Cellulose conversion into polyols catalyzed by reversibly formed acids and supported ruthenium clusters in hot water

Article abstract of DOI:10.1002/anie.200702661

(Chemical Equation Presented) A green approach to efficient conversion of cellulose into hexitols and other lighter alcohols through two steps is reported. In this process, cellulose is hydrolyzed to glucose by acids formed reversibly in situ from hot water, and the glucose is hydrogenated on Ru/C. Understanding of the green aqueous catalytic systems should lead to more efficient conversion of cellulose into fuels and chemicals.

Full text of DOI:10.1002/anie.200702661

Communications
DOI: 10.1002/anie.200702661
Sustainable Chemistry
Cellulose Conversion into Polyols Catalyzed by Reversibly Formed
Acids and Supported Ruthenium Clusters in Hot Water**
Chen Luo, Shuai Wang, and Haichao Liu*
Cellulose is the most abundant source of biomass, and holds
impressive potential as an alternative to fossil fuels for
sustainable production of fuels and chemicals.[1–5] To this end,
cellulose conversion into polyols, among the various primary
conversion routes explored to date,[1,3] is evolving as a very
viable option in terms of energy efficiency and atom economy.
The biomass-derived polyols such as sorbitol and glycerol are
being considered as new bioplatform molecules,[6–9] which can
be efficiently converted into H2, synthesis gas, alkanes, liquid
fuels, and oxygenates, for example, by the aqueous-phase
processing developed recently by Dumesic and co-work-
ers.[8,9]
acid-catalyzed reactions.[3,11] Such in situ formation of the acid
is reversible, and the protons automatically disappear at
ambient temperatures, thus leading to complete elimination
of the problems of acid recovery and waste disposal.[11]
Herein, we report an efficient conversion of cellulose into
polyols by combination of hydrolysis using H+ ions, reversibly
formed in situ in hot water, with instantaneous hydrogenation
on carbon-supported Ru clusters (Ru/C; Scheme 1).
Cellulose, a linear polymer of d-glucose with b-1,4-
glycosidic bonds, can be readily hydrolyzed by mineral acids
into glucose, which is then hydrogenated to form sorbitol and
other polyols.[1,10] However, this process is not green and
suffers from the common problems associated with the use of
liquid acids, for example, corrosion and acid recovery or
disposal. In attempts to solve these problems, Fukuoka and
Dhepe[4] recently showed that liquid acids can be replaced by
solid acids for cellulose conversion into sorbitol and mannitol,
but at relatively low yields, most likely owing to the robust
structure of crystalline cellulose and its limited accessibility to
surface acid sites. In parallel with this work, we reported a
one-step approach to cellulose conversion into polyols by
hydrogenation on soluble Ru clusters in ionic liquids,[5] but it
encounters difficulties associated with separation of the Ru
clusters and polyol products from the ionic liquids. These
problems render the two processes unfeasible for industrial
practice. For these reasons, it is apparent that hydrolysis of
cellulose by liquid acids is currently the best method,
provided that the existing acid problems can be circumvented.
It is known that liquid water at elevated temperatures
(above 473 K) can generate H+ ions capable of performing
Scheme 1. Catalytic cellulose conversion into polyols.
