Sulfonated lignin-derived ordered mesoporous carbon with highly selective and recyclable catalysis for the conversion of fructose into 5-hydroxymethylfurfural

Article abstract of DOI:10.1016/j.apcata.2019.02.008

Sulfonic acid-functionalized lignin-derived mesoporous carbon (LDM C-S O₃H) was prepared using phenolation and evaporation induced self-assembly method followed by sulfonation. The obtained LDM C-S O₃H bearing sulfonic acid density of 0.6528 mmol/g possessed a well-ordered two-dimensional hexagonal mesoporous characteristics. A 5-hydroxymethylfurfural (5-HMF) yield of 98.0% with a full fructose conversion was obtained using LDM C-S O₃H as catalyst at 140 °C for 2 h in DMSO. Reactive kinetics studies revealed that fructose conversion in DMSO without catalyst or catalyzed by LDMC-SO₃H may obey pseudo-first-order kinetics, and the activation energy of latter (72 kJ/mol) was much lower than that of former (114 kJ/mol). Adsorption kinetics studies indicated that almost no 5-HMF adsorbed onto LDM C-S O₃H probably had a great contribution to the high selectivity of up to 98.0%, while the fructose adsorption on LDM C-S O₃H was a diffusion-controlling adsorption process with more following Bangham kinetic model and Weber-Morris kinetic model owing to the characteristics of ordered mesostructure of LDM C-S O₃H. Moreover, LDM C-S O₃H exhibited superior reusability and stability in catalytic performance with a 5-HMF yield higher than 88.0% in six runs probably due to the synergistic effect of mesopore structure with a special surface and -SO₃H groups with a relatively high content. These research results will contribute to a better understanding of structure-performance relationship of LDM C-S O₃H used as an efficient catalyst in the fructose-to-5-HMF transformation as well as the high-value utilization of lignin in the field of catalysis.

Full text of DOI:10.1016/j.apcata.2019.02.008

Accepted Manuscript
Title: Sulfonated lignin-derived ordered mesoporous carbon
with highly selective and recyclable catalysis for the
conversion of fructose into 5-hydroxymethylfurfural
Authors: Linhuo Gan, Li Lyu, Tianruo Shen, Shuai Wang
PII:
S0926-860X(19)30063-8
DOI:
Reference:
https://doi.org/10.1016/j.apcata.2019.02.008
APCATA 16978
To appear in:
Applied Catalysis A: General
Received date:
Revised date:
Accepted date:
13 October 2018
1 February 2019
5 February 2019
Please cite this article as: Gan L, Lyu L, Shen T, Wang S, Sulfonated lignin-derived
ordered mesoporous carbon with highly selective and recyclable catalysis for the
conversion of fructose into 5-hydroxymethylfurfural, Applied Catalysis A, General
(2019), https://doi.org/10.1016/j.apcata.2019.02.008
This is a PDF file of an unedited manuscript that has been accepted for publication.
As a service to our customers we are providing this early version of the manuscript.
The manuscript will undergo copyediting, typesetting, and review of the resulting proof
before it is published in its final form. Please note that during the production process
errors may be discovered which could affect the content, and all legal disclaimers that
apply to the journal pertain.

Sulfonated lignin-derived ordered mesoporous carbon with highly selective and
recyclable catalysis for the conversion of fructose into 5-hydroxymethylfurfural
Linhuo Gan*, Li Lyu, Tianruo Shen, Shuai Wang
*Corresponding author: Linhuo Gan
College of Chemical Engineering, Huaqiao University, 668 Jimei Avenue, Xiamen 361021, China.
E-mail: lhgan401@126.com; lhgan@hqu.edu.cn.
1

GRAPHICALABSTRACT
HIGHLIGHTS
ò
ò
ò
ò
Sulfonic acid-functionalized well-ordered lignin-derived mesoporous carbon (LDMC-SO3H) was
synthesized using phenolation treatment and soft-template method followed by sulfonation.
LDMC-SO3H showed superior catalytic activity in fructose dehydration to 5-HMF in terms of fructose
conversion (100.0%), 5-HMF yield (98.0%) and selectivity (98.0%).
LDMC-SO3H possessed superior reusability and stability in terms of 5-HMF yield during six cycle
runs.
The well-ordered mesostructure and the suitable hydrophobic/hydrophilic balance of the surfaces of
LDMC-SO3H might play vital roles in the efficient catalytic dehydration of fructose to 5-HMF.
ABSTRACT
Sulfonic acid-functionalized lignin-derived mesoporous carbon (LDMC-SO3H) was prepared using
phenolation and evaporation induced self-assembly method followed by sulfonation. The obtained LDMC-
SO3H bearing sulfonic acid density of 0.6528 mmol/g possessed a well-ordered two-dimensional hexagonal
mesoporous characteristics. A 5-hydroxymethylfurfural (5-HMF) yield of 98.0% with a full fructose
2

conversion was obtained using LDMC-SO3H as catalyst at 140 °C for 2 h in DMSO. Reactive kinetics
studies revealed that fructose conversion in DMSO without catalyst or catalyzed by LDMC-SO3H may
obey pseudo-first-order kinetics, and the activation energy of latter (72 kJ/mol) was much lower than that
of former (114 kJ/mol). Adsorption kinetics studies indicated that almost no 5-HMF adsorbed onto LDMC-
SO3H probably had a great contribution to the high selectivity of up to 98.0%, while the fructose adsorption
on LDMC-SO3H was a diffusion-controlling adsorption process with more following Bangham kinetic
model and Weber-Morris kinetic model owing to the characteristics of ordered mesostructure of LDMC-
SO3H. Moreover, LDMC-SO3H exhibited superior reusability and stability in catalytic performance with a
5-HMF yield higher than 88.0% in six runs probably due to the synergistic effect of mesopore structure
with a special surface and -SO3H groups with a relatively high content. These research results will
contribute to a better understanding of structure-performance relationship of LDMC-SO3H used as an
efficient catalyst in the fructose-to-5-HMF transformation as well as the high-value utilization of lignin in
the field of catalysis.
Keywords: Kraft lignin, Mesoporous carbon, Solid acid catalyst, Kinetics analysis, Green chemistry
3

1 Introduction
5-hydroxymethylfurfural (5-HMF) has been considered as an important biomass-derived green
platform compound possessing highly reactive substructures including furan ring, hydroxymethyl group
and aromatic aldehyde. Its derivatives can be utilized widely as perfume, pharmaceutical intermediate, fuel
additive, and a substitute for benzene series applied in the synthesis of new polymer materials [1].Therefore,
the research on the dehydration of biomass to prepare 5-HMF using hetero- and homogeneous acidic
catalysts has drawn much attention in the world [2].
Ordered mesoporous carbons (OMCs), have received worldwide attention owing to their remarkable
properties such as mesoscopic structures, high surface area and good chemical stabilities, thus becoming a
kind of important material with potential application prospects in the fields of catalyst support [3],
adsorption [4],drug delivery [5] and supercapacitor electrode [6]. In recent years, combining the advantages
of mesoporous carbon, the acid-functionalized OMCs have been empolyed in the acid catalyzed
dehydration of fructose to produce 5-HMF owing to its excellent catalytic performance with high reaction
selectivity and stability [7,8]. Gallo et al [7] studied the activity and stability of sulfonated ordered
mesoporous carbons (OMCs-SO3H) for the conversion of fructose to 5-HMF. It was found that OMCs-
SO3H showed comparable selectivity and up to 3 times more stable than commercial Nafion SAC-13. Wang
et al [8] investigated the dehydration of fructose to 5-HMF using OMC-SO3H as catalyst, and result showed
that a high HMF yield of 89.4% in DMSO was obtained at 120 °C within 30 min. Furthermore, no apparent
loss of the activity of OMC-SO3H was observed after it was reused 3 times. Conventionally, nanocasting
method technique was widely used in synthesis of ordered mesoporous carbons. However, compared to its
disadvantages of complex process and environmental pollution, soft-template method using sacrificial
amphiphilic surfactants for synthesizing ordered mesoporous carbons has been adopted and further
4

