Hydrolysis of cellulose into glucose by magnetic solid acid

Article abstract of DOI:10.1002/cssc.201000300

The main attraction: A magnetic solid acid with mesoporous structure was synthesized for the hydrolysis cellulose into glucose. Glucose is generated efficiently from amorphous cellulose in the mesopores of the catalyst with a yield of 50 %. Moreover, catalyst separation can be readily achieved by magnetic force.

Full text of DOI:10.1002/cssc.201000300

DOI: 10.1002/cssc.201000300
Hydrolysis of Cellulose into Glucose by Magnetic Solid Acid
Da-ming Lai,[a] Li Deng,[a] Jiang Li,[a] Bing Liao,[b] Qing-xiang Guo,[a] and Yao Fu*[a]
Cellulose is the major component of plant biomass, which is
readily available and does not compete with the food supply.
Hydrolysis of cellulose as the entry point of biorefinery
schemes is an important process for chemical and biochemical
industries based on sugars, especially for fuel ethanol produc-
tion.[1] However, cellulose not only provides a renewable
carbon source, but also offers challenges to researchers due to
the structural recalcitrance.[1,2] Considerable efforts have been
devoted to the study of hydrolysis of cellulose by enzymes,
acids, and supercritical water.[3] Recently, acid-catalyzed hydrol-
ysis has attracted increasing attention. To overcome resistance
to degradation through breaking hydrogen bonds and b-1,4-
glycosidic bonds, ionic liquids have been employed to form
homogeneous solutions of cellulose prior to hydrolysis.[4] Al-
though the homogeneous hydrolysis can be carried out under
mild conditions with a high glucose yield, the workup for sepa-
ration of sugars, dehydrated products, and unreacted cellulose
from the ionic liquid is normally difficult.[1,4c,5] Hydrolysis of
cellulose in diluted acids has been practiced for a long time.
However, acid-waste generation and corrosion hazards are sig-
nificant drawbacks of this process. To move towards more en-
vironmentally sustainable approaches, Onda et al. demonstrat-
ed that sulfonated activated carbon can convert amorphous
ball-milled cellulose into glucose with a yield of 41%.[6] Almost
simultaneously, Hara et al. completely hydrolyzed cellulose to
water soluble b-1,4-glucans at 1008C using a more robust sul-
fonated activated carbon catalyst.[7] Hydrolysis of cellobiose
and cellulose catalyzed by layered niobium molybdate was
achieved by Takagaki et al., athough the yield of glucose from
cellulose was low.[8] Fukuoka et al. found that mesoporous-
carbon-supported Ru catalysts were also able to catalyze the
hydrolysis of cellulose into glucose.[9] More recently, sulfonated
silica/carbon cellulose hydrolysis catalysts were synthesized by
Jacobs et al., affording glucose in 50% yield.[10] Zhang et al.
employed sulfonated carbon with mesoporous structure for
hydrolysis of cellulose, giving a glucose yield of 75%, which is
the highest recorded yield on a solid acid.[11]
from the solid residues. Although cellulose can be converted
almost completely in some cases,[7,11] lignin components can
not be converted and humins are formed sometimes as solid
residues. Secondly, to achieve a high yield of glucose, the
reaction was usually conducted at low cellulose/liquid ratio
(ca. 1:100).[6,8–11] However, concentration of the glucose solu-
tion, prior to the production of ethanol or other compounds, is
energy-consuming. Thus effective treatment of high cellulose
loadings is required. In view of the importance of the cellu-
lose/liquid ratio, Jacobs et al. carried out hexitol production
from concentrated cellulose with heteropoly acid and Ru/C.[12]
We designed and synthesized a magnetic solid acid catalyst
for the hydrolysis of cellulose at high cellulose/liquid ratio
(1:10 or 1:15; Scheme 1). Sulfonic acid-functionalized mesopo-
Scheme 1. Magnetic solid acid-catalyzed hydrolysis of cellulose to form
glucose.
