Cellulose conversion with tungstated-alumina-based catalysts: Influence of the presence of platinum and mechanistic studies

Article abstract of DOI:10.1002/cssc.201200880

The performances of platinum supported on tungstated alumina (Pt/AlW) in the hydrothermal conversion of cellulose at 190 °C under H₂ pressure were evaluated and compared to that of Pt-free tungstated alumina (AlW). We show that the presence of Pt significantly increased the extent of conversion and led to a different product distribution with the formation of acetol and propylene glycol as the main products and a global yield of up to 40 %. Based on previous reports, we propose the formation of pyruvaldehyde on the Lewis acid sites of the tungstated alumina as a key intermediate. Pyruvaldehyde can then be transformed to acetol and propylene glycol or lactic acid depending on the presence or absence of supported Pt. Copyright

Full text of DOI:10.1002/cssc.201200880

CHEMSUSCHEM
FULL PAPERS
DOI: 10.1002/cssc.201200880
Cellulose Conversion with Tungstated-Alumina-Based
Catalysts: Influence of the Presence of Platinum and
Mechanistic Studies
[
a]
[a]
[a]
[b]
Flora Chambon, Franck Rataboul,* Catherine Pinel, Amandine Cabiac,
[b]
[a]
Emmanuelle Guillon, and Nadine Essayem
The performances of platinum supported on tungstated alumi-
na (Pt/AlW) in the hydrothermal conversion of cellulose at
a global yield of up to 40%. Based on previous reports, we
propose the formation of pyruvaldehyde on the Lewis acid
sites of the tungstated alumina as a key intermediate. Pyruval-
dehyde can then be transformed to acetol and propylene
glycol or lactic acid depending on the presence or absence of
supported Pt.
1908C under H2 pressure were evaluated and compared to
that of Pt-free tungstated alumina (AlW). We show that the
presence of Pt significantly increased the extent of conversion
and led to a different product distribution with the formation
of acetol and propylene glycol as the main products and
Introduction
Cellulose is considered as a promising renewable resource to
Al O , which is able to convert cellulose into sugar alcohols at
2
3
[
1–3]
produce fuels and chemicals.
The chemical valorisation of
1908C under H with a yield of 31%. Since then, various cata-
2
cellulose is a great challenge as this bio-polymer is very resist-
ant to chemical transformations under conventional condi-
tions. The number of studies on the conversion of cellulose
into valuable molecules has grown spectacularly over the past
few years. Different reaction media and conditions have been
reported for the conversion of cellulose, which includes hydrol-
ysis in water by liquid mineral acids or enzymes or alternatively
in non-conventional media such as ionic liquids and super-criti-
lytic systems have been reported. Principally, they are formed
[20]
[21,22]
of C-supported Ru associated with mineral acids, soluble
[23]
or insoluble hetero-poly acids or formed from Ru directly
[
24]
[25]
supported on heteropoly acids. Other systems such as Pt
[26]
or Ni phosphides supported on C and mineral acids associat-
[27]
ed with Ru in zeolites are also known. The use of 2-propanol
[28]
as a hydrogen source has also been reported. These new
systems allow the decrease of the reaction temperature (160–
1908C) while increasing the yields of the sugar alcohols (40–
80%). However, in some cases a cellulose ball-milling pre-treat-
ment was applied to increase the reactivity.
[
4–7]
cal solvents.
The use of heterogeneous catalysts under hydrothermal con-
ditions has been reported recently and is the subject of inten-
[
8–10]
sive research.
Various catalytic systems with specific proper-
In a recent study, we compared the efficiency of solid Lewis
and Brønsted acids on the conversion of cellulose in water. We
showed that the solid Lewis acids tungstated zirconia (ZrW)
and especially tungstated alumina (AlW) are remarkable cata-
lysts that promote cellulose conversion under hydrothermal
ties have been applied to produce different products. The
simple hydrolysis of cellulose in water into glucose has been
reported at mild temperatures with mono-functional solid-acid
[
11,12]
catalysts such as sulfonated C,
sulfonated silica/C nano-
[
13]
composites or with catalytic systems based on Ru supported
conditions to yield lactic acid as the main product with yields
[
14]
[29,30]
on mesoporous C. Alkane diols such as propylene and ethyl-
up to 27% when using AlW.
