Selectivity enhancement in the aqueous acid-catalyzed conversion of glucose to 5-hydroxymethylfurfural induced by choline chloride

Article abstract of DOI:10.1039/c3gc41495g

In this work we wish to show that choline chloride (ChCl), a cheap and safe quaternary ammonium salt industrially produced at a few thousand tons per year through a 100% atom economy process, is capable of enhancing the selectivity of metal chlorides such as AlCl₃, FeCl₃ and CuCl₂ in the aqueous tandem isomerization/dehydration of glucose to HMF. Under optimized conditions, 70% yield of HMF was obtained in a water/ methylisobutylketone (MIBK) biphasic system which is a competitive yield to those traditionally obtained in imidazolium-based ionic liquids in the presence of hazardous chromium salts. By means of counter experiments, we show that the selectivity enhancement is optimal for a ChCl content of 50 wt% in water. At higher loading of ChCl, strong molecular interaction occurs between ChCl and HMF making the extraction of HMF from the aqueous phase difficult with MIBK, thus enhancing side reactions of HMF with water and hexoses to unwanted products. Interestingly, this process can be transposed to the direct conversion of cellulose to HMF which is an even more challenging reaction. In this case, we show that combination of FeCl₃ with AlCl₃ allowed cellulose to be converted to HMF with 49% yield in a one pot reaction. From the viewpoint of sustainable chemistry, this work shows noticeable advantages such as the use of (1) water as a solvent, (2) ChCl as a cheap and safe additive, (3) cheap and naturally abundant metals (Al, Fe, and Cu) and (4) renewable raw materials.

Full text of DOI:10.1039/c3gc41495g

View Article Online
View Journal
Green Chemistry
Accepted Manuscript
This article can be cited before page numbers have been issued, to do this please use: F. Jerome, F. Liu, M. Audemar, K. Vigier,
D. Cartigny, J. Clacens, A. A.H. Padua, M. Costa Gomes and F. De Campo, Green Chem., 2013, DOI: 10.1039/C3GC41495G.
This is an Accepted Manuscript, which has been through the RSC Publishing peer
review process and has been accepted for publication.
Accepted Manuscripts are published online shortly after acceptance, which is prior
Cutting-edge research for a greener sustainable future
www.rsc.org/greenchem
Volume 12 | Number 9 | September 2010 | Pages 1481–1676
to technical editing, formatting and proof reading. This free service from RSC
Publishing allows authors to make their results available to the community, in
citable form, before publication of the edited article. This Accepted Manuscript will
be replaced by the edited and formatted Advance Article as soon as this is available.
To cite this manuscript please use its permanent Digital Object Identifier (DOI®),
which is identical for all formats of publication.
More information about Accepted Manuscripts can be found in the
Information for Authors.
Please note that technical editing may introduce minor changes to the text and/or
graphics contained in the manuscript submitted by the author(s) which may alter
content, and that the standard Terms & Conditions and the ethical guidelines
that apply to the journal are still applicable. In no event shall the RSC be held
responsible for any errors or omissions in these Accepted Manuscript manuscripts or
any consequences arising from the use of any information contained in them.
ISSN 1463-9262
COMMUNICATION
CRITICAL REVIEW
Luque, Varma and Baruwati
Dumesic et al.
Magnetically seperable organocatalyst Catalytic conversion of biomass
for homocoupling of arylboronic acids to biofuels
1463-9262(2010)12:9;1-U
www.rsc.org/greenchem
Registered Charity Number 207890

Green Chemistry
Page 1 of 9Journal Name
Dynamic Article Links ►
View Article Online
DOI: 10.1039/C3GC41495G
Cite this: DOI: 10.1039/c0xx00000x
www.rsc.org/xxxxxx
ARTICLE TYPE
Selectivity enhancement in the aqueous acid-catalyzed conversion of
glucose to 5-hydroxymethylfurfural induced by choline chloride
Fei Liu,a Maïté Audemar,a Karine De Oliveira Vigier,a Damien Cartigny,b Jean-Marc Clacens,b
Margarida F. Costa Gomes,c Agilio A. H. Pádua,c Floryan De Campob and François Jérôme*a
5
Received (in XXX, XXX) Xth XXXXXXXXX 20XX, Accepted Xth XXXXXXXXX 20XX
DOI: 10.1039/b000000x
In this work we wish to show that choline chloride (ChCl), a cheap and safe quaternary ammonium salt
industrially produced at thousand tons per year through a 100 % atom economy process, is capable of
enhancing the selectivity of metal chlorides such as AlCl3, FeCl3 and CuCl2 in the aqueous tandem
10 isomerization/dehydration of glucose to HMF. Under optimized conditions, 70% yield of HMF was
obtained in a water/methylisobutylketone (MIBK) biphasic system which is a competitive yield to those
traditionally obtained in imidazoliumꢀbased ionic liquids in the presence of hazardous chromium salts. By
means of counter experiments, we show that the selectivity enhancement is optimal for a ChCl content of
50 wt% in water. At higher loading of ChCl, strong molecular interaction occurs between ChCl and HMF
15 making the extraction of HMF from the aqueous phase difficult with MIBK, thus enhancing side
reactions of HMF with water and hexoses to unwanted products. Interestingly, this process can be
transposed to the direct conversion of cellulose to HMF which is an even more challenging reaction. In
this case, we show that combination of FeCl3 with AlCl3 allowed cellulose to be converted to HMF with
49% yield in a one pot reaction. From the view point of sustainable chemistry, this work gathers
20 noticeable advantages such as use of 1) water as a solvent, 2) ChCl as a cheap and safe additive, 3) cheap
and naturally abundant metal (Al, Fe, Cu) and 4) renewable raw materials
drawbacks (enzyme deactivation, large quantity of enzymes and
low activity of enzymes).5 Note that enzymatically produced
fructose can be then dehydrated to HMF in a sequential process
(fermentation followed by acid catalysis).6 In 2007, Zhang and
50 coꢀworkers reported that chromium chloride can efficiently
catalyze the tandem isomerization/dehydration of glucose to
HMF with an unprecedented yield of 70% in ionic liquids.7
Recently, variants of this process have been proposed especially
from cellulose and lignocellulosic biomass. Although from an
55 environmental point of view the sustainability of such process is
debatable, this work is however a benchꢀtop methodology that
allows a better understanding of the different elementary steps
involved in the catalytic isomerization of glucose to fructose.
