High-yielding one-pot synthesis of glucose from cellulose using simple activated carbons and trace hydrochloric acid

Article abstract of DOI:10.1021/cs300845f

High-yielding one-pot synthesis of glucose from cellulose and pentoses/hexoses from real biomass is achieved by using simple activated carbons and 0.012% HCl in water. Ball-milling cellulose and the carbon together created good physical contact between the solid substrate and solid catalyst before the reaction, selectively and drastically improving the depolymerization rate of cellulose to oligomers. Thus, our methodology overcomes a major obstacle in this type of reaction, namely, that the collision between a solid catalyst and a solid substrate is limited. Mechanistic studies have suggested that the active sites of the carbons are weakly acidic functional groups, in which vicinal carboxylic and phenolic groups synergistically work for the hydrolysis reaction.

Full text of DOI:10.1021/cs300845f

Research Article
pubs.acs.org/acscatalysis
High-Yielding One-Pot Synthesis of Glucose from Cellulose Using
Simple Activated Carbons and Trace Hydrochloric Acid
Hirokazu Kobayashi,† Mizuho Yabushita,† Tasuku Komanoya,† Kenji Hara,† Ichiro Fujita,‡
and Atsushi Fukuoka*,†
†Catalysis Research Center and Graduate School of Chemical Sciences and Engineering, Hokkaido University, Sapporo, Hokkaido
001-0021, Japan
‡Showa Denko K.K., Minato-ku, Tokyo 105-8518, Japan
S
* Supporting Information
ABSTRACT: High-yielding one-pot synthesis of glucose from
cellulose and pentoses/hexoses from real biomass is achieved
by using simple activated carbons and 0.012% HCl in water.
Ball-milling cellulose and the carbon together created good
physical contact between the solid substrate and solid catalyst
before the reaction, selectively and drastically improving the
depolymerization rate of cellulose to oligomers. Thus, our
methodology overcomes a major obstacle in this type of
reaction, namely, that the collision between a solid catalyst and
a solid substrate is limited. Mechanistic studies have suggested
that the active sites of the carbons are weakly acidic functional groups, in which vicinal carboxylic and phenolic groups
synergistically work for the hydrolysis reaction.
KEYWORDS: biomass, carbon catalyst, cellulose, glucose, hydrolysis
higher acidity than HZSM-517 have recently been developed to
1. INTRODUCTION
this end. Meanwhile, we and others demonstrated that
The production of renewable chemicals and fuels from biomass
has attracted worldwide interest for nurturing sustainable
mesoporous carbons such as CMK-318 catalyze the hydrolysis
of cellulose.19−21 The hydrolytic function of an oxidized carbon
societies.1−5 Cellulose, a polymer composed of glucose units, is
nanotube was also suggested in related reactions.22 Weakly
the most abundant and nonfood biomass resource, and glucose
is a versatile precursor to biodegradable plastics and fuels.4 The
acidic and water-tolerant carbons might be potentially useful for
the hydrolysis of cellulose; however, the yield of glucose needs
selective conversion of cellulose to glucose (Scheme 1)
to be improved, and the details of the catalysis have not been
typically requires large amounts of mineral acids, enzymes, or
clarified. Hence, the purpose of this study is the high-yielding
ionic liquid solvents because of the persistent and water-
insoluble properties of cellulose and the chemical instability of
one-pot synthesis of glucose from cellulose using simple
glucose.6−9 A prospective strategy to reduce the usage of
activated carbons. Mechanistic studies were also conducted to
soluble catalysts is the utilization of solid catalysts,10−12 and
estimate the active sites on carbons.
immobilized sulfonic acids13−16 and a silica catalyst possessing
2. EXPERIMENTAL SECTION
Scheme 1. Hydrolysis of Cellulose to Glucose
2.1. Reagents. Microcrystalline cellulose was purchased
from Merck, and cellobiose was from Kanto. The carbons used
in this study were a coke powder (Showa Denko), BA50
(Ajinomoto Fine-Techno), MSP20 (Kansai Netsu Kagaku),
BP2000 (Black Pearls 2000, Cabot), and XC72 (Vulcan XC72,
Cabot). All of the other reagents such as distilled water were
the finest grade, obtained from Wako, TCI (Tokyo Chemical
Industry), and Aldrich.
