Direct catalytic conversion of glucose and cellulose

Article abstract of DOI:10.1039/c7gc03318d

Biomass product 5-hydroxymethylfurfural (5-HMF) can be used to synthesize a broad range of value added compounds currently derived from petroleum. Thus, the effective conversion of glucose or cellulose (the major components of biomass) into fuels and chemical commodities has been capturing increasing attention. Previous studies have been extensively focused on a two-step process for producing 5-HMF from glucose or cellulose, i.e., the isomerization of glucose into fructose and then the dehydration of fructose. We herein discovered that heterogeneous sulfonated poly(phenylene sulfide) (SPPS) containing strong Br?nsted acid sites is able to convert glucose and cellulose into 5-HMF with a high yield in ionic liquids (ILs). The optimal activity of glucose conversion to 5-HMF achieves a yield of 87.2% after 4 h reaction at 140 °C. For direct cellulose conversion, a 5-HMF yield of 68.2% can be achieved. The reaction mechanism over the SPPS catalyst in ILs was studied by DFT calculations, and the results indicated that the SO₃H group of SPPS plays a crucial role in glucose conversion into 5-HMF, and it acts as a proton donor as a Br?nsted acid and functions as a proton acceptor as the conjugate base. Furthermore, the anions and cations of ILs together with SO₃H-SPPS helped in stabilizing the reaction intermediates and transition states, which also resulted in glucose facile conversion into 5-HMF. The new catalyst system highlights new opportunities offered by optimizing the production of 5-HMF directly from glucose and cellulose.

Full text of DOI:10.1039/c7gc03318d

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: Z. Li, K. Su, J. Ren,
D. Yang, B. Cheng, C. K. Kim and X. Yao, Green Chem., 2018, DOI: 10.1039/C7GC03318D.
This is an Accepted Manuscript, which has been through the
Volume 18 Number
7 7 April 2016 Pages 1821–2242
Royal Society of Chemistry peer review process and has been
accepted for publication.
Green
Chemistry
Accepted Manuscripts are published online shortly after
Cutting-edge research for a greener sustainable future
www.rsc.org/greenchem
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
author guidelines.
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, outlined
in our author and reviewer resource centre, 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.
ISSN 1463-9262
CRITICAL REVIEW
G. Chatel et al.
Heterogeneous catalytic oxidation for lignin valorization into valuable
chemicals: what results? What limitations? What trends?
rsc.li/green-chem

Page 1 of 33
Green Chemistry
View Article Online
DOI: 10.1039/C7GC03318D
Direct catalytic conversion of glucose and cellulose
Zhenhuan Li1*, Kunmei Su1, Jun Ren2, Dongjiang Yang3, Bowen Cheng1, Chan
Kyung Kim2, Xiangdong Yao3, 4
*
1School of Materials Science and Engineering, The State Key Laboratory of Hollow Fiber
Membrane Materials and Processes, Tianjin Polytechnic University, 300160 Tianjin, China.
Email: lizhenhuan@tjpu.edu.cn
2 Department of Chemistry and Chemical Engineering, Center for Design and Applications of
Molecular Catalysts, Inha University, Incheon, 22212, Republic of Korea.
3Queensland Micro and Nanotechnology, Griffith University, Nathan Campus, Qld 4111,
Australia. Email: d.yang@griffith.edu.au;
4School of Natural Sciences, Griffith University, Nathan, QLD 4111, Australia. Email:
x.yao@griffith.edu.au
1

Green Chemistry
Page 2 of 33
View Article Online
DOI: 10.1039/C7GC03318D
Abstract
Biomass product 5ꢀhydroxymethylfurfural (5ꢀHMF) can be used to synthesize a
broad range of value added compounds currently derived from petroleum. Thus, the
effective conversion of glucose or cellulose (the major components of biomass) into
fuels and chemical commodities has been capturing increasing attentions. Previous
studies have been extensively focused on a twoꢀstep process for producing 5ꢀHMF
from glucose or cellulose, i.e., the isomerization of glucose into fructose and then the
dehydration of fructose. We herein discovered that heterogeneous sulfonated
poly(phenylene sulfide) (SPPS) containing strong Brønsted acid sites is able to
convert glucose and cellulose into 5ꢀHMF with a high yield in ionic liquids (ILs). The
optimal activity of glucose conversion to 5ꢀHMF achieves a yield of 87.2% after 4 h
reaction at 140 °C. For direct cellulose conversion, a 5ꢀHMF yield of 68.2% can be
achieved. The reaction mechanism over SPPS catalyst in ILs was studied by DFT
calculations, and results indicated that the SO3H group of SPPS plays a crucial role on
glucose conversion into 5ꢀHMF, and it acts as a proton donor as a Brønsted acid and
functions as a proton acceptor as the conjugate base. Furthermore, the anions and
cations of ILs together with SO3HꢀSPPS help to stabilize the reaction intermediates
and transition states, which also resulted in glucose facile conversion into 5ꢀHMF.
The new catalyst system highlights new opportunities offered by optimizing the
production of 5ꢀHMF directly from glucose and cellulose.
2

