Importance of the synergistic effects between cobalt sulfate and tetrahydrofuran for selective production of 5-hydroxymethylfurfural from carbohydrates

Article abstract of DOI:10.1039/d0cy00225a

In this study, an effective catalytic system (CoSO₄·7H₂O/THF) for selective conversion of fructose to 5-hydroxymethylfurfural (HMF; yield: 88%) was developed. The synergistic effects among Co²⁺, SO₄^2-, crystal water and tetrahydrofuran (THF) were crucial for achieving selective dehydration of fructose to HMF. Co²⁺ worked as a Lewis acid for catalyzing mainly dehydration of fructose to HMF but not the further decomposition of HMF to levulinic acid. THF could help to retain HMF while CoSO₄ could coordinate with HMF, enhancing the thermal stability of HMF in THF. The crystal water in cobalt sulfate could help to coordinate with fructose, which facilitated the conversion of fructose via dehydration reactions. The CoSO₄·7H₂O/THF catalytic system could also catalyze the conversion of inulin and cellulose into HMF. The main advantages of the CoSO₄·7H₂O/THF catalytic system are the low cost, the easy recycling of the CoSO₄·7H₂O catalyst and the easy separation of HMF from volatile THF.

Full text of DOI:10.1039/d0cy00225a

View Article Online
View Journal
Catalysis
Science &
Technology
Accepted Manuscript
This article can be cited before page numbers have been issued, to do this please use: K. Sun, Y. Shao, Q.
Li, L. Zhang, Z. Ye, D. Dong, S. Zhang, Y. Wang, X. Li and X. Hu, Catal. Sci. Technol., 2020, DOI:
10.1039/D0CY00225A.
Volume
Number
21 January 2017
Pages 313-534
7
2
This is an Accepted Manuscript, which has been through the
Royal Society of Chemistry peer review process and has been
accepted for publication.
Catalysis
Science &
Technology
rsc.li/catalysis
Accepted Manuscripts are published online shortly after acceptance,
before technical editing, formatting and proof reading. Using this free
service, authors can make their results available to the community, in
citable form, before we publish the edited article. We will replace this
Accepted Manuscript with the edited and formatted Advance Article as
soon as it is available.
You can find more information about Accepted Manuscripts in the
Information for Authors.
Please note that technical editing may introduce minor changes to the
text and/or graphics, which may alter content. The journal’s standard
Terms & Conditions and the Ethical guidelines still apply. In no event
shall the Royal Society of Chemistry be held responsible for any errors
or omissions in this Accepted Manuscript or any consequences arising
from the use of any information it contains.
ISSN 2044-4761
PAPER
Chaoqiu Chen, Yong Qin et al.
Highly dispersed Pt nanoparticles supported on carbon
nanotubes produced by atomic layer deposition for hydrogen
generation from hydrolysis of ammonia borane
rsc.li/catalysis

Page 1 of 29
Catalysis Science & Technology
View Article Online
DOI: 10.1039/D0CY00225A
Importance of the synergistic effects between cobalt
sulfate and tetrahydrofuran for selective
production of 5-hydroxymethylfurfural from
carbohydrates
Kai Suna, Yuewen Shaoa, Qingyin Lia, Lijun Zhanga, Zhengmao Yea, Dehua Donga, Shu Zhangb, Yi Wangc,
Xueli Li*d, Xun Hu*a
aSchool of Materials Science and Engineering, University of Jinan, Jinan 250022, P. R. China.
bCollege of Materials Science and Engineering, Nanjing Forestry University, Nanjing, 210037,
P. R. China.
cState Key Laboratory of Coal Combustion, Huazhong University of Science and Technology,
Wuhan, 430074, P. R. China.
dShaanxi Key Laboratory of Chemical Reaction Engineering, Department of Chemistry and
Chemical Engineering, Yan′an University, Yan′an, 716000, P. R. China.
*Corresponding author. Tel. /fax: +86–531–89736201;
E-mail: Xueli Li: lzlixueli@163.com, Xun Hu: Xun.Hu@outlook.com
Submit to
Catalysis Science & Technology
For consideration for publication
1

Catalysis Science & Technology
Page 2 of 29
View Article Online
DOI: 10.1039/D0CY00225A
Abstract
In this study, an effective catalytic system (CoSO47H2O/THF) for selective
conversion of fructose to 5-hydroxymethylfurfural (HMF, yield: 88%), was developed.
2-
The synergistic effects among Co2+, SO4 , the crystal water and THF were crucial for
achieving selective dehydration of fructose to HMF. Co2+ worked as Lewis acid for
catalyzing mainly dehydration of fructose to HMF but not the further decomposition of
HMF to levulinic acid. THF could help to retain HMF while CoSO4 could coordinate
with HMF, enhancing the thermal stability of HMF in THF. The crystal water in cobalt
sulfate could help to coordinated with fructose, which facilitated the conversion of
fructose the dehydration reactions. CoSO47H2O/THF catalytic system could also
catalyze the conversion of inulin and cellulose into HMF. The main advantages for
CoSO47H2O/THF catalytic system are the low cost, the easy recycle of CoSO47H2O
catalyst and the easy separation of HMF from the volatile THF.
Keywords:
fructose/glucose/inulin/cellulose,
5-hydroxymethylfurfural,
CoSO47H2O/THF, synergistic effects, acid-catalysed conversion
2

Page 3 of 29
Catalysis Science & Technology
View Article Online
1. Introduction
DOI: 10.1039/D0CY00225A
5-Hydroxymethylfurfural (HMF) is an important platform chemical for manufacturing
the value-added products such as 2,5-diformylfuran [1], 2,5-furandicarboxylic acid (FDCA)
[2, 3] or be potentially used in the prevention and treatment of cardiovascular disease [4].
How to efficiently and selectively produce HMF from monosaccharides [5-7], cellulose [8, 9],
or biomass [10-12] has been a research foci in biorefinery. Catalyst and reaction medium are
essential parameters affecting selectivity for HMF production from biomass or biomass
derived sugars [13, 14].
In existing literatures, the acid catalysts employed for the conversion of the sugars to
HMF include mineral acids [15], solid acidic resin catalysts [16, 17], organic acids [18] and
some salts with Lewis acid sites [19-21]. Mineral acids and solid acidic resin catalysts like
Amberlyst 70 are Brønsted acids and they could not only catalyze the dehydration of fructose
to HMF, but also the further conversion of HMF to levulinic acid (LA), due to their strong
acidity [22]. The organic acids have lower acidity and the effectiveness for suppression of the
secondary decomposition of HMF strongly depends on the pH of the reaction medium [23].
Furthermore, how to separate the liquid organic acid and HMF is a remaining issue to be
considered. As for the use of salts as the Lewis acid catalysts for the conversion of sugars to
HMF, Zhao and the co-workers [24] found that CrCl2 could effectively transform glucose into
HMF in ionic liquid-water. CrCl2 dissolved in ionic liquid-water and interacted with ionic
-
liquid to form CrCl3 , which played an important role in the isomerization of glucose into
fructose through improving the hydride transfer. Nevertheless, how to separate HMF from
ionic liquid, which is not volatile, is also an issue to be resolved.
Distillation is a simple method for the separation of products from reactants, which,
however, would be energy-intensive if the boiling point of the reaction medium (solvent) is
3