2.2. Preparation of K26 and K20. The coke powder was
treated with KOH (3 equiv based on weight) under N2 at 973
Received: December 27, 2012
Revised: February 16, 2013
© XXXX American Chemical Society
581
dx.doi.org/10.1021/cs300845f | ACS Catal. 2013, 3, 581−587

ACS Catalysis
Research Article
a
Table 1. Hydrolysis of Cellulose by Carbon Catalysts
yield based on carbon (%)
sugar product
byproduct
b
c
entry
catalyst
None
ball-milling of cellulose T (K) conv. (%) glucose fructose mannose oligomers total levoglucosan 5-HMF
others
h
1
single
single
single
single
single
single
single
single
503
503
503
503
503
503
503
503
503
453
453
453
453
453
453
453
28
60
59
50
57
52
37
35
98
12
93
18
98
95
39
40
4.6
36
0.5
2.7
1.9
1.7
1.0
0.9
0.5
0.6
4.3
0.2
0.6
0.5
1.5
0.7
1.8
1.3
0.6
2.6
1.3
1.6
1.0
0.8
0.5
0.9
1.4
0.2
0.7
0.4
1.5
1.3
1.8
1.3
15
2.5
21
44
40
36
39
39
19
26
66
0.2
2.1
2.7
1.6
0.9
0.7
0.3
0.2
3.9
<0.1
0.7
0.1
3.0
2.2
1.0
1.1
1.8
3.4
2.9
2.1
3.5
2.5
1.8
1.9
5
11
14
11
14
10
16
7
h
h
h
h
h
h
h
h
i
2
K26
3
K20
35
26
17
1.7
6.3
4
MSP20
BA50
CMK-3
BP2000
XC72
5
20
j
6
12
25
12
19
7
6.4
5.8
57
1.3
20
2.9
88
8
d
d
9
K26
mix
3.7
6.6
70
10
2.7
12
16
3
10
11
12
13
14
15
16
None
single
8.3
0.2
1.0
d
d
i
K26
mix
91
14
94
80
35
37
<1
4
e
i
K26
single
<0.1
1.7
d
f
d
i
K26 , HCl
mix
<1
11
2
d
g
d
i
K26 , H2SO4
mix
69
27
30
8.6
3.9
4.3
1.8
f
i
HCl
single
single
1.6
e
f
i
K26 , HCl
0.7
1
a
b
c
Conditions: cellulose, 324 mg; carbon, 50 mg; distilled water, 40 mL. 5-Hydroxymethylfurfural. (conversion) − (total yield of the shown
d
e
f
products). K26 and cellulose were ball-milled together. K26 was ball-milled without cellulose. 0.012 wt % HCl aq. (pH 2.5) was used instead of
g
h
distilled water. 0.018 wt % H2SO4 aq. (pH 2.5) was used instead of distilled water. Rapid heating−cooling conditions (Supporting Information,
Figure S1). Temperature was kept for 20 min. CMK-3 was synthesized several times, and the samples gave glucose in the range of 12−16% yield.20
i
j
K. The resulting black powder was washed with water, 1 M HCl
aq., and boiling water until the washing water became neutral.
Finally, the purified carbon was filtrated by a 45 μm mesh in
water to remove large particles and dried at 353 K, and the
resulting sample was named K26 (80% yield based on the
weight of coke used). Similarly, K20 was synthesized by using
KOH (2.8 equiv) with the same procedures as above. K26 was
characterized by the Boehm titration,23 diffuse reflectance
infrared Fourier transform (DRIFT; Perkin-Elmer, Spectrum
100), temperature-programmed desorption (TPD; Bel, BEL-
CAT-A, mass spectrometer), NH3-TPD (Bel, BELCAT-A,
mass spectrometer, m/z = 16), N2 adsorption (Bel, Belsorp-
mini), laser diffraction (Nikkiso, Microtrac MT3300EXII), ion
chromatograph after the combustion, and energy dispersive X-
ray spectroscopy (EDX; Shimadzu EDX-720). EDX analysis
indicated that the concentration of residual potassium in K26
was only 120 ppm, which did not affect the catalytic activity.
2.3. Heat-Treatment of K26. K26 was charged in a quartz
fixed-bed flow reactor. The temperature, monitored by a
thermocouple, was elevated from 298 to 393 K in He flow (3
mL min−1), and the temperature was kept at 393 K for 1 h to
remove water adsorbed. Further treatment was carried out by
raising the temperature to 673, 873, 1073, or 1273 K by 10 K
min−1, and the temperature was kept for 2 h.
2.4. Ball-Milling Treatment of Cellulose. Single ball-
milling: microcrystalline cellulose (10.0 g) was milled in a
ceramic pot (3.6 L) with alumina balls (1.5 cm, 2 kg) at 60 rpm
for 2 days. Mix-milling: microcrystalline cellulose (10.0 g) and
K26 (1.54 g) were milled together in a ceramic pot (3.6 L) with
alumina balls (1.5 cm, 2 kg) at 60 rpm for 2 days. The milled
cellulose was analyzed by viscometry using an Ubbelohde
viscometer and 9 wt % LiCl/DMAc,24 laser diffraction, XRD
(Rigaku, MiniFlex), and 13C CP/MAS NMR (Bruker, MSL-
300, 13C 75 MHz, MAS: 8 kHz).
milled 324 mg of cellulose, 50 mg of carbon, and 40 mL of
distilled water (or 0.012% HCl aq.) were charged in the reactor.
