Catalytic Conversion of Carbohydrates to Levulinate Ester over Heteropolyanion-Based Ionic Liquids

Article abstract of DOI:10.1002/cssc.201601080

An efficient one-pot approach for the production of levulinate ester from renewable carbohydrates is demonstrated over heteropolyanion-based ionic liquid (IL-POM) catalysts with alcohols as the promoters and solvents. The relationships between the structure, acidic strength, and solubility of the IL-POM in methanol and the catalytic performance were studied intensively. A cellulose conversion of 100 % could be achieved with a 71.4 % yield of methyl levulinate over the catalyst [PyPS]₃PW₁₂O₄₀ [PyPS=1-(3-sulfopropyl)pyridinium] at 150 °C for 5 h. This high efficiency is ascribed to the reasonably high activity of the ionic liquid (IL) catalyst and reaction coupling with rapid in situ esterification of the generated levulinic acid with the alcohol promoter, which allows the insolubility of cellulose encountered in biomass conversion to be overcome. Furthermore, the present process exhibits high feedstock adaptability for typical carbohydrates and handy catalyst recovery by a simple self-separation procedure through temperature control.

Full text of DOI:10.1002/cssc.201601080

DOI: 10.1002/cssc.201601080
Full Papers
Catalytic Conversion of Carbohydrates to Levulinate Ester
over Heteropolyanion-Based Ionic Liquids
Changhua Song, Sijie Liu, Xinwen Peng, Jinxing Long,* Wenyong Lou, and Xuehui Li*[a]
An efficient one-pot approach for the production of levulinate
ester from renewable carbohydrates is demonstrated over het-
eropolyanion-based ionic liquid (IL-POM) catalysts with alco-
hols as the promoters and solvents. The relationships between
the structure, acidic strength, and solubility of the IL-POM in
methanol and the catalytic performance were studied inten-
sively. A cellulose conversion of 100% could be achieved with
for 5 h. This high efficiency is ascribed to the reasonably high
activity of the ionic liquid (IL) catalyst and reaction coupling
with rapid in situ esterification of the generated levulinic acid
with the alcohol promoter, which allows the insolubility of cel-
lulose encountered in biomass conversion to be overcome.
Furthermore, the present process exhibits high feedstock
adaptability for typical carbohydrates and handy catalyst re-
covery by a simple self-separation procedure through tempera-
ture control.
a
71.4% yield of methyl levulinate over the catalyst
[PyPS]3PW12O40 [PyPS=1-(3-sulfopropyl)pyridinium] at 1508C
Introduction
The overexploration of petroleum resources and the deteriora-
tion of the environment have spurred the exploration of green
processing from nonedible biomass to biofuels and biochemi-
cals.[1] Numerous studies have focused on the one-pot conver-
sion of carbohydrates to value-added and versatile chemical
platforms, such as 5-hydroxymethylfurfural (5-HMF),[2] levulinic
acid (LA),[3] and 2,5-furandicarboxylic acid (FDCA),[4] to support
the sustainable production of valuable compounds, fuel, and
power.[5] In this context, LA and its esters have been highlight-
ed for their high potential as substitutes for current petro-
chemicals.[6] For example, diphenolic acid, synthesized by the
condensation of LA with phenols, has been envisaged as a de-
sirable substitute for 4,4’-(propane-2,2-diyl)diphenol in the
polycarbonate and epoxy-resin industries.[7] Similarly, methyl
levulinate (ML) has found application in the synthesis of
g-valerolactone,[8] which is used as a solvent and an alternative
fuel additive.[9] Furthermore, ethyl levulinate (EL) is considered
as a new biobased diluent for biodiesel owing to its high satu-
rated fatty acid content.[10]
geneous catalysts have been pursued as alternatives. For ex-
ample, various solid acid catalysts, such as sulfonated metal
oxides, acidic resins, and zeolites, have been explored wide-
ly.[6b,12] However, it should be noted that solid acids showed
lower activities for cellulose in comparison with their superior
catalytic performances for glucose and fructose.[13] This can be
ascribed to the limited contact between the solid catalysts and
the water-insoluble cellulose,[14] and the insolubility is attribut-
ed to the presence of inter- and intramolecular hydrogen
bonding in the stabilized twofold-helix conformation of cellu-
lose.[15] Therefore, ecofriendly and easily regenerated catalysts
with desirable activities should be explored. Furthermore, an
appropriate acidic strength and reaction temperature are cru-
cial to the conversion of carbohydrates to LA and its esters, es-
pecially in the rate-determining opening of the H-bonded cel-
lulose structure. As the acidity of the catalyst employed and
the temperature increase, the hydrolysis rate can be enhanced.
However, a common problem is that an elevated temperature
and a higher acid concentration facilitate the formation of
humins,[16] which is regarded as a main byproduct in the carbo-
hydrate industry. Thus, an acid catalyst with a tunable acidic
strength should be devised and coupled with an intensified
process to replace the use of higher reaction temperatures to
accelerate the depolymerization of cellulose.
Generally, the conventional production of LA and its esters is
achieved through acid-catalyzed processes with mineral acids,
such as H2SO4, HCl, and HBr, as homogenous catalysts;[6b,11] un-
fortunately, problems such as environmental pollution, catalyst
separation issues, and corrosion are encountered. Thus, hetero-
Ionic liquids (ILs), which combine the advantages of homo-
genous and heterogeneous catalysts with tunable structures
and physiochemical properties, are considered to be a promis-
ing solution.[17] For example, the use of acidic ILs is an efficient
and sustainable approach for cellulose and lignocellulosic bio-
mass conversion, and ILs containing SO3H groups have attract-
ed considerable attention owing to their strong Brçnsted acidi-
ties.[18] In 2009, the first use of Brçnsted acidic ILs, namely,
1-methyl-3-(3-sulfopropyl)imidazolium chloride ([MIMPS]Cl) and
1-methyl-3-(3-sulfobutyl)imidazolium chloride ([MIMBS]Cl), to
[a] Dr. C. Song, Dr. S. Liu, Dr. X. Peng, Prof. Dr. J. Long, Dr. W. Lou,
Prof. Dr. X. Li
School of Chemistry and Chemical Engineering, Pulp & Paper Engineering
State Key Laboratory of China
South China University of Technology
Guangzhou, 510640 (China)
E-mail: cejxlong@scut.edu.cn
cexhli@scut.edu.cn
Supporting Information for this article can be found under:
http://dx.doi.org/10.1002/cssc.201601080.
ChemSusChem 2016, 9, 1 – 11
1
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
These are not the final page numbers! ÞÞ

Full Papers
dissolve cellulose was reported by Amarasekara and Owereh.[19]
Then, Liu et al. proposed the use of HSO3-functional ILs based
on imidazolium and quaternary ammonium cations in the hy-
drolysis of cellulose, and 1-(3-sulfopropyl)triethylammonium
hydrogen sulfate ([TEAPS]HSO4) gave a maximum total reduc-
ing sugar yield of 99% at 1008C.[20] The IL 1-methyl-3-(3-
sulfobutyl)imidazolium hydrogen sulfate ([MIMBS]HSO4) is
a robust and efficient catalyst for cellulose degradation and
biomass liquefaction that demonstrates better catalytic activity
compared with those of conventional imidazolium ILs.[21] Re-
cently, Siankevich et al. reported that the mixture of the ILs 1-
catalysts with tunable acidity were designed and synthesized
for the conversion of carbohydrates to levulinate ester in an ef-
ficient and intensified process. During the reaction, the alcohol
acts as both promoter and solvent and can esterify the gener-
ated LA rapidly to accelerate the whole carbohydrate conver-
sion process through reaction coupling. Additionally, the rela-
tionships between the structure, acidic strength, and solubility
in methanol of the IL and the catalytic activity are evaluated
and discussed.
