytic performances of the sup-
ported nanoparticles were tested in the l-sorbose oxidation by
using a Metrohm Titrino 718 apparatus with the ability to pro-
vide high precision control of the pH value (Æ0.01) and a glass
reactor that was especially designed to perform reactions
under pressure (maximum pressure of 4 bar) and controlled
pH values (Scheme S1 in the Supporting Information). The re-
action conditions were optimized to maximize the yield of 2-
KGUA (0.3m sorbose, pressure of O2 =2 bar, T=508C, metal/
sorbose=1:250 (mol/mol), pH 7.5).
Brçnnimann et al., by using Pd/Al2O3, reported a selectivity
of 67% at 58% conversion at a moderate temperature (508C)
and a pH value close to neutrality (7.3).[7] In this study, a high
metal/substrate molar ratio (1:30) was used as a result of the
low pH value required for the stability of the desired product.
It has been shown, indeed, that the presence of a base acceler-
ates H abstraction, which is the rate-determining step in alco-
hol oxidation.[1] Moreover, a strong deactivation phenomenon
occurred, mainly because of Pt passivation by O2 and the irre-
versible adsorption of products.[7] The same group showed
that the addition of Bi or Pb to Pt increased the reaction rate
but decreased the selectivity for 2-KGUA (31–38%, instead of
67% with unmodified Pt).[7] By contrast, the modification of Pt
with phosphines or amines could significantly enhance the se-
lectivity but slightly decreased the activity and the catalyst
durability.[8–11] The best reported results were achieved with
hexamethylenetetramine with a selectivity of 80% for 2-KGUA
at 50% conversion.[10] Subsequently, it has been reported that
Pt impregnated on a resin, such as hyper-cross-linked polystyr-
ene, is able to produce 2-KGUA with a selectivity of 98% and
full conversion at T=708C and pH 6.5.[12,13]
However, these latter results have recently been under
debate because of the reliability of using the iodometric
method to determine the yield of 2-KGUA.[15] It is therefore evi-
dent that the development of a selective and durable catalyst
for l-sorbose oxidation to 2-KGUA still remains a challenge.
It was reported that, by alloying Au with Pt, it is possible to
enhance the activity, selectivity, and stability of Pt catalysts in
the liquid-phase oxidation of glycerol and sorbitol, even in the
absence of a base.[16–18] The addition of Au has been shown to
improve the resistance of Pt to overoxidation by O2 and to
avoid or minimize the leaching of metals into the reaction so-
lution.[1,16–19] Furthermore, it was highlighted that the support
can also play a fundamental role in improving the selectivity,
by limiting the occurrence of side reactions.[16] The aim of this
work is to explore the extension of the use of alloyed AuPt cat-
alysts in the liquid-phase oxidation of l-sorbose and to deter-
mine if the use of these alloyed catalysts can also be beneficial
in this case in terms of activity, selectivity, and stability. In addi-
tion, the role of the support on the catalytic performance was
investigated by depositing the metal nanoparticles (NPs) on
supports with different morphologies and surface chemistries,
namely H-Mordenite, TiO2, activated carbon, and MgO.[16,18]
The catalytic data are presented in Table 1. As all of the Pt
catalysts showed similar mean diameter (3–4 nm; Table S1 in
the Supporting Information), the different activities can be
likely addressed to the different surface chemistries of the sup-
Table 1. Results of sorbose oxidation with Pt, Au, and AuPt catalysts at
30% conversion and after 8h of reaction time.[a]
Catalyst
Conver-
Selectivity [%]
Carbon
sion [%] 2-KGUA 5-KGLA C1/C2 others balance [%]
1%Pt/MgO
1%Pt/Al2O3
1%Pt/AC
30
24
8
7
5
12
12
10
8
14
15
12
12
18
15
11
10
50
54
35
37
43
46
20
21
24
24
19
26
4
4
6
8
8
7
8
5
5
12
12
3
3
4
85
73
96
96
98
95
98
98
96
96
98
98
99
99
81[b]
30
38
36
44
40
59
57
48
48
58
54
80
75
37[b]
30
59[b]
30
1%Pt/TiO2
51[b]
1%Pt/H-Mordenite 30
29[b]
1%AuPt/AC
30
89[b]
30
1%AuPt/TiO2
83[b]
9
5
[a] Reaction conditions: 0.3m Sorbose, O2 pressure=2 bar, T=508C,
metal/sorbose=1:250 mol/mol, pH 7.5. [b] Conversion after 8 h of reac-
tion time.
port and possible metal–support interactions. We found
a direct correlation between the activity and the acid/base
properties of the support. In general, with an increase in the
basicity of the support (MgO (pH 12.1)>Al2O3 (8.1)ꢀAC (7.6)>
TiO2 (5.5)>H-Mordenite (2.8); Table S2 in the Supporting Infor-
mation), a higher catalytic activity was observed: Pt/MgO
(81%)>Pt/AC (59%)>Pt/TiO2 (51%)>Pt/Al2O3 (37%)>Pt/H-
Mordenite (29%; Table 1) after 8 h of reaction time.
