Direct catalytic synthesis of 5-Methylfurfural from biomass-derived carbohydrates

Article abstract of DOI:10.1002/cssc.201000369

5-Methylfurfural (MF), an important chemical intermediate, can be obtained directly from carbohydrates using a metal catalyst, hydroiodic acid, and hydrogen in a biphasic reaction procedure. The catalyst system is robust and can be recycled without loss of reactivity.

Full text of DOI:10.1002/cssc.201000369

DOI: 10.1002/cssc.201000369
Direct Catalytic Synthesis of 5-Methylfurfural from Biomass-Derived
Carbohydrates
Weiran Yang and Ayusman Sen*[a]
The increasing cost of fossil fuels and concerns about their en-
vironmental impact are accelerating the transition to a bio-
mass-based economy. While fuel production from biomass has
gained much attention,[1] the use of renewable resources is
also very important for the production of chemicals.[2] Typically,
Scheme 2. Two-step process for the formation of 5-methylfurfural from bio-
mass-derived carbohydrates.
chemical intermediates have a much higher value than fuels.
These intermediates, traditionally derived from fossil fuels, can
be further transformed into solvents, polymers, and specialty
chemicals. Currently, only very few compounds of commercial
interest are directly synthesized from biomass-derived carbohy-
drates by using non-enzymatic approaches.[3] We recently re-
ported a one-step catalytic process to produce 2,5-dimethylte-
trahydrofuran (DMTHF) directly from cellulosic biomass, with
high yield and selectivity.[4] DMTHF is an important chemical in-
termediate that can be used as either liquid fuel or solvent.
Other important chemicals obtained by catalytic methods in-
clude 5-hydroxymethylfurfural (HMF);[5] 5-chloromethylfurfural
(CMF);[6] and, recently reported, lactic acid derivatives[7] and g-
valerolactone.[8] Herein, we report the direct synthesis of 5-
methylfurfural (MF) from biomass-derived carbohydrates by
using a multifunctional catalyst. The catalytic system is robust,
and the catalyst can be recycled without loss of activity.
palladium as catalyst under an atmosphere of hydrogen
(Scheme 2). The yield of the hydrogenation step is up to
98%.[13] However, CMF formation from hexoses (such as fruc-
tose) requires long reaction times (ca. 10 h) and large quanti-
ties of organic solvent and surfactants, making the process
commercially unattractive.[6a] Very recently, Mascal and Nikitin
have improved the CMF synthesis step by demonstrating that
not only fructose and glucose, but also cellulose and even raw
biomass can be converted into CMF in good yields (ca. 70–
80%).[6b,14] The procedure still requires
amount of organic solvents.
a relatively large
We report a method to convert biomass-derived carbohy-
drates directly to MF in 2 h or less and with a relatively high
yield by using a biphasic reaction system (Scheme 3).
MF is a useful intermediate for the production of pharma-
ceuticals, chemicals for agriculture, perfumes, and other appli-
cations.[9] It is also a common flavoring component in the food
industry,[10] and is even considered a potential anti-tumor
agent.[11] The large-scale production of MF in industry involves
5-methylfuran, N,N-dimethylformamide, and phosphorus oxy-
chloride or phosgene (Scheme 1).[12] During this process, signif-
icant excesses of expensive and strongly poisonous phospho-
rus oxychloride and N,N-dimethylformamide are used.
Scheme 3. Direct catalytic synthesis of 5-methylfurfural from fructose.
Our results are summarized in Table 1. The reactions were
carried out in a glass-lined stainless steel autoclave. A biphasic
reaction system with water and an organic extracting solvent
(benzene or toluene) was employed; we have shown earlier
that the use of an organic extractant results in higher yields by
preventing further reaction of the desired product.[4] With
RuCl3·xH2O as catalyst, the fructose, HI, and the catalyst were
all dissolved in water before the reaction. The autoclave was
then charged with 300 psi H2 (1 psi= 6.894ꢀ103 Pa) and
heated in an oil bath. The formed MF partitioned into the or-
ganic layer as the only major product while the catalyst and
the acid were left in the aqueous solution. For product isola-
tion, the organic layer was separated and pure MF was ob-
tained by either distillation or evaporation of the organic sol-
vent. Besides small amounts of the starting material (fructose)
a few side products [such as levulinic acid (5–10%) and 2,5-
hexanedione (<5%)] were found in the aqueous layer. Some
humin was also formed (5–10%).