(2-hydroxyethyl)-3-methylimidazolium chloride ([C2OHmim]Cl) Results and Discussion
and 1-methyl-3-(3-sulfobutyl)imidazolium trifluoromethanesul-
Determination of acidic strengths of IL-POM catalysts
fonate ([MIMBS]CF3SO3) was a powerful system for the produc-
tion of 5-HMF from cellulose.[22] Numerous studies of the pro-
duction of LA and its esters have also been reported. Ren et al.
showed that the IL 1-methyl-3-(3-sulfopropyl)imidazolium hy-
drogen sulfate ([MIMPS]HSO4) was an efficient catalyst for the
conversion of cellulose to LA with a yield of 55.0% under the
assistance of microwave heating.[23] Amarasekara and Wiredu
found that [MIMPS]Cl was an efficient catalyst in cellulose de-
composition, and an EL yield of 19.0% was obtained for a reac-
In this work, a series of typical IL-POMs with similar structures
were designed and synthesized (Scheme 1), and their acidities
were characterized by the Hammett method[31] and pH mea-
surement. As depicted in Table 1 and Figure S7, [PyPS]3PW12O40
[PyPS=1-(3-sulfopropyl)pyridinium], [EIMPS]3PW12O40 [EIMPS=
1-ethyl-3-(3-sulfopropyl)imidazolium], [TEAPS]3PW12O40, and
[MIMPS]3PW12O40 have strong Hammett acidic strengths,
whereas [PyPS]2HPW12O40, [Py]3PW12O40 (Py=pyridinium),
[i-PIMPS]3PW12O40 [i-PIMPS=1-isopropyl-3-(3-sulfopropyl)imida-
zolium], [IMPS]3PW12O40 [IMPS= 1-(3-sulfopropyl)imidazolium],
and [PyPS]H2PW12O40 have moderate Hammett acidic strengths.
The results from pH determination show a similar trend,
except for [Py]3PW12O40 and [IMPS]3PW12O40 owing to their
poor solubilities at room temperature.
tion performed at 1708C for 12 h in
a water–ethanol
medium.[24] Shen et al. showed that a 39.4% LA yield could be
achieved with the acidic IL catalyst [MIMBS]HSO4.[25] In our
recent work, we also demonstrated that 98.4% of cellulose
could be converted into 31.1% butyl levulinate (BL) with the
SO3H-functionalized [MIMBS]HSO4 IL catalyst.[26] However, the
yield and selectivity of the target products in the aforemen-
tioned processes should be enhanced.
Solubilities of IL-POM catalysts in methanol
Recently, heteropolyacid-based catalysts with agreeable cat-
alytic behavior have been applied widely in reactions such as
esterification,[27] hydrogenation,[28] and oxidation.[29] Owing to
their intrinsic structures and physiochemical properties, hetero-
polyanion-based ionic liquid (IL-POM) catalysts are generally
immiscible with polar solvents at ambient temperatures but
miscible at elevated temperatures;[30] thus, they can act as ho-
mogeneous catalysts at suitable temperatures and then precip-
itate as gelatinous solids at low temperatures to provide re-
markably convenient catalyst separation and recycling. Further-
more, the acidic strength of IL-POMs can be fine-tuned by var-
iation of the cation and anion. Therefore, a series of IL-POMs
The UV/Vis spectra of saturated solutions of various IL-POMs
are depicted in Figure 1, and the solubilities of the IL-POMs in
methanol are listed in Table 2. The prominent peaks at l
ꢀ210 nm are associated with the O!P transition, and those
at lꢀ265 nm are assigned to the ligand-to-metal charge trans-
fer (O2ꢁ!W6+).[32] It should be noted that the imidazole rings
of the IL-POMs generally show UV/Vis absorption at l=
210 nm,[33] which is overlapped by the O!P transition. There-
fore, the O2ꢁ!W6+ absorption band (l=260 nm) was selected
for the solubility determination of the IL-POMs in saturated
methanol solution. The UV/Vis absorbance results and those
Scheme 1. Structures of the as-synthesized IL-POM catalysts.
&
ChemSusChem 2016, 9, 1 – 11
www.chemsuschem.org
2
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÝÝ These are not the final page numbers!

Full Papers
Table 1. Hammett-function determination and pH values of IL-POMs at different sample concentrations.[a]
Entry
IL-POM
Amax
[I]
[%]
[IH+]
[%]
Ho
pH
C1 [0.10 mm]
C2 [1.0 mm]
C3 [10 mm]
1
2
3
4
5
6
7
8
9
–
3.210
1.607
1.625
1.640
1.623
1.480
1.610
1.632
1.638
1.620
1.579
1.635
100
50.07
0
–
–
–
–
[PyPS]H2PW12O40 (A1)
[PyPS]2HPW12O40 (A2)
[PyPS]3PW12O40 (A3)
[Py]3PW12O40 (A4)
[PyPS]HSO4 (A5)[b]
[IMPS]3PW12O40 (B1)
[MIMPS]3PW12O40 (B2)
[EIMPS]3PW12O40 (B3)
[i-PIMPS]3PW12O40 (B4)
[BnIMPS]3PW12O40 (B5)
[TEAPS]3PW12O40 (C)
49.93
49.37
48.92
49.45
53.88
49.83
49.16
48.97
49.53
50.81
49.06
0.991
1.001
1.009
0.999
0.922
0.993
1.005
1.008
0.998
0.976
1.006
3.22
3.30
3.28
4.32
3.88
4.98
3.33
3.32
3.36
3.32
3.35
2.30
2.39
2.39
3.80
2.70
4.55
2.41
2.36
2.50
2.47
2.38
1.44
1.48
1.47
3.63
1.71
4.27
1.50
1.47
1.67
1.69
1.43
50.63
51.08
50.55
46.12
50.17
50.84
51.03
50.47
49.19
50.94
10
11
12
[a] Ho =pKa+log[I]/[IH+]; indicator: 4-nitroaniline (pKa =0.99). For the determination of Hammett functions, IL-POM (0.1 mm) and 2.0 mm 4-nitroaniline in
methanol. [b] 10.0 mm in methanol for the determination of the Hammett function of [PyPS]HSO4.
from the determination of the W content are in accordance
with each other. Most of the IL-POMs have moderate solubili-
ties in the range 48.1 ([TEAPS]3PW12O40) to 68.7 gLꢁ1
([PyPS]3PW12O40) at room temperature. However, it is notewor-
thy that only a trace amount of W was detected in a saturated
methanol solution of [Py]3PW12O40, which exhibits a very weak
absorption band; this clearly indicates that [Py]3PW12O40 is only
slightly soluble in methanol at room temperature (Table 2,
Entry 4).
