Catalytic conversion of raw Dioscorea composita biomass to 5-hydroxymethylfurfural using a combination of metal chlorides in N,N-dimethylacetamide solvent containing lithium chloride

Article abstract of DOI:10.1007/s11164-016-2496-9

We synthesized 5-hydroxymethylfurfural (HMF) from carbohydrates using metal chloride catalysts. A 33.2?% yield of HMF was obtained from raw Dioscorea composita biomass with high starch by using a catalyst system composed of CrCl₃·6H₂O and LaCl₃·6H₂O at 120?°C for 4?h in N,N-dimethylacetamide containing lithium chloride. The catalyst system is also cost-effective for the conversion of soluble starch into HMF. In addition, levulinic acid was not formed in the reactions.

Full text of DOI:10.1007/s11164-016-2496-9

Res Chem Intermed
DOI 10.1007/s11164-016-2496-9
Catalytic conversion of raw Dioscorea composita biomass
to 5-hydroxymethylfurfural using a combination of metal
chlorides in N,N-dimethylacetamide solvent containing
lithium chloride
1
1
1
•
Feiyun Li
Taichun Guo
•
Gaofeng Shi
1
•
Guoying Wang
1
•
Xinming Lei
Received: 7 November 2015 / Accepted: 25 February 2016
Ó Springer Science+Business Media Dordrecht 2016
Abstract We synthesized 5-hydroxymethylfurfural (HMF) from carbohydrates using
metal chloride catalysts. A 33.2 % yield of HMF was obtained from raw Dioscorea
composita biomass with high starch by using a catalyst system composed of
CrCl Á6H O and LaCl Á6H O at 120 °C for 4 h in N,N-dimethylacetamide containing
3
2
3
2
lithium chloride. The catalyst system is also cost-effective for the conversion of soluble
starch into HMF. In addition, levulinic acid was not formed in the reactions.
Keywords Dioscorea composita Á Soluble starch Á Glucose Á Metal chlorides Á
5
-Hydroxymethylfurfural
Introduction
Diminishing fossil fuel reserves, severe environmental pollution, and a deteriorating
ecosystem have aroused the attention of people to renewable energy resources [1,
2
]. In chemical industries, abundant biomass resources such as carbohydrates are
being studied as potential renewable carbon sources for liquid fuels and other
valuable chemicals [3–6]. In particular, 5-hydroxymethylfurfural (HMF), a six-
carbon furan ring produced from carbohydrate dehydration, has gradually become a
key chemical platform [6–10]. Recent efforts have focused on converting
carbohydrates to HMF. Fructose is a preferred feedstock for high HMF yield.
Electronic supplementary material The online version of this article (doi:10.1007/s11164-016-2496-9)
contains supplementary material, which is available to authorized users.
&
Gaofeng Shi
gaofengshi_lzh@163.com
1
School of Petrochemical Engineering, Lanzhou University of Technology, Lanzhou 730050,
People’s Republic of China
123

F. Li et al.
Building of a biorefinery platform based on HMF requires huge amounts of
inexpensive, renewable, and raw non-food materials containing glucose [6], such as
non-food plants rich in starch. Starch is a polysaccharide that is widely available in
nature and at relatively lower cost than other sugars such as fructose and glucose;
And besides, hydrolysis of starch is favored over that of cellulose [11]. Dioscorea
composita is rich in starch (ca. 64–75 % by dry weight) and no systematic farming
is required. It could be a good sustainable biomaterial for the industrial production
of HMF. We aimed to exploit a catalytic system for direct transformation of raw D.
composita biomass into HMF.
Few studies have been conducted on the possibility of starch for potential HMF
production. Mineral acids [12], solid acids [13], and metallic compounds [14–19]
have been utilized as catalysts for starch dehydration in multiple reaction media,
including biphasic water–organic systems and ionic liquids. Chheda et al. [12]
claimed that 26 % HMF yield from starch can be achieved using HCl as a catalyst
under a biphasic system. Yang et al. [14] also studied starch transformation into
HMF using AlCl Á6H O under a biphasic system composed of an aqueous phase
3
2
saturated with NaCl and an organic extracting-phase tetrahydrofuran, and an HMF
yield of 50 % was achieved; however, the HMF yields are relatively low, primarily
because these catalysts are unsuitable for isomerization of glucose to fructose in an
aqueous phase. Ionic liquid catalytic systems have also been developed for starch
conversion to HMF in high yield, except for biphasic systems. Hu et al. [16] found
that HCl and CrCl in 1-octyl-3-methylimidazolium chloride was an effective
2
catalytic system that can efficiently convert starch to HMF with a yield of 73.0 %.