developed in recent years [9]. So far, owing to the excellent properties of good solubility in surfactant
solution, capability to interact with surfactant and self-assembly, stable three-dimensional rigid structure
and relatively high carbon content, phenolic-formaldehyde based resins mainly including formaldehyde
cross-linked phenol, resorcinol and phloroglucinol have been selected widely as carbon precursors applied
in soft-template synthesis of ordered mesoporous carbons [10].
The application of recyclable heterogeneous catalyst derived from renewable feedstock for chemical
productions is considered as one of the core principles of green chemistry [11]. As is known, lignin is a
natural biomass resource possessing obvious advantages of high carbon content, low cost, environmental
friendliness and sustainable utilization, which has been used widely as a superior carbon precursor in the
preparations of carbon materials such as activated carbon [12] and char [13]. In the meanwhile, lignin is
the only natural biopolymer consisting of plenty of aromatic rings or phenolic hydroxyl groups in its three-
dimensional network structure. Thus, to reduce the reliance on crude oil resources to some extent, lignin
can be employed as a promising natural substitute for phenols to crosslink with formaldehyde to synthesize
phenol resins [14]. However, the poor solubility and chemical activity of lignin restricted its addition
amount during the synthesis process of phenol resin utilized as adhesive. In recent years, thermochemical
technology has been adopted widely to improve the solubility and reactivity of lignin via phenolation
treatment, which make it more suitable for the synthesis of lignin-phenol-formaldehyde (LPF) resin [15,16].
For the same consideration, lignin also can be a superior carbon precursor for synthesis of mesoporous
carbon. To our best knowledge, some studies were performed on the synthesis of lignin-derived mesoporous
carbon (LDMC). Among them, the research group led by Naskar studied on LDMC synthesis and its
applications involving drug delivery and supercapacitor electrode material [17-20]. However, these LDMCs
were short of orderliness and the whole process was time-consuming (more than 7 days). Furthermore, only
5

the KL fraction extracted by methanol was employed in LDMC synthesis, thus leading to the under-
utilization of KL. More recently, using lignin extracted from soft wood or straw as carbon sources, Qin et
al [21] found that the low-molecular-weight lignin bearing more hydroxyl groups was favorable to the
formation of highly ordered mesoporous carbon as well as its utilization as catalyst support. Song et al [22]
synthesized LDMC for supercapacitors using a nano-sized MgO template coupled with soft template.
Besides, Chen et al [23] synthesized NiO-containing mesoporous carbon materials derived from sodium
lignosulphonate (LS) using sol-gel method. Using LS as carbon source and cetyltrimethylammonium
bromide as template agent, Zhao et al [24] prepared mesoporous carbon by hydrothermal and template
method. Considering the sustainable development and environmental protection, either KL or LS, being a
renewable biomass resource and the largest reservoir of natural aromatic polymer recovered from pulping
spent liquor, is a very promising carbon precursor for LDMC synthesis. KL is separated from spent liquor
of kraft pulping, which is the most dominant chemical method in pulp and paper industry. Compared to the
water-soluble LS, the industrial applications of KL in large quantity are restrained by its poor solubility to
a great extent. Therefore, finding an innovative approach for efficient utilization of KL in LDMC synthesis
is more urgent and valuable.
Herein, firstly, to overcome the obstacles of complex super-molecular structure and poor solubility of
KL, thus increasing the overall utilization of KL and the synthesis efficiency of LDMC as well as reducing
its preparation cost, a well-ordered mesoporous carbon was synthesized using KL as raw material via soft-
template method as shown in Scheme 1. Thus, we creatively present a reasonable synthesis route via using
KL degradation products liquefied in phenol under alkaline condition followed by crosslinking with
formaldehyde as carbon precursor. Compared to the results described by Naskar et al [17,19], KL could be
used fully as much as possible, and the preparation time was shortened largely (about 3 days). Furthermore,
6

KL could be degraded into fragments with lower molecular weight and more phenolic hydroxyl groups
were engrafted onto structure after phenolation treatment [16], thus facilitating the formation of LDMC
[21]. Secondly, LDMC prepared in the present work was utilized as catalyst support for the synthesis of
carbonaceous solid acid catalyst using traditional sulfonation method. The sulfonic acid-functionalized
lignin-derived mesoporous carbon (LDMC-SO3H) was fully characterized by small-angle X-ray scattering
(SAXS), transmission electron microscopy (TEM) and Brunauer-Emmett-Teller (BET) to confirm the
formation and stability of ordered mesoporous structure. The catalytic performance of LDMC-SO3H was
investigated in the fructose dehydration to 5-HMF. Finally, to clarify the catalytic mechanism when using
LDMC-SO3H as catalyst in fructose-to-5-HMF transformation, the studies on reaction kinetics and
adsorption kinetics onto LDMC-SO3H were carried out, respectively. This study will not only provide an
efficient solid acid catalyst for the conversion of fructose to 5-HMF, but also propose a new strategy for the
high value-added utilization of KL in the field of catalysis as well as apply a sustainable approach for the
synthesis of OMC.
2 Experimental
2.1 Materials
KL and Pluronic F127 were purchased from Sigma-Aldrich Chemical Co. Phenol, formaldehyde (an
aqueous solution of 37%), sodium hydroxyl (NaOH), concentrated hydrochloric acid (HCl) and
concentrated sulfuric acid (H2SO4) were purchased from Xilong Chemical Co. Ltd, China. Fructose and
dimethyl sulphoxide (DMSO) were purchased from Sinopharm Chemical Reagent Co. Ltd, China. 5-HMF
(95%) and 5-HMF (HPLC solvent) were purchased from Aladdin Industrial Inc. All chemicals in this work
were used as received without further purification.
7