Choline chloride (ChCl), a cheap, biodegradable and safe
60 quaternary ammonium salt industrially produced at thousand tons
per year through a 100 % atom economy process, is known to be
capable of producing a Deep Eutectic Solvent (DES) when mixed
with metal chlorides or hexoses such as glucose or fructose.8 This
particular property of ChCl suggests that this moiety has the
65 unique ability to stabilize or destabilize hydrogen bond donors
such as hexoses or HMF but also metalꢀderived catalytic species.
In this context, this work aims at identifying fundamental issues
on the way how ChCl may impact the catalytic cycles involved in
the tandem isomerisation/dehydration of glucose to HMF (i.e.
70 role as a ligand, stabilization of HMF, etc…). By comparing data
from the current literature with our experimental results, we wish
Introduction
The development of Chemistry Safety Rules, the proposition of
the twelve principles of green chemistry in 1998 by P. Anastas
25 and J. Warner as well as the price and scarcity of oils are now key
elements in establishing sustainable reactions/processes.1 Within
this framework, use of renewablyꢀsourced raw materials for the
production of innovative products has become a top priority in
chemical industry. In this context, much effort has been paid to
30 the synthesis 5ꢀhydroxymethylfurfural (HMF) as a chemical
platform from which solvents, monomers, fuel additives, among
others can be then produced.2 HMF is produced by an acidꢀ
catalyzed triple dehydration of hexoses. Although a plethora of
works have been dedicated to the acidꢀcatalyzed dehydration of
35 fructose to HMF, the direct use of glucose (directly available
from lignocellulosic biomass) is much less widespread mainly
because it requires tackling the isomerization of glucose to
fructose prior to the dehydration step to HMF. Isomerization of
glucose to fructose can be catalyzed by Lewis acids such as for
40 instance tinꢀcontaining zeolites but yields of fructose remained
however rather low.3 Basic catalysts were also proposed but
hexoses are relatively instable in alkaline conditions affording
fructose in low yield.4 Up to now, industrial processes involve
enzymes which are capable of producing up to 50% yield of
45 fructose from glucose although this process has also clearly some
This journal is © The Royal Society of Chemistry [year]
[journal], [year], [vol], 00–00 | 1

Green Chemistry
Page 2 of 9
View Article Online
DOI: 10.1039/C3GC41495G
Table 1 Acidꢀcatalyzed conversion of glucose to HMF in the presence of
Cu(NTf2)2
to show here that use of ChCl as an additive in water is a great
tool for assisting Lewis acid catalysts in the direct conversion of
glucose to HMF. Under optimized conditions, HMF was
produced with up to 70% yield in water which is a very
competitive yield to those obtained in traditional imidazoliumꢀ
based ionic liquids with chromiumꢀbased catalysts. Beside the
clear economic and environmental advantages stemming from the
use of ChCl (< 2€/Kg, biocompatible) and water, we wish to
demonstrate here that assistance of ChCl also allows using safe
5
Entry ChClꢀH2O/MIBK T (°C) Time
Conv. HMF Selectivity
weight ratio
(min) glucose yield (%)
(%)
(%)a
10 metal chlorides such as AlCl3, FeCl3 or CuCl2.
1
2
3
4
5
6
7
8
9
ꢀ
180
180
180
180
180
180
165
200
180
60
3
ꢀ
<5%
18
31
ꢀ
5ꢀ2/0
2ꢀ2/0
1ꢀ2/0
2ꢀ2/2
2ꢀ2/2
2ꢀ2/2
2ꢀ2/2
2ꢀ2/2c
82
74
90
72
85
70
89
83
22
42
23
64
39
43
33
46
Results and Discussion
20
20
15
30
30
7
Catalytic experiments
20
46
33b
30
In a first set of experiments, glucose was mixed with ChCl
(Glucose/ChCl weight ratio = 1/1, 2/1, 1/5). As described earlier,
15 when the solution was heated at 100°C a transparent and viscous
liquid DES was formed. Next, 3 mol% of Cu(NTf2)2 were added
to the solution and heated first at 180°C under microwave
(250W). The choice for this Lewis acid was motivated by our
recent results collected with metal triflates in the catalytic
20 etherification of glycerol.9 Note that catalyst screening is
described later in the manuscript. Unfortunately, under these
conditions, HMF was only produced as a trace presumably due to
the low reactivity of glucose when engaged in the formation of a
DES as similarly observed in the current literature by Kazlauskas
25 and coꢀworkers in a ChCl/glycerol eutectic mixture (Table 1,
entry 1).10 With the aim of lowering the interaction between
glucose and ChCl, water was progressively added to the reaction
medium. In all experiments, the glucose content was kept
constant to 10 wt%. As water was added, we were pleased to see
30 that formation of HMF occurred and yield of HMF reached 31%
at a ChCl/H2O weight ratio of 1 (i.e. 7.1 M) which correspond to
an HMF selectivity of 42% (Table 1, entry 3). Upon prolonged
reaction time, the yield of HMF slightly dropped due to its side
degradation, and condensation. Note that, under these conditions,
35 levulinic acid was not detected but formation of soluble black
materials soꢀcalled humins was also observed. When the amount
of water was further increased (ChCl/H2O weight ratio of 1/2),
the yield of HMF was also decreased suggesting that an optimal
ChCl/H2O ratio need to be respected (Table 1, entry 4). More
40 information about the role of the ChCl/water ratio is provided
later in the manuscript.
29
38
7
a molar yield; b along with 9% yield of levulinic acid; c from a glucose
60 content of 20 wt%
selectivity of 43% vs 64% at 180°C (Table 1, entry 7). Similarly,
an increase of the reaction temperature from 180°C to 200°C
(Table 1, entry 8) also lowered the HMF selectivity (from 64% to
33%) mainly due to the dominant formation of soluble and
65 insoluble black materials as well as other side products. Hence,
one may conclude that 180°C is the optimal temperature when
working with Cu(NTf2)2 as a catalyst. In a similar trend than
previous works,2 an increase of the glucose content from 10 wt%
to 20 wt% also lowered the selectivity of the reaction and HMF
70 was produced with 38% yield (vs 46% at a glucose content of 10
wt%, Table 1, entries 5, 9).
Although promising yields of HMF have been obtained in the
ChClꢀH2O/MIBK system in the presence of Cu(NTf2)2, we then
screened other Lewis acid catalysts in order to select the most
75 active ones. Otherwise noted, all conditions employed for
catalytic experiments are 10 wt% of glucose, a ChClꢀH2O/MIBK
(ratio of 1ꢀ1/1) biphasic system and 3 mol% of catalyst. Reaction
temperature was fixed to 165°C (vs 180°C with Cu(NTf2)2) in
order to reduce the reaction rate and to better select the most
80 active catalysts. To compare the activity of all catalysts, the
reaction time was first fixed to 15 min in all cases. Results are
summarized in Table 2.
