An ionic liquid-organics-water ternary biphasic system enhances the 5-hydroxymethylfurfural yield in catalytic conversion of glucose at high concentrations

Article abstract of DOI:10.1039/c5gc01039j

Increasing the glucose loading in the 1-butyl-3-methylimidazolium chloride ([BMIM]Cl) ionic liquid containing a dissolved CrCl₃ catalyst system led to excessive formation of humins and a serious decrease in the 5-hydroxymethylfurfural (5-HMF) yield. A biphasic system containing glycol dimethyl ether (GDE) as the extraction phase, and [BMIM]Cl/CrCl₃/glucose in combination with a partitioned amount of GDE and an appropriate amount of water as the reaction phase was found to be highly efficient for the reaction; CrCl₃ catalyzed the formation of 5-HMF in 64.5 mol% yield from a very high glucose concentration (80 wt% with respect to the ionic liquid) at 108 °C. This 5-HMF yield in the [BMIM]Cl-GDE-H₂O ternary biphasic system nearly doubled that obtained in the single [BMIM]Cl/CrCl₃/glucose reaction phase. Importantly, the GDE phase contained about 56% of the generated 5-HMF without detectable contamination by the ionic liquid or carbohydrates. GDE served multiple functions: as a hydrogen-bond acceptor, it exhibited excellent extraction performance for 5-HMF; due to its low boiling point and suitable solubility saturation point in the ionic liquid, a sustained GDE bubbling phenomenon in the ionic liquid phase was observed that promoted the rate of inter-phase mass-transfer of 5-HMF in reactions; and GDE mediated the [BMIM]Cl phase to a reduced viscosity. In addition, an appropriate amount of water in the ternary system promoted the extraction efficiency of 5-HMF and also lowered the viscosity of [BMIM]Cl/glucose. The ionic liquid-organics-water ternary biphasic system has been demonstrated for high 5-HMF productivity and separation efficiency.

Full text of DOI:10.1039/c5gc01039j

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: J. Zhou, Z. Xia, T.
Huang, P. Yan, W. Xu, Z. Xu, J. Wang and Z. C. Zhang, Green Chem., 2015, DOI: 10.1039/C5GC01039J.
This is an Accepted Manuscript, which has been through the
Royal Society of Chemistry peer review process and has been
accepted for publication.
Accepted Manuscripts are published online shortly after
acceptance, before technical editing, formatting and proof reading.
Using this free service, authors can make their results available
to the community, in citable form, before we publish the edited
article. We will replace this Accepted Manuscript with the edited
and formatted Advance Article as soon as it is available.
You can find more information about Accepted Manuscripts in the
Information for Authors.
Please note that technical editing may introduce minor changes
to the text and/or graphics, which may alter content. The journal’s
standard Terms & Conditions and the Ethical guidelines still
apply. In no event shall the Royal Society of Chemistry be held
responsible for any errors or omissions in this Accepted Manuscript
or any consequences arising from the use of any information it
contains.
www.rsc.org/greenchem

View Article Online
DOI: 10.1039/C5GC01039J
Green Chemistry
Page 1 ofG11 reen Chemistry
Dynamic Article Links ►
Cite this: DOI: 10.1039/c0xx00000x
www.rsc.org/xxxxxx
PAPER
Ionic Liquid-Organics-Water Ternary Biphasic System Enhances
5-Hydroxymethylfurfural Yield in Catalytic Conversion of Glucose at
High Concentration
Jinxia Zhou,a,b Zhi Xia,a Tingyu Huang, a Peifang Yan,a Wenjuan Xu,a Zhanwei Xu, a Jianji Wang c and Z.
Conrad Zhang*a
5
Received (in XXX, XXX) Xth XXXXXXXXX 20XX, Accepted Xth XXXXXXXXX 20XX
DOI: 10.1039/b000000x
Increasing glucose loading in the 1-butyl-3-methylimidazolium chloride ([BMIM]Cl) ionic liquid
containing dissolved CrCl3 catalyst system led to excessive formation of humins and serious decrease in
10 5-hydroxymethylfurfural (5-HMF) yield. A biphasic system containing glycol dimethyl ether (GDE) as
the extraction phase, and [BMIM]Cl/CrCl3/glucose in combination with a partitioned amount of GDE and
an appropriate amount of water as the reaction phase was found highly efficient for the reaction; CrCl3
catalyzed the formation of 5-HMF in 64.5 mol% yield from a very high glucose concentration (80 wt%
with respect to the ionic liquid) at 108°C. This 5-HMF yield in the [BMIM]Cl-GDE-H2O ternary biphasic
15 system nearly doubled that obtained in the single [BMIM]Cl/CrCl3/glucose reaction phase. Importantly,
the GDE phase contained about 56% of the generated 5-HMF without detectable contamination by the
ionic liquid or carbohydrates. GDE served multiple functions: as a hydrogen-bond acceptor, it exhibited
excellent extraction performance for 5-HMF; due to its low boiling point and suitable solubility saturation
point in the ionic liquid, a sustained GDE bubbling phenomenon in the ionic liquid phase was observed
20 that promoted the rate of inter-phase mass-transfer of 5-HMF under reaction; and GDE mediated the
[BMIM]Cl phase to a reduced viscosity. In addition, an appropriate amount of water in the ternary system
promoted the extraction efficiency of 5-HMF and also lowered the viscosity of [BMIM]Cl/glucose. The
ionic liquid-organics-water ternary biphasic system has been demonstrated for high 5-HMF productivity
and separation efficiency.
production of various non-petroleum based chemicals and
fuels.2,3 Up to now, the C6 sugars, such as glucose and fructose,
35 as well as polysaccharides, such as cellulose and starch, have
been used to produce 5-HMF.4,5 Many studies adopted fructose as
a starting material as it can be readily converted to 5-HMF in
high yield by acid catalyzed dehydration.3,6,7 However, fructose is
not a highly prevailing carbohydrate in the nature. In contrast,
40 glucose exists as the building unit of the most abundant
carbohydrate polymers, cellulose and starch, in nature.8 Hence,
the perspective to economically convert glucose as the reactant to
5-HMF has motivated tremendous amount of recent researches.3-5
Although fructose has been shown to be readily dehydrated to
45 5-HMF, the synthesis of 5-HMF from glucose has been met with
considerable technical challenge for needing an efficient catalytic
system.9 A technically viable route for the conversion of glucose
to 5-HMF is via isomerization of glucose to fructose, followed by
subsequent dehydration of fructose to 5-HMF. Considerable
50 efforts have been focused on one-pot tandem conversion
strategies from glucose.2 Besides the development of efficient
catalysts with high selectivity to produce 5-HMF, the selection of
25
1
Introduction
The diminishing fossil fuel reserves and the growing concern
about global warming have driven up the potential utilization of
biomass resources as renewable feedstock for the production of
fuels and chemicals.1 Among numerous possible routes currently
30 under investigation, one with potentially high viability for
commercial development is the synthesis of 5-HMF, which has
been widely recognized as a biomass-derived intermediate for the
a Dalian National Laboratory of Clean Energy and State Key Laboratory
of Catalysis, Dalian Institute of Chemical Physics, Chinese Academy of
Sciences, Dalian, China. Email: zczhang@yahoo.com; Tel: (+86)411-
84379462; Fax: (+86)411-84379371
bCollege of Environmental and Chemical Engineering, Dalian
University, Dalian, China;
cSchool of Chemistry and Chemical Engineering, Key Laboratory of
Green Chemical Media and Reactions, Ministry of Education, Henan
Normal University, Xinxiang, Henan, China
This journal is © The Royal Society of Chemistry [year]
[journal], [year], [vol], 00–00 | 1

