Microwave-assisted conversion of carbohydrates to levulinic acid: An essential step in biomass conversion

Article abstract of DOI:10.1039/c2gc36682g

Degradation of non-edible carbohydrates to levulinic acid (4-oxopentanoic acid) was studied by using dielectric heating with microwave energy. Levulinic acid and its reduced and dehydrated derivative, γ-valerolactone (GVL), can be used for the production of small-molecule, functionalized hydrocarbons, which might be potential platform molecules for the chemical industry. First, simple model compounds (fructose, glucose, saccharose and cellobiose) were hydrolyzed in order to find the optimum reaction conditions (e.g. reagent, reaction temperature, acid concentration, time) for the degradation and transformation of polysaccharides (cellulose, chitin, chitosan) by using controlled microwave irradiation. Cellulose, a non-edible biopolymer of plant origin, was successfully converted to levulinic acid under the optimized conditions (2 M H₂SO₄, 170 °C, 50 min) with a yield of 34.2% in a mono-mode Multisynth microwave reactor. The reactions proceeded with hydrochloric acid catalysis as well, and a slightly better yield was achieved, however, using HCl (a chlorine containing catalyst) raises serious environmental concerns. The hydrolysis of glucosamine-based glycans (d-glucosamine, N-Ac-d-glucosamine, LMw-chitosan, MMw-chitosan, chitin) was also studied and optimized with sulfuric acid as a catalyst in a mono-mode Multisynth microwave reactor. The highest yield of levulinic acid was obtained with 2 M H ₂SO₄ at 190 °C for 30 min. N-Ac-d-glucosamine, d-glucosamine, LMw-chitosan and MMw-chitosan resulted in levulinic acid with yields between 20.6% and 32.7%, the larger molecular weight chitin was degraded to levulinic acid with a yield of 37.8%.

Full text of DOI:10.1039/c2gc36682g

Green Chemistry
View Article Online
View Journal | View Issue
PAPER
Microwave-assisted conversion of carbohydrates to
levulinic acid: an essential step in biomass conversion†
Cite this: Green Chem., 2013, 15, 439
Ármin Szabolcs, Márk Molnár, Gábor Dibó* and László T. Mika*
Degradation of non-edible carbohydrates to levulinic acid (4-oxopentanoic acid) was studied by using
dielectric heating with microwave energy. Levulinic acid and its reduced and dehydrated derivative,
γ-valerolactone (GVL), can be used for the production of small-molecule, functionalized hydrocarbons,
which might be potential platform molecules for the chemical industry. First, simple model compounds
(fructose, glucose, saccharose and cellobiose) were hydrolyzed in order to find the optimum reaction
conditions (e.g. reagent, reaction temperature, acid concentration, time) for the degradation and trans-
formation of polysaccharides (cellulose, chitin, chitosan) by using controlled microwave irradiation. Cellu-
lose, a non-edible biopolymer of plant origin, was successfully converted to levulinic acid under the
optimized conditions (2 M H2SO4, 170 °C, 50 min) with a yield of 34.2% in a mono-mode Multisynth
microwave reactor. The reactions proceeded with hydrochloric acid catalysis as well, and a slightly better
yield was achieved, however, using HCl (a chlorine containing catalyst) raises serious environmental con-
cerns. The hydrolysis of glucosamine-based glycans (D-glucosamine, N-Ac-D-glucosamine, LMw-chitosan,
MMw-chitosan, chitin) was also studied and optimized with sulfuric acid as a catalyst in a mono-mode
Multisynth microwave reactor. The highest yield of levulinic acid was obtained with 2 M H2SO4 at 190 °C
for 30 min. N-Ac-D-glucosamine, D-glucosamine, LMw-chitosan and MMw-chitosan resulted in levulinic
acid with yields between 20.6% and 32.7%, the larger molecular weight chitin was degraded to levulinic
acid with a yield of 37.8%.
Received 2nd August 2012,
Accepted 31st October 2012
DOI: 10.1039/c2gc36682g
www.rsc.org/greenchem
chitin. Based on the biorefinery concept, sugars and starches
have already been successfully used.6 However, the selection
Introduction
The global chemical industry is almost exclusively based on and consumption of appropriate resources have become a con-
non-renewable resources, i.e. fossil fuels, which also provide troversial issue due to the dramatically increased utilization of
90% of all our energy needs. While it is very difficult to predict edible resources. In addition, the selective conversion of non-
the exact date of depletion of crude oil and natural gas, the edible biomass components, such as cellulose, chitin, etc.,
development of alternative strategies to supply sustainable into platform molecules plays a key role in sustainable develop-
fuels and carbon based feedstocks should be accelerated.1,2 ment.7 Consequently, the simple conversion of biomass
The global efforts to reduce the carbon dioxide emission also into carbon based chemicals6,8,9 and fuels10 has become the
demand new and innovative strategies for the green pro- focus of interest. It was established that saccharose can be con-
duction of fuels,3 platform molecules and value-added chemi- verted to various C5-oxygenates via levulinic acid (LA),11 which
cals.4 Hydrogenation of carbon dioxide could be one of the is one of the most important intermediates of the proposed
approaches to produce carbon-based building blocks and biomass based carbon cycle.12 Due to its reactive functional
intermediates.5 In addition, nature can help to convert carbon groups, levulinic acid has been identified as a valuable bio-
dioxide to biomass, one of the most preferred renewable based multipurpose building block of functionalized C3–C6
resources.