The temperature profiles of the reactions are shown below.
Hydrolysis of mix-milled cellulose: the mix-milled sample [374
mg, containing cellulose (324 mg) and K26 (50 mg)] and 40
mL of distilled water (or 0.012% HCl) were charged in the high
pressure reactor.
Two types of temperature profiles were used in the
hydrolysis reactions. Rapid heating−cooling conditions: the
reactor was heated to 503 K in 18 min, and then quickly cooled
down to less than 323 K by blowing air for 22 min (Supporting
Information, Figure S1). Reactions at T K for X min: the
reactor was heated to a certain temperature (T; typically 453 K
in 11 min), and the temperature was kept for X min. Then the
reactor was cooled down to less than 323 K by blowing air.
The reaction mixture was separated into solid and liquid by
centrifugation and decantation. The aqueous phase was
analyzed by high-performance liquid chromatography (col-
umns: Shodex Sugar SH-1011, Phenomenex Rezex RPM-
Monosaccharide Pb++).20 Yields of products were determined
by the absolute calibration method with an error of 1%. The
identification of products was checked by a LC/MS (Thermo
Fisher Scientific, LCQ-Fleet, APCI). Conversion of cellulose
was determined based on the weight difference of the solid part
before and after reaction.
2.6. Analysis of Composition of Bagasse Pulp. Bagasse
pulp was obtained in Okinawa Prefecture, and the composition
was analyzed by an established method (NREL, TP-510−
42618).25 The contents of cellulose, hemicellulose, and lignin
were 59%, 27% (xylan 25%, arabinan 2%), and 9%, respectively,
based on the dry weight.
3. RESULTS AND DISCUSSION
Several carbons were tested in the hydrolysis of ball-milled
cellulose in distilled water under the rapid heating−cooling
conditions (Supporting Information, Figure S1, Table 1, entries
1−8), and it was found that their catalytic activities strongly
2.5. Hydrolysis of Cellulose. Hydrolysis of ball-milled
cellulose: the reaction was conducted in a hastelloy C22 high-
pressure reactor (OM Lab-Tech, MMJ-100, 100 mL). The ball-
582
dx.doi.org/10.1021/cs300845f | ACS Catal. 2013, 3, 581−587

ACS Catalysis
Research Article
a
Table 2. Control Experiments on the Hydrolysis of Cellulose
yield based on carbon (%)
sugar product
mannose
byproduct
b
c
entry
reaction medium
distilled water
conv. (%)
glucose
fructose
oligomers
total
levoglucosan
5-HMF
others
1
28
29
26
22
39
4.6
4.3
3.8
3.1
0.5
0.8
0.8
0.5
0.6
0.8
0.8
0.6
15
17
17
13
21
23
22
17
0.2
0.2
0.2
0.1
1.8
1.9
2.0
1.4
5
4
1
3
17
18
19
20
10 μM H2SO4
50 μM acetic acid
filtrate of used K26
d
e
f
f
f
f
f
f
f
f
filtrate of used K26 and cellulose
n.d.
n.d.
n.d.
n.d.
n.d.
n.d.
n.d.
n.d.
a
b
Reaction conditions: ball-milled cellulose, 324 mg; solution, ca. 40 mL. Rapid heating−cooling conditions (Supporting Information, Figure S1). 5-
c
d
Hydroxymethylfurfural. (conversion) − (total yield of the shown products). The filtrate of K26 aqueous mixture subjected to the rapid heating−
e
cooling conditions was used as the solvent. The filtrate of the mixture of the cellulose hydrolysis by K26 (Table 1, entry 2) was used as the solvent.
f
Determination was not possible because products in the primary reaction (Table 1, entry 2) were contained, and they underwent the degradation
during this reaction.
depended on their natures. The most active catalyst, alkali-
activated carbon K26, afforded 60% conversion of cellulose and
36% yield of glucose (entry 2). This glucose yield was obviously
higher than that in the control experiment without catalysts
(4.6%, entry 1). The other products were fructose (2.7%),
mannose (2.6%), water-soluble cello-oligosaccharides (2.5%),
and byproducts (16%). K26 was reusable up to 4 times without
loss of activity (Supporting Information, Figure S2). The
production of concentrated glucose (>10 wt %) was also easily
achieved (Supporting Information, Table S1), which is
beneficial to avoid the energy-consuming condensation of the
products. Notably, an inexpensive steam-activated carbon BA50
produced glucose in 17% yield (entry 5), which was applied to
the conversion of bagasse pulp (vide infra).
We have checked if K26 functions as a solid catalyst that can
hydrolyze cellulose. The pH value of an aqueous dispersion of
K26 was 4.9 because of weakly acidic surface functional groups.