Green Chemistry
View ArticlPe Oanglinee 2 of 11
DOI: 10.1039/C5GC01039J
suitable solvent media is particularly important in enhancing 5-
partition coefficient and the 5-HMF concentration in the top
HMF yield and in facilitating 5-HMF separation. Generally, 60 organic phase. Second, although a large number of organic
water, aprotic organic solvents, and ionic liquids have been used
as the reaction media for the conversion of glucose to 5-HMF in
the presence of a catalyst.2-5 This reaction could hardly be
realized in pure water at a mild reaction condition.10 The reaction
solvents such as ethers, ketones, esters and alcohols are potential
extractants in biphasic reaction systems for 5-HMF production,
the most suitable type of the organics has to be determined, and
preferably with essentially predictable properties. Third, some
5
in biphasic organic/water system at high temperature (170-175◦C) 65 measures for enhancing the inter-phase mass transfer of 5-HMF
could generate 5-HMF, but the yield was unacceptable due to the
low selectivity to 5-HMF (26-33%).11,12 Some polar aprotic
must be taken to overcome the poor mixing under agitation due to
the large density difference between the organic phase and the
ionic liquid phase. Most importantly, 5-HMF production from
glucose in biphasic solvents should be evaluated in a reaction
10 solvents like dimethylsulfoxide (DMSO)13 or dimethylacetamide
(DMA)14 have been studied as the reaction media or as aqueous
solvent modifiers for the conversion of glucose into 5-HMF with 70 system with a high glucose concentration, so the results could be
13
AlCl3 or LiCl14 catalysts. However, it has been found that the
high boiling point of these organics makes the distillation process
15 energy-intensive and causes serious loss of the 5-HMF product
due to the reactive nature of 5-HMF at high temperature.
relevant for the design of a commercial 5-HMF production
process.
This work was specifically carried out for the synthesis of 5-
HMF from glucose in high concentration in [BMIM]Cl/CrCl3
Extractive separation faces great challenge because DMSO is 75 under mild reaction conditions. A series of organics and water
miscible with a wide range of organic solvents.15 Recently, some
ionic liquids have been reported to be highly effective reaction
20 media for one pot conversion of glucose to 5-HMF under mild
conditions.16,17 The first major breakthrough was reported based
were evaluated as components of the ternary biphasic system that
also effectively mediates the properties of [BMIM]Cl ionic liquid
to enhance the 5-HMF yield. The mechanism of the solvency
mediation on enhancing the 5-HMF yield and extraction
on the discovery in 1-ethyl-3-methylimidazolium chloride 80 efficiency in the high-concentration systems was investigated.
([EMIM]Cl) ionic liquid solvent.18 5-HMF yields of up to 70%
were obtained in this solvent with CrCl2 and CrCl3 as catalysts.
2
Material and Methods
25 Other forms of chloride salt or chromium ion based catalysts in 1-
alkly-3-methylimidazolium chloride, [AMIM]Cl, were also
effective. 19
2.1 Materials
5-Hydroxymethylfurfural (5-HMF, 98%) was obtained from
Aladdin (China). Tetrahydrofuran (THF, 99%), butyl ether (99%),
Despite of the significant progress over the recent years, a
challenge that must be addressed before commercial production 85 methyl tertiary butyl ether (MTBE, 99%), cyclohexanone (Anon,
30 of 5-HMF is the ability to adopt glucose in higher concentrations
in the process. Because a dilute system requires high capital
investment and operating expenses for a process, a low-
concentration process is economically unfavorable on
99%), glycerol (99%) and D-glucose (glucose, 99%) were
purchased from Sinopharm (China). Glycol dimethyl ether (GDE,
99%) and acetylacetone (Hacac, 99%) were supplied by Tianjin
Guangfu Chemical Reagent Co., Ltd (China). Ethyl acetate
commercial scale. However, most reported works over the past 90 (EtOAc, 99%) was purchased from Tianjin Kermel Chemical
35 were limited to researches on low glucose concentrations (in
loadings in ≤ 10 wt% with respect to the reaction solvent)2-6 with
the purpose of gaining insights on catalytic performance of
different catalytic systems. Reaction based on low feed
Reagent Co.,Ltd (China). n-butanol (99%) was bought from
Tianjin Damao Chemical Reagent Factory (China). Methyl
isobutyl ketone (MIBK, 99%) was supplied by Beijing Chemical
Reagent Factory (China). Deionized water (DI H2O) was supplied
concentration could achieve good results, such as high feed 95 by Hangzhou Wahaha Group Co., Ltd. (China). All the chemicals
40 conversion and high product selectivity, but it cannot effectively
predict the performance from high-concentration systems.20,21 Up
till now, the effect of higher glucose concentration has received
little attention in ionic liquid systems according to the published
above were used as received. [BMIM]Cl was purchased from
Lanzhou Greenchem ILS, LICP (China). It was degassed through
online-vacuum at 100 C for 10h. 1H-NMR spectral data of the
o
[BMIM]Cl were determined by using an AV-400 Bruker
literature. Therefore, research is critically needed to study catalyst 100 spectrometer and was found in good agreement with those
45 performance, solvent effects and process conditions for 5-HMF
production from glucose in high concentrations.
reported in the literature. 25
2.2 Viscosity measurement
In addition, efficient isolation of 5-HMF from the ionic liquid
solvents is also a critical part of the overall 5-HMFproduction
process. Separation of 5-HMF by distillation is not practical due
50 to the very low vapor pressure of 5-HMF. An alternative choice is
to extract 5-HMF formed in the reaction system into the organic
extraction phase. Although a few biphasic systems involving an
ionic liquid reaction phase and an organic extraction phase have
been reported,22-24 there are still several issues to be addressed.
55 First, these studies were limited to comparing the glucose
conversions and 5-HMF yields between the biphasic systems and
the monophase systems. Little attention was given to
systematically study the extraction performance, such as the
Capillary viscosities of [BMIM]Cl and its mixtures at
atmospheric pressure were measured using an GB/T265
105 Kinematic viscosity measuring unit (Dalian Zhineng Instrument
Group Co., Ltd., China) fitted with flat open Abramovich
Standard viscometers. The temperature of the thermostatic
glycerol bath was measured with a 100 Ω platinum resistance
thermometer with an overall estimated uncertainty of ±0.1 K. The
110 viscosity measurement for each sample was repeated three times
and the overall maximum uncertainty of these viscosity
measurements was estimated as ≤1%.
2.3 Extraction methods
2
|
Green Chem., [year], [vol], 00–00
This journal is © The Royal Society of Chemistry [year]

