Angewandte
Chemie
the reaction conditions, so a longer reaction time would lead
to lower yield of the desired product. Thus, if a low catalyst
loading of 1 mol% was to be employed, the reaction
conditions would have to be optimized to maximize the
HMF yield.
ligands prevent chromium from forming multiple coordina-
tions to the NHC ligand in BMIM+ClÀ, thereby reducing the
catalytic activity as in the case of 8/CrCl2. In contrast,
inhibition effects were not observed with the addition of the
bipyridine ligand as in the case of 6/CrCl2 (HMF yield of 76%
from glucose) (Table 1, entry 15). Additionally, in the CrCl2/
EMIM system, a NHC/Cr complex could be formed under the
reaction conditions and therefore serves as a catalyst.[13]
In summary, a new NHC–Cr/ionic liquid system has been
developed for the selective conversion of sugars into HMF.
Excellent efficiencies were achieved and we attained the
highest HMF yields reported thus far for both fructose and
glucose feedstocks. The HMF yields were as high as 96% and
81% for fructose and glucose, respectively. The new system is
tolerant towards high substrate loading, and the catalyst and
ionic liquid can be recycled. HMF is provided as the sole
product isolated after simple extraction.
The substrate/solvent weight ratio was also found to be
important for the overall efficiency of the reaction system (see
Figure 2d). When the fructose/BMIM+ClÀ weight ratio was
increased from 0.05:1 to 0.2:1, the HMF yield changed slightly
from 95% to 94%. As the fructose/BMIM+ClÀ weight ratio
increased from 0.2:1 to 0.5:1, the HMF yield decreased
substantially to 70%. Additional increases in the fructose/
ionic BMIM+ClÀ weight ratio did not lead to significant
variation in the HMF yield. Remarkably, the HMF yield
remained rather unaffected (81–77%) as the glucose/
BMIM+ClÀ weight ratio was varied from 0.05 to 0.67. The
HMF yield decreased slightly (73%) when the glucose/
BMIM+ClÀ weight ratio was increased to 1.0:1. In this case,
BMIM+ClÀ assisted in the reaction rather than serve as a
solvent.
Received: July 3, 2008
Revised: September 17, 2008
Published online: October 27, 2008
The different behavior of fructose and glucose in Fig-
ure 1d suggested different reaction mechanisms for the two
feedstocks. In the latter, glucose might be converted into
fructose first and subsequently into HMF[10] over the NHC/Cr
catalyst. In this case, the fructose concentration would be
relatively low even when the glucose substrate loading was
high since fructose would merely be an intermediate in the
conversion of glucose into HMF. Interestingly, HMF yields
were approximately 14% lower for the reaction conducted in
argon versus that conducted in air (Table 1, entry 14 versus
entry 6). The NHC/Cr catalysts were also tested in dimethyl-
sulfoxide (DMSO) as the solvent and lower HMF yields were
obtained both from fructose (28–52%) and glucose (25–32%;
see Table S1 in the Supporting Information). Again, catalysts
with bulky NHC ligands showed higher efficiency in the
DMSO system.
Keywords: biomass · carbenes · carbohydrates · catalysis ·
ionic liquids
.
[1] Special issue on Sustainability and Energy, Science 2007, 315,
781.
[2] a) A. J. Ragauskas, C. K. Williams, B. H. Davison, G. Britovsek,
J. Cairney, C. A. Eckert, W. J. Frederick, Jr., J. P. Hallett, D. J.
Leak, C. L. Liotta, J. R. Mielenz, R. Murphy, R. Templer, T.
[3] a) D. L. Klass in Encyclopedia of Energy, Vol. 1 (Ed.: C. J.
Cleveland), Elsevier, Dordrecht, 2004; b) L. R. Lynd, J. H.
c) Y. Romꢀn-Leshkov, C. J. Barrett, Z. Y. Liu, J. A. Dumesic,
Nature 2007, 447, 982.
This reaction process allows recycling of NHC/Cr catalyst
and the ionic liquid. HMF was the sole product from the
diethyl ether extraction when the conversions of glucose or
fructose were conducted at temperatures below 1008C (see
the Supporting Information). After the diethyl ether extrac-
tion, the reaction medium was preheated to 1008C for 2 hours
to remove the low boiling point components, such as diethyl
ether and water, and then directly used in the next run by
adding the sugar substrate. The recycled reaction system
retained high activity in the conversion of fructose into HMF
(Table 1, entries 16–18). The HMF yield from fructose in the
recycled system was even higher than that in the system with
the fresh catalyst. This might be because of the retention of
some small amount of unreacted fructose after 6 hours of
reaction in the previous cycle. For the glucose feedstock,
HMF yields decreased gradually over multiple runs with the
recycled catalyst.
[4] For recent reviews, see: a) J. N. Chheda, G. W. Huber, J. A.
[6] a) Y. Roman-Leshkov, J. N. Chheda, J. A. Dumesic, Science
Durand, S. Razigade, J. Duhamet, P. Faugeras, P. Rivalier, P. Ros,
G. Avignon, Appl. Catal. A 1996, 145, 211; f) X. Qian, M. R.
Nimlos, M. Davis, D. K. Johnson, M. E. Himmel, Carbohydr.
217, 71; i) C. Moreau, R. Durand, S. Razigade, J. Duhamet, P.
Faugeras, P. Rivalier, P. Ros, G. Avignon, Appl. Catal. A 1996,
145, 211; j) D. Mercadier, L. Rigal, A. Gaset, J. P. Gorrichon, J.
In the CrCl2/EMIM system,[10] it was proposed that the
À
CrCl3 anion rather than NHC/CrCl2 complex acted as the
catalyst, and the reaction was suppressed with additional
ligands, such as bipyridine or glyceraldehyde. However, our
results clearly suggest that NHC/CrClx complexes play the
key role in glucose dehydration in BMIM+ClÀ. Bulky NHC
Angew. Chem. Int. Ed. 2008, 47, 9345 –9348
ꢀ 2008 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
9347