Page 3 of 33
Green Chemistry
View Article Online
DOI: 10.1039/C7GC03318D
Introduction
Concerns about global warming and energy security have led to the exploration of
alternatives to fossil hydrocarbon resources to supply chemicals and energy1. For
example, efficient routes are being explored for the conversion of carbohydrates, the
main constituents of renewable biomass, into fuels and chemicals2ꢀ3. 5ꢀ
hydroxymethylfurfural (5ꢀHMF) is considered to be a key biorenewable platform
molecule and is one of the ten most important platform chemicals listed by U.S.
Department of Energy, since 5ꢀHMF and its derivatives are potential substitutes for
petroleumꢀbased chemicals that are consumed in high volume3ꢀ10. Although 5ꢀ
HMF is already produced at an industrial level by AVA chemical from fructose in
high yield by using Brønsted and Lewis acid catalysts3,8, the selective transformation
of glucose into 5ꢀHMF, the dominant sugar in cellulosic biomass, remains a great
challenge since the stable pyranꢀtype six membered ring structure makes it hard to
dehydrate in the presence of acid catalysts11. At present, the conversion of glucose
into 5ꢀHMF is achieved in two steps. Glucose is first isomerized to fructose using
catalysts such as an enzyme, a Lewis acid or a base, and then the fructose is
dehydrated to yield 5ꢀHMF under acidic conditions6ꢀ10. Recently, unprecedented 5ꢀ
HMF yields were reported for glucose dehydration by chromium (II) chloride (CrCl2)
in the ionic liquid (IL) 1ꢀethylꢀ3ꢀmethylimidazolium chloride ([EMIM]Cl). The yield
ꢀ
of 5ꢀHMF was in the range of 60–70%8. The CrCl3 anion coordinates with an
3

Green Chemistry
Page 4 of 33
View Article Online
DOI: 10.1039/C7GC03318D
enediolate intermediate to promote hydrogen transfer and thus accelerates the
isomerisation of glucose to fructose8ꢀ9. The ionicꢀliquid medium was thought to
provide a nonꢀcoordinating polar environment to stabilize catalytically active lowꢀ
coordinate metal species12. However, it is hard to further enhance the 5ꢀHMF yield
due to the inevitable isomerisation process, and it is impossible to achieve the scale of
production (100ꢀ200 kTon per year) at low cost (~$1.00/kg) required for commercial
impact11. In addition, the homogeneous chromium based catalysts have serious
environmental and operational problems for the largeꢀscale production of 5ꢀHMF
from glucose. The high toxicity of Cr catalysts is incompatible with a green and
environmentally benign biomass conversion process, and the high cost and
inefficiency of separating homogeneous chromium catalysts from the products make
their recovery impractical.
If an efficient and environmentally friendly catalyst system can be invented for the
effective dehydration of glucose into 5ꢀHMF, 5ꢀHMF synthesis from glucose or
glucoseꢀbased carbohydrates can be an economic and sustainable strategy to replace
fossilꢀbased chemicals or energy. However, acid catalysts for fructose conversion into
5ꢀHMF via fructose dehydration to the furanose form13 usually exhibit very low
activity in glucose conversion into 5ꢀHMF6ꢀ10. Those acid catalysts include
homogeneous mineral acids such as sulphuric acid14 or hydrochloric acid and
heterogeneous acids such as strong acid cation exchange resins5,15ꢀ16, Hꢀform
zeolites17ꢀ18, and supported heteropolyacids19. To achieve economically feasible
production of 5ꢀHMF from glucose, a new efficient catalyst system is highly desired.
4

Page 5 of 33
Green Chemistry
View Article Online
DOI: 10.1039/C7GC03318D
Herein, we report that sulfonated poly(phenylene sulfide), abbreviated as SPPS, in
ionic liquids can catalyse the effective conversion of glucose and cellulose to 5ꢀHMF
at a high yield (87.2% and 68.2%, respectively). SPPS is chemically and thermally
stable and has been employed as the base polymer for protonꢀconducting
electrolytes20ꢀ21. The SO3H group of SPPS acts as a proton donor as a Brønsted acid
and functions as a proton acceptor as the conjugate base, and ILs help to stabilize the
reaction intermediates or transition states and promote glucose facile conversion into
5ꢀHMF22. Furthermore, SPPS catalyst is heterogeneous, much more easily separated
from the product and recycled than homogeneous catalysts. This new catalyst system
enables a novel process for large scale synthesis of 5ꢀHMF directly from glucose and
cellulose in the future.
Experimental
Materials. Nꢀethyl bromide and 1ꢀmethylimidazole (> 99%), SnCl4, TiCl4, CrCl2,
CrCl3, benzenesulfonic acid, leucine, tyrosine, tryptophan, proline, H2SO4,
chlorosulfonic acid, cellulose, cellobiose, sucrose, lactose, maltose, glucose, fructose,
galactose et al. were all obtained with AR degree from SigmaꢀAldrich. 5ꢀHMF used
as the standard sample was purchased from SigmaꢀAldrich.
Anhydrous 1ꢀmethylꢀ3ꢀethyl imidazolium bromide ([EMIM]Br) was synthesized
from ethyl bromide and 1ꢀmethylimidazole at 25 oC for 24 h. The resulting solid was
purified by acetonitrile and ethyl acetate to wipe off the residual ethyl chloride and 1ꢀ
5