rous silica materials are excellent acid catalysts.[13] Inspired by
Stucky et al.[14], we prepared the magnetic solid acid catalyst
through a modified direct synthetic method involving the co-
condensation of tetraethoxysilane (TEOS) and mercaptopropyl
trimethoxysilane (MTPMS) and oxidation of mercapto groups
in the presence of magnetic Fe3O4 nanoparticles (MNPs), tri-
block copolymers and hydrogen peroxide. To our knowledge,
the sulfonic group functionalized magnetic SBA-15 catalyst
(Fe3O4–SBA–SO3H) is the first magnetic solid acid with ordered
mesoporous structure. By combining MNPs and SBA-15 meso-
porous structure, the solid acid is not only separable but also
durable under hydrothermal conditions which are usually em-
ployed for hydrolysis of cellulose. In the presence of Fe3O4–
SBA–SO3H, b-1,4-glycosidic bonds were hydrolyzed effectively
and the molar yield of glucose from amorphous cellulose
reached 50%. This catalyst was also highly active for starch
and corn cob feeds. More importantly, the catalyst could be
easily separated from the reaction system by magnetic force
and undergo repeated use without deactivation.
Considering the real biomass components and practical pro-
cess for glucose production, two challenges remain for a cata-
lytic system. Firstly, solid catalysts should be readily separated
[a] D.-m. Lai, L. Deng, J. Li, Prof. Q.-x. Guo, Prof. Dr. Y. Fu
Anhui Province Key Laboratory of Biomass Clean Energy
Department of Chemistry, University of Science and Technology of China
Hefei 230026, China
Fax: (+86)551-3606689
E-mail: fuyao@ustc.edu.cn
[b] B. Liao
Key Laboratory of Cellulose and Lignocellulosics Chemistry
Guangzhou Institute of Chemistry, Chinese Academy of Sciences
Guangzhou 510650, China
Figure 1 shows the small-angle and high-angle XRD patterns
of functionalized SBA-15 samples. Similar to pure SBA-15,
Supporting information for this article is available on the WWW under
http://dx.doi.org/10.1002/cssc.201000300.
ChemSusChem 2011, 4, 55 – 58
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
55

Figure 1. XRD patterns of solid acid catalysts.
Fe3O4–SBA–SO3H and SBA–SO3H also exhibit a peak at 2q
ꢀ0.6–0.88, corresponding to (100) diffraction. This peak indi-
cates that the ordered mesoporous structure was retained in
the acid catalysts. The peak intensity reveals that the degree of
ordering decreased due to the embedded Fe3O4 MNPs. Peaks
corresponding to (110) and (200) diffractions were eliminated
from the patterns of the acid catalysts by the addition of
MTPMS and hydrogen peroxide. At high angles, the typical dif-
fraction peaks of pure stoichiometric magnetite (Fe3O4) were
detected. The peaks at 2q=18–288 reveal the amorphous silica
matrix in Fe3O4–SBA–SO3H.
Figure 2. TEM images of fresh (a, b and c) and spent (d) Fe3O4–SBA–SO3H
catalysts.
SO3H. Compared to the performance of H2SO4 with the same
amount of H+, Fe3O4–SBA–SO3H exhibited much higher effi-
ciency. The conversion of cellobiose at 1208C for 1 h was 98%
or 54% in the presence of Fe3O4–SBA–SO3H or 0.05m H2SO4,
respectively. After increasing the concentration of H2SO4 to
0.11m to give the same amount of sulfates as sulfonic groups
in Fe3O4–SBA–SO3H, the solid acid catalyst was still superior to
H2SO4 (Table 1, entries 1–3). Regarding the pore volume and
amount of acid sites, the channels in Fe3O4–SBA–SO3H bearing
concentrated acids, about 2.2m acid sites in channels (amount
of acid/pore volume), may have facilitated the hydrolysis of
b-1,4-glycosidic bonds.