To the best of our knowledge,
ene glycols can be obtained under H with the use of metal-
this was the first example of high yields of lactic acid obtained
directly from cellulose by using solid catalysts.
2
[
15]
oxide-supported Ni catalysts,
C-supported Ru associated
[
16]
[17,18]
with tungstic acid or tungsten carbide catalysts.
ing work on the conversion of cellulose into sugar alcohols
Pioneer-
Herein, we report the influence of Pt supported on the solid
Lewis acid AlW on cellulose conversion and product distribu-
tion. We also propose a general reaction scheme that leads to
various products depending on the presence or absence of Pt
and hydrogen on the Lewis acid catalyst.
[19]
(sorbitol, mannitol) was reported by Fukuoka and Dhepe.
They evidenced the efficiency of the bi-functional catalyst Pt/g-
[
a] F. Chambon, Dr. F. Rataboul, Dr. C. Pinel, Dr. N. Essayem
Institut de Recherche sur la Catalyse et l’Environnement de Lyon
IRCELYON, Universitꢀ Lyon 1, CNRS, UMR5256
Results and Discussion
2
avenue Albert Einstein, 69626 Villeurbanne (France)
E-mail: franck.rataboul@ircelyon.univ-lyon1.fr
Cellulose conversion
[
b] Dr. A. Cabiac, Dr. E. Guillon
For comparison, in addition to AlW, we studied the performan-
ces of other catalysts with regard to the acidic nature of the
IFPEN-Lyon, BP3
6
9360 Solaize (France)
ꢀ
2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemSusChem 2013, 6, 500 – 507 500

CHEMSUSCHEM
FULL PAPERS
www.chemsuschem.org
Clearly, the presence of a Pt phase on AlW had
a strong influence on the final cellulose conversion
with a more rapid conversion principally after pro-
longed reaction times. This confirms the important
role played by Pt in the cellulose dissolution process
as already reported for Pt/g-Al O , but the explana-
Table 1. Main features of the different catalysts used in this study.
[
a]
[b]
Catalyst
A
BET
2
Composition
[wt%]
d
[nm]
particle
Acid-site density
Lewis-acid-site
proportion [%]
ꢀ1
ꢀ1
[c]
[
m g
]
[mmolg
]
AlW
Pt/AlW
Pt/g-Al
Pt/SiO
294
235
208
297
W: 18.8
Pt: 2.0
Pt: 2.5
Pt: 1.8
–
652
722
425
54
100
100
85
1.1
1.1
4.6
2
3
2 3
O
tion of this phenomenon is still a matter of debate.
According to Fukuoka and Dhepe, the acidic function
of alumina promotes cellulose hydrolysis and the H2/
Pt system enhances the solid catalyst acidity through
2
–
[
a] Mean diameter of the metal particles was determined by using TEM. [b] Deter-
mined by NH adsorption monitored by using thermogravimetric analysis (TGA).
c] Determined by performing pyridine adsorption–desorption experiments.
3
[
H splitting on the metal, which leads to a higher
2
support to evidence a possible peculiarity of the Pt/AlW
system. Silica was chosen because of its neutral character, and
g-Al O was included because of its previously reported effi-
2
3
[
19]
ciency as a support for cellulose conversion. The main char-
acteristics of all these catalysts are presented in Table 1. The
surface areas of the different supports evaluated in this reac-
2
ꢀ1
tion are in the same range (200–300 m g ). TEM measure-
ments (not shown here) demonstrate that on silica, the Pt par-
ticle size is approximately 4.6 nm, whereas much smaller parti-
cles were obtained on AlW and g-Al O3 supports (1.1 nm).
2
Solely AlW and Pt/AlW catalysts possess 100% Lewis acid sites
ꢀ
1
with a density of up to 700 mmolg .
The influence of the supports and the supported metallic
catalyst on the hydrothermal conversion of cellulose was stud-
ied to determine the role of the metal on cellulose reactivity
(
Table 2). The reaction was carried out under a H atmosphere
2
Figure 1. Cellulose conversion without (~) and with the catalysts Pt/AlW (*)
and AlW (*) as a function of time.
[
a]
Table 2. Cellulose conversion in the presence of the different catalysts.