Note that this result does not indicate the formation of soluble
acids from K26 because the pH of the solution returned to 5.8
after separation of K26 using a polytetrafluoroethylene
membrane. For comparison, 10 μM H2SO4 (pH 4.7) and 50
μM acetic acid (pH 4.6) were used in the hydrolysis of
cellulose, resulting in no enhancement of conversion or yield at
all (Table 2, entries 17, 18). As it is known that the hydrolysis
by H3O+ is negligible at pH higher than 4,26 the promotion of
the hydrolysis by K26 is not ascribed to the buffering effect
releasing H3O+ in the suspended state. Moreover, an aqueous
mixture of K26 was subjected to the reaction conditions at 503
K and filtered to remove the K26. The filtrate was used for the
hydrolysis of cellulose; however, the reaction was not
accelerated (entry 19). In addition, soluble acidic byproducts
formed from cellulose during the hydrolysis reaction degraded
only 11% of the cellulose, subtracting the conversion of
cellulose in the blank experiment (28%, entry 1) from that in
this reaction (39%, entry 20). Therefore, soluble compounds
would not be major active species in the hydrolysis of cellulose
by K26.
constant concentration of dissolved cellulose. However, the rate
increased almost linearly with increasing concentration of solid
cellulose (Supporting Information, Table S2). Although the
possibility of soluble active species or partial dissolution of the
substrate is not completely eliminated, these results suggest that
solid K26 hydrolyzes solid cellulose in addition to the
hydrolysis by hot-compressed water26 and small amounts of
acids derived from cellulose.
Considering that the hydrolysis of cellulose by carbon
catalysts also occurs at the solid−solid interface, their limited
collision is a major obstacle in this type of reaction. Therefore,
cellulose and K26 were ball-milled together, denoted as mix-
mill hereafter, as a pretreatment to improve their contact
(Supporting Information, Figure S3). The mixed state can
continue in the subsequent hydrolysis reaction thanks to the
insoluble properties of the catalyst and the substrate. The
hydrolysis of this mix-milled sample at 503 K resulted in 98%
conversion, and the products were glucose (57% yield),
fructose (4.3%), mannose (1.4%), oligomers (3.7%), and
byproducts (32%) (Table 1, entry 9). As the hydrolysis
reaction was already completed, milder conditions (453 K, 20
min) were chosen based on the effect of temperature
(Supporting Information, Figure S4) to produce sugars
selectively (entries 10−16). The modified reaction afforded
91% yield of sugars [glucose (20%), fructose (0.6%), mannose
(0.7%), and oligomers (70%)] with 98% selectivity (entry 11),
whereas the separately milled K26 and cellulose (viz., K26 and
cellulose, respectively, were singularly ball-milled) provided
only 14% yield of the products (entry 12). These two results
clearly indicate that the mix-milling pretreatment drastically and
selectively accelerates the hydrolysis of cellulose.
An extracted solution of the mix-milled sample with boiling
water was not active for the hydrolysis of singularly milled
cellulose (glucose yield 1.5%, oligomers 10%), thus showing
that the effect of mix-milling is not ascribed to the formation of
soluble active species from K26 during its ball-milling. The DP
of our mix-milled cellulose (690) and singularly milled one
(640) were similar because of the mild conditions (simple ball-
milling at 60 rpm). Furthermore, the extract of the mix-milled
sample with boiling water contained only a small amount of
soluble oligomers (2.2%). Hence, the enhancement of the
reaction performance by our mix-milling is not due to the
mechanocatalytic hydrolysis28,29 during the treatment but due
to the mixing. Accordingly, our pretreatment does not require
such high power to dissociate the glycosidic bonds.
We have also investigated whether or not cellulose is
hydrolyzed as a solid substrate. The ball-milled cellulose did not
contain a considerable amount of soluble oligomers (only
0.3%), verified by the extraction with boiling water, and the
milled cellulose had a high enough degree of polymerization
(DP = 640). The influence of the solubility of cellulose in water
was also estimated. If the slightly soluble portion of cellulose
(ca. 2 × 10−3 wt % at ambient temperature)27 corresponded to
the reaction performance, the hydrolysis rate should be
controlled by the saturated solubility of cellulose, giving a
Physicochemical properties of the milled samples were
further characterized to elucidate how the mix-milling pretreat-
583
dx.doi.org/10.1021/cs300845f | ACS Catal. 2013, 3, 581−587

ACS Catalysis
Research Article
ment enhances the hydrolysis reaction. Figure 1 displays 13C
CP/MAS NMR spectra of the mix-milled cellulose and the
almost negligible owing to the limited collision. Clearly, the
mix-milling treatment is essential.