Among tested metal triflates and metal triflimidates, Al is the
most active one affording a glucose conversion of 99% followed
85 by Fe and Cu (Table 2, entries 1ꢀ7). In the case of metal
chlorides, large differences of activity were observed. As
previously described in imidazoliumꢀbased ionic liquids7 or in a
water/NaCl mixture,11 AlCl3, CrCl2 and CrCl3 were the most
active ones (Table 2, entries 8, 10, 11). It is noteworthy that safe
90 metal chlorides such as FeCl3 and CuCl2 also exhibited an
acceptable activity leading to a glucose conversion of 58% and
56%, respectively (Table 2, entries 12, 13). Activity of metal
chlorides can be classified as follow AlCl3, CrCl2, CrCl3>FeCl3,
CuCl2>MgCl2, FeCl2>MnCl2, BiCl3>ZnCl2. The great catalytic
95 activity of AlCl3, FeCl3 and CuCl2 in an aqueous solution of ChCl
is attractive mainly because of their low price and natural
abundance. Hence, we selected these metal chlorides and we tried
to optimize the yield of HMF. As the yield of HMF passes
Next, in an attempt to increase the yield of HMF,
methylisobutylketone (MIBK) was added to the system in order
to “continuously” extract HMF from the catalytic phase and to
45 limit its acidꢀcatalyzed side reaction with water or hexoses to
unwanted products. As expected, when 10 wt% of glucose were
heated at 180°C in a ChClꢀH2O/MIBK (ratio of 1ꢀ1/1) biphasic
system for 15 min, the selectivity to HMF was improved and
HMF was produced with 46% yield (Table 1, entry 5). Note that
50 upon prolonged reaction time, the yield of HMF started to
decrease and, in this case, we clearly observed the formation of
levulinic acid (9% yield after 30 min, Table 1, entry 6). A
decrease of the reaction temperature from 180°C to 165°C
obviously lowered the reaction rate but unfortunately had also a
55 negative impact on the reaction selectivity. At 165°C, the
maximum yield of HMF was 30% which correspond to an HMF
2
| Journal Name, [year], [vol], 00–00
This journal is © The Royal Society of Chemistry [year]

Page 3 of 9
Green Chemistry
View Article Online
DOI: 10.1039/C3GC41495G
Table 2 Catalytic conversion of glucose to HMF over various Lewis acids
AlCl3 (43%) up to 70% conversion of glucose (see supporting
information, Fig S1, S2 and S3). However, upon prolonged
reaction time significant difference was observed between both
metal chlorides. In the presence of AlCl3 the yield of HMF
35 gradually increased up to 70% (at 90% conversion). Reversely, in
the presence of FeCl3, the yield of HMF dramatically dropped at
conversion higher than 70%. This surprising difference of
behaviour between both metal chlorides prompted us to more
deeply investigate the stability of catalytically active species in
40 our system.
Entry
Catalyst
Conv. Glucose Optimized conditions to get the
(%) after 15
min
maximum molar yield of HMFa
Time (min) Conv (%) HMF
yield
(%)
Recent works have shown that, in the presence of water, metal
chlorides such as CrCl2 and AlCl3 are in equilibrium with metal
hydroxide species M(OH)x(H2O)y and HCl.12 Metal hydroxide is
suspected to be the catalytically active species in the
45 isomerization of glucose to fructose while released HCl would
catalyze the dehydration of fructose to HMF. Assuming that
metal hydroxide species are suspected to be the catalytically
active sites in the isomerization of glucose to fructose, we first
heated FeCl3 and AlCl3 in neat water (30 min at 165°C under
50 microwaves) prior to add ChCl and glucose. The pH of the
solution was measured before and after heating of the
corresponding metal chloride in neat water (Table 3). As
expected, in both cases, an important drop of the pH was
observed after heating in water which is in accordance with a
55 partial hydrolysis of metal chlorides accompanied by the release
of HCl in the medium. Note that measured pH were in the same
range between AlCl3 and FeCl3 suggesting that the higher activity
of AlCl3 cannot be ascribed to a greater acidity. However,
noticeable visual differences were observed between preꢀheated
60 aqueous solution of AlCl3 and FeCl3. Whereas the solution
remained homogeneous in the case of AlCl3, we clearly observed
the formation of a rustꢀcoloured precipitate in the case of FeCl3
which was attributed to the formation of hydrated iron oxide as
supported by RAMAN and XRD analyses (see supporting
65 information, Fig S4 and S5) (Fig. 1). Hence, we suspected here
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
Al(NTf2)3
Fe(NTf2)3
Cu(NTf)2
Al(OTf)3
Cu(OTf)2
99
63
23
99
47
44
22
99
ꢀ
99
99
58
56
26
23
15
12
5
3
87
63
72
93
54
66
nd
89
90
99
93
72
69
nd
nd
nd
nd
nd
74
50
56
34
46
54
23
24
nd
67
70
49
57
43
37
nd
nd
nd
nd
nd
34
22
15
15
3
20
15
nd
3
15
15
3
25
30
nd
nd
nd
nd
nd
10
5
b
Fe(OTf)3
Bi(OTf)3
AlCl3
c
AlCl3
CrCl2
CrCl3
FeCl3
CuCl2
MgCl2
FeCl2
MnCl2
BiCl3
ZnCl2
BHCd
HCld
ꢀ
ꢀ
a nd= not determined; b result collected at 180°C, c result collected
at150°C, d BHC = betain hydrochloride, pH=1
through a maximum during the catalytic process, we first
monitored the HMF yield versus time. At 165°C, kinetic studies
confirmed the high activity of AlCl3 and we found that HMF was
produced with a maximum yield of 67% after only 3 min of
reaction which corresponds to a glucose conversion of 89% and
an HMF selectivity of 75% (Table 2, entry 8). AlCl3 is so active
5
Table 3 Stability of catalytic speciesa
10 that we were able to decrease the temperature to 150°C (entry 9).