oxygenates (Fig. 1).13 One of the most important derivatives is
Carbohydrates, as the main part of biomass, are present in γ-valerolactone (GVL),12 which could be considered as a sus-
the form of sugars, starches, cellulose, lignocellulose, and tainable liquid and can be produced by catalytic hydrogen-
ation of levulinic acid.14 GVL has been successfully used for
the production of ionic liquids,15 pentane-1,4-diol,11,14b
butene isomers,18a alkanes,11 adipic acid,16 polymers,17 trans-
Eötvös Loránd University, Institute of Chemistry, Pázmány Péter sétány 1/A,
Budapest, Hungary. E-mail: ltmika@elte.hu; Fax: +36 30 372 2592;
portation fuels,18 etc.
Tel: +36 30 372 2500
The synthesis of levulinic acid from saccharose in the pres-
ence of HCl as a catalyst was first reported by Mulder.19 This
†Electronic supplementary information (ESI)
available. See DOI:
10.1039/c2gc36682g
This journal is © The Royal Society of Chemistry 2013
Green Chem., 2013, 15, 439–445 | 439

View Article Online
Paper
Green Chemistry
concentration, reaction time) on the degradation of different
carbohydrates to levulinic acid using microwave technology.
Results and discussion
The catalytic dehydration of non-edible carbohydrates to levuli-
nic and formic acid is a particularly attractive approach for
carbohydrate-based biomass conversion. Since formic acid can
be catalytically decomposed to carbon dioxide and hydrogen
by using palladium, silver, copper,26 or even more efficiently
in the presence of tris[(2-diphenylphosphino)ethyl]phosphine
[P(CH2CH2PPh2)3] modified iron catalyst,27 the latter could be
used for efficient GVL production.14a Alternatively, transfer
hydrogenation of levulinic acid with formic acid, as a hydrogen
source, in the presence of the Shvo-catalyst results in the for-
mation of 4-hydroxyvaleric acid, which spontaneously turns to
GVL via ring closure dehydration.28 Horváth et al.11 found that
levulinic acid was obtained by dehydrating saccharose under
acidic conditions (1.8 M sulfuric acid) at 140 °C for 8 h with a
yield of 35%. Additionally, under the same conditions using a
green and highly acidic catalyst (Nafion-NR-50), a similar yield
was obtained, however, much longer time (40 h) was necess-
ary.11 Since the seminal papers of Gedye29 and Giguere,30
microwave dielectric heating has become a popular tool in
common organic laboratories. It has been shown that con-
trolled microwave heating dramatically decreases reaction
time, increases product yield, and enhances product purity by
reducing unwanted side reactions.31
Fig. 1 Levulinic acid based platform molecules.
Scheme 1
transformation involves hydrolysis of the glycosidic bond and
two consecutive dehydration steps in the formation of equi-
molar amounts of levulinic acid and formic acid (Scheme 1).
Recently, several papers were published on the conventional
acid catalyzed conversion of different carbohydrates to levuli-
nic acid.9,20
In this work, we are focusing on the application of this
modern technology in order to find milder and more efficient
methods for the hydrolysis and transformation of glycans into
small organic platform molecules.
Beyond cellulose, chitin is the second most abundant bio-
polymer on the Earth. It is an industrial waste of the food
industry, it is obtained from crustaceans such as shrimps,
crabs, etc. Similarly to cellulose, chitin could be a sustainable
green feedstock for production of valuable platform molecules.
Recently, a few papers have been published on the microwave-
assisted degradation of polysaccharides.21 However, the
majority of these works was performed by using carbohydrates
of plant origin.22,23 A most recent study by Kerton et al. reports
the degradation of chitosan (a partially deacetylated form of
the natural chitin) to levulinic acid by using microwave
irradiation under aqueous conditions.24 Although different
Lewis-acids including lanthanide-trifluoromethanesulfonates
were found to be active catalysts, SnCl4·5H2O showed the
highest activity for the transformation of chitosan into LA with
an average yield of 24 wt%. It is noteworthy that selective
formation of levulinic acid and 5-hydroxymethyl-2-furfural
(5-HMF) can be influenced by the concentration of the catalyst
as well. Kerton et al. also reported that the microwave-assisted
decomposition of N-Ac-D-glucosamine under non-aqueous con-
ditions results in the formation of 3-acetamido-5-acetylfuran,
which could be a nitrogen containing platform molecule.25
We report here a detailed investigation of the effect of
different reaction parameters (e.g. reagent, temperature, acid
We started systematically to optimize the reaction con-
ditions. Firstly, the effect of temperature on the microwave-
assisted degradation of saccharose to levulinic acid was exam-
ined. By varying the temperature from the initial 100 °C to
140 °C for 10 min, the yield of levulinic acid increased from
2.1% to 34%. To compare the effect of different heating
methods on the degradation of carbohydrates, parallel
samples were treated as follows: 400 mg (2.22 mmol) of D-fruc-
tose was dissolved in 8 mL 2 M H2SO4 and heated in the
microwave oven at 170 °C. The yield of levulinic acid was
found to be 42.7% using microwave irradiation for 30 min.