2.2 Preparations of LDMC and LDMC-SO3H
Typically, 10.0 g of KL, 30.0 g of phenol, 2.0 g of NaOH and 50 mL of 50 v% ethanol solution were
added into a three-necked, round-bottomed flask provided with a mechanical stirrer, and KL was phenolated
in alkaline medium at 160 °C for 2 h. Then, 10.0 mL of 37 wt% formaldehyde solution was added dropwise,
and the mixture was stirred and heated at 70 °C for 60 min (donated as LPF solution). After cooling reaction
mixture to room temperature, the pH value of LPF solution was adjusted to 7.0 using 2.0 mol/LHCl solution.
Afterward, solvent was removed by reduced pressure distillation, and the final LPF product was dissolved
in ethanol to precipitate salt.
Pluronic F127 was dissolved in 20 mL ethanol under magnetic stirring at 40 °C, which was added into
the above ethanol solution containing LPF prepolymer, further stirred for 12 h. The solution was transferred
onto a Petri dish, the solvent of ethanol evaporated gradually at room temperature for 8 h to form a
transparent membrane first, and then the membrane was further thermo-polymerized at 100 ºC for 24 h in
oven. The obtained materials were scraped from the dishes and carbonized in a tube furnace under nitrogen
atmosphere with the heating profile being heating to 400 °C at a rate of 1 °C/min and then heating up to
900 °C at a rate of 2 °C/min followed by maintenance at 900 °C for 60 min. Finally, LDMC was obtained
after being cooled to room temperature in nitrogen flow.
The as-synthesized LDMC was added into the concentrated sulfuric acid (m/v, 1 g/20 mL) and
sulfonated under nitrogen atmosphere at 180 °C for 12 h. After cooling to room temperature, the black
2-
slurry was washed with hot water (>80 °C) until no detection of SO4 in the filtrate, then the sulfonated
product was dried at 60 °C in oven for 8 h to obtain LDMC-SO3H.
2.3 Characterization
Thermal gravimetric analyses (TGA) and differential thermal analyses (DTA) of F127, LPF and the
8

mixture of F127 and LPF were carried out on Shimadzu DTG-60H thermogravimetric analyzer in a
temperature range of 30-800 °C under nitrogen atmosphere at a constant heating rate of 10 °C/min. Fourier
transform infrared spectroscopy (FTIR) spectra of KL and LDMC-SO3H were recorded on a Nicolet iS50
FTIR spectrometer for KBr pellets in a range of 4000-400 cm-1 and 32 scans per sample. X-ray diffraction
(XRD) pattern of LDMC was recorded on Rigaku Smart Lab X-ray diffractometer with Cu Kα radiation
source at 40 kV and 30 mA from 10 to 80 ° (2θ) at a scanning speed of 4 °/min. The regular mesopore
arrangement in a 2θ range of 0.6-5° was identified by using SAXS profile recorded by the Rigaku Smart
Lab X-ray diffractometer with Cu Kα radiation at 40 kV and 40 mA with a scanning speed of 1 °/min. The
pore textural properties of LDMC and LDMC-SO3H were investigated by nitrogen adsorption−desorption
at -196 °C in a Micromerities ASSP 2020 gas adsorption analyzer. The specific surface area (SBET) was
calculated using BET method, and the pore size distribution curve was obtained from adsorption branch by
using BJH method and DFT model. TEM image of LDMC-SO3H was obtained using a FEI Tecnai G2 F20
200 kV field emission transmission electron microscope.
Using elemental analysis on an Elementar Vario Micro elemental analyzer, the sulfur contents of KL
and LDMC-SO3H were measured to be 1.84 wt% and 2.30 wt%, respectively. The loading amounts
(mmol/g) of -SO3H group in LDMC-SO3H was estimated using the acid-base back titration on a WDDY-
2008J automatic potentiometric titrator (China). Firstly, 50 mg of LDMC-SO3H was added into 20 mL, 2
mol/LNaCl aqueous solution, then the mixture was stirred at room temperature for 12 h and was centrifuged.
The filtrate was titrated with 0.01 mol/L NaOH aqueous solution. As a result, the loading amount of -SO3H
group was determined to be 0.6528 mmol/g in this work.
2.4 Fructose conversion into 5-HMF
The dehydration reaction of fructose was carried out in a 25 mL digestion tube. 200 mg of fructose,
9

20 mg of catalyst and 2 mL of DMSO were added into the digestion tube. The mixture was shaken
intensively (200 rpm) in shaking table by purging with nitrogen for 2 min. Subsequently, the reaction was
heated at desired reaction temperature for certain time. After reaction, the reactor was cooled to the room
temperature immediately. The suspension was then filtered using 0.22 µm filter membrane, and the
remaining amount of fructose and 5-HMF yield in filtrate were determined using high performance liquid
chromatography (HPLC).
For the recycling experiments, 200 mg of fructose, 17.5 mg of LDMC-SO3H and 2 mL of DMSO were
added to a digestion tube. The mixture was heated at 140 ºC for 120 min under nitrogen atmosphere. After
the reaction, LDMC-SO3H was readily recovered by filtering, washing with ethanol three times and dried
at 80 °C overnight. The recovered LDMC-SO3H was used to catalyze the dehydration of fructose into 5-
HMF over ten successive reactions by adding fresh fructose (200 mg) and DMSO solvent (2 mL) in each
run.
2.5 Analysis of product
The concentration of 5-HMF was measured by HPLC analysis performed on an Elite P-1201 (Dalian,
China) equipped with a UV-Vis detector (λ=284 nm) and an Agilent ZORBAX Eclipse Plus C18 analysis
column (Ø4.6 × 250 mm). The mixture solution of methanol and water (1/4, v/v) was used as the mobile
phase at a flow rate of 0.8 mL/min, and the column temperature was maintained at 50 ºC. In the meanwhile,
the concentration of fructose was tested by HPLC analysis performed on an Agilent 1260 Infinity II
equipped with a refractive index detector and an Agilent ZORBAX carbohydrate analysis column (Ø4.6 ×
250 mm). The mixture solution of acetonitrile and water (3/1, v/v) was used as the mobile phase at a flow
rate of 1.0 mL/min, and the column temperature was maintained at 50 ºC. The amounts of fructose and 5-
HMF were calculated according to the external standard curves constructed with authentic standards.
10

Finally, yield of 5-HMF (Y), fructose conversion (x) and reaction selectivity (S) were calculated based on
the Eqs. (1)-(3), respectively.
nH cHVM F
Y(%) =
×100
(1)
mM H
nF cFV
x(%) = (1−
)×100
(2)
(3)
m
Y
S(%) = ×100
x
where cF and cH are the concentrations of fructose and 5-HMF (mg/mL) determined by HPLC; m is the
original amount of fructose (mg) before reaction; V is the volume of solution (mL); MF and MH are the
molecular weights of fructose and 5-HMF (mg/mol); nF and nH are the times of volume dilution for fructose
and 5-HMF, respectively.
In addition, the important byproduct of levulinic acid generated from the further acid catalyzed
degradation of 5-HMF was determined using gas chromatography (GC) performed on an Agilent 7890N
GC equipped with a flame ionization detector and an Agilent 19091N-116 HP-INNOWax chromatographic
column (60 m × 0.32 mm × 0.25
µ
m). Nitrogen was used as carrier gas. The 10
µ
L of sample was injected
1. The GC oven
and the injector temperature was set at 250 °C in split mode with split ratio of 20
:
temperature was programmed from 100 °C (3 min) to 220 °C (10 min) at 30 °C/min. The detector
temperature was maintained at 250 ºC.
2.6 Batch adsorption experiments
Adsorption kinetic experiments of fructose and 5-HMF on LDMC-SO3H were carried out in a series
of 25 mL digestion tubes. 200 mg of fructose, 20 mg of catalyst and 2 mL of DMSO were added into the
digestion tube at 45±0.2 °C. All digestion tubes were placed in a thermostatic orbital shaker (THZ-C-1,
Qiangle Lab Instruments, China) and agitated continuously at a constant speed of 150 rpm for different
11