At this temperature, a slight increase of the HMF yield to 70%
was observed. Note that this yield value is similar to that
previously reported by Zhang in the presence of 6 mol% of CrCl2
in 1ꢀbutylꢀ3ꢀmethylimidazolium chloride (reference work in this
15 field of chemistry),7 thus showing the potential of such catalytic
system for the sustainable production of HMF. In the case of
FeCl3, a maximum yield of HMF of 43% (entry 12) was obtained
after 25 min of reaction, which is quite close to the yields of
HMF collected with chromiumꢀbased catalysts in our conditions
20 (49% and 57% yield in the presence of CrCl2 and CrCl3,
respectively, entries10ꢀ11). Remarkably, whereas Bronsted acid
usually provides very low yields of HMF from glucose (also
observed here in the absence of ChCl), we were pleased to see
here that, in the presence of ChCl, HMF was produced with 34%
25 and 22% yield with BHC (betaïn hydrochloride) and HCl,
respectively (Table 2, entries 19ꢀ20), reinforcing the key role
played by ChCl in the conversion of glucose to HMF.
Without preꢀheating
With preꢀheating
Catalyst pHb
Conv.
glucose
(%)
HMF molar
pHc
Conv.
HMF molar
yield (%)a
glucose (%) yield (%)a
AlCl3
FeCl3
3.54
2.10
73
72
44
43
1.86
1.18
77
72
42
22
a 3 mol % of catalyst, 150°C, reaction time was adjusted in compared to
compare data at same conversion; b initial pH of the solution after
addition of the catalyst; c initial pH of the solution after preꢀheating metal
70 chlorides in water for 30 min at 165°C
Next, we tried to rationalize the greater selectivity of AlCl3 as
compared to that of FeCl3 in our conditions. Interestingly, we
30 found that yields of HMF were comparable between FeCl3 and
This journal is © The Royal Society of Chemistry [year]
Journal Name, [year], [vol], 00–00 | 3

Green Chemistry
Page 4 of 9
View Article Online
DOI: 10.1039/C3GC41495G
Table 4 Effect of various salts in the acidꢀcatalyzed dehydration of
55 glucose to HMFa
5
Entry
1
Salt
Reaction
media
Conv.
HMF molar
(a) Before heating
(b) after heating
glucose (%) yield (%)
Fig. 1 Pictures of the solution of FeCl3 and AlCl3 (a) before and (b) after
heating in water at 165°C
H2O
58b
33
10 that the lower efficiency of FeCl3 might be ascribed to the
precipitation of iron oxide species during the reaction resulting in
a dramatic decrease of the glucose/fructose isomerization rate to
the benefit of side reactions catalyzed by released HCl. To further
support this hypothesis, ChCl and glucose were added to the preꢀ
15 heated aqueous solution of AlCl3 and FeCl3. Whereas a similar
yield of HMF was recovered using a preꢀheated aqueous solution
of AlCl3, the selectivity to HMF was dramatically decreased in
the case of FeCl3 further demonstrating that precipitation of iron
oxides species has a detrimental effect on the catalyst activity and
20 selectivity. Note that when ChCl was added during the preꢀ
heating step of metal chlorides in water, a rustꢀcoloured
precipitate was also observed from FeCl3 while all species
remained soluble with AlCl3 suggesting that ChCl did not prevent
the deactivation of FeCl3.
87b
15b
9
1
2
3
[BMIM]Cl
H2O
ꢀ
ꢀ
4
5
DMSO
H2O
45b
57b
1
18
6
7
H2O
H2O
50c
56b
12
15
25 Role of ChCl
NaCl
a FeCl3, 3 mol %, 165°C, 10 wt% of glucose; b 15 min; c 30 min.
To collect more information on the synergistic role of ChCl and
water in our system, the reaction was carried out without ChCl
either in neat 1ꢀbutylꢀ3ꢀimidazolium chloride or in neat water. All
reactions were performed at 165°C (250W under microwaves)
30 with a glucose loading of 10 wt% and a H2O/MIBK ratio of 2/2.
Although AlCl3 is more stable than FeCl3 in our conditions, FeCl3
has been selected in the following experiments mainly because
this metal chloride is generally considered as a poorly active
catalyst in such reaction. Hence, FeCl3 appears as a good
35 candidate to clearly highlight the promoting effect of ChCl on the
catalysts selectivity. In the absence of ChCl, HMF was produced
with a very low yield of 9% and 1% in [BMIM]Cl and in neat
water, respectively, and an important formation of black materials
was observed (Table 4, entries 2, 3). Similarly, when the catalytic
40 reaction was conducted without water in a mixture ChCl/DMSO,
HMF was also produced with only 1% yield (Table 4, entry 4)
suggesting that combination of ChCl with H2O is one of the key
for explaining the selectivity improvement observed with FeCl3.
ChCl was also replaced by other water soluble quaternary
45 ammonium salts to confirm its effect on the reaction selectivity.
In such case, the amount of salt was adjusted to keep constant the
mol% of salt introduced in water. Reactions were all stopped and
compared at a conversion range of 50ꢀ60% to evaluate the role of
choline on the HMF selectivity. When ChCl was replaced by
50 dimethyldibutylammonium chloride, methyltributylammonium
chloride or NaCl, the yield of HMF dramatically dropped to
18%,12% and 15%, respectively (vs 33% with ChCl, Table 4
entry 1). Additionally, when ChCl was replaced by ethanol, the
selectivity of the reaction was shifted towards the formation of
fructose (18 % yield at 63 % conversion) in agreement with the
recent work of Riisager.13 From these results, one may also
60 conclude that both the ꢀOH and ammonium groups of ChCl play
a crucial role in the reaction mechanism.
To get more insight on the role of ChCl in such catalytic reaction,
we carried out extra experiments. ChCl is capable of producing
eutectic mixtures with hydrogen bond donors such glucose or
65 fructose.14 In a first approximation, one may suspect that, in
water, ChCl interact stronger with fructose than with glucose thus
facilitating the isomerization of glucose to fructose. By means of
calculations (see ESI), we found that the total energy of the
solutions, including terms due to the solutes (sugars), the solvent
70 (ChCl/H2O) and also soluteꢀsolvent interactions, were similar
between the glucose/ChCl/H2O and fructose/ChCl/H2O systems.
Additionally, decompositions of the energy into different terms,
either by nature (electrostatic, van der Waals, intramolecular) or
by component (solute, solvent), gave very close values for the
75 two systems. Hence, from these results, one may conclude that
the two systems have a similar energy and thus rule out the
possible “assisting role” of ChCl in the glucose/fructose
isomerization step.
