The study of factors influencing the depolymerisation of cellulose using a solid catalyst in ionic liquids

Article abstract of DOI:10.1016/j.carbpol.2010.01.039

Cellulose is the most abundant biomass in the world and constitutes a large fraction of plant biomass. The tight hydrogen-bonding network and van der Waals interactions greatly stabilise cellulose, making it notoriously resistant to hydrolysis. Hence, more efficient pre-treatment procedures are required for the conversion of cellulose to monosaccharides. In this study, the depolymerisation of cellulose and wood in ILs using a solid catalyst is performed successfully. Depolymerisation produces three types of substances: total reducing sugar (TRS), glucose and ethanol. With Avicel, yields of TRS, glucose and ethanol are 76.3% (w/w), 17.2% (w/w) and 4.2% (w/w), respectively. With wood, yields of TRS, glucose and ethanol are 25.6% (w/w), 11.5% (w/w) and 7.7% (w/w), respectively. The time courses of product yields indicates that the depolymerisation of cellulose in [C₄mim]Cl, using Dowex, is similar to the depolymerisation with concentrated H₂SO₄.

Full text of DOI:10.1016/j.carbpol.2010.01.039

Carbohydrate Polymers 80 (2010) 1168–1171
Contents lists available at ScienceDirect
Carbohydrate Polymers
journal homepage: www.elsevier.com/locate/carbpol
The study of factors influencing the depolymerisation of cellulose using
a solid catalyst in ionic liquids
*
Hisanori Watanabe
Kanagawa Environmental Research Center, 1-3-39 Shinomiya, Hiratsuka, Kanagawa 254-0014, Japan
a r t i c l e i n f o
a b s t r a c t
Article history:
Cellulose is the most abundant biomass in the world and constitutes a large fraction of plant biomass. The
tight hydrogen-bonding network and van der Waals interactions greatly stabilise cellulose, making it
notoriously resistant to hydrolysis. Hence, more efficient pre-treatment procedures are required for
the conversion of cellulose to monosaccharides. In this study, the depolymerisation of cellulose and wood
in ILs using a solid catalyst is performed successfully. Depolymerisation produces three types of sub-
stances: total reducing sugar (TRS), glucose and ethanol.
With Avicel, yields of TRS, glucose and ethanol are 76.3% (w/w), 17.2% (w/w) and 4.2% (w/w), respec-
tively. With wood, yields of TRS, glucose and ethanol are 25.6% (w/w), 11.5% (w/w) and 7.7% (w/w),
respectively.
Received 1 December 2009
Received in revised form 15 January 2010
Accepted 20 January 2010
Available online 25 January 2010
Keywords:
Ionic liquid
Cellulose
Solid catalyst
Glucose
The time courses of product yields indicates that the depolymerisation of cellulose in [C4mim]Cl, using
Dowex, is similar to the depolymerisation with concentrated H2SO4.
Ethanol
Ó 2010 Elsevier Ltd. All rights reserved.
1. Introduction
ment procedures are required for the conversion of cellulose to
monosaccharides.
In the past decade, there has been extensive economic develop-
ment in developing countries, e.g. BRICs. These developments
cause to extinguish numerous fossil fuels and exhaust large
amount of greenhouse gas, e.g. carbon dioxide and would elevate
surface temperature of earth. The use of renewable energy, espe-
cially biomass, instead of fossil fuels would help to address this cri-
sis. Today, many bioethanol plants in which sugarcane and bagasse
can be fermented are being constructed.
Ionic liquids (ILs) that have a cation and anion pair and are liq-
uids under 373 K have been investigated for their use as solvents in
organic reactions in recent years (Rogers & Seddon, 2002). Swatlo-
ski, Spear, Holbrey, and Rogers (2002) reported that cellulose dis-
solved in 1-butyl-3-methylimidazolium chloride ([C4mim]Cl)
forms a homogeneous solution. This dissolution might make the
cellulose microfibrils accessible to catalysts. Rinaldi, Palkovits,
and Schuth (2008) reported that solid catalysts, e.g. Amberlyst 15
DRY, are powerful tools for the hydrolysis of cellulose in IL to pro-
duce monosaccharides, disaccharides, reducing sugars, furfural and
5-hydroxymethylfurfural. However, the manner in which other
products and factors influence the reaction, e.g. reaction tempera-
ture and type of IL, was not shown.