Green Chemistry
Page 6 of 33
View Article Online
DOI: 10.1039/C7GC03318D
methylimidazole, and the purified [EMIM]Br was dried in vacuum drying oven at 70
oC for 24 h. Other ILs was synthesized by similar methods.
Sulfonated poly(thiophenylene) preparation. PPS was synthesized in a stainless
steel 5 L autoclave. The autoclave was equipped with a variable speed mechanical
stirrer, external heating chamber, cooling coils, N2 inlet and outlet. After 1, 4ꢀ
dichlorobenzene, Na2S, NMP and Na3PO4 were loaded into the reactor, the air in the
autoclave was flushed out using N2. The autoclave was heated to its target
temperature and maintained at this value for the required time, and the stirring speed
was maintained at 300 r/min. When the reaction was finished, the autoclave was
cooled with water, and PPS was washed by NMP, ethanol and water respectively.
The PPS molecular weight was characterized using HTGPC (PLꢀGPC220) at
Huatong Sichuan Research Center in China. Samples were injected in slurry form at
ambient temperature and reꢀdissolved by an inꢀline preꢀcolumn heater at 250 oC. A
viscometer consisting of a capillary tube with inlet and outlet taps connected to a
sensitive differential pressure transducer was used as the sole detector, with
deflections converted to a concentration using the column calibration. The columns
o
and viscometer were operated at 210 C, and 1ꢀchloronaphthalene was used as mobile
phase.
Sulfonation of PPS (Mw 2.69 ×104 g molꢀ1) was performed by the following
procedure. PPS was dissolved in 10 mL of 10% SO3ꢀH2SO4. The suspension was
o
heated to 80 C, and stirred for 3 h under a dry nitrogen atmosphere. After the
6

Page 7 of 33
Green Chemistry
View Article Online
DOI: 10.1039/C7GC03318D
reaction, the mixture was poured into 300 mL of distilled water to precipitate a white
powder. The powder was dried under reduced pressure.
Characterization of SPPS sulfonation degree. Titration Procedure: The IEC (Ionꢀ
Exchange Capacity) value for each SPPS was determined through the backꢀtitration
o
method. Sulfonated PPS in acid (H+) form was dried overnight at 100 C under
vacuum, weighed, and then immersed in saturated NaCl for 24 h. The amount of H+
ion released from the samples was determined by titration with 0.01 M NaOH using
phenol phthalein as an indicator. The IEC value of the SPPS was calculated with the
following equation.
CxV
IEC =
M
Where C and V are the concentration and volume of NaOH, respectively, and M is
the weight of SPPS.
XPS Procedure: XPS spectra were acquired using a Kratos Axis Ultra spectrometer,
equipped with a monochromatic Al KR source (hv = 1486.6 eV), operated at 225 W
(15 mA, 15 kV) and a base pressure of 1 ×109 Torr. Samples were degassed overnight
at 1× 108 Torr prior to analysis. Photoelectrons were counted with an 8ꢀchannel
hemispherical detector. Photoelectron pass energies were set at 160 eV for survey
spectra and 40 eV for high resolution spectra. Survey spectra are the average of two
sweeps from 600 to 4 eV, with a 0.5 eV step size and a 100 ms dwell time at each
step. Highꢀresolution S 2p spectra are the average of five sweeps, with a 0.1 eV step
7

Green Chemistry
Page 8 of 33
View Article Online
DOI: 10.1039/C7GC03318D
size and an 800 ms dwell time. Spectra were processed and deconvoluted using Casa
XPS version. A background subtraction was performed using the Shirley method.
Binding energies were calibrated to extrinsic/adventitious carbon at 284.8 eV. The
Marquardt rootꢀmeanꢀsquares method was used to fit components in the highꢀ
resolution spectra.
FTꢀIR spectrum was obtained on a Nicolet Nexus 470 FTꢀIR analyzer using the
KBr method. Reaction products were analyzed by LCꢀMS (ThermofinniganLCQ
Advantage) and GCꢀMS (HP5972) (capillary column: 30 m SEꢀ30, 0.25 mm ID, and
0.25 µm film thickness) to identify products and polycarbonate precursors.
5-HMF yield characterization. All reaction products were analyzed using a high
performance liquid chromate graph (Waters 1525 equipped with UV and a 996
photodiode array detector) and quantified with 1ꢀchloronaphthalene as an interior
standard (or calibration curves generated from commercially available standards).
Following a typical reaction, the product mixture was diluted with a known mass of
CH3OH, centrifuged to sediment insoluble products, and 1ꢀchloronaphthalene was
added into the product solution. The concentrations of products were calculated from
HPLCꢀpeak integrations and used a 1ꢀchloronaphthalene interior standard to calculate
molar yields.
5ꢀHMF and other products were analyzed by reversedꢀphase chromatography on a
Varian MicrosorbꢀMV100ꢀ5 C18 column (250×4.6 mm, 93:7water/acetonitrile, 1
mL/min, 308 K). The reaction product was also analyzed by gas chromatographyꢀ
8

Page 9 of 33
Green Chemistry
View Article Online
DOI: 10.1039/C7GC03318D
mass spectrometry (GCꢀMS) and high performance liquid chromatography using an
evaporative light scattering detector (HPLCꢀELSD).
General procedure for the conversion of carbohydrates to 5-HMF. The catalytic
experiments were performed in a 20 ml flask. In a typical experiment, 0.9 mg glucose
and 10 mass% of various catalysts with respect to glucose were added into 5 ml
solvent. After it had been purged with nitrogen, the reaction mixture was heated in an
oilꢀbath at elevated temperature and stirred at 500 rad/min. After the desired reaction
time elapsed, the reaction mixture was cooled to room temperature immediately.
DFT calculations. All the geometries of the βꢀglucose and the corresponding
intermediates were optimized in vacuo using hybrid density functional methods with
23
B3LYP/6ꢀ311+G(d,p) level of theory implemented in Gaussian 09 . The same level
of theory and basis sets were used for the computations of vibrational frequencies, in
order to verify the optimized structures (zero imaginary frequencies) as minima or
firstꢀorder saddle points (one imaginary frequency for transition states structures).
Then, the intrinsic reacton coordinate (IRC)24, 25 pathways of transition states have
been traced to confirm that each of them actually connects the desired reactant and
product. Dispersionꢀcorrected B3LYP optimisations incorporating Grimme’s D3
correction with BeckeꢀJohnson damping (keyword empiricaldispersion=gd3bj) were
carried out on the geometry optimizations. To consider the effects of water and ionic
liquid at the same time, we decided to use water as our model solvent, which is easier
to introduce in the form of discrete molecules, moreover, because of its high polarity
9

