Selective aqueous phase oxidation of 5-hydroxymethylfurfural to 2,5-furandicarboxylic acid over Pt/C catalysts: Influence of the base and effect of bismuth promotion

Article abstract of DOI:10.1039/c3gc40727f

5-Hydroxymethylfurfural (HMF) was quantitatively oxidized to 2,5-furandicarboxylic acid (FDCA) at 100 °C under 40 bar air in moderately basic aqueous solution in the presence of active carbon supported platinum and bismuth-platinum catalysts. The transformation of HMF into FDCA proceeded via the 5-hydroxymethyl-2-furancarboxylic acid (HMFCA) and 2,5-diformylfuran (DFF) intermediates; both of these were very reactive and rapidly oxidized to 5-formylfurancarboxylic acid (FFCA), the subsequent oxidation of which was found to be the rate-limiting step. The preparation method of the platinum catalysts influenced the particle size of metallic platinum and modified the surface of the support, therefore determining the activity. The addition of a carbonate base (Na₂CO₃/HMF molar ratio = 2) led to faster overall conversion than bicarbonate (NaHCO₃/HMF = 4) by maintaining an appropriate pH for the oxidation reaction. The ex situ or in situ addition of a bismuth promoter still accelerated the reaction; the highest activity was observed for a Bi/Pt molar ratio of ca. 0.2. Furthermore, the promoter helped to prevent some deactivation of the Pt catalyst upon recycling experiments. Quantitative conversion of HMF (0.1 M) and >99% yield of FDCA were achieved using a molar ratio of HMF to Pt of 100 and Na₂CO₃ as the homogeneous base in less than 2 h.

Full text of DOI:10.1039/c3gc40727f

Green Chemistry
View Article Online
View Journal
PAPER
Selective aqueous phase oxidation of
-hydroxymethylfurfural to 2,5-furandicarboxylic
5
Cite this: DOI: 10.1039/c3gc40727f
acid over Pt/C catalysts: influence of the base and
effect of bismuth promotion†
Hicham Ait Rass, Nadine Essayem and Michèle Besson*
5
1
-Hydroxymethylfurfural (HMF) was quantitatively oxidized to 2,5-furandicarboxylic acid (FDCA) at
00 °C under 40 bar air in moderately basic aqueous solution in the presence of active carbon supported
platinum and bismuth–platinum catalysts. The transformation of HMF into FDCA proceeded via the
-hydroxymethyl-2-furancarboxylic acid (HMFCA) and 2,5-diformylfuran (DFF) intermediates; both of these
5
were very reactive and rapidly oxidized to 5-formylfurancarboxylic acid (FFCA), the subsequent oxidation
of which was found to be the rate-limiting step. The preparation method of the platinum catalysts
influenced the particle size of metallic platinum and modified the surface of the support, therefore
2 3
determining the activity. The addition of a carbonate base (Na CO /HMF molar ratio = 2) led to faster
overall conversion than bicarbonate (NaHCO /HMF = 4) by maintaining an appropriate pH for the oxi-
3
dation reaction. The ex situ or in situ addition of a bismuth promoter still accelerated the reaction; the
highest activity was observed for a Bi/Pt molar ratio of ca. 0.2. Furthermore, the promoter helped to
prevent some deactivation of the Pt catalyst upon recycling experiments. Quantitative conversion of HMF
Received 17th April 2013,
Accepted 10th June 2013
DOI: 10.1039/c3gc40727f
(
0.1 M) and >99% yield of FDCA were achieved using a molar ratio of HMF to Pt of 100 and Na
2 3
CO as
www.rsc.org/greenchem
the homogeneous base in less than 2 h.
access to 2,5-furandicarboxylic acid (FDCA), a potential bio-
renewable substitute for petroleum-derived terephthalic acid
Introduction
Currently, there is intensive research on the utilization of used in the manufacture of poly(ethylene terephthalate) (PET)
biomass for the production of chemicals and fuel because of plastics. The FDCA counterpart of PET called poly(ethylene
1
,2
its abundance and renewability. Biomass consists mainly of 2,5-furanoate) (PEF) has been shown to present comparable
8
–10
carbohydrate components stored as sugar polymers (starch, properties.
cellulose, or hemicellulose). A range of potential platform Several routes using air over different catalysts have been
compounds can be obtained, of which 5-hydroxymethyl- explored in different solvents. The use of combined Co acetate/
furfural (HMF) is synthesized by acid-catalyzed dehydration of Mn acetate/Br catalysts in acetic acid at 125 °C under 70 bar air
1
1
C6-sugar monomers (typically fructose rather than glucose). pressure has been patented for the oxidation of HMF deriva-
Extensive efforts have been made to find improved economi- tives such as ethers and esters and applied for the Avantium
1
2
cally acceptable ways of production of HMF, and the conver- YXY process. The oxidative esterification in methanol at
sion of biomass into this platform chemical has been the 130 °C in the presence of 1 wt% Au/TiO catalyst with sodium
2
3
–7
13
subject of recent reviews.
methoxide as promoter under 4 bar O2 or in the presence of
without the pres-
gave >98% yield to the
HMF can be subsequently transformed into a variety of bio- gold supported on nanoparticulate nCeO
2
1
4
based chemicals. The selective oxidation of both functional ence of a base under 10 bar O
2
groups in this molecule is of particular interest, since it gives dimethylfuroate. In particular, catalytic systems that permit
the reaction in an aqueous solvent using molecular oxygen or
air have been investigated, using supported metallic nanopar-
IRCELYON, Institut de recherches sur la catalyse et l’environnement de Lyon,
UMR5256 CNRS-Université Lyon 1, 2 Avenue Albert Einstein, 69626 Villeurbanne
Cedex, France. E-mail: michele.besson@ircelyon.univ-lyon1.fr;
Fax: +33(0)472445399; Tel: +33(0)472445358
ticles, mainly Pt-, Ru- and more recently Au-based catalysts.
In general, Pt, Pd and Au-supported catalysts require the
addition of a homogeneous base in aqueous solution to obtain
high yields in the acid, although some recent studies proposed
aerobic alcohol oxidation (glycerol, HMF) under base free
†
Electronic supplementary information (ESI)
0.1039/c3gc40727f
available. See DOI:
1
This journal is © The Royal Society of Chemistry 2013
Green Chem.