Scheme 1. Industrial process for 5-methylfurfural production.
Another reported method to produce MF starts from bio-
mass-derived carbohydrates (hexoses) via a two-step process.
First, 5-chloromethylfurfural (CMF) is obtained by acidic de-
composition of carbohydrates in high concentrations of chlo-
ride ion. MF is then obtained by hydrogenation of CMF with
[a] Dr. W. Yang, Prof. A. Sen
Department of Chemistry
The Pennsylvania State University
University Park, PA 16802 (USA)
Fax: (+1)814-863-5319
E-mail: asen@psu.edu
ChemSusChem 2011, 4, 349 – 352
ꢁ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
349

(entries 8 and 9) twice, with no decrease in observed catalyst
activity. With the homogenous ruthenium catalyst, the humin
formed during the reaction could be removed by filtration.
With the heterogeneous palladium on carbon catalyst, humin
separation was harder and, in the long run, the humin may de-
crease the activity of the catalyst by blocking its active sites.
Other biomass-derived carbohydrates were also tested for
this transformation. The results are shown in Table 2. With
Table 1. Catalytic conversion of fructose to 5-methylfurfural.[a]
Entry
Catalyst
Amount
[equiv%]
HI
[mmol]
T
[8C]
t
[h]
Yield
[%]
Conv.
[%]
1
2
3[b]
4[c]
5
RuCl3
RuCl3
RuCl3
RuCl3
Pd/C
Pd/C
Pd/C
Pd/C
Pd/C
Pd/C
PdCl2
RuCl3
5
5
5
5
5
1
1
1
1
1
1
1
6
3
3
3
6
6
3
3
3
3
3
3
75
90
90
90
75
75
90
90
90
90
90
90
2
1
1
1
2
2
0.5
1
1
67
64
63
63
63
63
68
69
68
59
63
15
98
97
98
97
95
94
97
96
95
93
–
6
7
8[d]
9[e]
10[f]
11[g]
12[h]
Table 2. Conversion of other carbohydrate-based biomasses to 5-methyl-
furfural.[a]
1
1
1
Carbohydrate Catalyst Amount HI
[equiv%] [mmol]
T
Yield Conv.
96
[8C] [%]
[%]
[a] Conditions: fructose (1 mmol), HI (57 wt%, in water), H2 (300 psi;
1 psi= 6.894ꢀ103 Pa), H2O (1.8 mL), benzene (2 mL). [b] Catalyst recycled
from entry 2. [c] Catalyst recycled from entry 3. [d] Catalyst recycled from
entry 7. [e] Catalyst recycled from entry 8. [f] H2 (100 psi). [g] Pd mirror
found at the end of the reaction. [h] No organic solvent was added,
water was the only solvent.
Inulin
RuCl3
RuCl3
Pd/C
Pd/C
5
5
1
1
6
6
3
75 61
75 35
105 31
96
45
81
Sucrose
Glucose
Cellulose
0.1+1 mmol NaI 115 42
<90
[a] Conditions: carbohydrate (1 mmol), catalyst (1% or 5 equiv%, relative
to the carbohydrate), HI (57 wt%, in water), H2 (300 psi), H2O (1.8 mL),
benzene (4 mL), t=2 h.