According to N2 adsorption–desorption isotherms, the aver-
age pore diameter of SBA–SO3H was 4.7 nm, the average pore
volume was 0.56 cm3 gÀ1 and the BET surface area was
603 m2 gÀ1. These values for Fe3O4–SBA–SO3H were 4.8 nm,
0.49 cm3 gÀ1, and 464 m2 gÀ1. The above results suggest that
Fe3O4 MNPs did not occupy or block the available intrachannel
space, allowing oligosaccharides to enter the channels and
contact with acid sites. (see the Supporting Information,
Figure S1, Table S1).[15,16]
When amorphous cellulose was used as feed, the yield of
glucose decreased considerably to 50%. This result can be
attributed to the poor solubility, which makes the acid sites in-
accessible to b-1,4-glycosidic bonds on the cellulose chains. To
yield glucose, cellulose had to be hydrolyzed to soluble glu-
cose or oligosaccharides in water by the protons diffused from
the SBA-15 channels; oligosaccharides then underwent hydrol-
ysis to glucose, as cellobiose does. The former step taking
place on the surface of cellulose is much more difficult than
the latter, which occurs in solution. Furthermore, the yield of
glucose increased with the amount of acid in SBA–SO3H
(Table 1, entries 5–7).
TEM micrographs of Fe3O4–SBA–SO3H clearly show that
Fe3O4 MNPs were embedded in the ordered silica matrix over
large domains (Figure 2a). In good agreement with the XRD
results, the images along the (001) and (100) directions also
indicate the ordered hexagonal mesoporous structure in
Fe3O4–SBA–SO3H (Figure 2b,c).
The XPS spectrum (see the Supporting Information,
Figure S3) exhibits only one peak (169.01 eV) in the binding
energy range 158–177 eV, which demonstrates that all mer-
capto groups were oxidized to sulfonic groups by hydrogen
peroxide. This finding is consistent with the results of Stucky
et al.[14]
In terms of the barriers for hydrolysis, such as crystallinity,
surface area, hydrogen bonds, and b-1,4-glycosidic bonds, mi-
crocrystalline cellulose is the most stable carbohydrate under
acidic conditions.[4c] When microcrystalline cellulose was hydro-
lyzed, the maximum yield of glucose was 26%, which is close
to the content of the amorphous region in the feed (degree of
crystallinity is about 75%). Compared to the amorphous
region, the microcrystalline region was densely packed forming
an ordered matrix that inhibited the contact of acids with b-
We initially tested the catalysts to convert b-1,4-glycosidic
bonds by using as a feed cellobiose, which is commonly used
as model compound for cellulose. The molar yield of glucose
reached 96%, indicating that the efficient hydrolysis of b-1,4-
glycosidic bonds was achieved in the presence of Fe3O4–SBA–
56
www.chemsuschem.org
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemSusChem 2011, 4, 55 – 58

Table 1. Hydrolysis of carbohydrates to glucose or total reducing sugars.
Entry[a]
Substrate
Catalyst
Amount of acid sites [mmolgÀ1
]
Yield [mol%]
1[b]
2
3
4
5
6
7
8
cellobiose (1.0 g)
cellobiose (1.0 g)
cellobiose (1.0 g)
Fe3O4–SBA–SO3H
0.05 M H2SO4
0.11 M H2SO4
Fe3O4–SBA–SO3H
SBA–SO3H
1.09
–
–
96 (98)[c]
32 (54)[c]
36 (60)[c]
50
24
31
52
amorphous cellulose (1.0 g)
amorphous cellulose (1.0 g)
amorphous cellulose (1.0 g)
amorphous cellulose (1.5 g)
microcrystalline cellulose (1.5 g)
microcrystalline cellulose (1.5 g)
microcrystalline cellulose (1.5 g)
microcrystalline cellulose (1.5 g)
microcrystalline cellulose (1.5 g)
microcrystalline cellulose (1.5 g)
starch (1.5 g)
1.09
0.44
0.56
1.50
1.09
1.09
0.72
4.40
0.05
0
SBA–SO3H
SBA–SO3H
Fe3O4–SBA–SO3H
Fe3O4–SBA–SO3H
AC–SO3H
Amberlyst-15
g-Al2O3
SBA-15
Fe3O4–SBA–SO3H
Fe3O4–SBA–SO3H
26
9
12 (42)[d]
21
10
11
12
13
14
15
15
3
<1
95
45
1.09
1.09
corn cob (1.5 g)
[a] Reaction conditions: Solid acid catalyst (1.5 g), carbohydrate, H2O (15 mL), T=1508C, t=3 h. [b] 1.5 g solid acid catalyst, 15 mL H2O at 1208C for 1 h.
