Chemocatalytic hydrolysis of cellulose at 37 °c, 1 atm

Article abstract of DOI:10.1039/c5cy01677k

The metal salt-Br?nsted acidic ionic liquid system composed of ZnCl₂·1.74H₂O-1-(1-propylsulfonic)-3-methylimidazolium chloride can directly hydrolyze untreated cellulose in 78% total reducing sugar and 19% glucose yield at 37 °C, 1 atm in 4.0 days. The new chemical catalyst is used at a temperature lower than typical cellulase enzyme operating conditions.

Full text of DOI:10.1039/c5cy01677k

Catalysis
Science &
Technology
View Article Online
View Journal
COMMUNICATION
Chemocatalytic hydrolysis of cellulose at 37 °C, 1
atm†
Cite this: DOI: 10.1039/c5cy01677k
*
Ananda S. Amarasekara and Bernard Wiredu
Received 1st October 2015,
Accepted 9th November 2015
DOI: 10.1039/c5cy01677k
www.rsc.org/catalysis
The metal salt – Brønsted acidic ionic liquid system composed of
ZnCl2·1.74H2O-1-(1-propylsulfonic)-3-methylimidazolium chloride
can directly hydrolyze untreated cellulose in 78% total reducing
sugar and 19% glucose yield at 37 °C, 1 atm in 4.0 days. The new
chemical catalyst is used at a temperature lower than typical
cellulase enzyme operating conditions.
research group introduced the incorporation of the –SO3H
acid catalyst function to IL, thus eliminating the use of a sep-
arate acid and considerably improving the efficiency of cellu-
lose hydrolysis.8–10 Where we first reported that Brønsted
acidic ionic liquid (BAIL) 1-(1-propylsulfonic)-3-methyl-
imidazolium chloride (PSMIMCl) can be used as an acid cata-
lyst as well as the solvent; yielding 62% TRS after 1.5 h, at 70
°C, 1 atm.8 Later we found that these sulfuric acid derivatives
with IL characteristics can be used as catalysts in aqueous
phase as well.9,11 For example a dilute aqueous solution of
PSMIMCl was shown to be a better catalyst than aq. H2SO4 of
the same H+ concentration for the degradation of cellulose
at 150–160 °C at 0.38–0.52 MPa.9 This enhanced catalytic
activity of the PSMIMCl when compared to H2SO4 can be
explained as a result of an interaction or binding of the BAIL
on the cellulose surface, which facilitates the approach of
–SO3H group for hydrolysis of the glycosidic link.12,13 This
observation can be seen as an important lead for the develop-
ment of a BAIL based cellulase mimic type catalyst for cellu-
lose hydrolysis.
Then there are new directions, such as the incorporation
of metal ions like Mn2+, Fe3+, Cu2+, and Zn2+ as Lewis acid co-
catalysts to the Brønsted acid system, shown to improve the
cellulose pretreatment and acid hydrolysis efficiencies.14–16 In
addition, we have recently found that addition of a catalytic
amount of Mn2+, Fe3+ or Co2+ chloride salt can be used to
improve the sugar yields during the acid catalyzed saccharifi-
cation of corn stover in water at 140–170 °C.17 As these new
discoveries point out the advantages of mixing Lewis and
Brønsted acid catalysts for cellulose hydrolysis, we hypothe-
sized that catalytic activity of the BAILs could also be further
enhanced by incorporating –OH chelating metal ions and
anions like Cl− to the catalyst. Consequently we have studied
the cellulose hydrolysis in a number of metal chloride–BAIL
systems to test this hypothesis. Where ZnCl2 based systems
showed much promise and here we report the first catalyst
that can hydrolyze untreated cellulose under conditions closer
to ambient temperature.