Considering that ionic liquids are expensive, a catalytic system that can convert
starch to HMF in high yield at low energy cost may be developed.
Chromium catalysts in alkylimidazolium chloride ionic liquids or an N,N-
dimethylacetamide (DMA) system enable catalyzed production of HMF from
glucose, sucrose, and cellulose in good yields [20–23]. Lanthanide (III) ions can
also efficiently catalyze dehydration of fructose into HMF in organic solvents, such
as dimethyl sulfoxide (DMSO) [24]. A work by Yi et al. [17] on the efficient
conversion of raw tapioca root biomass with high starch to HMF using chromium
(
III) chloride as the catalyst provided further information on HMF formation. In the
present paper, we report a cost-effective catalytic system for HMF synthesis from D.
composita biomass based on dehydration of glucose and maltose into HMF. The
catalytic system is composed of a combination of chromium (III) chloride and
lanthanide (III) chloride in DMA-LiCl. This system facilitates access to fuels and
chemicals derived from HMF.
Experimental
General
Commercial chemicals were of analytical grade or better and were used without
further purification. All reactions were performed in glass flasks heated in a
temperature-controlled oil bath with magnetic stirring.
1
23

Catalytic conversion of raw Dioscorea composita biomass…
D. composita was purchased from Xishuangbanna Samost Technology Co., Ltd.
Xishuangbanna, China). Soluble starch, maltoseÁH O, glucose, CrCl Á6H O,
(
2
3
2
LaCl Á6H O, AlCl , LiCl, DMA, N,N-dimethylformamide (DMF), DMSO, and
3 2 3
levulinic acid were of analytical grade and obtained from Tianjin Chemical Reagent
Company (Tianjin, China). A standard sample of HMF (99 %) was obtained from
Sigma-Aldrich Company (USA). D. composita, soluble starch, maltose, and glucose
were dried to constant weight at 60 °C prior to use.
HMF synthesis from sugars
In a typical run, glucose (1 g, 5.6 mM), CrCl Á6H O (5 mol% to glucose) and
3
2
LaCl Á6H O (2.5 mol% to glucose) were added into DMA (10 mL). The mixture
3
2
was heated at 100 °C for 0.5–2.5 h with a magnetic stirrer. Reactions with maltose
were similar except that maltose (1 g, 5.6 mM glucose units), CrCl Á6H O
3
2
(
(
10 mol% with respect to glucose contained in the maltose) and LaCl Á6H O
3
2
5 mol% with respect to glucose contained in the maltose) were mixed in DMA.
HMF synthesis from soluble starch and D. composita
Soluble starch (1 g, 5.84 mmol glucose units) was suspended in 25 mL of DMA and
kept at 160 °C for 1 h under continuous stirring. After the slurry had been allowed
to cool to 100 °C, 2 g of anhydrous LiCl was added [25–27]. The mixture was
heated at 80 °C for 0.5 h under continuous stirring to form a viscous solution. Then,
CrCl Á6H O (30 mol% with respect to glucose contained in the soluble starch) and
3
2
LaCl Á6H O (15 mol% with respect to glucose contained in the soluble starch) were
3
2
mixed in the solution. The reaction mixture was stirred at 100–140 °C for 1–5 h.
Reactions with D. composita were similar except that D. composita (1 g, 64.64 %
starch, 4 mmol glucose units) was dissolved into 50 mL of DMA containing 4 g of
anhydrous LiCl. For soluble starch and D. composita, HMF yields are defined as
follows: HMF yield = moles of HMF produced/moles of starting glucose units.