Ru/C catalysts were chosen in this work as a result of their
reported superior activity for glucose hydrogenation to
sorbitol.[12] As shown in Table 1, cellulose reaction occurred
rapidly on Ru/C in hot water; 38.5% conversion and 22.2%
yield of hexitols (sorbitol and mannitol at a molar ratio of
about 3.6:1) were obtained in 5 min at 518 K and 6 MPa H2,
which sharply increased to 85.5% and 39.3%, respectively,
upon prolonging the reaction time to 30 min (Table 1,
entry 1). Cellulose was also converted under these conditions
into dehydration products of sorbitan and into degradation
products of xylitol, erythritol, glycerol, propylene glycol,
ethylene glycol, and methanol, as well as trace amounts of
undesired CH4; the reaction selectivity slightly increased as
the reaction time was increased from 5 to 30 min. These
degradation products appear to originate predominantly from
the decomposition of glucose, as based on our results from
glucose and sorbitol reactions under the same conditions,
consistent with the general finding that glucose is much less
stable toward further reactions than the corresponding
polyols.[10] This result suggests that high selectivity for hexitols
[*] C. Luo, S. Wang, Prof. Dr. H. Liu
Beijing National Laboratory for Molecular Sciences
State Key Laboratory for Structural Chemistry of Stable and
Unstable Species, Green Chemistry Center
College of Chemistry and Molecular Engineering
Peking University
Beijing 100871 (China)
Fax: (+86)10-6275-4031
E-mail: hcliu@pku.edu.cn
[**] This work was supported by the National Basic Research Project of
China (No. 2006CB806100) and NSFC (Grant Nos. 20673005 and
20573004). This work was also supported in part by the Program for
New Century Excellent Talents in University,(NECT-05-0010), State
Education Ministry.
Supporting information for this article is available on the WWW
under http://www.angewandte.org or from the author.
7636
ꢀ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2007, 46, 7636 –7639

Angewandte
Chemie
Table 1: Cellulose conversion and selectivity at 518 K.[a]
Entry
Conversion Hexitol
Selectivity [%]
[%]
yield [%]
Sorbitol Mannitol Sorbitan Xylitol Erythritol Glycerol Propylene Ethylene CH3OH CH4
glycol
glycol
1[a]
2[a]
3
H2O+4 wt%
Ru/C
H2O or H2O+C
38.5
22.2
(39.3)
–
45.2
(34.6)
–
12.6
(11.4)
–
11.1
(13.4)
–
4.2
5.9
5.3
(7.5)
–
5.9
(7.9)
–
5.3
(6.6)
–
3.5
(4.0)
–
1.1
(1.5)
–
(85.5)[b]
38.6
(6.0) (7.1)
–
–
–
–
(87.5)[b]
C2H5OH or diox- 0 (0)
ane +
–
–
–
–
–
–
–
–
–
4 wt% Ru/C
4
25 mL
10.2
6.4
43.9
18.9
19.1
3.1
3.2
2.1
2.3
3.2
3.2
1.0
C2H5OH+25 mL
H2O+4 wt%
Ru/C
5
6
7
H2O+8 wt%
Ru/C
H2O+2 wt%
Ru/C
H2O+1 wt%
Ru/C
37.9
38.2
38.1
21.1
23.2
13.7
43.7
47.4
24.2
12.0
13.4
11.8
11.3
11.4
–
4.1
4.1
2.9
5.9
5.5
–
6.6
4.7
–
6.0
4.1
–
5.9
5.0
–
3.3
3.1
–
1.2
1.3
–
[a] 5 min reaction time, 6 MPa H2, 50 mL H2O, 1 g cellulose, 0.04 mmol Ru. [b] Data in parentheses refer to a reaction time of 30 min.
necessitates immediate hydrogenation of glucose once it is
formed from cellulose hydrolysis.
Such hydrolysis did not lead to any changes in the
cellulose crystal structure as evidenced from the identical
XRD patterns in Figure S1 of the Supporting Information,
which show that the native cellulose I structure[13] remained
intact after the reactions even at 85.5% cellulose conversion
(at 518 K for 30 min). This result indicates that cellulose
hydrolysis occurs at its crystal surface, with no swelling or
dissolution, which otherwise would lead to formation of the
cellulose II crystal form, as generally observed under near-
supercritical or supercritical conditions, which require much
higher temperatures and pressures (593–673 K and
25 MPa).[3]
Recycling the Ru/C catalyst over five runs did not lead to
any significant decline in cellulose conversion and selectivity
(see the Supporting Information, Figure S2). Analysis of the
aqueous reaction solutions by ICP after each cycle showed no
detectable leaching of Ru. Characterization of the catalyst by
TEM (Figure 1) showed no essential change in the mean
diameters of the Ru clusters and their size distributions after
the five reaction cycles. These results demonstrate that the
catalyst is stable and reusable under the conditions in this
work.