Compared with traditional heating, the same sample was
treated in a conventional Schlenk tube in an oil bath at the
same temperature for 8 h, with a 31.3% yield. It is worthy of
note that a similar yield was obtained with the dehydration of
saccharose (35%, 8 h).11 As expected, if dielectric and conven-
tional heating are compared the reaction time was significantly
shorter and the yield was always higher when microwave
heating was applied. The monosaccharide, D-glucose (400 mg,
2.22 mmol), was also transformed under identical conditions.
In this case, the yield of LA was moderate (15%) under conven-
tional heating and 2.7 times higher (40.5%) when dielectric
heating was applied, moreover, the reaction time was much
440 | Green Chem., 2013, 15, 439–445
This journal is © The Royal Society of Chemistry 2013

View Article Online
Green Chemistry
Paper
Fig. 2 Influence of sulfuric acid concentration on the yield of interconversion
of saccharose (400 mg) to levulinic acid (in 8 mL solution; t = 10 min; T =
Fig. 3 Influence of reaction temperature on the yield of interconversion of cel-
lobiose (400 mg) to levulinic acid (in 8 mL 2 M sulfuric acid; t = 10 min); yield
140 °C); yield (%) = 100% × nlevulinic acid/nsaccharose
.
(%) = 100% × nlevulinic acid/ncellobiose
.
the conclusion of recent studies.32,33 It is well known that the
formation of 5-HMF from glucose proceeds via isomerization
to fructose. The five fructose isomers (D-fructoketose, D-fructo-
pyranoses, D-fructofuranoses) are present in equilibrium. In an
acidic environment, both the furanoses and the pyranoses,
through the appropriate oxocarbenium intermediate, lose two
water molecules. However, the products are completely
different, furanoses form exclusively 5-HMF and pyranoses via
an unknown mechanism are converted to humins.32 It was
also proposed that by increasing temperature, the equilibria
are shifted to the protonated pyranose form, the precursor of
humins. Hydrolysis of cellobiose provides two D-glucose mole-
cules, which must be firstly converted to D-fructose, which is
transformed into 5-HMF, then into levulinic acid. On the other
hand, hydrolysis of saccharose gives one molecule of D-glucose
and D-fructose each, the fructose is directly converted to
5-HMF, then to levulinic acid under an acidic environment. It
was verified by the 13C-NMR analysis of the neat aqueous dehy-
dration mixture of 400 mg (2.22 mmol) D-fructose (ESI
Fig. S1†). While the presence of 5-HMF cannot be detected, the
equimolar formation of formic and levulinic acid (13C-NMR, δ,
ppm: HCOOH: 166.2; CH3(CO)CH2CH2COOH: 27.9, 29.3, 37.8,
177.6, 213.5) was clearly indicated. Furthermore, the amount
of possible unreacted substrates or water-soluble by-products
was under the detection limit of our instrument. The calcu-
lated mass balance was found to be 89.5% (ESI Fig. S4†). It
should be noted that the filtered and dried black tar
(100.5 mg) cannot be dissolved either in organic solvents or
water. Based on the result of elementary microanalysis, this
solid material was pure carbon powder (>99%). The pyrolysis
analysis of the same sample resulted in <1% solid residue,
without the observation of any melting. The NMR analysis of
the EtOAc soluble unidentified brown residue (67.8 mg in ESI
Fig. S4†) indicated the formation of high carbon containing
“humins” (ESI Fig. S5 and S6†). It was proposed that
these oligo- and polymers might be formed by the acid
catalyzed oligomerization of the reaction intermediates con-
taining CvC double bonds.34
shorter (30 min). These results are in accord with the present
scientific opinion of the effect of microwave irradiation on the
chemical reactions in general.31b
Subsequently, the effect of the chemical composition and
concentration of the acid catalyst was studied. Although dehy-
dration of carbohydrates was successfully performed by using
various acids, mineral acids were the most efficient ones.6 The
conversion of carbohydrates to levulinic acid is highly affected
by the acid concentration as well. It was recently reported by
Horváth et al.32 that low acid concentration (10−5–10−1 M)
favors the formation of 5-hydroxymethyl-2-furaldehyde (5-HMF)
with a maximum in the case of 0.5 M acid catalyst con-
centration. It was also established that the acid concentration
has a significant effect on the selective formation of both
5-HMF and levulinic acid from chitosan in the presence of
a metal based catalyst.24 Subsequently, the influence of the
concentration of sulfuric acid on the formation of levulinic
acid was tested in the range of 0.1–2.5 M (Fig. 2). Working
below 1 M sulfuric acid concentration, the formation of levuli-
nic acid was slowly increasing (simultaneously, the amount of
5-HMF was decreasing) by increasing acid concentration.