Cellulose is the most abundant biomass in the world (Ragauskas
et al., 2006) and constitutes a large fraction of plant biomass, e.g.
wood, wheat bran, straw and cotton. It is a highly crystalline poly-
mer of
D-anhydroglucopyranose units joined together in long
chains by b-1,4-glycosidic bonds. The tight hydrogen-bonding net-
work and van der Waals interactions greatly stabilise cellulose
(Nishiyama, Sugiyama, Chanzy, & Lnagan, 2003), making it notori-
ously resistant to hydrolysis.
Herein, we demonstrate ethanol production through the hydro-
lysis of cellulose using a solid catalyst in different ILs and reveal the
factors that influence the reaction.
However, yeast produces bioethanol by using only glucose, nor
cellulose. Therefore, cellulose must be pre-treated, e.g. degraded
using acids or enzymes, to produce bioethanol.
2. Materials and methods
Cellulose gives rise to monomeric glucose after complete hydro-
lysis with various processes that are catalyzed by mineral acid or
enzymes (Zhang & Lynd, 2004). Hence, more efficient pre-treat-
2.1. Material
The cellulose sample Avicel was purchased from Sigma (St.
Louis, USA) and was dried under vacuum at 373 K for 24 h before
use.
Wood from sugi (Cryptomeria japonica), which is one of the
most common softwood in Japan, was selected as feedstock. Sugi
* Tel.: +81 (0) 463 24 3311; fax: +81 (0) 463 24 3300.
E-mail address: watanabe@k-erc.pref.kanagawa.jp.
0144-8617/$ - see front matter Ó 2010 Elsevier Ltd. All rights reserved.
doi:10.1016/j.carbpol.2010.01.039

H. Watanabe / Carbohydrate Polymers 80 (2010) 1168–1171
1169
wood was ground with a Wiley mill. The component analysis and
characteristics of sugi wood (Cryptomeria japonica) were shown in
Table 1.
cose and ethanol productions might be related to the dissolution
rate of cellulose in each IL. A lesser amount of glucose was obtained
from wood than from Avicel.
As a solid catalyst, Dowex 50WX8 was purchased from Wako
Pure chemical industries, Co. Ltd. (Tokyo, Japan). Ionic liquids were
purchased from Kanto chemical Co. Ltd. (Tokyo, Japan) and used
without further purification.
The two ILs that included the lactate anion yielded less glucose
and ethanol than ILs with chloride and bromide. Inter- and intra-D-
anhydroglucopyranose hydrogen bonds are formed in crystalline
cellulose and these bonds resist dissolution in common solvents,
e.g. water, alcohol and benzene. An electron-donating anion, such
as chloride, is considered to distort the hydrogen bonds in crystal-
2.2. Reaction procedure
line cellulose, which is connected to the OH function of D-anhy-
droglucopyranose. This causes molecular level structural changes
in the form of cellulose i.e. from crystalline to amorphous. In con-
trast, lactate seemed to have a weaker affect on the hydrogen
bonds in crystalline cellulose than chloride and bromide. From
the above observations it can be suggested that ILs that have a lac-
tate anion resulted in lesser glucose and ethanol production than
ILs with a chloride or bromide ion. Glucose yields from Avicel
and wood at 373, 383, 393, 403 and 413 K are represented in
Fig. 1. Glucose production increased with an increase in reaction
temperatures from 373 K to 393 K and decreased with an increase
in reaction temperatures from 393 K to 413 K. The highest glucose
yield, 17.2% (w/w), was obtained at 393 K from Avicel and 11.5%
(w/w) from wood. Wood consists of cellulose, hemicellulose and
lignin. Cellulose microfibrils in wood are filled with hemicellulose
and lignin. Glucose yields from wood were lower than those from
Avicel, which might be because the cellulose component of wood is
less accessible to solid catalysts than in Avicel.
Cellulose (0.1 g) in [C4mim]Cl was heated with stirring at 373 K,
ambient pressure until clear solution was formed. To this cellulose
solution was quickly added Dowex 50WX8 (0.1 g; water content is
53% (w/w)) and a reactor was capped tightly. The reactor was
heated at 373, 383, 393, 403 or 413 K. At different reaction time
intervals, samples were withdrawn, quenched with air and poured
9 ml of deionised water. The aqueous solutions were centrifuged at
10,000 rpm for 5 min, subjected to total reducing sugar (TRS), glu-
cose and ethanol analysis.