Catalysis Science & Technology
Page 4 of 29
View Article Online
too high. Unfortunately, the separation of HMF from ionic liquid via distiDllOaI:ti1o0.n1039is/D0nCoY0t0225A
cost-effective. Furthermore, the ionic liquid used in the work is not an abundantly available
feedstock with cheap price. It is necessary to find a simple, low cost while robust catalytic
system for the effective conversion of sugars to HMF. The catalytic system should include
both the catalyst that could achieve the selective conversion of the sugars to HMF and the
reaction medium that could achieve the cost-effective separation of HMF from the reaction
medium. Although dimethyl sulfoxide (DMSO) could effectively inhibit the secondary
decomposition reactions to achieve high yield of HMF [25, 26], DMSO has a high boiling
point and was difficult to be separated from HMF.
With this in mind, we made an effort to develop the effective catalytic system for the
conversion of sugars to HMF by selecting the organics with the low boiling point as the
reaction medium and the salts with the properties of Lewis acid as the catalyst. The use of the
Lewis acid catalyst, instead of Brønsted acids, was to minimize the secondary decomposition
of HMF. Initially, we screen the catalytic activity of some sulfates for the conversion of
fructose/glucose in tetrahydrofuran (THF) and found that CoSO47H2O/THF was an effective
catalytic system for the selective conversion of fructose to HMF, with the yield of HMF up to
88%. The coordination between CoSO47H2O and THF played an essential role for the
selective conversion of fructose into HMF. In addition, we found that the CoSO47H2O/THF
could also effectively catalyse the conversion of inulin and cellulose into HMF with no
organic acid by-products like LA formed. More importantly, CoSO47H2O has a negligible
solubility in THF and can be separated from the liquid mixture via simple filtration, while
THF and HMF can also be separated via distillation at low cost, due to the high volatility of
THF.
4

Page 5 of 29
Catalysis Science & Technology
View Article Online
2.Materials and Methods
DOI: 10.1039/D0CY00225A
2.1 Materials
Fructose, glucose, the metal salts (Na2SO4, K2SO4, Fe2(SO4)3, NiSO46H2O, CuSO4,
CoSO47H2O, CoCl26H2O, Co(NO3)26H2O, Co(C2H3O2)26H2O) and the organic solvents
(THF, DMSO, methanol, ethanol, isopropanol, acetic acid, acetone and furfural) were
analytical grade and were supplied from Sinopharm Chemical Reagent Co., Ltd. Inulin was
purchased from Shandong west Asia chemical industry Co., Ltd. Cellulose (AR, partical size:
65 μm) and HMF (99%) were purchased from Shanghai Macklin Biochemical Co., Ltd.
2.2 Catalytic tests
The catalytic conversion of sugars/inulin/cellulose to HMF with the salts as catalyst and
the different solvent as reaction medium was conducted in an autoclave reactor with a volume
of 10 mL. After loading feedstock, catalyst, and solvent, the reactor was purged with
pressurized N2 (99.99%) for removing the residual air for 3 times, which was then heated to
the targeting temperature in 10 min and maintained for 120 min. After finishing the
experiments, the reactor was quenched, and the liquid products as well as the solid products
were then collected and separated for further analysis.
2.3 Characterizations
The distribution of the liquid products was analysed by a Shimadzu GC-MS-2020 plus
equipped with a Wax pillar column (30 m of length, 0.25 mm of internal diameter, 0.25 um of
film thickness). The temperature of the column was initially maintained at 50C for 3 min and
then raised to 250C with a ramping rate of 10C/min and hold at 250C for 3 min. The
products were identified with a standard library (NIST MS Search 2014). The yields of LA
5

Catalysis Science & Technology
Page 6 of 29
View Article Online
DOI: 10.1039/D0CY00225A
and HMF were calculated via the following equations:
Mole of HMF produced
YHMF-1 (%) =
 100%
(1)
Mole of glucose/fructose loaded
Mass of HMF produced
YHMF-2 (%) =
YLA-1 (%) =
 100%
 100%
(2)
(3)
(4)
Mass of inulin/cellulose loaded
Mole of LA produced
Mole of glucose/fructose loaded
Mass of LA produced
YLA-2 (%) =
 100%
Mass of inulin/cellulose loaded
YHMF-1: the yield of HMF from glucose/fructose;
YHMF-2: theoretical yields of HMF from inulin/cellulose;
YLA-1: the yield of LA from glucose/fructose;
YLA-2: theoretical yields of LA from inulin/cellulose;
The concentration of glucose or fructose after the catalytic tests was detected by a
high-performance liquid chromatography (HPLC, Shimadzu) equipped with a size-exchange
column (Shodex, SH1011 of type, 30 m of length, 8 mm of internal diameter). The procedures
were detailed as follows. The liquid sample, without further dilution, was fed into the SH1011
column directly for the analysis with a mobile phase (0.01 mol/L of H2SO4). The temperature
of column was set at 50°C and a flow speed of 0.6 mL/min for the mobile phase was
employed. Conversion of glucose/fructose was calculated via the following equation:
Mole of sugar consumed
(5)
The conversion of sugars （%）=
 100%
Mole of glucose/fructose loaded
A UV-fluorescence spectrometer (Shimadzu, RF-6000) was used to characterize the soluble
polymer with the π-conjugated structure. The testing conditions were set as below: the range
6