As shown in Table 1, entry 1 a 67% yield of MF can be di-
rectly obtained from fructose in 2 h at 758C. MF is not stable
under the reaction conditions; longer reaction times lead to
MF decomposition and decreased yield. When increasing the
temperature to 908C, less acid can be employed and 64% of
MF was obtained in 1 h (entry 2). Because palladium is also a
very good hydrogenation catalyst, 5% palladium on carbon
was also employed for this transformation. At 758C, a 63%
yield of MF was obtained in 2 h (entry 5). The catalyst ratio
could be decreased from 5% to 1% without any loss in the re-
activity (entries 5 and 6). At 908C, a 68% yield of MF from fruc-
tose was obtained in 0.5 h. Usually, 300 psi H2 was used; when
decreasing the hydrogen pressure to 100 psi, the yield of MF
decreased from 68% to 59% (entry 10). As shown in entry 11 a
soluble palladium salt (PdCl2) could also be employed; howev-
er, this led to the formation of a palladium mirror. Finally, and
consistent with our earlier observation,[4] omitting the organic
extractant resulted in a very low yield of MF and the formation
of a significant amount of humin (entry 12).
inulin, a polymer mainly composed of fructose and a glucose
chain end, a 61% yield of MF was obtained at 758C in 2 h.
Only 35% yield was achieved with sucrose, a dimer with one
fructose and one glucose, as starting material at 758C in 2 h.
The unreacted fraction was mainly the glucose part of sucrose
because of its significantly lower reactivity. By increasing the
temperature to 1058C, 81% conversion and 31% yield of MF
could be obtained from glucose. Longer reaction times led to
decreased yields due to MF decomposition. More humin was
obtained in this case. Cellulose, a polysaccharide consisting of
a linear chain of several hundred to >10000 b(1!4) linked d-
glucose units, required more harsh conditions for conversion.
With a combination of HI and NaI at 1158C, a 42% yield of MF
was obtained in 2 h. Thus, fructose and fructose-derived poly-
saccharides are superior to glucose and glucose-derived poly-
saccharides for MF formation under these reaction conditions.
The presence of iodide is critical to the conversion of fruc-
tose to MF. When using HCl instead of HI and reaction condi-
tions otherwise identical to Table 1, entry 7, almost no MF was
formed (Scheme 4, reaction 1) in 1 h. The products obtained
from fructose were identified as DMTHF (8%), 2,5-dimethylfur-
an (3%), 2,5-hexanedione (17%), levulinic acid (10%), and
After separation of the organic layer, the aqueous layer con-
taining the catalyst and HI could be re-used. Additional fruc-
tose and fresh organic extractant were added to the aqueous
solution and the reaction was continued. We recycled both the
ruthenium (Table 1, entries 3 and 4) and palladium catalysts
small amounts of unidentified compounds.
A significant
Scheme 4. Fructose conversion in the presence of HCl.
350
www.chemsuschem.org
ꢁ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemSusChem 2011, 4, 349 – 352

probe operating at 360.13 MHz), GC (HP Hewlett Packard-5890
series II with a FID detector; 95% dimethyl/5% diphenyl-polysil-
oxane column), and GC-MS (Waters GC-TOF with Agilent 6890 GC;
20 meter 150 mm i.d., 0.15 um 95% dimethyl/5% diphenyl-polysil-
oxane film column; 70 eV electron ionization). HNMR spectra and
the GC retention times of the products were also compared with
authentic samples.
amount of humin was formed in this case. In addition, a large
amount of cyclohexane was obtained due to hydrogenation of
the organic extractant benzene, as has been discussed previ-
ously.[4]
1
The proposed reaction pathway for the synthesis of MF from
fructose is shown in Scheme 5. It is well-known that 5-hydroxy-
methylfurfural can be obtained by acid dehydration of fruc-
GC and GC-MS analysis methods:
The initial oven temperature was
408C; the temperature was then
ramped at 38Cminꢀ1 until 1008C
was reached; after that, the tem-
perature
was
ramped
at
108Cminꢀ1 until 2008C was
reached and held for 5 min. For
Scheme 5. Reaction pathway for the conversion of fructose to 5-methylfurfural.
GC-MS analysis, the initial oven
temperature was 408C. This tem-
perature was held for 1 min, and
tose.[5] When a high concentration of Clꢀ is present, 5-chloro-
methylfurfural (CMF) is the major product.[6] In the present
case, with high concentrations of Iꢀ present, the formation of
5-iodomethylfurfural (IMF) as an intermediate is reasonable. Be-
cause iodide is a good leaving group, IMF can quickly convert
into MF under hydrogenation conditions. IMF was not ob-
served under the reaction conditions, presumably due to its in-
stability. Because chloride is a poorer leaving group than
iodide, the subsequent reaction of CMF proceeds by a different
pathway (Scheme 4, reaction 2).
subsequently ramped at a rate of 158Cminꢀ1 until 2908C was
reached, held for 7 min. The total elapsed time was 25 min. The in-
jector temperature was 2908C with a split of 20:1. The helium flow
rate was 0.5 mLminꢀ1. The transfer line temperature was 2208C.
The mass scan was 35–650 Dasꢀ1
.
Quantification methods: Product yields were determined from
1H NMR spectra and GC analysis of the organic layer, using nitro-
methane as the internal standard. The yields reported were repro-
ducible to within 2%. Conversions were calculated based on
1HNMR analysis of the aqueous layer, using DMSO as internal stan-
dard.
In conclusion, we have developed a simple method to di-
rectly produce 5-methylfurfural from biomass-derived carbohy-
drates, by using a biphasic reaction system. With fructose as
starting material, 68% MF is obtained in 0.5 h at 908C. The ho-
mogeneous catalyst RuCl3.xH2O and heterogeneous catalyst
Pd/C show similar reactivity for this transformation, and both
can be easily recycled.
Acknowledgement
We thank the US Department of Energy, Office of Basic Energy
Sciences for financial support.
Keywords: 5-methylfurfural
· biomass · carbohydrates ·
catalysis · renewable resources · sustainable chemistry
Experimental Section
Materials: Ruthenium(III) chloride hydrate was purchased from
Pressure Chemical. 5% palladium on carbon was purchased from
Johnson Matthey. 5-chloromethylfurfural was purchased from
Alchem Laboratories. All of the carbohydrates were purchased
from either Sigma–Aldrich or Alfa Aesar. Cellulose was in the
powder form with a particle diameter of approximately 20 micro-
meter. High-pressure hydrogen was obtained from GT&S, Inc. and
used without further purification. Isotopically enriched chemicals,
such as C6D6 and D2O, were obtained from Cambridge Isotope Lab-
oratories and used without further purification.
[1] Reviews and books: 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. 2007, 119, 7298–7318; Angew. Chem. Int. Ed. 2007, 46,
7164–7183; c) M. Hambourger, G. F. Moore, D. M. Kramer, D. Gust, A. L.
Moore, T. A. Moore, Chem. Soc. Rev. 2009, 38, 25–35; d) A. J. Ragauskas
et al., Science 2006, 311, 484–489; e) D. L. Klass, in Biomass for Renewa-
ble Energy, Fuels, and Chemicals, Academic Press, San Diego, 1998; f) Bi-
orefineries-Industrial Processes and Products (Eds.: B. Kamm, P. R. Gruber,
M. Kamm), Wiley-VCH: Weinheim, 2006; g) R. D. Perlack, L. L.Wright,
A. F. Turhollow, R. L. Graham, B. J. Stokes, D. C. Erbach, Biomass as Feed-
stock for a Bioenergy and Bioproducts Industry: The Technical Feasibility
of a Billion-Ton Annual Supply; Report DOE/GO-102995–102135, US De-
partment of Energy and US Department of Agriculture, Oak Ridge, TN,
2005. Available at http://www1.eere.energy.gov/biomass/pdfs/final_bil-
lionton_vision_report2.pdf.
[2] Reviews and books: a) A. Corma, S. Iborra, A. Velty, Chem. Rev. 2007,
107, 2411–2502; b) H. Danner, R. Braun, Chem. Soc. Rev. 1999, 28, 395–
405; c) J. Clark, F. Deswarte, Introduction to Chemicals from Biomass.
John Wiley & Sons, Hoboken, 2008; d) M. N. Belgacem, A. Gandini, in
Monomers, Polymers and composites from Renewable resources, Elsevier,
2008; e) T. Werpy, G. Petersen, A. Aden, J. Bozell, J. Holladay, J. White, A.
Manheim, Top Value-Added Chemicals from Biomass, Volume I: Results of
Screening for Potential Candidates from Sugars and Synthesis Gas, Report
DOE/GO-102004–101992, Office of Scientific and Technical Information,
US Department of Energy, Oak Ridge, TN, 2004; www1.eere.energy.gov/
Typical procedure for transformation of fructose to MF, and catalyst
recycling: Fructose (1 mmol, 0.18 g), HI (57 wt% in water, 3 mmol,
0.4 mL), 5% Pd/C (0.01 mmol, 20 mg), water (1.8 mL), and benzene
(2 mL) were added to a glass reaction vial in open air. The vial was
placed in a high pressure stainless steel reactor, flushed with H2,
and charged with 300 psi of H2. The reactor was finally put in an
oil bath and heated to 908C for 0.5 h. After the reaction was com-
plete, the top organic layer was directly sampled for analysis. More
starting material and fresh benzene were added to the aqueous
solution for the next cycle.
Analysis methods: Products were analyzed by 1HNMR spectroscopy
(Bruker Avance-360 spectrometer equipped with a quadnuclear
ChemSusChem 2011, 4, 349 – 352
ꢁ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemsuschem.org
351

biomass/pdfs/35523.pdf (accessed January 2011); f) J. J. Bozell, G. R. Pe-
tersen, Green Chem. 2010, 12, 539–554.
[3] G. Centi, R. A. van Santen, Catalysis for Renewables, Wiley-VCH, Wein-
heim, 2007.
[4] W. Yang, A. Sen, ChemSusChem 2010, 3, 597–603.
[5] a) Y. Romꢂn-Leshkov, J. N. Chheda, J. A. Dumesic, Science 2006, 312,
1933; b) H. Zhao, J. E. Holladay, H. Brown, Z. C. Zhang, Science 2007,
316, 1597.
[6] a) K. Hamada, T. Nagase, G. Suzukamo, US. Patent 4076733, 1979; b) M.
Mascal, E. B. Nikitin, Angew. Chem. 2008, 120, 8042–8044; Angew. Chem.
Int. Ed. 2008, 47, 7924–7926.
Org. Chem. 2009, 74, 8739–8753; c) H. Geneste, D. Sauer, W. Braje, W.
Amberg, M. Mezler, M. Bakker, M. Henrica, IPN WO 2008/145616; d) N. J.
Press, S. J. Watson, D. Porter, IPN WO 2008/148790; e) D. Mcconnell, M.
Impagnatiello, D. Kessler, O. Kraemer, S. Schneider, L. Van Der Veen, U.
Weyerczernilofsky, T. Wunberg, IPN WO 2009/112565.
[10] J. M. De Man, Principles of Food Chemistry, 3rd Ed., Springer, Heidelberg,
1999.
[11] K. Michail, V. Matzi, A. Maier, R. Herwig, J. Greilberger, H. Juan, O.
Kunert, R. Wintersteiger, Anal. Bioanal. Chem. 2007, 387, 2801.
[12] J. J. Miller, P. L. De Benneville, J. Org. Chem. 1957, 22, 1268–1269.
[13] K. Hamada, G. Suzukamo, K. Fujisawa, US Patent 4335049, 1982.
[14] M. Mascal, E. B. Nikitin, ChemSusChem 2009, 2, 859–861.
[7] M. S. Holm, S. Saravanamurugan, E. Taarning, Science 2010, 328, 602–
605.
[8] L. Deng, J. Li, D. M. Lai, Y. Fu, Q. X. Guo, Angew. Chem. 2009, 121, 6651–
6654; Angew. Chem. Int. Ed. 2009, 48, 6529–6532.
Received: October 28, 2010
Revised: December 1, 2010
[9] Selected examples: a) L. Tradtrantip, N. D. Sonawane, W. Namkung, A. S.
Verkman, J. Med. Chem. 2009, 52, 6447–6455; b) M. E. Jung, G-Y. J. Im, J.
Published online on January 26, 2011
352
www.chemsuschem.org
ꢁ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemSusChem 2011, 4, 349 – 352