This trend confirms the observation that a basic support is
able to accelerate the H abstraction, which is the rate-deter-
mining step in the oxidative dehydrogenation of alcohols and
polyols.[1] Pt/Al2O3 represents an exception by showing an ac-
tivity that is lower than that of more acidic catalysts. We as-
cribed this behavior to the bigger Pt particle size (4.4 nm) rela-
tive to that of the other catalysts (3.4–3.7 nm). For a correct
comparison, the selectivity was first evaluated at iso conver-
sion (30%; Table 1). In terms of selectivity, Pt/MgO, the most
active catalyst, showed a very low selectivity for 2-KGUA (24%
at 30% conversion) with the promotion of the formation of C1
Results and Discussion
Pt catalysts were synthesized with a sol immobilization tech-
nique.[20] Pt nanoparticles with a mean diameter of 3.2 nm
(Table S1 in the Supporting Information) were synthesized in
the presence of polyvinyl alcohol (PVA) as the protective agent
and NaBH4 as the reducing agent. The Pt NPs were then immo-
bilized on the different metal oxides (H-Mordenite, TiO2, Al2O3,
MgO) or on activated carbon (AC). A small increase of the par-
ticle size was observed for Pt nanoparticles after the immobili-
zation on the support, in particular for Pt/Al2O3, for which a Pt
ChemSusChem 2015, 8, 4189 – 4194
www.chemsuschem.org
4190
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Papers
and C2 products derived from CÀC cleavage. Indeed, at higher
conversion (81%), the formation of degradation products is
more evident and a selectivity of only 8% was obtained
(Table 1). Moreover, the carbon balance of around only 85%
was ascribed to the formation of CO2. Pt NPs supported on
slight basic catalysts (Al2O3 and AC) showed a higher selectivity
to 2-KGUA (38 and 44%, respectively) with a limited formation
of liquid C1/C2 products and CO2 relative to Pt/MgO (Table 1).
The best selectivity was observed with acidic supports (TiO2
and H-Mordenite) that were able to minimize the formation of
C1/C2 products. Pt/TiO2 showed the best results with a selectivi-
ty for 2-KGUA of 59% (Table 1). The selectivity did not signifi-
cantly change at higher conversion (Table 1 and Figure S1a in
the Supporting Information).
Figure 1. HAADF STEM images of the AuPt/TiO2 catalyst and the correspond-
ing Au-La and Pt-La elemental maps from individual particles.
cles were bimetallic (AuPt) in nature. A high-magnification
HAADF STEM image and the corresponding XEDS spectrum
from a representative AuPt/TiO2 particle that demonstrates the
AuPt alloy structure is shown in Figure 2.
From an examination of the reaction profiles (Figure S2 in
the Supporting Information), it can be observed that the ma-
jority of the catalysts, except Pt/MgO, did not show good dur-
ability, as evidenced by a strong deactivation phenomenon
after 4 h of reaction time. This deactivation could be explained
by an irreversible chemisorption of the carboxylic acids on the
Pt active sites. Moreover, inductively coupled plasma (ICP) anal-
ysis showed a moderate Pt leaching (2–6%).
To improve the catalyst durability, we modified the Pt cata-
lysts by adding Au, in a nominal ratio of Au:Pt=6:4, and se-
lected the Pt/TiO2 and Pt/AC catalysts that showed the highest
selectivity for 2-KGUA. AuPt catalysts were prepared by a two-
step procedure, in which the Pt PVA-protected salt was slowly
reduced by H2 in presence of supported Au. A similar mean di-
ameter for the AuPt nanoparticles (3.4–3.7 nm) was obtained
in both cases. The detailed morphology of AuPt/AC is reported
elsewhere[19] and reveals the presence of AuPt alloy, whereas
a full characterization of AuPt/TiO2 is described herein.