Cellulose is the most abundant biopolymer on earth and the
resourceful utilization of cellulose is a paramount factor in a
sustainable carbon based future. The hydrolysis of this
polysaccharide is a formidable challenge due to its complex
molecular architecture with stiff polymeric chains and close
packing via strong inter and intra-molecular hydrogen
bonds.1 The current technology for this process is high
temperature–pressure aq. sulfuric acid pretreatment followed
by the use of cellulase enzymes for the hydrolysis.1,2 The alter-
native method of mineral acid catalyzed hydrolysis gives poor
sugar yields and requires even higher temperature–pressure
conditions.3 While these operations are highly energy con-
suming and expensive, there is an urgent need for a more
efficient catalytic method for harvesting the full potential of
the most abundant renewable carbohydrate polymer.
Since the 2002 report on the use of 1-butyl-3-
methylimidazolium salts, which are room temperature ionic
liquids (ILs) for the dissolution of cellulose, considerable
efforts have been devoted to improve the solubility of cellu-
lose in ILs for processing of cellulosic biomass.4–6 Then, in
2007 Zhao et al. reported the first use of ILs as a solvent to
hydrolyze cellulose.7 Where they attained a 77% total reduc-
ing sugar (TRS) yield after 9 h at 100 °C by using conc. H2SO4
as the acid catalyst for cellulose dissolved in 1-butyl-3-
methylimidazolium chloride.7 In
a follow up work our
Department of Chemistry, Prairie View A&M University, Prairie View, Texas
77446, USA. E-mail: asamarasekara@pvamu.edu; Fax: +1 936 261 3117;
Tel: +1 936 261 3107
† Electronic supplementary information (ESI) available: Experimental procedures
and Ea calculation. See DOI: 10.1039/c5cy01677k
This journal is © The Royal Society of Chemistry 2015
Catal. Sci. Technol.

View Article Online
Communication
Catalysis Science & Technology
We have examined the cellulose hydrolysis by measuring
the TRS and glucose yields in mediums containing variable
amounts of PSMIMCl in ZnCl2. These mixtures contained
controlled amount of water as well, since water attack at the
glycosidic link is the key step in the hydrolysis. Our earlier
work using neat BAIL has shown that ∼2.0 equivalents of
water per glucose unit is sufficient for cellulose hydrolysis.8
As well as preliminary tests have revealed that anhydrous
ZnCl2 can be converted to a viscous, colorless, clear liquid of
hydrated ZnCl2 by the addition of 1.74 equivalents of water at
23 °C. This ZnCl2·1.74H2O was used throughout the study. In
the initial stages a mixture of 30% (w/w) BAIL in ZnCl2
·1.74H2O was used to find the optimum weight of cellulose
that can be hydrolyzed in a gram of solvent/catalyst medium.
During these experiments TRS produced from hydrolysis of
25–150 mg cellulose in 1.000 g of solvent/catalyst at 37 °C
were measured using 3,5-dinitrosalicylic acid (DNS) method10
and the results are shown in Fig. 1. Which revealed that 100
mg of cellulose in 1.000 g of 30% (w/w) BAIL–ZnCl2·1.74H2O
produces the highest TRS % yield; thus 100 mg of cellulose
per 1.000 g of catalyst/solvent were used in the next set of
experiments.
This optimization analysis shows that 30% (w/w) BAIL in
ZnCl2·1.74H2O produced the highest TRS % yield of 78% after
4.0 d at 37 0.1 °C. Moreover, a reduction in TRS yields are
observed in samples of more than 4.0 d, and this may be due
to the decomposition of sugars in the highly acidic
medium.18 The highest glucose yield of 19% was also
achieved in reactions carried out in the same 30% (w/w) BAIL
in ZnCl2·1.74H2O at 4.0 d. Additionally, the glucose yield also
decreased after 4.0 d as TRS (Fig. 2B). With the aim to check
the cellulose hydrolysis in only ZnCl2·1.74H2O medium we
have prepared a mixture containing 10% (w/w) cellulose in
ZnCl2·1.74H2O and incubated at 37 °C for 4 days. However no
reducing sugars were detected. Similar experiment with 10%
(w/w) cellulose in BAIL also failed to produce any reducing
sugars. The experiments carried out in ZnCl2–BAIL mixtures
with water contents higher than 1.74 H2O per mol of ZnCl2
also failed to produce sugars.