For evaluating the durability of K26 in this system, a high
concentration of the mix-milled sample (cellulose: 200 g L−1,
K26: 31 g L−1) was used in the hydrolysis. The used K26 and
unreacted cellulose were recovered, milled again with fresh
cellulose, used for the hydrolysis reactions in 0.012% HCl, and
the reuse experiments were repeated for 3 times (total 4 runs).
The additive amount of fresh cellulose was just equivalent to
that consumed in the previous run, keeping a constant amount
of cellulose in each run. Conversion of cellulose was 88−91%,
and glucose was obtained in 71%, 72%, 70%, and 67% yields,
respectively, in the reuse experiments with the concentrations
of 13 to 14 wt % (Figure 2). In addition, the degradation rate of
Figure 1. 13C CP/MAS NMR spectra of the mix-milled K26 and
cellulose (solid line) and singularly milled cellulose (dashed line). C4c
shows the chemical shift of C4 in crystalline cellulose and C4a does
that in the amorphous one.
singularly milled one, which are almost the identical. The broad
peaks are assigned to amorphous cellulose as indicated in
Figure 1,30 and the crystallinity indexes determined from the
C4 signals for both samples were less than 5%. Furthermore,
these samples show almost the same XRD patterns of
amorphous cellulose (Supporting Information, Figure S5) and
similar distributions of particle sizes (Supporting Information,
Figure S6). Thus, a possible role of the mix-milling pretreat-
ment is making better physical contact between the cellulose
and the carbon.
Figure 2. Reuse experiments of mix-milled K26 for the hydrolysis of
cellulose (200 g L−1) in 0.012% HCl. First, 473 K for 2 min, then 423
K for 60 min. Red, glucose; blue, oligomers; yellow, fructose and
mannose; gray, others.
The hydrolysis of mix-milled cellulose mainly produces
soluble oligomers as described above (Table 1, entry 11). For
further converting the formed oligomers to glucose in a one-pot
reaction, we have utilized 0.012% HCl, which neither corrodes
common stainless steel reactors nor has a negative economic
impact because of the very low concentration and the low
price.31 The hydrolysis of mix-milled cellulose by K26 with
trace HCl afforded as high as 88% yield of glucose with 90%
selectivity based on the conversion of cellulose (98%) (entry
13). The yield of humins was at most 2% in this case as the
recovered solid after the reaction was only 2%. Time course of
this experiment indicated that oligomers were produced prior
to the formation of glucose in 2 min after reaching 453 K
(Supporting Information, Figure S7). Subsequently, the yield of
glucose increased to 88% in 20 min and then decreased because
of the degradation. H2SO4 was also used for the cellulose
hydrolysis, but it was not as effective as HCl at the same pH
cellulose did not decrease in another series of reuse
experiments at lower cellulose conversions (Supporting
Information, Figure S8A). Particle size of the mix-milled
sample slightly increased in the first run because of aggregation
during the recovery process, and smaller particles were not
formed (Supporting Information, Figure S8B). Hence, K26 is
fairly stable in the mix-milling as well as in the presence of HCl
and high concentrations of products. The yields of glucose in
Figure 2 were lower than that in the standard conditions (88%,
Table 1 entry 13), which will be improved by the optimization
for the high-concentration reactions in our future work.
We also tried the saccharification of bagasse pulp as a real
biomass substrate by an inexpensive steam-activated carbon
BA50, which showed catalytic activity in cellulose conversion
(Table 1, entry 5), for practical use. Taking account of the
different reactivity of hemicellulose and cellulose, the mix-
milled sample was subjected to the reaction at 453 K for 1 min
and subsequently at 483 K for 2 min. As a result, hemicellulose
was converted to xylose (84% yield) and arabinose (8.1%) and
cellulose was converted to glucose (76%), fructose (2.5%), and
mannose (1.5%) (Figure 3). In addition, pentoses were mainly
produced in the first step, whereas hexoses were produced in
the second one. This separate production of sugars will be
advantageous for further use.
2−
(entry 14), which was due to the negative effect of SO4 on
the hydrolysis reaction (Supporting Information, Table S3).
Thus, we focused on HCl, and control experiments were
conducted to evaluate the contributions of HCl and K26.
Hydrolysis of singularly milled cellulose by HCl provided 27%
yield of glucose and 39% conversion (entry 15). The yield of
oligomers was only 3.9%, showing the quick transformation of
oligomers to glucose by HCl. Accordingly, HCl works for the
hydrolysis of cellulose and oligomers, and the further
improvement from entry 15 to 13 is accounted for the
hydrolysis by K26. Besides, the hydrolysis of separately milled
cellulose and K26 in the trace HCl gave a similar yield of
glucose (30%) and a conversion (40%, entry 16) to those of
entry 15 (yield 27%, conversion 39%), indicating that the
hydrolysis by K26 without the mix-milling pretreatment was
Next, the active sites on K26 were estimated by several
physicochemical measurements. We checked that the hydrolysis
activity did not result from the catalysis of sulfonic groups. The
content of sulfur in K26 was less than 0.01%, corresponding to
only 10−5% in the whole reaction mixture. Even 10−4% H2SO4
did not promote the hydrolysis of cellulose (Table 2, entry 17).