High-angle annular dark-field (HAADF) images of the AuPt/
TiO2 catalyst showed well-dispersed metal nanoparticles on
TiO2 (Figure S3a,b in the Supporting Information). The catalyst
had a normal distribution of particles with sizes in the 1–7 nm
range, with a mean diameter of (3.7Æ0.1) nm and a total
metal dispersion of around 30% (Table S1 and Figure S4 in the
Supporting Information). Individual atoms and sub-nanometer
clusters were also detected in the catalyst, along with a limited
number of larger agglomerated structures (Figure S3c,d in the
Supporting Information).
A representative selection of HAADF scanning transmission
electron microscopy (STEM) images of the AuPt particles sup-
ported on TiO2 and the corresponding energy-dispersive X-ray
spectroscopy (XEDS) elemental maps are presented in Figure 1.
The spatial distribution of Au and Pt shown in Figure 1 re-
vealed the coexistence of Pt and Au elements within the nano-
particle, which suggests the formation of random gold–plati-
num alloy structures. Quantitative XEDS analysis was per-
formed in about 26 nanoparticles. The resulting particle-size-
composition diagram showed the presence of very small plati-
num nanoparticles (around 3.5 nm), gold nanoparticles varying
between 3 and 9 nm in size, and bimetallic nanoparticles in
the composition range of 35–65 wt% Pt (Figure S5 in the Sup-
porting Information). The XEDS analyses from individual parti-
cles confirmed that the majority of the supported nanoparti-
Figure 2. Representative high-magnification HAADF STEM image of the
AuPt/TiO2 catalyst and the corresponding XEDS spectrum obtained from the
individual nanoparticle.
The bimetallic catalysts showed an enhanced conversion rel-
ative to the corresponding monometallic counterparts
(Table 1). Moreover, the presence of Au increased the durability
of the catalysts. Indeed, in a comparison of the reaction pro-
files of Pt and AuPt catalysts (Figure 3), the bimetallic catalysts
appeared less prone to deactivate during the reaction.
In terms of selectivity, AuPt catalysts showed higher selectiv-
ity for 2-KGUA than the corresponding monometallic catalysts
(Table 1). At approximately 90% conversion, AuPt/AC gave a se-
lectivity of 54% with the formation of 26% of C1/C2 products,
whereas AuPt/TiO2 showed a selectivity of 75%, one of the
ChemSusChem 2015, 8, 4189 – 4194
www.chemsuschem.org
4191
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Papers
(TEM) performed on the used catalyst did not show any appre-
ciable changes in AuPt particle size (Table S1 in the Supporting
Information); therefore, the deactivation was attributed to the
presence of adsorbed species. Indeed, after thorough washing
of the catalyst with distilled water, the original activity was re-
stored and kept at least for another two runs. No leaching of
Au or Pt was observed from the ICP results, which confirmed
the stability of the AuPt/TiO2 catalyst.
Conclusions
In summary, we have demonstrated that, by alloying Pt to Au,
it is possible to enhance significantly the selectivity of Pt-based
catalysts in the selective oxidation of sorbose to 2-keto-gulonic
acid (2-KGUA), an important precursor for vitamin C. In particu-
lar, by choosing the appropriate support (TiO2), it is possible to
increase the selectivity for 2-KGUA, probably by decreasing the
reactivity of H2O2 with l-sorbose and 2-KGUA to form C1/C2
products. Moreover, the AuPt/TiO2 catalyst showed good stabil-
ity in recycling. For the restoration of the initial activity after
four runs, the catalyst needs to be washed with an excess of
water, probably to remove strongly adsorbed species from the
active sites.
Figure 3. Reaction profiles for Pt and AuPt catalysts supported on activated
carbon and TiO2.
best reported results in the literature, and showed a significant
decrease in the formation of CÀC cleavage products (9%). It
was reported that the CÀC bond cleavage can be a result of
the formation of H2O2 through O2 reduction by a metal hy-
dride.[16,21–23] Therefore, we determined the amount of H2O2
present at the end of the reaction, by titration with KMnO4. In
the presence of AuPt/TiO2, a concentration of 0.089 mmolLÀ1
of H2O2 was detected, whereas a ten times higher amount of
H2O2 was detected with AuPt/AC (0.813 mmolLÀ1), which is di-
rectionally consistent with the selectivity data obtained. Experi-
ments on H2O2 decomposition with the two catalysts showed
a similar degradation rate (Figure S6 in the Supporting Infor-
mation). Therefore, we ascribed the different selectivity of the
two catalytic systems to the higher tendency of AuPt/AC to
catalyze the reaction between H2O2 and sorbose, which thus
promotes the CÀC scission.