Green Chemistry
Page 10 of 33
View Article Online
DOI: 10.1039/C7GC03318D
(dielectric constant, 78) will provide a stronger solvation effect than ionic liquid.
Water solvation effects will give an upper limit to the effect of a conventional solvent
the reaction rate. Water solvation effects are often introduced in a satisfactory and
computationally economic way through continuum models like PolarizedꢀContinuum
Model (PCM).26,ꢀ28 (Computational details can be found in Supplementary Materials).
Results and discussion
Sulfonated poly(thiophenylene) preparation and its usage in biomass conversion
into 5-HMF
Poly(phenylene sulfide) (PPS), which was synthesized from Na2S and 1,4ꢀ
dichlorobenzene in Nꢀmethylpyrrole, was the starting point for the synthesis of SPPS
by a simple sulfonation process in oleum under a dry nitrogen atmosphere. The molar
ratio of basic divalent sulfide (S) to strong Brønsted acid (SO3H) in SPPS can be wellꢀ
controlled by the sulfonation temperature and time. SPPS catalyst has a hydrophobic
PPS backbone matrix and hydrophilic SO3H ionic sites. We can characterize the
degree of sulfonation of SPPS by titration, XPS and FTꢀIR (Figure 1). In the XPS
spectra of SPPS, four major signals with different intensities are observed in the sulfur
2p region, at approximately 170, 168, 166 and 164 eV. On the basis of their binding
energies, we assign them to the sulfone (ꢀSO2ꢀ), sulfate (ꢀSO3H), sulfoxide (or sulfite ꢀ
SOꢀ) and sulfide groups (ꢀSꢀ), respectively (Figure S1). Two minor components at
170 and 166 eV are formed by partial oxidation of the thioether during the sulfonation
of PPS. The SO3H groups were also detected in the IR spectrum, where strong
10

Page 11 of 33
Green Chemistry
View Article Online
DOI: 10.1039/C7GC03318D
characteristic peaks at 1201 and 1055 cmꢀ1, assigned to the S=O stretching vibration,
were observed. These peaks were not observed in the IR spectrum of the pristine
PPS29. In general, the hydrophobic PPS matrix and hydrophilic SO3H group are
sarepartially or completely phaseꢀseparated in SPPS24.
(Figure 1)
In the present study, we use SPPS as an environmentally friendly heterogeneous
catalyst for 5ꢀHMF synthesis directly from fructose, glucose and cellulose in an IL
solvent. 5ꢀHMF yields from fructose, glucose and cellulose, catalysed by a SPPS
catalyst with 26.2% sulfation in an anhydrous [EMIM]Br solvent, are shown in Figure
2.
(Figure 2)
It can be seen that the yield of 5ꢀHMF from glucose in anhydrous [EMIM]Br
o
solvent at moderate temperatures (80ꢀ110 C) ranges from 5 to 40% in 4 h (Figure
2A). At the temperatures between 100 and 140 oC, the yield of 5ꢀHMF from the direct
conversion of cellulose is in a range of 10ꢀ20%. These yields are lower than that of
conversion from fructose (30ꢀ45%). Nonetheless, when the reaction temperature was
o
o
raised to between 140 C and 180 C, excellent 5ꢀHMF yields from glucose and
cellulose of 87.2% and 64.3% respectively, were achieved. Only trace unwanted
levulinic acid (LA) (<0.5 mol%) was detected according to the GCꢀMS and HPLCꢀ
ELSD data (Figure S2 and Figure S3). A very high selectivity (>99%) was achieved
in the SPPS catalyzed system. The glucose turnover number (TON) and turnover
11

Green Chemistry
Page 12 of 33
View Article Online
DOI: 10.1039/C7GC03318D
frequency (TOF) over SPPS reach 1.95 ×103 and 0.14/s, respectively. These yields are
much higher than the highest 5ꢀHMF yield (45.2%) from the fructose dehydration
catalyzed by SPPS at the optimal conditions. To the best of our knowledge, this is the
highest yield from nonꢀtoxic heterogeneous catalysts for 5ꢀHMF production from
glucose and cellulose.
A high yield of 5ꢀHMF was achieved on the SPPS catalyst when macromolecular
cellulose was used as feed stock. This is surprising because there is a steric hindrance
between the SPPS polymer and cellulose macromolecules, which should impede their
reaction. This steric hindrance is possibly weakened by proton exchange between the
IL and the SPPS30. We also investigated the performance of SPPS in other ILs such as
1ꢀbutylꢀ3ꢀmethylimidazolium bromide ([BMIM]Br), 1ꢀbutylꢀ3ꢀmethylimidazolium
chloride ([BMIM]Cl), 1ꢀallylꢀ3ꢀmethylimidazolium chloride ([AMIM]Cl, 1ꢀbutylꢀ3ꢀ
methylimidazolium hydrate ([BMIM]OH) and Nꢀethylꢀpyridinium bromide ([EPy]Br).
As presented in Figure 2C, the catalytic reaction over SPPS in other neutral ionic
liquids shows similar catalytic results, namely that the 5ꢀHMF yields from glucose are
much higher than those from fructose. An exception appeared in the basic
[BMIM]OH, where the 5ꢀHMF yield from fructose was little higher than that from
glucose or cellulose, since the Brønsted acidic centre of SPPS could be neutralized by
the basic [BMIM]OH.
The formation of 5ꢀHMF from fructose, glucose and cellulose catalyzed by SPPS in
o
[EMIM]Br mixtures at 100, 140, and 180 C, respectively, was investigated (see
12