[19]
conversion. In one of our previous studies, we re-investigat-
Catalyst
Differential cellulose conversion owed Cellulose conversion
[
b]
ed the catalytic properties of Pt/g-Al O and proposed that fast
2
3
to the metallic phase [pp]
[%]
hydride transfer promoted by the H /Pt system contributed to
2
none
–
31
39
28
41
62
55
70
31
[31]
the acceleration of cellulose conversion. In a previous study,
we observed enhanced cellulose solubilisation when solid
Lewis acids were used. We proposed that Lewis centres might
coordinate the soluble oligomers that issue from the cellulose
hydrolysis initiated by the hot-water medium and that the co-
operation between the Brønsted acidity of the hot water and
SiO
Pt/SiO
g-Al
2
2
ꢀ11
+21
+15
2
O
3
2 3
Pt/g-Al O
AlW
Pt/AlW
none
[29]
the Lewis centres explains this enhancement. All these ex-
planations might still hold true in the presence of Pt/AlW.
A recent report highlighted the importance of the Pt precursor
[a] Cellulose (1.6 g), catalyst (0.68 g), H
2
O (65 mL), PH2 =5 MPa, 1908C,
24 h. [b] pp=percentage points.
on the Pt/g-Al O3 system. It was proposed that the use of
2
for all the materials tested. The results show that the conver-
sions significantly increased in the presence of Pt on the solids,
except for the neutral silica support. With g-Al O and AlW, the
H PtCl could shift the dissociation equilibrium of water to en-
2
6
hance the hydrothermal conversion of cellulose by increasing
[32]
the acidity of the aqueous reaction medium.
Iglesia et al.
2
3
cellulose conversion increased by 21 and 15 percentage
points, respectively. The highest cellulose conversion was ob-
tained in the presence of Pt/AlW (70%).
proposed that if Pt was supported on ZrW (Pt/ZrW), the hydro-
gen activated by the supported Pt participates in the forma-
[33,34]
tion of protonic sites on the surface of the material.
As it is
Cellulose conversions over time for AlW, Pt/AlW and without
catalyst are presented in Figure 1, and similar profiles are seen
in these three cases. Nevertheless, the conversion increase was
slightly faster with AlW and even more with Pt/AlW to give rise
to a considerably higher conversion after 30 h of reaction.
a similar catalyst, this may be also true for Pt/AlW, which partly
explains the faster and higher cellulose conversion obtained
by increasing the proton density in the reaction medium.
In the case of SiO and Pt/SiO , the conversions were in the
2
2
same range as those observed without a catalyst. This can be
explained by the very low acidity of the support for which the
ꢀ
2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemSusChem 2013, 6, 500 – 507 501

CHEMSUSCHEM
FULL PAPERS
www.chemsuschem.org
above propositions on the presence of Pt should not be valid,
which leads to no value-added catalyst.
fluence of the amount of supported W species on the selectivi-
[35]
ties was highlighted.
Finally, the global yield of the other products (glucose, or-
ganic acids, un-identified, etc) is always in the 20–25% range
independent of the catalytic system. Although their global se-
lectivity is predominant with SiO and g-Al O , the presence of
Product selectivity
2
2
3
The product yields obtained with each system under a H at-
2
the solid AlW decreased this selectivity in favour of lactic acid
or acetol and propylene glycol. Therefore, the presence of
pure Lewis acid AlW leads to more selective transformations.
mosphere are presented in Figure 2. In some cases, the results
obtained are as expected and in agreement with that already
[29]
This tendency was already reported with the support alone,
and we demonstrate in this paper that it is also true in the
presence of supported Pt.
Mechanistic studies
If cellulose was treated under H in the presence of the Lewis
2
acid AlW, the formation of lactic acid was predominant, where-
as the additional presence of Pt led to a mixture of acetol and
propylene glycol. Here, we aim to propose a mechanistic view
to explain both the higher extent of cellulose dissolution in
the presence of AlW and the formation of the main products
depending on the conditions (Scheme 1).
First, the solid Lewis acid (L) suspended in water can form
Figure 2. Product yields [black: sorbitol; white: propylene glycol; dark grey;
acetol; shaded from grey to white: lactic acid; light grey; others (glucose,
levulinic acid, formic acid, acetic acid, unidentified)] of the hydrothermal cel-
ꢀ
supported negatively charged hydroxide species (LꢀOH ). This
+
also forms protons, which increases the H density and could
lulose conversion with the various catalysts. Conditions: 1908C, 5 MPa H
4 h.