Above 1 M sulfuric acid concentration, a higher yield of levuli-
nic acid was observed reaching a maximum at 2 M H2SO4.
Comparing the results obtained under conventional and
microwave conditions, the heating method practically has no
influence on the optimum acid concentration. Secondly, cello-
biose, which is a disaccharide (and considered as a dimeric
building block of cellulose), was used as a model substrate for
the optimization of reaction temperature using 2 M H2SO4 for
10 min. The amount of levulinic acid gradually became higher
by increasing reaction temperature up to 190 °C (the practical
upper limit of our instrument) without any maximum, as
expected (Fig. 3). Although a higher amount of levulinic acid
was obtained at higher temperature, over 170 °C a significantly
increasing amount of humin formation was observed. It was
found that to obtain a similar yield of levulinic acid, higher
temperature was necessary for the hydrolysis of cellobiose
than that of saccharose. This observation can be explained by
This journal is © The Royal Society of Chemistry 2013
Green Chem., 2013, 15, 439–445 | 441

View Article Online
Paper
Green Chemistry
Table 1 Acid catalyzed transformation of biomass carbohydrates into levulinic
acid
Yield of LA
Entry
Substrate
Additive
t [min]
[mol%]
wt%
1
2
3
4
5
6
7
8
D-Fructose
D-Fructose
D-Glucose
D-Glucose
Cellobiose
Cellobiose
Cellulose
Cellulose
HCl
H2SO4
HCl
H2SO4
HCl
H2SO4
H2SO4
HCl
H2SO4
HCl
H2SO4
HCl
H2SO4
HCl
H2SO4
HCl
30
30
30
30
30
30
30
50
50
10
10
10
10
20
20
20
20
30
30
49.4
42.7
48.6
40.5
44.0
41.3
21.4
46.0
34.2
36.4
25.6
22.4
20.6
31.9
19.3
37.0
32.1
32.7
37.8
31.8
27.5
31.4
26.1
29.9
28.0
15.4
31.0
23.0
19.6
13.8
11.8
10.8
22.7
13.7
26.3
22.8
18.7
21.6
9
Cellulose
10
11
12
13
14
15
16
17
18
19
D-Glucosamine
D-Glucosamine
N-Ac-D-glucosamine
N-Ac-D-glucosamine
LMw-chitosan
LMw-chitosan
MMw-chitosan
MMw-chitosan
Chitin
Fig. 4 Influence of reaction temperature on the yield of interconversion of cel-
lulose (400 mg) to levulinic acid (in 8 mL 2 M H2SO4; t = 30 min); yield (%) =
100% × nlevulinic acid/ncellulose
.
H2SO4
HCl
H2SO4
Chitin
Reaction temperature for carbohydrates of plant origin (entries 1–9):
170 °C, for animal origin carbohydrates (entries 10–19): 190 °C; (LMw:
low-molecular-weight, MMw: medium-molecular-weight, both 75–85%
deacetylated, Sigma-Aldrich); yield (mol%) = 100% × nlevulinic
/
acid
nsubstrate; yield (wt%) = 100% × mlevulinic acid/msubstrate
.
Initially, based on the results of our studies with model dis-
accharides, 2 M sulfuric acid as a reaction medium and 170 °C
as the reaction temperature were set for the hydrolysis of cellu-
lose, a high-molecular weight, insoluble polymer. After a
10 min microwave irradiation the yield of levulinic acid was
very low (3.7%). To optimise the yield, the reaction time was
increased stepwise. Although the yield increased by using
longer reaction time (Table 1, entries 8 and 9), the amount of
humins as by-products sharply increased. Thus, 30 min is con-
Fig. 5 Influence of sulfuric acid concentration on the yield of interconversion
of cellobiose (400 mg; t = 10 min) and cellulose (400 mg; t = 30 min) to
levulinic acid (T = 170 °C); yield (%) = 100% × nlevulinic acid/nsubstrate
.