contact time. The suspension was then filtered using 0.22 µm filter membrane, and the molar concentrations
of fructose remaining and 5-HMF in filtrate were determined using HPLC mentioned above. The amounts
adsorbed of fructose (qFt) and 5-HMF (qHt) were determined using the following Eqs. (4),(5):
(cF −cF )×180.16×V
0
t
qF =
(4)
(5)
t
mL
(cH
−
cH )
×
126.11
×V
0
t
qH
=
t
mL
where qFt and qHt are the amounts adsorbed of fructose and 5-HMF after adsorption at different contact time
(mg/g), cF0 and cH0 are the initial molar concentrations of fructose and 5-HMF (mol/L), cFt and cHt are the
molar concentrations of fructose and 5-HMF after a certain adsorption time (mol/L), V is the volume of
DMSO added into adsorption system (mL) and mL is the mass of LDMC-SO3H as adsorbent (g).
3 Results and discussion
3.1 Sulfonic acid-functionalized LDMC
To ensure an appropriate carbonization temperature for removing soft temple (F127) completely
during the preparation process of catalyst, the pyrolysis analyses of main substances involved in
experiments were conducted by means of the methods of TGA and DTA, and the results were shown in Fig.
1. It can be seen from Fig. 1a that all of the LPF, F127 and their compound start to lose weight at 190 °C.
Especially for F127, its weight loss increases dramatically at 315 °C, and then becomes gentle at 410 °C,
finally leading to zero in weight. In the meantime, the DTA curve in Fig. 1b further exhibits the maximum
weight loss rate of F127 pyrolysis occurs at 380 °C, highlighting the complete removal of F127 at
temperature higher than 400 °C. While the weight loss rate of LPF pyrolysis is relatively stable in the
temperature range from room temperature to 1000 °C. Furthermore, it is also observed in Fig. 1b that the
12

maximum weight loss rate of the pyrolysis of compound of LPF and F127 appears at 395 °C, which is close
to the result of F127, thus suggesting the weight loss of compound is mainly caused by F127 combustion.
Thus, F127 can be removed absolutely at the selected carbonization temperature of 900 °C in this
experiment.
At carbonization temperature of 900 °C, the XRD pattern in Fig. 2 of as-prepared LDMC is
characterized mainly by two peaks appeared around at 22° and 43°, respectively. The former broad
diffraction peak corresponds to the (002) reflection of graphite, illustrating the composition of non-regular
multi-ring aromatic carbon thin layers. And the latter peak with less intense is attributed to the (110) and/or
(100) planes of graphite, suggesting an occurrence of graphitic structure in LDMC.
Fig. 3 illustrates the FTIR spectra of KL and LDMC-SO3H. The FTIR spectrum of KL displays the
typical structural characteristics of KL. A strong and broad peak is observed around 3420 cm-1, which is
originated from O-H stretching, illustrating the presence of a large amount of aliphatic hydroxyl group and
phenolic hydroxyl group in KL structure. The absorption bands located around 1600 and 1515 cm-1 are
assigned to vibrations of aromatic rings in lignin. More absorption bands are observed at 1450 cm-1
(deformation vibration of the CH- group in the aromatic ring), 1370 cm-1 (symmetric deformation vibrations
of C-H in methoxyl groups), 1270 and 1230 cm-1 (stretching vibrations of guaiacyl rings), 1130 cm
(vibrations of syringyl ring) and 1030 cm-1 (deformation vibrations of the C-H bonds in the guaiacyl ring
and C-O bonds) [25]. Furthermore, it is found that the intensity of the absorption band at 1515 cm-1 is
greater than that at 1600 cm-1, and the intensity of the absorption band at 1270 cm-1 is also greater than that
at 1230 cm-1, as well as there is an intensive absorption band at 1030 cm-1, all of which are very typical for
the FTIR spectrum of softwood lignin containing more guaiacyl units [26]. For LDMC-SO3H, there occurs
a strong and broad adsorption peak around 1084 cm-1 originating from S=O bond (~1190 cm-1) and C-S
13

bond (~1040 cm-1) in its FTIR spectrum, confirming a successful introduction of -SO3H group into LDMC
structure after sulfonation.
Fig. 4 shows the SAXS patterns of LDMC prepared by the method of solvent evaporation induced
self-assembly (EISA) and the sulfonated LDMC after sulfonation treatment. The SAXS patterns of the
synthesized carbon materials from Fig. 4 reveal the well resolved (100) ( 2θ=1.0°) and the (110) and (200)
peaks, indicating an existence of ordering at long range distance associated with the two-dimensional (2D)
hexagonal symmetry (p6m mesostructure) [27]. Because the N2 adsorption-desorption isotherms at a low
relative pressure of approximately 0.01-0.04 is considered as exhibiting microporous characteristics [21],
Fig. 5 depicts the N2 adsorption-desorption isotherms in the relative pressure at P/P0 >0.05 and pore size
distribution of LDMC before and after sulfonation. Fig. 5a reveals that carbon samples are characterized
by a broad hysteresis loop at a high relative pressure (P/P0>0.4). The isotherm is type IV, a typical
mesoporous material. Meanwhile, a narrow peak in pore size distribution occurs at about 3.42 nm shown
in Fig. 5b. Notably, there is no obvious difference in SAXS pattern between LDMC and LDMC-SO3H (Fig.
4), and the same result occurs in the results of BET measurement as shown in Fig. 5. Furthermore, seen
from Table 1, the specific surface area and the mesopore volume (Vmeso) reflecting textural properties of
mesopore in LDMC are similar to those of LDMC-SO3H, indicating a good maintenance of mesoporous
structure in carbon material after sulfonation treatment. Compared with the lignin-derived solid acid (LPC-
SO3H) prepared via incomplete carbonization (at 450 °C) followed by sulfonation in our previous work
[28], there is no obvious structural shrinkage for LDMC-SO3H after the treatment of concentrated sulfuric
acid probably due to its higher crystallinity after carbonization at high temperature (900 °C used in this
work) shown in Fig. 2.
The morphology and microstructure of LDMC-SO3H inspected by transmission electron microscopy
14

is shown in Fig. 6. It is observed clearly that LDMC-SO3H exhibits hexagonally arranged and stripe-like
patterns, further authenticating a well-ordered 2D hexagonal mesoporous characteristics [27].
3.2 Conversion of fructose into 5-HMF
To evaluate the catalytic performance, LDMC-SO3H was used as solid acid catalyst for the dehydration
of fructose into 5-HMF in DMSO. Fig. 7 exhibits the influences of reaction temperature and time on the
dehydration of fructose into 5-HMF in terms of fructose conversion, 5-HMF yield and reaction selectivity
with LDMC-SO3H and without catalyst (blank tests). Seen from Fig. 7a,c,e, it is observed that fructose
conversion, 5-HMF yield and selectivity increase with prolonging reaction time in the absence of catalyst,
reflecting a certain catalytic reactivity of DMSO (solvent) in the conversion process of fructose.
Furthermore, higher temperature more exacerbates this change trend. However, by comparing the
experimental results between dehydration fructose to 5-HMF with LDMC-SO3H (Fig. 7b,d,f) and those
without catalyst (Fig. 7a,c,e), it is found that LDMC-SO3H can accelerate the conversion of fructose
significantly. For example, the fructose conversion increases from 53.0% to 92.2% with adding LDMC-
SO3H (catalyst load of 10 wt%) at 140 °C for 60 min. More importantly, it is found that in the absence of
catalyst, the 5-HMF yield (8.1%) and selectivity (10.3%) are very low at the initial reaction time, and low
temperature can prolong this initial period dramatically. It is widely believed that the dehydration of glucose
or fructose to 5-HMF catalyzed by an acid catalyst is via an open-chain or a cyclic furanose intermediate
pathway [29]. A key intermediate in the dehydration reaction of fructose to 5-HMF in DMSO at 150 °C
without catalyst was identified by Amarasekara et al [30]. It was found that there was a synergistic catalysis
between DMSO solvent and acid catalyst for this reaction. In this work, it is proposed that due to the longer
time spent on the dehydration of fructose to intermediates in the absence of catalyst compared to that in the
present of LDMC-SO3H as catalyst, 5-HMF is not formed during the first period of time. In other word, the
15