2.3. Analysis method
The amount of TRS was measured using the DNS method (Li &
Zhao, 2007). The glucose concentration was determined using a
high-performance liquid chromatograph (Shimadzu LC-9A, Kyoto,
Japan) with an shim-pack SCR-101C column, column oven (Shima-
dzu CTO-6A, Kyoto, Japan) and refractive index detector (Shimadzu
RID-10A, Kyoto, Japan).
The solid catalyst, Dowex 50WX8, is a cation exchanger that
contains sulfonic groups. Usually, strong interaction of the sulfonic
groups with the imidazolium cation of IL is observed in bulk water.
Partial existence of a three-dimensional structure in bulk IL, a
hydrogen-bonding network involving the anion and the ring as
well as the n-alkyl hydrogen atoms, creates a strong hydrophobic
interaction between the two n-alkyl groups of the imidazolium
cations and a unique corrugated sheet structure of the imidazolium
rings (Saha, Hayashi, Kobayashi, & Hamaguchi, 2003; Hayashi, Oza-
wa, & Hamaguchi, 2003). This structure might restrict the mobility
of the imidazolium cation of IL. As a result, the ether linkages of
carbohydrate polymer dissolved in the ILs could be easily attacked
by the sulfonic groups. From the preceding explanation, it appears
that the interactions of the sulfonic groups with the imidazolium
cation of IL in bulk IL are weaker than those seen in bulk water.
The time courses of TRS produced from Avicel and wood are
represented in Fig. 2. These reveal that the TRS yield from Avicel
increased with time until 0.5 h and then decreased from 1 h. In
contrast, the TRS yield from wood maintained almost the same va-
lue at all reaction times. The maximum TRS yields from Avicel and
wood were 76% (w/w) and 26% (w/w), respectively. Rinaldi et al.
(2008) reported that the TRS from microcrystalline cellulose and
wood (spruce) were approximately 27% (w/w) and 23% (w/w) in
5 h, respectively, and Li and Zhao (2007) reported that the TRS
from Avicel and spruce were 73% (w/w) and 71% (w/w), respec-
tively. The TRS yield from Avicel reported in this study is higher
than the values reported by Rinaldi et al. (2008) and Li and Zhao
(2007).
The ethanol concentration was determined using a gas chro-
matograph with FID detector (Shimazu GC-15A, Kyoto, Japan)
and DB-WAX column.
2.4. Calculation of TRS, glucose and ethanol yields
TRS, glucose and ethanol yields were calculated by the Eqs. (1)–
(3), respectively.
TRS yield ð%Þ ¼ Amount of TRS production ðgÞ=0:1 ðgÞ ꢀ 100
ð1Þ
Glucose yield ð%Þ ¼ Amount of glucose production ðgÞ=0:1 ðgÞ ꢀ 100
ð2Þ
Ethanol yield ð%Þ ¼ Amount of ethanol production ðgÞ=0:1 ðgÞ ꢀ 100:
ð3Þ
3. Results and discussion
Table 2 summarised glucose and ethanol yields from Avicel and
wood (Cryptomeria japonica) in different ILs. Maximum glucose and
ethanol productions were obtained using [C4mim]Cl. The highest
dissolution rate of cellulose was obtained using [C4mim]Cl. Glu-
To better understand the solid acid catalyzed hydrolysis of cel-
lulose and wood in [C4mim]Cl, regression analyses of the experi-
mental data were performed by the non-linear least squares
curve fitting method with the Origin 7.0 software. The analyses
indicated that the kinetics of Avicel most likely follows a consecu-
tive first-order reaction sequence, where the k1 for TRS formation
and the k2 for TRS degradation were 0.0875 minꢁ1 and
0.007 minꢁ1, respectively. Therefore, according to this study, TRS
degradation was slower than TRS formation. Moreover, the k1 va-
lue was higher than the values reported by Li and Zhao (2007).
Table 1
Component composition and chemical analysis of sugi wood used in this study.
Cellulose (wt.%)
Hemicelluloses
(wt.%)
Lignin
(wt.%)
Ash
(wt.%)
(a) Component composition
43
27
25
S
0.5
C (wt.%)
H
(wt.%)
O
N
Cl
(wt.%)
(wt.%)
(wt.%)
(wt.%)
(b) Elementary analysis
51.13
6.44
42.13
0.30
<0.01
<0.01

1170
H. Watanabe / Carbohydrate Polymers 80 (2010) 1168–1171
Table 2
Glucose and ethanol yield from Avicel and wood (Cryptomeria japonica) in different ILs. Reaction conditions : Avicel (0.1 g, corresponding to approximately 0.62 mmol of
anhydroglucose, C6H10O5) or wood (0.1 g) dissolved in 2 g of IL, Dowex 50WX8 (0.1 g), reaction time is 3 h, reaction temperature is 393 K.