Considering that the yield of HMF is optimal at a ChCl/H2O ratio
80 of 1/1, we next focused our attention on the partition coefficient
of HMF in these biphasic systems. Dumesic and coꢀworkers have
4
| Journal Name, [year], [vol], 00–00
This journal is © The Royal Society of Chemistry [year]

Page 5 of 9
Green Chemistry
View Article Online
DOI: 10.1039/C3GC41495G
previously showed that a “continuous” extraction of HMF from
the aqueous phase allowed its side reactions with water or
results. At a ChCl content lower than 50 wt%, metal hydroxide
species are produced in water which is accompanied by the
hexoses (i.e. rehydration to levulinic and formic acids, 60 release of HCl. The coordination sphere of these metal hydroxide
condensation to humins, etc…) to be limited.15 To this end, we
reproduced the composition of the reaction media obtained at the
end of the reaction (see composition entry 9 of Table 2). In a
typical experiments, HMF (a quantity equivalent to 70% yield)
species may be modified by increasing the amount of ChCl in the
reaction media as previously shown by K. Binnemans and R. Van
Deun. Note that the possible role of ChCl as a ligand is consistent
with our aboveꢀdescribed experiments summarized in Table 4. In
5
was heated in a biphasic water/MIBK system (each phase = 10 65 particular, by increasing the concentration of ChCl in water, the
mL) at 100°C for 10 min. The MIBK and aqueous phase were
nature and concentration of catalytic species in the reaction media
might change which would be consistent with a change of the
selectivity to HMF. Concomitantly, the possible extraction of
HMF from the aqueous phase with MIBK allowed limiting side
10 then separated and analyzed by HPLC. The partition coefficient
was defined as the mass ratio between the amount of HMF
detected in the MIBK and water phases, respectively. Then, same
experiments were reproduced by progressively substituting water 70 reactions of HMF with water and hexoses. In this context, best
by ChCl. Result of this study is presented on Fig. 2.
selectivity to HMF was obtained in a biphasic aqueous solution of
ChCl (50 wt%)/MIBK. At a ChCl content higher than 50 wt%,
strong molecular interaction between ChCl and produced HMF
became dominant and HMF was more strongly retained in the
15 Interestingly, up to a ChCl content of nearly 50 wt%, the partition
coefficient did not significantly vary and remained around 1.2
(i.e. about 53 wt% of HMF was extracted with MIBK).
Reversely, at a higher ChCl content, the partition coefficient 75 aqueous phase. As a consequence, HMF is subjected to side
significantly and continuously decreased up to 0.6 (i.e. about 38
20 wt% of HMF in the MIBK phase) at a ChCl content of 80 wt%.
This result suggests that, at a ChCl content higher than 50 wt%,
HMF is strongly retained in the aqueous phase, presumably
reactions with water or hexoses leading to a drop of the reaction
selectivity. Note that this hypothesis is supported by our recent
findings that suggested the formation of a possible DES between
HMF and ChCl.17 Hence, one may conclude that 50 wt% of ChCl
through hydrogen bond interaction. As a consequence, in this 80 in water is an acceptable compromise between the catalyst
scenario, side reactions of HMF with water (and hexoses) should
25 be more favourable and should result in a drop of the reaction
selectivity which is consistent with our experimental observation.
activity/selectivity and the extraction of HMF from the aqueous
phase.
One should comment that when all the aboveꢀdescribed reactions
were performed in conventional heating, HMF was produced
85 with a much lower selectivity (in this case insoluble black
materials were produced in a higher amount) and that whatever
the reaction conditions (temperature, time, fructose and catalyst
loading, etc…). In conventional heating, it has been previously
demonstrated that the isomerization rate of glucose to fructose
90 was nearly 4 times lower than the dehydration rate of fructose to
HMF.18 Because of the relative instability of HMF in aqueous
acidic conditions, it is clear that the production of HMF is in
competition with its own degradation, rehydration or
condensation. Although we are fully convinced that microwaves
95 also contributed to a better control of the reaction selectivity, all
attempts to identify which elementary step was affected by
microwaves in a greater extent unfortunately failed mainly due to
the complexity of the system. Note that during the course of our
investigations, similar observation has been reported by Rinaldi,
100 Dumesic and coꢀworkers.19
1,4
1,2
1
120
100
80
60
40
20
0
30
0,8
0,6
0,4
0,2
0
35
Partition coefficient
HMF yield
0
20
40
60
80
100
wt % of ChClin water
Fig. 2 Partition coefficient of HMF in biphasic ChClꢀH2O/MIBK systems
40 and HMF yield (see conditions Table 2, entry 9) as a function of the ChCl
concentration
Although consideration of the partition coefficients allowed
explaining the behaviour of our system at high loading of ChCl, it
does not explain why the selectivity to HMF was increased when
45 the ChCl content was concomitantly raised from 0 to 50 wt% (in
this range of concentration, the partition coefficients are similar,
Fig. 2). Recently, K. Binnemans and R. Van Deun have
investigated the speciation of copper chlorides in ChCl/water
mixtures.16 This work clearly showed by EXAFS, electrochemical
50 measurements and crystallography that the environment of the
cupric ion is modified according to the amount of ChCl in water.
At a water concentration of 39 wt% in ChCl, ChCl is coordinated
on copper while at a concentration higher than 95%, the cupric
ion is fully hydrated. At an intermediate water/ChCl composition,
55 mixed chloroꢀaquo complexes are dominant species. Inspired by
data from the current literature and by our experimental results,
we propose hereafter a plausible explanation for rationalizing our
Recyclability of the system
Having all these results in hand, we then evaluated the
recyclability of the ChClꢀH2O phase containing the acid catalyst.
Recycling experiments were performed at 150°C for 15 min in
105 the presence of 3 mol % of AlCl3. In all cases, MIBK was added
to continuously extract HMF from the reaction media (ChClꢀ
H2O/MIBK ratio = 2ꢀ2/2). After the first catalytic run, the
reaction media was cooled down to room temperature and
decanted. The MIBK phase was separated and the aqueous phase
110 containing ChCl and AlCl3 was washed again with MIBK to
ensure the complete removal of HMF from the catalytic phase.