prepared LDMC-SO3H can accelerate the intermediate reactions to shorten the time for generating 5-HMF
from fructose in DMSO solvent.
In the meanwhile, it was observed obviously in Fig. 7b that the conversion of fructose increases with
prolonging time until it reaches a high equilibrium conversion and is even completely transformed. Notably,
increasing reaction temperature can shorten the reaction time required to achieve the equilibrium
conversion or the full conversion of fructose. Within the scope of experiments, fructose conversion, 5-HMF
yield and selectivity reach the maximum values of 100.0%, 98.0% and 98.0% when reaction temperature
and time are 140 °C and 120 min, respectively. It is particularly noteworthy that 5-HMF yield as well as
reaction selectivity increase dramatically to the maximum values first but then decline gradually with time.
Research over the years has revealed that in addition to the main reaction of dehydration of fructose into 5-
HMF, the transformation process of fructose catalyzed by acids is complicated owing to the simultaneous
existence of side reactions including tandem and/or parallel reactions [2,31,32]. With prolonging reaction
time to some extent, side reactions could occur under enough reaction time. For example, the accumulation
of 5-HMF produced from fructose-to-5-HMF transformation will accelerate the further degradation of 5-
HMF to other substances with smaller molecular weight such as levulinic acid [32-35], thus leading to the
gradual decreases in 5-HMF yield and reaction selectivity. Under the experimental conditions of reaction
temperature and time of 140 °C and 120 min, 5-HMF yield and selectivity are only 8.1% and 10.3% without
catalyst, but both of them increase dramatically to the maximum 5-HMF yield (98.0%) with adding LDMC-
SO3H (catalyst load of 10 wt%). It is suggested that LDMC-SO3H can accelerate the fructose conversion
and has highly selective catalysis for the conversion of fructose to 5-HMF in DMSO within 120 min. And
when reaction time is prolonged from 120 min to 180 min, some 5-HMF is rehydrated to levulinic acid (4.0
mol%), thus slightly decreasing 5-HMF yield from the maximum value of 98.0% to 93.8% (Fig. 7d), which
16

basically corresponds to the stoichiometry of rehydration reaction of 5-HMF to levulinic acid. In addition,
based on the mass balance of 5-HMF, a small amount of 5-HMF (about 0.2 mol%) is probably transformed
into the insoluble humins.
3.3 Reaction kinetics study
Kinetics study of fructose dehydration into 5-HMF using LDMC-SO3H as a catalyst was investigated.
Previous studies have certified the assumption that the conversion of fructose into 5-HMF is inclined to
obey a pseudo-first order reaction process when fructose is dehydrated into 5-HMF (the sole and stable
degradation product) with high selectivity [11,36]. Thus, under experimental conditions, the rate of fructose
conversion can be expressed as follows (Eq. (6)) according to the first order reaction:
dcF
= −kcF
(6)
dt
which, integrated, lead to Eq. (7):
cF
1
0
kt = ln
= ln
(7)
cF
1− xt
t
where k is the pseudo-first-order rate constant of fructose dehydration at a certain reaction temperature
(min-1); t is the reaction time (min); and xt is the conversion of fructose at a certain reaction time (%).
Results of the data fitting are listed in Table 2. The correlation coefficient R2 shows relatively good
linearity at different reaction temperature, which supports the hypothesis of pseudo-first order reaction for
the dehydration of fructose into 5-HMF. Furthermore, as listed in Table 2, the obtained rate constant (k) of
fructose-to-5-HMF transformation increases when the reaction temperature increases from 120 to 160 °C,
thus quantitatively confirming an obvious promotion of fructose conversion into 5-HMF via raising reaction
temperature. In addition, the reaction rate constant follows Arrhenius equation (Eq. (8)), and its logarithmic
form is shown as Eq. (9).
17

Ea
)
k = Ae(−
(8)
(9)
RT
Ea
lnk = −
+ ln A
RT
where A is the pre-exponential factor (min-1) and Ea is the activation energy (kJ/mol). Results of the data
fitting for the plot of (ln k) as a function of (1/T) based on the Arrhenius equation are also listed in Table 2.
As shown in Table 2, the obtained Ea and A for the fructose conversion without catalyst and with LDMC-
SO3H as catalyst are 114 kJ/mol, 3.13 × 1012 min-1 and 72 kJ/mol, 7.04 × 107 min-1, respectively. It further
affirms that the addition of LDMC-SO3H as catalyst can decrease the reaction activation energy and
promote the efficient conversion of fructose significantly, and it is suggested that the synthesized LDMC-
SO3H in this work has a comparable efficiency as a catalyst for fructose dehydration to 5-HMF in DMSO
over other efficient porous solid acids reported in literature [11], such as sulfonic acid-functionalized metal-
organic frameworks (55 kJ/mol), HZSM-5 (Si/Al=25) (70 kJ/mol), bifunctional catalyst prepared by
sulfonic acid- and ionic liquid-functionalization over mesoporous silica nanoparticles (68 kJ/mol) under
the reaction conditions of temperature of 120 °C, time of 60 min and catalyst load of 60 wt% in DMSO.
Fig. 8a exhibits the comparison of catalytic performance of different catalysts including three typical
commercial solid acids of HZSM-5, Amberlyst 15 and phosphotungstic acid (PTA) for the conversion of
fructose into 5-HMF under the previous optimized conditions. As shown in Fig. 8a, fructose conversion
(100%), 5-HMF yield (98.0%) and reaction selectivity (98.0%) obtained with LDMC-SO3H as catalyst are
much higher than those with HZSM-5 as catalyst. In the meanwhile, the catalytic performance of LDMC-
SO3H for production of 5-HMF from fructose dehydration is comparable to that of Amberlyst 15 or PTA,
highlighting a good prospect for catalytic application of LDMC-SO3H in the conversion of fructose into 5-
HMF. Furthermore, to verify the behavior of catalyst, the effect of reaction time on the fructose conversion
reaction with three different catalyst (Amberlyst 15, PTA and LDMC-SO3H) were investigated, and the
18