Page 13 of 33
Green Chemistry
View Article Online
DOI: 10.1039/C7GC03318D
o
Figure 2B). The 5ꢀHMF yield from the fructose conversion is a maximum at 100 C;
further increase in temperature does not improve the yield. The highest 5ꢀHMF yields
o
from glucose and cellulose were achieved at 140 and 180 C, respectively. The
activation energies of fructose and glucose conversion into 5ꢀHMF were estimated
from the kinetic data at different temperatures to be 25.5 kcal/mol and 34.6 kcal/mol,
respectively. It is also noted that the highest yield of the fructose and cellulose
conversions was achieved within the first 2 h of the reaction, but within 4 h for
glucose conversion. The further prolonged reaction resulted in a declining yield of 5ꢀ
HMF, due to the formation of byꢀproducts similar to the results of conventional
catalytic processes31ꢀ33
.
The effects of SPPS sulfonation degree on biomass conversion into 5-HMF
The sulfonation degree of SPPS catalyst is found to influence its catalytic activity. For
example, 65ꢀ88% 5ꢀHMF yields from glucose were obtained in [EMIM]Br solvent
when the sulfonation degree was between 10.0 and 35.0 mol% (Figure 3). At a
sulfonation degree of 26.2 mol%, the SPPS catalyst displays the best activity for 5ꢀ
HMF yield of 87.2%. When cellulose was the reactant and the reaction was conducted
o
at 180 C, using the SPPS catalyst with a sulfonation degree of 21.8 mol% led to a
high 5ꢀHMF yield of 68.2%.
(Figure 3)
Biomass conversion into 5-HMF over different catalysts
13

Green Chemistry
Page 14 of 33
View Article Online
DOI: 10.1039/C7GC03318D
The performance of various other common acidic catalysts such as metal chlorides,
amino acids and supported acids, were also investigated for the production of 5ꢀHMF
from fructose, glucose and cellulose in [EMIM]Br. As shown in Figure 4, the SPPS
catalysts display much higher activity for the conversions of glucose and cellulose to
5ꢀHMF although they have much smaller BET specific surface areas than those of the
2ꢀ
2ꢀ
heterogeneous solid supported acid catalysts, such as SO4 /Al2O3, SO4 /Al2O3 and
SO4 /SBAꢀ15. For instance, the specific surface areas are 0.23, 1.56, and 3.23 m2/g
2ꢀ
for the catalyst with 40.5, 26.2, and 21.8 mol% sulfation degree, respectively.
Apparently, the porosity degree of SPPS decreases with the increase of sulfonation.
This is due to the fact that the hydrogen bonds resulting from the introduced ꢀSO3H
groups leads to a stronger interaction among SPPS macromolecules, which makes the
SPPS volume shrink. The yields of 5ꢀHMF from glucose conversion catalyzed by
homogeneous catalysts of CrCl2, CrCl3 and SnCl4 are merely 19.5, 26.0 and 18.9%,
respectively, at 140 °C for 4 h. These yields can be improved to 50.1, 48.5 and 44.9%
when the reaction was conducted at 100 °C for 4 h (Figure 4B, the bars indicated with
a prefixed number of 100). But such yields are still substantially lower than those
achieved by the SPPS catalysts. The only exception is the yield of 5ꢀHMF from
fructose catalyzed by CrCl2 and H2SO4, which is 58.3 and 55.1%, respectively, which
is higher than that catalyzed by SPPS (45.2%). Moreover, in our experiment the
process using homogeneous CrCl2, CrCl3 and SnCl4 catalysts produced many more
byꢀproducts than that using the SPPS catalysts. For instance, the most important side
reactions are the rehydration of 5ꢀHMF into levulinic and formic acids, and either
14

Page 15 of 33
Green Chemistry
View Article Online
DOI: 10.1039/C7GC03318D
polymerization of 5ꢀHMF or cross polymerization of 5ꢀHMF and carbohydrate34.
These polymeric humins consume initial carbohydrate, leading to an increase in
viscosity and thus limiting mass transfer.
(Figure 4 )
Biomass conversion mechanism into 5-HMF catalyzed by SPPS in presence of
ILs
The prevailing view in the literature is that converting glucose and cellulose directly
into 5ꢀHMF is much more difficult than converting fructose to 5ꢀHMF. Hence, the
isomerization of glucose or cellulose into fructose is an essential step for producing 5ꢀ
HMF from glucose or cellulose and has been extensively studied. However, the
present study demonstrates that in the SPPS catalysed systems using ILs as solvent,
the conversion of glucose and cellulose into 5ꢀHMF is much easier and more efficient
than the conversion of fructose.
In order to explore the glucose conversion mechanism over SPPS at high
temperature (at which excellent yields were achieved), high performance liquid
chromatography mass spectrometry (HPLCꢀMS) was used to characterize the product
distribution of the reaction at 140 °C. Three products were detected and their base
peaks in the mass spectrum corresponding to molecular ions (M+) at m/z=126, 144
and 162, respectively (see Figure S4). No fructose was detected. However, fructose
was detected in the systems catalyzed by the SPPS catalyst at low temperature (<100
oC) or catalyzed by conventional acid catalysts31. These systems exhibit relatively low
15