sidered as the optimum reaction time which provided levulinic carbohydrates, the presence of chlorine is an environmental
acid with a moderate yield (Table 1, entry 7). Subsequently, the concern. Generally speaking, the difference between chlorine-
effect of temperature on the formation of levulinic acid was free (H2SO4) and chlorine-containing (HCl) catalyzed reactions
investigated from 130 °C to 180 °C. The optimum temperature was not significant (except for LMw-chitosan, Table 1, entries
was found to be 170 °C (Fig. 4). Above this value, the formation 10 and 11). In addition, hydrochloric acid can easily desorb
of humins became more significant, which can be explained from the reaction mixture to the atmosphere, while sulfuric
by the higher equilibrium concentration of the protonated pyr- acid cannot.
anose forms. In order to verify the optimum acid concen-
The interconversion pathway of carbohydrates of plant
tration, dehydration of cellulose and cellobiose was performed origin is depicted in Scheme 2. It was found that dehydration
in a concentration range of 0.1–2.5 M H2SO4. The optimum of fructofuranose (4a) gave the highest yield of levulinic acid
conditions for cellobiose were found to be 2 M H2SO4 at (42.7%, Table 1, entry 2). It is important to note that the trans-
170 °C for 10 min; and for cellulose 2 M H2SO4 at 170 °C for formation of fructose (4a–c) into 5-HMF (7) starts from proto-
30 min (Fig. 5).
nated furanose intermediate (6).32
The optimized reaction conditions (e.g. acid concentration,
Starting from D-glucopyranose (3a), which is in equilibrium
time, temperature, yield) for the production of levulinic acid with D-fructofuranose (4a) under the applied reaction con-
from carbohydrates of plant origin are summarized in Table 1, ditions,33 slightly lower conversion (40.5%, Table 1, entry 4)
entries 1–9. For a comparison, the reactions were performed was observed. Presumably, glucose (3a–c) to fructose (4a–c) iso-
by using 2 M H2SO4 and 2 M HCl as well under identical merization is fast enough to result in 6, which is a key inter-
conditions. Although aqueous hydrochloric acid was found to mediate of the whole process. Intermediate
6 either
be a slightly more efficient catalyst for the dehydration of irreversibly dehydrates to 5-HMF (7) or reversibly isomerizes
442 | Green Chem., 2013, 15, 439–445
This journal is © The Royal Society of Chemistry 2013

View Article Online
Green Chemistry
Paper
Fig. 6 Influence of temperature on the yield of transformation of chitosan
(300 mg) to levulinic acid (in 12 mL 2 M H2SO4; t = 20 min); yield (mol%) =
100% × nlevulinic acid/nsubstrate
.
the presence of hydrochloric and sulfuric acid. Based on the
results obtained from cellulose and cellobiose (Fig. 5), 2 M sul-
furic acid (or hydrochloric acid) was applied. By using LMw-
chitosan (low-molecular weight chitosan) as a model com-
pound, the effect of the irradiation time and reaction temp-
erature on the levulinic acid formation was measured in the
range of 10–30 min and 120–190 °C (Fig. 6; 190 °C is the temp-
erature limit of our instrumentation). After 10 min, only a
trace amount of the product was detected, and over 30 min an
increasing amount of tar formation was observed, thus, the
reaction time was set to 20 min in further experiments. By
increasing the reaction temperature, in parallel the amount of
Scheme 2 Decomposition of carbohydrates of plant origin to levulinic acid
based on Horváth’s proposed mechanism from ref. 32.
via (4a–c), at 170 °C to protonated fructopyranose intermediate levulinic acid increased as well, however no temperature
(5), which is responsible for the formation of humin by- maximum was observed in the temperature range applied
products.32
(cf. hydrolysis of cellobiose). Consequently, further investi-
Due to the relatively weak β(1 → 4) glycosidic bond between gations were performed at 190 °C (Table 1, entries 10–19).
the two D-glucose units, dehydration of cellobiose (2) provided
Importantly, applying 20 min irradiation time for the con-
a similar amount of levulinic acid (41.3%, Table 1, entry 6).
version of monomers (D-glucosamine, N-Ac-D-glucosamine)
Expectedly, the high molecular weight cellulose (1) resulted resulted in extensive humin formation, thus, in the presence
in the lowest yield of levulinic acid using 30 min reaction time, of 2 M sulfuric acid, the reactions were performed only for
however by allowing a longer reaction time (50 min) the yield 10 min. Moreover, the yield of levulinic acid (Table 1, entries
could be improved (Table 1, entries 7 and 9). Cellulose was 11 and 13) was practically the same for N-Ac-D-glucosamine
insoluble in our hydrolysis cocktail, thus its partial hydrolysis and D-glucosamine, which indicates that N-acetylation has neg-
to soluble oligomers should precede its further degradation to ligible influence on the levulinic acid formation. In order to
D-glucose. Theoretically, D-glucose could be derived from 1 by compare the result obtained (59.4%) for D-glucosamine24 in
terminal cleavage of the polymer chain. The relatively low con- the presence of a tin based catalyst, the degradation was
centration of fructose has an unfavorable effect on the levuli- carried out using H2SO4 as a catalyst for 30 min, resulting in a
nic acid formation. On the other hand, the longer reaction similar yield of levulinic acid of 64.5% (ESI Fig. S7†). It is
time favors humin and tar formation from the corresponding clearly proven that the catalyst has no influence on the equili-
C6-isomers (5).13,32 The reproducibility of the catalytic reaction brium of humin and/or tar formation, as expected. The reac-
was confirmed by repeating the degradation of cellulose in the tion was monitored by NMR spectroscopy, neither 5-HMF nor
presence of H2SO4 at 170 °C for 50 min. The yield of levulinic amide by-products were detected in the reaction mixture. It
acid was found to be 34.2% (Table 1, entry 9) and 33.9% in the can be explained that under acidic conditions interconversion
repeated experiment.