View Article Online
DOI: 10.1039/C5GC01039J
Page 3 of 11
Green Chemistry
To establish the extraction parameters, model compounds and
(2 )
Phase volume ratio  V e / V f
solvents were used in the absence of a catalyst. For each
extraction sample, a specified amount of [BMIM]Cl (preheated at
100 oC and kept in liquid state before dispensing), water, 5-HMF,
glucose and organic solvent were added into a glass vial with
scale lines. The glass vial was sealed with a cap and its volume
was calibrated. Chemical dispense was conducted on a Core
Module 3 (CM3) high throughput instrument (Freeslate Co.,
USA), a bench-top platform system that performs complex
Where Ve and Vf stand for the volumes of the extraction phase
60 and the raffinate phase, respectively.
The extraction efficiency was the ratio of the 5-HMF
partitioned in the extraction phase and the total 5-HMF in the
biphasic system. It was calculated according to the equation (3):
5
Extraction
efficiency
 C e V e /( C e V e  C f V f )  100 %
(3 )
65
The saturated solubility (mg/g) of a organic solvent in
[BMIM]Cl at a given temperature was conducted as follow
equation (4):
10 automated liquid and powder weighing, mixture stirring and
temperature monitoring. The dispense accuracies of the powder
and the liquid are 0.5 mg and 0.5 µl, respectively. The extraction
o
experiments were operated at room temperature, 24~26 C. The
(4)
Where msample is the mass (mg) of the sample collected from the
Saturated
solubility
 m solute /( m sample  m solute )  1000
compounds in the vial were mixed by sufficient shaking for 30
15 min, and at last the vial was settled in the static stage for phase
separation for 3 h. After several experiments and taking samples
at different time intervals, it was found that increasing the
agitation time (minimum time 30 min) and the rest time
(minimum time 3 h) had no effect on equilibrium phase
20 compositions. After complete phase segregation, the volumes of
the top and bottom phases were measured and the samples from
the two phases were collected. Then glycerol as an internal
analytical standard was added to the samples, and the mixtures
were analyzed by HPLC. A mass balance was made between the
25 initial mass of 5-HMF and the amounts in the extraction phase
and the raffinate phase in equilibrium. The relative error in the
mass balance was within 2 %.
The saturated solubilities of the organic solvents in [BMIM]Cl
at a given temperature were conducted as follows: 2 g [BMIM]Cl
30 and 1 ml organic solvent (ensuring excessive amount for
dissolution) were loaded in a 4 ml vial by CM3. And then the
mixture was sealed, put on the CM3 heating bath, stirred and
heated to a specified temperature for 1 h, during which the
mixture was shaken manually at 5 min intervals. Finally, the vial
35 was settled in the static stage on the CM3 heating bath for phase
separation for 3 h. After that the sample was collected from the
raffinate phase for analysis by HPLC.
70 raffinate phase and msolute is the mass (mg) of the organic solvent
contained in the sample.
2.4 Measurement of ternary phase diagram
The coexistence curve values were determined by means of the
cloud point method. The measurement was made in a glass
75 conical flask equipped with a glass stopper at room temperature,
o
24–26 C. A known composition, homogeneous and transparent,
of ([BMIM]Cl + water) was prepared in a glass conical flask. The
mixture was recorded by mass using a Mettler electronic balance
(ML204, accurate to 0.0001 g). Then GDE was progressively
80 added to the mixture using a micro-burette. The cloud point was
determined by observing the transition from a homogeneous to a
heterogeneous mixture as indicated by the mixture turbidity. The
added GDE was also recorded by mass using a balance. The mass
distributions of the three components were calculated for the
85 binodal curve measurement. The procedure was repeated by
alteration of [BMIM]Cl and water mass ratio. Repeated
experiments showed that the maximum deviations from the
average value were less than 0.5%.
The tie line for experimental results was determined by using a
90 halogen moisture analyzer (Mettler Toledo V20 Compact
Volumetric KF Titrator). The mixtures were prepared inside a
glass vial and were vigorously agitated for 30 min at least and left
to rest for 3 h or longer. After the necessary rest time, the mixture
splits into two clear and transparent liquid phases with a well
95 defined interface. Next, the extraction phase and the reffinate
phase were separated and weighted. The extraction phase mixture
was sampled and its water content was examined on the halogen
moisture analyzer. The water amount of the raffinate phase was
calculated by subtracting the water in the extraction phase from
100 the total water initially added in the mixture. The tie line values
were obtained by locating the two dots on the coexistence curve
with the given water contents.
All samples collected from the extraction phase and the
raffinate phase were analyzed on an Agilent 1260 HPLC
40 equipped with a refractive index detector (RID) and a diode array
detector (DAD). Components were separated with a Hi – Piex H
(300×7.7 mm) column and with H2SO4 (5 mM) eluent at a flow
rate of 0.6 ml·min-1 ， and the column temperature was
o
maintained at 65 C. Glycerol was added as an internal standard
45 for quantitative calculations.
The characteristics of the dispersions of solutes in both phases
were determined by the following parameters:26,27
The partition coefficient was determined as the ratio of the
equilibrium concentrations of the partitioned solute in the
50 extraction phase and the raffinate phase. It was calculated
according to the equation (1):
2.5 Catalytic reactions
All reactions were carried out on the CM3 high throughput
105 instrument. A test protocol consisting of the following steps was
used. First, each 4 ml vial was dispensed with given amounts of
[BMIM]Cl (in liquid, preheated at 100 oC), CrCl3.6H2O (in
powder), glucose, water and organic solvents automatically by a
robotic arm of the CM3. Then the vials were sealed with caps.
110 Finally, the bottom parts (one third of the vial height) of the vials
were heated up on the CM3 heating board (temperature resolution,
Partition
coefficien t  C e / C f
(1 )
Where Ce and Cf are equilibrium concentrations of the partitioned
solute in the extraction phase and the raffinate phase, respectively.
The phase volume ratio was defined as the volume ratio of the
extraction phase to the raffinate phase. It was calculated
according to the equation (2):
55
This journal is © The Royal Society of Chemistry [year]
Journal Name, [year], [vol], 00–00 | 3

Green Chemistry
View ArticlPe Oanglinee 4 of 11
DOI: 10.1039/C5GC01039J
o
0.5 C) with a stirring at 300 RPM. When the bottom reaction
an order higher than one.28 Therefore, when the glucose
concentration is raised, the rate of humins formation is expected
to accelerate over that of 5-HMF formation. In addition, the low
o
temperature reached to 108 C (in about 15 min), reaction time
was started (Time = 0 min). The CM3 monitored the reaction
time, removed the vial away from the heating board and 60 5-HMF yield at high glucose concentration may be partly
5
quenched the reaction in a cool board immediately after reaching
the designed time. For the mixtures that generated two phases
after reaction, the volumes of the extraction phase and the
reaction phase (in bottom) were measured according to the scale
attributed to poor mass transfer due to increased viscosity. The
kinematic viscosity of [BMIM]Cl at 80 oC is 109.4 mm2/s.
Addition of 80 wt% glucose to [BMIM]Cl sharply increases the
o
kinematic viscosity of the [BMIM]Cl/glucose mixture at 80 C to
lines on the wall of the vail. Then, the two phases were separated 65 454.6 mm2/s, which is more than 4 times that of the pure
10 into two vails and analyzed individually. Glycerol was added as
an internal standard for quantitative calculations. The mixtures
with the addition of the internal standard were diluted by water,
filtered with a 0.2 mm syringe filter and analyzed by HPLC
(adopting the same procedure as described in the extraction
15 method).
[BMIM]Cl. The increased viscosity slows the desired catalytic
molecular reaction while favors the localized side reactions as the
rates of heat and mass transfers are largely dependent on the fluid
flow dynamics of the system.
70
The conversion of glucose, the selectivity and yield of the
products were the factors used to evaluate the performance of the
total reaction system. For those generated two phases after
reaction, besides the partition coefficient as defined in equation
20 (1), the 5-HMF yield in the extraction phase was also analyzed.
They are defined by the following equations. The data variation
in this work was no more than ± 2% (relative value).
75
mole of glucose
mole of glucose initially charged
mole of product produced
reacted
Conversion
(mol% ) 
 100 %
(5)
(6)
(7)
(8)
80
25
Yield (mol% ) 
 100 %
mole of glucose initially charged
Yield
Selectivit y (mol% ) 
 100 %
Conversion
85
5 - HMF yield in extraction
phase (mol%
)
30
Figure 1. Effect of glucose loading on CrCl3 catalyzed synthesis of 5-
HMF in [BMIM]Cl monophasic system. (Reaction conditions: 250 mg
[BMIM]Cl; molar ratio of CrCl3:glucose = 1:10; 108 oC; 60 min; 300
RPM).
mole of 5 - HMF in the extract phase

