Synergizing graphene oxide with microwave irradiation for efficient cellulose depolymerization into glucose

Article abstract of DOI:10.1039/c7gc01691c

The abundance in hydroxyl groups neighboring the hydrolytic site (β-1,4 glycosidic bonds) in cellulose forms a tightly packed crystalline structure, and hence, is a major challenge in cellulose depolymerization into fermentable sugars. Herein, we report on the synergistic effect of the carbocatalyst, graphene oxide (GO) and microwave irradiation (MW) on cellulose depolymerization in the absence of pretreatment. Results showed that microcrystalline cellulose (MCC) can be effectively depolymerized into glucose in as short as 30 s at 473 K (800 W). Generated yields without pretreatments were as high as 61 ± 4% within 60 min (200 W and 453 K), which has never been reported before to the best of our knowledge. The synergy of GO and MW was demonstrated by high heating rates and which allowed for the activation of in situ crystalline-to-amorphous transformations and suppressed immediate formation of degradation products. The activity of GO under MW was much higher than those of other solid acid catalysts such as Amberlyst 15, sulfated zirconia, and phosphotungstic acid due to the variety of surface functionalities and its high microwave absorptivity. The depolymerization mechanism was also discussed in terms of surface attrition of MCC as observed via scanning electron microscopy. On this basis, the proposed process offers an important strategy for cellulose depolymerization.

Full text of DOI:10.1039/c7gc01691c

View Article Online
View Journal
Green
Chemistry
Accepted Manuscript
This article can be cited before page numbers have been issued, to do this please use: E. G. G. Mission,
A. T. Quitain, M. Sasaki and T. Kida, Green Chem., 2017, DOI: 10.1039/C7GC01691C.
This is an Accepted Manuscript, which has been through the
Volume 18 Number
7 7 April 2016 Pages 1821–2242
Royal Society of Chemistry peer review process and has been
accepted for publication.
Green
Chemistry
Accepted Manuscripts are published online shortly after
Cutting-edge research for a greener sustainable future
www.rsc.org/greenchem
acceptance, before technical editing, formatting and proof reading.
Using this free service, authors can make their results available
to the community, in citable form, before we publish the edited
article. We will replace this Accepted Manuscript with the edited
and formatted Advance Article as soon as it is available.
You can find more information about Accepted Manuscripts in the
author guidelines.
Please note that technical editing may introduce minor changes
to the text and/or graphics, which may alter content. The journal’s
standard Terms & Conditions and the ethical guidelines, outlined
in our author and reviewer resource centre, still apply. In no
event shall the Royal Society of Chemistry be held responsible
for any errors or omissions in this Accepted Manuscript or any
consequences arising from the use of any information it contains.
ISSN 1463-9262
CRITICAL REVIEW
G. Chatel et al.
Heterogeneous catalytic oxidation for lignin valorization into valuable
chemicals: what results? What limitations? What trends?
rsc.li/green-chem

Please do not adjust margins
Page 1 of 13
Green Chemistry
View Article Online
DOI: 10.1039/C7GC01691C
Journal Name
ARTICLE
Synergizing graphene oxide with microwave irradiation for
efficient cellulose depolymerization into glucose
Elaine G. Missiona, Armando T. Quitain*b,c, Mitsuru Sasakid and Tetsuya Kida*b
Received 00th January 20xx,
Accepted 00th January 20xx
The abundance in hydroxyl groups neighboring the hydrolytic site (β-1,4 glycosidic bonds) in cellulose forms a tightly
packed crystalline structure, and hence, is a major challenge in cellulose depolymerization into fermentable sugars. Herein,
we report on the synergistic effect of the carbocatalyst, graphene oxide (GO) and microwave irradiation (MW) on cellulose
depolymerization in the absence of pretreatment. Results showed that microcrystalline cellulose (MCC) can be effectively
depolymerized into glucose in as short as 30 s at 473 K (800 W). Generated yields without pretreatments were as high as
61±4% within 60 min (200W and 453 K), which has never been reported before to the best of our knowledge. The synergy
of GO and MW was demonstrated by high heating rates and which allowed for the activation of in situ crystalline-to-
amorphous transformations and suppressed immediate formation of degradation products. The activity of GO under MW
was much higher than those of other solid acid catalysts such as Amberlyst 15, sulfated zirconia, and phosphotungstic acid
due to the variety of surface functionalities and its high microwave absorptivity. The depolymerization mechanism was
also discussed in terms of surface attrition of MCC as observed via scanning electron microscopy.