2
,
explain partly the conversion increase to produce glucose
through the hydrolysis of the polysaccharides solubilised by
2
[29,31]
water auto-protolysis.
Ellis and Wilson demonstrated that
reported. For example, in the absence of Pt, AlW had a com-
the reaction of glucose in aqueous NaOH formed pyruvalde-
hyde through cleavage and dehydration as the main steps;
pletely different behaviour to SiO2 and g-Al O . Indeed, al-
2
3
ꢀ
[36]
though SiO and g-Al O formed lactic acid and acetol in simi-
both reactions were catalysed by the OH species. If we con-
2
2
3
ꢀ
lar yields, with AlW, acetol was absent and lactic acid was pre-
sider in our case that the surface species LꢀOH also possesses
[
29]
dominant.
basic sites, we can suppose this transformation to be possible,
which releases pyruvaldehyde into the medium. Therefore, the
Next, the presence of a metallic phase on the supports influ-
enced the distribution of the products and this different be-
haviour depends dramatically on the nature of the support.
ꢀ
LꢀOH species would also be an active site of this overall
transformation. Pyruvaldehyde has also been identified as an
intermediate product in a recent report on glucose reactivity
With Pt/SiO , we can see that the yield of lactic acid decreased
2
[37]
significantly, whereas that of acetol remained unchanged. With
in sub-critical water. Here, we can explain again the impor-
Pt/g-Al O , the yield of acetol and lactic acid remained similar;
tance of the support: if Pt/g-Al O was used, these types of
2
3
2
3
however, sorbitol produced by the hydrogenation of glucose
sites probably would not have been formed (or in a lower
amount) and glucose would not have been transformed into
sorbitol by fast hydrogenation (see above); this transformation
was not predominant with the AlW support.
[
19]
emerged as the main product (14%). Other C products (pro-
3
pylene glycol) were also observed. The main difference is that,
in the presence of Pt/AlW, acetol (28%) and propylene glycol
(
20%) were formed predominantly at the expense of lactic
Then, we expect the formed pyruvaldehyde to be very reac-
tive under our conditions, and it would be subject to various
transformations depending on the catalyst. In the absence of
Pt, it will finally yield lactic acid through coordination of the
carbonyl groups on the Lewis sites of the catalyst similar to
acid, which was obtained in a very low yield. Importantly, sor-
bitol was also obtained but in a low amount, which confirms
the significantly different behaviour of g-Al O and AlW.
2
3
The difference in selectivities obtained for these products
showed the importance of the nature of the acid sites. If we
consider the acidic features of Pt-based catalysts (Table 1), it
appears that the cellulose conversion into acetol and propyl-
ene glycol is enhanced both by a high number of acid sites
and Lewis acidity. As this paper was in preparation, the per-
formances of catalytic systems composed of AlW and Ru/C for
the formation of propylene glycol along with ethylene glycol
[38]
that proposed by Vogel et al. In the presence of supported
Pt, pyruvaldehyde leads to acetol through the hydrogenation
of the aldehyde function and acetol is transformed into pro-
pylene glycol by the reduction of the remaining carbonyl
function.
These hypotheses were confirmed by separate experiments.
Pyruvaldehyde was treated under H with AlW (1008C, 2 h) or
2
and sorbitol under a H atmosphere were reported, and the in-
Pt/AlW (1908C, 8 h). In the first case, lactic acid was obtained
2
selectively with 53% yield and 66% conversion. In the second
ꢀ
2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemSusChem 2013, 6, 500 – 507 502

CHEMSUSCHEM
FULL PAPERS
www.chemsuschem.org
2
Scheme 1. Proposed mechanism for the formation of lactic acid, acetol and propylene glycol from cellulose with AlW or Pt/AlW catalysts under H .
case, propylene glycol was obtained with 80% yield along
with lactic acid (4% yield) for full conversion. Moreover, it
seems that these transformations are faster when pyruvalde-
hyde is used as the reactant than if cellulose is present
Figure 3 represents the evolution of the selectivity as a func-
tion of the cellulose conversion for AlW and Pt/AlW.