To confirm the in situ formation of acid and its role in the
Figure 1. TEM images and histograms of Ru particle size distribution
of the Ru/C catalyst (4 wt%Ru) before (a) and after (b) five cycles of
cellulose reaction at 518 K.
aqueous system, several experiments were performed with no
Ru/C catalyst or no water. Similar cellulose conversions
(about 38.6% and 87.5% after 5 min and 30 min, respec-
tively) were obtained in water in the absence of Ru/C
(Table 1, entry 2), but such conditions led to brown solutions
(versus colorless solutions in the presence of Ru/C), and coke-
like precipitates apparently formed from the acid-catalyzed
condensation reactions of the primary product glucose and its
derivatives (Table 1). Substitution of either protic solvents
(for example, ethanol) or aprotic solvents (for example,
dioxane) for water did not give any cellulose conversion on
Ru/C under similar conditions (Table 1, entry 3). However,
addition of water to ethanol (1:1 volume ratio) led to a 10.2%
cellulose conversion with 62.8% selectivity for hexitols in
5 min under similar conditions (Table 1, entry 4). This lower
conversion compared to that with the pure water system
reflects the lower dielectric constant of the mixed solvent
system and in turn its lower acidity.[11] Taken together, these
Angew. Chem. Int. Ed. 2007, 46, 7636 –7639
ꢀ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org
7637

Communications
results demonstrate that water is required for in situ forma-
increased in the range 478–513 K, and then decreased to
44.5% at 533 K. Selectivity for the degradation alcohol
products monotonically increased from 14.6% to 44.2%,
while selectivity for sorbitan decreased from 19.8% to 10.2%
upon increasing the temperature from 478 K to 533 K. It is
noteworthy that selectivity for CH4 was as low as about 1%
even at the highest temperature (533 K) after 20 min, and
these reaction conditions give rise to a 100% conversion of
cellulose.
These results show that our approach can lead to almost
complete conversion of cellulose into hexitols and other
useful alcohol products, including sorbitan and methanol.
Such aqueous alcohol solutions can be directly converted into
H2 and synthesis gas, among other products, by an aqueous-
phase reforming (APR) process,[1,8,9b] in which it is reported
that selectivity for H2 production is higher from lighter
alcohols like ethylene glycol and methanol than from heavier
ones like sorbitol.[8] Therefore, it appears that our approach is
suitable for combination with the APR process for production
of H2 and synthesis gas directly from cellulose. Nonetheless,
such mixed-alcohol solutions are not desirable as feedstocks
for the synthesis of pure chemicals. For this purpose, our
current studies are focused on increasing the product
selectivity (to 100% hexitol) by designing more selective
and efficient hydrogenation catalysts and by choosing optimal
reactor configurations and reaction conditions.
tion of the acid that effects the cellulose hydrolysis step
responsible for determining the cellulose conversion, and
Ru/C is required for instantaneous hydrogenation of the
hydrolyzed product glucose to form polyols (Scheme 1)
instead of the above-mentioned condensation products or
cokes.
It was indeed found that the product distributions depend
on the activities of the hydrogenation catalysts. When the Ru
loading was changed from 8% to 2%, the combined hexitol
selectivity increased slightly from 55.7% to 60.8% with a
concurrent decrease in the combined selectivity for the
degradation alcohol products from 31.8% to 26.5%
(Table 1, entries 1, 5, and 6). Further decreasing the Ru
loading to 1% led to a much lower hexitol selectivity of
36.0% with a large fraction of unsaturated products (Table 1,
entry 7) that, although not yet identified, were found to
=
=
contain C O and C C bonds as tested by Fehlingꢀs and
KMnO4 solutions. These results agree well with the perform-
ances in glucose hydrogenation carried out by loading glucose
as the reactant under the same conditions. TEM character-
ization shows that these catalysts have narrow unimodal size
distributions of Ru particles with diameters of 1.5, 3.2, 3.7, and
4.2 nm from 1% to 8% loading (see the Supporting
Information, Figures S3–S5), thus indicating that Ru particle
size for the glucose hydrogenation is important, as generally
found with nanoparticle catalysts.[14]
In conclusion, we have presented a green approach to
efficient conversion of cellulose into hexitols together with
other lighter polyols through two steps: cellulose hydrolysis to
glucose by acids that are reversibly formed in situ from hot
water and subsequent glucose hydrogenation by supported
Ru clusters. Further advances in understanding these green
aqueous catalytic systems will lead to rational control of the
polyol distributions, and thus to efficient conversion of
cellulose into renewable fuels and chemicals.