On this basis, the proposed process offers an important strategy for cellulose depolymerization.
DOI: 10.1039/x0xx00000x
www.rsc.org/
proton away via competitive protonation from β-1,4 glycosidic
bonds5. Third, cellulose comprises of crystalline and
amorphous or low ordered domains, and this heterogeneity
adds complexity to the system6. Additionally, even after
successful protonation of the glycosidic O-site, conformational
changes are necessary for cleaving to take place5.
1. Introduction
Conversion of lignocellulosic biomass provides an alternative
stable source of biofuel and biochemical precursors as it
captures the biomass fractions that would have otherwise
been eliminated via open burning, consequently reducing the
carbon dioxide being emitted into the atmosphere. Biomass
contains about 35-50% cellulose, a non-edible biopolymer
found in the primary cell wall of green plants and most algae.
It has an estimated global availability of 70 billion tons1, and
which could be depolymerized into its glucose monomers
through hydrolysis.
Cellulose depolymerization into glucose often entails
elevated temperatures and strong acids due to the recalcitrant
structure of cellulose. First, the D-glucose units are connected
in a criss-cross conformation, concealing the β-1,4 glycosidic
site for immediate scission. Second, the hydroxyls neighboring
the β-1,4 glycosidic bonds develop intra- and inter-chain
hydrogen bonds forming the tightly packed crystalline
structure, rendering cellulose insoluble to water and most
organic compounds2-4. These hydroxyls would tend to sift the
Hence, an effective cellulose depolymerization requires
decrystallization, dissolution and hydrolytic scission of the β-
1,4 glycosidic bonds. Enzymatic hydrolysis has been the
traditional pathway, however, this method suffers from slow
conversion rates and is generally expensive due to enzyme
handling and mutant strains development for increased
selectivity7,8. Homogeneous acid catalysis with sulfuric and
hydrochloric9,10 has been a prominent alternative since they
can easily permeate into the bulk crystalline structure.
Unfortunately, their use impacts the quality of hydrolysis
products and poses problems on catalyst recovery,
neutralization waste and reactor corrosion11. Subcritical and
supercritical conditions4,12-15 were also applied to weaken
hydrogen bonds and increase the solvating property of water.
Yet, this process suffers from poor selectivity to glucose and
requires higher energy to operate. These issues have been
addressed through the development of various types of solid
acid catalysts that includes metal oxides, polymer-based acids,
sulfonated carbonaceous acids, heteropolyacids, zeolites,
a Graduate School of Science and Technology, Kumamoto University, Kumamoto,
Japan
b Faculty of Advanced Science and Technology
supported metal catalysts and magnetic solid acids16-21
.
c International Research Organization for Advanced Science and Technology
d Institute of Pulsed Power Science, Kumamoto University, Kumamoto, Japan
Email: quitain@kumamoto-u.ac.jp; tetsuya@kumamoto-u.ac.jp
Electronic Supplementary Information (ESI) available: Characterization of graphene
oxide, microwave heating profile, TRS analysis, Formation of degradation products;
spent GO characterization, Microcystalline and ballmilled cellulose XRD spectra.
See DOI: 10.1039/x0xx00000x
However, catalytic performance of aforementioned solid
acids was found to be relatively low due to the following
reasons: (i) limited interaction with bulk cellulose, (ii) presence
of strong acid sites but absence of binding sites and (iii)
This journal is © The Royal Society of Chemistry 20xx
J. Name., 2013, 00, 1-3 | 1
Please do not adjust margins

Please do not adjust margins
Green Chemistry
Page 2 of 13
ARTICLE
Journal Name
View Article Online
DOI: 10.1039/C7GC01691C
thermal instability22,23
.
For these reasons, several phosphotungstic acid and functionalized GO. The influence of
pretreatments that would increase the reactivity of cellulose process variables such as temperature, reaction time, catalyst
prior to hydrolysis have been developed, such as dissolution and solvent loading on glucose yields and hydrolysis products
with ionic liquids and decrystallization through ball milling. For were also studied. Furthermore, we elucidated a potential
instance, Rinaldi et al. reported the dissolution of cellulosic reaction mechanism that describes the depolymerization
substrates in 1-butyl-3-methylimidazolium chloride (BMIMCl) process.
for conversion to reducing sugars in the presence of Amberlyst
15DRY resin24. However, the process was relatively inclined to
cello-oligomer production with poor fermentable sugars yield
2. Experimental
of ~3% after 4 h at 373 K. On the other hand, Yabushita et al.
ball milled microcrystalline cellulose (MCC) with alumina to
increase contact surface between the two25. After 48 hr
ballmilling and 24 h reaction time at 418 K, about 72% glucose
yield was achieved. Unfortunately, with the pretreatment
process much longer than the actual hydrolysis process, scaling
up is rather impractical.