of 5-HMF to levulinic acid is faster than its formation.
In order to evaluate the utilization of polysaccharides of Although the formation of a ketimine intermediate and NH3
animal origin for the production of levulinic acid, various elimination from D-glucosamine were proposed recently, none
N-containing carbohydrates (chitin, chitosan, and their mono- of them could be detected from chitin and chitosan.24 Interest-
mers: D-glucosamine and N-Ac-D-glucosamine) were studied in ingly, the yield of transformation after 20 min microwave
This journal is © The Royal Society of Chemistry 2013
Green Chem., 2013, 15, 439–445 | 443

View Article Online
Paper
Green Chemistry
irradiation was 19.3% for LMw-chitosan (Table 1, entry 15), Euriso-top SAS, Saint-Aubin Cedex, France. NMR spectra were
and almost double, 32.1%, for MMw-chitosan (Table 1, entry recorded in a 5-mm NMR tube using a Bruker Avance-250
17). The two kinds of chitosans are commercial products spectrometer. Hydrolysis and dehydration reactions were per-
(Sigma-Aldrich), no data are available on their exact compo- formed in a 10-mL quartz reactor with a Milestone Multisynth
sition. It was recently reported24 that hydrolysis of the glycosi- AFC-FO 300 microwave instrument (Labsystem Kft., Budapest,
dic bonds of chitosan is a slower reaction than the subsequent Hungary) equipped with an automated safety release system.
dehydration steps of D-glucosamine which go through the
The Multisynth microwave reactor was used always in the
open-chain aldohexose form to levulinic acid, furthermore the single-mode option. The reactions were conducted between
reaction rate of the hydrolysis of glycosidic bonds can be influ- 110 and 190 °C by using the automatic temperature control
enced by the acid concentration. To obtain the same yield of system of the instrument. The upper limit of the energy
levulinic acid, transformation of chitosan was performed for a applied was set to 250 W and the reaction time was varied
much longer time (50 min instead of 30) than that necessary from 10 minutes to 50 minutes. In a typical experiment, the
for the degradation of the glucosamines. This can be explained carbohydrate sample solution was mixed or suspended
by the slow formation of carbonium ion in the rate-determin- (depending on the solubility of the sample) with 4 mL 2 M sul-
ing step when more concentrated acid is used.35 During the furic or hydrochloric acid. After completing the reaction, the
degradation of chitin (polymer of N-acetyl-D-glucosamine) at reactor was cooled down by applying the air-cooling system at
190 °C, only a trace amount of levulinic acid was detected after maximum power. Due to the small size of the reaction flask all
20 min. However, by increasing the reaction time to 30 min, a reactions were repeated three more times to collect enough
significant amount of levulinic acid (37.8%, Table 1, entry 19) material for further analysis. The combined reaction mixture
was formed.
(ca. 16 mL) was passed through a laboratory filter paper in
order to remove the insoluble, brownish impurities, which
formed occasionally. This solid by-product was washed first
with distilled water (3 × 3 mL) and ethyl acetate (3 × 3 mL),
then all filtrates were combined (ca. 34 mL). The phases were
Conclusions
This study demonstrates that polysaccharides of both plant separated, and levulinic acid was completely extracted from
and animal origin can be successfully converted to levulinic the solution with ethyl acetate (4 × 15 mL). The combined
acid in the presence of sulfuric acid by using microwave dielec- organic phase was dried over MgSO4. After evaporating the
tric heating. Investigating the effect of acid concentration on solvent a brownish-yellowish, viscous liquid remained. It was
1
the formation of levulinic acid showed that the maximum identified as levulinic acid by H-NMR and 13C-NMR, its yield
yield of levulinic acid was obtained in the presence of 2 M sul- was calculated from the 1H-NMR spectrum using benzene as
furic acid for saccharose, cellobiose and cellulose. The reac- an internal standard (ESI Fig. S2 and S3†). Yield was calculated
tions were performed in the presence of hydrochloric acid as as yield (%) = nlevulinic acid/nsubstrate
well. It is proposed that chlorine containing catalysts can be
.