View Article Online
Paper
Green Chemistry
1
5–23
38,39
conditions in water.
The role of the base for Pt catalysts is bismuth or lead, in the case of platinum.
When used
ascribed to the neutralization of the acid functions formed alone, bismuth is inactive for the oxidation reaction; when
which otherwise will strongly adsorb on the metallic surface associated with Pt or Pd, it can significantly enhance the
and substantially deactivate the catalyst. Moreover, the oxi- overall catalytic performances (activity, sometimes selectivity)
dation of an aldehyde is believed to proceed through the and/or the lifetime of the catalyst for the selective oxidation of
4
0–46
geminal diol intermediate; the rate of reversible hydration is alcohols or aldehydes in aqueous solutions.
This promot-
accelerated at elevated pH. For gold catalysts, the hydroxide is ing effect was scarcely investigated in the oxidation of HMF.
2
4
helpful for the first deprotonation reaction.
Using a PtPb catalyst and strongly alkaline conditions, FDCA
In the oxidation of HMF, the nature of the base (hydroxyl or was formed with 99% purity in 2 h in a semi-batch reactor
carbonate bases) and the amount of base that should be with dioxygen continuous flow at ambient temperature
added do not reach a consensus. FDCA was obtained with (2.5 mL min− of O2). Starting directly from fructose was pro-
1
27
almost quantitative yield using a Pt/Al
2 3
O catalyst at 60 °C posed in a membrane reactor where the oxidation reaction of
under 0.2 bar partial oxygen pressure at regulated pH in the HMF occurs in the presence of PtBi/C after the diffusion of
2
5,26
range 8–11 with KOH.
Over a PtPb/C catalyst, Verdeguer HMF formed from fructose dehydration from the water phase
2
7
et al. observed that hydroxyl bases were more effective than in the MIBK phase, or in the presence of a silicone-encapsu-
4
7
carbonate bases. However, using stoichiometric aqueous lated, MIBK swollen PtBi catalyst in water in a batch reactor.
2
8,29
Na CO , Lilga et al.
obtained nearly quantitative yields of However, whatever the approach, the yield in FDCA never
FDCA in the oxidation of 1 wt% HMF over Pt/C and Pt/Al in exceeded 25% at 80 °C under ambient pressure.
a continuous reactor at 100 °C under 7 bar air (LHSV = 4.5 h
2
3
2 3
O
−
1
,
The main objective of the present work was to further inves-
GHSV = 600 h− ). For gold catalysts, a minimum NaOH/HMF tigate Pt catalysts and examine in more detail the influence of
atomic ratio of 4 is generally required. Below this loading the the base (the nature and amount) and the possible promotion
selectivity to FDCA decreased, and in the absence of base, only by bismuth. The reusability of the catalysts was also studied by
negligible yields of FDCA are recovered over Au/TiO2, successive reuses.
1
30–32
Au/nCeO
2
or Au-Cu/TiO
2
.
A comparison of the supported Pt,
Pd and Au catalysts under basic aqueous conditions (2 equi-
valents of NaOH) at room temperature under 7 bar O in a semi-
batch reactor revealed that Pd/C and Pt/C were more selective
2
Experimental section
Catalyst preparation
(71–79%) to FDCA compared with Au/C at total conversion
after 6 h; higher base concentrations and also pressures of The activated carbons used as supports were supplied by CECA
2
−1
oxygen were required for Au, resulting in 80% yield at total and correspond to the trade names L3S (900 m g , pH 6) and
33
2
−1
conversion after 22 h.
3SW (950 m g , pH 10). The supported platinum catalysts
Thus, a high pH achieved with the addition of a base is were prepared either by cationic exchange (Pt /C) of the oxi-
ex
necessary for the oxidation of HMF to FDCA in water. dized support with Pt(NH
However, since the need for a base is a serious limitation, as with an aqueous solution of H
an alternative to the use of homogeneous bases solid bases as
supports for the metal were explored. The reaction was (pre-oxidized with a NaOCl solution) in suspension in 100 mL
claimed to be possibly performed in aqueous media in the of a 1 M ammoniacal solution under N at room temperature.
absence of base over hydrotalcite-supported gold catalysts at After 2 h, an aqueous solution of Pt(NH ) Cl was added drop-
3
)
4
Cl
2
or by impregnation (Ptimp/C)
4
8
2
6
PtCl .
Cationic exchange was performed on active carbon L3S
2
2
1
3
4
2
9
5 °C under atmospheric pressure of O
2
, or on ruthenium wise. After stirring under N
2
overnight, the suspension was fil-
2
2,23
hydroxide Ru(OH)
x
on different supports.
However, pH tered and washed to neutrality. The solid was dried under
−
1
measurement or elemental analysis of the product solution vacuum at 40 °C and reduced in H flow (250 mL min
,
2
indicated that magnesium hydroxides were formed upon dis- heating rate 2 °C min− up to 300 °C, 3 h at 300 °C) before
solution of the highly basic materials by the product acids and purging with Ar and passivation in 1% O /N . In the impreg-
While the base is necessary, one must also take nation method, an aqueous solution of H PtCl containing the
1
2
2
2
3,34
diacid.
into account that high pH can lead to the gradual degradation required amount of Pt was added dropwise under stirring to
2
6
3
0,35,41
of HMF.
more stable in alkaline medium than HMF.
important to limit the degradation reactions by operating at 37 wt% formaldehyde solution was added and then a 30 wt%
moderate pH. KOH solution was added. After stirring under N overnight,
Another concern in selective oxidation is the durability of the suspension was filtered and washed with water. The cata-
The partially oxidized intermediates are much an aqueous suspension of carbon L3S or 3SW. After 5 h of
3
1,33
It is therefore impregnation, the slurry was cooled down in an ice bath and a
2
3
0–32
the catalysts. Monometallic gold stability is usually low
and Pt catalysts suffer from irreversible deactivation problems
lyst was dried under vacuum at 40 °C.
The bimetallic Pt–Bi/C catalysts were prepared via a surface
in selective oxidations with air which are due to overoxidation redox reaction ensuring the deposition of bismuth on plati-
36
of the metal surface. A solution to this problem which has num in the liquid phase from the monometallic catalysts and
been proposed is the addition of another metal, such as an aqueous solution of BiONO using glucose as the reducing
3
3
2,37
41,49
copper in the case of gold,
or a heavy metal, particularly agent and according to a previously reported procedure.
Green Chem.
This journal is © The Royal Society of Chemistry 2013