Product analysis
The analysis of product HMF was conducted with an HP6890/5973 N gas
chromatograph–mass spectrometer (GC–MS) equipped with an HP-5MS capillary
column (30 m 9 0.25 mm 9 0.25 lm), with the following analysis conditions:
detector temperature, 280 °C; injection temperature, 280 °C; carrier gas, argon;
flow rate of carrier gas, 1.0 mL/min. The chemical structure of product HMF was
characterized by Fourier transform infrared spectroscopy (FT-IR, Nicolet Nexus
-
1
6
70, Nicolet, USA) at an optical range of 400–4000 cm and a resolution of
-
1
8
cm
(Fig. S5 in Supporting Information). The product HMF was further
characterized by nuclear magnetic resonance (NMR) analysis (Avance III 400,
Bruker, USA; Figs. S3, S4 in Supporting Information). The sugars from the
dehydration of soluble starch were also characterized by FT-IR and the sugar
samples were precipitated by an ethanol extraction method. Levulinic acid was
estimated by using vanillic aldehyde as the indicator of the spectral measure study.
123

F. Li et al.
Vanillic aldehyde can react with levulinic acid, thereby resulting in a change in
colour from colorless to blue-green.
Results and discussion
HMF synthesis from glucose
The reactivity of glucose in DMA was first explored with a single metal chloride
catalyst (CrCl Á6H O, AlCl ), and reactions were performed at 80–100 °C for
3
2
3
0
.75 h. The results are shown in Table 1. Glucose was converted to HMF in
Table 1 HMF synthesis from glucose
Entry Solvent Catalyst (mol%)
Co-catalyst (mol%) T (°C) Time (h) Molar yield (%)
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
DMSO
DMF
DMA
DMA
DMA
DMA
DMA
DMA
DMA
DMA
DMA
DMA
DMA
DMA
DMA
DMA
DMA
DMA
DMA
DMA
DMA
DMA
DMA
DMA
DMA
DMA
DMA
DMA
CrCl
CrCl
CrCl
CrCl
CrCl
AlCl
AlCl
AlCl
CrCl
CrCl
CrCl
3
3
3
3
3
3
3
3
3
3
3
Á6H
Á6H
Á6H
Á6H
Á6H
Á6H
Á6H
Á6H
Á6H
Á6H
Á6H
2
2
2
2
2
2
2
2
2
2
2
O, 5
O, 5
O, 5
O, 5
O, 5
O, 5
O, 5
O, 5
O, 5
O, 5
O, 5
100
100
100
80
0.75
0.75
0.75
0.75
1.5
1.5
1.5
30.4
6.1
100
80
56.9
1.5
0.75
0.75
1.5
100
100
80
7.3
16
LaCl
LaCl
LaCl
3
3
3
Á6H
Á6H
Á6H
2
2
2
O, 2.5
O, 2.5
O, 2.5
0.75
0.75
1.5
1.5
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
100
100
100
100
80
40.7
77.6
1.5
LaCl
LaCl
3
Á6H
Á6H
2
O, 2.5
0.75
1.5
3
2
O, 2.5
6.3
CrCl
CrCl
CrCl
CrCl
CrCl
CrCl
CrCl
CrCl
CrCl
CrCl
CrCl
CrCl
CrCl
CrCl
CrCl
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
Á6H
Á6H
Á6H
Á6H
Á6H
Á6H
Á6H
Á6H
Á6H
Á6H
Á6H
Á6H
Á6H
Á6H
Á6H
2
O, 5
O, 5
O, 5
AlCl
AlCl
AlCl
3
3
3
Á6H
Á6H
Á6H
2
2
2
O, 2.5
O, 2.5
O, 2.5
0.75
0.75
1.5
8.2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
100
100
100
100
100
100
100
100
100
100
100
100
100
100
17.4
73.1
28.7
27.8
31.9
25
O, 6.82 LaCl
O, 6.25 LaCl
O, 5.62 LaCl
O, 3.75 LaCl
3
3
3
3
3
3
3
3
3
3
3
3
Á6H
Á6H
Á6H
Á6H
Á6H
Á6H
Á6H
Á6H
Á6H
Á6H
Á6H
Á6H
2
O, 0.68
O, 1.25
O, 1.88
O, 3.75
O, 5
0.75
0.75
0.75
0.75
0.75
0.75
0.75
0.5
2
2
2
2
2
2
2
2
2
2
2
O, 2.5
O, 1.5
O, 10
O, 5
LaCl
LaCl
LaCl
LaCl
LaCl
LaCl
LaCl
LaCl
24.4
18.5
41.5
15.3
40.8
77.6
44.1
40
O, 6
O, 5
O, 2.5
O, 2.5
O, 2.5
O, 2.5
O, 2.5
O, 5
1
O, 5
1.5
O, 5
2
O, 5
2.5
Glucose (1 g, 5.6 mM), solvent (10 mL). Catalyst loading and co-catalyst loading is relative to glucose.