Experimental Section
Materials
NaAuCl4·2H2O and K2PtCl4 were from Aldrich (99.99% purity). H-
Mordenite and TiO2 P25 were from Degussa, Al2O3 from Condea,
MgO from Alfa Aesar, and activated carbon (X40S; surface area=
1086 m2 gÀ1; pore volume (PV)=1.5 mLgÀ1, intensity ratio of the D
and G bands (ID/IG)=2.4) was from Camel. NaBH4 (purity>96%)
from Fluka and polyvinylalcohol (PVA; mw =13000–23000, 87–89%
hydrolyzed) from Aldrich were used. A 1wt% PVA solution in water
was prepared. The gaseous oxygen from SIAD was 99.99% pure.
To evaluate the resistance of AuPt to deactivation, stability
tests have been performed on AuPt/TiO2, the catalyst present-
ing the highest selectivity for 2-KGUA (75%; Figure 4). Recy-
cling experiments were performed by filtering and using the
catalyst in the next run without any further purification. The
catalyst was stable until the fourth run, and then a gradual
loss of activity was observed. Transmission electron microscopy
Catalyst preparation
AuPVA: Solid NaAuCl4·2H2O (0.051 mmol) and PVA (1wt%) solution
(Au/PVA=1:0.5 w/w) were added to H2O (100 mL). After 5 min,
NaBH4 (0.1m) solution (Au/NaBH4 =1:4 mol/mol) was added to the
yellow solution with vigorous magnetic stirring. Within a few mi-
nutes of sol generation, the colloid was immobilized by adding the
support with vigorous stirring. The amount of support was calcu-
lated as having a total final metal loading of 1 wt%. After 2 h, the
slurry was filtered, and the catalyst was washed thoroughly with
distilled water (neutral mother liquors) and dried at T=808C for
4 h.
PtPVA: Solid K2PtCl4 (0.051 mmol) and PVA (1wt%) solution (Pt/
PVA=1:0.5 w/w) were added to H2O (100 mL). After 5 min, H2 was
bubbled through (50 mLminÀ1) under atmospheric pressure at
room temperature for 2 h. The colloid was immobilized by adding
the support with vigorous stirring. The amount of support was cal-
culated as having a total final metal loading of 1 wt%. After 2 h,
the slurry was filtered, and the catalyst was washed thoroughly
with distilled water (neutral mother liquors) and dried at T=808C
for 4 h.
Figure 4. Results of the stability test for AuPt/TiO2.
ChemSusChem 2015, 8, 4189 – 4194
www.chemsuschem.org
4192
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Papers
AuPtPVA: NaAuCl4·2H2O (Au: 0.031 mmol) was dissolved in H2O
(60 mL), and PVA (1wt%) was added (Au/PVA=1:0.5 w/w). The
yellow solution was stirred for 3 min, after which 0.1m NaBH4 (Au/
NaBH4 =1:4 mol/mol) was added with vigorous magnetic stirring.
Within a few minutes of sol generation, the gold sol was immobi-
lized by adding the support with vigorous stirring. The amount of
support was calculated as having a gold loading of 0.60 wt%. After
2 h, the slurry was filtered, and the catalyst was washed thoroughly
with distilled water (neutral mother liquors). The Au/support was
dispersed in H2O (40 mL), with K2PtCl4 (Pt: 0.021 mmol) and PVA so-
lution (Pt/PVA=1:0.5 w/w) added. H2 was bubbled through
(50 mLminÀ1) under atmospheric pressure at room temperature for
2 h. The slurry was filtered, and the catalyst was washed thorough-
ly with distilled water (neutral mother liquors) and dried at T=
808C for 4 h. The total metal loading was 1 wt%.
Samples for TEM characterization were prepared by depositing an
ethanol suspension of the catalyst onto lacey-carbon-coated
300 mesh copper grids. The particle morphology of the supported
nanoparticles, specifically the nanoparticle size, was first investigat-
ed by TEM with a Philips LaB6 electron microscope operated at
200 kV and equipped with a Gatan charge-coupled-device camera.