Then, glucose was directly added to the catalytic phase and
heated for another 15 min at 150°C. One can see from Fig. 3 that
the yield of HMF remained constant during at least three
115 consecutive recycling. After the third recycling, the yield of HMF
This journal is © The Royal Society of Chemistry [year]
Journal Name, [year], [vol], 00–00 | 5

Green Chemistry
Page 6 of 9
View Article Online
DOI: 10.1039/C3GC41495G
Table 5 Catalytic conversion of cellulose to HMF in an aqueous solution
of ChCl
80
70
60
50
40
30
20
10
0
Extraction
of ChCl
5
Entry
Cellulose
Catalyst
ChCl Glucose molar HMF molar
cristallinity
yield (%)
yield (%)
Higha
Higha
Higha
Higha
Higha
Lowe
Lowe
FeCl3
AlCl3
FeCl3
no
no
yes
yes
yes
no
41
ꢀ
26
ꢀ
ꢀ
70
ꢀ
6
8b
17
14c
27
7
10
1
2
3
4
5
6
7
1
2
3
4
5
6
Catalytic run
AlCl3
d
d
FeCl3/AlCl3
FeCl3
FeCl3/AlCl3
Fig. 3 Recycling experiments (150°C, 15 min, 3 mol% of AlCl3, ChClꢀ
H2O/MIBK ratio = 2ꢀ2/2.
yes
49
60 a microcrystalline cellulose PH AVICEL 101; b along with 2% of glucose
and 3% of levulinic acid; c similar yields was obtained with CrCl2 (17%
yield); d reaction was performed in two steps. First 200°C for 5 min in
neat water containing 4 wt% (vs cellulose) of FeCl3 followed by addition
of 3 mol% of AlCl3 (vs in-situ produced glucose) and 50 wt% of ChCl
65 (heating at 135°C for 90 min); e cellulose with low cristallinity index was
obtained by a ballꢀmilling or a dissolution/regeneration from 1ꢀbutylꢀ3ꢀ
imidazolium chloride. Note that similar results were obtained whatever
the preꢀtreatment.
15 however continuously dropped mainly due to the accumulation of
impurities in the catalytic phase. After the fifth run, we attempted
to recover ChCl to reuse it. To this end, the reaction media was
diluted with 20 mL of water resulting in the precipitation of black
materials which were filtered off. Water was then evaporated and
20 the recovered residue was dried overnight at 150°C prior to be
extracted with ethanol. After removal of ethanol under reduced
pressure, ChCl was recovered (> 90 wt%) with a sufficient purity
to be reused (see Fig. S6 for copies of NMR spectra). Indeed,
after extraction and reuse of ChCl, a yield of HMF of 60% was
25 recovered demonstrating the relative stability of ChCl in our
working conditions.
precipitated (under the form of hydrated iron oxides) from the
70 reaction media leading to an important release of HCl in the
solution, a Bronsted acid known to efficiently catalyze the
depolymerization of cellulose to glucose. Reversely, from AlCl3,
all species (HCl and aluminium hydroxides) remained soluble
and, without ChCl, the reaction suffered from an important lack
Transposition to cellulose
Encouraged by all these results, we then attempted the direct 75 of selectivity.
conversion of cellulose to HMF which is an even more
30 challenging reaction. This cascade reaction requires first a
hydrolysis of cellulose to glucose followed by the
isomerization/dehydration of glucose to HMF (Fig. 4). Reactions
Next, ChCl was added to a solution containing FeCl3 and
cellulose. In this case, the selectivity of the reaction was changed
and glucose was produced with 26% yield along with 17% yield
of HMF, further demonstrating the beneficial effect of ChCl for
were first performed at 200°C (under microwave irradiations) in 80 the production of HMF (Table 5, entry 3). Note that the reaction
the presence of
4
wt% of FeCl3 or AlCl3 for
5
min.
selectivity was more difficult to control from cellulose than from
glucose. In such conditions, HMF was indeed also readily
converted to levulinic and formic acids. Similar trend was also
observed using AlCl3 and CrCl2 (14% and 17% yield of HMF,
35 Microcrystalline cellulose (PH AVICEL 101) was used in all
experiments. Interestingly, in the absence of ChCl, cellulose was
selectively converted to glucose (41% yield) in the presence of
FeCl3 whereas AlCl3 afforded glucose, HMF and levulinic acid in 85 respectively, entry 4).
2%, 8% and 3% yield respectively (along with many other side
40 products) (Table 5, entries 1, 2). This difference of selectivity
between FeCl3 and AlCl3 is consistent with our previous
observations. During the reaction with FeCl3, iron hydroxide
To increase the selectivity to HMF, cellulose was first heated in
the presence of FeCl3 (hydrolysis of cellulose to glucose) prior to
add AlCl3 (conversion of glucose to HMF). As observed above,
glucose was produced with 40% yield in the presence of FeCl3
90 only. Then AlCl3 was directly added to the solution (without
intermediate purification) and the resulting mixture was heated at
135°C for 90 min. Interestingly, in this case, the in-situ produced
glucose was converted to HMF with similar selectivity (68%)
than that described above from glucose leading to HMF with a
95 yield of 27% (Table 5, entry 5). Please note that combination of
HCl and AlCl3 did not provide similar results suggesting that
FeCl3 still participate to some extent to the conversion of glucose
to HMF. To further increase the yield of HMF, microcrystalline
cellulose was subjected to a preꢀtreatment process prior to its
100 catalytic conversion to HMF. To this end the crystallinity index
of microcrystalline cellulose was decreased either by
dissolution/regeneration in 1ꢀbutylꢀ3ꢀmethylimidazolium chloride
([BMIM]Cl) or by ballꢀmilling. As expected, after these two preꢀ
45
50
55
Fig. 4 Catalytic conversion of cellulose to HMF in an aqueous solution of
ChCl containing FeCl3 and AlCl3
6
| Journal Name, [year], [vol], 00–00
This journal is © The Royal Society of Chemistry [year]

Page 7 of 9
Green Chemistry
View Article Online
DOI: 10.1039/C3GC41495G
treatments, cellulose was found to be much less recalcitrant.
During the first step in the presence of FeCl3, cellulose was
deconstructed to glucose with a yield of 70% versus 40% without
FMC BioPolymer (degree of polymerization about 250, glucose
content > 99%, water content = 5wt %, particle size < 40µm).