Yields are based on GC–MS analysis
1
23

Catalytic conversion of raw Dioscorea composita biomass…
moderate yields (0–30.4 %; entries 3–4, 6–7). The results were negative to some
extent even though the yields were higher at 100 °C than at 80 °C. As a next
attempt, reactions for longer reaction time (1.5 h) were investigated with increased
HMF yields (entries 5, 8). In particular, CrCl Á6H O yielded 56.9 % HMF at 100 °C
3
2
for 1.5 h (entry 5). By contrast, a lesser amount of HMF was obtained with AlCl
3
catalyst (entry 8). Nonetheless, the catalyst system, composed of CrCl Á6H O and
3
2
DMA, was effective for HMF synthesis compared with yields of up to 70 %
achieved in 1-alkyl-3-methylimidazolium chloride [20]. In another attempt, glucose
transformation into HMF was catalyzed by CrCl Á6H O in DMF or DMSO at
3
2
1
00 °C for 0.75 h. The HMF yields were nearly 1.5 % in both cases (entries 1–2).
Based on the above results, glucose conversion into HMF under single metal
chloride conditions was not perfect. Given the recent results on the efficient
catalytic activity of lanthanide (III) ions for saccharide conversion to HMF [24],
LaCl Á6H O was used with CrCl Á6H O (molar ratio between two metal chlorides is
3
2
3
2
CrCl Á6H O:LaCl Á6H O = 2:1) to enhance the ability of glucose transformation
3
2
3
2
into HMF. The reactions were operated at 80–100 °C for 0.75–1.5 h (entries 9–11).
The HMF yields were expectedly increased, and all the HMF yields were higher
than those in the presence of a single metal chloride. Moreover, levulinic acid was
not formed (no change in color from colorless to blue-green) when LaCl Á6H O was
3
2
added as a co-catalyst. Intrigued by the LaCl Á6H O effect, we investigated glucose
3
2
dehydration catalyzed by single LaCl Á6H O (entries 12–13). All the HMF yields
3
2
were unexpectedly not more than 6.3 %. The known catalytic activity of lanthanide
III) ions for glucose conversion to HMF [18, 24] along with our observations of
(
glucose reactivity catalyzed by single LaCl Á6H O indicated that CrCl Á6H O leads
3
2
3
2
to glucopyranose isomerization into fructofuranose. This process is followed by
dehydration catalyzed by a combination of CrCl Á6H O and LaCl Á6H O (Fig. 1).
3
2
3
2
However, the reason why the catalytic performance was enhanced when addition of
LaCl as co-catalysts still needs further investigation. In another attempt, glucose
3
conversion was investigated in the presence of CrCl Á6H O and AlCl at 80–100 °C
3
2
3
for 0.75–1.5 h (entries 14–16). AlCl is less effective for glucose conversion to
3
HMF than LaCl Á6H O. (AlCl only effectively catalyzes the isomerization of
3
2
3
glucose to fructose in aqueous solution [14]).
The effect of molar ratio between CrCl Á6H O and LaCl Á6H O (10 mol% to
3
2
3
2
glucose, total equivalent of two metal chlorides) was studied next. A series of molar
ratios from 10:1 (CrCl Á6H O:LaCl Á6H O) to 1:4 were screened. The reactions
3
2
3
2
were conducted at 100 °C for 0.75 h (entries 17–22). Based on the results, either the
amount of LaCl Á6H O was gradually increased from a molar ratio of 2:1 to 1:4 or
3
2
reduced from 2:1 to 1:10, and all HMF yields were lower than 40.7 %. The optimal
molar ratio of CrCl Á6H O and LaCl Á6H O was apparently 2:1 (entry10). These
3
2
3
2
results further confirmed that CrCl Á6H O can more effectively lead to glucopy-
3
2
ranose isomerization to fructofuranose than LaCl Á6H O, and the perfect glucose
3
2
conversion to HMF depends on the combination of CrCl Á6H O and LaCl Á6H O. A
3
2
3
2
4
1.5 % HMF yield was then obtained when the amount of catalyst was doubled
entry 23). In addition, time-course experiments of HMF synthesis using
CrCl Á6H O and LaCl Á6H O were performed. The maximum HMF yield can be
(
3
2
3
2
123