Catalytic conversion of cellulose to levulinate ester
Microcrystalline cellulose was selected for the IL-POM catalytic
activity test, and the results are listed in Table 3. In the absence
of catalyst, the cellulose conversion was only 3.6% without
Figure 1. UV/Vis spectra of methanol solutions saturated with IL-POMs. A1:
any detectable ML production (Table 3, Entry 1); therefore, the
[PyPS]H2PW12O40; A2: [PyPS]2HPW12O40; A3: [PyPS]3PW12O40; A4: [Py]3PW12O40
B1: [IMPS]3PW12O40; B2: [MIMPS]3PW12O40; B3: [EIMPS]3PW12O40; B4: [i-
PIMPS]3PW12O40; B5: [BnIMPS]3PW12O40; and C: [TEAPS]3PW12O40.
;
thermal chemical reaction at 1508C and the autocatalytic per-
formance of the solvent are negligible. As shown in Table 3, it
is interesting that the conversion of cellulose and the yield of
ML are promoted significantly by the addition of the IL-POMs
containing cationic sulfonic groups coupled with POM func-
tional anions. For example, 100% cellulose conversion was
achieved with 71.4% ML yield and 79.7% process efficiency
over [PyPS]3PW12O40 (Table 3, Entry 4). Generally, the domino
processes for the formation of ML from cellulose, including the
hydrolysis of cellulose to glucose, the dehydration/rehydration
process from glucose to LA, and the subsequent esterification
of LA with methanol, can be enhanced by acidic catalysts. On
the other hand, direct cellulose methanolysis, which also con-
tributes to ML production, is promoted in the presence of an
acid catalyst. Hence, the integration of these typical acid-
catalyzed processes over IL-POMs with high acidic strengths
can promote cellulose degradation and ML generation.
Table 2. Solubilities of IL-POMs in methanol, as determined by ICP-OES
and UV/Vis spectroscopy methods.
Entry
IL-POM
Solubility [gLꢁ1
]
SICP
[a]
[b]
SUV
1
2
3
4
5
6
7
8
9
[PyPS]H2PW12O40 (A1)
[PyPS]2HPW12O40 (A2)
[PyPS]3PW12O40 (A3)
[Py]3PW12O40 (A4)
[IMPS]3PW12O40 (B1)
[MIMPS]3PW12O40 (B2)
[EIMPS]3PW12O40 (B3)
[i-PIMPS]3PW12O40 (B4)
[BnIMPS]3PW12O40 (B5)
[TEAPS]3PW12O40 (C)
[PyPS]3PW12O40 (A3)
58.4
62.5
68.7
1.98
53.1
50.8
54.5
50.3
60.6
48.1
0
59.8
63.1
71.3
2.88
53.1
51.5
55.5
52.3
62.0
49.4
0
10
11[c]
The substitution of the IL-POM cation has a pronounced
effect on the ML yield, as is demonstrated clearly by two sys-
tems with the same anion, that is, [PyPS]3PW12O40 and [MIM-
PS]3PW12O40, which yielded similar apparent acidic strengths
(Table 1). However, the catalytic activities of these IL-POMs are
remarkably different. The highest ML yield of 71.4% was ach-
[a] Solubility of IL-POM from the
W content detected by ICP-OES
(Table S1). [b] Solubility of IL-POM from the UV/Vis spectra result.
[c] [PyPS]3PW12O40 content in the chloroform-soluble fraction from the
product separation (Scheme S1); the “0” value is defined as “not detect-
ed”.
ChemSusChem 2016, 9, 1 – 11
www.chemsuschem.org
3
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
These are not the final page numbers! ÞÞ

Full Papers
Table 3. Catalytic performances of different IL-POMs for the conversion of microcrystalline cellulose.[a]
Entry
Catalyst
Promoter/solvent
Conversion [%]
Yield of levulinate ester
YLA [%]
Yglucose [%]
Efficiency [%]
product
yield [%]
1
2
3
4
5
6
7
8
9
10
11
12
13
14[b]
15
16
17
18
19
blank
methanol
methanol
methanol
methanol
methanol
methanol
methanol
methanol
methanol
methanol
methanol
methanol
methanol
methanol
n-hexane
H2O
3.6ꢂ0.1
100
100
ML
ML
ML
ML
ML
ML
ML
ML
ML
ML
ML
ML
ML
ML
LA
not detected
48.4ꢂ1.9
53.4ꢂ2.9
71.4ꢂ1.7
not detected
11.1ꢂ0.3
26.2ꢂ1.1
36.5ꢂ1.8
45.8ꢂ2.9
38.8ꢂ2.1
33.9ꢂ1.4
29.0ꢂ0.9
29.7ꢂ1.3
41.6ꢂ1.9
not detected
–
0.9ꢂ0.01
6.8ꢂ1.0
6.5ꢂ1.2
8.3ꢂ0.5
1.2ꢂ0.1
1.2ꢂ0.2
1.7ꢂ0.03
4.4ꢂ0.1
8.8ꢂ0.4
6.0ꢂ0.9
13.5ꢂ0.9
2.9ꢂ0.1
3.0ꢂ0.8
5.0ꢂ1.1
not detected
18.1ꢂ1.5
3.9ꢂ0.1
4.0ꢂ0.1
9.4ꢂ0.1
0.2ꢂ0.02
1.5ꢂ0.2
1.6ꢂ0.4
0.3ꢂ0.02
1.0ꢂ0.1
0.2ꢂ0.01
15.5ꢂ0.4
3.1ꢂ0.5
1.1ꢂ0.2
1.9ꢂ0.2
0.5ꢂ0.2
3.5ꢂ0.3
3.4ꢂ0.2
3.1ꢂ0.1
not detected
32.9ꢂ0.5
0.5ꢂ0.1
1.5ꢂ0.4
1.0ꢂ0.01
0.9ꢂ0.01
55.2ꢂ2.0
59.9ꢂ2.9
79.7ꢂ1.8
1.2ꢂ0.1
12.3ꢂ0.4
27.9ꢂ1.1
40.9ꢂ1.7
54.6ꢂ3.0
44.8ꢂ1.9
47.4ꢂ1.6
31.9ꢂ0.9
32.7ꢂ1.2
46.6ꢂ2.2
–
[PyPS]H2PW12O40 (A1)
[PyPS]2HPW12O40 (A2)
[PyPS]3PW12O40 (A3)
[Py]3PW12O40 (A4)
[PyPS]HSO4 (A5)
[IMPS]3PW12O40 (B1)
[MIMPS]3PW12O40 (B2)
[EIMPS]3PW12O40 (B3)
[i-PIMPS]3PW12O40 (B4)
[BnIMPS]3PW12O40 (B5)
[TEAPS]3PW12O40 (C)
H3PW12O40
100
3.2ꢂ0.2
88.9ꢂ1.3
100
100
100
100
100
100
100
H2SO4
100
[PyPS]3PW12O40 (A3)
[PyPS]3PW12O40 (A3)
[PyPS]3PW12O40 (A3)
[PyPS]3PW12O40 (A3)
[PyPS]3PW12O40 (A3)
0.3ꢂ0.1
71.4ꢂ1.3
100
100
100
–
EL
n-PL
i-PL
18.1ꢂ1.5
61.4ꢂ2.8
41.0ꢂ1.4
30.9ꢂ1.3
ethanol
n-propanol
2-propanol
57.5ꢂ2.8
37.0ꢂ1.4
21.5ꢂ1.4
Conditions: [a] microcrystalline cellulose (0.162 g); catalyst (0.5 mmol); alcohol, n-hexane, or water (20 mL); 2 MPa N2; 1508C; 5 h. Efficiency [%]=YML
[%]+YLA [%]. n-PL: n-propyl levulinate; i-PL: isopropyl levulinate. [b] H2SO4 (1.5 mmol).