2.1 Chemicals
Avicel microcrystalline cellulose (MCC) was purchased from
Merck, USA while glucose, levulinic acid, formic acid and 5-
hydroxymethylfurfural standard reagents were purchased
from Wako Pure Chemical Industries (Osaka, Japan). Graphite
powder, sulfuric acid (98%), formic acid (99%), acetic acid
(99%), sodium nitrate, potassium permanganate, hydrogen
peroxide (30%), hydrochloric acid (35%), dinitrosalicylic acid
and potassium tartrate were all purchased from Wako Pure
Chemical Industries (Osaka, Japan). Amberlyst 15 and
phosphotungstic acid were purchased from Sigma Aldrich
(France) and Alfa Aesar (USA), respectively while sulphated
zirconia (JRC-ZRO-1) was obtained from Japan Reference
Catalyst. Single-walled carbon nanotubes were obtained from
Sigma Aldrich, USA. Carbonic acid and amine functionalized GO
were prepared in-house using protocols developed in our lab,
as stated elsewhere35. All chemicals were used as received.
In this light, we have synergized the carbocatalyst
graphene oxide (GO) with microwave irradiation (MW) to carry
out one-pot decrystallization, dissolution and hydrolysis of
cellulose, thus eliminating the need for a separate pre-catalytic
stage. GO is the water dispersible precursor of graphene
produced from the chemical exfoliation of graphite powders. It
has been attracting interest in catalysis26-28 due to its layered
soft-structure and abundant functional groups. Unlike
conventional solid catalysts that typically contains single type
of functionalities, the presence of hydroxyls, carboxyls,
carbonyls and epoxy groups along the basal plane and edges29-
31
of GO would allow weak hydrogen interactions and can
2.2 Graphene oxide synthesis and characterization
potentially overcome competitive protonation that arises
during protonation of β-1,4 glycosidic bonds. Moreover, GO
has the advantage of being a versatile nanoscaffold, with its
surface allowing for manipulation of functionalities and
construction of nanostructures with tailored properties32. On
the other hand, MW, as a form of electromagnetic energy,
converts to thermal energy by heating polar materials, such as
water33, which then dissipates heat to the bulk of the system.
This leads to the reverse heating phenomena, where heat
emanates from within the system and so it circumvents the
induction stage prevailing under conventional heating. MW
would likely accelerate reactions by enhancing the interaction
between cellulose and GO, providing conformational changes
for effective molecular interaction and delivering the
necessary energy to cleave the hydrolytic bond, significantly
cutting down reaction time and energy use34. In addition,
when used for biomass depolymerization, MW can take
advantage of the bound water originally contained in the
substrate, eliminating the need for drying or an additional
solvent.
Graphene oxide was synthesized following the modified
Hummer’s method36. First, graphite (4 g) and NaNO3 (4 g) were
mixed with H2SO4 (184 ml) in an ice bath followed by the slow
addition of KMnO4 (20 g) for oxidation. The mixture was then
stirred for 2 h in an oil bath maintained at 308 K and was
transferred back to the ice bath where distilled water (184 ml)
was introduced dropwise. The mixture was then stirred at 368
K for 1 h. Distilled water (400 ml) was added dropwise while in
ice bath and the reaction was terminated by the addition of
about 50ml H2O2. The resulting yellow mixture was centrifuged
at 4000 rpm for 10min. The residue was washed with 5% HCl
and centrifuged at 4000 rpm for 10 min thrice followed by
three-time distilled water washing with centrifugation at 4000
rpm for 30 min. The supernatant was disposed of while the
thick suspension was diluted with water and sonicated for 4 to
6 h. Finally, the suspension was centrifuged for 30 min at
10000 rpm, and the liquid was dried at 333 K for 3 to 4 days.
The resulting GO sheets were pulverized to obtain the GO
powder.