Detailed high-resolution HAADF STEM imaging and XEDS analyses
were performed with an FEI Titan3 microscope operated at 200 kV
accelerating voltage for a deeper investigation of the AuPt-alloy
structure of the AuPt/TiO2 catalyst. This microscope was equipped
with double aberration correctors, to provide ultra-high-resolution
HAADF STEM images, and an Oxford Inca energy-dispersive X-ray
spectrometer equipped with a 30 mm2 ultrathin window Si/Li X-ray
detector. XEDS data were collected either as spectrum images, in
which a focused electron probe was scanned across a region of in-
terest during data collection, or in stationary spot mode, in which
an emitted X-ray spectrum from 0–20 kV was acquired from a spe-
cific point on a particle with a probe size less than 0.5 nm. Spectra
were acquired with a probe current of approximately 0.5 nA and
dwell times of between 200 and 400 ms per pixel in the case of
maps and 20–30 seconds per analysis in spot mode. STEM digital
images were acquired by using the FEI TIA software, and the
Oxford Inca microanalysis software was used for XEDS acquisition
and analysis. The atomic fractions of gold and platinum were
quantified by the Cliff–Lorimer method on relative intensities of
the Pt-La and Au-La peaks with k-factors provided by the XEDS
system manufacturer. The AuPt particle size distribution and the
total metal dispersion were determined by counting 250 particles
in the HAADF STEM images with GAUSS software.
Oxidation procedure
Reactions were performed in a thermostated glass reactor (50 mL)
provided with an electronically controlled magnetic stirrer connect-
ed to a large reservoir (5000 mL) containing oxygen at 3 bar. The
oxygen uptake was followed by a mass flow controller connected
to a computer through an analog-to-digital board, which plotted
a flow–time diagram. The reactor was specially designed to mea-
sure the pH value under pressure. The pH value was automatically
controlled by a Metrohm Titrino 718 apparatus and maintained by
adding NaOH (0.5m). Sorbose (0.3m) and the catalyst (metal/sor-
bose=1:250 mol/mol) were mixed in distilled water (total volume
20 mL). NaOH (0.5m) was added until the desired pH value (7.5)
was reached. The reactor was pressurized at 2 bar of N2 and ther-
mostated at the appropriate temperature. Once the required tem-
perature (508C) was reached, the gas supply was switched to
oxygen and the monitoring of the reaction was started.
The acid/base properties of the supports were measured with
a Metrohm Titrino 718 instrument. Typically, sample (100 mg) was
dispersed in a 10À3 m KCl solution (50 mL). The mixture was kept
overnight at room temperature with vigorous stirring and the pH
value of the solution was measured.
Recycling tests were performed under the same experimental con-
ditions. The catalyst was recycled in the subsequent run after filtra-
tion without any further treatment. Alternatively, the catalyst was
regenerated by washing with distilled water before reusing it in
the successive run. The recovery of the catalyst was always >98%.
Samples were removed periodically and analyzed by high-per-
formance liquid chromatography with an Alltech OA-10308 column
(300”7.8 mm) with UV (set at l=210 nm) and refractive index de-
tection to analyze the mixture of the samples. An aqueous H3PO4
solution (0.1 wt%) was used as the eluent. Products were identified
by comparison with the original samples.
Nitrogen adsorption isotherms were measured at T=À1968C with
a TriStar 3000 volumetric adsorption analyzer manufactured by Mi-
cromeritics Instrument Corp. (Norcross, GA). Before adsorption
measurements, the carbon powders were degassed in flowing ni-
trogen for 1–2 h at T=2008C. The specific surface area of the sam-
ples was calculated with the Brunauer–Emmett–Teller method
within the relative pressure range of 0.05–0.20.
Keywords: carbohydrates · gold · oxidation · platinum ·
supported catalysts
Hydrogen peroxide detection and degradation test
[1] T. Mallat, A. Baiker, Chem. Rev. 2004, 104, 3037–3058.
[2] S. E. Davis, M. S. Ide, R. J. Davis, Green Chem. 2013, 15, 17–45.
[3] M. Sankar, N. Dimitratos, P. J. Miedziak, P. P. Wells, C. J. Kiely, G. J. Hutch-
ings, Chem. Soc. Rev. 2012, 41, 8099–8139.
[4] M. Besson, P. Gallezot, C. Pinel, Chem. Rev. 2014, 114, 1827–1870.
[5] A. Villa, N. Dimitratos, C. E. Chan-Thaw, C. Hammond, L. Prati„ G. J.
Hutchings, Acc. Chem. Res. 2015, 48, 1403–1412.
[6] H. Zhang, N. Toshima, Catal. Sci. Technol. 2013, 3, 268–278.