We have also explored the effects of reaction temperature
on cellulose conversion and selectivity. As shown in Figure 2,
Experimental Section
Ru/C catalysts were prepared by impregnating activated carbon with
acetone solutions of RuCl3, subsequent drying at 393 K, and then
reduction at 673 K in a 20% H2 flow. Cellulose (Alfa Aesar,
microcrystalline; relative crystallinity of about 84%, as estimated
from its XRD pattern shown in Figure S1a in the Supporting
Information, according to the method reported in reference [3a])
reactions were carried out in a teflon-lined stainless steel autoclave
(150 mL) typically at 518 K and 6 MPa H2 for 5 min with vigorous
stirring. Liquid-phase products were analyzed by HPLC and ESI-MS,
and gas-phase products were analyzed by GC. Cellulose conversions
were determined by the change in the weight of cellulose loaded
before and after the reactions, and selectivities are reported on a
carbon basis; carbon mass balance is better than 98 Æ 3% in this work.
XRD patterns for the cellulose were measured on a Rigaku
D/Max-2000 diffractometer using CuKa radiation (l = 1.5406 Š),
operated at 30 kV and 100 mA, in the range of 10–408. TEM images
for the Ru/C catalysts were taken on a Philips Tecnai F30 FEG-TEM
operated at 300 kV. The samples were prepared by uniformly
dispersing Ru/C catalysts in ethanol and then placing them onto
carbon-coated copper grids. The average size of the Ru particles and
Figure 2. Cellulose conversion and selectivity on Ru/C (4 wt%Ru) as a
function of reaction temperature in the range 478–533 K after 5 min
(6 MPa H2, 50 mL H2O, 1 g cellulose, 0.04 mmol Ru). Left-hand axis:
&
^
conversion ( ); right-hand axis: selectivity for hexitols ( ), for C1–C5
~
*
alcohols ( ), for sorbitan ( ), and for CH4 (”).
cellulose conversion on 4 wt% Ru/C in water increased
sharply from 5.6% to 83.1% upon increasing the temperature
from 478 K to 533 K. This result is consistent with the stronger
acidity at higher reaction temperatures,[15] which are required
for cellulose hydrolysis, as also reflected by the disappearance
of the partially hydrolyzed products cellobiose and cellotriose
(not included in Figure 2) above 503 K. Hexitol selectivity
increased slowly from 55.1% to 61.1% as the temperature
7638
www.angewandte.org
ꢀ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Angew. Chem. Int. Ed. 2007, 46, 7636 –7639

Angewandte
Chemie
their size distributions were calculated by averaging of at least 200
particles randomly distributed in TEM images.
[8] a) R. D. Cortright, R. R. Davda, J. A. Dumesic, Nature 2002,
418, 964 – 967; b) G. W. Huber, J. W. Shabaker, J. A. Dumesic,
Science 2003, 300, 2075 – 2077; c) G. W. Huber, J. A. Dumesic,
Cata. Today 2006, 111, 119 – 132.
Received: June 18, 2007
Revised: July 8, 2007
Published online: August 31, 2007
[9] a) G. W. Huber, R. D. Cortright, J. A. Dumesic, Angew. Chem.