GO characterization was carried out using x-ray diffraction
(XRD), Fourier transform infrared spectroscopy (FTIR),
elemental analysis (EA), Brunauer-Emmett-Teller (BET) and
thermogravimetric analysis (TGA). Structural changes were
determined by XRD analysis using Rigaku miniflex 600. XRD
In this paper, we report on the synergism between MW
and GO for microcrystalline cellulose (MCC) depolymerization
process with water as the sole solvent. The catalytic
performance of GO was compared with other solid acid
catalysts such as Amberlyst 15, sulfated zirconia,
2 | J. Name., 2017, 00, 1-13
This journal is © The Royal Society of Chemistry 20xx
Please do not adjust margins

Page 3 of 13
Green Chemistry
Please do not adjust margins
Journal Name
ARTICLE
View Article Online
DOI: 10.1039/C7GC01691C
ꢁ
⁄
퐺푙푢푐표푠푒 ꢀ푛 ꢂℎ푒 ℎ푦푑푟표푙푦푠푎ꢂ푒∗0.9 (
ꢁ
)
patterns were generated with CuKα radiation (λ = 0.154 nm, U
= 40 kV, I = 40 mA) in the 2θ range of 2-600, a scan speed and
scan step of 0.020. The obtained XRD patterns were analyzed
using PDXL-2 analysis software (Rigaku Corporation). The
presence of functional groups were determined by FTIR
technique using JASCO FTIR 4100. The samples were milled
with KBr and compacted into KBr disks. Transmittance were
scanned at the wavenumber range of 4000-400cm-1. The
surface area, mean pore diameter and total pore volume of
the solid catalysts were measured using Belsorp mini II (BEL
Japan Inc.). Samples were filled in a tube under N2 atmosphere
퐿
퐺푌(푤푡%) =
푥 100
[1]
[2]
⁄
퐼푛ꢀꢂꢀ푎푙 푐푒푙푙푢푙표푠푒 ꢀ푛푝푢ꢂ (
)
퐿
ꢁ
⁄
퐺푙푢푐표푠푒 (
)
ꢁ
퐿
퐺푟푎푝ℎ푒푛푒 표ꢃꢀ푑푒 (
퐶푎푡푎푙푦푠푡 푃푟표푑푢푐푡푖푣푖푡푦 =
⁄
)
퐿
Yields of degradation products (DP) such as formic acid, 5-
hydroxymethylfurfural and levulinic acid have been
individually calculated from the initial substrate concentration
according to Eq. 3,
ꢁ
⁄
퐶표푛푐푒푛ꢂ푟푎ꢂꢀ표푛 ꢀ푛 ℎ푦푑푟표푙푦푠푎ꢂ푒(
)
퐿
(
)
퐷푃 푤푡% =
푥 100
[3]
ꢁ
and then degassed for 24
h at 333 K prior to the
⁄
퐼푛ꢀꢂꢀ푎푙 푐푒푙푙푢푙표푠푒 ꢀ푛푝푢ꢂ (
)
퐿
measurements. Thermal stability of GO was determined by
TGA and was carried out with Seiko Instruments Inc. Exstar
6000 TG/DTA 6200. About 400 mg of samples placed on
ceramic pan were heated from 308 to 1253 K at 283 K/min
heating rate under N2 gas.
For uncatalyzed runs with no glucose detection, TOC
measurement was performed to determine the water soluble
products that were not converted into glucose using Shimadzu
TOC-V CSN Total organic carbon analyzer. Total reducing
sugars (TRS) were analyzed using dinitrosalicylic acid (DNS)
method38. A mixture of 1 ml of hydrolysate diluted with
distilled water to 3 ml and 3 ml of DNS was heated for 5 min at
373 K then cooled to room temperature. The absorbance was
measured at 540 nm using JASCO V660 spectrophotometer.
The concentration was quantified using a 5-point calibration
curve with glucose as standard and the yield was obtained
using Eq. 4,
2.2 Depolymerization reactions
Experiments were conducted using Microwave Assisted
Reactor system (MARS5, CEM Matthews NC Company, USA)
equipped with a fiber optic thermocouple and pressure sensor.
In a typical run, MCC, GO powder and distilled water at
prescribed amounts were placed in the XP1500 reactor,
sealed, mounted on to the holder and subjected to varying
MW conditions. The samples were irradiated from room
temperature until the desired reaction temperature was
reached. Under these conditions, the pressure of the system
was slightly higher than the saturated vapor pressure of water
at the specified temperature. At the end of each reaction,
samples were cooled down and the liquid components
(hydrolysates) and solid residues were separated via filtration
in cold water bath. The residues were washed with 20 ml
distilled water and dried at 333 K for 48 h. The duration for
heating up was recorded, and corresponding heating rates
were calculated in K/s. Conventional hydrothermal methods
were carried out using a batch type SUS reactor (inner volume
= 8.8 ml) and a heating electric furnace37.