ieved with [PyPS]3PW12O40, whereas a yield of only 36.5% was
obtained in the presence of [MIMPS]3PW12O40. Imidazole is con-
siderably more basic than pyridine, though both of them are
aromatic heterocyclic compounds. As the lone pair of the H-
bearing N atom in imidazole is required to maintain aromatici-
ty, the other N atom with a lone pair in its sp2 orbital, perpen-
dicular to the aromatic system, is the basic one.[34] Thus, com-
pared with that of [PyPS]3PW12O40, the acidity of [MIM-
PS]3PW12O40 is partly compensated by its basic N atom, and
this results in lower catalytic activity. To verify this assumption,
H, ethyl, isopropyl, and benzyl substituents were grafted onto
the basic N atom to adjust the basicity of the IL in the control
experiments (Table 3, Entries 7 and 9–11, respectively). As the
electron-donating ability increases in the order benzyl<H<
methyl<ethyl<isopropyl,[35] theoretically, the basicity of the
mentioned N atom in imidazole should increase in the order
cation can enhance the acidic strength of the as-synthesized
IL-POMs, though the acidity of heteropolyacid is normally
stronger than that of a sulfonated acid. The results indicate
that a stronger acidic strength of the IL-POM is desirable to
trigger the reaction. The H3PW12O40 catalyst exhibits a signifi-
cantly lower ML yield of 29.7%, which implies that protons
from the sulfonic-containing cation favor ML formation more
than those from the parent heteropolyacid. On the other hand,
the catalytic activity of the IL-POM without a propanesulfonate
group ([Py]3PW12O40) was also investigated, and this catalyst re-
sulted in 3.2% cellulose conversion and no detectable ML for-
mation (Table 3, Entry 5). The poor solubility of [Py]3PW12O40 is
considered to be the main reason for this result owing to
solid-to-solid contact limitations. However, it is also plausible
that the weaker acidic strength (Table 1, Entry 5) may also con-
tribute to the seemingly poor catalytic activity exhibited by
this system.
[BnIMPS]+ >[IMPS]+ >[MIMPS]+ >[EIMPS]+ >[i-PIMPS]+.
The
catalytic performance of the IL-POMs containing an
imidazole ring follows the sequence [EIMPS]3PW12O40 >
[i-PIMPS]3PW12O40 ꢀ[MIMPS]3PW12O40 >[IMPS]3PW12O40 >
[BnIMPS]3PW12O40, which accords well with their acidic-strength
sequence (Table 1 and Figure S7). The catalytic activity of
[i-PIMPS]3PW12O40 is lower than that of [EIMPS]3PW12O40, and
this can be ascribed to the steric hindrance of the isopropyl
group in cellulose conversion.
To determine whether the heteropolyanion is indispensable
to the achievement of a high yield of ML, the catalytic per-
formance of [PyPS]HSO4 was tested. In general, the Keggin het-
eropolyacids and their salts are strong acids,[36] and the rela-
tively higher acidic strength of the heteropolyacid anion
makes a bigger contribution to the acidities of these IL-POMs
ꢁ
than the HSO4 ion; therefore, [PyPS]3PW12O40 has a much
higher acidic strength than that of [PyPS]HSO4. In the catalytic-
performance comparison, 11.1% yield of ML was obtained at
88.9% cellulose conversion with [PyPS]HSO4 as the catalyst
(Table 3, Entry 6), whereas all of the PW12O403ꢁ-coupled IL-POMs
exhibited complete cellulose conversions with moderate-to-
high ML yields. Thus, on the basis of the aforementioned re-
sults, it is hypothesized that intensive synergistic effects be-
tween the sulfonic-functionalized cation and the heteropolyan-
ion of [PyPS]3PW12O40 are responsible for the high ML yield.
The effect of the SO3H functional group in the cation of the
pyridine-based IL-POMs on cellulose conversion and ML yield
was also studied. Clearly, more ML production can be achieved
with an increasing number of coordinated [PyPS]+ groups. For
example, the ML yield increases from 48.4 to 53.4 and 71.4%
as the number of coordinated [PyPS]+ groups increases from
one to two and three. As shown in Table 1 (Entries 2–4), the
substitution of the H+ ion in the heteropolyacid by the [PyPS]+
&
ChemSusChem 2016, 9, 1 – 11
www.chemsuschem.org
4
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÝÝ These are not the final page numbers!

Full Papers
Furthermore, this IL-POM with an appropriate acidic strength
accelerates cellulose hydrolysis, LA generation, and finally
esterification to realize a coupled and intensified process. The
rapid in situ esterification of LA not only slows the depolymeri-
zation equilibrium of cellulose but also protects ML from side
reactions owing to the immiscibility of [PyPS]3PW12O40 with
apolar esters;[30a] therefore, it is of benefit for enhanced ML se-
lectivity. Furthermore, [PyPS]3PW12O40 shows significantly en-
hanced catalytic activity relative to that of the conventional
acid H2SO4 (Table 1, Entry 14). This suggests that the present IL
catalysts have great potential as promising alternatives to con-
ventional acids in ML production owing to their high activities
and reduced corrosiveness.
Entry 1). However, the conversion of cellulose increased signifi-
cantly to 100%, and the ML yield increased to 30.3% with
0.3 mmol of [PyPS]3PW12O40 catalyst (Figure 2). As illustrated in
Figure 2, the yield of ML peaks at 71.4% if the catalyst loading
is 0.5 mmol. Further increments in catalyst quantity resulted in
the formation of dark brown floccule polymers and reduced
ML yields. This can be ascribed to the idea that excess catalytic
sites simultaneously accelerate cellulose conversion to ML and
facilitate undesirable side reactions.
Effects of reaction temperature and time
The effects of reaction temperature and time on the conver-
sion of cellulose to ML are depicted in Figure 3. This ML pro-
duction process is highly temperature-dependent. At 1308C,
the cellulose conversion increases gradually from 39.0 to 100%
from 3 to 6 h reaction time. However, at elevated tempera-
tures, 3 h is sufficient for complete cellulose conversion (Fig-
ure 3a). Glucose is the dominant product at 1308C (Figure 3b),
whereas ML is the major product at temperatures above this
(Figure 3d). Generally, the thermal effect is helpful for the deg-
radation of the b-1,4-glycoside bonds in cellulose[37] and accel-
erates the reaction but is not advantageous for the exothermic
esterification reaction. The use of excess alcohol not only leads
to the in situ esterification of LA but also pushes the esterifica-
tion equilibrium towards the product side and effectively mini-
mizes the LA yield. At 140 and 1508C, the LA yields increase in-
itially and then decrease owing to the consumption of LA. In
contrast, higher temperatures (160 and 1708C) result in small
increases to the LA yields as the time increases, whereas the
ML yields increase and decrease accordingly (Figure 3c and d).