[7] C. Brçnnimann, Z. Bodnar, P. Hug, T. Mallat, A. Baiker, J. Catal. 1994, 150,
199–211.
[8] C. Brçnnimann, T. Mallat, A. Baiker, J. Chem. Soc. Chem. Commun. 1995,
1377–1378.
A 1 mm solution of H2O2 (15 mL) was stirred at the appropriate
temperature under a N2 atmosphere. The pH value of the solution
was adjusted by adding Na2CO3 (0.5m, pH 7.5). The amount of the
catalyst added was the same as that in the sorbose oxidation
(about 200 mg). Hydrogen peroxide was quantified by sampling by
permanganate titration. The following procedure has been set up:
a sample (5 mL) of the filtered reaction solution was titrated with
a 0.01n KMnO4 solution at a constant pH value of 7.5Æ0.1 (by ad-
dition of concentrated H2SO4). The detection limit of H2O2 was
0.01 mm.
[9] C. Brçnnimann, Z. Bodnar, R. Aeschimann, T. Mallat, A. Baiker, J. Catal.
1996, 161, 720–729.
[10] T. Mallat, C. Brçnnimann, A. Baiker, J. Mol. Catal. A 1997, 117, 425–438.
[11] T. Mallat, C. Brçnnimann, A. Baiker, Appl. Catal. A 1997, 149, 103–112.
[12] S. N. Sidorov, I. V. Volkov, V. A. Davankov, M. P. Tsyurupa, P. M. Valetsky,
L. M. Bronstein, R. Karlinsey, J. W. Zwanziger, V. G. Matveeva, E. M.
Characterization
The metal content and the metal leaching were checked by ICP
analysis of the filtrate, on a Jobin Yvon JY24 instrument.
ChemSusChem 2015, 8, 4189 – 4194
www.chemsuschem.org
4193
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Papers
Sulman, N. V. Lakina, E. A. Wilder, R. J. Spontak, J. Am. Chem. Soc. 2001,
123, 10502–10512.
[18] A. Villa, S. Campisi, K. M. Mohammed, N. Dimitratos, F. Vindigni, M. Man-
zoli, W. Jones, M. Bowker, G. J. Hutchings, L. Prati, Catal. Sci. Technol.
2015, 5, 1126–1132.
[19] N. Dimitratos, A. Villa, D. Wang, F. Porta, D. Su, L. Prati, J. Catal. 2006,
244, 113–121.
[13] L. M. Bronstein, G. Goerigk, M. Kostylev, M. Pink, I. A. Khotina, P. M. Valet-
sky, V. G. Matveeva, E. M. Sulman, M. G. Sulman, A. V. Bykov, N. V. Lakina,
R. J. Spontak, J. Phys. Chem. B 2004, 108, 18234–18242.
[20] L. Prati, A. Villa, Acc. Chem. Res. 2014, 47, 855–863.
[21] W. C. Ketchie, Y. Fang, M. S. Wong, M. Murayama, R. J. Davis, J. Catal.
2007, 250, 94–101.
[22] L. Prati, P. Spontoni, A. Gaiassi, Top. Catal. 2009, 52, 288–296.
[23] M. Sankar, N. Dimitratos, D. W. Knight, A. F. Carley, R. Tiruvalam, C. J.
Kiely, D. Thomas, G. J. Hutchings, ChemSusChem 2009, 2, 1145–1151.
[14] T. Reichstein, A. Gruessner, Helv. Chim. Acta 1934, 17, 311 –328.
[15] E. Grünewald, U. Prüße, K.-D. Vorlop, Catal. Today 2011, 173, 76–80.
[16] A. Villa, G. M. Veith, L. Prati, Angew. Chem. 2010, 122, 4601–4604;
Angew. Chem. Int. Ed. 2010, 49, 4499–4502.
[17] G. L. Brett, Q. He, C. Hammond, P. J. Miedziak, N. Dimitratos, M. Sankar,
A. A. Herzing, M. Conte, J. A. Lopez-Sanchez, C. J. Kiely, D. W. Knight,
S. H. Taylor, G. J. Hutchings, Angew. Chem. 2011, 123, 10318–10321;
Angew. Chem. Int. Ed. 2011, 50, 10136–10139.
Received: September 4, 2015
Published online on November 27, 2015
ChemSusChem 2015, 8, 4189 – 4194
www.chemsuschem.org
4194
ꢀ 2015 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim