High-yield conversion of plant biomass into the key value-added feedstocks 5-(hydroxymethyl)furfural, levulinic acid, and levulinic esters via 5-(chloromethyl)furfural

Article abstract of DOI:10.1039/b918922j

5-(Hydroxymethyl)furfural, levulinic acid, ethyl levulinate and butyl levulinate are produced by the solvolysis of 5-(chloromethyl)furfural (CMF). Since CMF can be derived in high yield from sugars, cellulose, or lignocellulosic feedstocks, the process de

Full text of DOI:10.1039/b918922j

View Article Online / Journal Homepage / Table of Contents for this issue
COMMUNICATION
www.rsc.org/greenchem | Green Chemistry
High-yield conversion of plant biomass into the key value-added feedstocks
5-(hydroxymethyl)furfural, levulinic acid, and levulinic esters via
5-(chloromethyl)furfural
Mark Mascal* and Edward B. Nikitin
Received 11th September 2009, Accepted 26th November 2009
First published as an Advance Article on the web 15th December 2009
DOI: 10.1039/b918922j
5-(Hydroxymethyl)furfural, levulinic acid, ethyl levulinate
and butyl levulinate are produced by the solvolysis of 5-
(chloromethyl)furfural (CMF). Since CMF can be derived
in high yield from sugars, cellulose, or lignocellulosic
feedstocks, the process described here presents an efficient
entry into the value-added manifold of biomass-derived
products of relevance to the organic materials and fuel
industries.
cannot cover the breadth of chemical feedstock classes available
from petroleum.
Recently, we described the digestion of sugars, cellulose,
or cellulosic biomass with hydrochloric acid in a biphasic
reactor which resulted in the isolation of a mixture of 5-
(chloromethyl)furfural 1 (CMF) and levulinic acid 2 in up to
95% combined yield (Scheme 1).4 The major product of the
reaction is by far CMF 1, which accounts for between 70 and
90% of the organic material isolated, depending on the feedstock
loading, while levulinic acid comprises less than 10% of the
product mixture.
While our earlier work focused on CMF 1 as a precur-
sor to the new generation biofuels 5-(ethoxymethyl)furfural
3 and 5-methylfurfural 4 by reaction of 1 with ethanol and
hydrogen, respectively,5 in this report we describe the effi-
cient conversion of CMF 1 into the value-added products 5-
(hydroxymethyl)furfural 5 (HMF), levulinic acid 2 (LA), ethyl
levulinate 6a (EL) and n-butyl levulinate 6b (BL) by reaction
with water or alcohols at elevated temperatures.
HMF 5 is a versatile platform chemical for the synthe-
sis of a wide range of industrially important materials, in-
cluding biofuels.6 For example, 5 is a starting material in
Dumesic’s approach to the production of biomass-derived
hydrocarbons.7 It can also be reduced to 2,5-dimethylfuran
7 or 2,5-dimethyltetrahydrofuran 8, both high energy liquids
which have been proposed as fuel additives.8 HMF 5 is also
As and when the global petroleum economy draws to a close,
the portfolio of derived products from biomass will need to
diversify in order to satisfy feedstock requirements across a
range of markets. Bioethanol and biobutanol, for example,
have primarily uses as fuel oxygenates, and can be used in the
production of biodiesel from vegetable oil, but they can also
be dehydrated to give the major industrial feedstocks ethylene
and 1-butene. Similarly, biogenic methane can be catalytically
converted into methanol which, like ethanol and butanol, can be
used to prepare biodiesel, but can also be submitted to oxidative
coupling processes to give ethylene and higher olefins.1-3 Yet
as versatile as these mainstream biomass derivatives are, they
Department of Chemistry and Bioenergy Research Group, University of
California, Davis, 1 Shields Avenue, Davis, CA, 95616, USA
Scheme 1
370 | Green Chem., 2010, 12, 370–373
This journal is
The Royal Society of Chemistry 2010
©

View Article Online
Scheme 2
a precursor to the polyester monomer furan-2,5-dicarboxylic rate of the conversion, such that after only 30 s heating in
acid 9, a molecule that has the distinction of appearing on the
list of Top Value-Added Chemicals From Biomass published
by the US Department of Energy’s National Renewable Energy
Laboratory (NREL). This list of twelve, mainly sugar-derived
products was assembled in order to identify major opportunities
for “the production of value-added chemicals from biomass
that would economically and technically support the production
of fuels and power in an integrated biorefinery, and identify
the common challenges and barriers of associated production
technologies.”9 Finally, the core structure of 5 also appears in
dilute HCl the ratio of LA to HMF approaches 1 : 1. However,
sampling the reaction mixture at subsequent intervals shows
little in the way of further progress over the course of the
next hour. Heating CMF in water at temperatures >100 ◦C
in a closed vessel was also shown to accelerate the reaction, and
it was by a combination of higher temperature and acid catalysis
that a high yielding route to 2 was first developed. Thus, heating
1 in dilute HCl at 150 ◦C for 5 h gave a 94% isolated yield
of LA. In an attempt to further shorten the reaction time, the
temperature was raised to 190 ◦C, which resulted in the complete
conversion of 1 into 2 within 20 minutes without the need for
additional acid (Scheme 3).
R
the blockbuster antiulcer drug Ranitidine (Zantacꢀ) 10.
Although HMF 5 can be prepared from fructose in good
yield,10 fructose itself is sourced from the food chain, and it
would be far too expensive to produce 5 on an industrial
scale in this manner. Interestingly, a report as far back as
1912 describes the hydrolysis of 1 to 5 in unspecified yield.11
This paper was cited by Haworth and Jones in 1944 in their
repetition of the procedure,12 which again lacked experimental
detail. We thus undertook a careful study of this conversion
with the benefit of modern analytical techniques. The results
are quantified as shown in Scheme 2. Thus, under optimized
conditions (hydrolysis in boiling water for 30 s), HMF 5 is
produced in 86% yield from 1 alongside 10% LA 2. Formation
of the latter product 2, observed across a range of conditions,
appears to be unavoidable in this reaction. In the case, however,
where LA 2 is actually the desired product, it can be produced
with high efficiency (see below).
Scheme 3
When CMF 1 is stirred with ethanol at room temperature,
5-(ethoxymethyl)furfural 3 is produced in high yield.5 As was
the case with the CMF 1 to HMF 5 conversion, more severe
conditions shift the product distribution from the substituted
furan toward the levulinate. Thus, heating CMF 1 in ethanol at
160 ◦C for 30 min gave EL 6a in 85% isolated yield (Scheme 4).
LA 2, like furan-2,5-carboxylic acid 9, also appears on the
NREL Top 12 Value-Added Chemical list.9 In its candidate
summary biography, LA 2 is referred to as “one of the more
recognized building blocks available from carbohydrates,” the
derivatives of which “address a number of large volume chemical
markets.” These derivatives include levulinate esters and 2-
methyltetrahydrofuran (fuel additives), d-aminolevulinic acid
(herbicide), and b-acetylacrylic acid, diphenolic acid and 1,4-
pentanediol (polymer building blocks).9 The technical barrier
to development of an LA-based chemical intermediate trade is
indicated to be the moderate selectivity of the process used to
derive LA 2 from biomass.
When CMF 1 is heated for longer periods in water, the amount
of LA 2 grows at the expense of HMF 5, but even after 15 h
reflux the reaction is not complete, and increasing amounts of
decomposition products (dark solids) are seen to accumulate.
The addition of acid to the reaction mixture increases the initial
Scheme 4
The analogous reaction of 1 with n-butanol at 110 ◦C for
2 h likewise gave BL 6b (Scheme 5). In the conversion of C6
furfurals to levulinates, there is a loss of a one-carbon fragment
at the formate oxidation state, i.e. formic acid in the hydrolysis of
CMF 1 or the corresponding formate ester in the alcoholysis. In
the reactions in Schemes 3 and 4, no attempt was made to recover
the formic acid or ethyl formate by-products. In the reaction
with n-butanol, however, n-butyl formate 11 (BF) was isolated
in approximately equal amounts to that of butyl levulinate 6b,
although it could not be completely separated from the excess
butanol by distillation.13
Scheme 5
This journal is
The Royal Society of Chemistry 2010
Green Chem., 2010, 12, 370–373 | 371
©

View Article Online
Table 1 Yields of value-added products from carbohydrate and
biomass feedstocks determined as the following expression: (yield of
CMF 1 and LA 2 from either glucose, sucrose, cellulose, or corn
stover)4 ¥ (yield of HMF 5, LA 2, EL 6a, BL 6b, or BF 11 from CMF 1)
a central organic platform chemical, and biphasic acid/solvent
carbohydrate digesters as the method of choice for exploiting
cellulosic biomass.
feedstock
Experimental
derivative
glucose
sucrose
cellulose
corn stover
Hydrolysis of CMF 1 to HMF 5
HMF 5 (LA 2)a 70.0 (12.6) 77.3 (14.2) 72.0 (13.6) 69.1 (15.8)
LA 2a
78.7
72.7
87.1
80.5
81.5
75.2
81.0
74.6
CMF 1 (0.9490 g, 6.565 mmol) was added in a single portion
to boiling water (900 mL) in a 2 L round-bottomed flask
with fast stirring. After 30 s the reaction was quickly cooled
to room temperature in an ice/water bath. The mixture was
extracted with ethyl acetate (5 ¥ 100 mL), the aqueous layer
was saturated with sodium chloride, and extraction with ethyl
acetate was continued (5 ¥ 100 mL). The combined organic
extracts were dried (MgSO4) and the solvent was evaporated.
Column chromatography (silica, 1 : 1 Et2O–CH2Cl2) gave HMF
5 (0.7137 g, 86.2%) and LA 2 (0.0751 g, 9.9%).
EL 6ab
BL 6b c(BF 11) 72.4 (70.4) 80.2 (77.8) 74.9 (72.4) 74.4 (69.5)
a The LA 2 yield is a combination of that produced by both reactions.
b Assumes conversion of both CMF 1 and LA 2 into EL 6a in the same
percent yield. c Assumes conversion of both CMF 1 and LA 2 into BL
6b in the same percent yield.
If the results in Scheme 2–5 are superimposed upon the high
yields of CMF 1 and LA 2 from glucose, sucrose, cellulose,
and corn stover recently described by our group,4 the overall
yield of value-added products from these substrates can be
reckoned as shown in Table 1. To our knowledge, these levels of
conversion from carbohydrate feedstocks are largely unrivalled
in the literature.
Hydrolysis of CMF 1 to LA 2
A 150 mL sealed glass vessel was charged with CMF (0.9889 g,
6.84 mmol) and water (30 mL) and the mixture was heated in
an oil bath at 190 ◦C with stirring for 20 min. The reaction was
allowed to cool to room temperature and filtered. The filtrate was
extracted with ethyl acetate (5¥100 mL), the aqueous layer was
saturated with sodium chloride, and extraction with ethyl acetate
was continued (5¥100 mL). The combined organic extracts were
dried (MgSO4) and the solvent was evaporated to give LA 2
(0.7248 g, 91.2 %).
While a comprehensive review of developments in the area
of biomass conversion cannot be given here, the results given in
Table 1 can be shown to compare favorably to approaches to LA
2 and HMF 5 at the forefront of the field. Thus, a recent report
by Yong and co-workers described the conversion of glucose
into HMF 5 in up to 81% yield (by GC and NMR analysis)
using a chromium(II) catalyst and N-heterocyclic carbene ligand
system.14 However, the method requires the use of expensive
1-butyl-3-methylimidazolium chloride ([BMIM]Cl) ionic liquid
as the solvent. Practical questions remain about this process,
not least of which being the activity of the catalytic system
after multiple cycles. Along the same lines, Dumesic and co-
workers have published a study in which HMF 5 is derived
from glucose with 53% selectivity at high conversion in 60%
aqueous DMSO in a biphasic reactor.15 While promising, the
separation of DMSO from the HMF product remains an issue.
Most recently, Binder and Raines have shown that corn stover
can be processed directly into HMF 5 in 48% yield (by HPLC
analysis) in a medium loaded with 10 mol% CrCl3, 10 mol%
HCl, and 60 wt% [EMIM]Cl in N,N-dimethylacetamide-LiCl
solution.16
As concerns LA 2, although a multitude of routes have been
described over the years, the highest yields to date from biomass
are claimed for the “Biofine Process,” a two-stage protocol
involving high pressures and temperatures, for which yields
of between 70–80% of 2 are reported.17 The technology we
have developed, however, operates under substantially milder
conditions and is more versatile with respect to product output,
i.e. includes furfurals such as 3, 4, and 5 and their derivatives.
In conclusion, we have described a simple, efficient processes
for the conversion of biomass-derived 5-(chloromethyl)furfural
1 into the mainstream value-added products 5-(hydroxymethyl)
furfural 5, levulinic acid 2, ethyl levulinate 6a and butyl levulinate
6b, which may be applied to a variety of purposes in the materials
and fuel industries that would otherwise involve the expenditure
of petroleum. We foresee a time when CMF 1 may emerge as
Ethanolysis of CMF 1 to EL 6a
A 150 mL sealed glass vessel was charged with CMF (3.1662 g,
21.90 mmol) and absolute EtOH (80 mL) and the mixture was
heated in an oil bath at 160 ◦C with stirring for 30 min. The
reaction was allowed to cool to room temperature and the excess
ethanol was evaporated. Chromatography (silica, 1 : 1 hexane–
ethyl acetate) gave EL 6a (2.6746 g, 84.7%).
Butanolysis of CMF 1 to BL 6b and BF 11
A solution of CMF (9.2211 g, 63.79 mmol) in n-BuOH (50 mL)
was heated at 110 ◦C with stirring for 2 h. Distillation gave a
mixture of BF 11 (5.65 g, 86.7%) and n-BuOH (16.40 g) (by
NMR integration) between 104–110 ◦C and BL 6b (9.2770 g,
84.4%) between 90–91 ◦C at 2 Torr.
Acknowledgements
This work was supported by the US Department of Energy
(award number DE-FG36-08GO88161) and the Nevada Insti-
tute for Renewable Energy Commercialization (award number
2008/11/002).
References
1 L. Chen, X.-W. Zhang, L. Huang and L.-C. Lei, Chemical Engineer-
ing and Processing, 2009, 48, 1333.
2 T. Ren, M. K. Patel and K. Blok, Energy, 2008, 33, 817.
3 J. H. Lunsford, Catal. Today, 2000, 63, 165.
4 M. Mascal and E. B. Nikitin, ChemSusChem, 2009, 2, 859.
372 | Green Chem., 2010, 12, 370–373
This journal is
The Royal Society of Chemistry 2010
©

View Article Online
5 M. Mascal and E. B. Nikitin, Angew. Chem., Int. Ed., 2008, 47, 7924.
6 J. Lewkowski, ARKIVOC, 2001, (i), 17.
7 G. W. Huber, J. N. Chheda, C. J. Barrett and J. A. Dumesic, Science,
2005, 308, 1446.
8 Y. Roma´n-Leshkov, C. J. Barrett, Z. Y. Liu and J. A. Dumesic, Nature,
2007, 447, 982.