F. Li et al.
OH
O
OH
O
O
O
OH
O
HO
OH
O
O
n
OH
+
H
OH
O
Soluble starchand Dioscorea composita rich in starch
levulinic acid
formic acid
s
i
s
y
l
o
r
d
y
h
OH
O
OH
HO
HO
h
y
O
O
d
OHC
OH
OH
o r
O
OH
l
y
HO
s
maltose
OH
i
s
5-Hydroxymethylfurfural
metal chlorides
metal chlorides
OH
XX
Cr
3-
X
X
OH
O
O
OH
OH
HO
HO
O
O
HO
HO
OH
OH
OH
OH OH
OH
glucose
putative enediolate intermediate
fructose
Fig. 1 Hydrolysis of carbohydrates into glucose using a combination of metal chlorides in DMA.
Chromium species, depicted here as a hexacoordinate chromium (III) complex, catalyze the isomerization
of glucose into fructose through a putative enediolate intermediate, followed by dehydration to HMF
using a combination of metal chlorides in solvent. And levulinic acid wasn’t formed in the reactions
achieved at 77.6 % after 1.5 h (entry 26), and prolonged reaction led to decreased
HMF yields because of HMF conversion to byproducts such as humins [22].
HMF synthesis from maltose
In our study, HMF synthesis without any catalysts was first attempted at 120 °C for
2
h, and the HMF yield was only 4.8 % (entry 10; Table 2). Subsequently, the HMF
synthesis catalyzed by the combination of CrCl Á6H O and LaCl Á6H O (7.5 mol%
3
2
3
2
to glucose unit in maltose; entry 1) was performed, and only 31.7 % HMF yield was
obtained at 100 °C for 1.5 h. Given that disaccharide conversion to HMF involves
the following steps: (1) hydrolysis breaks the glycosidic bond, converting
disaccharides into glucose and fructose; (2) glucose is isomerized into fructose;
and (3) fructose is dehydrated to form HMF (Fig. 1) [6]. The lower yield for maltose
is attributed to the to difficultly of breaking the glycosidic bond as compared with
the glucose (entry 26; Table 1). Therefore, we sought to utilize the catalytic system
1
23

Catalytic conversion of raw Dioscorea composita biomass…
Table 2 HMF synthesis from maltose
Entry
Solvent
Catalyst (mol%)
Co-catalyst
mol%)
T (°C)
Time (h)
Molar
yield (%)
(
1
2
3
4
5
6
7
8
9
DMA
DMA
DMA
DMA
DMA
DMA
DMA
DMA
DMA
DMA
DMA
DMA
DMA
DMA
CrCl
CrCl
CrCl
CrCl
CrCl
CrCl
CrCl
CrCl
CrCl
3
3
3
3
3
3
3
3
3
Á6H
Á6H
Á6H
Á6H
Á6H
Á6H
Á6H
Á6H
Á6H
2
2
2
2
2
2
2
2
2
O, 5
LaCl
LaCl
LaCl
LaCl
LaCl
LaCl
LaCl
LaCl
LaCl
3
3
3
3
3
3
3
3
3
Á6H
Á6H
Á6H
Á6H
Á6H
Á6H
Á6H
Á6H
Á6H
2
2
2
2
2
2
2
2
2
O, 2.5
O, 3.75
O, 5
100
100
100
100
100
100
100
100
100
120
100
100
120
120
1.5
1.5
1.5
1.5
1.5
0.5
1
31.7
38.1
45.2
44.7
43.4
3.1
O, 7.5
O, 10
O, 12.5
O, 15
O, 10
O, 10
O, 10
O, 10
O, 6.25
O, 7.5
O, 5
O, 5
8.3
O, 5
2
27.1
26.1
4.8
O, 5
2.5
2
1
1
1
1
1
0
1
2
3
4
HCl, 5
HCl, 20
HCl, 5
HCl, 20
1.5
1.5
1.5
1.5
3.8
23.3
14.5
32.8
Maltose (1 g, 5.6 mM glucose units), DMA (10 mL). Catalyst loading and co-catalyst loading is relative
to glucose units contained in the maltose. Yields are based on GC–MS analysis
of CrCl Á6H O and LaCl Á6H O to convert maltose into HMF in DMA. Next, the
3
2
3
2
effect of catalyst and co-catalyst loading on maltose dehydration was investigated
(
molar ratio of CrCl Á6H O and LaCl Á6H O was 2:1; entries 2–5). The HMF yields
3
2
3
2
were increased from 31.7 to 45.2 % when the catalyst and co-catalyst loadings
varied from 7.5 to 15 mol%. Moreover, the HMF yields did not significantly differ
when the catalyst and co-catalyst loadings were higher than 15 %. The effect of
reaction time on maltose dehydration was also tested to better understand the
catalytic system (entries 6–9). The HMF yields became higher along with increasing
time from 0.5 to 1.5 h. However, when the reaction time was extended to 1.5 h or
longer, the HMF yield decreased to some extent because of HMF conversion
byproducts such as humins [22]. Furthermore, the HMF synthesis from maltose
using hydrochloric acid as the catalyst was reported, and a 32.8 % HMF yield was
achieved in the presence of 20 mol% hydrochloric acid at 120 °C for 1.5 h (entry
1
4). By contrast, the best performance of HMF yield at 45.2 % was realized under
milder and less acidic (corrosive) condition (entry 3).