2004, 116, 1575 – 1577; Angew. Chem. Int. Ed. 2004, 43, 1549 –
1551; b) R. R. Soares, D. A. Simonetti, J. A. Dumesic, Angew.
Chem. 2006, 118, 4086 – 4089; Angew. Chem. Int. Ed. 2006, 45,
3982 – 3985.
[10] a) J. M. Robinson, C. E. Burgess, M. A. Bently, C. D. Brasher,
B. O. Horne, D. M. Lillard, J. M. Macias, H. D. Mandal, S. C.
Mills, K. D. OꢀHara, J. T. Pon, A. F. Raigoza, E. H. Sanchez, J. S.
Villarreal, Biomass Bioenergy 2004, 26, 473 – 483; b) V. I.
Sharkov, Angew. Chem. 1963, 75, 805 – 812; Angew. Chem. Int.
Ed. Engl. 1963, 2, 405 – 409.
[11] a) T. S. Chamblee, R. R. Weikel, S. A. Nolen, C. L. Liotta, C. A.
Eckert, Green Chem. 2004, 6, 382 – 386; b) S. A. Nolen, C. L.
Liotta, C. A. Eckert, R. Gläserb, Green Chem. 2003, 5, 663 – 669.
[12] a) B. Kusserow, S. Schimpf, P. Claus, Adv. Synth. Catal. 2003, 345,
289 – 299; b) B. W. Hoffer, E. Crezee, P. R. M. Mooijman, A. D.
Langeveld, F. Kapteijn, J. A. Moulijn, Catal. Today 2003, 79, 35 –
41.
[13] a) S. Y. Oh, D. I. Yoo, Y. Shin, H. C. Kim, H. Y. Kim, Y. S.
Chung, W. H. Park, J. H. Youk, Carbohydr. Res. 2005, 340, 2376 –
2391; b) H. L. Chen, A. Yokochi, J. Appl. Polym. Sci. 2000, 76,
1466 – 1471.
Keywords: alcohols · cellulose · heterogeneous catalysis ·
.
hydrogenation · hydrolysis
[1] G. W. Huber, S. Iborra, A. Corma, Chem. Rev. 2006, 106, 4044 –
4098.
[2] D. Klemm, B. Heublein, H. Fink, A. Bohn, Angew. Chem. 2005,
117, 3422 – 3458; Angew. Chem. Int. Ed. 2005, 44, 3358 – 3393.
[3] a) M. Sasaki, T. Adschiri, K. Arai, AIChE J. 2004, 50, 192 – 202;
b) M. Sasaki, Z. Fang, Y. Fukushima, T. Adschiri, K. Arai, Ind.
Eng. Chem. Res. 2000, 39, 2883 – 2890; c) S. Saka, T. Ueno,
Cellulose 1999, 6, 177 – 191.
[4] A. Fukuoka, P. L. Dhepe, Angew. Chem. 2006, 118, 5285 – 5287;
Angew. Chem. Int. Ed. 2006, 45, 5161 – 5163.
[5] N. Yan, C. Zhao, C. Luo, P. J. Dyson, H. Liu, Y. Kou, J. Am.
Chem. Soc. 2006, 128, 8714 – 8715.
[6] A. J. Ragauskas, C. K. Williams, B. H. Davison, G. Britovsek, J.
Cairney, C. A. Eckert, W. J. Frederick, J. P. Hallett, D. J. Leak,
C. L. Liotta, J. R. Mielenz, R. Murphy, R. Templer, T. Tscha-
plinski, Science 2006, 311, 484 – 489.
[7] Y. Shen, S. Wang, C. Luo, H. Liu, Prog. Chem. 2007, 19, 431 –
436.
[14] A. T. Bell, Science 2003, 299, 1688 – 1691.
[15] A. V. Bandura, S. N. Lvov, J. Phys. Chem. Ref. Data 2006, 35, 15 –
30.
Angew. Chem. Int. Ed. 2007, 46, 7636 –7639
ꢀ 2007 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.angewandte.org 7639