results are shown in Fig. 8b. Seen from Fig. 8b, the reaction time to reach the maximum fructose conversion
and 5-HMF yield is within 40 min using PTA as catalyst, which is shorter than that using Amberlyst 15 or
LDMC-SO3H as catalyst, highlighting the catalytic efficiency of PTA. Notably, it is observed that although
the variation trend of fructose conversion with reaction time using Amberlyst 15 is similar to that using
LDMC-SO3H, the 5-HMF yield obtained using Amberlyst 15 is much lower than that using LDMC-SO3H
during the experimental time (~120 min). As is known, both of Ameberlyst 15 and PTA have very high acid
density [37]. In addition, Bispo et al [38] proposed that when the loading of -SO3H group on sulfonic acid-
functionalized periodic mesoporous organosilicas was not less than 0.56 mmol/g, other particular factors
such as the problem of acid sites accessibility might also influence the activity of catalyst. Therefore, the
result in this work implies that the ordered mesoporous structure of LDMC-SO3H bearing less acid sites
(0.6528 mmol/g of -SO3H group density) compared to Ameberlyst 15, could presumably contribute a lot to
promote the conversion of fructose into 5-HMF via providing pore channels for the acid sites accessibility.
3.4 Catalytic mechanism
It has been reported that mass transfer limitations often result in a reduced selectivity for exothermic
reactions, but the similar result was also found in the dehydration of fructose into 5-HMF as an endothermic
reaction [11]. To confirm the contribution of pore channels of the lignin-derived mesoporous carbon to the
heterogeneous catalytic system, adsorption studies of fructose and 5-HMF onto LDMC-SO3H in DMSO
solvent were carried out, respectively. Fig. 9 illustrates the adsorption kinetic curves of fructose and 5-HMF
on LDMC-SO3H under the experimental conditions of LDMC-SO3H of 20 mg as adsorbent, DMSO of 2
mL, initial concentration of fructose or 5-HMF of 100 mg/mL and adsorption temperature of 45 °C.
Interestingly, it is observed that there are different stages occurred during adsorption process. In the first
stage, a small amount of fructose is adsorbed gradually onto LDMC-SO3H during the initial 50 min. And
19

then, the adsorbed amount of fructose increases dramatically with prolonging adsorption time to 90 min in
the second stage. Finally, in the third stage, the adsorption kinetic curve of fructose on catalyst reaches a
platform within the experimental range of adsorption time from 90 to 240 min, and the equilibrium
adsorption capacity of fructose on LDMC-SO3H is determined to 691.4 mg/g. However, compared to
fructose, 5-HMF is barely adsorbed onto catalyst during its whole adsorption process, highlighting an
obvious difference in adsorption behavior on LDMC-SO3H between fructose and 5-HMF.
Based on the data shown in Fig. 7, the fructose conversion without catalyst (42.8%) is lower than that
with LDMC-SO3H (82.7%) at 30 min; at the same time, the former 5-HMF yield is close to zero, but the
latter 5-HMF yield is obtained to 46.9%. It is suggested that although fructose can be converted in DMSO
at high temperature, the fructose-to-5-HMF transformation only quickly occurs on the LDMC-SO3H when
fructose is gradually adsorbed onto LDMC-SO3H in the first adsorption stage. With prolonging time from
60 to 120 min, the 5-HMF yield without catalyst varies slightly from 1.0% to 8.1% which can be neglected,
but the 5-HMF yield in the presence of LDMC-SO3H increases dramatically from 77.7% to 98.0%.
Interestingly, it is found that the difference in 5-HMF yield increases significantly (from 46.8% at 30 min
to 76.7% at 60 min to 89.9% at 120 min), which accords with the result that larger amount of fructose is
adsorbed onto LDMC-SO3H and contacts its acid sites for achieving higher 5-HMF yield in the second
adsorption stage mentioned above. It can be inferred that the yield of 5-HMF increases significantly at the
initial reaction stage until reaching the maximum value (98.0%), which is probably due to the increasing
adsorbed amount of fructose onto LDMC-SO3H. The result also demonstrates that the dehydration process
of fructose to intermediates can be sped up dramatically and 5-HMF can be obtained quickly through the
contact between fructose and LDMC-SO3H, highlighting the efficient catalytic activity of LDMC-SO3H in
the dehydration of fructose to 5-HMF.
20

In addition, because LDMC-SO3H can hardly adsorb 5-HMF within the experimental time range of
0~240 min, the 5-HMF generated during the dehydration process of fructose can be removed away from
the acid sites of LDMC-SO3H into DMSO solvent instantaneously, thus enhancing the reaction selectivity
significantly via greatly reducing the further acid catalyzed degradation of 5-HMF into byproducts [32-35].
Moreover, due to the more hydrophobic nature of HMF than fructose, it can be speculated reasonably that
the as-prepared LDMC-SO3H has a suitable hydrophobic/hydrophilic balance of the surfaces, which is more
favorable for the faster departure of the generated 5-HMF, thus retarding the rehydration of 5-HMF to
byproducts and improving selectivity significantly [39].
To further clarify the correlation between mesoporous structure of LDMC-SO3H and its catalytic
mechanism for fructose-to-5-HMF transformation, four adsorption kinetic models including pseudo-first-
order, pseudo-second-order, Weber-Morris (intra-particle diffusion) and Bangham (two-constant rate)
equations were employed to fit the adsorption kinetic curve of fructose on LDMC-SO3H shown in Fig. 9.
The linear fitting results are illustrated in Fig. 10.
The integral expression of pseudo-first-order kinetic model proposed by Lagergren as Eq. (10):
k1
log(qF −qF ) = logqF −
t
(10)
e
t
e
2.303
where k1 is the pseudo-first-order rate constant (min-1) and qFe is the equilibrium amount adsorbed of
fructose after equilibrium adsorption (mg/g). Value of k1 was calculated from the slope of the plots of
2.303×log (qFe/(qFe-qFt)) versus t (Fig. 10a). The pseudo-second-order rate equation can be followed as Eq.
(11):
t
1
t
=
+
2
(11)
(12)
qF
qF
k2qF
t
e
e
2
h = k2qF
e
where k2 is the pseudo-second-order rate constant (g/(mg·min)), h expresses the initial adsorption rate
21

(mg/(g·min)) (Eq. (12)). Values of k2 and qFe were calculated from the plots of t/qFt versus t (Fig. 10b).
In general, the adsorption process onto porous adsorbents includes four major steps: (a) external
diffusion in bulk liquid-phase, (b) boundary layer diffusion, (c) intra-particle (pore) diffusion within particle
and (d) adsorption on active sites [40]. Among them, adsorption occurs instantaneously and its effect on
adsorption rate can be neglected. Thus, the main resistances to mass transfer are liquid film diffusion
(including steps (a) and (b)) and intra-particle diffusion. To further predict the rate-limiting step of
adsorption process, Weber-Morris model (intra-particle diffusion model) is suggested, which can be
described by the following Eq. (13):
qF = kipt0.5 + C
(13)
t
where kip is the intra-particle diffusion rate constant (mg/(g·min0.5)), C is a constant being closely related
with the thickness of boundary layer (mg/g). If C
≠ 0, indicates that intra-particle diffusion is not the only
controlling step and more than one diffusive resistance are involved in the adsorption process. Notably,
multilinearities can be observed from the linear plot of qFt versus t0.5 (Fig. 10c) and the plots do not pass
through the origin, demonstrating that the intra-particle diffusion controls the adsorption process but not
the only rate-limiting step. In other words, the mass transfer in adsorption process is mainly influenced and
controlled by liquid film diffusion and intra-particle diffusion. As defined in Eq. (13), the slopes of the first
linear section (ki1) and the second linear section (ki2) are the diffusion rate constants for different adsorption
stages.
The Bangham kinetic model is expressed as Eq. (14), which is often known as the two-constant rate
equation and usually employed to describe the mechanism of pore diffusion in the complex heterogeneous
diffusion process [41,42].
lnqF =lnν +kb lnt
(14)
t
22