[c] The value in parentheses is the conversion of cellobiose. [d] At 1508C for 12 h; the value in parentheses is the yield of levulinic acid.
1,4-glycosidic bonds in the bulk phase. However, this finding
does not mean that Fe3O4–SBA–SO3H can not decompose cel-
lulose. Prolonging reaction time gave rise to levulinic acid
(42% yield at t=12 h), which is a main product of acid-cata-
lyzed dehydration of hexose. The overall yield of glucose and
levulinic acid reached 54%, indicating that the crystalline
region of cellulose was also partially hydrolyzed and that fur-
ther decomposition of glucose at high temperature lowered
the selectivity for glucose.
sive than H2SO4. Therefore, it is of importance to establish a
catalyst separation procedure for the hydrolysis of cellulose.
Used Fe3O4–SBA–SO3H was easily separated from the resulting
mixture in a magnetic field and, after a simple regeneration
procedure, can be used repeatedly without deactivation (Fig-
ure 3a). The yields of glucose from amorphous cellulose were
48% and 49% in the second and third recycling runs. The levu-
linic acid was generated as by-product with a yield of 4%. Ac-
Other solid catalysts were also tested for hydrolysis of micro-
crystalline cellulose. Sulfonated active carbon (ACÀSO3H) and
macroporous resin Amberlyst 15 can also decompose cellulose,
giving 21% and 15% yields of glucose, respectively. Although
they also possess sulfonic groups, their performances were not
as good as that of Fe3O4–SBA–SO3H. Two catalysts without sul-
fonic groups, g-Al2O3 and pure SBA-15, showed lower activities
for the hydrolysis of microcrystalline cellulose.
To demonstrate that Fe3O4–SBA–SO3H is capable of hydrolyz-
ing lignocellulosic biomass, a typical biomass corn cob was
employed. The molar yield of total reducing sugars was 45%,
based on the contents of cellulose and hemicellulose
(45 wt%+35 wt%) in corn cob.[17] Notably, trace amounts of
dehydrated products, such as levulinic acid and furfural with
concentrations of 1.7 gLÀ1 and 0.9 gLÀ1, respectively, were
generated by this method. Ethanolic fermentation seemed un-
inhibited by such low byproduct concentrations.[18] Despite the
fermentation process, the chemical applications are also attrac-
tive. In principle, solid acid-catalyzed hydrolysis may afford glu-
cose as a feed or intermediate for the syntheses of sorbitol,[5,12]
HMF,[19] levulinic acid, and g-valerolactone.[20] Moreover,
efficient hydrolysis of starch can also be achieved by Fe3O4–
SBA–SO3H, giving 95% glucose.
In the heterogeneous reaction system, full liquefaction of
cellulosic material is not commonly achieved, which has re-
stricted solid catalyst recovery. Furthermore, the amount of
solid catalysts used for hydrolysis is rather high, even more
than the amount of cellulose and they are much more expen-
Figure 3. a) Catalytic performance over three cycles; b) separation of the
magnetic solid acid.