Page 7 of 29
Catalysis Science & Technology
of excitation wavelength: 200–400 nm; the range of emission wavelength: 250–50V0iewnArmticl;e Online
DOI: 10.1039/D0CY00225A
scanning speed: 600 nm/min. Prior to the measurement, the residual liquid after the reaction
was diluted to four hundred ppm with anhydrous ethanol as the solvent. The different colours
represent the different fluorescence intensity.
The hydroxyl and other groups in the mixed of solvent and sugars were characterized by
using a Nicolet iS50 FT-IR spectrometer at room temperature. 1 mg of the sugar and 199 mg
of potassium bromide powder were mixed and grounded evenly, and then added a drop of
solvent under an infrared light. The mixture was pressed into a thin disc at the pressure of 15
MPa. The thin disc was tested by using an infrared spectrometer.
UV-Vis spectra of various solvents dissolved CoSO47H2O was measured by a Metash
UV-visible photometer (UV-8000S). The scanning wavelength was from 800 to 300 nm. The
right solution was measured as soon as it was prepared.
3.Results and discussion
3.1 Conversion of glucose/fructose with various metal sulfates in THF
In the initial trials, a series of metal sulfates were selected as the catalysts to estimate
their capabilities for the conversion of glucose/fructose into HMF in THF, an aprotic solvent
resembling DMSO but with a much lower boiling point (189ºC versus 66ºC). As shown in
Table 1, the full-conversions of glucose/fructose were obtained for all the runs, while the
different metal sulfates displayed distinct catalytic behaviors. For the blank experiments in the
absence of salts, the yield for HMF from fructose (13.8%) was higher than that from glucose
(4.9%) (Entry 1 and 2, Table 1). This was expected as the dehydration of the furanose form of
fructose directly produced HMF, while glucose has to isomerize into fructose form firstly
before being further transformed into HMF [27].
The typical alkali metal sulfates (Na2SO4, K2SO4) could neither enhance the
isomerization of glucose to fructose nor improve the formation of HMF from either glucose or
fructose. Conversely, Na2SO4 or K2SO4 coordinated with the sugars, interfering with their
7

Catalysis Science & Technology
Page 8 of 29
View Article Online
conversion to HMF and leading to their polymerization to form the insoluble polymers (the
DOI: 10.1039/D0CY00225A
yield of coke: glucose-Na2SO4: 41.5%, frucose-Na2SO4: 77.8%, glucose-K2SO4: 50.9%,
fructose-K2SO4: 63.3%).
Fe2(SO4)3 can hydrolyze partially in water and create an acidic reaction medium, which,
however, was not the case in THF. Instead of hydrolysis, it was believed Fe2(SO4)3
coordinated with the sugars, promoting the retrograde aldol condensation of the sugar to form
acetic acid and the polymers (the yield of coke: glucose-Fe2(SO4)3: 26.5%, glucose-Fe2(SO4)3:
50.8%,). CuSO4, in comparison to Fe2(SO4)3, could not only improve the dehydration of
fructose into HMF in THF (Entry 9 and 10, Table 1), but also the further decomposition of
HMF to LA. CuSO4 clearly could catalyse the hydrolysis reactions, as also evidenced by the
conversion of glucose to LA (Entry 9, Table 1). Encouraged by this interesting result, other
divalent transition-metal salts were also used in the conversion of glucose/fructose.
NiSO46H2O presented significant catalytic activity for the conversion of glucose/fructose into
both HMF and LA, especially for the formation of HMF from fructose (Entry 11 and 12,
Table 1). However, NiSO46H2O could also catalyse the retrograde aldol condensation of the
sugars to form acetic acid.
CoSO47H2O also catalysed the formation of acetic acid and selectively promoted the
dehydration of fructose to HMF with the yield reaching 79.3% (Entry 13 and 14, Table 1).
The cobalt ions were probably active for the dehydration of fructose to HMF. However, if
CoCl26H2O, Co(NO3)26H2O and Co(C2H3O2)26H2O were used as the catalysts, the yield of
HMF decreased significantly or the dehydration of fructose could not take place at all (Entry
15-20, Table 1). The results indicated the synergistic effects between cobalt ion and the
sulfate in CoSO4 must exist. Either the cation or the anion or both could coordinate with
fructose to facilitate the dehydration of fructose. Nevertheless, whether the synergistic effects
between CoSO4 and THF for the conversion of fructose existed or not has not been cleared
8

Page 9 of 29
Catalysis Science & Technology
View Article Online
from the above results. Our previous studies showed reaction medium played significant roles
DOI: 10.1039/D0CY00225A
in affecting the catalytic behaviors of the solid acid resin catalysts [28, 29]. For the salt
catalysts, they have different solubilities in different solvents and coordination status. How
the solvent with the distinct properties affect the catalytic activity of the CoSO4 was further
investigated.
3.2 Conversion of glucose/fructose in various solvents
Some typical aprotic solvents (acetone, DMSO THF, furan and furfural) and protic
solvents (methanol, ethanol, isopropanol, water and acetic acid) were employed in the
conversion of glucose/fructose. The results displayed in Table 2 indicated that solvents
significantly impacted the activity of CoSO47H2O catalyst. DMSO is a well-known solvent
for achieving high yields of HMF from fructose dehydration, as DMSO could coordinate with
HMF and prevent the further decomposition of HMF to LA and formic acid [25].
Nevertheless, in DMSO, the yields of HMF were negligible with CoSO47H2O as the catalyst.
This result clearly suggested, from an indirect angle, that the synergistic effects between
CoSO4 and THF existed and was crucial for achieving of the high yields of HMF.
The UV-fluorescence results for characterizing the soluble polymers showed (Figure 1c)
the significant polymerization of fructose in DMSO, while the polymerization reactions in
THF were negligible (Figure 1c and d).
Although the yield of HMF reached 40.9% in acetone, acetone was not an ideal solvent
for dehydration of fructose. Acetone contains α-hydrogen and carbonyl functionalities, which
could trigger aldol condensation and polymerization with the fragment of sugars [30, 31],
leading to the decrease of HMF production. Alcohols could protect the sugars or furans from
polymerization [32-34]. Nevertheless, alcohols could react with HMF via acetalization and/or
etherification reactions [35]. As shown in Table 2 (Entry 7-12), CoSO4 could effectively
9