Type of IL
Glucose yield (%)
Avicel
Ethanol yield (%)
Avicel
Wood
Wood
1-Ethyl-3_methylimidazolium chloride
1-Butyl-3_methylimidazolium chloride
1-Hexyl-3-methylimidazolium chloride
1-Ethyl-3-methylimidazolium bromide
1-Ethyl-2,3-dimethylimidazolium bromide
1-Butyl-2,3-dimethylimidazolium bromide
1-Hexyl-3-methylimidazolium bromide
1-Hexyl-2,3-diethylimidazolium bromide
9.4
17.2
0.9
0.5
0.9
<0.5
<0.5
0.5
2.7
11.5
1.1
1.9
4.2
0.6
<0.5
<0.5
<0.5
1.0
<0.5
7.7
<0.5
0.6
<0.5
<0.5
6.3
5.2
<0.5
<0.5
3.2
2.9
0.6
0.6
1-Ethyl-3-methylimidazolium
1-Butyl-3-methylimidazolium
N,N-Diethyl-N-methyl-N-(2-methoxyethyl)ammonium bis(trifluoromethanesulufonyl)imide
L
-lactate
-lactate
<0.5
<0.5
<0.5
1.0
<0.5
<0.5
<0.5
<0.5
<0.5
<0.5
<0.5
<0.5
L
from wood was faster than k2 for TRS degradation from Avicel,
which might due to the macro and microscale structural differ-
ences between Avicel and wood.
The time courses of glucose yields from Avicel and wood are
represented in Fig. 3. The results reveal that the glucose yield from
Avicel increases with an increase in the reaction time until 2.5 h
and then decreases with time from 3 h. The glucose yield from
wood increases with an increase in reaction time until 3 h, but
which decreases with an increased in reaction time from 4 h. The
maximum glucose yield was 27.7% (w/w) from Avicel and 11.5%
(w/w) from wood.
The depolymerisation of cellulose with concentrated H2SO4 fol-
lowed a random hydrolysis mechanism, in which both endoglycos-
idic and exoglycosidic scission occurred, but the endoglycosidic
product, oligoglucose, was the major product during the initial
hydrolysis stage.
In this study, shorter reaction times favoured TRS yield, and the
glucose yield increased with an increase in the reaction time,
which suggests that the mechanism of cellulose depolymerisation
in [C4mim]Cl, using Dowex, is similar to the depolymerisation
mechanism with concentrated H2SO4.
Fig. 1. Glucose yields from Ç Avicel and s wood (Cryptomeria japonica) determined
by HPLC at 373, 383, 393, 403 and 413 K. Reaction conditions: Avicel (0.1 g) or wood
(0.1 g) dissolved in 2 g [C4mim]Cl, Dowex 50WX8 (0.1 g), reaction time is 3 h.
The liquids, which were depolymerized in ILs by using Avicel at
each reaction time (0.25, 0.5, 1, 2, 2.5, 3, 4 and 4.5 h), presented
brown colors. However, the liquids produced by the depolymerisa-
tion of wood with shorter reaction times (0.25, 0.5 and 1 h) were
light yellow in color; longer reaction times (2, 2.5, 3, 4 and 4.5 h)
gave brown colors. These color differences suggest that the reac-
tion with wood was more difficult than that with Avicel.
Wood is composed of cellulose, hemicellulose and lignin. Hemi-
cellulose and lignin are filled between the cellulose microfibrils.
The glucose yield from wood was less than that from Avicel, which
Fig. 2. Time cource of TRS from Ç Avicel and s wood (Cryptomeria japonica)
determined by DNS assay. Reaction conditions: Avicel (0.1 g) or wood (0.1 g)
dissolved in 2 g [C4mim]Cl, Dowex 50WX8 (0.1 g), reaction temperature is 393 K.
Regression analysis of the experimental data by non-liner least squares curve fitting
with the software Origin 7.0. Time-course of TRS concentration from Avicel (solid
line) and wood (dash line) calculated by an equation: TRS concentration = Aꢀ(k1/
(k1 ꢁ k2))ꢀ(exp(ꢁk2ꢀt) ꢁ exp(ꢁk1ꢀt)), where, A: 41.98 (Avicel), 8,239,000 (wood);
k1: 0.0875 (Avicel), 0.00095 (wood) k2: 0.007 (Avicel), 757.8 (wood).