Moreover, no strong acid peaks were observed in the NH3-
584
dx.doi.org/10.1021/cs300845f | ACS Catal. 2013, 3, 581−587

ACS Catalysis
Research Article
as ΔConversion and ΔGlucose (see the caption of Figure 4).
Both ΔConversion and ΔGlucose gradually decreased when
elevating the HT temperature, suggesting the decomposition of
active sites of K26 in the HT. Then, the specific amounts of
oxygenated groups of heat-treated K26s were determined by
the Boehm titration23 (Figure 4B). The amount of carboxylic
acids first decreased at about 673 K, followed by the reduction
of lactones. Phenolic groups were slightly more stable than
lactones, which was consistent with previous results.32 The
elimination of oxygenated groups was also supported by
DRIFT and TPD measurements (Supporting Information,
Figures S10, S11). From these results, we concluded that the
weakly acidic oxygenated groups were removed in the
temperature range of the deactivation of K26. Although the
lactone is not an acid, it may be in an equilibrium with the
carboxylic acid and the phenol in hot-compressed water.
Morphological change by the HT was also checked by N2
adsorption and laser diffraction measurements. The heat-
treated K26s provided type I isotherms, indicating similar
microporous structures (Figure 5). The Brunauer−Emmett−
Figure 3. Hydrolysis of bagasse pulp mix-milled with BA50. Red,
glucose; yellow, fructose and mannose; green, xylose; purple,
arabinose.
TPD measurement (Supporting Information, Figure S9).
Hence, the hydrolytic activities of K26 might be ascribed to
weakly acidic groups such as carboxylic acids and phenols on
the surface.
To elucidate the contributions of the functional groups for
the hydrolytic activity, they were partially removed by a heat
treatment32 under He flow, denoted as HT, at 673 to 1273 K.
Figure 4A depicts the hydrolytic activities of the treated K26s
for the hydrolysis of singularly ball-milled cellulose against the
HT temperature. For evaluating the activities, conversion and
glucose yield obtained in the blank reaction without catalysts
were subtracted from those of the catalytic reactions, indicated
Figure 5. N2 adsorption−desorption isotherms for heat-treated K26s
at 77 K. Red, nontreated; orange, pretreated at 673 K; green, 873 K;
cyan, 1073 K; purple, 1273 K. Baselines are shifted to avoid overlap of
the respective isotherms.
Teller (BET) specific surface area decreased from 2270 m2 g−1
to 1730 m2 g−1 along with the decrease of the pore volume
from 1.00 cm3 g−1 to 0.75 cm3 g−1 when increasing the HT
temperature (Table 3). The distributions of particle diameters
Table 3. Morphological Properties of Heat-Treated K26s
total pore
heat treatment BET specific surface
volume
median particle
diameter (μm)
temp. (K)
area (m2 g−1
)
(cm3 g−1
)
no treatment
673
2270
2240
1970
1980
1730
1.00
0.99
0.86
0.86
0.75
13
13
13
12
12
873
1073
1273
were maintained during the HT (Figure 6), providing almost
constant median particle diameters (12−13 μm) (Table 3). It is
not probable that these small morphological changes of K26 in
the HT are the origin of the deactivation of K26. Instead, we
propose that the catalytic activity is correlated with the specific
amounts of oxygenated groups. However, the contributions of
respective functional groups were not clarified in these
Figure 4. (A) Hydrolysis of cellulose by heat-treated K26s and (B) the
specific amounts of functional groups. ΔGlucose = (yield of glucose in
the catalytic reaction) − (yield of glucose in the control experiment
without catalysts, 4.6%). ΔConversion = (conversion in the catalytic
reaction) − (conversion in the control experiment without catalysts,
28%).
585
dx.doi.org/10.1021/cs300845f | ACS Catal. 2013, 3, 581−587

ACS Catalysis
Research Article
phenolic groups in the hydrolysis of glycosidic bonds. Capon
demonstrated that the hydrolysis of a glycosidic bond is
dramatically accelerated by an intramolecular adjacent carbox-
ylic acid,33 while opposite-side ones do not work. Katz et al.
reported that even silanols (pKa = ca. 7) can hydrolyze cellulose
by optimizing the configuration of functional groups; ether
bonds are formed between hydroxyl groups of cellulose chains
and silanols, and then glycosidic bonds of the immobilized
cellulose are activated by free silanols located near the
substrate.34,35 They also indicated that dense hydroxyl groups
on alumina worked as an array of weak acidic groups for the
hydrolysis of cellulose in a similar fashion to those of silanols.36
Moreover, Hara et al. indicated that phenolic groups can adsorb
glucans.37,38 Hence, we tentatively propose that phenolic
groups form ether or hydrogen bonds with hydroxyl groups
of a cellulose chain, and that adjacent carboxylic acids have an
opportunity to hydrolyze glycosidic bonds (Supporting
Information, Figure S12). Mechanistic details will be shown
in future work.
