Conversion of cellulose to glucose and levulinic acid via solid-supported acid catalysis

Article abstract of DOI:10.1016/j.tetlet.2010.02.148

Cellulose is hydrolyzed to glucose, which is further converted to levulinic acid in the presence of surface-supported Br?nsted and Lewis acid catalysts. Nafion catalysts, in particular, have the potential to be recycled or applied to a continuous flow reactor for the synthesis of these biofuel precursors.

Full text of DOI:10.1016/j.tetlet.2010.02.148

Tetrahedron Letters 51 (2010) 2356–2358
Contents lists available at ScienceDirect
Tetrahedron Letters
journal homepage: www.elsevier.com/locate/tetlet
Conversion of cellulose to glucose and levulinic acid via solid-supported
acid catalysis
*
*
Jessica Hegner, Kyle C. Pereira, Brenton DeBoef , Brett L. Lucht
Department of Chemistry, University of Rhode Island, Kingston, Rhode Island 02881, United States
a r t i c l e i n f o
a b s t r a c t
Article history:
Cellulose is hydrolyzed to glucose, which is further converted to levulinic acid in the presence of surface-
supported Brønsted and Lewis acid catalysts. Nafion catalysts, in particular, have the potential to be recy-
cled or applied to a continuous flow reactor for the synthesis of these biofuel precursors.
Ó 2010 Elsevier Ltd. All rights reserved.
Received 17 December 2009
Revised 22 February 2010
Accepted 23 February 2010
Available online 1 March 2010
Cellulose, the principle structural component of plants, is the
most abundant organic compound on earth. It is a biopolymer that
consists of glucose units joined by b(1?4) linkages. This bond, by
itself, is not partially strong, as it can be readily hydrolyzed by sim-
ple acids such as HCl or by cellulolytic enzymes. However, the cel-
lulose polymer, as a whole, is very robust. The individual glucose
monomers are so densely packed together by an extensive network
of hydrogen bonds that enzyme catalysts and most aqueous sol-
vents cannot penetrate the structure. Consequently, cellulose is
virtually insoluble in water and is generally regarded as a difficult
material to work with. The development of methods for hydrolyz-
ing raw biomass from renewable, non-food bearing plants, either
to glucose or directly to biofuels, is highly desirable. These difficult
transformations are not readily feasible using the current best
technology, which employs cellulolytic enzymes, as they require
water solvents and mild conditions under which cellulose is not
readily soluble. Herein, we report an improved chemical method
to convert raw cellulose to biofuel precursors.
acid) in the presence of LiCl, resulting in an 80% conversion. The
primary product of the reaction is 5-(chloromethyl)furfural, which
is not a biofuel, but this intermediate can be readily converted to
6
5-(ethoxy)furfural, which can be used as a diesel additive.
7
Ionic liquids have been used to dissolve cellulose and even raw
8
wood chips. These solvents have the added benefit that they are
generally considered to be ‘green’ due to their low vapor pressure.
These solvents have been used with Brønsted acid catalysts to
9
convert fructose to hydroxymethylfuran. Recently, Schüth and
co-workers degraded cellulose in [BMIM][Cl], an imidazolium-
based ionic liquid, with an Amberlyst catalyst. A combined yield
of less than 4% was observed for the production of both mono-
8
and disaccharides. Subsequently, Binder and Raines described a
reaction medium containing dimethylacetamide, LiCl, CrCl
2
, and
an ionic liquid that is capable of processing both sawdust and corn
1
0
stover to furfurals in good yields.
Minimal attention has been given to the use of solid-supported
acid catalysis for the depolymerization of cellulose in water. Recent
reports describe the hydrolysis of cellulose by acid-modified amor-
The chemical conversion of cellulose to fuel feedstocks has re-
ceived significant attention recently.1 A method for converting
C6-sugars such as glucose and sorbitol to fuels suitable for use in
automobiles and aircraft has recently been reported. However,
the process employs a Pt–Re catalyst on a graphite support, and
is energy intensive, requiring temperatures >500 K and pressures
11
12
phous carbon, layered transition metal oxides, or Amberlyst re-
8
sin. These surface acidic species were quite efficient for the
conversion of cellulose into smaller, water-soluble b(1?4) glucans,
but conversion to glucose or other small biofuel precursors was
limited.
2
–4
of 27 bar.
While sugars are cheap and readily available com-
In this Letter we report the chemical conversion of cellulose to
glucose and levulinic acid by hydrolysis of silica-supported acid
catalysts under relatively mild conditions. The reaction is catalyzed
modities, their refinement from biomass adds to the overall energy
consumption of the process. Reducing conditions have also been
applied to the conversion of cellulose to sugar alcohols. Either plat-
inum or ruthenium catalysts at high temperatures and under high-
by both Nafion and FeCl
3
supported on amorphous silica.13 Nafion
SAC 13 (Nafion polymer on amorphous silica), in particular, has the
potential to be recycled or applied to a continuous flow reactor,
providing that residual, unreacted cellulose can be removed from
the system.
Our initial experiments were conducted with cellobiose as a
substrate (Scheme 1). Cellobiose is a water-soluble single repeat
unit of cellulose and is frequently used as a cellulose model
compound. A Teflon-capped glass reaction flask was charged with
2
pressure of H can convert cellulose to a mixture of polyols, whose
5
main component (80%) was sorbitol.
The current best method for converting biomass to a fuel was
reported in 2008. The process employs refluxing HCl (a Brønsted
*
Corresponding authors. Tel.: +1 401 874 9480; fax: +1 401 874 5072.
E-mail address: blucht@chm.uri.edu (B.L. Lucht).
0
040-4039/$ - see front matter Ó 2010 Elsevier Ltd. All rights reserved.
doi:10.1016/j.tetlet.2010.02.148