Distilled water was used as in all catalytic reactions.
preꢀtreatment (Table 5, entry 6). Then, after addition of AlCl3, in- 60 Catalytic experiments: All reactions were performed in an
5
situ produced glucose was readily converted to HMF with 49%
yield (i.e. 70% selectivity), (Table 5, entry 7). Although the large
energy input required by the preꢀtreatment of cellulose by ballꢀ
milling or by dissolution in ionic liquids nowadays represents a
automated multimode microwave oven MicroSynth from
Milestone s.r.l. (Italy). This oven operates at 2.45 GHz and is
temperature controlled by a fiberoptic sensor. Temperature was
monitored by inserting the calibrated fiberoptic sensor in the
major drawback, the coupling of our methodology with the recent 65 reaction mixture while stirring properly. The maximum power
10 work reported by F. Schüth20 in the field of mechanocatalytic
deconstruction of cellulose to water soluble celloꢀoligomers may
open a more ecoꢀefficient route for the production of HMF from
cellulose.
input could be adjusted between 0 and 1000 Watt. Under
microwave heating, the internal temperature of the reaction vessel
reached the reaction setꢀpoints in 20 min, and was maintained at
± 5°C using a power compensation feedback control.
70 Catalytic conversion of glucose to HMF: In
a typical
experiment, 0.4 g of glucose was dissolved in 2 g of water
containing 2 g of ChCl and 3 mol% of a Lewis acid. Then 2g of
MIBK were added and the biphasic solution was heated under
microwave (250W) at the desired temperature for 1ꢀ60 min (see
75 Tables for conditions).
Recycling procedure: At the end of the reaction (conditions: 10
wt% of glucose, 3 mol% of AlCl3, T =150°C, 15 min, 6g of
ChClꢀH2O/MIBK (ratio 2ꢀ2/2)), the reaction was cooled down to
room temperature and the MIBK phase containing the HMF was
80 separated by decantation. Next, the catalytic phase was washed
two times with 2.5 mL of MIBK to remove the residual HMF.
Then, glucose was added to the catalytic phase and the reaction
was heated again at 150°C for 15 min.
Catalytic conversion of cellulose to HMF: 200 mg of cellulose
85 was first suspended in 2 mL of water containing 18 mg of FeCl3.
The solution was then heated under microwave (250W) for 5 min
at 200°C. Then the reaction was cooled down and 16 mg of
AlCl3, 2 g of ChCl and 2 mL of MIBK was directly added to the
solution. The mixture was then heated for 30ꢀ180 min at 120ꢀ
90 135°C under microwave (250W).
Conclusion
15 We report here that ChCl can be considered as a safe and cheap
additive capable of assisting the catalytic conversion of glucose
to HMF in water. Under optimized conditions, HMF was
produced with 70% yield which is a competitive yield to those
traditionally obtained in imidazoliumꢀbased ionic liquids in the
20 presence of hazardous chromium salts. A screening of various
Lewis acids revealed that ChCl has the ability to improve the
selectivity of various safe and abundant metal chlorides such as
AlCl3, FeCl3 and CuCl2, thus increasing the sustainability of this
process. It should be noted that the promoting effect of ChCl on
25 the HMF selectivity is however closely depending on the ChCl
content. Although the exact elucidation of the reaction
mechanism involved in the presence of ChCl is difficult to be
determined with accuracy at this stage, we suggest that at a ChCl
lower than 50 wt% in water, the selectivity is governed by 1) the
30 coordinating ability of ChCl that presumably modifies the nature
and the concentration of catalytic species in water, and 2) the
extraction of HMF from the aqueous phase, thus limiting side
reactions of HMF with water or hexoses. At a ChCl content
higher than 50 wt%, strong molecular interaction between HMF
35 and ChCl became dominant making extraction of HMF from the
aqueous phase more difficult which resulted in a decrease of the
reaction selectivity.
The scientific strategy of this work can be transposed to the direct
production of HMF from cellulose which is a much more
40 challenging reaction. In this case, combination of FeCl3 and
AlCl3 allowed cellulose to be converted to HMF with 49% yield
in a one pot process.
From the view point of sustainable chemistry, this work gathers
noticeable advantages such as use of 1) water as a solvent, 2)
45 ChCl as a cheap and safe additive, 3) cheap and naturally
abundant metal (Al, Fe, Cu) and 4) renewable raw materials. We
do believe that this work, together with the recent progress
reported in the field of process intensification (including biphasic
catalysis) will definitely contribute to increase again the ecoꢀ
50 efficiency and sustainability of this work.
Analytical: Conversion of glucose was calculated by external
calibration using a Shimadu Prominence LC equipped with a
degasser DGUꢀ20A3,
a
pump system (5LCꢀ20AQ),
a
thermostated autosamplor SILꢀ20AC and an oven CTOꢀ20AC
95 maintained at 40°C. Glucose was separated on a Shodex KSꢀ802
column, using water as an eluent, and quantified by a RID. Yields
of HMF were determined by external calibration at 25°C using an
HPLC equipped with a Nucleosil 100–5C18 column (250 x 4.6
mm), a Shimadzu LCꢀ20AT pump, a Shimadzu RIDꢀ10A
100 detector, and an acetonitrile/water (10:90) mixture as mobile
phase (0.8 mL.minꢀ1).
Acknowledgements
Authors are grateful to the CNRS, the French Ministry of
Research and SOLVAY for their financial supports.
105 Notes and references
a Institut de Chimie des Milieux et Matériaux de Poitiers,
CNRS/Université de Poitiers/ENSIP, 1 rue Marcel Doré, 86022
Poitiers/FRANCE Fax: +33 5 49 45 33 49; Tel: +33 5 49 45 40 52;
E-mail: francois.jerome@univ-poitiers.fr
Experimental Section
Chemicals: Fructose, glucose, choline chloride and
methyisobutylketone (MIBK) were purchased from Sigma
Aldrich as well as anhydrous and hydrated Lewis acids. All
55 chemicals were used as received without further purification.
Microcrystalline cellulose PH AVICEL 105 was purchased to
110 b Eco-Efficient Products and Processes Laboratory, Unité Mixte de
Recherche UMI 3464 CNRS/RHODIA, 3966 Jin Du Road, Shanghai
201108, China, Fax: +86 21 54425911 ; Fax, Tel : +86 21 24089391, E-
mail: floryan.decampo@ap.rhodia.com
This journal is © The Royal Society of Chemistry [year]
Journal Name, [year], [vol], 00–00 | 7

Green Chemistry
Page 8 of 9
View Article Online
DOI: 10.1039/C3GC41495G
c a Institut de Chimie de Clermont-Ferrand, CNRS/Université Blaise
Pascal Clermont-Ferrand, 24 avenue des Landais, BP 80026, 63171
Aubiére, France. E-mail : agilio.padua@univ-bpclermont.fr
‡ Footnotes should appear here. These might include comments relevant
to but not central to the matter under discussion, limited experimental and
spectral data, and crystallographic data.