 100 %
mole of glucose initially
charged
90
3
Results and Discussion
3.1 Issues in the conversion of glucose in high concentration
35 Results in Figure 1 show that increasing the initial glucose
concentration has a significantly negative impact on the 5-HMF
yield in pure [BMIM]Cl solvent. The almost complete conversion
of glucose even at a loading as high as 100 wt% with respect to
[BMIM]Cl indicates that the CrCl3 catalyst was highly active.
40 However, the selectivity to 5-HMF showed a gradual decrease
with the increase of glucose loading. The yield of 5-HMF at 20
wt% glucose loading is 66.4 mol%, and it quickly decreases to
43.6 mol% when the glucose loading is doubled to 40 wt%.
Further increase in the glucose loading continues to lower the 5-
45 HMF yield. For instance, when the initial glucose loading is 80
wt%, the yield of 5-HMF is only 30.3 mol%.
The lower yield of 5-HMF at high glucose concentration may
be attributed to the higher rates of side-reactions to undesired by-
products. When the mixtures after reaction were diluted by water,
50 most of the humins were observed as insoluble solid that adhered
to the wall of the vial. 5-HMF is formed from glucose via two
steps, CrCl3 catalyzed isomerization of glucose to fructose and
acid catalyzed dehydration of fructose to 5-HMF. The preferred
reaction pathway to 5-HMF is a unimolecular transformation in
55 each of the steps. It has been postulated that 5-HMF is formed via
a first order reaction, while humins are formed via reactions with
95
100
105 Figure 2. Effect of the reaction time on the glucose conversion and
product selectivity under high glucose concentration in
a
[BMIM]Cl/CrCl3 monophasic system. (Reaction conditions: 250 mg
[BMIM]Cl; 200 mg glucose; molar ratio of CrCl3:glucose = 1:10; 108 oC;
300 RPM).
110
The influence of reaction time on glucose conversion and
product selectivity at 80wt% glucose loading was investigated
and the results are shown in Figure 2. The conversion of glucose
quickly reaches 76.4 mol% in 15 min, but the selectivity to 5-
4
|
Green Chem., [year], [vol], 00–00
This journal is © The Royal Society of Chemistry [year]

View Article Online
DOI: 10.1039/C5GC01039J
Page 5 of 11
Green Chemistry
HMF is only 36.0 mol%. The results suggest that oligomerization
dominates from the very beginning of the reaction with the high
reaction phase, simulating a scenario of 75 mol% glucose
conversion into 5-HMF and H2O based on 80 wt% loading with
glucose concentration. Further increase in the reaction time 60 respect to [BMIM]Cl. The partition coefficient and the extraction
results in further-increased conversion of glucose, but also
resulted in concomitant decrease in 5-HMF selectivity. In this
reaction the primary by-products is humins, along with glucose
dimmers, anhydroglucose and trance amount of fructose. The
efficiency are plotted to manifest the amount of extractable 5-
HMF from [BMIM]Cl. The results in Figure 3 show that the
extraction performances of the solvents vary widely even if
among the same type. Among the ethers, GDE shows the best
5
mechanism of humins formation is still not fully understood, but 65 extraction performance, as indicated by the high extraction
it has been proposed that humins are formed due to
efficiency and the high partition coefficient, 37.9% and 0.28,
respectively. THF, another ether molecule, also performs well,
even though a little less effective than GDE in extracting 5-HMF.
The rest of the evaluated ether molecules, butyl ether and MIBE,
10 oligomerization of glucose, fructose (an intermediate in the
conversion of glucose to 5-HMF), 5-HMF, and/or other reaction
intermediates.2,6,29,30 At high glucose concentration, bimolecular
and sequential oligomerization are favored, so side-reactions to 70 show much poorer extraction performances. The extraction
form oligomers and humins become serious. The detected
15 glucose dimmers manifested that some of the glucose has
condensed into glycopolymers, thus further reaction can result in
excessive humins formation.
efficiency of the ketones follows the order of hacac >
cyclohexone > MIBK. EtOAc as an ester shows moderate
extraction performance. n-butanol is fully miscible with the
[BMIM]Cl solvent. The hydrogen bonding between the oxygen in
Even though the [BMIM]Cl/CrCl3 system is known to be 75 the extraction solvents and the proton in the hydroxyl group of 5-
efficient in converting glucose to 5-HMF at 10 wt% glucose
20 loading, the 5-HMF yield is quite low at a high glucose
concentration. One strategy to enhance the 5-HMF yield at a high
glucose concentration is by constructing a [BMIM]Cl-organics-
HMF appears important; the extraction efficiency of a solvent
largely dependent on its capability of accepting protons. GDE
with a higher O/C ratio can theoretically provide more hydrogen-
bond-acceptor sites for 5-HMF, so it exhibits stronger extraction
water ternary biphasic system to facilitate the continuous 80 than MIBE or butyl ether, both of which have smaller O/C ratios
recovery of 5-HMF to an extraction phase during the reaction.
in their molecules. Hacac and MIBK follow similar patterns, that
is, hacac has a higher O/C ratio and hence shows higher
extraction efficiency as compared to MIBK. However, although
THF and cyclohexone have relatively low O/C ratios, as cyclic
85 compounds they both exhibit strong extraction efficiency. The
cyclic structures of the two organics may be favorable in
exposing the O sites for hydrogen bonding with the hydroxyl
proton of the 5-HMF molecule. n-butanol and isobutanol have
been used as phase modifier to promote production of 5-HMF
90 from fructose in a two-phase system involving an aqueous phase
and an organic phase.6,7,31 However, these butanols are not
suitable extractant, since they form homogenous phase with the
[BMIM]Cl solvent. According to the results above, some
organics such as GDE, THF, hacac and cyclohexone show good
95 5-HMFextraction efficiencies from the model reaction phase.
25 3.2 Effect of organics
3.2.1 Organics as potential extractants
30
35
40
3.2.2 Effect of organics on 5-HMF yield
The organics discussed above have been combined with
[BMIM]Cl to form functional solvent systems for the CrCl3
catalyzed conversion of glucose in high concentration to 5-HMF,
100 as shown in Figure 4. The results in Fig 4 (a) indicate that all the
CrCl3 catalyzed systems have catalytic activities in converting
glucose, but the 5-HMF yield varies in a wide range depending
on the organics used. A high 5-HMF yield of 64.5 mol% was
achieved in the GDE/[BMIM]Cl solvent system. This yield
Figure 3. The partition coefficients and extraction efficiencies of 5-HMF
45 in different organics / [BMIM]Cl systems(GDE—glycol dimethyl ether;
THF—Tetrahydrofuran; MIBE—methyl tertiary butyl ether; Hacac—
acetylacetone; MIBK—Methyl isobutyl ketone; EtOAc—Ethyl acetate)
(Extraction conditions: 500 mg [BMIM]Cl; 2 ml extractant; 210 mg 5- 105 nearly doubled that obtained in the monophasic [BMIM]Cl/CrCl3
HMF, 100 mg glucose; 90 mg H2O; 25 oC.)
system. Besides the GDE/[BMIM]Cl solvent system, the use of
THF, MIBK and MIBE over the glucose/[BMIM]Cl/CrCl3
reaction phase also resulted in higher 5-HMF yield than the single
50
In a biphasic reaction system, the organic phase over
[BMIM]Cl reaction phase is expected to act as an extraction
glucose/[BMIM]Cl/CrCl3
reaction
phase.
The
other
phase for continuous transfer of 5-HMF into the organic phase 110 organic/[BMIM]Cl solvent systems are close to or even less
after its formation in the reaction phase. The extraction behaviors
effective than the single glucose/[BMIM]Cl/CrCl3 system.
Although hacac and cyclohexone exhibit good extraction
behaviors in the model systems as shown in Figure 3, little or a
very low 5-HMF yield was obtained from the reaction systems
55 of the organics are of great importance. In this section, a series of
organic solvents, including ethers, ketones, ester and alcohol, are
evaluated as extractants for extracting 5-HMF from the model
This journal is © The Royal Society of Chemistry [year]
Journal Name, [year], [vol], 00–00 | 5