J. Hegner et al. / Tetrahedron Letters 51 (2010) 2356–2358
2357
Table 1
Conversion of cellulose by solid-supported acid catalysts at different temperatures
Catalyst
Temp. % Yield cycle 1a % Yield cycle 2a % Yield cycle 3a
Nafion SAC 13 130 °C 2 (0)
Nafion SAC 13 160 °C 6 (0)
Nafion SAC 13 190 °C 9 (2)
1 (0)
8 (0)
4 (4)
1 (0)
1 (0)
2 (1)
1 (0)
2 (0)
2 (5)
1 (0)
1 (0)
1 (0)
Scheme 1. Hydrolysis of cellobiose using solid-supported Brønsted and Lewis acid
catalysts.
FeCl
FeCl
FeCl
3
3
3
/Silica
/Silica
/Silica
130 °C 3 (0)
160 °C 7 (0)
190 °C 9 (1)
cellobiose (1.0 g, 3 mmol), Nafion SAC 13 (0.47 g), and 15 mL of
water. The flask was heated to 130 °C for 24 h followed by filtration
to remove the catalyst and concentrated in vacuo to remove the
water, providing a quantitative conversion of cellobiose to glucose.
The catalyst can then be reused under identical reaction conditions
to quantitatively convert cellobiose to glucose confirming the recy-
clability of the catalyst. Identical experiments were conducted
with FeCl /silica with similar quantitative conversions and recycla-
3
bility. However, reaction with the Lewis acid provided a 1:2 mix-
ture of glucose and levulinic acid. Control experiments with
a
Isolated yield of glucose (and levulinic acid). The ratio between the two com-
pounds was determined by 1H NMR.
conversions for both the second (8%) and third (7%) reaction cycles.
Though Nafion SAC 13 is generally regarded to be a robust catalyst
at moderate temperatures, the side-chain rearrangement of Naf-
ion-type polymers has been observed above its glass transition
temperature (100–110 °C). Additionally, the decrease in yield as
a result of catalyst cycling may be attributed to the presence of
residual insoluble cellulose on the catalyst surface. Even though
these yields are low when compared to most organic reactions,
the data in Table 1 are notable when compared to the majority
of work in the field of cellulose degradation chemistry. In particu-
lar, our yields are superior to those that were observed with solid-
supported catalysts in ionic liquids, and with more complex cata-
1
5
silica, FeCl
the acidic species (FeCl
3
, and an aqueous dispersion of Nafion confirmed that
and Nafion) were the reactive species
3
while the silica is a relatively inert surface under the reaction
conditions. Furthermore, control experiments in which aqueous
glucose solutions were heated in the presence of FeCl produced
3
a quantitative conversion to levulinic acid; indicating that the
reaction proceeds by a cellobiose ? glucose ? levulinic acid se-
1
4
8,11
quence. The formic acid by-product of the final step of this
process was presumably removed during the rotary evaporation.
We then expanded our investigation to probe the solid-sup-
ported acid-catalyzed hydrolysis of cellulose. A stainless steel Parr
reactor was charged with Nafion SAC 13 (0.94 g), cellulose (2.0 g,
lysts, such as acid-modified graphite.
At higher temperatures glucose is not the only product ob-
served in the reaction mixture. Significant concentrations of levu-
linic acid are also observed. Interestingly, the ratio of glucose to
levulinic acid is dependent upon the storage time. As the storage
times are gradually increased from one to four days the concentra-
tion of levulinic acid gradually increases while the concentration of
glucose decreases. This observation is consistent with two mecha-
nistic hypotheses. First, the glucose generation is limited to a frac-
tion of the total available cellulose. This percentage is dependent
upon temperature and is likely related to a soluble or accessible
component of the cellulose structure. Second, the levulinic acid is
only generated via the reaction of glucose and the rate of glucose
conversion to levulinic acid is slower than the rate of hydrolysis
of cellulose. This was confirmed by a simple control reaction, in
which glucose was heated to 190 °C in the presence of Nafion
SAC 13 for one day, resulting in a 1:1 mixture of unreacted glucose
and levulinic acid. This suggests that the reaction conditions can be
optimized for the production of either glucose or levulinic acid.
6
mmol), and water (30 mL). The reactor was heated in an oil bath
for one day followed by two sequential three day periods. After
each cycle, the reaction mixture was filtered to separate the solids,
which contained both the catalyst and unreacted cellulose, which
had been partially hydrated during the course of the reaction.
The solid was washed with warm water (5 mL) to extract any
water-soluble components. The water was removed by rotary
evaporation followed by high vacuum, leaving a residue of the
hydrolysis products, which were analyzed by 1H and C NMR
spectroscopy and compared to commercial standards. The water-
insoluble residue, which contained both the catalyst and cellulose,
was combined with additional cellulose (1.0 g) and water (30 mL)
and returned to the reaction conditions for the next cycle. This pro-
cess was repeated to determine the recyclability of the catalyst.
13
Similar procedures were utilized for the FeCl
hydrolysis of cellulose. In all cases, the only water-soluble products
observed were glucose and levulinic acid (Table 1).
3
/silica-catalyzed
Hydrolysis of cellulose by FeCl
(Table 1, 1 day). However, the recyclability of the catalyst is less
adequate for the FeCl
/silica catalyst. This is likely due to Fe3+
3
/silica provides similar results
3
Hydrolysis of cellulose by Nafion SAC 13 is highly dependent
upon the reaction temperature. At lower temperatures (130 °C)
there is little conversion of cellulose to glucose (2%) after one
day. As the temperature is increased, the conversion rates improve.
The increased conversion is likely related to increased solubility of
cellulose at higher temperatures. The conversion at 190 °C is very
good (11%) after the first cycle (1 day), In addition, the catalyst
shows good recyclability. Reuse of the catalyst provides efficient
leaching from the silica surface under hydrolytic conditions and
limits the utility of these catalysts.
In summary, we have demonstrated that solid-supported acid
3
catalysts (Nafion SAC 13 and FeCl /silica) are active for the hydroly-
sis of cellulose to glucose and levulinic acid at 190 °C. The recyclabil-
ity of the Nafion SAC 13 is good; suggesting that this is a promising
catalyst for the conversion of cellulose into liquid biofuel precursors.
Assuming that the solid catalyst can be mechanically separated from