where kb is the rate constant (mg/(g·min)) and
can be determined from the linear plot of ln qFt versus ln t (Fig. 10d).
Seen from Fig. 10, comparing the correlation coefficients (R2) of four adsorption kinetic equations,
ν is also a constant indicating the adsorption intensity. kb and
ν
the adsorption process of fructose onto LDMC-SO3H more follows the Bangham kinetic model (R2 =
0.90337) and the Weber-Morris kinetic model (R2 = 0.81561 for the first adsorption stage, R2 = 0.90580 for
the second adsorption stage). Furthermore, for Weber-Morris model, the calculated value of ki1 (12.61999)
is much lower than that of ki2 (252.3087), highlighting that the adsorption rate of fructose onto LDMC-
SO3H is more controlled by liquid film diffusion (as rate-limiting step) in its initial adsorption process (the
first stage). Based on the combination of the comprehensive analyses above and the structural
characteristics of LDMC-SO3H, it is suggested that the adsorption kinetic fitting results conform to the
diffusion-controlling adsorption of reactant onto porous catalyst during the heterogeneous catalysis process.
Both of liquid film diffusion and intra-particle (pore) diffusion contribute to the adsorption mechanism of
fructose on LDMC-SO3H. Moreover, the mass transfer rate of fructose may be increased via accelerating
stirring to decrease the resistance of liquid film and/or enlarging pore size as well as porosity of catalyst to
facilitate the intra-particle (pore) diffusion.
Combining the analyses of reaction kinetics and adsorption kinetics above, the catalytic mechanism
of LDMC-SO3H in the dehydration of fructose into 5-HMF with high conversion and selectivity is proposed
as shown in Scheme 2. As can be seen, the well-ordered mesostructure provides LDMC-SO3H with larger
area for grafting more -SO3H groups to improve the fructose contact with acid sites and the suitable paths
(pore channels) for the diffusion of the fructose and the generated 5-HMF, which contributes to the high
catalytic performance of LDMC-SO3H in the conversion of fructose to 5-HMF.
3.5 Stability in catalytic performance
23

Besides catalytic activity, the stability of catalyst is another vital property for demonstrating its
catalytic performance under the background of green and sustainable chemistry. To investigate the
reusability of LDMC-SO3H in the dehydration of fructose to 5-HMF in DMSO, a ten-run experiment was
performed, and the results are shown in Fig. 11. Notably, under the conditions of catalyst load of 8.75 wt%
(mass ratio of catalyst to fructose), reaction temperature and time of 140 °C and 2 h, the fructose conversion
decreases slightly from 98.0% to 95.8% along with a decrease in 5-HMF yield from 92.1% to 88.2% in
former six runs owing to the existence of minor deactivation of catalyst, but both of fructose conversion
and 5-HMF yield have obvious decreases in later four runs. First of all, compared to other sulfonic acid-
functionalized porous materials used as catalyst in fructose-to-5-HMF transformation in DMSO listed in
Table 3 [2,8,11,43] , the as-prepared catalyst in this work exerts better stability in catalytic performance in
terms of 5-HMF yield with a high 5-HMF selectivity (>90%) in former six runs. The effect of catalyst load
on the catalytic performance in fructose dehydration to 5-HMF is investigated, and the corresponding result
demonstrated in Fig. S1 shows that 5-HMF yield does not decrease linearly with decreasing catalyst load.
As shown in Fig. S1, when catalyst load decreases from 8.75 wt% to 3.75 wt%, 5-HMF yield decreases
slightly from 91.9% to 88.1%, the variation curve is relatively smooth. However, with further decreasing
catalyst load from 3.75 wt% to 1.25 wt%, 5-HMF yield decreases from 88.1% to 74.7%, and even decreases
to 8.1% in the absence of solid catalyst. Interestingly, 5-HMF yield increases significantly from 8.1%
(without adding anything) to 34.3% with adding LDMC before sulfonation, suggesting that LDMC without
–SO3H groups also exerts the catalytic performance in fructose dehydration to 5-HMF probably because
LDMC with an ordered mesostructure provides mesopore channels with special surfaces for the
dehydration reaction of fructose in DMSO. Yang et al [44] indicated that the hydrophobicity of
functionalized silica nanoparticles was beneficial to both of 5-HMF yield and selectivity. Inagaki et al [45]
24

proposed that the catalyst with orderly arranged hydrophobic-hydrophilic surfaces could enclose the guest
molecules or clusters with structural orientation in its pores, thus leading to an increase in the activity and
selectivity. Similarly, for LDMC-SO3H in this work, the structure with suitable hydrophobic/hydrophilic
balance of its surfaces can be beneficial to a large amount of capture of the fructose as well as a fast
departure of the 5-HMF produced from fructose dehydration reaction within the pore channels of catalyst,
which finally improves catalytic activity and reaction selectivity. Thus, it is suggested that the high 5-HMF
yield and selectivity in fructose dehydration is a synergistic effect of mesopore structure with a special
surface and -SO3H groups with a relatively high content in LDMC-SO3H.
The deactivation of functionalized solid acid in fructose dehydration to 5-HMF could be due to the
obstruction of insoluble humins generated in side reaction or the leaching of active acid sites after repeated
catalytic cycles [2,44]. As discussed in 3.2 Conversion of fructose into 5-HMF, there is almost no humins
produced under the optimized conditions. Therefore, the loss of -SO3H groups (acid sites) may be the main
reason for the deactivation of LDMC-SO3H, which has a much more obvious effect on the catalytic
performance in later four runs than that in former six runs in our work (Fig. 11), thus resulting in a nonlinear
variation trend for ten-run experiment. It is proposed that when the content of -SO3H group in LDMC-
SO3H is not very low, the problems of acid sites accessibility and reaction interface probably also influence
the catalytic performance of catalyst, which is similar to the description by Bispo et al [38]. Conversely,
the density of -SO3H group in LDMC-SO3H is the key factor affecting its catalytic performance in the
conversion of fructose to 5-HMF.
4 Conclusions
In summary, owing to the characteristics of lignin chemical structure, sulfonic acid-functionalized
well-ordered mesoporous carbon derived from kraft lignin was prepared successfully by adopting the
25

methods of phenolation, EISA and sulfonation. Compared with other solid acid catalysts reported, the as-
prepared LDMC-SO3H exhibited superior catalytic performance in the conversion of fructose into 5-HMF.
A high 5-HMF yield of up to 98.0% with a full fructose conversion was obtained under the conditions of
fructose of 200 mg, LDMC-SO3H of 20 mg (catalyst load of 10 wt%), DMSO of 2 mL, reaction temperature
and time of 140 °C and 2 h. Results revealed that the conversion reaction of fructose catalyzed by LDMC-
SO3H in DMSO solvent might obey pseudo-first-order kinetics with an observed activation energy of 72
kJ/mol which was lower than the activation energy of 114 kJ/mol for the reaction in the absence of catalyst.
Interestingly, it was found that LDMC-SO3H was inclined to adsorb fructose rather than 5-HMF, which
probably contributed a lot to the efficient catalysis with a high selectivity of 98.0% for fructose-to-5-HMF
transformation. Moreover, adsorption kinetic analysis indicated that the adsorption of fructose onto LDMC-
SO3H more followed Bangham kinetic model and Weber-Morris kinetic model, which accorded with the
result of diffusion-controlling adsorption on porous catalyst, further discovering the probable catalytic
mechanism of LDMC-SO3H in the dehydration of fructose into 5-HMF besides its sulfonic acid groups as
Brønsted acid sites. Most importantly, LDMC-SO3H possessed superior reusability with excellent stability
of the relatively high 5-HMF yield for fructose dehydration in DMSO.
Declarations of interest: none
Acknowledgements
The authors gratefully acknowledge the financial support by the National Natural Science Foundation
of China (grant No. 21706085).
References
[1] I.K.M. Yu, D.C.W. Tsang, Conversion of biomass to hydroxymethylfurfural: A review of catalytic
26