ꢁ
⁄
푇푅푆 푐표푛푐푒푛ꢂ푟푎ꢂꢀ표푛 ꢀ푛 ℎ푦푑푟표푙푦푠푎ꢂ푒(
)
퐿
(
)
=
푇푅푆 푤푡%
푥 100
[4]
ꢁ
⁄
퐼푛ꢀꢂꢀ푎푙 푐푒푙푙푢푙표푠푒 ꢀ푛푝푢ꢂ (
)
퐿
Powder and residue analyses. MCC and residues were
characterized by XRD and TGA using the same equipment
mentioned earlier. Crystallinity index (CrI%) was estimated
using Segal’s peak height method39 following Eq. 5.
퐼200−퐼퐴푚
퐶푟퐼 (%) =
푥 100
[5]
퐼200
where I200 = maximum intensity above baseline at 2θ = 220 to
230 and IAm is the minimum intensity above baseline
corresponding to amorphous content at 2θ = 180 to 190. The
crystallite size was estimated following the Scherrer equation15
2.4 Analytical Methods
in Eq. 6:
퐾휆
Liquid product analyses. The liquid products were analyzed
using high performance liquid chromatography (HPLC). Total
organic carbon (TOC) and total reducing sugars (TRS) were also
measured. Compounds were detected using a Shodex Sugar
SH1011 column at 333 K and refractive index (RI) detector. The
eluent, 3mM HClO4, was pumped at 0.5 mL/min. Samples were
filtered using a 0.45 μm syringe filter, after which 10 μL was
injected into the system. Peaks were identified by comparing
retention times of the standard compounds with the sample
then product quantification was done using individual
calibration curves. Glucose yield (GY) and catalytic productivity
are defined in Eq. 1 and 2 as follows:
퐷(푛푚)
=
[6]
훽1⁄ 푐표푠휃
2
where D is the crystallite size in nm; θ is the Bragg angle; λ is
the wavelength of the radiation; K is a constant and β1/2 is
corrected width of the peak given by the specimen.
Surface morphology of the samples was analyzed with a
JEOL JSM-7600F field emission scanning electron microscopy
(FESEM). Images were taken at x100 and x1000 magnifications
at 2 kV filament voltage. Samples were sputter coated with
gold using Quorum Technologies SC7620 Mini Sputter Coater
prior to the analysis.
This journal is © The Royal Society of Chemistry 20xx
J. Name., 2017, 00, 1-13 | 3
Please do not adjust margins

Please do not adjust margins
Green Chemistry
Page 4 of 13
ARTICLE
Journal Name
View Article Online
DOI: 10.1039/C7GC01691C
3.2 Synergism between microwave irradiation and graphene
oxide during ramping time
3. Results and discussions
3.1 Graphene oxide characterization
The typical heating profile under microwave irradiation
consists of two periods as shown in Fig. S2. The ramping time,
which is the duration it takes to heat up the irradiated material
from room temperature until the desired temperature, is
synonymous to heating up time in conventional heating;
whereas, the holding time, which is the duration at which
reaction is carried out at the desired temperature. In this
manuscript, reaction time refers to the combined ramping
time, and holding time measured from the time when the
desired temperature was reached.
GO synthesized in house through modified Hummer’s
method resulted in a dark brown-black powder. Structural
characterization was performed through various analyses such
as XRD, FTIR, EA and BET in comparison with graphite (Gr). The
XRD pattern (Fig. S1a) of GO shows a main reflection at ~10.2O,
corresponding to the 001 planes of GO in contrast to the peak
Gr exhibited at 26O. The disappearance of the 26O peak in GO
indicates complete exfoliation during the preparation
method36,40. It has an interlay space (d-spacing) of 0.82 nm
that indicates good intercalation of oxygen functionalities
between layers compared with Gr having d = 3.4 nm. The FTIR
spectra (Fig. S1b) confirmed the presence of functional groups
on its surface. The broad peak at ~3393 cm-1 and 1406 cm-1
corresponds to the O-H bond stretching vibration indicating
the presence of hydroxyl groups from adsorbed water. The
peaks at 1736cm-1, 1626cm-1, 1328cm-1, 1221 cm-1 and
1046cm-1 could be assigned to the oxygen containing groups
C=O (carbonyl/ carboxylic groups), C=C (aromatics), C-O
(carboxylic acid groups), C-O-C (epoxy) and C=OH (alkoxy),
respectively41. The degree of oxidation, represented by C/O
ratio based from EA (Table S1), is about 0.97 while the S
content is about 0.6%. From the BET-N2 analysis, its specific
surface area is 35m2/g. Based from the aforementioned
properties, the synthesized GO is comparable to that of
commercially available GO powders such as by DFJ
Nanotechnologies China.