2
358
J. Hegner et al. / Tetrahedron Letters 51 (2010) 2356–2358
residual cellulose material, solid-supported acidic catalysts, such as
those described herein, could be employed en masse for the conver-
sion of cellulose to biofuel precursors.
4. West, R. M.; Liu, Z. Y.; Peter, M.; Gaertner, C. A.; Dumesic, J. A. J. Mol. Catal. A
008, 296, 18.
2
5.
6.
7.
Fukuoka, A.; Dhepe, P. L. Angew. Chem., Int. Ed. 2006, 45, 5161.
Mascal, M.; Nikitin, E. B. Angew. Chem., Int. Ed. 2008, 47, 7924.
Swatloski, R. P.; Spear, S. K.; Holbrey, J. D.; Rogers, R. D. J. Am. Chem. Soc. 2002,
1
24, 4974.
Acknowledgments
8
9
.
.
Rinaldi, R.; Palkovits, R.; Schüth, F. Angew. Chem., Int. Ed. 2008, 47, 8047.
Hu, S.; Zhang, Z.; Zhou, Y.; Han, B.; Fan, H.; Li, W.; Song, J.; Xie, Y. Green Chem.
2008, 10, 1280.
0. Binder, J. B.; Raines, R. T. J. Am. Chem. Soc. 2009, 131, 1979.
1. Suganuma, S.; Nakajima, K.; Kitano, M.; Yamaguchi, D.; Kato, H.; Hayashi, S.;
Hara, M. J. Am. Chem. Soc. 2008, 130, 12787.
This work has been partially funded by the University of Rhode
Island (URI) Partnership for Energy (http://www.uri.edu/cels/ceoc/
ec/index.html). J.H. is an undergraduate URI Energy Fellow.
1
1
1
1
1
2. Takagaki, A.; Tagusagawa, C.; Domen, K. Chem. Commun. 2008, 5363.
3. Keinan, E.; Mazur, Y. J. Org. Chem. 2002, 43, 1020.
4. The conversion of saccharides to levulinic acid is well-known: Feather, M. S.;
Harris, J. F.. In Adv. Carbohydr. Chem. Biochem.; Tipson, R. S., Horton, D., Eds.;
Academic Press: New York, 1973; vol. 28, pp 212–217.
References and notes
1.
2.
3.
Huber, G. W.; Iborra, S.; Corma, A. Chem. Rev. 2006, 106, 4044.
Huber, G. W.; Chheda, J. N.; Barrett, C. J.; Dumesic, J. A. Science 2005, 308, 1446.
Kunkes, E. L.; Simonetti, D. A.; West, R. M.; Serrano-Ruiz, J. C.; Gaertner, C. A.;
Dumesic, J. A. Science 2008, 322, 417.
1
5. Zook, L. A.; Leddy, J. Anal. Chem. 1996, 68, 3793.