Green Chemistry
View ArticlPe Oanglinee 6 of 11
DOI: 10.1039/C5GC01039J
contained them. These two types of ketones were found to carry
enhanced 5-HMF yield are well correlated. As expected, we
out Aldol reactions32 with glucose and 5-HMF in the 60 found that almost 56% of the produced 5-HMF was transferred
[BMIM]Cl/CrCl3 catalytic system. Addition of n-butanol not only
cannot form two phase with [BMIM]Cl solvent but also
suppresses 5-HMF yield. These results manifests that the real
reaction systems are more multi-factorial and complicated.
into the GDE phase due to the excellent GDE extraction
efficiency. Moreover, the 5-HMF concentration in the top GDE
phase of the GDE/[BMIM]Cl solvent system reaches 0.215
mmol/ml, which is higher than those reported for glucose
65 dehydration in MIBK/water biphasic systems11,33 and in
THF/water biphasic system,34 which only reached a low 5-HMF
concentration of about 0.05 mmol/ml at the optimized conditions.
Because 5-HMF in the extraction phase can be recovered by
distillation of the extraction solvent, the much increased 5-HMF
70 concentration in the extraction phase can be expected to result in
significant energy saving in the production of 5-HMF.
It was reported that the stability of 5-HMF was reduced in
[BMIM]Cl/CrCl3 system when glucose was introduced.29 The
loss of 5-HMF due to humins formation from its oligomerization
75 with glucose in the current glulocse/[BMIM]Cl/CrCl3 system is
expected to occur in a significant extent at the high glucose
concentration. The organics that efficiently extract 5-HMF away
from the [BMIM]Cl solvent to the top extraction phase
effectively protect 5-HMF from humins formation with high
80 concentration glucose and also promote the positive reaction, thus
provide a pathway for 5-HMF yield in the current systems.
5
10
15
20
25
30
35
3.2.3 Properties of the organics saturated [BMIM]Cl and
effects of the organics on glucose conversion
Table 1 Relevant physical properties of the organics and [BMIM]Cl
85 containing dissolved organics.
Items
GDE
THF
204
MIBK MIBE
Saturated solubility of solute in [BMIM]Cl 202
at 80 oC (mg/g)
103
10
Boiling point of the solute (oC)
83
65
116
55
Bubbling state of [BMIM]Cl with dissolved Strongly Strongly None Weakly
solute at 108 oC
Kinematic viscosity of [BMIM]Cl with
dissolved solute at 80 oC (mm2/s)
26
29
52
105
Based on the results in Figure 4, GDE, THF, MIBK and MIBE
are potential extractants. These organics have different boiling
points and different saturated solubilities in [BMIM]Cl, as listed
90 in Table 1. Careful observation of the biphasic systems revealed
that abundant bubbles rose up from the ionic liquid phase to the
organic phase in the GDE/[BMIM]Cl and THF/[BMIM]Cl
Figure 4. Synthesis of 5-HMF from high-concentration glucose in
40 [BMIM]Cl catalyzed by CrCl3 in the presence of different organics, of
which GDE, THF, MIBE and MIBK form separate extraction phases with
[BMIM]Cl and hacac, cyclohexone, EtOAc and n-butanol
form
homogeneous solutions with [BMIM]Cl. (a) Glucose conversion, 5-HMF
selectivity and yield in the total system; (b) 5-HMF concentration and
45 yield in the extraction phase ( Reaction conditions: 250 mg [BMIM]Cl; 2
ml solvent; 200 mg glucose; 80 mg water; molar ratio of CrCl3:glucose =
1:10; 108 oC; 60 min; 300 RPM ).
o
systems during reaction in the CM3 heating bath at 108 C. The
bubbling phenomenon in the MIBE/[BMIM]Cl system was very
95 weak, but no bubbling was observed in the MIBK/[BMIM]Cl
system. The heated reaction portion of the vial is about 1/3 of the
vial height. The temperature in this portion (the reaction phase) is
o
Among the organics studied, it was found that only four
50 organic solvents can form two phase with the [BMIM]Cl solvent,
as shown in Figure 4 (b). It is remarkable that the amount of 5-
HMF in the top GDE phase of the GDE/[BMIM]Cl biphasic
system corresponds to a 5-HMF yield of 36.1 mol%. The amount
of 5-HMF in other organics/[BMIM]Cl biphasic systems follows
55 the order of THF > MIBK > MIBE in the top phase. This order of
5-HMF yield among the organics in the top phase (shown in
Figure 4 (b)) agrees with that in the whole system (shown in
Figure 4 (a)), indicating that the extraction performance and the
108 C. The extraction phase occupies the remaining 2/3 height
of the vial and the temperature in the top part of the vial is about
100 88 oC. Therefore, a temperature gradient exists between the
reaction phase and the extraction phase in the vial. The wall of
the vial and the rotor provide the nucleation sites. The boiling
points of GDE, THF and MIBE are lower than the temperature
o
(108 C) in the reaction phase, while that of MIBK is above 108
105 oC. Because their boiling points are lower than the reaction
temperature, during the reaction, the GDE, THF and MIBE
partitioned in the [BMIM]Cl phase are dynamically gasified from
6
|
Green Chem., [year], [vol], 00–00
This journal is © The Royal Society of Chemistry [year]