Catalysis Science & Technology
Page 10 of 29
View Article Online
catalyse the conversion of fructose, but the HMF formed further reacted with the alcohols or
DOI: 10.1039/D0CY00225A
further decomposed to form the levulinates. Acetic acid and water are very polar, which
affected the coordination between CoSO4 and fructose, as CoSO4 could dissolve in these polar
solvents via the solvation effect, leading to the negligible formation of HMF.
The substances with π-conjugated structures could be detected by UV-fluorescence
spectrophotometer [36,37]. The UV-fluorescence spectra (Figure 1a) showed fructose in
water mainly polymerized to form the soluble polymers with π-conjugated structures. The
aprotic solvent like furan or furfural could facilitate the dehydration of fructose to HMF.
Nevertheless, these furans are very reactive to polymerization reactions [38] and could
cross-polymerize with the sugars or HMF [39], diminishing the production of HMF. The
liquid products from the conversion of the sugars in methanol, ethanol, THF and DMSO were
characterized with FT-IR. The results (Figure 2a and 2b) were generally in line with those in
Table 2. The carbonyl group at 1670 cm-1 belonged to HMF while that at 1713 cm-1 was
assigned to LA. In THF, the carbonyl group belonging to HMF was the dominate one.
Interestingly, in DMSO the absorption peak ca. 1100 cm-1 was weakened.
DMSO possibly coordinate with the sugars, interfering with the coordination of the
sugars with CoSO4 and then negatively affecting the dehydration of the sugars. The above
results showed that THF as the reaction medium could facilitate the selective conversion of
fructose to HMF, while water could not. The dehydration of fructose produces water, while
the CoSO47H2O catalysts used contains water. The water from fructose or the catalysts thus
might negatively impact the production of HMF, which was further investigated.
3.3 Roles of water on the conversion of glucose/fructose
The amount of fructose loaded determined the amount of water produced from the
dehydration of fructose. Hence, the effects of fructose loading on the formation of HMF were
10

Page 11 of 29
Catalysis Science & Technology
View Article Online
investigated and the results were displayed in Table S1. If the amount of fructose loaded was
DOI: 10.1039/D0CY00225A
lower than 0.1 g, LA, derived from the ring-opening and decomposition of HMF, could not be
formed. HMF was the main product and its yield was higher with the loading of 0.04 g than
with 0.02 g (88% versus 56.9%). The further increase of loading of fructose led to a
remarkable decrease of HMF production, while enhanced the production of LA. THF is not a
good solvent to dissolve CoSO4 while water is (3.6 g THF only dissolved 0.54 mg CoSO4).
The presence of a small amount of water seemed to aid the catalytic activity, while too much
water produced from the dehydration of fructose facilitated the formation of LA. The
solvation or hydration of CoSO4 with water would change its coordination status and impact
its catalytic activity.
To further understand the effects of water on catalytic activity and the decomposition of
HMF to LA, the conversion of glucose/fructose in the mixture of THF/water with varied ratio
of water was further investigated. The results in Table S2 showed, after removing the crystal
water in the cobalt sulfate, the CoSO4 catalyst showed little activity for the conversion of
glucose to HMF, while the yields of HMF from fructose were also much lower over CoSO4
than that over CoSO47H2O (3.1% versus 25.7%). The crystal water in CoSO47H2O was
clearly crucial for maintaining activity of the catalysts. Nevertheless, with external water
added, glucose and fructose behaved differently towards the formation of HMF. For glucose,
the addition of more water promoted its conversion to HMF, and only with the proportion of
water reaching 70%, the HMF started to decompose to LA. With fructose as a feedstock, the
addition of water led to a substantial decrease of the yields of HMF, while no LA were
detected. For glucose, it needs to isomerize to fructose for further conversion into HMF [40].
With the addition of external water, CoSO4 would be dissolved and the generated cation
and/or anions might promote the isomerization of glucose. For fructose, the only step required
for its conversion to HMF is the dehydration reactions, and the externally added water,
11

Catalysis Science & Technology
Page 12 of 29
View Article Online
obviously, did not favor the dehydration of fructose. Further to this, under the
DOI: 10.1039/D0CY00225A
sub-supercritical conditions, water acted as both acid and base [41], which probably promoted
the polymerization of fructose/HMF. The above results indicated that the crystal water in
cobalt sulfate was essential for the conversion of fructose into HMF. The synergistic effects
among cobalt ion, the sulfate functionality and the crystal water were crucial for maintaining
the high activity for the dehydration of fructose to HMF. In addition, the loading of CoSO4
also significantly impacted the conversion of glucose/fructose to HMF, as shown in Table 3.
With either glucose or fructose as the feedstock, the increase of the loading of the cobalt
sulfate initially enhanced the yields of HMF while diminished the formation of HMF if the
loading of cobalt sulfate was too high. The higher loading of the cobalt sulfate would
introduce too much crystal water, which might negatively impact the formation of HMF. The
results herein and in the Section 3.2 clearly indicated that THF was a favourable reaction
medium for the production of HMF from fructose, the detailed reason for this was further
explored.
3.4 The role of THF in the high-yield catalytic system
HMF is an intermediate product from the conversion of fructose into LA [42], and how to
suppress the conversion of HMF to LA is crucial to achieve a high selectivity of HMF. In
DMSO, a higher yield of HMF could be achieved as DMSO could effectively prevent its
further decomposition to LA [43]. THF might also have such effect, the thermal treatments of
HMF in THF and other solvents in both the absence and presence of cobalt sulfates, with the
purpose of comparison, were conducted. The results were shown in Table 4 and Figure 3.
More interestingly, with the presence of CoSO4, the recovery rate of HMF is even higher
(95.8%) (Entry 6, Table 4), indicating that CoSO4 could not catalyse the further
decomposition of HMF and even could coordinate with HMF, enhancing its stability in THF.
12