Figure 6. Distributions of particle diameters of heat-treated K26s in
water. Red, nontreated; orange, pretreated at 673 K; green, 873 K;
cyan, 1073 K; purple, 1273 K. Baselines are shifted to show respective
distributions.
4. CONCLUSIONS
experiments because the amounts of respective oxygenated
groups were similarly decreased in the HT.
The results represented here show that simple activated
carbons are effective for the hydrolysis of cellulose to water-
soluble sugars, and that combining with trace HCl affords 88%
yield of glucose. The good physical contact between solid
substrates and solid catalysts is particularly important to achieve
high performance. Mechanistic studies suggest that the active
sites of the carbons are weakly acidic functional groups, in
which neighboring carboxylic and phenolic groups synergisti-
cally work for the hydrolysis reaction. Notably, large-scale and
quick millings are already practically available,29 for example, in
the cement and food industries, thus showing applicability of
our method in the biorefinery.
The roles of carboxylic acids and phenols on K26 were
studied in model reactions, in which aromatic compounds were
used as molecular catalysts and cellobiose as a substrate by
quantitatively evaluating the hydrolytic activities at 443 K in
water (Figure 7). Phenol (pKa = 10) and benzoic acid (pKa =
ASSOCIATED CONTENT
* Supporting Information
■
S
Additional characterization data, reaction results, and figures.
This material is available free of charge via the Internet at
http://pubs.acs.org.
AUTHOR INFORMATION
Corresponding Author
*Fax: +81-11-706-9139. E-mail: fukuoka@cat.hokudai.ac.jp.
■
Notes
The authors declare no competing financial interests.
Figure 7. Hydrolysis of cellobiose by molecular catalysts at 443 K.
Cellobiose 25 mM, catalyst 0.5 mM. TOF = [(mole of glucose
produced in the catalytic reaction) − (mole of glucose produced
without catalysts)]/[2(mole of catalyst)·(reaction time)].
ACKNOWLEDGMENTS
■
This work was supported by a Grant-in-Aid for Scientific
Research (KAKENHI, 20226016) and a Grant-in-Aid for
Young Scientists (KAKENHI, 23760734) from the Japan
Society for the Promotion of Science (JSPS).
4.2) were almost inactive (TOF = 0 h−1, 4.5 h−1, respectively),
whereas o-hydroxybenzoic acid (salicylic acid, pKa = 3.0)
bearing both a carboxylic acid and a phenolic group gave a TOF
as high as 28 h−1. However, m- and p-hydroxybenzoic acids
(pKa = 4.1, 4.6, respectively) were nearly inactive (TOF = 5.7
h−1, 2.4 h−1), and o-chlorobenzoic acid (pKa = 2.9, TOF = 17
h−1) was also less active than salicylic acid regardless of their
similar acidities. The catalytic activities can be explained by two
factors: one is pKa because the order of activity is the same as
that of pKa except for salicylic acid. The other is the structure of
acids as indicated by the unexpectedly high activity of salicylic
acid, which suggests the synergy of neighboring carboxylic and
REFERENCES
■
(1) Alonso, D. M.; Bond, J. Q.; Dumesic, J. A. Green Chem. 2008, 12,
1493−1513.
(2) Climent, M. J.; Corma, A.; Iborra, S. Green Chem. 2011, 13, 520−
540.
(3) Gallezot, P. Chem. Soc. Rev. 2012, 41, 1538−1558.
́ ́
(4) Tuck, C. O.; Perez, E.; Horvath, I. T.; Sheldon, R. A.; Poliakoff,
M. Science 2012, 337, 695−699.
(5) Kamm, B.; Gruber, P. R.; Kamm, M. Biorefineries - Industrial
Processes and Products; Wiley-VCH: Weinheim, Germany, 2010.
(6) Rinaldi, R.; Schuth, F. ChemSusChem 2009, 2, 1096−1107.
̈
586
dx.doi.org/10.1021/cs300845f | ACS Catal. 2013, 3, 581−587

ACS Catalysis
Research Article
(7) Zhang, Y.-H. P.; Lynd, L. R. Biotechnol. Bioeng. 2004, 88, 797−
824.
(8) Su, Y.; Brown, H. M.; Huang, X.; Zhou, X.-D.; Amonette, J. E.;
Zhang, Z. C. Appl. Catal. A: Gen. 2009, 361, 117−122.