View Article Online
Green Chemistry
Paper
The monometallic Pt-catalyst was added to an aqueous solu- at regular times via a sample diptube. The samples were fil-
tion of glucose. An appropriate amount of BiONO dissolved tered (using a 0.45 μm PVDF membrane), diluted 15 times and
3
in 1 M HCl was added into the mixed slurry under a nitrogen then analyzed. After the reaction, the autoclave was cooled to
atmosphere and the pH was increased to 9 with NaOH. The room temperature and the catalyst was filtered off.
catalyst was filtered, washed, and dried under vacuum at
4
0 °C.
Analysis of the reaction medium
The composition of the reaction medium was determined by
HPLC equipped with a PDA detector and an R.I. detector in
series. The reaction products were separated on an ICE-Coragel
Catalyst characterization methods
The platinum and bismuth content in the solids was deter-
mined after digestion and dissolution of the solids in H SO
2 4
and aqua regia at 250–300 °C by analyzing the solution using
inductively coupled plasma-optical emission spectroscopy
1
2 4
07H column thermostatted at 65 °C. 10 mM H SO was used
−1
as the mobile phase at a fixed flux of 0.6 mL min . Retention
times and calibration curves for the observed products were
determined by injecting known concentrations of reference
commercial products. The total organic carbon (TOC) in solu-
tion was measured using a TOC analyzer (Shimadzu 5050A)
and the measured value was compared to the molar balance
calculated from HPLC analysis. The pH value was measured
using a MeterLab PHM 240 pH-meter.
(ICP-OES, Activa Jobin-Yvon).
The specific surface area (SBET) was measured by N adsorp-
tion/desorption at liquid nitrogen temperature using a Micro-
meritics 2020 apparatus after outgassing under vacuum at
2
3
50 °C for 3 h.
Powder X-ray diffraction patterns were obtained with a D8
Advance A25 diffractometer (Bruker) using Cu Kα radiation
λ = 0.154184 nm). The crystalline phases were identified by
The possible presence of platinum or bismuth in the reac-
tion mixtures during the catalytic reactions was detected by
ICP-OES.
(
reference to the JCPDS data files. Crystal size values were calcu-
lated where possible using the Scherrer equation from the
width at half maximum intensity.
Platinum dispersion was determined by volumetric CO chemi-
sorptions at 35 °C using a Micromeritics ASAP 2010M
apparatus, after reduction at 300 °C during 2 h and outgassing
at 300–350 °C for 4 h. CO chemisorption tests on catalysts con-
taining Bi alone showed no chemisorptions, and then it was
applied to determine Pt accessibility on the support surface.
Transmission electron microscopy coupled to energy-dis-
persive X-ray analysis (TEM-EDX) of the supported catalysts
was carried out using a JEOL 2010 microscope with an LaB6
source operating at 200 kV. The samples were suspended in
ethanol and agitated ultrasonically, a drop of the suspension
was placed on a copper mesh grid with carbon film coating,
and ethanol was allowed to evaporate. EDX was done using an
Oxford Link ISIS instrument. For the analysis of individual
particles, the electron beam was focused to a diameter close to
the diameter of the particle under investigation.
Results and discussion
Characterization of catalysts
The XRD patterns for the monometallic catalysts prepared on
carbon CL3S and C3SW and some promoted catalysts (Bi/Pt
molar ratio = 0.25 or 0.20, respectively) are presented in Fig. 1.
Representative TEM images are shown in Fig. 2, as well as his-
tograms evaluated on 200 to 450 particles over different
regions of the samples. The results of ICP-OES analysis, BET
analysis, mean sizes determined from TEM, and metallic dis-
persion from CO chemisorption experiments are listed in
Table 1.
Standard procedure for HMF oxidation
HMF of 95% purity was purchased from Interchim. The
aqueous phase reactions were performed in a 300 mL batch
high pressure reactor made of Hastelloy equipped with a mag-
netically driven gas inducing stirrer. In the standard reaction,
2
g HMF (95%) dissolved in 150 mL of the alkaline aqueous
solution containing the desired amount of base expressed as
base/HMF molar ratio were loaded into the reactor with an
appropriate amount of catalyst (HMF/Pt molar ratio = 100).
The initial concentration was chosen low because the low solu-
bility of FDCA in aqueous solution could result in precipi-
tation. The reactor was then sealed and pressurized with 40
bar air at time zero. The mixture was heated to the reaction
temperature of 100 °C (total pressure in the reactor then
attained ca. 51 bar) under gentle stirring (950–1000 rpm). The
Fig. 1 XRD patterns of Pt and PtBi catalysts prepared. (a) 3.3% Ptex/CL3S, (b)
3.6% Ptimp/CL3S, (c) 3.6% Ptimp 0.9% Bi/CL3S, (d) 5.1% Ptimp/C3SW, and (e) 5.1%
progress of the reaction was followed by taking liquid samples Ptimp 1% Bi/C3SW
.
This journal is © The Royal Society of Chemistry 2013
Green Chem.

View Article Online
Paper
Green Chemistry
Fig. 2 TEM images of (a) 3.3% Ptex/CL3S, (b) 3.6% Ptimp/CL3S, (c) 3.6 % Ptimp 0.9 % Bi/CL3S, (d) 5.1% Ptimp/C3SW and (e) 5.1% Ptimp 1% Bi/C3SW, and the corres-
ponding particle size distribution.
Table 1 Results of ICP-OES, SBET, XRD, TEM, and CO chemisorption for the pre- patterns (Fig. 1) centered at ca. 25° and 40° come from the
pared catalysts
diffraction of the active carbon supports. In the 3.3% Ptex/CL3S
catalyst the three main peaks at 39.7°, 46.3° and 68.4° were
detected which were assigned to Pt; a crystallite size of 5 nm
was estimated using the Scherrer equation. The XRD patterns
of all other catalysts prepared showed no significant additional
peaks as compared to the support itself, indicating a high dis-
persion of Pt. No obvious diffraction peaks related to Bi or Bi
oxides were observed because the Bi content is low in the bi-
metallic catalyst and these species may exist as amorphous
phases.
c
S
BET
Crystallite Pt average Dispersion
a b
(m g−1) size (nm) size (nm) (%)
2
Catalyst
3
3
5
3
5
.3% Ptex/CL3S
.6% Ptimp/CL3S
.1% Ptimp/C3SW
.6% PtimpBi 0.9%/CL3S 780
.1% Ptimp 1% Bi/C3SW 935
790
788
930
5
<2
<2
<2
<2
5.9 ± 1.4
1.9 ± 0.7
1.9 ± 0.6
2.1 ± 0.6
2.0 ± 0.5
5
60
38
22
14
a
b
c
Estimated from XRD. Determined by TEM. Determined by CO
chemisorption.
By TEM (Fig. 2 and S1 in ESI†), the three monometallic
samples investigated appeared uniformly covered with Pt
The deposition of platinum and/or bismuth did not modify nanoparticles of a relatively homogeneous size. However, par-
the specific surface area of the solids (Table 1). For all the ticle sizes of platinum varied depending on the synthesis
materials studied, the broad diffraction peaks in the XRD method. The cationic exchange method produced a larger
Green Chem.
This journal is © The Royal Society of Chemistry 2013