Next we have investigated the effects of reaction time and
the composition of the BAIL–ZnCl2 on TRS and glucose
yields. The cellulose samples were mixed in 0–40% BAIL in
ZnCl2·1.74H2O and incubated at 37.0 0.1 °C for 1–8 days.
After the specified period the samples were diluted with
water, neutralized (NaOH), unreacted cellulose and precipi-
tated zinc salts were removed by centrifugation (1700g) to
give clear sugar solutions; in which TRS and glucose were
measured10 (ESI†). The average TRS and glucose % yields
produced in 0–40% BAIL in ZnCl2·1.74H2O medium for 0–
8 days at 37 0.1 °C are presented in the 3D plots 2A and 2B
respectively.
Fig.
2 (A) Total reducing sugar (TRS) and (B) glucose % yields
produced during the cellulose hydrolysis using 0–40% BAIL in ZnCl2
·1.74H2O for 0–8 d, 37 0.1 °C. 100 mg Sigmacell cellulose – type 101
(DP ∼ 450, from cotton linters) was used. Averages of duplicate
experiments.
Fig. 1 The total reducing sugar (TRS) % yields produced during the
hydrolysis of Sigmacell cellulose – type 101 (DP ∼ 450, from cotton
linters) in 30% (w/w) BAIL in ZnCl2·1.74H2O at 37 °C, 4.0 d.
Catal. Sci. Technol.
This journal is © The Royal Society of Chemistry 2015

View Article Online
Catalysis Science & Technology
Communication
Furthermore, we have carried out an experiment at 50 °C,
for 4 days to check if the glucose yield improves at a higher
temperature. Unfortunately at 50 °C, glucose yield slightly
decreased to 16% from earlier 19%; nevertheless TRS yield
remained the same. Even though the new catalyst system
gives a good TRS yield, it produces only a low glucose yield.
This may be due to well-known19 dehydration of glucose into
5-hydroxymethylfurfural and subsequent decomposition and
polymerization under highly acidic reaction conditions.
In order to compare the new catalyst system with cellulose
hydrolysis using the standard enzymatic hydrolysis under
NREL/TP-5100-63351 protocol20 we have carried out an addi-
tional experiment. Cellulase from Trichoderma reesei ATCC
26921, which catalyzes the breakdown of cellulose into glu-
cose, cellobiose, and higher glucose polymers was used as
the enzyme cocktail. The TRS and glucose in the hydrolyzate
were measured using DNS method and glucose oxidase–per-
oxidase assay as in the BAIL in ZnCl2·1.74H2O catalyst system
experiments. The results of this experiment showing the
changes in TRS and glucose yields with time are shown as
Fig. 3. The results of the standard NREL protocol enzymatic
hydrolysis experiment carried out at 50 °C shows higher TRS
and glucose yields than the new catalyst system presented in
this work. However, the new BAIL in ZnCl2·1.74H2O catalyst
works at 37 °C, and does not require a pre-treatment of cellu-
lose before the hydrolysis.
Fig. 4 13C-NMR spectrum of α/β D-cellobiose in ZnCl2·4D2O (50 mg/
0.6 mL) at 23 °C.
mediums. The 13C spectrum in D2O was used as a reference
and the chemical shifts changes (Δδ) were calculated for each
carbon in α/β-D-cellobiose (Fig. 4) for the solvent change from
D2O to ZnCl2·4D2O and these Δδ values are shown in
Fig. 5.22,23 Which shows that changing the solvent to ZnCl2
·4D2O results in significant chemical shift changes in a num-
ber of carbons, suggesting a strong interaction between cello-
biose and ZnCl2. The largest decreases in chemical shifts of
−0.549, −0.696 and −0.851 ppm were observed for C6, C6′-α
and C6′-β respectively. The major increases in δ of 0.66,
0.495, 0.513 and 0.359 ppm were observed for C5, C1′-α, C1′-
β and C5′-β. This clearly indicates that Zn2+ cations and/or
Cl− anions interacts or binds with a specific region of cellobi-
ose. The highest chemical shift changes occur in C6, addi-
tionally all C6 peaks showed peak broadening as well.