View Article Online
DOI: 10.1039/C5GC01039J
Page 7 of 11
Green Chemistry
the overheated reaction portion of the vial and are condensed in
In addition, when [BMIM]Cl is used as the reaction medium
the extraction phase below the dew point of the vapor. MIBE 60 for the conversion of glucose in high-concentrationss, the high
with low solubility in [BMIM]Cl generated little bubbling. MIBK
partitioned in [BMIM]Cl shows no bubbling at 108 oC due to its
higher boiling point. According to the reflux phenomena, it can
be established that the organics dissolved in [BMIM]Cl,
viscosity of the [BMIM]Cl/ CrCl3/glucose system is expected to
negatively impact on heat and mass transfers, especially at high
glucose concentration. The dissolved organics help to reduce
[BMIM]Cl viscosity to a varying degree. As shown in Table 1,
5
especially GDE and THF with higher solubilities and lower 65 the saturated solubility of GDE in [BMIM]Cl at 80 oC is 202
boling points, exhibit dynamic equilibrium between the two
phase by constant gasification from and dissolution in the
10 [BMIM]Cl phase. The bubbling is beneficial to facilitate mixing.
The organic layer of the biphasic system acts as an extraction
mg/g, and the corresponding kinematic viscosity of [BMIM]Cl
o
with dissolved GDE at 80 C is 26 mm2/s, which is nearly one-
quarter the value of pure [BMIM]Cl (109.4 mm2/s). THF showed
a similarly pronounced decrease in viscosity as GDE. In
phase for continuous removal of 5-HMF from the reaction phase. 70 comparison, MIBK and MIBE are much less effective in
The bubbles increase the interface area and also the bubbling
dynamically enhances the turbulence of the mixture; both are
15 beneficial for the transfer of 5-HMF from the reaction phase to
the extraction phase. In addition, the low boiling points of GDE
lowering the viscosity due to their lower saturated solubility in
[BMIM]Cl. According to the results in Table 1, the decrease in
viscosity due to saturated dissolution of the organic extractants in
[BMIM]Cl is well correlated with the solubility of the organics in
and THF can allow significant energy saving in the recovery of 5- 75 [BMIM]Cl. The reduced viscosity of the mediated [BMIM]Cl
HMF by distillation.
solvent due to the dissolved organics is expected to effectively
facilitate the mass and heat transfers of the reaction system. It is
important to note that, although [BMIM]Cl is able to dissolve the
organics to a varying degree, the organic phase in excess of the
80 soluble amount in [BMIM]Cl does not show [BMIM]Cl
contamination. In the real biphasic reaction systems as listed in
Figure 4, [BMIM]Cl and glucose weren’t detected by HPLC in
the samples from the extraction phase of the four biphasic
systems. Freeing of [BMIM]Cl contamination in the extraction
85 phase is a highly desirable feature of the current biphasic systems
that show largely enhanced 5-HMF yield and make it feasible to
recover 5-HMF in high-purity.
20
25
30
It has been reported that addition of some organics in aqueous
solvent greatly reduces the solubility of sugars.35,36 Since
90 [BMIM]Cl can dissolve GDE up to their solubility as shown in
Table 1, it is necessary to determine the impact of the dissolved
GDE on glucose solubility in [BMIM]Cl, especially to maintain a
high glucose concentration. The concentration of glucose in the
35 Figure 5. Influence of bubbling and stirring on the glucose conversion
and product selectivity under high glucose concentration in
o
a
GDE saturated [BMIM]Cl at 80 C was determined to be above
GDE/[BMIM]Cl biphasic system. (Reaction conditions: 250 mg 95 1.0g/g [BMIM]Cl. A transparent and homogeneous liquid was
[BMIM]Cl; 2 ml GDE; 200 mg glucose; 80 mg water; molar ratio of
CrCl3:glucose = 1:10; 108 oC; 300 RPM).
prepared even when 1.1g glucose was mixed with 1g [BMIM]Cl
saturated with GDE. Therefore, glucose solubility in GDE
saturated [BMIM]Cl solvent is not a limiting factor. Importantly,
it was found that the organics like GDE dissolved in [BMIM]Cl
40
In order to quantify the contribution of bubbling to 5-HMF
production, special reaction conditions with or without bubbling 100 can reduce insoluble humins formation. After reaction at the
and stirring were studied at a high glucose concentration (Figure
conditions as described in Figure 4, no apparent insoluble
substances were observed in the mixture from the
GDE/[BMIM]Cl/CrCl3 catalytic system when it was diluted by
water. Similar phenomena have been reported with the addition
5). ―Reaction 1‖ was carried out in an incubator at 108 oC. There
45 was no temperature gradient between the extraction phase and the
reaction phase in the vial of Reaction 1. No bubbling
phenomenon of bubbling was observed during the reaction period. 105 of DMSO37 and acetone35 in the aqueous reactions. The organics
―Reaction 2‖ and ―Reaction 3‖ were carried out on the CM3
heating board with temperature gradient in the vials as described
50 above. The experimental conditions of ―Reaction 2‖ were similar
to those of ―Reaction 3‖ except that the former was not stirred.
As shown in Figure 5, the 5-HMF yield of Reaction 2 is about
triple that of Reaction 1, which means that bubbling plays a
critical role in improving reaction performance. Stirring also has
55 positive effect since the 5-HMF yield of Reaction 3 is about 10
mol% higher than that of Reaction 2. It is most remarkable that
internal GDE buddling in the reaction phase is much more
effective than mechanic stirring for improving the 5-HMF yield.
reduced humins formation by protecting the hexose from
oligomerization and thereby increasing the selectivity to 5-HMF.
3.3 Effect of water addition to the biphasic system
3.3.1 Effect of water loading on reactions
110 The influence of water on the catalytic performance of the
GDE/[BMIM]Cl biphasic reaction systems with high glucose
concentration was investigated and the results are shown in
Figure 6. The conversion of glucose shows a small decrease with
the increase of water from no added water to 6:1 mole ratio of
115 water:glucose. The selectivity to 5-HMF in the biphasic system
This journal is © The Royal Society of Chemistry [year]
Journal Name, [year], [vol], 00–00 | 7