The results indicate that moderate temperatures are appropri-
ate for the hydrolysis of cellulose and simultaneous in situ
esterification; therefore, more complete conversion of cellulose
and higher ML yields can be achieved at optimized reaction
times. Excessively high reaction temperatures and times result
in severe undesirable side reactions, such as repolymerization
and the formation of humins,[38] and, therefore, decrease the
ML yield. For example, at 1508C, the ML yield increases gradu-
ally from 33.0 to 71.4% as the reaction time increases from 3
to 5 h and then decreases to 54.9% at 7 h. Thus, an appropri-
ate temperature (1508C) and time (5 h) are required to achieve
excellent process efficiency.
The effects of the reaction medium and promoter were also
investigated. As shown in Table 3, if the cellulose depolymeri-
zation is conducted in the nonpolar solvent n-hexane, both
the cellulose conversion and the LA yield are negligible
(Table 3, Entry 15). The presence of water is beneficial for cellu-
lose hydrolysis, and 71.4% cellulose conversion and 18.1% LA
yield could be attained (Table 3, Entry 16). However, if alcohols
are used, the cellulose is converted completely at process effi-
ciencies in the range 30.9 to 79.7% (Table 3, Entries 17–19). In
these systems, the in situ esterification of the generated LA
with the alcohol promoter can be catalyzed by the acid cata-
lyst, and this shifts the equilibrium of LA production through
reaction coupling and, thus, enhances the process. This intensi-
fication effect of the in situ esterification can be further con-
firmed by cellulose depolymerization with various alcohols. For
example, 71.4% ML yield was obtained in a methanol medium,
whereas 2-propanol presents a relatively low isopropyl levuli-
nate yield of 21.5% owing to steric hindrance from the
branched alkyl chain.
Effect of catalyst loading
The effect of the catalyst loading on the ML generation is illus-
trated in Figure 2. The cellulose conversion was only 3.6% and
no ML was detected in the control experiment (Table 3,
Conversion of carbohydrates over [PyPS]3PW12O40 catalyst
The catalytic activity of the optimum IL-POM [PyPS]3PW12O40
for ML production was studied further with various carbohy-
drate substrates and real biomass (Table 4 and Figure S9). The
nature of the biomass has a substantial effect on the conver-
sion and ML yield. Mono- and polysaccharides can be convert-
ed completely with suitable ML yields (51.4–76.1%) and pro-
cess efficiencies (60.8–82.7%) under the optimized reaction
conditions (1508C for 4–5 h) with very low glucose yields
(Table 4 and Figure S9). The raw biomasses corn straw and
bagasse are partly converted at 1708C for 4–4.5 h with ML
yields of approximately 50%.
Figure 2. Effect of catalyst loading on ML production. Conditions: microcrys-
talline cellulose (0.162 g), [PyPS]3PW12O40, methanol (20 mL), 2 MPa N2,
1508C, 5 h.
ChemSusChem 2016, 9, 1 – 11
www.chemsuschem.org
5
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
These are not the final page numbers! ÞÞ

Full Papers
Figure 3. Effects of temperature and time on (a) cellulose conversion, (b) glucose yield, (c) LA yield, and (d) ML yield. Conditions: microcrystalline cellulose
(0.162 g), [PyPS]3PW12O40 (0.5 mmol), methanol (20 mL), 2 MPa N2.
the hydrolysis process, and this results in the promo-
tion of undesired side reactions[16,40] and, thereby,
Table 4. Conversion of various carbohydrates over [PyPS]3PW12O40 catalyst.[a]
lower ML yields. If sucrose is used as the substrate,
the hydrolysis of sucrose leads to the generation of
fructose, which undergoes dehydration readily to
yield ML, whereas glucose isomerizes before dehy-
dration.[41]
Substrate
Time [h]
Conversion [%]
YML [%]
YLA [%]
Efficiency [%]
glucose[b]
sucrose[b]
cellobiose[b]
starch
cellulose
corn straw[c]
bagasse[c]
4
100
100
100
100
58.7ꢂ1.0
76.1ꢂ1.1
58.6ꢂ1.4
51.4ꢂ1.3
71.4ꢂ1.7
50.3ꢂ1.1
50.0ꢂ1.3
6.9ꢂ0.8
6.6ꢂ0.5
6.5ꢂ0.4
9.4ꢂ1.1
8.3ꢂ0.5
7.8ꢂ0.5
8.9ꢂ0.6
65.6ꢂ0.9
82.7ꢂ1.3
65.1ꢂ1.5
60.8ꢂ1.3
79.7ꢂ1.9
58.1ꢂ1.1
58.9ꢂ1.1
4.5
4.5
5
5
4.5
4
100
Summarily, with the optimum [PyPS]3PW12O40 load-
ing of 0.5 mmol, cellulose can be degraded com-
pletely with 71.4% ML yield and 79.7% process effi-
ciency at 1508C for 5 h. This result shows the clear
advantages of the IL-POM catalysts over the current
technologies (Table S2), that is, the use of milder con-
ditions (lower temperature), higher product yield,
and less catalyst corrosion.
45.9ꢂ2.9
50.2ꢂ1.8
Conditions: [a] Cellulose or starch (0.162 g, containing 1 mmol C6H6O5 unit),
[PyPS]3PW12O40 (0.5 mmol), methanol (20 mL), 2 MPa N2, 1508C. Efficiency [%]=YML
[%]+YLA [%]. [b] 1 mmol substrate. [c] 0.5 g substrate, 1708C. The ML yield was calcu-
lated from the cellulose content.
Hemicellulose is converted to furfural and its derivatives,
whereas the conversion of lignin is negligible. The three main
Reusability of [PyPS]3PW12O40 catalyst
components of real biomass (cellulose, hemicellulose, and
lignin) are tightly bound through a mixture of hydrogen bond-
ing and covalent linkages, which provide structural rigidity to
the lignocellulosic matrix.[18,39] Thus, the lower ML yields ob-
served for bagasse and corn straw are attributed to the hin-
dered cellulose hydrolysis that occurs because of the recalci-
trant structure of real biomass. Interestingly, the ML yields for
starch, cellobiose, and glucose were lower than that for cellu-
lose. This may be attributed to the significantly higher solubili-
ties of starch, cellobiose, and glucose relative to that of cellu-
lose during the reaction; the increased solubility can accelerate
Under the optimized conditions, the reusability of the IL-POM
catalyst [PyPS]3PW12O40 was investigated, and the results are
presented in Figure 4. The procedure could be repeated up to
ten times without any clear loss of cellulose conversion. How-
ever, the ML yield decreased from 71.1 to 57.6% after the 10th
run. Inductively coupled plasma optical emission spectroscopy
(ICP-OES) analysis detected no tungsten leaching into the
chloroform phase (Table S1, Entry 11); therefore, [PyPS]3PW12O40
could not be extracted in this process. Nevertheless, there is
approximately 20% weight loss of the catalyst after the 10th
&
ChemSusChem 2016, 9, 1 – 11
www.chemsuschem.org
6
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÝÝ These are not the final page numbers!

Full Papers
1
at d=1.83 and 3.70 ppm in the H NMR spectrum (Figure 6b)
and at d=25.04 and 67.87 ppm in the 13C NMR spectrum (Fig-
ure 7b). This suggests the formation of the furan ring structure
of humins,[42] a main byproduct in carbohydrate conversion.