avoided without significant compromise on the yield of trans-
formation. The reaction time was significantly reduced by
using microwave irradiation as a more effective heating
Acknowledgements
method compared to conventional heating. However, the This work was supported by the Hungarian Scientific Research
heating method had no influence on the product yield, only Fund (Grant no. OTKA-CNK 78065). The European Union and
the rate of hydrolysis was accelerated. It was found that the the European Social Fund have provided financial support to
optimum temperature for the acid catalyzed decomposition of the project under the grant agreement no. TÁMOP 4.2.1./B-09/
cellulose was 170 °C, and the reaction time was 50 min. The KMR-2010-0003. Special thanks to Prof. István T. Horváth (City
systematic investigation on carbohydrates of animal origin University of Hong Kong) for useful discussion.
shows that chitosan and chitin can be used as renewable feed-
stocks for the production of levulinic acid with a yield of
20–39%. This study proves that non-edible biomass can be a
possible source and/or raw material of platform molecules
References
such as levulinic acid, i.e. it clearly corresponds to the prin-
ciples of biorefinery.
1 (a) D. L. Klass, Biomass for Renewable Energy, Fuels, and
Chemicals, Elsevier, Amsterdam, 1998; (b) D. J. C. MacKay,
Sustainable Energy – Without the Hot Air, UIT Cambridge
Ltd, Cambridge, 2009.
2 J. Chow, R. J. Kopp and P. R. Portney, Science, 2003, 302,
1528.
Experimental
All chemicals were purchased from Sigma-Aldrich Kft., Buda-
pest, Hungary (except ethyl acetate which was obtained from
Molar Chemicals Kft., Budapest, Hungary) and used without
further purification. Deuterated solvents were purchased from
3 A. J. Ragauskas, C. K. Williams, B. H. Davison, G. Britovsek,
J. Cairney, C. A. W. J. Frederick Jr., J. P. Hallett, D. J. Leak,
C. L. Liotta, J. R. Mielenz, R. Murphy, R. Templer and
T. Tschaplinski, Science, 2006, 311, 484.
444 | Green Chem., 2013, 15, 439–445
This journal is © The Royal Society of Chemistry 2013

View Article Online
Green Chemistry
Paper
4 (a) S. Fernando, S. Adhikari, C. Chandrapal and N. Murali,
Energy Fuels, 2006, 20, 1927; (b) G. W. Huber and A. Corma,
R. M. West and J. A. Dumesic, Langmuir, 2010, 26,
16291.
Angew. Chem., Int. Ed., 2007, 46, 7184; (c) J. N. Chheda, 19 G. J. Mulder, J. Prakt. Chem., 1840, 21, 203.
G. W. Huber and J. A. Dumesic, Angew. Chem., Int. Ed., 20 (a) C. Chang, X. Ma and P. Cen, Chin. J. Chem. Eng., 2006,
2007, 46, 7164; (d) M. Stöcker, Angew. Chem., Int. Ed., 2008,
47, 9200.
5 H. Arakawa, M. Aresta, J. N. Armor, M. A. Barteau,
E. J. Beckman, A. T. Bell, J. E. Bercaw, C. Creutz, E. Dinjus,
D. A. Dixon, K. Domen, D. L. DuBois, J. Eckert, E. Fujita,
14, 708; (b) B. Girisuta, B. Danon, R. Manurung, L. P. B.
M. Janssen and H. J. Heeres, Bioresour. Technol., 2008, 99,
8367; (c) J. Hegner, K. C. Pereira, B. DeBoef and B. L. Lucht,
Tetrahedron Lett., 2010, 51, 2356; (d) C. Hongzhang, Y. Bin
and J. Shengying, Bioresour. Technol., 2011, 102, 3568.
D. H. Gibson, W. A. Goddard, D. W. Goodman, J. Keller, 21 (a) L. V. Budarin, H. J. Clark, A. B. Lanigan, P. Shuttleworth
G. J. Kubas, H. H. Kung, J. E. Lyons, L. E. Manzer,
T. J. Marks, K. Morokuma, K. M. Nicholas, R. Periana,
L. Que, J. Rostrup-Nielson, W. M. H. Sachtler,
L. D. Schmidt, A. Sen, G. A. Somorjai, P. C. Stair,
B. R. Stults and W. Tumas, Chem. Rev., 2001, 101, 953.
6 J. J. Bozell and G. R. Petersen, Green Chem., 2010, 12, 539.
7 I. T. Horváth and P. Anastas, Chem. Rev., 2007, 107, 2169.
and D. J. Macquarrie, Bioresour. Technol., 2010, 101, 3776;
(b) Z. Zhang and Z. K. Zhao, Bioresour. Technol., 2010, 101,
1111; (c) Z. Zhang and Z. K. Zhao, Carbohydr. Res., 2009,
344, 2069; (d) C. Z. Li, Z. H. Zhang and Z. B. K. Zhao, Tetra-
hedron Lett., 2009, 50, 5403; (e) A. Orozco, M. Ahmad,
D. Rooney and G. Walkeret, Process Saf. Environ. Prot., 2007,
85, 446.