HMF synthesis from soluble starch and D. composita
Starch is a polyglucan linked with glucose units through a-glycosidic linkages. In
addition to the challenge of glucose isomerization to fructose, two most important
factors must be considered for industrial applications; that is, the reduced solubility
in almost all solvents and glycosidic bonds are difficult to break. However, we
suspected that the solubility of soluble starch in DMA-LiCl and the efficient
conversion of glucose and maltose into HMF with the combination of CrCl Á6H O
3
2
123

F. Li et al.
Table 3 HMF synthesis from soluble starch and D. composita
Entry
Solvent
Catalyst
mol%)
Co-catalyst
(mol%)
T
Time
(h)
Molar
yield (%)
(
(°C)
1
2
3
4
5
6
7
8
9
DMA-LiCl (8.5 %wt)
DMA-LiCl(8.5 %wt)
DMA-LiCl (8.5 %wt)
DMA-LiCl (8.5 %wt)
DMA-LiCl (8.5 %wt)
DMA-LiCl (8.5 %wt)
DMA-LiCl (8.5 %wt)
DMA-LiCl (8.5 %wt)
DMA-LiCl (8.5 %wt)
DMA-LiCl (8.5 %wt)
DMA-LiCl (8.5 %wt)
DMA-LiCl (8.5 %wt)
DMA-LiCl (8.5 %wt)
DMA-LiCl (8.5 %wt)
DMA-LiCl (8.5 %wt)
DMA-LiCl (8.5 %wt)
DMA-LiCl (8.5 %wt)
CrCl
CrCl
CrCl
CrCl
CrCl
CrCl
CrCl
CrCl
CrCl
CrCl
CrCl
CrCl
CrCl
CrCl
CrCl
CrCl
CrCl
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
Á6H
Á6H
Á6H
Á6H
Á6H
Á6H
Á6H
Á6H
Á6H
Á6H
Á6H
Á6H
Á6H
Á6H
Á6H
Á6H
Á6H
2
O, 10
2
O, 10
2
O, 10
2
O, 10
2
O, 20
2
O, 30
2
O, 40
2
O, 30
2
O, 30
2
O, 30
2
O, 30
2
O, 30
2
O, 30
2
O, 30
2
O, 30
2
O, 30
2
O, 30
LaCl
LaCl
LaCl
LaCl
LaCl
LaCl
LaCl
LaCl
LaCl
LaCl
LaCl
LaCl
LaCl
LaCl
LaCl
LaCl
LaCl
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
Á6H
Á6H
Á6H
Á6H
Á6H
Á6H
Á6H
Á6H
Á6H
Á6H
Á6H
Á6H
Á6H
Á6H
Á6H
Á6H
Á6H
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
2
O, 5
100
120
120
140
120
120
120
120
120
120
120
120
120
120
120
120
120
1.5
1.5
3
2.3
4.3
O, 5
O, 5
19
O, 5
1.5
1.5
1.5
1.5
1
18.6
6.5
O, 10
O, 15
O, 20
O, 15
O, 15
O, 15
O, 15
O, 15
O, 15
O, 15
O, 15
O, 15
O, 15
8.7
8.8
3.5
2
12.4
52.2
54.5
53.4
8.6
1
1
1
1
1
1
1
1
0
1
2
3
4
5
6
7
3
4
5
1
2
12.9
31.7
33.2
33.3
3
4
5
Soluble starch (1 g, 5.84 mM glucose units, entries 1–12), D. composita (1 g, 64.64 % starch, 4 mM
glucose units, entries 13–17) The weight percent of LiCl is relative to DMA (entries 1–12; DMA, 25 mL;
entries 13–17, DMA, 50 mL). Catalyst loading and co-catalyst loading is relative to glucose units con-
tained in the polysaccharides. Yields are based on GC–MS analysis
Dioscorea composita
1160
1080
Soluble starch
1080
1160
9
85
1
800 1700 1600 1500 1400 1300 1200 1100 1000 900 800 700 600 500
-1
Wavenumber (cm )
Fig. 2 The FT-IR spectra of soluble starch and D. composita
and LaCl Á6H O can achieve superior results. The HMF yield from soluble starch
3
2
was up to 54.5 % within 5 h at 120 °C (entries 8–12; Table 3). These yields agreed
well with the reports of HMF synthesis from starch by Yang et al. [14] and Hu et al.