The ESI-MS spectra of the fresh and spent catalysts present
similar m/z values (Figure 8). The elemental analyses show
a slight increase in the carbon content (from 8.15 to 9.50%)
for the spent catalyst after the 10th run in comparison with
that of the fresh [PyPS]3PW12O40 (Table S3), and this confirms
further the existence of coating of humins on the IL catalyst.
Additionally, the W content, determined by ICP-OES, demon-
strates that the purity of the spent catalyst is 95.1% (Table S4).
Therefore, we can conclude that the IL-POM [PyPS]3PW12O40 is
highly reusable. The slight loss of activity for ML production is
caused by mechanical loss during the IL-POM separation and
recrystallization. On the other hand, the formation of humins,
a nonvolatile and conventional byproduct from carbohydrate
conversion, also contributes to the loss of catalytic activity.
Figure 4. The recyclability of [PyPS]3PW12O40. Conditions: microcrystalline cel-
lulose (0.162 g), [PyPS]3PW12O40 (0.5 mmol), methanol (20 mL), 2 MPa N2,
1508C, 5 h.
Conclusions
run. Therefore, 20 wt% of the fresh IL-POM [PyPS]3PW12O40 was
supplemented in the eleventh run. Surprisingly, a ML yield of
70.1% was obtained; this suggests that mechanical loss during
the separation of the product rather than catalyst deactivation
is one of the reasons for the reduction of process efficiency.
To determine the structural changes between the fresh and
spent catalyst after the 10th run, a series of comparative char-
acterizations were performed through FTIR spectroscopy, 1H
and 13C NMR spectroscopy, ESI-MS, elemental analysis, and ICP-
OES. As shown in Figure 5, the difference between the FTIR
A new ecofriendly and intensified strategy for the efficient con-
version of carbohydrates to levulinate esters is achieved
through a one-pot process over heteropolyanion ionic liquids
(IL-POMs). Under mild conditions (1508C for 5 h), a cellulose
conversion of 100% yields 71.4% methyl levulinate (ML) at
a process efficiency of 79.7%; therefore, this process shows
clear advantages over current technologies. This high efficiency
is due to the excellent catalytic activity of [PyPS]3PW12O40
[PyPS=1-(3-sulfopropyl)pyridinium] for the simultaneous cellu-
lose depolymerization and in situ esterification of the generat-
ed levulinic acid (LA) with methanol. Furthermore, this process
is efficient for the conversion of other typical carbohydrates
and the production of levulinate esters. Moreover,
[PyPS]3PW12O40 can be recovered readily by self-separation
through temperature control. In addition, the IL-POM catalyst
shows excellent reusability, and a satisfactory ML yield can be
obtained even after the 10th run. Therefore, the presented
technique will be a good reference for future biomass valoriza-
tion owing to its high efficiency, simple technology, mild con-
ditions, and simple catalyst recycling and reusability.
Experimental Section
Materials
Figure 5. FTIR spectra of (a) fresh and (b) spent [PyPS]3PW12O40 catalyst after
the 10th run.
Microcrystalline cellulose (Avicel PH-101, particle size ꢀ50 mm),
soluble starch, and d-(+)-glucose were purchased from Sigma–
Aldrich. The corn straw and bagasse were collected from Liaoning
and Guangdong provinces, China, respectively, and then air-dried
and ground to 40–50 mesh. The cellulose contents (w/w) of the
corn straw and bagasse were 44.5 and 34.7%, respectively, as de-
termined by the National Renewable Energy Laboratory (NREL) an-
alytical method. d-(+)-Cellobiose, sucrose, 1,3-propanesultone, imi-
dazole, 1-methylimidazole, 1-benzylimidazole, pyridine, trimethyla-
mine, methanol (ACS/HPLC reagent), n-propanol, 2-propanol, THF,
LA, ML, EL, and D2O (99.8 at% D) were purchased from J&K Chemi-
cal Ltd. Phosphotungstic acid hydrate (H3PW12O40·xH2O), 1-ethylimi-
spectra of the fresh and spent IL-POM [PyPS]3PW12O40 is insig-
nificant, and the characteristic absorptions of the pyridinium
cation (n˜ =1633, 1488, 1124, and 1181 cmꢁ1) and the Keggin
heteropolyacid anion (n˜ =1080, 978, 896, and 804 cmꢁ1) can
be observed clearly in the spectrum of spent [PyPS]3PW12O40.
1
The H and 13C NMR spectra of the spent IL-POM also exhibit
similar chemical shifts to those of the fresh one (Figure 6 and
7). However, it should be noted that extra signals are apparent
ChemSusChem 2016, 9, 1 – 11
www.chemsuschem.org
7
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
These are not the final page numbers! ÞÞ

Full Papers
Figure 6. 1H NMR of spectra (a) fresh and (b) spent [PyPS]3PW12O40 catalyst after the 10th run.
Figure 7. 13C NMR spectra of (a) fresh and (b) spent [PyPS]3PW12O40 catalyst after the 10th run.
dazole, n-octanol (standard for GC), potassium bromide, and 4-ni-
troaniline were purchased from Aladdin. 1-Isopropylimidazole was
purchased from Aamas-beta. Sulfuric acid (98%), diethyl ether, eth-
anol, toluene, and chloroform were supplied by Guanghua Chemi-
cal Factory Co., Ltd. All chemicals were of analytical-grade and
were used without further purification.
tection of a nitrogen atmosphere. The obtained white precipitate,
1-(3-sulfopropyl)pyridinium (PyPS), was collected by filtration,
washed with diethyl ether (3ꢁ100 mL), and finally dried under
vacuum overnight. PyPS (0.06 mol) was added to an aqueous solu-
tion containing H3PW12O40 (0.02 mol). The mixture was stirred vigo-
rously at room temperature for 24 h and then dried under vacuum
to give the final product [PyPS]3PW12O40 as a light yellow powder.
The other IL catalysts [PyPS]H2PW12O40, [PyPS]2HPW12O40,
Preparation of IL-POM catalysts
[Py]3PW12O40, [PyPS]HSO4,
[EIMPS]3PW12O40, [i-PIMPS]3PW12O40,
[TEAPS]3PW12O40 were prepared accordingly.
[IMPS]3PW12O40,
[MIMPS]3PW12O40,
and
The general preparation procedure for the IL-POM catalyst[30]
(Scheme 1) was as follows. For [PyPS]3PW12O40, pyridine (0.11 mol)
and 1,3-propanesultone (0.10 mol) were dissolved in toluene, and
the mixture was stirred vigorously at 508C for 24 h under the pro-
[BnIMPS]3PW12O40,
&
ChemSusChem 2016, 9, 1 – 11
www.chemsuschem.org
8
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÝÝ These are not the final page numbers!

Full Papers
Typical procedure for levulinate ester production
In a typical process for the conversion of cellulose to ML, micro-
crystalline cellulose (0.162 g), methanol (20 mL), and a given
amount of the IL-POM catalyst were mixed in a 50 mL stainless-
steel autoclave. The reactor was first purged with N2 three times to
displace the air and then heated to the designated temperature
and stirred with a magnetic stirrer (300 rpm). After the desired
time, the mixture was cooled to room temperature in an ice–water
bath.