systems and underlying mechanisms, Bioresour. Technol. 238 (2017) 716-732.
[2] H. Hafizi, A. Najafi, Chermahini, M. Saraji, G. Mohammadnezhad, The catalytic conversion of
fructose into 5-hydroxymethylfurfural over acid-functionalized KIT-6, an ordered mesoporous silica,
Chem. Eng. J. 294 (2016) 380-388.
[3] B. Karimi, H.M. Mirzaei, H. Behzadnia, H. Vali, Novel ordered mesoporous carbon based sulfonic
acid as an efficient catalyst in the selective dehydration of fructose into 5-HMF: the role of solvent and
surface chemistry, ACS Appl. Mater. Interfaces 7 (2015) 19050-19059.
[4] Z. Wu, Y. Meng, D. Zhao, Nanocasting fabrication of ordered mesoporous phenol-formaldehyde resins
with various structures and their adsorption performances for basic organic compounds, Microporous
Mesoporous Mater. 128 (2010) 165-179.
[5] X. Wang, P. Liu, Y. Tian, Ordered mesoporous carbons for ibuprofen drug loading and release behavior,
Microporous Mesoporous Mater. 142 (2011) 334-340.
[6] S. Tanaka, A. Doi, T. Matsui, Y. Miyake, Mass transport and electrolyte accessibility through
hexagonally ordered channels of self-assembled mesoporous carbons, J. Power Sources 228 (2013) 24-
31.
[7] J.M.R. Gallo, R. Alamillo, J.A. Dumesic, Acid-functionalized mesoporous carbons for the continuous
production of 5-hydroxymethylfurfural, J. Mol. Catal. A: Chem. 422 (2016) 13-17.
[8] J. Wang, Z. Zhang, S. Jin, X. Shen, Efficient conversion of carbohydrates into 5-hydroxylmethylfurfan
and 5-ethoxymethylfurfural over sufonic acid-functionalized mesoporous carbon catalyst, Fuel 192
(2017) 102-107.
[9] W. Libbrecht, A. Verberckmoes, J.W. Thybaut, P.V.D. Voort, J.D. Clercq, Soft templated mesoporous
carbons: Tuning the porosity for the adsorption of large organic pollutants, Carbon 116 (2017) 528-
27

546.
[10]Y. Meng, D. Gu, F. Zhang, Y. Shi, L. Cheng, D. Feng, Z. Wu, Z. Chen, Y. Wan, A. Stein, D. Zhao, A
family of highly ordered mesoporous polymer resin and carbon structures from organic-organic self-
assembly, Chem. Mater. 18 (2006) 4447-4464.
[11]J. Chen, K. Li, L. Chen, R. Liu, X. Huang, D. Ye, Conversion of fructose into 5-hydroxymethylfurfural
catalyzed by recyclable sulfonic acid-functionalized metal-organic frameworks, Green Chem. 16 (2014)
2490-2499.
[12]Suhas, P.J.M. Carrott, M.M.L.R. Carrott, Lignin - from natural adsorbent to activated carbon: a review,
Bioresour. Technol. 98 (2007) 2301-2312.
[13]R.K. Sharma, J.B. Wooten, V.L. Baliga, X.H. Lin, W.G. Chan, M.R. Hajaligol, Characterization of
chars from pyrolysis of lignin, Fuel 83 (2004) 1469-1482.
[14]S. Zhao, M.M. Abu-Omar, Synthesis of renewable thermoset polymers through successive lignin
modification using lignin-derived phenols, ACS Sustainable Chem. Eng. 5 (2017) 5059-5066.
[15]J. Podschun, B. Saake, R. Lehnen, Reactivity enhancement of organosolv lignin by phenolation for
improved bio-based thermosets, Eur. Polym. J. 67 (2015) 1-11.
[16]S. Yang, J. Wen, T. Yuan, R. Sun, Characterization and phenolation of biorefinery technical lignins for
lignin-phenol-formaldehyde resin adhesive synthesis, RSC Adv. 4 (2014) 57996-58004.
[17]D. Saha, K.E. Warren, A.K. Naskar, Soft-templated mesoporous carbons as potential materials for oral
drug delivery, Carbon 71 (2014) 47-57.
[18]D. Saha, K.E. Warren, A.K. Naskar, Controlled release of antipyrine from mesoporous carbons,
Microporous Mesoporous Mater. 196 (2014) 327-334.
[19]D. Saha, E.A. Payzant, A.S. Kumbhar, A.K. Naskar, Sustainable mesoporous carbons as storage and
28

controlled-delivery media for functional molecules, ACS Appl. Mater. Interfaces 5 (2013) 5868-5874.
[20]D. Saha, Y. Li, Z. Bi, J. Chen, J.K. Keum, D.K. Hensley, H.A. Grappe, H.M. Meyer, III, S. Dai, M.P.
Paranthaman, A.K. Naskar, Studies on supercapacitor electrode material from activated lignin-derived
mesoporous carbon, Langmuir 30 (2014) 900-910.
[21]H. Qin, R. Jian, J. Bai, J. Tang, Y. Zhou, B. Zhu, D. Zhao, Z. Ni, L. Wang, W. Liu, Q. Zhou, X. Li,
Influence of molecular weight on structure and catalytic characteristics of ordered mesoporous carbon
derived from lignin, ACS Omega 3 (2018) 1350-1356.
[22]Y. Song, J. Liu, K. Sun, W. Xu, Synthesis of sustainable lignin-derived mesoporous carbon for
supercapacitors using a nano-sized MgO template coupled with Pluronic F127, RSC Adv. 7 (2017)
48324-48332.
[23]F. Chen, W. Zhou, H. Yao, P. Fan, J. Yang, Z. Fei, M. Zhong, Self-assembly of NiO nanoparticles in
lignin-derived mesoporous carbons for supercapacitor applications, Green Chem. 15 (2013) 3057-3063.
[24]W. Zhao, X. Lin, H. Cai, T. Mu, X. Luo, Preparation of mesoporous carbon from sodium lignosulfonate
by hydrothermal and template method and its adsorption of uranium(VI), Ind. Eng. Chem. Res. 56
(2017) 12745-12754.
[25]L. Gan, M. Zhou, D. Yang, X. Qiu, Preparation and evaluation of carboxymethylated lignin as
dispersant for aqueous graphite suspension using turbiscan lab analyzer, J. Disper. Sci. Technol. 34
(2013) 644-650.
[26]I. Bykov, Characterization of natural and technical lignins, Luleå Tekniska Universitet (2008).
[27]W. Libbrecht, F. Deruyck, H. Poelman, A. Verberckmoes, J. Thybaut, J.D. Clercq, P.V.D. Voort,
Optimization of soft templated mesoporous carbon synthesis using definitive screening design, Chem.
Eng. J. 259 (2015) 126-134.
29