Green Chemistry
View ArticlPe Oanglinee 8 of 11
DOI: 10.1039/C5GC01039J
without the addition of water is 55.5 mol%, and it gradually
3.3.2 Ternary phase diagram of [BMIM]Cl/water/GDE
approaches 68.0 mol% when the reaction is conducted at 5:1
mole ratio of water:glucose, above which the selectivity 5-HMF
shows a slight decrease. The trend of 5-HMF yield with the added
water follows a similar pattern as that of 5-HMF selectivity. The
maximum 5-HMF yield was obtained with H2O:glucose molar
ratio of 4:1 for the reaction. As to the extraction performance
(Figure 6 b), the 5-HMF concentration in the GDE phase does not
reach its maximum plateau until at H2O:glucose ratio of 4:1.
10 However, with the increase in the water loading in the range of
0:1 to 6:1 H2O:glucose molar ratio, the amount of 5-HMF in the
extraction phase exhibits a volcano trend with a maximal value
achieved at 4:1 molar ratio, corresponding to a 5-HMF yield of
36.1 mol% in the GDE phase. The amount of extracted 5-HMF
15 corresponds to about 56% of the total 5-HMF produced in the
biphasic system. The above results manifested that appropriate
amount of added water in the reaction system is significantly
beneficial for promoting the 5-HMF yield of both the total system
and its partitioning in the extraction phase.
60
65
70
75
80
5
Figure 7. A ternary phase diagram of the [BMIM]Cl/water/GDE system
with tie lines at room temperature.(The original compositions of GDE
(ml) : [BMIM]Cl (mg) : water (mg) for establishing tie line 1, 2 , 3 and 4
are 2:500:40, 2:500:80, 2:500:160 and 2:500:250, respectively).
20
Figure
7
shows the ternary phase diagram of the
[BMIM]Cl/water/GDE system. Water and GDE are miscible with
each other at any proportion, and so are water and [BMIM]Cl.
While GDE and [BMIM]Cl form two phases, a pure GDE phase
25
30
35
40
45
50
85 and a [BMIM]Cl dominant phase containing17 wt% GDE.
Addition of water increases the miscibility of [BMIM]Cl and
GDE. The tie lines 1, 2, 3 and 4 were obtained via gradual
increase of water in the biphasic systems. The equilibrium
compositions of the two phases in each tie line are provided in the
90 endpoint positions of the tie line. It was found that the GDE
dominant extraction phases of tie line 1 to 3 are located on the
water-GDE side line and are close to the GDE point in the ternary
phase diagram. The GDE dominant extraction phase of tie line 4
contains a small amount of [BMIM]Cl. The [BMIM]Cl dominant
95 raffinate phases of tie line 1 to 4 are shifted from containing a
small amount to a large amount of water and GDE with the
increasing water content.
The [BMIM]Cl/water/GDE phase behavior is very beneficial
for the GDE/[BMIM]Cl biphasic reaction system. In particular,
100 addition of water up to 32 wt% of [BMIM]Cl, do not result in the
contamination of the GDE extraction phase by [BMIM]Cl. This
behavior was also confirmed by the absence of [BMIM]Cl peak
in the chromatographic analysis of the extraction phase of the
GDE/[BMIM]Cl biphasic reaction system originally loaded with
105 glucose:water in 1:4 molar ratio. In addition, the added water is
mainly retained in the [BMIM]Cl dominant raffinate phase. This
behavior is particularly beneficial because water can further
lower the viscosity of the [BMIM]Cl/CrCl3/glucose mixture, in
addition to GDE. For instance, the kinematic viscosity of the
110 water/[BMIM]Cl mixture (water content is 32 wt% of [BMIM]Cl)
at 80 oC is only 4.0 mm2/s. , It is also important to emphasize that
water also promotes the phase transfer of 5-HMF from the
[BMIM]Cl phase to the extraction phase. Under the experimental
conditions (with added water) described in Figure 3, the partition
115 coefficient and the extraction efficiency of GDE for 5-HMF from
[BMIM]Cl are 0.281 and 37.9%, respectively. However, the two
Figure 6. Effect of water on CrCl3 catalyzed synthesis of 5-HMF from
glucose in high-concentration in GDE/[BMIM]Cl biphasic systems. (a)
Glucose conversion, 5-HMF selectivity and yield in the total system with
55 varying molar ratio of water to glucose; (b) 5-HMF concentration and
yield in the GDE phase with varying molar ratio of water to glucose
(Reaction conditions: 250 mg [BMIM]Cl; 2 ml GDE; 200 mg glucose;
molar ratio of CrCl3:glucose = 1:10; 108 oC; 60 min; 300 RPM ).
8
|
Green Chem., [year], [vol], 00–00
This journal is © The Royal Society of Chemistry [year]

View Article Online
DOI: 10.1039/C5GC01039J
Page 9 of 11
Green Chemistry
values fall to 0.125 and 21.4%, respectively, with no added water.
completely consumed after 90 min. Similar to the monophasic
The significant difference between the values with or without 60 reactions in Figure 2, appreciable amount of glucose dimmers and
water demonstrates the positive role of water in promoting 5-
HMF extraction. The possible promoting mechanism by water is
rationalized as follows: In the [BMIMCl]/5-HMF mixture, 5-
HMF is strongly associated with [BMIM]Cl mainly through
anhydroglucose form from the beginning of the reaction. A small
amount of fructose can be detected. Further increase in the
reaction time results in gradual decrease of the byproducts.
However, unlike the monophasic reaction in which the 5-HMF
5
hydrogen bonding between the Cl- anion and the hydroxyl proton 65 stops growing in 15min even as the remaining
of 5-HMF. Added water can also form hydrogen bonding with
[BMIM]Cl. Water to form strong and abundant hydrogen
10 bonding with [BMIM]Cl effectively weakens the 5-HMF-
[BMIM]Cl interaction force. As a result, 5-HMF is liberated from
glucose continues to be converted, in the biphasic reaction system
the selectivity to 5-HMF does not reach its maximum until 60
min. Further reaction beyond 60 min is not beneficial as the
conversion of glucose barely increases and the selectivity of 5-
the [BMIM]Cl phase, resulting in the enhanced extraction 70 HMF starts to decrease.
efficiency by the organic extractant phase.
Figure 9 shows the effect of the catalyst loading on 5-HMF
synthesis from 80 wt% glucose in the GDE/[BMIM]Cl biphasic
systems. The total yield of 5-HMF at 1:100 catalyst loading was
only 36.8 mol% due to the low glucose conversion (82.3 mol%)
75 and the poor selectivity to 5-HMF (44.7 mol%). With the increase
in catalyst loading in the range of 1:100 to 20:100, the selectivity
toward 5-HMF shows a volcano trend with a maximal value of
66.4 mol% achieved when the catalyst loading is at 10:100 CrCl3 :
glucose, above which the selectivity slightly decreases. Because
80 both of the two tandem reactions in glucose conversion to 5-HMF,
which include an isomerization step and a dehydration step,
depend on CrCl3 catalyst, a larger amount of catalyst loading
favors simultaneous conversion of each step. But too much
catalyst may also catalyze undesired side reactions. With all these
85 considered, 10:100 is considered the optimal amount for the
catalyst loading.
In our experiments, we did not observe obvious reduction in
the efficiency of the reaction solvent and the catalyst in the first
re-use of the [BMIM]Cl/CrCl3 system (data not shown here)
90 without removing the humins from the first run. But successive
cycles can be expected to reduce the glucose solubility, and
therefore the efficiency of the reaction system due to
accumulation of humins. It will be a subject of continued study to
evaluate the methods of humins separation from this system.
3.4 Effects of reaction time and catalyst loading
15
20
25
30
35
40
45
50
55
Figure 8. Effect of reaction time on CrCl3 catalyzed synthesis of 5-HMF
from 80 wt% glucose in the GDE/[BMIM]Cl biphasic system. (Reaction
conditions: 250 mg [BMIM]Cl; 2 ml GDE; 200 mg glucose; 80 mg water;
CrCl3:glucose = 1:10; 108 oC; 300 RPM).
95 Conclusions
In summary, the enhanced 5-HMF yield from glucose in high
concentration and the efficient separation of the product have
been realized by constructing a [BMIM]Cl-organics-water ternary
biphasic system. Increased glucose loading in the single
100 [BMIM]Cl/CrCl3/glucose reaction phase led to increasingly
excessive formation of humins and serious decrease in 5-HMF
yield. Mediation of the ionic liquid property by suitable organics
and water efficiently facilitated the glucose conversion to 5-HMF
and the continuous recovery of 5-HMF to an extraction phase
105 during the reaction. GDE as a hydrogen-bond acceptor not only
exhibited excellent extraction performance for 5-HMF, but also
played an important role in mediating the [BMIM]Cl phase. Due
to its low boiling point and suitable solubility saturation point in
the ionic liquid, a sustained GDE bobbling phenomenon in the
110 ionic liquid phase was observed, which promoted the rate of
inter-phase mass-transfer of 5-HMF during the reaction. In
addition, GDE helped to reduce the viscosity of the [BMIM]Cl
phase. Water present in the [BMIM]Cl-organics-water ternary
biphasic systems is mainly retained in the [BMIM]Cl phase.
115 Appropriate amount of water in the solvent systems promoted the
Figure 9. Effect of CrCl3 catalyst loading on 5-HMF synthesis from 80
wt% glucose in the GDE/[BMIM]Cl biphasic systems. (Reaction
conditions: 250 mg [BMIM]Cl; 2 ml GDE; 200 mg glucose; 80 mg water;
108 oC; 60 min; 300 RPM).
Figure 8 shows the results obtained at different reaction time
for CrCl3 catalyzed conversion of 80 wt% glucose in the
GDE/[BMIM]Cl biphasic system. The glucose is almost
This journal is © The Royal Society of Chemistry [year]
Journal Name, [year], [vol], 00–00 | 9