The cellulose hydrolysis method described in this study is a pow-
erful tool for TRS formation.
The kinetics for wood did not follow a consecutive first-order
reaction sequence; k1 and k2 were 0.00095 minꢁ1 and 757.8 minꢁ1
,
respectively. Therefore, TRS degradation was faster than TRS for-
mation. Moreover, k1 for TRS formation from wood was slower
than k1 for TRS formation from Avicel, and k2 for TRS degradation
Fig. 3. Time courses of glucose yield from Ç Avicel and s wood (Cryptomeria
japonica) determined by HPLC. Reaction conditions: Avicel (0.1 g) or wood (0.1 g)
dissolved in 2 g [C4mim]Cl, Dowex 50WX8 (0.1 g), reaction temperature is 393 K.

H. Watanabe / Carbohydrate Polymers 80 (2010) 1168–1171
1171
might be because the cellulose in wood is less accessible to solid
catalysts than that in Avicel.
the depolymerisation of cellulose in [C4mim]Cl, using Dowex, is
similar to the depolymerisation with concentrated H2SO4.
Rinaldi et al. (2008) reported that approximately 3% (w/w)
mono- and disaccharides were produced from
a-cellulose. Li and
References
Zhao (2007) reported that 39% (w/w) glucose was produced from
Avicel and 36% (w/w) from spruce. The glucose production in this
study is higher than the mono- and disaccharides reported by Ri-
naldi et al. (2008) and smaller than the values reported by Li and
Zhao (2007). Therefore, the depolymerisation of cellulose and
wood in ILs using a solid catalyst, as performed in this study, would
be more suitable for TRS formation rather than glucose formation.
Hayashi, S., Ozawa, R.,
& Hamaguchi, H. (2003). Raman spectra, crystal
polymorphism, and structure of a prototype ionic-liquid [bmim]Cl. Chemistry
Letters, 32, 498–499.
Li, C., & Zhao, Z. K. (2007). Efficient acid-catalyzed hydrolysis of cellulose in ionic
liquid. Advanced Synthesis & Catalysis, 349, 1847–1850.
Nishiyama, Y., Sugiyama, J., Chanzy, H., & Lnagan, P. (2003). Crystal Structure and
hydrogen bonding system in cellulose Ia from synchrotron X-ray and neutron
fiber diffraction. Journal of American Chemical Society, 125, 14300–14306.
Ragauskas, A. J., Williams, C. K., Davison, B. H., Britovsek, G., Cairney, J., Eckert, C. A.,
et al. (2006). The path forward for biofuels and biomaterials. Science, 311,
484–489.
4. Conclusion
Rinaldi, R., Palkovits, R., & Schuth, F. (2008). Depolymerization of cellulose using
solid catalysts in ionic liquids. Angewandte Chemie, Int. Ed, 47, 8047–8050.
Rogers, R. D., & Seddon, K. R. (2002). Ionic liquids: Industrial applications for green
chemistry. Oxford Univ. Press.
Saha, S., Hayashi, S., Kobayashi, A., & Hamaguchi, H. (2003). Crystal structure of 1-
butyl-3-methylimidazolium chloride. A clue to the elucidation of the ionic
liquid structure. Chemistry Letters, 32, 740–741.
Swatloski, R. P., Spear, S. K., Holbrey, J. D., & Rogers, R. D. (2002). Dissolution of
cellose with ionic liquids. Journal of American Chemical Society, 124, 4974–4975.
Zhang, Y.-H. P., & Lynd, L. R. (2004). Toward an aggregated understanding of
enzymatic hydrolysis of cellulose: Noncomplexed cellulase systems.
Biotechnology and Bioengineering, 88, 797–824.
Depolymerisation of cellulose in ILs using Dowex produces
three types of substances: total reducing sugar (TRS), glucose and
ethanol.
With Avicel, yields of TRS, glucose and ethanol were 76.3% (w/
w), 17.2% (w/w) and 4.2% (w/w), respectively. With wood, yields
of TRS, glucose and ethanol were 25.6% (w/w), 11.5% (w/w) and
7.7% (w/w), respectively.
Shorter reaction times favoured TRS yield and glucose yield in-
creased with an increase in the reaction times, which indicates that