Product separation
The whole procedure for product separation is shown in
Scheme S1. The solid residue was first collected by filtration,
washed with H2O and THF (five times each), and then dried under
vacuum overnight to enable the calculation of the conversion. De-
ionized water (25 mL) was added to the filtrate, and then chloro-
form (5ꢁ5 mL) was used for the product extraction. After careful
separation, an organic phase (containing levulinate ester) and an
aqueous phase (containing the water-soluble fraction and the IL-
POM catalyst) were obtained. The organic phase was diluted with
chloroform to 50 mL for qualitative and quantitative analysis. The
aqueous phase was diluted with deionized water to 100 mL for the
measurement of the yields of the water-soluble products such as
LA and glucose. After the removal of water at 508C under reduced
pressure, the obtained fraction was diluted with methanol (50 mL)
to dissolve the IL-POM catalyst. The solvent was first removed from
the methanol solution with a rotary evaporator until a light yellow
solid appeared. Then, the suspension was placed in a refrigerator
(ꢁ108C) to precipitate the IL-POM catalyst and to separate the re-
sidual methanol-soluble organic chemicals. The recycled and re-
crystallized catalyst was obtained by centrifugation and then dried
at 708C under vacuum overnight. The catalyst was used directly
for the next run without further purification.
Figure 8. Negative-ion-mode ESI-MS spectra of (a) fresh and (b) spent
[PyPS]3PW12O40 catalyst after the 10th run.
Characterization of the ILs
1
The H and 13C NMR spectra were recorded with a Bruker AV-400
spectrometer with D2O as the solvent. The ESI-MS spectra were ob-
tained with an Agilent 1290/Bruker maxis impact instrument. The
FTIR spectra (KBr pellets) were recorded with a Bruker Equinox-55
FTIR instrument. The C, H, S, and N contents were measured with
a Vario EL III elemental analyzer. The thermal stabilities of the ILs
were studied with a NETZSCH TG 209 F3 analyzer from 30 to
8008C at the rate of 108Cminꢁ1 under a nitrogen atmosphere. The
detail catalyst characterization data can be found in the Support-
ing Information.
IL acidic strengths and their solubilities in methanol
The Brçnsted acidic strengths of the as-synthesized ILs were deter-
mined by the Hammett method[31] and measurement of the pH
values. The Amax values for the Hammett functions were obtained
with a UV/Vis spectrometer (UV-3010) in the range l=200 to
800 nm. 4-Nitroaniline was used as the indicator, and the Hammett
acidity function (Ho) was calculated with Equation (1). The pH
values of the IL catalysts were measured with a Precision PHS-25
pH meter.
Qualitative and quantitative analysis of the products
The volatile products in the chloroform (organic phase) were iden-
tified by GC–MS and quantified by GC with an Agilent 7890 GC in-
strument with both a 5977A MS detector and a flame ionization
detector (FID). An Agilent HP-5 MS [(5% phenyl)methylpolysilox-
ane] capillary column (30 mꢁ250 mmꢁ0.25 mm) was used for
chemical separation. The FID was operated at 2708C, and the carri-
er gas was helium at a flow rate of 1.0 mLminꢁ1. The initial oven
temperature was 508C (held for 3 min) and then it was ramped to
2508C (held for another 1 min) at 158Cminꢁ1. The chemicals in the
aqueous phase were analyzed with an Agilent 1200 HPLC instru-
ment equipped with an HPX-87H column (300 mmꢁ7.8 mm, 5 mm)
and a refractive index detector (RID). A solution of 5 mm H2SO4
was used as the mobile phase at a flow rate of 0.6 mLminꢁ1. The
column oven temperature was 658C. Commercial chemicals were
used as the standard compounds for the qualitative and quantita-
tive analyses.
Ho ¼ pKaþlog ½Iꢃ=½IHþꢃ
ð1Þ
In Equation (1), I represents the indicator base, and [IH+] and [I] are
the molar concentrations of the protonated and unprotonated
forms of the indicator, respectively. The pK(I)a value of
4-nitroaniline is 0.99.
The solubilities of the IL-POMs in methanol were obtained by two
methods: (1) The solubilities of the IL-POMs were calculated from
the W content in their saturated methanol solutions, as measured
by ICP-OES analysis with a NexION 300 analyzer. (2) A standard
curve-based method was employed with the intensity of the ab-
sorption band at lꢀ260 nm in UV/Vis spectra, and H3PW12O40 was
employed as the standard compound for quantitative analysis. Sa-
turated solutions of the ILs were diluted 1000 times before analy-
sis.
The conversion of carbohydrate was defined by Equation (2).
mc ꢁ ms
Conversion ½%ꢃ ¼
ꢄ100 %
ð2Þ
mc
mc is the mass of feed carbohydrate, and ms is the mass of the un-
reacted feedstock collected by filtration, washing, and drying.
ChemSusChem 2016, 9, 1 – 11
www.chemsuschem.org
9
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
These are not the final page numbers! ÞÞ

Full Papers
[15] D. Klemm, B. Philipp, T. Heinze, U. Heinze, W. Wagenknecht, Comprehen-
sive Cellulose Chemistry: Functionalisation of Cellulose, Wiley-VCH, Wein-
heim, 1998.
The yields of levulinate esters, LA, and glucose and the process ef-
ficiency were defined by Equations (3)–(6), respectively.
nLA ester
[16] I. van Zandvoort, Y. Wang, C. B. Rasrendra, E. R. van Eck, P. C. Bruijnincx,
H. J. Heeres, B. M. Weckhuysen, ChemSusChem 2013, 6, 1745–1758.
[17] J. P. Hallett, T. Welton, Chem. Rev. 2011, 111, 3508–3576.
[18] A. M. Da Lopes, R. Bogel-Łukasik, ChemSusChem 2015, 8, 947–965.
[19] A. S. Amarasekara, O. S. Owereh, Ind. Eng. Chem. Res. 2009, 48, 10152–
10155.
[20] Y. Liu, W. Xiao, S. Xia, P. Ma, Carbohydr. Polym. 2013, 92, 218–222.
[21] a) J. Long, B. Guo, X. Li, Y. Jiang, F. Wang, S. C. Tsang, L. Wang, K. M. K.
Yu, Green Chem. 2011, 13, 2334–2338; b) J. Long, G. Bin, J. Teng, Y. Yu,
L. Wang, X. Li, Bioresour. Technol. 2011, 102, 10114–10123.
[22] S. Siankevich, Z. Fei, R. Scopelliti, P. G. Jessop, J. Zhang, N. Yan, P. J.
Dyson, ChemSusChem 2016, 9, 2089–2096.
YieldLA ester ½%ꢃ ¼
YieldLA ½%ꢃ ¼
ꢄ100 %
ð3Þ
ð4Þ
nðC6H6O5 unitÞ
nLA
ꢄ100 %
nðC6H6O5 unitÞ
nglucose
Yieldglucose ½%ꢃ ¼
ꢄ100 %
ð5Þ
ð6Þ
nðC6H6O5 unitÞ
Process efficiency ½%ꢃ ¼YieldLA esterþYieldLA
nLA ester, nLA, and nglucose are the molar amounts of levulinate ester,
LA, and glucose, respectively, and n(C6H6O5 unit) is the molar
amount of C6H6O5 units in the initial feedstock.
[23] H. Ren, Y. Zhou, L. Liu, Bioresour. Technol. 2013, 129, 616–619.