In an attempt to explain the unprecedented catalytic activ-
ity in the ZnCl2–BAIL system we have studied the interaction
of cellulose model compound cellobiose with ZnCl2 and
ZnCl2–BAIL using NMR spectroscopy. The initial attempts to
1
record H and 13C NMR of cellobiose in ZnCl2·1.74D2O at 23
°C was not successful due to high viscosity of the medium.
However, dilution of the ZnCl2 up to ZnCl2·4D2O allowed
recording good 13C spectra after ∼15k scans (Fig. 4). Though,
1H NMR still produced broad unresolved peaks. With this
tool we have studied the 13C spectra of cellobiose in D2O,21
These effects are most probably due to chelation with the
C6 hydroxyl groups. In the next stage we have recorded three
additional 13C spectra after adding increasing amounts of
BAIL into the same NMR tube to provide cellobiose : BAIL
mole ratios 1 : 0.06, 1 : 0.48 and 1 : 1.14 respectively. The Δδ
values in these three mediums are also shown in Fig. 5. The
BAIL modified ZnCl2 medium Δδ curves are very close to the
first Δδ curve obtained without BAIL. This may be due to a very
small or negligible effect by the addition of the BAIL on the
interaction/binding of ZnCl2 with carbohydrate. Therefore,
ZnCl2·4D2O and in three different ZnCl2·4D2O
+ BAIL
Fig. 3 The changes in TRS and glucose yields with time for Sigmacell
cellulose type 101 (DP 450, from cotton linters) hydrolyzed
according to NREL/TP-5100-63351 protocol20 using cellulase enzyme
from Trichoderma reesei ATCC 26921 at 50 °C, pH = 5.0, sodium
citrate buffer.
–
∼
Fig. 5 13C NMR chemical shift changes (Δδ) of D-cellobiose due to
different solvent mediums at 23 °C. Chemical shifts changes are
measured relative to D-cellobiose in D2O.
This journal is © The Royal Society of Chemistry 2015
Catal. Sci. Technol.

View Article Online
Communication
Catalysis Science & Technology
the NMR data suggests that ZnCl2 plays an important role by
interacting with carbohydrate and probably changing the
conformation of the polysaccharide, facilitating the BAIL cat-
alyzed hydrolysis.
4 R. P. Swatloski, S. K. Spear, J. D. Holbrey and R. D. Rogers,
J. Am. Chem. Soc., 2002, 124, 4974–4975.
5 C. Z. Liu, F. Wang, A. R. Stiles and C. Guo, Appl. Energy,
2012, 92, 406–414.
Furthermore, we have calculated the activation energy (Ea)
for the hydrolysis of cellulose in 30% (w/w) BAIL in ZnCl2
·1.74H2O at 25–37 °C by using the first order rate constants
6 H. F. N. De Oliveira, C. Farès and R. Rinaldi, Chem. Sci.,
2015, 6, 5215–5224.
7 C. Li and Z. K. Zhao, Adv. Synth. Catal., 2007, 349,
1847–1850.
as 121
2 kJ mole−1 (ESI†). The generally accepted Ea for
H2SO4 catalyzed cellulose hydrolysis in water is 170–185 kJ
mole−1.24,25 Therefore the low Ea in the new catalyst further
supports the synergistic effect of BAIL and hydrated ZnCl2.
The mechanism of hydrolysis may involve the transfer of
acidic proton of the BAILs –SO3H group to glycosidic oxygen
of cellulose, resulting protonation of the oxygen. The co-
catalyst ZnCl2 may facilitate the protonation by interaction
with the polysaccharide and changing the conformation, as
evident from the 13C NMR study of the model compound.