The final main products were formed, even at low conver-
sions, during the first minutes of the reaction. This would
ꢀ
2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemSusChem 2013, 6, 500 – 507 503

CHEMSUSCHEM
FULL PAPERS
www.chemsuschem.org
absent at the lowest conversions. These data indicate clearly
that propylene glycol was formed from acetol by the H /Pt/
2
AlW system. This was confirmed independently by treating
acetol with Pt/AlW under 5 MPa of H for 24 h, which resulted
2
in complete conversion with full selectivity into propylene
glycol.
To summarise (Scheme 2), it appears that from the glucose
intermediate, in the presence of AlW alone, route 1 is favoured
with the largely predominant formation of lactic acid from the
proposed key intermediate pyruvaldehyde. In the presence of
Pt/AlW, this route is less favoured and the selectivity shifts to-
wards the faster formation of acetol and propylene glycol
(
route 2). The selectivity balance between acetol and propyl-
ene glycol could be tuned by adapting the reaction time, and
we will probably obtain a much higher selectivity into propyl-
ene glycol after prolonged reaction times. Route 3, which
yields sorbitol from glucose, is also observed but is not
a major route. Certainly the cleavage of glucose by proton
transfer is faster than its hydrogenation. Route 4, which gives
rise to propylene glycol from lactic acid, is less probable as the
reduction of carboxylic acids requires different conditions than
those used here. We confirmed this by performing a separate
experiment, which showed that no propylene glycol was ob-
served if an aqueous lactic acid solution was treated in the
presence of Pt/AlW under H at 1908C.
2
Figure 3. Evolution of the selectivities (* lactic acid;
&
acetol; ~ propylene
glycol) as a function of cellulose conversion with AlW (top) and Pt/AlW
bottom).
(
agree with the high disappear-
ance rate of the first intermedi-
ates (glucose, pyruvaldehyde).
In the presence of AlW, the se-
lectivity in lactic acid was quite
low at the beginning of the reac-
tion but increased rapidly and
continuously, which showed that
lactic acid accumulated with re-
action time and was an end-
product for the transformation
of cellulose with AlW. Acetol was
also detected and its selectivity
was constantly low, which indi-
cates that its formation was
never dominant here.
In the presence of Pt/AlW, we
can see that the formation of
acetol was predominant at low
conversion and attained its max-
imum selectivity after 4 h (corre-
sponding to 20% conversion).
A decrease of the selectivity of
acetol then occurred along with
an increase in the formation rate
of propylene glycol, which was Scheme 2. Possible transformations of glucose with AlW or Pt/AlW catalysts.
ꢀ
2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemSusChem 2013, 6, 500 – 507 504

CHEMSUSCHEM
FULL PAPERS
www.chemsuschem.org
Catalyst stability and recycling
We studied the stability of the catalysts first by determination
the extent of leaching (Table 3) through elemental analysis of
the recovered solutions. Pt/AlW and Pt/g-Al O are quite stable,
2
3
Table 3. Stability of the different catalysts.
[
a]
Catalyst
Metal dissolution [%]
Pt
other
Si: 6
Pt/SiO
Pt/g-Al
Pt/AlW
2
0.25
<0.05
<0.05
2
O
3
W: 0.8
[a] In reference to the initial amount of metal in the catalyst.
Figure 5. Recycling experiment with Pt/AlW (black: conversion; white: acetol
yield; grey: propylene glycol yield). Conditions: cellulose (1.6 g), catalyst
with leaching of less than 0.05% of initial Pt and 0.8% of initial
W. The participation of these solubilised species on cellulose
conversion is negligible. Indeed, if a solution recovered after
the first reaction was treated in the presence of fresh cellulose,
a conversion similar to that obtained in the absence of any cat-
alyst was obtained.
(
2
0.65 g), water (65 mL), 5 MPa H , 1908C, 24 h.
to reduce the metallic sites. A catalytic test with fresh cellulose
was performed in the presence of this solid phase. We ob-
tained a slightly lower conversion with the used catalyst,
which can probably be explained by the weak sintering of the
metallic catalyst after the first run and/or by the presence of
the residual initial cellulose, which could prevent the full recy-
clability of the catalyst. However, the respective yields of acetol
and propylene glycol were maintained, which shows the po-
tential recyclability of the Pt/AlW catalyst.
Pt/SiO emerges as an unstable catalyst under hydrothermal
2
conditions. This probably explains its poor activity for cellulose
transformation, which is similar to that obtained without any
catalyst (Table 2, Figure 2).