View Article Online
Green Chemistry
DOI: 10.1039/C5GPC0a1g03e9J10 of 11
extraction efficiency of 5-HMF and further lowered the viscosity
14 B. Binder and R. T. Raines, J. Am. Chem. Soc., 2009, 131, 1979-1985.
of [BMIM]Cl/glucose. Particularly, [BMIM]Cl mediated with 45 15 H. Wang, T. Deng, Y. Wang, Y. Qi, X. Hou and Y. Zhu, Bioresour.
GDE and water still possess high capacity in dissolving glucose
to high concentrations. In a biphasic system containing GDE as
the extractant, [BMIM]Cl with partitioned GDE and an
appropriate amount of water and CrCl3 as the reaction medium,
an unprecedented 5-HMF yield of 64.5 mol% was achieved from
glucose in high concentration (80 wt% with respect to the ionic
liquid). This yield nearly doubled that obtained in the single
Technol., 2013,136, 394–400.
16 A. Stark, Energy Environ. Sci., 2011,4, 19-32.
17 Z.C. Zhang, Catalytic transformation of carbohydrates and lignin in
ionic liquids. Wiley Interdisciplinary Reviews: Energy and
Environment., 2013, 2, 655–672.
5
50
18 H. Zhao, J.E. Holladay, H. Brown and Z.C. Zhang, Science, 2007,
316,1596–1600.
10 [BMIM]Cl/CrCl3/glucose reaction phase. Remarkably, the GDE
phase contained about 56% of the generated 5-HMF without
detectable contamination by the ionic liquid or carbohydrates.
19 G. Yong, Y. Zhang and J.Y. Ying, Angew. Chem., Int. Ed., 2008, 47,
9345–9348.
55 20 J. Wang, J. Ren, X. Liu, J. Xi, Q. Xia, Y. Zu, G. Lu and Y. Wang,
Green Chem., 2012, 14, 2506–2512.
21 N. Shi, Q. Liu, Q. Zhang, T. Wang and L. Ma, Green Chem., 2013,
15, 1967–1974.
Acknowledgements
We acknowledge financial support from Chinese Central
15 Government "Thousand Talent Program" Funding and National
Natural Science Foundation of China (Grant No. G 21203015).
22 J.Y.G. Chan and Y. Zhang, ChemSusChem, 2009, 2, 731–734.
60 23 S. Lima, P. Neves, M.M. Antunes, M. Pillinger, N. Ignatyev and
A.A.Valente, Appl. Catal. A: Gen., 2009, 363, 93–99.
24 F. Tao, H. Song and L. Chou, J. Mol. Catal. A Chem., 2012, 357,11–
18.
References
25 T. Erdmenger, J. Vitz, F, Wiesbrock and U. S. Schubert, J. Mater.
1 T. Werpy and G. Peterson, Top Value Added Chemicals from
Biomass, Pacific Northwestern National Laboratory, 2004, vol. 1.
65
Chem., 2008, 18, 5267–5273.
26 Y. Pei, J. Wang, K. Wu, X. Xuan and X. Lu, Sep. Purif. Technol.,
20
2
R.J. van Putten, J. C. van der Waal, E. de Jong, C. B. Rasrendra, H. J.
2009, 64, 288–295.
Heeres and J. G. de Vries, Chem. Rev., 2013, 113, 1499–1597.
27 J. Gao, L. Chen and Z.C. Yan, Sep. Purif. Technol., 2014,124, 107–
3 T. Wang, M.W. Nolte and B.H. Shanks, Green Chem., 2014, 16,
548–572.
116.
70 28 B. Girisuta, L.P.B.M. Janssen and H.J. Heeres, Chem. Eng. Res. Des.,
2006, 85, 339–349.
4 B. Saha and M.M. Abu-Omar, Green Chem., 2014, 16, 24–38.
25 5 S.P. Teong, G. Yi and Y. Zhang, Green Chem., 2014, 16, 2015–2026.
6 Y. Román-Leshkov, J. N. Chheda and J. A. Dumesic, Science, 2006,
312, 1933–1937.
29 H. Li, W, Xu, T. Huang, S. Jia, Z. Xu, P. Yan, X. Liu, and Z. C.
Zhang, ACS Catal., 2014, 4 (12), 4446–4454.
30 J.B. Binder, A.V. Cefali, J.J. Blank and R. T. Raines, Energy Environ.
7 A.Torres, P. Daoutidis and M. Tsapatsis, Energy Environ. Sci., 2010,
3, 1560-1572.
75
Sci., 2010,3, 765-771.
31 Y. Roman-Leshkov, C. Barrett, Z. Liu and J. Dumesic, Nature, 2007,
30 8 D. Klemm, B. Heublein, H. Fink and A. Bohn, Angew. Chem. Int. Ed.,
2005, 44, 3358–3393.
47, 982–986.
32 F. Rodrigues, Y. Canac and A. Lubineau, Chem. Commun., 2000,
9 O.O. James, S. Maity, L.A. Usman, K.O. Ajanaku, O.O. Ajani, T.O.
Siyanbola, S. Sahu and R. Chaubey, Energy Environ. Sci., 2010, 3,
1833–1850.
2049–2050.
80 33 V.V. Ordomsky, J. van der Schaaf, J.C. Schouten, and T.A. Nijhuis,
ChemSusChem, 2013, 6, 1697 – 1707.
35 10 S. Jia, K. Liu, Z. Xu, P. Yan, W. Xu, X. Liu and Z.C. Zhang, Catal.
Today, 2014, 234, 83–90.
34 Y. Yang, C. Hu and M. M. Abu-Omar, Green Chem., 2012, 14, 509–
513
11 I. Jiménez-Morales, M. Moreno-Recio, J. Santamaría-González, P.
Maireles-Torres and A. Jiménez-López, Appl. Catal. B: Environ.,
2014, 154–155,190–196.
35 M. Bicker, J. Hirth and H. Vogel, Green Chem., 2003, 5, 280–284.
85 36 M. Bicker, D. Kaiser, L. Ott and H. Vogel, J. Supercrit. Fluids.,
2005, 36, 118–126.
40 12 D.A. Cantero, M.D. Bermejo and M.J. Cocero, Bioresour. Technol.,
2013, 135, 697–703.
37 R.M. Musau and R. M., Munavu, Biomass. 1987, 13, 67–74.
13 S. Xiao, B. Liu, Y. Wang, Z. Fang and Z. Zhang, Bioresour. Technol.,
2014,151, 361–366.
10
|
Green Chem., [year], [vol], 00–00
This journal is © The Royal Society of Chemistry [year]

Page 11 of 11
Green Chemistry
View Article Online
DOI: 10.1039/C5GC01039J
Graphical abstract
Ionic liquid-organics-water ternary biphasic system has high 5-HMF productivity
and separation efficiency in CrCl3 catalyzed conversion of glucose in high
concentration.