ChemSusChem 2011, 4, 55 – 58
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.chemsuschem.org
57

cording to the TEM micrograph (Figure 2d), the Fe3O4–SBA–
SO3H used three times still contained Fe3O4 MNPs and ordered
mesoporous channels are also preserved. The XRD pattern ex-
hibits that Fe3O4 MNPs in the used catalyst were presented in
the state of stoichiometric magnetite as before (see the Sup-
porting Information, Figure S1). Based on N2 adsorption–de-
sorption isotherms, the BET surface area was 346 m2 gÀ1. The
average pore diameter was 4.0 nm, indicating that the ordered
mesoporous structure was retained (Table S1). The above re-
sults reveal that Fe3O4–SBA–SO3H is tolerant to the employed
hydrothermal conditions.
Keywords: biomass · heterogeneous catalysis · hydrolysis ·
magnetic nanoparticles · solid acids
[1] R. Rinaldi, F. Schuth, ChemSusChem 2009, 2, 1096.
[2] M. E. Himmel, S. Y. Ding, D. K. Johnson, W. S. Adney, M. R. Nimlos, J. W.
Brady, T. D. Foust, Science 2007, 315, 804.
[3] a) Y. H. P. Zhang, L. R. Lynd, Biotechnol. Bioeng. 2004, 88, 797; b) G. W.
Huber, S. Iborra, A. Corma, Chem. Rev. 2006, 106, 4044; c) M. Sasaki, Z.
Fang, Y. Fukushima, T. Adschiri, K. Arai, Ind. Eng. Chem. Res. 2000, 39,
2883.
In summary, both efficient hydrolysis of concentrated cellu-
lose to glucose and easy separation of catalyst were achieved
by using solid acid Fe3O4–SBA–SO3H with ordered mesopores
and MNPs. The sulfonic acids in channels were found to be suf-
ficiently robust to hydrolyze 1,4-glycosidic bonds. In the pres-
ence of Fe3O4–SBA–SO3H, cellobiose gave glucose in 96%
yield, whereas amorphous cellulose was hydrolyzed to glucose
with a medium yield. Owing to its high hydrothermal stability,
the catalyst can be used repeatedly without obvious deactiva-
tion. Furthermore, Fe3O4–SBA–SO3H can be a powerful tool for
the chemical transformation of cellulose into valuable chemi-
cals, including sorbitol, HMF, g-valerolactone, and, especially,
levulinic acid. The optimization of hydrolysis and dehydration
for levulinic acid production are underway.
[4] a) C. Z. Li, Z. K. B. Zhao, Adv. Synth. Catal. 2007, 349, 1847; b) C. Z. Li, Q.
Wang, Z. K. Zhao, Green Chem. 2008, 10, 177; c) R. Rinaldi, R. Palkovits,
F. Schꢂth, Angew. Chem. 2008, 120, 8167; Angew. Chem. Int. Ed. 2008,
47, 8047; d) J. B. Binder, R. T. Raines, Proc. Natl. Acad. Sci. USA 2010, 107,
4516.
[5] R. Palkovits, K. Tajvidi, J. Procelewska, R. Rinaldi, A. Ruppert, Green
Chem. 2010, 12, 972.
[6] a) A. Onda, T. Ochi, K. Yanagisawa, Green Chem. 2008, 10, 1033; b) A.
Onda, T. Ochi, K. Yanagisawa, Top. Catal. 2009, 52, 801.
[7] a) S. Suganuma, K. Nakajima, M. Kitano, D. Yamaguchi, H. Kato, S. Haya-
shi, M. Hara, J. Am. Chem. Soc. 2008, 130, 12787; b) M. Kitano, D. Yama-
guchi, S. Suganuma, K. Nakajima, H. Kato, S. Hayashi, M. Hara, Langmuir
2009, 25, 5068; c) D. Yamaguchi, M. Kitano, S. Suganuma, K. Nakajima,
H. Kato, M. Hara, J. Phy. Chem. C 2009, 113, 3181.
[8] A. Takagaki, C. Tagusagawa, K. Domen, Chem. Commun. 2008, 5363.
[9] H. Kobayashi, T. Komanoya, K. Hara, A. Fukuoka, ChemSusChem 2010, 3,
440.