1
23

Catalytic conversion of raw Dioscorea composita biomass…
Sample
Glucose
1020
1020
Maltose
1030
1
800 1700 1600 1500 1400 1300 1200 1100 1000 900 800 700 600 500
-1
Wavenumber (cm )
Fig. 3 The FT-IR spectra of analytical grade sugars and the hydrolysis product of starch. A sample from
the reaction mixture containing sugars (Table S1, entry 2) is precipitated by an ethanol extraction method
[15]. We then discovered another desirable attribute of the present process: the
ready conversion of D. composita with high starch content. For the conversion of
untreated D. composita, 33.2 % HMF yield was significantly obtained at 120 °C for
4
h (entry 16). Levulinic acid was not formed in all experiments. The
physicochemical changes of the extractant derived from soluble starch were
characterized using FT-IR to explore the scheme of HMF formation from starch in
DMA-LiCl. Figure 2 shows the FT-IR spectral profiles of the D. composita and
analytical grade soluble starch. The absorption bands of the raw biomass by the
C–O–C and C–OH stretching modes that are typically observed in carbohydrates
-
1
such as starch were identified in the 900–1200 cm regions [28, 29]. The banding
pattern of the raw biomass was very similar with that of soluble starch, indicating
that the main biomass composition is starch. Based on the determination of total
123

F. Li et al.
reducing sugar through the dinitrosalicylic acid (DNS) method (Table S1 in
Supporting Information), the sugars in the reaction mixture (Table S1, entry 2) were
characterized using an FT-IR spectrometer. As shown in Fig. 3, the new absorption
-
1
peak at 1020 cm for the sample is attributed to the bending vibration of the
hydroxyl group in mono- and disaccharides [30]. This shift in the absorbance signals
is probably caused by starch depolymerization into smaller molecules such as
-
1
glucose or maltose (the absorbance band around the 1020 cm region is frequently
found from glucose and maltose standards; Fig. 3). Given the results on sugar
conversion and reports in the literature, we propose that the supposed scheme of
HMF formation from starch in DMA-LiCl is the following (Fig. 1): (1) starch
hydrolysis into maltose and glucose catalyzed by the combination of CrCl Á6H O
3
2
and LaCl Á6H O; (2) isomerization of glucose monomers into fructose catalyzed by
3
2
CrCl Á6H O; and (3) fructose dehydration to form HMF catalyzed by the
3
2
combination of CrCl Á6H O and LaCl Á6H O. However, further investigation is
3
2
3
2
needed to confirm this mechanism.
Conclusion
Effective catalytic conditions for D. composita transformation into HMF were
achieved using CrCl Á6H O and LaCl Á6H O in DMA-LiCl. The cost-effective
3
2
3
2
catalytic system has an excellent potential for industrial application. Further
mechanistic studies of the reactions involved are beneficial for the exploitation of
non-food plants containing starch as raw materials for renewable energy and
chemical industries.
Compliance with ethical standards
Conflict of interest The authors have declared no conflict of interest.