View Article Online
Green Chemistry
Paper
average particle size (5.9 ± 1.4 nm) than those of catalysts 5.1% these conditions, HMF suffered from serious degradation from
Ptimp/C3SW and 3.6% Ptimp/CL3S prepared by impregnation the beginning of the reaction and the solution became heavily
which exhibited a similar particle size (<2 nm). A slightly nar- coloured.
rower distribution was observed for 5.1% Ptimp/C3SW. The esti-
The oxidation reactions were therefore conducted under
mated crystalline size of Pt in 3.3% P /C (5 nm) fits moderately basic conditions in a NaHCO3 aqueous solution
ex L3S
3
relatively well with that calculated by TEM (5.9 nm). The CO (NaHCO /HMF molar ratio = 4). A typical reaction result with
chemisorption data confirm a better dispersion of the catalysts the evolution of HMF and products as a function of time is
prepared by impregnation (60% and 38% compared with 5%, shown in Fig. 3.
respectively).
In the presence of 3.3% Ptex/CL3S, HMF underwent relatively
The metal particles in the Bi promoted catalysts are also fast oxidation. The partially oxidized DFF and HMFCA were
homogeneously distributed on the activated carbons. Fig. 2c formed, supporting a reaction route involving competitive oxi-
and 2e show that the deposition procedure of bismuth on the dation of the alcohol and aldehyde groups. However, the
Ptimp/C catalysts preserved the mean diameters which were maximum concentrations of both intermediates were gener-
2
.1 ± 0.6 nm and 2.0 ± 0.5 nm, respectively. EDX spectra of 3.6% ated in small amounts and DFF and HMFCA were very quickly
PtimpBi0.9%/CL3S (Bi/Pt molar ratio 0.25) revealed that the par- further oxidized to FFCA. The subsequent oxidation of FFCA
ticles contained both Pt and Bi with molar ratios between 0.2 was much slower, showing that the rate-controlling step
and 0.25. Owing to the preparation method, it may be expected appears to be the final oxidation of the aldehyde group of
that Bi lies mainly on top of Pt. This was confirmed by the FFCA into acid in FDCA. It has to be noted that FFCA was prac-
decrease of metallic accessibility measured by CO chemisorp- tically converted only once; HMF was nearly completely con-
tions after deposition of bismuth, from 60% and 38% to 22% sumed after 3 h, suggesting a stronger adsorption of the initial
and 14%, respectively (Table 1).
HMF substrate on the catalyst surface. Moreover, in this par-
ticular experiment, conversion was not yet complete after 24 h
Monometallic Pt catalyst optimization in HMF oxidation
The catalysts prepared using the two different synthesis
methods were tested in the batch reactor under standard test
conditions: 100 °C, 40 bar air, 0.1 M HMF solution, catalyst
loading to get an HMF/Pt molar ratio of 100. The oxidation of
HMF involves both the oxidation of an aldehyde and an
alcohol function. Following the evolution of the reaction with
time, different compounds were detected (Scheme 1).
The oxidation can proceed through the partially oxidized
intermediates
5-hydroxymethyl-2-furancarboxylic
acid
(HMFCA) and 2,5-diformylfuran (DFF) which are both sub-
sequently oxidized to 5-formyl-2-furancarboxylic acid (FFCA)
before further oxidation to FDCA. It is generally believed that
the hydroxymethyl group undergoes oxidative dehydrogenation
to form the aldehyde, and that the aldehyde groups are first
hydrated to the geminal diols before oxidative dehydrogena-
3
8,39,50
tion to the acids.
Fig. 3 Evolution of concentrations of HMF and products DFF, HMFCA, FFCA,
Initially, the reaction was performed using NaOH as the
base and a NaOH/HMF molar ratio of 4, in the presence of
and FDCA as HMF was oxidized over 3.3% Ptex/CL3S in the presence of NaHCO
3
.
Reaction conditions: 100 °C, 40 bar air, 0.1 M HMF, HMF to Pt molar ratio 100,
3
.3% Pt /CL3S. This ratio corresponds to twice the stoichio-
NaHCO /HMF = 4. ◆ HMF, △ DFF, □ HMFCA, FFCA, FDCA, + molar balance,
●
3
ex
metric amount needed to neutralize the acid groups. Under × TOC and --- pH.
Scheme 1 Reaction pathway of HMF oxidation to FDCA.
This journal is © The Royal Society of Chemistry 2013
Green Chem.

View Article Online
Paper
Green Chemistry
of reaction. The amount of oxygen introduced with 40 bar air
was estimated and found to be at least four times the amount
of hydrogen adsorbed on the catalyst surface by dehydrogena-
tion reactions, i.e. it was sufficient to ensure that O concen-
2
tration was not limiting for the reaction to go to completion in
this discontinuous batch reactor.
Moreover, one may note that at the beginning of the reac-
tion the molar carbon balance in the liquid phase, either from
HPLC measurements or from TOC analysis, did not match
with the initial concentration introduced into the reactor
∼100 mmol L− ). However, after complete disappearance of
1
(
HMF and DFF, the molar balance was >95% up to the end of
the reaction. This phenomenon was observed in all reactions
described in this work. This defect in the initial balance was
attributed to a strong adsorption of HMF on the catalyst
through the furan ring. Conversely, in neutral aqueous solu-
tion, the relative order of adsorption onto carbon was found to
2
8
increase in the order HMF < DFF < FFCA < FDCA. The pH
during the reaction decreased from 8.5 to 7.3 at the end of the
reaction.
The results in the presence of 3.6% Ptimp/CL3S or 5.1%
Ptimp/C3SW using the same reaction conditions with NaHCO
3
as the base (NaHCO /HMF molar ratio = 4) are shown in Fig. 4.
3
The reaction profiles over the Pt catalysts prepared by
impregnation on both supports are similar to the one over the
previous catalyst prepared by exchange under the applied con-
ditions. However, these two catalysts displayed an improved
activity; over both catalysts, HMF was fully converted after 1 h
and conversion of FFCA formed as a further intermediate was
nearly complete after 12 h of reaction. DFF and HMFCA were
Fig. 4 Evolution of the concentrations of HMF and products DFF, HMFCA,
observed with very low concentrations and the final product of FFCA, and FDCA as a function of time over (a) 3.6% Ptimp/CL3S and (b) 5.1%
Ptimp/C3SW, in the presence of NaHCO
HMFCA, ● FFCA, FDCA, + molar balance, × TOC and --- pH.
3 3
(NaHCO
/HMF = 4). ◆ HMF, △ DFF,
oxidation of FFCA was FDCA with an excellent selectivity. More-
over, it should be mentioned that 3.6% Ptimp/CL3S and 5.1%
Ptimp/C3SW which were prepared on two different active
□
carbons by the same method and presenting quite similar dis- modification of the surface of the support by oxidation with a
persions showed the same reaction rates. The pH value sodium hypochlorite aqueous solution to introduce mainly
decreased from 8.6 to 7.5 and 8 because of the neutralization carboxylic acids which act as anchoring groups for the
of the acidic functions by the base.
exchange. Even after the deposition of platinum and thermal
at 300 °C, the solid still contains polar
To understand the different reaction rates shown by the two reduction in flowing H
2
different preparation methods, it was verified that a thermal oxygenated groups which determine the carbon surface hydro-
treatment at 300 °C under H of 3.6% Ptimp/CL3S (which was philic properties and therefore the catalytic behavior. The
2
initially only reduced in liquid phase at ambient temperature) importance of functional groups in carbonaceous supports in
did not modify the physical characteristics and catalytic per- influencing the activity of metallic nanoparticles deposited on
formances of the Pt catalyst prepared by impregnation. carbon in liquid phase oxidation of polyols has been
4
8,51–55
Indeed, comparison of the reaction profile in Fig. S2 in ESI† mentioned.
In some alcohol oxidation reactions, activity
and that of the initial catalyst (Fig. 4a) showed that the cata- was enhanced with removal of oxygen-containing functions
lysts behaved similarly. The XRD patterns (Fig. S5 in ESI†) from the support over supported Pt or Au catalysts, while in
showed no visible growth of the crystallite size of platinum.
other studies the activity was enhanced by increasing the con-
Regarding the activity of the catalysts, the lower activity of centration of phenolic groups. Since in our case the metal par-
3
C
.3% Pt /C
compared to 3.6% Ptimp/CL3S and 5.1% Ptimp
/
ticle size and the surface properties are both changed, further
If we now consider the reaction pathway observed on the
ex L3S
3SW could then be ascribed to the lower metallic dispersion studies are necessary for distinguishing the effects.
as determined by XRD, TEM and CO chemisorptions (Table 1),
since the smaller particles are generally more active. However, supported Pt catalysts under our reaction conditions, the par-
another factor may play some role, i.e. the presence of oxygen tially oxidized products DFF (alcohol oxidation) and HMFCA
functional groups on 3.3% Ptex/CL3S. It should be noted that (aldehyde oxidation) were produced from the beginning of the
this catalyst was prepared by cationic exchange after reaction. Both intermediates were formed with low maximum
Green Chem.
This journal is © The Royal Society of Chemistry 2013