After reaction, the catalysts were collected by filtration and
washed extensively before analysis. The XRD pattern of used
Pt/AlW was similar to that of the fresh material, except for the
presence of residual adsorbed cellulose. The signals attributed
to the support are not altered and no signal that corresponds
to Pt is observed, which indicates that the particles were still
well dispersed (Figure 4). This also indicates that re-precipita-
tion of leached AlW should be marginal.
Conclusions
In this study, we compared the reactivity of AlW and Pt/AlW
for the conversion of cellulose under hydrothermal conditions
under H pressure. We showed that the presence of Pt on AlW
2
TEM measurements (not shown here) indicated that the
mean particle size of Pt increased from 1.1 to 2.1 nm after re-
action.
significantly increases the conversion and changes the product
distribution of the reaction. Based on literature data, we pro-
pose that the presence of Lewis acid sites in the water
medium would lead to pyruvaldehyde as key intermediate,
which would react differently depending on the catalyst. With
AlW, pyruvaldehyde would be mainly transformed into lactic
acid, whereas the presence of supported Pt on AlW would lead
to acetol and propylene glycol through hydrogenation of pyru-
valdehyde by the metallic phase. Pt/AlW has been shown to
be an efficient and recyclable catalyst for the formation of
Finally, we performed a recycling experiment with Pt/AlW
(Figure 5). To minimise catalyst modification before recycling,
we treated the recovered solid, which consisted of the catalyst
and the un-reacted cellulose, at 3008C for 2 h under a H flow
2
these two important C molecules, and an interesting global
3
yield of 40% was obtained, which corresponds to a 60%
global selectivity after 24 h.
Experimental Section
Materials
Microcrystalline cellulose purchased from Sigma–Aldrich was used
as received (estimated polymerisation degree 250, crystallinity
index 70%, and particle size 20 mm, estimated 4 wt% water).
[29]
[31]
AlW
and Pt/g-Al O3
were prepared as reported previously.
2
Figure 4. XRD spectra of Pt/AlW (a) before and (b) after reaction.
Pt/SiO and Pt/AlW were prepared by incipient wetness impregna-
2
ꢀ
2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemSusChem 2013, 6, 500 – 507 505

CHEMSUSCHEM
FULL PAPERS
www.chemsuschem.org
2
ꢀ1
tion of SiO (Alfa Aesar, 300 m g ) by Pt(NH ) Cl , and of AlW by
TEM and XRD measurements were performed by using a JEOL
2010LaB6 microscope and a Bruker D8 Advance A25 diffractometer,
respectively.
2
3 4
2
H PtCl (8 wt% in H O). Calcination (air, 5008C, 2 h) and reduction
H , 3008C, 2 h) gave rise to catalysts with 2 wt% Pt.
2
6
2
(
2
Cellulose transformation
Acknowledgements
The reactions were performed in a 100 mL Parr Hastelloy autoclave
equipped with a Rushton turbide. The autoclave was filled with
cellulose (1.6 g), catalyst (0.68 g) and de-ionised water (65 mL).
The authors thank IFPEN, the CNRS and the University Lyon 1 for
funding and particularly the analytical services of IRCELYON-
CNRS.
Pt-containing catalysts were reduced ex situ (H , 3008C, 2 h). The
2
autoclave was flushed three times with H at room temperature
2
ꢀ1
and then heated to 1908C (58Cmin ). The pressure was adjusted
to 5 MPa using H2, and these conditions were maintained for 24 h.
The reaction was stopped by cooling with an ice bath. The pressur-
ised gas was evacuated, and the reaction mixture filtered for sepa-
rate analyses of the liquid and the residual solid.
Keywords: heterogeneous catalysis · hydrothermal synthesis ·
lewis acids · platinum · tungsten
[
[
[
1] G. W. Huber, S. Iborra, A. Corma, Chem. Rev. 2006, 106, 4044–4098.
2] R. Rinaldi, F. Schꢁth, ChemSusChem 2010, 2, 1096–1107.