Page 13 of 29
Catalysis Science & Technology
This is very different from the Brønsted acid catalysts, which could catalyse boVtihew Atrhticele Online
DOI: 10.1039/D0CY00225A
dehydration of fructose to HMF and the further decomposition of HMF to LA [44]. In water,
in the presence of CoSO4, the recovery rate of HMF was also higher than that in absence of
CoSO4. The significant degradation of HMF in water is mainly due to the acid/base roles of
water played under the sub-supercritical conditions [45]. Acetic acid is a carboxylic acid,
which catalyzed the degradation of HMF via probably the polymerization reactions, as
evidenced by the deep color of the resultant solution with acetic acid or water as the reaction
medium (Figure 3).
The recovery rates of HMF in ethanol, isopropanol and DMSO were also high but only in
the absence of CoSO4 (Entry 1-3, Table 4). In ethanol and isopropanol, the presence of CoSO4
resulted in the conversion of HMF via etherification to form 5-ethoxy-methyl-furfural or
5-isopropoxy-methyl-furfural, respectively (Figure S1). The colour of the resulting solution
became deeper, indicating the occurrence of the polymerization reactions (Figure 3). In the
presence of Brønsted acid, DMSO could protect HMF and prevent its further conversion,
while, in the presence of CoSO4, the degradation of HMF proceeded remarkably (Figure 3).
THF could coordinate with HMF, while the presence of CoSO4 further strengthen the
coordination effects, enhancing the stability of HMF in the catalytic system. This is a
pre-requisite for achieving high yields of HMF. The coordination between cobalt sulfate and
the reaction medium could be further confirmed with the UV-Vis absorption characterization.
3.5 The coordination between cobalt sulfate and THF
UV-Vis absorption spectra for the mixture of CoSO47H2O with ethanol, DMSO, H2O or
THF and the test conditions were shown in Figure S2. H2O-CoSO47H2O had a characteristic
peak at 502 nm, which originated from the d-d transition from cobalt itself.
DMSO-CoSO47H2O showed the absorption at 555 nm in the visible light region, while the
13

Catalysis Science & Technology
Page 14 of 29
View Article Online
absorption of the THF-CoSO47H2O showed a distinct absorption at ca. 510 nm. For
DOI: 10.1039/D0CY00225A
ethanol-CoSO47H2O, no clear absorption peaks were visible. The different peaking
wavelength indicated the distinct patterns for the coordination between cobalt sulfate and the
solvent. The different coordination also led to the changed colours of the mixture of
CoSO4/solvent, as shown in Figure S3. CoSO4 fully dissolved in water or DMSO, partially
dissolved in ethanol while slightly dissolved in THF [46]. Nevertheless, the small amount of
CoSO4 in THF substantially changed the coordination status with THF, which was further
investigated with FT-IR study.
The FT-IR spectra of CoSO4 in THF showed three types of hydroxyl group in the region
of 3600-3200 cm-1 (Figure 4b). The first peak at 3222 cm-1 belonged to O–H of the hydrogen
bonds in water, as compared with the spectra of the pure water. The second peak at 3350 cm-1
belonged to the hydroxyl group coordinated with cobalt ions but having blue shift from 3222
to 3350 cm-1, as also compared with the absorption spectra for CoSO4-H2O. The third peak at
3509 cm-1 was attributed to the hydroxyl group which was affected by the hydrogen bond
between water and THF. If CoSO4 could coordinate with the hydroxyl group in water, it could
also coordinate with the hydroxyl group in fructose and in HMF, impacting the dehydration of
fructose and the further degradation of HMF.
In addition to this, the coordination between the solvents and sulfate ions were clearly
displayed in Figure 4a. The S=O group of CoSO4 had a stretching vibration at 1094 cm-1. The
addition of water forms the hydrogen band between water and shift the absorption of the S=O
to a lower wavenumber (from 1094 to 1082 cm-1). The same scenario happened for
CoSO4-ethanol. The dissolution of cobalt sulfates, as discussed earlier, was improved owing
to the coordination between sulfate ions and ethanol or water.
Besides, DMSO itself had an S=O group, which had two peaks at 1019 and 996 cm-1
(Figure 4a) and shifted to 1042 and 1020 cm-1 when CoSO4 coordinated with DMSO.
14

Page 15 of 29
Catalysis Science & Technology
Meanwhile, the S=O group belonging to CoSO4 shifted from 1124 to 1094 cm-1. ThisVirewesAruticllte Online
DOI: 10.1039/D0CY00225A
indicated the coordination between DMSO and cobalt ions was stronger than hydrogen band
so that the special peak of S=O group shift to a higher wavenumber. THF could not interact
with cobalt or sulfate ions directly and only form the hydrogen band with crystal water of
CoSO4 so that there was a slight blue shift for the special peak of S=O from 1094 to 1096
cm-1. That is to say, the coordination between CoSO4 and THF depended on the hydrogen
bond formed from crystal water and CoSO4 or THF. The organic phase with higher volume
ratio to aqueous phase was responsible for the excellent catalytic capability [47]. Here also,
the catalytic system without water did not have the coordination between CoSO4 and THF and
thus did not aid catalytic performance (Entry 1 and 8, Table S2). Nevertheless, too much
water could destroy the balance between CoSO4, crystal water and THF and negatively
affected the conversion of fructose to HMF.
The above discussions showed that the interaction between the cobalt sulfate and the
crystal water could improve the dehydration of fructose. The synergies among cobalt sulfate,
crystal water and THF in CoSO4/THF catalytic system promoted the formation of HMF from
fructose.
3.6 The conversion of inulin and cellulose in CoSO4/THF catalytic system
To further investigate the catalytic behaviors of CoSO4/THF catalytic system, the
conversions of inulin and cellulose were conducted and the results were compared with that in
sulfuric acid/THF system. The yields of HMF from the conversion of fructose, inulin and
cellulose with 0.1 M of H2SO4 as the catalyst in THF were zero (Entry 1-3, Table 5).
The resulting products showed a black color, indicating the dominance of the
polymerization reactions. The concentration of sulfuric acid applied was too high, leading to
the quick polymerization of the reactants. With the decreasing concentration of sulfuric acid,
15