Green Chemistry
Page 16 of 33
View Article Online
DOI: 10.1039/C7GC03318D
5ꢀHMF yields. In addition to the target 5ꢀHMF (m/z = 126), the byꢀproducts were
identified as 3,4ꢀdihydroxyꢀ5ꢀ(hydroxymethyl)ꢀtetrahydrofuranꢀ2ꢀcarbaldehyde
(intermediate A, m/z = 162), and 4ꢀhydroxyꢀ5ꢀ(hydroxymethyl)ꢀ4,5ꢀdihydrofuranꢀ2ꢀ
carbaldehyde (intermediate B, m/z = 144), respectively.
Similar to the spherulite compounds of pristine PPS20ꢀ21, 29, 35, SPPS displays a
spherical particle structure in ILs, and the hydrophilic sulfonic groups mainly
distribute on the outer surface of SPPS spheres because the sulfonation took place on
the PPS surface in oleum. The ionꢀexchange of [BMIM]Br with the Brønsted acid
sites of SPPS promote the release of H+ ions36, and the surface of the catalyst is partly
modified by [EMIM]+ cations. The sugarꢀsolubilizing [EMIM]+ cations improve the
affinity of glucose for SPPS heterogeneous catalysts8, making the catalytically active
centres readily accessible to glucose molecules. To verify our assumptions, DFT
calculations (Supplementary Materials) were carried out to gain further insights into
the mechanism for the transformation of glucose to 5ꢀHMF in the systems of SO3Hꢀ
SPPS and [EMIM]Br. According to previous experimental36ꢀ40 and theoretical
studies38ꢀ44 on the catalytic conversion of glucose to HMF by Brønsted acids and
Lewis acidic metal salts, It is generally acknowledged that the first step is the
formation of an enediol intermediate (IM) from glucose, and then the enediol
intermediate converts into fructose and further to 5ꢀHMF. Based on our DFT
calculation, we suppose that the reaction may follow the mechanism shown in
Scheme 1 and Figure 5ꢀ8 with SPPS acting as Brønsted acid catalyst and [EMIM]Br
as solvent. Firstly, the H atom of sulfonic acid attacks the ring O of the glucose 6ꢀ
16

Page 17 of 33
Green Chemistry
View Article Online
DOI: 10.1039/C7GC03318D
member ring structure, resulting in the protonation of the ring O and the breakage of
C1−O6 bond (intermediate IM1), which has a very low barrier with only about 10.7
kcal/mol. Subsequently, the openꢀring aldehyde form is transformed in to 1,2ꢀenol by
a ketoneꢀenol isomerization, which is slowly elementary step with an energy barrier
of 21.5 kcal/mol. Then, the 1,2ꢀenol to ketose occurs via the second enolꢀketone
isomerization step with an barrier of 17.4 kcal/mol. The ketose form closes the ring
structure to form a 5ꢀmember ring fructose with an energy barrier of 9.3 kcal/mol. The
results indicates that it is relatively difficult for the formation of 1,2ꢀenediol to ketoneꢀ
enol isomerization reaction. Generally, the transformation of fructose to 5ꢀHMF
includes three dehydration processes. The calculated dehydration mechanism details
are shown in Figure 6ꢀ7. Along the reaction coordinate, the first dehydration process
is key step with a barrier of 13.6 kcal/mol, leading to C−OH bond cleavage.
Throughout the whole reaction pathway, we found that the ketolꢀisomerization and
the first dehydration are two rateꢀdetermining elementary steps to be carried out under
thermal conditions.
(Scheme 1) and (Figure 5-8 )
To verify the role of SPPS, sulfonated polysulfone (SPSF), benzenesulfonic acid
(BSA) and acidic ionꢀexchange resin (AIER) were used to catalyze the synthesis of 5ꢀ
HMF from fructose and glucose, respectively. As expected, SPSF, BSA and AIER
displayed higher activity in fructose conversion to 5ꢀHMF than in glucose conversion.
In contrast SPPS showed much better activity in glucose conversion to 5ꢀHMF
17

Green Chemistry
Page 18 of 33
View Article Online
DOI: 10.1039/C7GC03318D
(Figure S5). Moreover, the unique structure of the SPPS catalyst is believed to
contribute to the high 5ꢀHMF selectivity. The catalyst possesses a hydrophobic
backbone matrix and hydrophilic SO3H sites. The hydrophobic backbone matrix
minimizes the attraction to the nearby SPPS catalyst of H2O molecules formed in the
reaction, and thus avoids further reaction to form the byꢀproduct levulinic acid (Figure
S2 and S6).
The SPPS catalyst also possesses a number of advantages for practical application
in biomass conversion. For instance, the SPPS catalyst can be well dispersed in an
ionic liquid though the ionꢀexchange of [BMIM]Cl with SO3H, and can be readily
separated after the reaction simply by filtration, sedimentation, or centrifugation
(Figure S7). The synthesis of the SPPS catalyst can be scaled up without difficulty
because of the facile sulfonation of commercial PPS bulky particles. More
importantly, the SPPS catalyst could be recycled and displayed no decline in activity
after 5 recycles (Figure S8). Therefore, SPPS is an ideal heterogeneous catalyst for
the conversion of a biomass (glucose) into a fine chemical (5ꢀHMF).
Conclusions
This study demonstrates that SPPS are highly efficient for the direct conversion of
glucose or cellulose to 5ꢀHMF. The environmentally benign conversion involves
some new mechanism and is feasible for the practical valorization of glucose and
cellulose. Thus, knowledge acquired in this study may inspire further research on new
catalysts for this purpose.
18