View Article Online
Green Chemistry
Paper
concentrations, since they were subsequently and quickly con- less than 10% in the presence of 4 eq. of NaHCO
3
. The evolu-
verted to FFCA. The latter intermediate was less reactive and tion of the pH during the reactions clearly showed that with
accumulated in the reaction medium to attain a maximum only two equivalents of base, the value was 5.5 instead of 7.5 at
selectivity of 70–75%. The catalyst method of preparation the end of the reaction after 24 h. This shows that the base is
slightly influenced the respective initial selectivities to HMFCA not only required for neutralization of the acid but also to help
and DFF. The selectivity for DFF was slightly higher than for in the hydration of the aldehyde. If pH is too low, then
HMFCA over Ptex/CL3S indicating a preferential oxidation of the hydration of the aldehyde to the geminal diol is lowered.
alcohol function in the presence of the aldehyde; it was very
A reaction was also carried out in the absence of a base,
low in the presence of the two catalysts prepared by impreg- starting from a less concentrated aqueous solution ([HMF] =
nation. It is generally expected that the aldehyde is more easily 0.02 mol L− , HMF/Pt = 100). The results are shown in Fig. S4
1
3
8,39
oxidized in the presence of a hydroxymethyl group.
oxidation of an aldehyde function proceeds through the reac- rapidly, as the acids were formed and pH became more acidic.
tive geminal diol, the formation of which may be disfavored in The nature of the carbonate base was also investigated,
the case of a conjugation. However, Vinke et al. who carried using 5.1% Ptimp/C3SW. The replacement of 4 eq. of NaHCO by
The in ESI.† As expected, the reaction rate of HMF decreased
2
5
3
out the reaction at 60 °C under atmospheric pressure at a con- 4 eq. of KHCO
trolled pH of 9 over 5% Pt/Al observed a higher initial strating that the cation of the bicarbonate has no effect. On
selectivity to DFF, which they explained by a conjugation of the the other hand, the addition of 2 eq. of K CO (or Na CO , not
3
gave perfectly superimposable profiles, demon-
2 3
O
2
3
2
3
aldehyde group with the aromatic furan ring. Conversely, shown) clearly displayed an improved activity, as shown in
during the reaction carried out over a Pt/C catalyst at room Fig. 5 over 5.1% Ptimp/C3SW
temperature under strongly basic conditions (1.25 M NaOH, The higher performances of the catalyst with the addition
NaOH/HMF ca. 10) under O flow, the main intermediate was of the CO
HMFCA formed by oxidation of the aldehyde via the hemiace- explained by the strength of these bases with pK
.
2
−
−
2
3
base compared to the HCO
3
base can be
of 3.65 and
tal to the acid; the oxidation of the remaining hydroxymethyl 7.65, respectively. The pH remains at a higher value with the
b
5
6
2
7
group in HMFCA occurred in a second step to the diacid.
carbonate base which facilitates the hydration of the aldehyde.
The higher selectivity for HMFCA at high pH values can, in No formation of by-products was observed, in contrast to the
that case, be explained by a higher conversion of the aldehyde preliminary experiments performed with NaOH. It is worth
3
1,33
function to the geminal diol which would undergo oxidative noting that, as mentioned in the literature,
the catalyst
dehydrogenation which is favored under these very alkaline must be present in the reaction medium from the beginning
conditions. In our experiments, the situation is a relatively of the reaction to achieve rapid conversion of HMF to the oxi-
moderate pH (pH varying from 8.5 to 7.5–8.1 in the presence dized intermediates. Indeed, the blank oxidative treatment of
of NaHCO ) and a temperature of 100 °C high enough to acti- a Na CO (2 eq.) aqueous solution of HMF, either in the
3 2 3
vate the primary alcohol group. Finally, FDCA was sub- absence of the platinum catalyst or in the presence of the C3SW
sequently formed in quantitative yields from oxidation of support, yielded 50% HMF degradation within 2 h with no
FFCA.
However, it is noteworthy that the reaction pathway over the
detected formation of the oxidized intermediates.
Analysis of the final reaction mixtures by ICP-OES spec-
Pt catalysts is very different from the one generally observed on trometry showed that a small amount of the platinum metal
gold systems under very basic conditions. The latter involve on the catalyst was leached (ca. 2–3%).
the fast oxidation of HMF via a hemiacetal to HMFCA, while
Influence of the addition of bismuth at different Bi/Pt molar
ratios
DFF is absent, implying that the aldehyde moiety of HMF is
more easily oxidized. HMFCA is usually formed in high
amounts in the reaction medium before its fast subsequent We then considered the promotion of the Pt catalysts by
oxidation to FDCA after all HMF has been consumed. The bismuth. They were prepared with different Bi/Pt ratios varying
FFCA intermediate is generally not detected. This indicates between 0.07 and 0.96. Since the monometallic catalysts 3.6%
that the controlling step over gold is the oxidation of the Ptimp/CL3S and 5.1% Ptimp/C3SW showed identical perform-
2
1,30,31
hydroxymethyl group of HMFCA.
ances, the effect of the Bi was examined independently of the
parent Pt catalyst.
Table 2 compares the catalytic performances of the mono-
metallic catalyst and bimetallic catalysts containing different
Influence of the base (the base/HMF ratio, the nature of the
base)
The reactions using 3.6% Ptimp/CL3S and 5.1% Ptimp/C3SW were Bi/Pt ratios after 1 h and 6 h of reaction using 4 eq. of NaHCO
carried out with a different NaHCO /HMF ratio of 2 (Fig. S3 for for HMF conversion and FFCA and FDCA formation. The reac-
3
3
5
.1% Pt/C3SW in ESI†) instead of 4 (Fig. 4). Over both catalysts, tion in the presence of 5.1% Ptimp 1% Bi/C3SW (Bi/Pt = 0.2) was
the decrease of this ratio to 2 did not significantly affect the remarkably accelerated with respect to the reaction over the
reaction rate of HMF conversion; however, the addition of the monometallic parent catalysts. The total conversion of HMF
base in just the quantity equal to the stoichiometry gave was attained in less than 1 h and total conversion of FFCA was
slightly lower reaction rates for FFCA oxidation. After 12 h of obtained after 6 h compared to more than 12 h without
reaction, 20% of FFCA remained to be converted instead of bismuth promotion (Fig. 4). Similarly, modification of 3.6%
This journal is © The Royal Society of Chemistry 2013
Green Chem.