3] A. M. Ruppert, K. Weinberg, R. Palkovits, Angew. Chem. 2012, 124,
The monitoring of the conversion over short reaction times (3–
1
0 h) was performed according to the following procedure. The au-
toclave was equipped with a 20 mL stainless steel dropping funnel,
which can be pressurised. The autoclave was filled with the cata-
lyst (0.68 g) and de-ionised water (50 mL). A suspension of cellu-
lose (1.6 g) in de-ionised water (15 mL) was introduced into the
dropping funnel connected to the autoclave. The autoclave was
flushed three times with H2 and heated to 1908C. The cellulose
suspension was introduced into the hot autoclave, and the pres-
2
614–2654; Angew. Chem. Int. Ed. 2012, 51, 2564–2601.
[4] S. Van De Vyver, J. Geboers, P. A. Jacobs, B. F. Sels, ChemCatchem 2011,
3, 82–94.
5] I. A. Ignatyev, P. G. N. Mertens, C. Van Doorslaer, K. Binnemans, D. E.
de Vos, Green Chem. 2010, 12, 1790–1795.
6] A. Cabiac, E. Guillon, F. Chambon, C. Pinel, F. Rataboul, N. Essayem,
Appl. Catal. A 2011, 402, 1–10.
7] F. Rataboul, N. Essayem, Ind. Eng. Chem. Res. 2011, 50, 799–805.
8] S. Van de Vyver, J. Geboers, L. Peng, F. de Clippel, M. Dusselier, T. Vosch,
L. Zhang, G. Van Tendeloo, C. J. Gommes, B. Goderis, P. A. Jacobs, B. F.
Sels, WIT Trans. Ecol. Environ. 2011, 154, 129–140.
[
[
[
[
sure was adjusted to 5 MPa of H . This experimental set-up pre-
2
vents possible cellulose reactivity during the heating period.
Aliquots of the reaction mixture were collected over time.
For catalyst leaching and recycling studies, the solid phase that
consisted of the catalyst and the un-reacted cellulose was collected
by filtration and washed with water. The amounts of metallic spe-
cies in the liquid phase were determined by elemental analysis.
[9] P. L. Dhepe, A. Fukuoka, ChemSusChem 2008, 1, 969–975.
[10] H. Kobayashi, T. Komanoya, S. K. Guha, K. Hara, A. Fukuoka, Appl. Catal.
A 2011, 409–410, 13–20.
[
[
11] A. Onda, T. Ochi, K. Yanagisawa, Green Chem. 2008, 10, 1033–1037.
12] S. Suganuma, K. Nakajima, M. Kitano, D. Yamaguchi, H. Kato, S. Hayashi,
M. Hara, J. Am. Chem. Soc. 2008, 130, 12787–12793.
A treatment at 3008C for 2 h under H flow was applied to the re-
2
covered solid before re-use.
[
13] S. Van de Vyver, L. Peng, J. Geboers, H. Schepers, F. de Clippel, C. J.
Gommes, B. Goderis, P. A. Jacobs, B. F. Sels, Green Chem. 2010, 12,
1
560–1563.
[14] H. Kobayashi, T. Komanoya, K. Hara, A. Fukuoka, ChemSusChem 2010, 3,
40–443.
Analytical methods
4
The textural and acidity features of the solids were determined as
previously reported.
[
15] X. Wang, L. Meng, F. Wu, Y. Jiang, L. Wang, X. Mu, Green Chem. 2012,
14, 758–765.
[29]
The liquid phase was analysed by using an Agilent HPLC system
with an RI detector (ICE COREGEL 107H column, 0.00 L% H SO ,
[16] Z. Tai, J. Zhang, A. Wang, M. Zheng, T. Zhang, Chem. Commun. 2012, 48,
052–7054.
[17] N. Ji, T. Zhang, M. Zheng, A. Wang, H. Wang, X. Wang, J. G. Chen,
7
2
4
ꢀ1
0
.5 mLmin , 808C). The liquid phase was also analysed by using
Angew. Chem. 2008, 120, 8638–8641; Angew. Chem. Int. Ed. 2008, 47,
a Shimadzu TOC-VSCH total organic carbon (TOC) analyser (7208C,
8510–8513.
Pt/Al O catalyst, IR detector).
2
3
[
[
18] Y. Zhang, A. Wang, T. Zhang, Chem. Commun. 2010, 46, 862–864.
19] A. Fukuoka, P. L. Dhepe, Angew. Chem. 2006, 118, 5285–5287; Angew.
Chem. Int. Ed. 2006, 45, 5161–5163.