References
1. A.E. Farrell, R.J. Pelvin, B.T. Turner, A.D. Jones, M. O’hare, D.M. Kammen, Science 311, 506
(2006)
2
3
4
5
6
. P. Szuromi, B. Jasny, D. Clery, J. Austin, B. Hanson, Science 315, 781 (2007)
. M. Bicker, J. Hirth, H. Vogel, Green Chem. 5, 280 (2003)
. D.A. Simonetti, J.A. Dumesic, ChemSusChem 1, 725 (2008)
. J.N. Chheda, G.W. Hubera, J.A. Dumesic, Angew. Chem. Int. Ed. 46, 7164 (2007)
. R.-J. van Putten, J.C. van der Waal, E. de Jong, C.B. Rasrendra, H.J. Heeres, J.G. de Vries, Chem.
Rev. 2013(113), 1499 (2013)
7. Y. Roman-Leshkov, C.J. Barrett, Z.Y. Liu, J.A. Dumesic, Nature 2007(447), 982 (2007)
8. M. Mascal, E.B. Nikitin, Angew. Chem. Int. Ed. 47, 7924 (2008)
9. T. Stahlberg, W.J. Fu, J.M. Woodley, A. Riisager, ChemSusChem 4, 451 (2011)
10. T. Buntara, S. Noel, P.H. Phua, I. Melian-Cabrera, J.G. de Vries, H.J. Heeres, Angew. Chem. Int. Ed.
50, 7083 (2011)
1
1
1
1. C.H. Zhou, X. Xia, C.X. Lin, D.S. Tong, J. Beltramini, Chem. Soc. Rev. 40, 5588 (2011)
2. J.N. Chheda, Y. Roman-Leshkov, J.A. Dumesic, Green Chem. 9, 342 (2007)
3. Y. Yang, X. Xiang, D.M. Tong, C.W. Hu, M.M. Abu-Omar, Bioresource Technol. 116, 302 (2012)
1
23

Catalytic conversion of raw Dioscorea composita biomass…
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
4. Y. Yang, C.W. Hu, M.M. Abu-Omar, Green Chem. 14, 509 (2012)
5. S.Q. Hu, Z.F. Zhang, J.L. Song, Y.X. Zhou, B.X. Han, Green Chem. 11, 1746 (2009)
6. J.A. Chun, J.W. Lee, Y.B. Yi, S.S. Hong, C.H. Chung, Starch/St a¨ rke 62, 326 (2010)
7. Y.B. Yi, M.G. Ha, J.W. Lee, C.H. Chung, Chem. Eng. J. 180, 370 (2012)
8. T.F. Wang, Y.J. Pagan-Torres, E.J. Combs, J.A. Dumesic, B.H. Shanks, Top. Catal. 55, 657 (2012)
9. X.L. Tong, L.H. Yu, G.X. Nie, Z.M. Li, J.B. Liu, S. Xue, Environ. Prog. Sustain. 34, 1136 (2015)
0. H.B. Zhao, J.E. Holladay, H. Brown, Z.C. Zhang, Science 316, 1597 (2007)
1. J.B. Binder, R.T. Raines, J. Am. Chem. Soc. 131, 1979 (2009)
2. C. Wang, L. Fu, X.L. Tong, Q.W. Yang, W.Q. Zhang, Carbohydr. Res. 347, 182 (2012)
3. B. Kim, J. Jeong, D. Lee, S. Kim, H.J. Yoon, Y.S. Lee, J.K. Cho, Green Chem. 13, 1503 (2011)
4. K.I. Seri, Y. Inoue, H. Ishida, Chem. Lett. 1, 22 (2000)
5. T. Heinze, P. Talaba, U. Heinze, Carbohydr. Polym. 42, 411 (2000)
6. M. Terbojevich, A. Cosani, G. Conio, A. Ciferri, Macromolecules 18, 640 (1985)
7. C.L. McCormick, P.A. Callais, Macromolecules 18, 2394 (1985)
8. J.W. Lee, M.G. Ha, Y.B. Yi, C.H. Chunget, Carbohydr. Res. 346, 177 (2011)
9. V. Bellon-Maurel, C. Vallat, D. Goffinet, Appl. Spectrosc. 49, 556 (1995)
0. Y.B. Yi, M.G. Ha, J.W. Lee, C.H. Chung, Chem. Eng. J. 180, 370 (2012)
123