[10] S. Van de Vyver, L. Peng, J. Geboers, H. Schepers, F. de Clippel, C. J.
Gommes, B. Goderis, P. A. Jacobs, B. F. Sels, Green Chem. 2010, 12, 1560.
[11] J. F. Pang, A. Q. Wang, M. Y. Zheng, T. Zhang, Chem. Commun. 2010, 46,
6935.
[12] J. Geboers, S. Van de Vyver, K. Carpentier, K. de Blochouse, P. Jacobs, B.
Sels, Chem. Commun. 2010, 46, 3577.
[13] W. M. Van Rhijn, D. E. De Vos, B. F. Sels, W. D. Bossaert, P. A. Jacobs,
Chem. Commun. 1998, 317.
[14] D. Margolese, J. A. Melero, S. C. Christiansen, B. F. Chmelka, G. D. Stucky,
Chem. Mater. 2000, 12, 2448.
[15] K. C. Souza, G. Salazar-Alvarez, J. D. Ardisson, W. A. Macedo, E. M. B.
Sousa, Nanotechnology 2008, 19.
[16] R. Rinaldi, F. Schꢂth, Energy Environ. Sci. 2009, 2, 610.
[17] Y. Sun, J. Y. Cheng, Bioresour. Technol. 2002, 83, 1.
[18] S. Larsson, E. Palmqvist, B. H. Hagerdal, C. Tengborg, K. Stenberg, G.
Zacchi, N. O. Nilvebrant, Enzyme Microb. Technol. 1999, 24, 151.
[19] a) H. B. Zhao, J. E. Holladay, H. Brown, Z. C. Zhang, Science 2007, 316,
1597; b) J. B. Binder, R. T. Raines, J. Am. Chem. Soc. 2009, 131, 1979;
c) H. P. Yan, Y. Yang, D. M. Tong, X. Xiang, C. W. Hu, Catal. Commun.
2009, 10, 1558.
Experimental Section
Cellulose, cellobiose, and catalysts were dried at 808C under
vacuum before use. All catalytic experiments were carried out in a
50 mL Parr autoclave made of 361 stainless steel. For a typical run,
the catalyst (1.5 g) and the carbohydrate (1.5 g) were taken up in
water (15 mL). The mixture was then heated to the desired temper-
ature within 20 min with vigorous stirring. To recover the magnetic
catalyst prior to recycling, a glass tube equipped a permanent
magnet was used. The concentrations of glucose were determined
by a HPLC system consisting of a Waters 1525 pump, a Wa-
ters 5C18-PAQ column (4.6ꢁ250 mm) or a d-sugar column, and a
Waters 2414 refractive index detector. H2SO4 (5 mm) was used as
mobile phase at flow rate of 0.6 mLminÀ1. The concentration of
total reducing sugars was determined by 3,5-dinitrosalicylic acid
(DNS) assay with a Shimadzu DUV-3700 spectrometer (see the
Supporting Information for experimental details)
[20] a) H. Mehdi, V. Fabos, R. Tuba, A. Bodor, L. T. Mika, I. T. Horvath, Top.
Catal. 2008, 48, 49; b) L. Deng, J. Li, D. M. Lai, Y. Fu, Q. X. Guo, Angew.
Chem. 2009, 121, 6651; Angew. Chem. Int. Ed. 2009, 48, 6529; c) J. P.
Lange, R. Price, P. M. Ayoub, J. Louis, L. Petrus, L. Clarke, H. Gosselink,
Angew. Chem. 2010, 122, 4581; Angew. Chem. Int. Ed. 2010, 49, 4479.
Acknowledgements
The authors are grateful to the 973 Program (2007CB210205),
CAS (KGCX2-YW-3306), NCET (08–0519), and NSFC–Guangdong
(U0834005) for financial support.
Received: September 14, 2010
Published online on December 9, 2010
58
www.chemsuschem.org
ꢀ 2011 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemSusChem 2011, 4, 55 – 58