Catalysis Science & Technology
Page 16 of 29
View Article Online
the yields of HMF was increased significantly, but was much lower than that in CoSO4/THF
DOI: 10.1039/D0CY00225A
catalytic system (Entry 2, Table S3). Furthermore, the coordination between HMF and THF
prevented the further decomposition of HMF to LA with sulfuric acid as the catalysts. In
CoSO4/THF catalytic system, both the reaction temperature and the amount of CoSO4 loaded
affected the yields of HMF produced from inulin. Under the experimental conditions
investigated, the maximum yield of HMF reached 40.1% from the hydrolysis of inulin, which
was much higher than that in sulfuric acid/THF at the same temperature (Entry 9-14, Table
S4).
As for cellulose, under the conditions employed the maximum yield of HMF was 35.3%
(Entry 16, Table S4), which was also much higher than that in sulfuric acid-THF system and
was even comparable to that from glucose. Inulin is mainly composed by fructose unit while
cellulose is composed by glucose unit. Nevertheless, although the yields of HMF from
fructose and glucose were significantly different (Table 3), the yields of HMF from the two
macro sugar polymers were not significantly different. It seemed the depolymerization of
inulin and cellulose determine the yields of HMF. How to enhance the catalytic efficiency of
CoSO4/THF catalytic system deserves further attention.
4. Conclusions
In summary, this study investigated catalytic conversion of fructose and glucose for the
production of HMF using cobalt sulfate as the catalyst in various reaction medium. The
results showed that in THF the sulfates like Na2SO4, K2SO4 and Fe2(SO4)3 strongly
coordinated with the sugars, interfering with their dehydration to HMF and promoting their
polymerization. In comparison, the coordination of CoSO47H2O with especially fructose in
THF mainly catalyzed the dehydration reactions for the selective production of HMF with a
maximum yield of 88%. There existed synergistic effects between cobalt ion and the sulfate in
16

Page 17 of 29
Catalysis Science & Technology
View Article Online
CoSO47H2O, as other cobalt salts such as CoCl2, Co(NO3)2 and Co(C2H3O2)2 showed
DOI: 10.1039/D0CY00225A
insignificant activity for catalyzing the dehydration of fructose. Nevertheless, CoSO47H2O
was only effective for the dehydration of fructose in THF, but not in other solvents such as
DMSO, water, acetone, acetic acid, furan, furfural, methanol, ethanol and iso-propanol. This
is because THF could help to retain HMF in the reaction medium while CoSO4 did not
catalyze the further decomposition of HMF and even could coordinate with HMF, enhancing
its stability in THF. Furthermore, the characterizations showed that the crystal water in cobalt
sulfate could help to coordinated with fructose, which facilitated the conversion of fructose
the dehydration reactions. The synergistic effects among cobalt ion, the sulfate functionality,
the crystal water and THF were crucial for achieving the high selectivity for the dehydration
of fructose to HMF. The CoSO47H2O/THF catalytic system also worked for the conversion
of inulin and cellulose into HMF, but the yields of HMF from these macro sugar polymers
have much room to be further improved. The biggest advantage for the CoSO47H2O/THF
catalytic system is the low cost, the easy recycle of CoSO47H2O catalyst and the easy
separation of HMF from the volatile THF. The potential of such catalytic system for the
conversion of other catalytic systems could also be further explored.
Conflicts of interest
There are no conflicts of interest to declare.
Acknowledgements
This work was supported by National Natural Science Foundation of China (No.
51876080), the Strategic International Scientific and Technological Innovation Cooperation
Special Funds of National Key R&D Program of China (No. 2016YFE0204000), the Program
for Taishan Scholars of Shandong Province Government and the Recruitment Program of
17

Catalysis Science & Technology
Page 18 of 29
View Article Online
DOI: 10.1039/D0CY00225A
Global Young Experts (Thousand Youth Talents Plan). Collaborative Innovation Center of
Technology and Equipment for Biological Diagnosis and Therapy in Universities of
Shandong.
References
1
2
3
4
J. Artz, S. Mallmann, R. Palkovits, ChemSusChem, 2015, 8, 672-679.
Z. Zhang, G. W. Huber, Chem Soc Rev, 2018, 47, 1351-1390.
J. Wang, J. Xi, Q. Xia, X. Liu, Y. Wang, Sci. China Chem., 2017, 60, 870-886.
A.J. Jang, J-H. Lee, M. Yotsu-Yamashita, J. Park, S. Kye, R. L. Benza, M. J. Passineau,
Y.J. Jeon, T. Nyunoya, Scientific Reports, 2018, 8, 11504-11514.
P. Zhou, Z. Zhang, Catal. Sci. Technol., 2016, 6, 3694-3712.
F. H. Richter, K. Pupovac, R. Palkovits, F. Schüth, ACS Catal., 2013, 3, 123-127.
H. Ning, J. Song, M. Hou, D. Yang, H. Fan, B. Han, Sci China Chem, 2013, 56,
1578-1585.
5
6
7
8
9
S. Jing, X. Cao, L. Zhong, X. Peng, R. Sun, J. Liu, Ind. Crop Prod., 2018, 126, 151-157.
Y-B. Huang, T. Yang, Y-T. Lin, Y.Z. Zhu, L. C. Li, Pan H. Green Chem., 2018, 20,
1323-1334.
10 F. Menegazzo, E. Ghedini, M. Signoretto, Molecules, 2018, 23, 2201-2218.
11 S. Rivas, C. Vila, J. L. Alonso, V. Santos, J. C. Parajó, J. J. Leahy, Ind. Crop Prod., 2019,
134, 309-317.
12 X. Li, R. Xu, J. Yang, S. Nie, D. Liu, Y. Liu, C. Si, Ind. Crop Prod., 2019, 130, 184-197.
13 Y. Zhang, P. Jin, M. Meng, L. Gao, M. Liu, Y. Yan, Nano, 2018, 13, 1850132.
14 I. K. M. Yu, D. C. W. Tsang, A. C. K. Yip, A. J. Hunt, J. Sherwood, J. Shang, H. Song,
Y.S. Ok, C.S. Poon, Green Chem., 2018, 20, 2064-2074.
18