[24] A. S. Amarasekara, B. Wiredu, Bioenergy Res. 2014, 7, 1237–1243.
[25] Y. Shen, J. Sun, Y. Yi, B. Wang, F. Xu, R. Sun, Bioresour. Technol. 2015,
192, 812–816.
[26] H. Ma, J. Long, F. Wang, L. Wang, X. Li, Acta Phys.-Chim. Sin. 2015, 31,
973–979.
Acknowledgements
[27] D. Song, S. An, Y. Sun, Y. Guo, J. Catal. 2016, 333, 184–199.
[28] B. Zhang, H. Asakura, J. Zhang, J. Zhang, S. De, N. Yan, Angew. Chem.
Int. Ed. 2016, 55, 8319–8323; Angew. Chem. 2016, 128, 8459–8463.
[29] J. Albert, P. Wasserscheid, Green Chem. 2015, 17, 5164–5171.
[30] a) Y. Leng, J. Wang, D. Zhu, X. Ren, H. Ge, L. Shen, Angew. Chem. Int. Ed.
2009, 48, 168–171; Angew. Chem. 2009, 121, 174–177; b) K. Li, L. Bai,
P. N. Amaniampong, X. Jia, J. M. Lee, Y. Yang, ChemSusChem 2014, 7,
2670–2677.
The authors gratefully acknowledge the financial support of the
Natural Science Foundation of China (Grant Nos. 21336002 and
21676108) and the Key Natural Science Foundation of Guang-
dong Province, China (No. 2015A030311048).
Keywords: biomass · carbohydrates · esters · ionic liquids ·
polyoxometalates
[31] C. Thomazeau, H. Olivier-Bourbigou, L. Magna, S. Luts, B. Gilbert, J. Am.
Chem. Soc. 2003, 125, 5264–5265.
[32] T. Rajkumar, G. R. Rao, Mater. Lett. 2008, 62, 4134–4136.
[33] F. Peral, E. Gallego, J. Mol. Struct. 1997, 415, 187–196.
[34] R. B. Grossman, The Art of Writing Reasonable Organic Reaction Mecha-
nisms, Springer, New York, 2003.
[35] F. A. Carey, R. J. Sundberg, Advanced Organic Chemistry: Part A: Structure
and Mechanisms, Springer, New York, 2007.
[1] a) G. W. Huber, S. Iborra, A. Corma, Chem. Rev. 2006, 106, 4044–4098;
b) J. N. Chheda, G. W. Huber, J. A. Dumesic, Angew. Chem. Int. Ed. 2007,
46, 7164–7183; Angew. Chem. 2007, 119, 7298–7318.
[2] S. Siankevich, Z. Fei, R. Scopelliti, G. Laurenczy, S. Katsyuba, N. Yan, P. J.
Dyson, ChemSusChem 2014, 7, 1647–1654.
[3] Z. Sun, L. Xue, S. Wang, X. Wang, J. Shi, Green Chem. 2016, 18, 742–
752.
[36] a) Y. Izumi, M. Ogawa, K. Urabe, Appl. Catal. A 1995, 132, 127–140; b) J.
Yang, M. J. Janik, D. Ma, A. Zheng, M. Zhang, M. Neurock, R. J. Davis, C.
Ye, F. Deng, J. Am. Chem. Soc. 2005, 127, 18274–18280.
[37] B. Girisuta, L. P. B. M. Janssen, H. J. Heeres, Ind. Eng. Chem. Res. 2007, 46,
1696–1708.
[4] Y. Wang, S. De, N. Yan, Chem. Commun. 2016, 52, 6210–6224.
[5] P. Gallezot, Chem. Soc. Rev. 2012, 41, 1538–1558.
[6] a) J. J. Bozell, G. R. Petersen, Green Chem. 2010, 12, 539–554; b) A. Dꢂ-
molis, N. Essayem, F. Rataboul, ACS Sustainable Chem. Eng. 2014, 2,
1338–1352.
[38] ꢃ. Szabolcs, M. Molnꢄr, G. Dibꢅ, L. T. Mika, Green Chem. 2013, 15, 439–
445.
[7] A. Corma, S. Iborra, A. Velty, Chem. Rev. 2007, 107, 2411–2502.
[8] X. Tang, Z. Li, X. Zeng, Y. Jiang, S. Liu, T. Lei, Y. Sun, L. Lin, ChemSusChem
2015, 8, 1601–1607.
[9] D. M. Alonso, S. G. Wettstein, J. A. Dumesic, Green Chem. 2013, 15, 584–
595.
[39] a) M. E. Himmel, S. Y. Ding, D. K. Johnson, W. S. Adney, M. R. Nimlos,
J. W. Brady, T. D. Foust, Science 2007, 315, 804–807; b) S. P. S. Chunda-
wat, B. S. Donohoe, L. da Costa Sousa, T. Elder, U. P. Agarwal, F. Lu, J.
Ralph, M. E. Himmel, V. Balan, B. E. Dale, Energy Environ. Sci. 2011, 4,
973–984.
[10] H. Joshi, B. R. Moser, J. Toler, W. F. Smith, T. Walker, Biomass Bioenergy
2011, 35, 3262–3266.
[11] A. Morone, M. Apte, R. A. Pandey, Renewable Sustainable Energy Rev.
2015, 51, 548–565.
[12] F. D. Pileidis, M. M. Titirici, ChemSusChem 2016, 9, 562–582.
[13] a) J. Hegner, K. C. Pereira, B. DeBoef, B. L. Lucht, Tetrahedron Lett. 2010,
51, 2356–2358; b) L. Peng, L. Lin, H. Li, Q. Yang, Appl. Energy 2011, 88,
4590–4596; c) S. Zhao, G. Xu, J. Chang, C. Chang, J. Bai, S. Fang, Z. Liu,
Bioresources 2015, 10, 2223–2234.
[40] S. J. Dee, A. T. Bell, ChemSusChem 2011, 4, 1166–1173.
[41] Y. Kuwahara, W. Kaburagi, K. Nemoto, T. Fujitani, Appl. Catal. A 2014,
476, 186–196.
[42] H. E. Gottlieb, V. Kotlyar, A. Nudelman, J. Org. Chem. 1997, 62, 7512–
7515.
Received: August 7, 2016
[14] A. Fukuoka, P. L. Dhepe, Angew. Chem. Int. Ed. 2006, 45, 5161–5163;
Angew. Chem. 2006, 118, 5285–5287.
Published online on && &&, 0000
&
ChemSusChem 2016, 9, 1 – 11
www.chemsuschem.org
10
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ÝÝ These are not the final page numbers!

FULL PAPERS
Iconic ionic: Heteropolyanion-based
ionic liquids are used as catalysts for the
conversion of renewable carbohydrates
to levulinate esters. The process can be
operated under mild conditions, and
the catalyst can be recovered readily by
self-separation through temperature
control. The catalyst shows excellent re-
usability, and a satisfactory yield can be
obtained even after the 10th run.
C. Song, S. Liu, X. Peng, J. Long,* W. Lou,
X. Li*
&& – &&
Catalytic Conversion of Carbohydrates
to Levulinate Ester over
Heteropolyanion-Based Ionic Liquids
ChemSusChem 2016, 9, 1 – 11
www.chemsuschem.org
11
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
&
These are not the final page numbers! ÞÞ