View Article Online
Paper
Green Chemistry
Ptimp/CL3S by bismuth with a Bi/Pt ratio of 0.25 accelerated concentrations after 1 h of reaction for different Bi/Pt ratios
strongly the reaction rates. However, there is a limit to Bi varying between 0.07 and 0.96.
content, and the results show that increasing the Bi amount to
a Bi/Pt ratio of 0.36 became less favorable; the overall conver- with the bicarbonate base. Increasing the Bi/Pt up to 0.2 dra-
sion to FDCA took more time. matically facilitates the oxidation reaction rates of HMF and of
The influence of the Bi/Pt ratio was also investigated in the FFCA to FDCA and leads to a significant decrease of the time
presence of Na CO (2 equivalents). The results are summar- necessary for obtaining a quantitative yield in FDCA. The reac-
Using the carbonate base, the same trend was detected as
2
3
ized in Table 3 which gives the HMFCA, FFCA and FDCA tion was complete within less than 3 h. However, a further
increase in the Bi/Pt ratios decreased the overall reaction rate.
The volcanic trend can be explained by too high coverages of
the Pt particles with high Bi loadings, which block access to
platinum.
Analysis of the filtered reaction media by ICP-OES after the
reactions showed no detectable leaching of platinum (Pt <
0.5%) in contrast to the monometallic catalysts, and no
bismuth (Bi < 0.5%).
Taken together, these results indicate that promotion by
bismuth is beneficial, and that the optimum Bi/Pt ratio is
around 0.2. An optimum Bi/Pt ratio has been reported in the
literature for oxidation reactions, although the value differed
depending on the oxidation reaction considered between 0.1
4
4
and 0.7. A plausible mechanism, consistent with previous
results in the literature for the oxidation reaction with Bi pro-
motion, could be an interaction between the π electrons of the
furanic ring and bismuth. The presence of oxophilic bismuth
can facilitate the chemical adsorption of the geminal diol
form over the catalyst surface for the dehydrogenation on
platinum.
Table 3 Comparison of the catalytic performances after 1 h of reaction of
bismuth promoted platinum catalysts with varying Bi/Pt molar ratio using
2 3 2 3
Na CO (Na CO /HMF = 2)
a
Yields (%)
Bi/Pt
0
HMFCA
FFCA
FDCA
10
4
2
0
4
35
33
21
17
29
47
51
62
77
83
66
44
0
0
0
0
0
.07
.13
.20
.52
.96
Fig. 5 Influence of the nature of the base (filled symbols: NaHCO
open symbols: K CO /HMF = 2) in the oxidation of HMF over 5.1% Ptimp/C3SW
HMF, ● FFCA, ■ FDCA and ▶ pH.
3
/HMF = 4 or
8
2
3
.
a
⍰
Full conversion of HMF in 1 hour.
Table 2 Comparison of the catalytic performances of bismuth promoted platinum catalysts in the presence of NaHCO
3
(NaHCO
3
/HMF = 4)
Reaction time
1
h
2 h
6 h
[FFCA] (mmol L−1
b
[FFCA] (mmol L−1
[FDCA] (mmol L−1
[FDCA] (mmol L−1
)
Catalyst
Bi/Pt
HMF conv (%)
)
)
)
5
3
5
3
4
.1% Ptimp/C3SW
.6% Ptimp/CL3S
.1% Ptimp 1% Bi/C3SW
.6% Ptimp 0.9% Bi/CL3S
.5% Ptimp 1.65% Bi/C3SW
0
0
0.2
0.25
0.36
99
99
100
100
100
59
53
28
38
34
39
70
60
40
28
23
1
3
16
69
72
98
96
81
a
53
a
Contains also 4 mmol L−1 HMFCA. Molar ratio.
b
Green Chem.
This journal is © The Royal Society of Chemistry 2013

View Article Online
Green Chemistry
Paper
Effect of the nature of the base for bimetallic catalysts
Influence of the addition of dissolved bismuth
As for the monometallic catalysts, the nature of the base and The bimetallic catalysts investigated up to now were prepared
the amount of base were investigated for the bimetallic cata- ex situ via a redox reaction. It has been previously reported
lysts (Fig. 6).
that the addition of a metallic salt of Pb, Sn, Cd, and Bi to a
These results demonstrate again that the nature of the suspension of Pt/C in an alkaline aqueous solution of HMF
cation Na or K has no effect, but that the replacement of (Pb/Pt atomic ratio 2.55) led to a drastic promoting effect. The
4
equivalents of NaHCO3 or KHCO3 by 2 equivalents of a analysis of the solid recovered after the reaction indicated that
stronger weak base such as Na CO or K CO leads to a sig- lead, for instance, was deposited on the solid and that it could
nificant increase of the reaction rate. The time to reach com- be successfully recycled without requiring further addition of
2
3
2
3
2
7,57
plete conversion under the conditions used is now less than the promoter salt.
h, and the selectivity to FDCA was >99%. The addition of 4 catalyst could be prepared in situ by addition of a bismuth
eq. of Na CO still accelerates the reaction which is then soluble salt. A ratio of Bi to Pt of 0.2 was chosen, since the
This prompted us to verify whether the
3
2
3
finished in less than 2 h, without affecting the final yield in above results show that the bimetallic catalysts with this ratio
FDCA. displayed the highest activity.
Fig. 6 Effect of the nature of the base and of the base/HMF ratio, catalyst 5.1% Ptimp 1% Bi/C3SW (Bi/Pt = 0.2): (a) 4 NaHCO
3 3 2 3
, (b) 4 KHCO , (c) 2 Na CO ,
(
d) 2 K CO , (e) 4 Na CO , (f) catalyst prepared in situ (2 Na CO
2
3
2
3
2
3
). ◆ HMF, △ DFF, □ HMFCA, ● FFCA, FDCA, + molar balance, × TOC and --- pH.
This journal is © The Royal Society of Chemistry 2013
Green Chem.