8 S. Van de Vyver, J. Gebores, P. A. Jacobs and B. F. Sels, 22 M. Mascal and E. B. Nikitin, ChemSusChem, 2009, 2, 859.
ChemCatChem, 2011, 3, 82.
9 B. Girisuta, L. P. B. M. Janssen and H. J. Heeres, Ind. Eng.
Chem. Res., 2007, 46, 1696.
23 R. Xing, S. Liu, H. Yu, Z. Guo, P. Wang, C. Li, Z. Lia and
P. Lia, Carbohydr. Res., 2005, 340, 2150.
24 K. W. Omari, J. E. Besaw and F. M. Kerton, Green Chem.,
2012, 14, 1480.
10 (a) J.-P. Lange, R. Price, P. M. Ayoub, J. Louis, L. Petrus,
L. Clarke and H. Gosselink, Angew. Chem., Int. Ed., 2010, 25 (a) K. W. Omari, L. Dodot and F. M. Kerton, ChemSusChem,
49, 4479; (b) D. M. Alonso, J. Q. Bond and J. A. Dumesic,
Green Chem., 2010, 12, 1493; (c) J. C. Serrano-Ruiz,
2012, 5, 1767; (b) M. W. Drower, K. W. Omari, J. N. Murphy
and F. M. Kerton, RSC Adv., 2012, 2, 4642.
D. J. Braden, R. M. West and J. A. Dumesic, Appl. Catal., B, 26 N. Aas, Y. Li and M. Bowker, J. Phys.: Condens. Matter, 1991,
2010, 100, 184. 3, S281.
11 H. Mehdi, V. Fábos, R. Tuba, A. Bodor, L. T. Mika and 27 A. Boddien, D. Mellmann, F. Gärtner, R. Jackstell, H. Junge,
I. T. Horváth, Top. Catal., 2008, 48, 49.
12 I. T. Horváth, H. Mehdi, V. Fábos, L. Boda and L. T. Mika,
Green Chem., 2008, 10, 238.
13 B. Girisuta, L. P. B. M. Janssen and H. J. Heeres, Chem. Eng.
Res. Des., 2006, 84, 339.
14 (a) J. M. Tukacs, D. Király, A. Strádi, G. Novodárszki, Z. Eke,
P. J. Dyson, G. Laurenczy, R. Ludwig and M. Beller, Science,
2011, 333, 1733.
28 V. Fábos, I. T. Horváth, N. Kaposy and H. Mehdi, HU2009-
00276-A1, 2009.
29 R. Gedye, F. Smith, K. Westaway, H. Ali, L. Baldisera,
L. Laberge and J. Rousell, Tetrahedron Lett., 1986, 27, 279.
G. Dibó, T. Kégl and L. T. Mika, Green Chem., 2012, 14, 30 R. J. Giguere, T. L. Bray, S. M. Duncan and G. Majetich,
2057; (b) F. M. A. Geilen, B. Engendahl, A. Harwardt, Tetrahedron Lett., 1986, 27, 4945.
W. Marquardt, J. Klankermayer and W. Leitner, Angew. 31 (a) C. O. Kappe, Angew. Chem., Int. Ed., 2004, 43, 6250;
Chem., Int. Ed., 2010, 49, 5510.
(b) D. Dallinger and C. O. Kappe, Chem. Rev., 2007, 107,
15 D. Fegyverneki, L. Orha, L. Láng and I. T. Horváth, Tetra-
hedron, 2010, 66, 1078.
16 P. K. Wong, C. Li, L. Stubbs, M. van Menurs,
2563.
32 G. R. Akien, L. Qi and I. T. Horváth, Chem. Commun., 2012,
48, 5850.
D. G. A. Kumband, C. Y. Lim and E. Drent, PTC/SG2012/ 33 E. A. Pidko, V. Degirmenci, R. A. van Santen and
000107, 2012. E. J. M. Hensen, Angew. Chem., Int. Ed., 2010, 49, 2530.
17 M. Chalid, H. J. Heeres and A. A. Broekhuis, J. Appl. Polym. 34 J. Horvat, B. Klaic, B. Metelko and V. Sunjic, Tetrahedron
Sci., 2012, 123, 3556. Lett., 1985, 26, 2111.
18 (a) J. Q. Bond, D. M. Alonso, R. M. West and J. A. Dumesic, 35 K. M. Vårum, M. H. Ottøy and O. Smidsrød, Carbohydr.
Science, 2010, 327, 1110; (b) J. Q. Bond, D. M. Alonso,
Polym., 2001, 46, 89.
This journal is © The Royal Society of Chemistry 2013
Green Chem., 2013, 15, 439–445 | 445