Next, the attack of water at glycosidic carbon can cleave the
C–O bond, completing the hydrolysis. As further studies of
the present approach we are proposing to study the recycling
of the catalysts after separation from sugars by using a com-
bination of precipitation of Zn salts and ion exchange chro-
matography methods. In addition we are currently working
on improving the glucose yield.
8 A. S. Amarasekara and O. S. Owereh, Ind. Eng. Chem. Res.,
2009, 48, 10152–10155.
9 A. S. Amarasekara and B. Wiredu, Ind. Eng. Chem. Res.,
2011, 50, 12276–12280.
10 A. S. Amarasekara and P. Shanbhag, BioEnergy Res., 2013, 6,
719–724.
11 A. Amarasekara and B. Wiredu, BioEnergy Res., 2014, 7,
1237–1243.
12 R. S. Payal and S. Balasubramanian, Phys. Chem. Chem.
Phys., 2014, 16, 17458–17465.
13 K. M. Gupta and J. Jiang, Chem. Eng. Sci., 2015, 121,
180–189.
14 S. Zhao, P. Li, Q. Zhang, J. Zhang and L. Kong, Res. Chem.
Intermed., 2013, 39, 3803–3812.
15 S. R. Kamireddy, J. Li, M. Tucker, J. Degenstein and Y. Ji,
Ind. Eng. Chem. Res., 2013, 52, 1775–1782.
16 S. Monavari, M. Galbe and G. Zacchi, Bioresour. Technol.,
2011, 102, 1103–1108.
17 B. Wiredu and A. S. Amarasekara, Bioresour. Technol.,
2015, 189, 405–408.
18 H. Ren, B. Girisuta, Y. Zhou and L. Liu, Carbohydr. Polym.,
2015, 117, 569–576.
19 S. Dutta, S. De and B. Saha, Biomass Bioenergy, 2013, 55,
355–369.
20 J. O. Resch, M. G. Baker and S. R. Decker, Low solids
enzymatic saccharification of lignocellulosic biomass, NREL
Report, NREL/TP-5100-63351, 2015.
Conclusions
The high catalytic activity observed in this study may be due
to interaction of ZnCl2 with cellulose, disrupting the
H-bonding network and similar interactions of carbohydrates
with Mn2+ (ref. 26) and Fe3+ (ref. 27) are reported in the liter-
ature. As far as we are aware the mildest conditions reported
for a chemocatalytic cellulose hydrolysis is 70 °C, 1 atm, for
1.5 h, in neat PSMIMCl.8 Therefore the present result is a sig-
nificant improvement and the first example of a cellulose
hydrolysis by a chemically catalyzed system in conditions
close to room temperature.
21 A. S. Amarasekara, O. S. Owereh and B. Ezeh, Carbohydr.
Res., 2011, 346, 2820–2822.
22 M. U. Roslund, P. Tähtinen, M. Niemitz and R. Sjöholm,
Carbohydr. Res., 2008, 343, 101–112.
Acknowledgements
23 J. C. Gast, R. H. Atalla and R. D. McKelvey, Carbohydr. Res.,
1980, 84, 137–146.
We thank NSF grants CBET-1336469 and HRD-1036593 for
financial support.
24 L. Shuai and X. Pan, Energy Environ. Sci., 2012, 5, 6889–6894.
25 L. V. A. Gurgel, K. Marabezi, M. D. Zanbom and A. A. D. S.
Curvelo, Ind. Eng. Chem. Res., 2012, 51, 1173–1185.
26 F. Tao, H. Song, J. Yang and L. Chou, Carbohydr. Polym.,
2011, 85, 363–368.
References
1 A. S. Amarasekara, Handbook of Cellulosic Ethanol, John
Wiley & Sons, 2013.
2 D. B. Wilson, Appl. Microbiol. Biotechnol., 2012, 93, 497–502.
3 R. Rinaldi and F. Schüth, ChemSusChem, 2009, 2, 1096–1107.
27 X. Kästele, C. Sturm and P. Klüfers, Eur. J. Pharm. Biopharm.,
2014, 86, 469–477.
Catal. Sci. Technol.
This journal is © The Royal Society of Chemistry 2015