9 “Top Value Added Chemicals From Biomass. Volume I: Results of
Screening for Potential Candidates from Sugars and Synthesis Gas.”
Technical report identifier PNNL-14804, Pacific Northwest National
Laboratory and the National Renewable Energy Laboratory, August
2004 (http://www1.eere. energy.gov/biomass/pdfs/35523.pdf).
10 B. F. M. Kuster, Starch/Staerke, 1990, 42, 314.
6b (238 ◦C) and 11 (107 ◦C), and the butanol solvent (117 ◦C).
Hence, separation of these minor products, which are themselves
commercially traded organic intermediates, would be uncomplicated.
If desired, the occurrence of haloalkanes could be avoided altogether
by practicing the hydrolysis of 1 to levulinic acid 2 (Scheme 3)
followed by standard esterification of 2 with ethanol or butanol using
a non halogen-containing catalyst.
14 G. Yong, Y. Zhang and J. Y. Ying, Angew. Chem., Int. Ed., 2008, 47,
9345.
15 J. N. Chheda, Y. Roma´n-Leshkov and J. A. Dumesic, Green Chem.,
2007, 9, 342.
16 J. B. Binder and R. T. Raines, J. Am. Chem. Soc., 2009, 131,
11 W. F. Cooper and W. H. Nuttall, J. Chem. Soc. Trans., 1912, 101,
1074.
1979.
17 D. J. Hayes, S. W. Fitzpatrick, M. H. B. Hayes and J. R. H. Ross, The
Biofine Process–Production of Levulinic Acid, Furfural, and Formic
Acid from Lignocellulosic Feedstocks. In: B. Kamm, P. R. Gruber
and M. Kamm, (ed.), Biorefineries–Industrial Processes and Products:
Status Quo and Future Directions, Wiley-VCH, Weinheim, 2006,
vol. 1, pp. 144–160.
18 A. Regnault, J.-P. Sachetto, H. Tournier, T. Hamm and J. M.
Armanet, US Patent 4,304,608, 1981 (Chem. Abstr. 89 : 217099).
19 U. Schuchardt, I. Joekes and H. C. Duarte, J. Chem. Technol.
Biotech., 1988, 41, 51.
12 W. N. Haworth and W. G. M. Jones, J. Chem. Soc., 1944, 667.
13 It should be pointed out here that the solvolytic reactions described
in Schemes 2–5 involve the liberation of a molecule of hydrogen
chloride into the reaction medium, i.e. water (Schemes 2 and 3),
ethanol (Scheme 4), or butanol (Scheme 5). In terms of atom
economy, the hydrogen chloride produced can be recovered for
recycling back into the acid/solvent digestion reaction wherein the
biomass was first processed into CMF 1. Industrial HCl recovery
modules may be based on distillation,18 pervaporation,19 acid–base-
couple extraction,20 solvent extraction,21,22 or electrodialysis.23 In
the reactions where alcohols were both reagent and solvent, small
quantities of the corresponding chloroalkane substitution products
could be detected in the reaction mixture (by NMR and GC-MS).
Chloroethane (bp 12 ◦C) is much more volatile than the desired
product 6a (204 ◦C) and the ethanol solvent (78 ◦C). Likewise,
chlorobutane (bp 78 ◦C) is more volatile than the desired products
20 A. Baniel and A. Eyal, PCT Int. Appl., 2009(Chem. Abstr.
2009 : 1263571).
21 J. L. Gaddy and E. C. Clausen, US Patent 4,645,658 (Chem. Abstr.
106 : 140514).
22 A. V. Forster, L. E. Martz, D. E. Leng and D. Ellis, Eur. Pat. Appl.,
1980(Chem. Abstr. 94 : 176995).
23 J. Wis´niewski and G. Wis´niewska, Desalination, 1997, 109, 187.
This journal is
The Royal Society of Chemistry 2010
Green Chem., 2010, 12, 370–373 | 373
©