75
80
5
10
15
20
25
30
35
1
P. T. Anastas, J. C. Warner, in Green ChemistryꢀTheory and Practice,
Oxford University Press: New York, 1998.
2
M. E Zakrzewska, E. BogelꢀLukasik, R. BogelꢀLukasik, Chem. Rev.,
2001, 111 (2), 397ꢀ417; b) J. Lewkowski ArkiVoc 2001, 2, 17ꢀ54; c)
T. Ståhlberg, W. Fu, J. M. Woodley, A. Riisager, ChemSusChem,
2011, 4(4), 451ꢀ458; d) A.A. Rosatella, S. P. Simeonov, R. F. M.
Frade, C. A. M. Afonso, Green Chem., 2011, 13, 754ꢀ793.
As representative selected works see M. Molinier, Y. Romanꢀ
Leshkov, M. E. Davis, Proc. Natl. Acad. Sci., 2010, 107 (14), 6164ꢀ
6168.
As selected works see a) R. O. L Souza, P. D. Fabiano; C. Feche, F.
Rataboul, D. Cardoso, N. Essayem, Catal. Today, 2012, 195(1), 114ꢀ
119J; b) S. Yu, E. Kim, S. Park, I. K. Song, J. C. Jung, Catal.
Commun., 2012, 29, 63ꢀ67 ; c) B. Y. Yang, R. Mongomery,
Carbohyd. Res., 1996, 280 (1), 27ꢀ45
3
4
85
90
5
6
As an example see S. Bhosale, M. Rao, V. Deshpande, Microbiol.
Rev., 1996, 60(2), 280ꢀ300; b) H. Yu, Y. Guo, D. Wu, W. Zhan, G.
Lu, J. Mol. Catal. B Enzym., 2011, 72 (1ꢀ2), 73ꢀ76; c) A. M.
Dehkordi, M. S. Tehrany, I. Safari, Ind. Eng. Chem. Res., 2009,
48(7), 3271ꢀ3278
As a recent work see P. M. Grande, C. Bergs, P. Dominguez de
Maria, ChemSusChem, 2012, 5(7), 1203ꢀ1206 and references cited
therein
95
100
7
8
H. Zhao, J. E. Holladay, H. Brown, Z. C. Zhang, Science, 2006, 316,
15971599.
a) Q. Zhang, K. De Oliveira Vigier, S. Royer, F. Jérôme, Chem. Soc.
Rev., 2012, 41, 7108ꢀ7146 ; b) A. P. Abbott, J. C. Barron, K. S.
Ryder and D. Wilson, Chem.–Eur. J., 2007, 13, 6495–6501; c) D. 105
Carriazo, M. C. Serrano, M. C. Gutierrez, M. L. Ferrer, F. del Monte,
ChemSocRev, 2012, 1(14), 4996ꢀ5014 ; d) P. D. de Maria, Z.
Maugeri, Curr. Opin. Chem. Biol, 2011, 15(2), 220ꢀ225.
9
F. Liu, K. De Oliveira Vigier, M. PeraꢀTitus, Y. Pouilloux, J.ꢀM.
Clacens, F. Decampo, F. Jérôme, Green Chem, 2013, 15, 901ꢀ909
110
40 10 J. T. Gorke, F. Srienc and R. J. Kazlauskas, Chem. Commun., 2008,
1235–1237
11 Y. J. PagánꢀTorres, T. Wang, J. M. R. Gallo, B. H. Shanks, J. A.
Dumesic, ACS Catal., 2012, 2, 930−934
12 As a very recent work see V. Choudhary, S. H. Mushrif, C. Ho, A. 115
Anderko, V. Nikolakis, N. S. Marinkovic, A. I. Frenkel, S. I. Sandler,
D. G. Vlachos J. Am. Chem. Soc., 2013, 135, 3997ꢀ4006 and
references cited therein.
45
13 S. Saravanamurugan, M. Paniagua, J. A. Melero, A. Riisager, J. Am.
Chem. Soc., 2013, 135 (14), 5246–5249.
50 14 F. Ilgen, D. Ott, D. Kralish, C. Reil, A. Palmberger, B. König, Green
Chem., 2009, 11, 1948–1954.
15 as a selected comprehensive review from authors see S. G Wettstein,
D. Martin Alonso, E. I Gürbüz, J. A Dumesic, Curr Opin. Chem.
Eng., 2012, 1, 218ꢀ224 and references cited therein.
55 16 P. De Vresse, N. R. Brooks, K. Van Hecke, L. Van Meervelt, E.
Matthijs, K. Binnemans, R. Van Deun, Inorg. Chem., 2012, 51, 4972ꢀ
4981.
17 F. Liu, J. Barrault, K. De Oliveira Vigier, F. Jérôme, ChemSusChem,
2012, 5 (7), 1223ꢀ1226
60 18 A) Y. Yang, C.ꢀW. Hu, M. M. AbuꢀOmar, ChemSusChem, 2012, 5,
405; b) Y. Yang, C.ꢀW. Hu, M. M. AbuꢀOmar, Green Chem., 2012,
14, 509.
19 R. CarrasquilloꢀFlores, M. Käldstöm, F. Schüth, J. A. Dumesic, R.
Rinaldi, ACS Catal., 2013, 3, 993ꢀ997
65 20 N. Meine, R. Rinaldi, F. Schüth, ChemSusChem, 2012, 5, 1449ꢀ1454
120
125
130
135
140
70
8
| Journal Name, [year], [vol], 00–00
This journal is © The Royal Society of Chemistry [year]

Page 9 of 9
Green Chemistry
View Article Online
DOI: 10.1039/C3GC41495G
Graphical abstract
5
Selectivity enhancement in the aqueous acid-catalyzed conversion of
glucose to 5-hydroxymethylfurfural induced by choline chloride
Fei Liu,a Maïté Audemar,a Karine De Oliveira Vigier,a Damien Cartigny,b Jean-Marc Clacens,b
10 Margarida F. Costa Gomes,c Agilio A. H. Pádua,c Floryan De Campo,b and François Jérôme*a
15
20
25
Here, we wish to show that addition of choline chloride to an aqueous solution of glucose containing a catalytic amount of AlCl3 or FeCl3 promotes the
30 tandem isomerization/dehydration of glucose to 5ꢀhydroxymethylfurfural with yield up to 70%. Such strategy can be also transposed to cellulose.
35
This journal is © The Royal Society of Chemistry [year]
Journal Name, [year], [vol], 00–00 | 9