View Article Online
Paper
Green Chemistry
The procedure was as follows: the catalyst 5.1% Ptimp and account for the deactivation of Pt/C, as generally found in the
3
6,38,39
an appropriate amount of Bi(NO ) ·5H O were added to the literature.
In fact, during the reaction under air the
3
3
2
alkaline aqueous solution (2 eq. of Na
After stirring under an inert atmosphere during 1 h, solid Vinke et al.
HMF was added to the suspension. The protocol was then the during the reaction. These results demonstrate that the Pt cata-
standard one. Fig. 6f shows the results. The curves in Fig. 6c, lyst should be reactivated under H before recycling or at least
d and 6f clearly show that this way of preparation of the cata- not dried under oxidation conditions.
2 3
CO ) in the reactor. strong interaction between HMF and platinum suggested by
2
5,26
certainly prevents overoxidation of the metal
2
6
lyst afforded exactly the same results as in the case of the
We also tested a bimetallic catalyst in successive experi-
CO . The results of the
of the used catalyst demonstrated that all bismuth introduced reuse experiments for 5.1% Ptimp 1% Bi/C3SW are shown in
ex situ prepared catalyst. Analysis of the reaction medium and ments in the presence of 2 eq. of Na
2
3
has been deposited on the solid during the test.
Fig. 8. It is worth emphasizing that the catalyst used in a reac-
tion was recovered by filtering off the solution of the previous
run after 8 h of reaction, washing and drying under vacuum at
Recycling of catalysts
40 °C and reused without reactivation under hydrogen. Upon
Catalyst recyclability was investigated by successive reuse of
recycling of the bismuth promoted catalyst, only a slight
decrease of the rate was exhibited after 5 runs. Interestingly,
the deactivation was much less important in comparison to
3
.6% Ptimp/CL3S and 5.1% Ptimp 1% Bi/C3SW. As concerns the
monometallic catalyst, after the first run, the catalyst was
simply recovered by vacuum evacuation, washed with water to
remove products adsorbed on its surface, and dried overnight
in an air oven at 110 °C.
Fig. 7 compares the evolution of the selectivity for the three
successive experiments after 6 h at total conversion of HMF
using the monometallic catalyst in the presence of 4 eq. of
nd
rd
the loss of activity observed between the 2 and 3 runs with
the monometallic catalyst when the latter was not pre-reduced.
The reusability of the catalyst has been improved by the Bi pro-
motion, and this is consistent with the known increased resist-
3
8,39
ance of PtBi catalysts to oxygen poisoning.
Neither
platinum nor bismuth could be found within the solution. In
fact, the small decrease in activity of the re-used bimetallic
catalyst can be ascribed to the oxidative reaction condition
which leads to over-oxidation of the Pt catalyst, but to a much
lower extent than for the non-promoted catalyst. A small loss
of the catalyst can also occur during the workup.
3
NaHCO .
Before the second run, the monometallic catalyst was
freshly reduced at 300 °C under hydrogen, while in the third
run the catalyst was not reactivated under hydrogen. With the
Pt catalyst, the catalyst showed no deactivation upon reuse of
the reduced used catalyst. As observed in the first run, full con-
version of HMF after 1 h and a nearly complete conversion of
FFCA after 12 h were attained. The product distribution
remained unchanged. However, in the second recycling, the
catalyst which has not been reduced before the recycling
exhibited a significant decrease of activity. In fact, total oxi-
dation of HMF was largely delayed, and after 12 h, a significant
amount of FFCA has not yet been converted. As no Pt was ana-
lyzed in the solution, metal leaching is not the cause of the
observed deactivation. Oxygen covering or oxidation of the
platinum surface during the oven-drying in air at 110 °C may
Some samples withdrawn from the reaction medium at
100 °C during the reaction were also randomly analyzed for
possible leaching of the metals under the reaction conditions.
No Pt or Bi could be detected within the limits of the analysis
method, demonstrating the chemical stability of these catalysts.
Effect of temperature and pressure
Finally, to get information on the behavior of the catalysts, oxi-
dation experiments were carried out at different reaction temp-
eratures (80–120 °C). The study was performed for the 5.1%
Ptimp 1% Bi/C3SW showing high activity with 2 eq. of Na CO .
2
3
st
nd
Fig. 7 Successive reuse of 3.6% Ptimp/CL3S. 1 run, 2 run after reduction of
rd
the catalyst, 3 run without reduction. Reaction conditions: 100 °C, 40 bar air,
Fig. 8 Successive reuse of 5.1% Ptimp 1% Bi/C3SW without pretreatment. Reac-
−1
−1
HMF 0.1 mmol L , NaHCO
bars: FDCA; grey bars: FFCA.
3
/HMF/Pt loading molar ratio 4/1/0.01, 6 h. Black
2 3
tion conditions: 100 °C, 40 bar air, HMF 0.1 mmol L , Na CO /HMF/Pt loading
molar ratio 2/1/0.01, 3 h. Black bars: FDCA; grey bars: FFCA.
Green Chem.
This journal is © The Royal Society of Chemistry 2013

View Article Online
Green Chemistry
Paper
As expected, the increase in temperature caused a significant
improvement in the HMF and FFCA conversions. At 120 °C, a
7 R.-J. van Putten, J. C. van der Waal, E. de Jong,
C. B. Rasrendra, H. J. Heeres and J. G. de Vries, Chem. Rev.,
2013, 113, 1499–1597.
>
99% yield in FDCA was achieved after 2 h, without the for-
mation of any by-products. The temperature of 100 °C was
chosen for the study of the pressure (40–60 bar) while main-
8 C. Moreau, M. N. Belgacem and A. Gandini, Top. Catal.,
2004, 27, 11–30.
taining the Na
2
CO
3
/HMF ratio at 2. An increase of air pressure
9 A. Gandini, A. J. D. Silvestre, C. P. Neto, A. F. Sousa and
M. Gomes, J. Polym. Sci., Part A: Polym. Chem., 2009, 47,
295–298.
did not affect the catalytic conversion, remaining practically
constant.
1
0 E. de Jong, M. A. Dam, L. Sipos and G.-J. Gruter, in Bio-
based Monomers, Polymers, and Materials, ed. P. B. Smith
et al., ACS Symp. Series, ACS, Washington, DC, 2012, ch. 1,
vol. 1105, pp. 1–13.
Conclusions
The oxidation of 0.1 M HMF to FDCA in moderately basic
aqueous solutions was studied using active carbon supported
Pt and PtBi catalysts. Observed initial intermediate products
observed were HMFCA and DFF which were rapidly oxidized to
FFCA. The hydration and oxidative dehydrogenation of the
aldehyde function are then the rate-limiting step. A strong
effect of the method of preparation on the catalytic activity was
observed. In addition to the particle size effect, the presence of
surface oxygenated groups was found to be possibly prejudicial
for the catalytic activity. We showed the importance of the
nature of the base. The use of 2 equivalents of carbonate with
respect to HMF displayed an increase of activity compared
with the addition of 4 equivalents of bicarbonate, by maintain-
ing the pH at a value favorable for the oxidation of FFCA.
Bismuth promoted Pt catalysts still displayed superior activity
compared to the monometallic catalysts; the catalyst contain-
ing a Pt/Bi molar ratio of ca. 0.2 (prepared ex situ or in situ)
presented the best catalytic activity which allowed under opti-
mized basic conditions the oxidation of HMF to FDCA in
nearly quantitative yield at 100 °C under 40 bar air in very
short reaction times. Additionally, the Bi-promoted system
seems to improve the ability for reuse and therefore leads to
improved productivity.
1
1
1
1
1
1
1
1
1 W. Partenheimer and V. V. Grushin, Adv. Synth. Catal.,
2001, 343, 102–111.
2 A. J. J. E. Eerhart, A. P. C. Faaij and M. K. Patel, Energy
Environ. Sci., 2012, 5, 6407–6422.
3 E. Taarning, I. S. Nielsen, K. Egeblad, R. Madsen and
C. H. Christensen, ChemSusChem, 2008, 1, 75–78.
4 O. Casanova, S. Iborra and A. Corma, J. Catal., 2009, 265,
109–116.
5 D. Liang, J. Gao, J. Wang, P. Chen, Z. Hou and X. Zheng,
Catal. Commun., 2009, 10, 1586–1590.
6 D. Liang, J. Gao, H. Sun, P. Chen, Z. Hou and X. Zheng,
Appl. Catal., B, 2011, 106, 423–432.
7 A. Villa, G. M. Veith and L. Prati, Angew. Chem., Int. Ed.,
2010, 49, 4499–4502.
8 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
and G. J. Hutchings, Angew. Chem., Int. Ed., 2011, 50,
10136–10139.
1
2
2
2
2
2
2
2
9 D. Liang, J. Gao, J. Wang, P. Chen, Y. Wei and Z. Hou,
Catal. Commun., 2011, 12, 1059–1062.
0 A. Tsuji, K. T. V. Rao, S. Nishimura, A. Takagaki and
K. Ebitani, ChemSusChem, 2011, 4, 542–548.
1 N. K. Gupta, S. Nishimura, A. Takagaki and K. Ebitani,
Green Chem., 2011, 13, 824–827.
2 Y. Y. Gorbanev, S. Kegnaes and A. Riisager, Top. Catal.,
011, 54, 1318–1324.
3 Y. Y. Gorbanev, S. Kegnaes and A. Riisager, Catal. Lett.,
011, 141, 1752–1760.
4 B. N. Zope, D. D. Hibbits, M. Neurock and R. J. Davis,
Science, 2010, 330, 74–78.
5 P. Vinke, H. E. van Dam and H. van Bekkum, Stud. Surf.
Sci. Catal., 1990, 55, 147–157.
6 P. Vinke, W. van den Poel and H. van Bekkum, Stud. Surf.
Sci. Catal., 1991, 59, 385–394.
Acknowledgements
2
This work was supported by FUI (Fonds unique interministér-
iel), FEDER (Fonds européen de développement régional),
Conseil Régional de Picardie, and Conseil Régional de la
Réunion within the PENTOVAL project.
2
References
1
G. W. Huber, S. Iborra and A. Corma, Chem. Rev., 2006,
06, 4044–4098.
1
27 P. Verdeguer, N. Merat and A. Gaset, J. Mol. Catal., 1993,
85, 327–344.
28 M. A. Lilga, R. T. Hallen, J. Hu, J. F. White and M. J. Gray,
US Pat., 2008 103 318, 2008.
29 M. A. Lilga, R. T. Hallen and M. Gray, Top. Catal., 2010, 53,
1264–1269.
30 Y. Y. Gorbanev, S. K. Klitgaard, J. M. Woodley,
C. H. Christensen and A. Riisager, ChemSusChem, 2009, 2,
672–675.
2
3
4
P. Gallezot, Chem. Soc. Rev., 2012, 41, 1538–1558.
X. Tong, Y. Ma and Y. Li, Appl. Catal., A, 2010, 385, 1–13.
A. A. Rosatella, S. P. Simeonov, R. F. M. Frade and
C. A. M. Afonso, Green Chem., 2011, 13, 754–793.
5
6
M. J. Climent, A. Corma and S. Iborra, Green Chem., 2011,
1
3, 520–540.
R. Lopes de Souza, H. Yu, F. Rataboul and N. Essayem,
Challenges, 2012, 3, 212–232.
This journal is © The Royal Society of Chemistry 2013
Green Chem.