Page 19 of 29
Catalysis Science & Technology
View Article Online
DOI: 10.1039/D0CY00225A
15 Y. Y. Lee, K. C. W. Wu, Phys. Chem. Chem. Phys., 2012, 14, 13914-13917.
16 B. Karimi, H. M. Mirzaei, H. Behzadnia, H. Vali, ACS Appl. Mater. Interfaces, 2015, 7,
19050-19059.
17 S. P. Teong, G. Yi, X. Cao, Y. Zhang, ChemSusChem, 2014, 7, 2120-2124.
18 D. Ray, N. Mittal, W. J. Chung, Carbohydr. Res., 2011, 346, 2145-2148.
19 K. R. Enslow, A. T. Bell, Catal. Sci.Technol., 2015, 5, 2839-2847.
20 Y. Wang, C. M. Pedersen, Y. Qiao, T. Deng, J. Shi, X. Hou, Carbohyd. Polym., 2015,
115, 439-443.
21 Y. Wang, C. M. Pedersen, T. Deng, Y. Qiao, X. Hou, Bioresource Technol., 2013, 143,
384-390.
22 C. Antonetti, A. M. R. Galletti, S. Fulignati, D. Licursi, Catal. Commun, 2017, 97,
146-150.
23 F. Salak Asghari, H. Yoshida, Ind. Eng. Chem. Res., 2006, 45, 2163-2173.
24 H. Zhao, J. E. Holladay, H. Brown, Z. C. Zhang, Science, 2007, 316, 1597–1599.
25 M. Shaikh, S. K. Singh, S. Khilari, M. Sahu, K. V. S. Rangnath, Catal. Commun., 2018,
106, 64-67.
26 X. Tong, Y. Li, ChemSusChem, 2010, 3, 350-355.
27 H. Xia, H. Hu, S. Xu, K. Xiao, S. Zuo, Biomass Bioenerg., 2018, 108, 426-432.
28 X. Hu, R. J. M. Westerhof, D. Dong, L. Wu, C-Z. Li, ACS Sustain. Chem. Eng., 2014, 2,
2562-2575.
29 X. Hu, S. Wang, R. J. M. Westerhof, L. Wu, Y. Song, D. Dong, C-Z. Li, Fuel, 2015, 141,
56-63.
30 D. Luo, W. Yin, S. Liu, N. Yang, S. Xia, P. Ma, Fuel, 2018, 230, 368-375.
31 X. Hu, D. Mourant, R. Gunawan, L. Wu, Y. Wang, C. Lievens, C. -Z. Li, RSC Advances
2012, 2, 9366-9370.
19

Catalysis Science & Technology
Page 20 of 29
View Article Online
DOI: 10.1039/D0CY00225A
32 X. Hu, S. Jiang, L. Wu, S. Wang, C.-Z. Li, Chem. Commun. 2017, 53, 2938-2941
33 X. Hu, C-Z. Li, Green Chem., 2011, 13, 1676-1679.
34 R. Gunawan, X. Li, A. Larcher, X. Hu, D. Mourant, W. Chaiwat, H Wu, C-Z. Li, Fuel,
2012, 95, 146-151.
35 M. J. Climent, A. Corma, S. Iborra, Green Chem., 2014, 16, 516-547.
36 K. Sun, Q. Xu, Y. Shao, L. Zhang, Q. Liu, S. Zhang, Y. Wang, Hu, X. Energy Fuel.,
2019, 33, 7480-7490.
37 Z. Zhang, X. Zhang, L. Zhang, Y. Wang, X. Li, S. Zhang, Q. Liu T. Wei, G. Gao, X. Hu,
Energ. Convers. Manage. 2020, 205, 112301.
38 Y. Jiang, D. Ding, S. Zhao, H. Zhu, H. I. Kenttämaa, M. M. Abu-Omar, Green Chem.,
2018, 20, 1131-1138.
39 X. Hu, S. Kadarwati, S. Wang, Y. Song, M. D. M Hasan., C-Z. Li, Fuel Process.
Technol., 2015, 137, 212-219.
40 R. Huang, W. Qi, R. Su, Z. He, Chem. Commun., 2010, 46, 1115-1117.
41 H. Ramsurn, R. B. Gupta, Energ. Fuel., 2012, 26, 2365-2375.
42 Y. Han, L. Ye, X. Gu, P. Zhu, X. Lu, Ind. Crop. Prod., 2019, 127, 88-93.
43 N. L. Mulik, P. S. Niphadkar, K. V. Pandhare, V. V. Bokade, ChemistrySelect, 2018, 3,
832-836.
44 S. S. Chen, L. Wang, I. K. M. Yu, D. C. W. Tsang, A. J. Hunt, F. Jérôme, S. Zhang, Y. S.
Ok, C.S. Poon, Bioresour. Technol., 2018, 247, 387-394.
45 J. Yu, C. Jiang, Q. Guan, J. Gu, P. Ning, R. Miao, Q. Chen, J. Zhang, Fuel, 2018, 217,
275-284.
46 N. A. Bogachev, A. O. Gorbunov, A. A. Tikhomirova, O. S. Pushikhina, M. Y. Skripkin,
A. B. Nikolskii, Russ. J. Gen. Chem., 2015, 85, 2509-2512.
47 N. Shi, Q. Liu, Q. Zhang, T. Wang, L. Ma, Green Chem., 2013, 15, 1967-1974.
20

Page 21 of 29
Catalysis Science & Technology
View Article Online
DOI: 10.1039/D0CY00225A
Table 1 The conversion of glucose or fructose in THFa
Yield (%)
Con.
Entry
Catalyst
Sugar
Acetic
acid
0
(%)b
HMF
LA
Furfural
0.3
0.9
0
1
2
No catalyst
No catalyst
Na2SO4
glucose
fructose
glucose
fructose
glucose
fructose
glucose
fructose
glucose
fructose
glucose
fructose
glucose
fructose
glucose
fructose
glucose
fructose
glucose
fructose
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
100
0
0
4.9
13.8
0.1
0.1
0.1
0.1
0
0
3
0
0
4
Na2SO4
1.8
0
0
0
5
K2SO4
0
0
6
K2SO4
1.3
0
0
0
7
Fe2(SO4)3
4.7
7.2
0
0
8
Fe2(SO4)3
0.4
0.3
2.1
1.0
0.9
1.7
1.9
0.1
0.2
0.5
0.2
0.1
0.3
0.1
8.4
8.6
6.4
2.4
0
9
CuSO4
0
10
11
12
13
14
15
16
17
18
19
20
CuSO4
0
14.1
12.2
32.7
5.2
11.8
6.8
7.1
2.3
3.1
0.6
2.1
1.1
1.3
NiSO46H2O
NiSO46H2O
CoSO47H2O
CoSO47H2O
CoCl26H2O
CoCl26H2O
Co(NO3)26H2O
Co(NO3)26H2O
Co(C2H3O2)26H2O
Co(C2H3O2)26H2O
14.5 14.7
3.9
0
79.3
16.1
20.7
0
0
0
0
0
0
0
0
0
aReaction condition: glucose or fructose: 0.4 g, THF: 3.6 g, sulfates: metal ions had the same
mole number with glucose/fructose, T = 170C, reaction time: 2 h, N2: 3 MPa.
bCon.: conversion.
21