Page 19 of 33
Green Chemistry
View Article Online
DOI: 10.1039/C7GC03318D
ACKNOWLSDGEMENTS
The authors are grateful for the financial support of the National and Tianjin Natural
Science Foundation of China (Nos. 21376177, 201676202 and 15JCZDJC7000). This
work is also supported by China National Textile and Apparel Council (J201406) and
China petroleum and chemical co., LTD (Nos. 215038 and 216090).
References
1 Y. C. Lin and G. W. Huber, Energy Environ. Sci. 2009, 2, 68.
2 L. D.Schmidt and P. J.Dauenhauer, Nature 2007, 447, 914.
3 J. N. Chheda, G. W.Huber and J. A.Dumesic, Angew. Chem. 2007, 119, 7298.
4 J. N. Chheda, G. W. Huber and J. A.Dumesic, Angew. Chem. Int. Ed. 2007, 46,
7164.
5 Y. RomanꢀLeshkov, J. N. Chheda and J. A. Dumesic, Science 2006, 312, 1933.
6 M. Moliner, Y. RomanꢀLeshkov and M. E.Davis, Proc. Natl. Acad. Sci. U. S. A.
2010, 107, 6164.
7 M. E. Zakrzewska, E. BogelꢀLukasik and R. BogelꢀLukasik, Chem. Rev. 2011, 111,
397.
8 H. B. Zhao, J. E. Holladay, H. Brown and Z. C. Zhang, Science 2007, 316, 1597.
9 G. Yong, Y. Zhang and J. Y. Ying, Angew. Chem. Int. Ed. 2008, 47, 9345.
10 J. B. Binder and R. T. Raines, J. Am. Chem. Soc. 2009, 131, 1979.
19

Green Chemistry
Page 20 of 33
View Article Online
DOI: 10.1039/C7GC03318D
11 R. J. Van Putten, J. C. Van der Waal, E. De Jong, C. B. Rasrendra, H. J. Heeres,
and J. G. De Vries, Chem. Rev. 2013, 113, 1499.
12 Z. C. Zhang, Adv. Catal. 2006, 49, 153.
13 S. J. Angyal, Adv. Carbohydr. Chem. Biochem. 1984, 42, 15.
14 J.Y. He, M.J. Liu, K.F. Huang, T. W. Walker, C. T. Maravelias, J. A.
Dumesica, G. W. Huber, Green Chem. 2017, 19, 3642.
15 X. H. Qi, M. Watanabe, T. M. Aida and R. L.Smith, Green Chem. 2008, 10, 799.
16 C. LansalotꢀMatras and C. Moreau, Catal. Commun. 2003, 4, 517.
17 C. Moreau, R. Durand, S. Razigade, J. Duhamet, P. Faugeras, P. Rivalier, P.
Ros and G. Avignon, Appl. Catal., A 1996, 145, 211.
18 V. V. Ordomsky, J. Van der Schaaf, J. C. Schouten and T. A.Nijhuis, J. Catal.
2012, 287, 68.
19 Y. RomanꢀLeshkov, C. J. Barrett, Z. Y. Liu and J. A. Dumesic, Nature 2007, 447,
982.
20 P. Hruszka, J. Jurga and B. Brycki, Polymer 1992, 33, 248.
21 K. Miyatake, H. Iyotani, K. Yamamoto and E. Tsuchida, Macromolecules 1996,
29, 6969.
22 E. A. Pidko, V. Degirmenci, R. A. Van Santen and E. J. M. Hensen, Angew.
Chem. Int. Ed. 2010, 49, 2530.
20

Page 21 of 33
Green Chemistry
View Article Online
DOI: 10.1039/C7GC03318D
23 Gaussian 09 (Revision D.01), Gaussian Inc., Wallingford, CT, 2009. (References
can be found in Supplementary Materials).
24 K. Fukui, 1981, 14, 363.
25 V. Barone, M. Cossi, J. Phys. Chem. A 1998, 102, 1995.
26 S. Miertus, E. Scrocco, J. Ttomasi, J. Chem. Phys. 1981, 74, 117.
27 F. Bessac, F. Maseras. J. Comput. Chem. 2008, 29, 892.
28 G.W. Amarante, M. Benassi, H.M.S. Milagre, A.A.C. Braga, F. Maseras, et al. Chem.
Eur. J. 2009, 15, 12460.
29 M. A. Barique, L. B.Wu, N. Takimoto, K. Kidena and A. Ohira, J. Phys. Chem.
B 2009, 113, 15921.
30 Zhao, Q. A.; Wang, L.; Zhao, S.; Wang, X. H.; Wang, S. T. Fuel 2011, 90, 2289ꢀ
2293.
31 P. Lanzafame, D. M. Temi, S. Perathoner, A. N. Spadaro and G. Centi, Catal.
Today 2012, 179, 178.
32 F. L. Yang, Q. S. Liu, X. F. Bai and Y. G. Du, Bioresour. Technol. 2011, 102,
3424.
33 V. Choudhary, S.H. Mushrif, C. Ho, A. Anderko, V. Nikolakis, N.S. Marinkovic,
A.I. Frenkel, S.I. Sandler and D.G.Vlachos, J. Am. Chem. Soc. 2013, 135, 3997.
21