The cellulose solubilisation percentage, also defined as the conver-
sion percentage, was calculated as the ratio of the total mass of C
solubilised in the liquid phase obtained from TOC analysis and the
initial mass of C in the charged cellulose:
[
[
20] R. Palkovits, K. Tajvidi, J. Procelewska, R. Rinaldi, A. Ruppert, Green
Chem. 2010, 12, 972–978.
21] J. Geboers, S. Van de Vyver, K. Carpentier, K. de Blochouse, P. Jacobs, B.
Sels, Chem. Commun. 2010, 46, 3577–3579.
mgCliquid phase
conversion ½%ꢁ ¼ solubilisation ½%ꢁ ¼ 100 % ꢂ mgCinitial
[22] R. Palkovits, K. Tajvidi, A. M. Ruppert, J. Procelewska, Chem. Commun.
011, 47, 576–578.
23] J. Geboers, S. Van de Vyver, K. Carpentier, P. Jacobs, B. Sels, Green Chem.
011, 13, 2167–2174.
24] M. Liu, W. Deng, Q. Zhang, Y. Wang, Y. Wang, Chem. Commun. 2011, 47,
717–9719.
cellulose
2
[
[
[
[
[
The C yields of the products detected by HPLC were calculated as
the molar ratio of the product i and the initial glucosyl units pres-
ent in the cellulose, corrected by the number of C atoms:
2
9
25] H. Kobayashi, Y. Ito, T. Komanoya, Y. Hosaka, P. L. Dhepe, K. Kasai, K.
Hara, A. Fukuoka, Green Chem. 2011, 13, 326–333.
26] P. Yang, H. Kobayashi, K. Hara, A. Fukuoka, ChemSusChem 2012, 5, 920–
nCi
ni
yield ½C %ꢁ ¼ 100 % ꢂ 6 ꢂ n
i
glucosyl units
926.
in which nCi is the number of C atoms in product i, n is the
i
27] J. Geboers, S. Van de Vyver, K. Carpentier, P. Jacobs, B. Sels, Chem.
Commun. 2011, 47, 5590–5592.
number of mols of product i determined by HPLC analysis and
nglucosyl units is the initial number of moles of glucosyl units in the cel-
lulose sample (mcellulose/162).
[28] H. Kobayashi, H. Matsuhashi, T. Komanoya, K. Hara, A. Fukuoka, Chem.
Commun. 2011, 47, 2366–2368.
ꢀ
2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemSusChem 2013, 6, 500 – 507 506

CHEMSUSCHEM
FULL PAPERS
www.chemsuschem.org
[
[
[
[
[
[
29] F. Chambon, F. Rataboul, C. Pinel, A. Cabiac, E. Guillon, N. Essayem,
Appl. Catal. B 2011, 105, 171–181.
30] F. Chambon, N. Essayen, F. Rataboul, C. Pinel, A. Cabiac, E. Guillon, PCT
Int. Appl. 2011, WO 2011098683.
31] V. Jollet, F. Chambon, F. Rataboul, A. Cabiac, C. Pinel, E. Guillon, N. Es-
sayem, Green Chem. 2009, 11, 2052–2060.
32] R. M. Ravenelle, F. Z. Diallo, J. C. Crittenden, C. Sievers, ChemCatChem
[35] Y. Liu, C. Luo, H. Liu, Angew. Chem. 2012, 124, 3303–3307; Angew.
Chem. Int. Ed. 2012, 51, 3249–3253.
[36] A. V. Ellis, M. A. Wilson, J. Org. Chem. 2002, 67, 8469–8474.
[37] M. Mçller, P. Nilges, F. Harnisch, U. Schrçder, ChemSusChem 2011, 4,
566–579.
[38] M. Bicker, S. Endres, L. Ott, H. Vogel, J. Mol. Catal. A: Chem. 2005, 239,
151–157.
2
012, 4, 492–494.
33] D. G. Barton, S. L. Soled, G. D. Meitzner, G. A. Fuentes, E. Iglesia, J. Catal.
999, 181, 57–72.
1
34] E. Iglesia, D. G. Barton, J. A. Biscardi, M. J. L. Gines, S. L. Soled, Catal.
Today 1997, 38, 339–360.
Received: November 19, 2012
Published online on February 20, 2013
ꢀ
2013 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemSusChem 2013, 6, 500 – 507 507