Catalysis Science & Technology
Page 22 of 29
View Article Online
Table 2 The conversion of glucose or fructose with CoSO47H2O in different solDvOeIn: 1t0s.1a039/D0CY00225A
Yield (%)
Entry
Solvent
Sugar
Furfural LM/LE/LIPb
LA
14.5
3.9
0
Ethers/estersc HMF
1
2
THF
THF
glucose
fructose
glucose
fructose
glucose
fructose
1.7
1.9
0
-
-
14.7
79.3
3.6
4.0
-
-
-
3
DMSO
DMSO
Acetone
Acetone
-
-
4
0
-
0
-
-
5
0
-
0
6
0.4
0.7
1.2
0
-
0.9
0
-
40.9
0
7
Methanol glucose
Methanol fructose
2.0
3.4
56.3
13.5
45.8
13.7
21.8
5.6
12.1
-
8
18.7
2.3
0
0
9
Ethanol
Ethanol
glucose
fructose
3.0
0
10
11
12
13
14
15
16
17
18
19
20
1.4
0.6
0.3
0
0.1
1.8
0
0
Isopropanol glucose
Isopropanol fructose
Acetic acid glucose
Acetic acid fructose
6.4
12.8
24.5
1.1
2.5
0
10.8
0
-
-
-
-
-
-
-
-
0.7
0.8
0.4
1.1
0
0
Waterd
Water
glucose
fructose
glucose
fructose
glucose
fructose
0.02
0.04
0.1
0.1
0
-
0.2
7.0
37.2
20.4
21.9
Furane
Furan
-
0
-
Furfural
Furfural
0
0
0
0
0
aReaction conditions: glucose or fructose: 0.4 g, solvent: 3.6 g, CoSO4: cobalt ions had the
same mole number with the sugar, T = 170C, reaction time: 2 h, N2: 3 MPa. All the
conversions were 100%.
bLM: methyl levulinate, LE: ethyl levulinate, LIP: isopropyl levulinate.
cThe product in methanol: 5-methoxy-methyl-furfural (MMF); the product in ethanol:
5-ethoxy-methyl-furfural (EMF); the product in isopropanol: 5-isopropoxy-methyl-furfural
(IPMF); the product in acetic acid: FFA: methyl-furfuryl-acetate.
dT = 200C, the other conditions were the same with a.
eReaction conditions: sugar: 0.04 g, solvents: 3.60 g, T = 170C, reaction time: 2 h, N2: 3
MPa. Catalyst: the mole number of CoSO47H2O was 1.5 times that of monosaccharide. All
the conversions were 100%.
22

Page 23 of 29
Catalysis Science & Technology
View Article Online
DOI: 10.1039/D0CY00225A
Table 3 Yields of HMF from sugars in THF with different amounts of CoSO47H2Oa
Yield (%)
CoSO47H2Ob
Entry
Sugar
LA
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
HMF
0
1
2
Glucose
Glucose
Glucose
Glucose
Glucose
Glucose
Glucose
Glucose
Glucose
Fructose
Fructose
Fructose
Fructose
Fructose
Fructose
Fructose
Fructose
Fructose
0
0.1n
0.5n
n
10.2
13.0
23.4
33.4
35.7
38.5
28.9
19.7
3.1
3
4
5
1.5n
2n
6
7
5n
8
10n
12.5n
0
9
10
11
12
13
14
15
16
17
18
0.1n
0.5n
n
44.7
41.7
52.5
70.3
64.1
63.0
88.0
39.2
1.5n
2n
5n
10n
12.5n
aReaction conditions: sugar: 0.04 g, THF: 3.60 g, T = 170C, reaction time: 2 h, N2: 3 MPa.
All the conversions were 100%.
bn was the mole number of the sugar.
23

Catalysis Science & Technology
Page 24 of 29
View Article Online
DOI: 10.1039/D0CY00225A
Table 4 The thermal-treatment of HMF in different solventsa
Recovery rate with
Recovery rate without
Entry
Solvent
CoSO47H2O
CoSO47H2O (%)
(%)
0
1
2
3
4
5
6
Ethanol
Isopropanol
DMSO
93.2
95.6
89.7
2.0
47.8
73.3
4.8
Acetic acid
Water
58.7
92.2
66.7
95.8
THF
aReaction conditions: HMF: 0.05 g, CoSO47H2O: 0.11 g, solvents: 3.60 g, T = 170C,
reaction time: 2 h, N2:3 MPa.
24

Page 25 of 29
Catalysis Science & Technology
View Article Online
DOI: 10.1039/D0CY00225A
a
Table 5 The conversion of raw materials with classic H2SO4
Concentration of H2SO4 Yield of LA
Yield of
Entry
Reactant
(mol/L)
0.1
(%)
0
HMF (%)
1
2
3
4
5
6
Fructose
Inulin
0
0
0.1
0
Cellulose
Fructose
Inulin
0.1
0
0
0.01
0.01
0.01
0
34.1
8.2
6.7
0
Cellulose
0
aReaction conditions: fructose/inulin/cellulose: 0.05 g, THF: 5 g, T = 200C, reaction time = 6
h, N2 = 3 MPa. All the conversions were 100%.
25

Catalysis Science & Technology
Page 26 of 29
View Article Online
DOI: 10.1039/D0CY00225A
Figure 1 UV-fluorescence spectra for the soluble polymers from the liquid products of
fructose-conversion in various solvents (a), 3D fluorescence spectra of soluble polymers in
acetone (b), DMSO (c) and in THF (d).
26

Page 27 of 29
Catalysis Science & Technology
View Article Online
DOI: 10.1039/D0CY00225A
Figure 2 FT-IR spectra of liquid products from the conversion of fructose or glucose with
different solvents.
27

Catalysis Science & Technology
Page 28 of 29
View Article Online
DOI: 10.1039/D0CY00225A
Figure 3 The change in color of solution before and after the reaction. Reaction conditions:
HMF: 0.05 g, CoSO47H2O: 0.11g, solvents: 3.60 g, T = 170C, reaction time: 2 h, N2: 3
MPa.
28

Page 29 of 29
Catalysis Science & Technology
View Article Online
DOI: 10.1039/D0CY00225A
Figure 4 FT-IR spectra of CoSO47H2O in THF, ethanol, DMSO and H2O, respectively. (a)
Full range spectra of solvents and CoSO4-solvent; (b) Spectra of CoSO4-solvent in the range
from 4000 to 2500 cm-1.
29