Green Chemistry
Page 22 of 33
View Article Online
DOI: 10.1039/C7GC03318D
34 G. Tsilomelekis, M. J. Orella, Z.X. Lin, Z.W. Cheng, W.Q. Zheng, V.
Nikolakisa, D. G. Vlachos, Green Chem. 2016,18, 1983.
35 S. Bose, T. Kuila, X. L. N. Thi, N. H. Kim, K. T. Lau and J. H. Lee, Prog.
Polym. Sci. 2011, 36, 813.
36 L. Hu, Z. Wu, J.Xu, Y. Sun, L. Lin and S.J. Liu, Chem. Eng. J. 2014, 244, 137.
37 E. J. Barreiro, A. E. Kummerle and C. A. M. Fraga, Chem. Rev. 2011, 111, 5215.
38 R. C. Remsing, R. P. Swatloski, R. D. Rogers and G. Moyna, Chem. Commun.
2006, 12, 1271.
39 W.H. Brown, C.S. Foote, B.L. Iverson and E.V. Anslyn, Organic Chemistry
(Fifth edition) , Lockwood, L., Kiselica, S., Summers, A., Woods, L., Eds.
Chapter 19, Enolate anions and enamines, pp. 712.
40 X. H. Qian, Top. Catal. 2012, 55, 218.
41 E. A. Pidko, V. Degirmenci, R. A. van Santen, E. J. M. Hensen. Angew. Chem.,
Int. Ed. 2010, 49, 2530.
42 J. Guan, Q. A. Cao, X. C. Guo, X. D. Mu. Comput. Theor. Chem. 2011, 963, 453.
43 R. S. Assary, T. Kim, J. J. Low, J. Greeley, L. A. Curtiss. Phys. Chem. Chem.
Phys. 2012, 14, 16603.
44 J. J. Li, J. H. Li, D. J. Zhang, C.B. Liu. J. Phys. Chem. B. 2015, 119, 13398.
22

Page 23 of 33
Green Chemistry
View Article Online
DOI: 10.1039/C7GC03318D
LEGEND
Figure 1 | XPS and FTIR survey spectra of SPPS. The sulfonation degree was
calculated from Figure 1A, and S binding energy (eV) located at around 164 eV,
while the S binding energy of SO3H is located at 168ꢀ169 eV.
Figure 2 | 5-HMF synthesis from fructose, glucose and cellulose. A: 0.9 g
carbohydrate, 0.09 g SPPS (26.2% sulfation degree), 5 mL [EMIM]Br, 4 h; B: 0.9 g
carbohydrate, 0.09 g SPPS (26.2% sulfation degree), 5 mL [EMIM]Br. C: Glucose
conversion to 5ꢀHMF over SPPS in different solvent (reaction condition: 0.18 g SPPS
with 26.2 mol% sulfonation degree, solvent10 ml, 140 °C for 4 h)
Figure 3 | 5-HMF yield from different carbohydrates. (Condition: SPPS catalyst
0.09 g, carbohydrate 0.9 g, 5 ml [EMIM]Br)
Figure 4 | 5-HMF yield from fructose over different catalysts in [EMIM] Br.
(Conditions: A: 0.9 fructose, 0.09 g catalyst, SPPS with 40.5% sulfation degree, 5mL
°
[EMIM]Br, 100 C, 4 h); B: 5ꢀHMF yield from glucose over different catalysts in
[EMIM]Br (Conditions: 0.9 g glucose, 0.09 g catalyst, SPPS with 26.2% sulfation
°
degree, 5 mL [EMIM]Br, 140 C,4 h); C: 5ꢀHMF yield from celluloseover different
catalysts in [EMIM]Br (Conditions: 0.9 g cellulose, 0.09 g catalyst, SPPS with 21.8%
sulfation degree, 5 mL [EMIM]Br, 180 °C, 2 h).
Scheme 1. The mechanism of glucose conversion in the presence of SPPS
23

Green Chemistry
Page 24 of 33
View Article Online
DOI: 10.1039/C7GC03318D
Figure 5. The computed Gibbs free energy profile of intermediates and transition
states for the formation of the 1,2ꢀenediol intermediate (IM3). Solvent effects were
considered by polarizable continuum model (PCM) singleꢀpoint calculations in water.
Figure 6. The computed Gibbs free energy profile of the enolꢀketone isomerization
and ringꢀclosing reactions to form fructose. Solvent effects were considered by
polarizable continuum model (PCM) singleꢀpoint calculations in water.
Figure 7. The computed Gibbs free energy profile of the continuous dehydration
process to form 5ꢀHMF. Solvent effects were considered by polarizable continuum
model (PCM) singleꢀpoint calculations in water.
Figure 8. Transition states for the reaction of Glucose to 5ꢀHMF. Red: oxygen,
yellow: sulfur, gray: carbon, white: hydrogen.
24

Page 25 of 33
Green Chemistry
View Article Online
DOI: 10.1039/C7GC03318D
Figure 1
25

Green Chemistry
Page 26 of 33
View Article Online
DOI: 10.1039/C7GC03318D
Figure 2
26

Page 27 of 33
Green Chemistry
View Article Online
DOI: 10.1039/C7GC03318D
Figure 3
27

Green Chemistry
Page 28 of 33
View Article Online
DOI: 10.1039/C7GC03318D
Figure 4
28

Page 29 of 33
Green Chemistry
View Article Online
DOI: 10.1039/C7GC03318D
29
Scheme 1
29