View Article Online
Paper
Green Chemistry
3
1 O. Casanova, S. Iborra and A. Corma, ChemSusChem, 2009, 43 H. Bönnemann, W. Brijoux, A. S. Tilling and K. Siepen,
, 1138–1144. Top. Catal., 1997, 4, 217–227.
2 T. Pasini, M. Piccinini, M. Blosi, R. Bonelli, S. Albonetti, 44 M. Wenkin, P. Ruiz, B. Delmon and M. Devillers, J. Mol.
N. Dimitratos, J. A. Lopez-Sanchez, M. Sankar, Q. He, Catal., 2002, 180, 141–159.
C. J. Kiely, G. J. Hutchings and F. Cavani, Green Chem., 45 S. Karski, J. Mol. Catal. A: Chem., 2006, 253, 147–154.
2
3
2
011, 13, 2091–2099.
46 C. Mondelli, J. D. Grunwaldt, D. Ferri and A. Baiker, Phys.
Chem. Chem. Phys., 2010, 12, 5307–5316.
47 M. Kröger, U. Prüsse and K.-D. Vorlop, Top. Catal., 2000,
13, 237–242.
3
3
3
3
3
3 S. E. Davis, L. R. Houk, E. C. Tamargo, A. K. Datye and
R. J. Davis, Catal. Today, 2011, 160, 55–60.
4 B. N. Zope, S. E. Davis and R. J. Davis, Top. Catal., 2012, 55,
2
4–32.
5 K. R. Vuyyuru and P. Strasser, Catal. Today, 2012, 195, 144–
54.
6 M. Besson and P. Gallezot, Catal. Today, 2000, 57, 127–
41.
7 S. Albonetti, T. Pasini, A. Lolli, M. Blosi, M. Piccinini,
48 P. Korovchenko, C. Donzé, P. Gallezot and M. Besson,
Catal. Today, 2007, 121, 13–21.
49 A. B. Crozon, M. Besson and P. Gallezot, New J. Chem.,
1998, 269–273.
50 S. E. Davis, B. N. Zope and R. J. Davis, Green Chem., 2012,
14, 143–147.
1
1
N. Dimitratos, J. A. Lopez-Sanchez, D. J. Morgan, 51 E. G. Rodrigues, M. F. R. Pereira, X. Chen, J. J. Delgado and
A. F. Carley, G. J. Hutchings and F. Cavani, Catal. Today,
012, 195, 120–126.
8 M. Besson and P. Gallezot, in Fine Chemicals through Het-
J. J. M. Orfao, J. Catal., 2011, 281, 119–127.
52 E. G. Rodrigues, J. J. Delgado, X. Chen, M. F. R. Pereira and
J. J. M. Orfeo, Ind. Eng. Chem. Res., 2012, 51, 15884–15894.
2
3
erogeneous Catalysis, ed. R. A. Sheldon and H. van Bekkum, 53 Z. Huang, F. Li, B. Chen, F. Xue, Y. Yuan, G. Chen and
Wiley-VCH, Weinheim, New York, Chichester, Brisbane,
Singapore, Toronto, 2001, pp. 491–518.
9 T. Mallat and A. Baiker, Chem. Rev., 2004, 104, 3037–
G. Yuan, Green Chem., 2011, 13, 3414–3422.
54 L. Prati, A. Villa, A. R. Lupini and G. M. Veith, Phys. Chem.
Chem. Phys., 2012, 14, 2969–2978.
3
4
4
4
3
058.
55 A. Villa, M. Schiavoni and L. Prati, Catal. Sci. Technol.,
2012, 2, 673–682.
56 V. L. Snoeyink and D. Jenkins, Water Chemistry, Wiley and
Sons, New York, Chichester, Brisbane, and Toronto, 1990,
p. 462.
0 C. Brönnimann, Z. Bodnar, P. Hug, T. Mallat and A. Baiker,
J. Catal., 1994, 150, 199.
1 M. Besson, F. Lahmer, P. Gallezot, P. Fuertes and G. Fleche,
J. Catal., 1995, 152, 116–121.
2 M. Wenkin, R. Touillaux, P. Ruiz, B. Delmon and 57 N. Merat, P. Verdeguer, L. Rigal, A. Gaset and M. Delmas,
M. Devillers, Appl. Catal., A, 1996, 148, 181–199. to FURCHIM, FR Patent, 2 669 634, 1990.
Green Chem.
This journal is © The Royal Society of Chemistry 2013