(9) Binder, J. B.; Raines, R. T. Proc. Natl. Acad. Sci. U.S.A. 2010, 107,
4516−4521.
(10) Fukuoka, A.; Dhepe, P. L. Angew. Chem., Int. Ed. 2006, 45,
5161−5163.
(11) Rinaldi, R.; Schuth, F. Energy Environ. Sci. 2009, 2, 610−626.
̈
(12) Kobayashi, H.; Ohta, H.; Fukuoka, A. Catal. Sci. Technol. 2012,
2, 869−883.
(13) Van de Vyver, S.; Geboers, J.; Jacobs, P. A.; Sels, B. F.
ChemCatChem 2011, 3, 82−94.
(14) Suganuma, S.; Nakajima, K.; Kitano, M.; Yamaguchi, D.; Kato,
H.; Hayashi, S.; Hara, M. J. Am. Chem. Soc. 2008, 130, 12787−12793.
(15) Onda, A.; Ochi, T.; Yanagisawa, K. Green Chem. 2008, 10,
1033−1037.
(16) Rinaldi, R.; Palkovits, R.; Schuth, F. Angew. Chem., Int. Ed. 2008,
̈
47, 8047−8050.
(17) Wang, H.; Zhang, C.; He, H.; Wang, L. J. Environ. Sci. 2012, 24,
473−478.
(18) Jun, S.; Joo, S. H.; Ryoo, R.; Kruk, M.; Jaroniec, M.; Liu, Z.;
Ohsuna, T.; Terasaki, O. J. Am. Chem. Soc. 2000, 122, 10712−10713.
(19) Kobayashi, H.; Komanoya, T.; Hara, K.; Fukuoka, A.
ChemSusChem 2010, 3, 440−443.
(20) Komanoya, T.; Kobayashi, H.; Hara, K.; Chun, W.-J.; Fukuoka,
A. Appl. Catal. A: Gen. 2011, 407, 188−194.
(21) Pang, J.; Wang, A.; Zheng, M.; Zhang, Y.; Huang, Y.; Chen, X.;
Zhang, T. Green Chem. 2012, 14, 614−617.
(22) Deng, W.; Liu, M.; Tan, X.; Zhang, Q.; Wang, Y. J. Catal. 2010,
271, 22−32.
(23) Boehm, H. P. Carbon 1994, 32, 759−769.
(24) McCormick, C. L.; Callais, P. A.; Hutchinson, B. H., Jr.
Macromolecules 1985, 18, 2394−2401.
(25) Sluiter, A.; Hames, B.; Ruiz, R.; Scarlata, C.; Sluiter, J.;
Templeton, D.; Crocker, D. NREL/TP-510-42618, 2011; http://
www.nrel.gov/biomass/pdfs/42618.pdf (accessed Oct. 11, 2012).
(26) Bobleter, O. Prog. Polym. Sci. 1994, 19, 797−841.
(27) Strachan, J. Nature 1938, 141, 332−333.
(28) Hick, S. M.; Griebel, C.; Restrepo, D. T.; Truitt, J. H.; Buker, E.
J.; Bylda, C.; Blair, R. G. Green Chem. 2010, 12, 468−474.
(29) Meine, N.; Rinaldi, R.; Schuth, F. ChemSusChem 2012, 5, 1449−
̈
1454.
(30) Kono, H.; Yunoki, S.; Shikano, T.; Fujiwara, M.; Erata, T.;
Takai, M. J. Am. Chem. Soc. 2002, 124, 7506−7511.
(31) Geboers, J.; Van de Vyver, S.; Carpentier, K.; Jacobs, P.; Sels, B.
Chem. Commun. 2011, 47, 5590−5592.
́
(32) Figueiredo, J. L.; Pereira, M. F. R.; Freitas, M. M. A.; Orfao, J. J.
̃
M. Carbon 1999, 37, 1379−1389.
(33) Capon, B. Tetrahedron Lett. 1963, 4, 911−913.
(34) Gazit, O. M.; Katz, A. Langmuir 2012, 28, 431−437.
(35) Gazit, O. M.; Charmot, A.; Katz, A. Chem. Commun. 2011, 47,
376−378.
(36) Gazit, O. M.; Katz, A. J. Am. Chem. Soc., in press, http://dx.doi.
org/10.1021/ja311918z.
(37) Nakajima, K.; Hara, M. ACS Catal. 2012, 2, 1296−1304.
(38) Kitano, M.; Yamaguchi, D.; Suganuma, S.; Nakajima, K.; Kato,
H.; Hayashi, S.; Hara, M. Langmuir 2009, 25, 5068−5075.
587
dx.doi.org/10.1021/cs300845f | ACS Catal. 2013, 3, 581−587