In order to isolate MW inherent effects, we first analyzed
reactions with or without catalysts during ramping period from
room temperature to 473 K. We have included graphite (Gr)
and carbon nanotubes (CNT) for comparison. Heating rates
were found to demonstrate the synergism between MW and
GO (Fig. 1). In the absence of GO, only water acted as MW
absorber since MCC interacts poorly with MW. Because of this,
at 200 W, the system failed to reach 473 K during ramping
even past 60 min. This may be ascribed to the change in the
dielectric loss factor of water as temperature increases,
affecting its ability to absorb MW42. Such was not observed
with MW+GO, demonstrating the cumulative interaction of GO
and water with MW. As water decreases its capability to
absorb MW, GO acted as an additional MW receptor boosting
heating within the system. Besides, the ability of a material to
dissipate heat energy, expressed in terms of dielectric loss
constants (tan δ), of GO43 was reported to be ~0.577 at 2.45
GHz and 298K, higher than that of distilled water44 (0.118), by
which GO has supplemented any change caused by the
temperature rise. Increase in MW power generally resulted to
shorter ramp times and higher heating rates, with MW+GO
systems producing three times higher heating rates than
uncatalyzed ones. Gr and CNT, both being carbon based
materials, also demonstrated interaction with MW. Gr and
CNT has tan δ of 0.55645 and 0.69544, respectively,
8
MW + GO
7
MW only
MW+ Gr
MW + CNT
6
5
4
3
2
1
0
In order to understand the mechanism of MCC
depolymerization during ramping period, CrI and crystallite
size of the residue after reactions were analyzed. In the
absence of GO, only a slight increase in CrI (Fig. 2b-d) was
observed, corresponding to a small portion of amorphous
cellulose that has been removed and dissolved, leaving the
bulk crystalline regions untouched. The almost unaltered
crystallite size conforms to this findings (Table 1). In contrast,
CrI was found to significantly decrease in MW+GO systems
(Fig. 2e-h), i.e. from 68 to 63 nm at 200 W, owing to the partial
disruption of crystalline regions and conversion to amorphous
200
400
600
800
Microwave power (W)
Fig. 1 Comparison of heating rates between uncatalyzed and catalysed
reactions during ramping from room temperature to 473 K at various MW
power levels (Reaction conditions: 0.1 g MCC, 0.1 g GO, 10 ml distilled
water, zero holding time. Ramping to 473 K at 200 W for uncatalyzed
reaction has not reached the desired temperature even after 60 min.)
cellulose46,47 This is likely indicative of an effective interface
.
between GO and the MCC bulk which may be initiated at the
reducing ends. The phenolic -OH groups on the basal plane of
GO could have formed intermediate hydrogen bonds with the
4 | J. Name., 2017, 00, 1-13
This journal is © The Royal Society of Chemistry 20xx
Please do not adjust margins

Page 5 of 13
Green Chemistry
Please do not adjust margins
Journal Name
ARTICLE
View Article Online
DOI: 10.1039/C7GC01691C
CrI %
CrI %
±
a) Avicel : 68 2
162 s*
±
e) 200 W: 63 4
±
f) 400 W: 62 2
±
b) 400 W: 69 2
63 s
262 s*
±
c) 600 W: 70 1
g) 600 W: 64 1
±
41 s
30 s
145 s
94 s
±
d) 800 W: 69 1
±
h) 800 W: 60 2
10
20
30
40
10
20
30
40
2θ, deg
2θ, deg
Fig 2. X-ray diffraction patterns of unreacted MCC [a] in comparison with residual cellulose after depolymerization reactions using microwave (MW)
alone [b-d] and the synergism of MW and graphene oxide (GO) [e-h] during ramping from room temperature to 473 K at various MW power levels. Values
with asterisk (*) indicates ramping times. Crystallinity index (CrI) was estimated using Segal’s method. (Reaction conditions: 0.1 g MCC, 0.1 g GO, 10ml
distilled water, zero holding time).
initially at MCC surface has been removed and consequently
Table 1 Crystallite sizes (nm) as a function of microwave power levels for
both uncatalyzed (MW) and catalyzed (MW+GO) reactions
dissolved. This was expected because MW energy (1 J/mol) is
much lower than the energy required to break typical chemical
bonds such as C-O (361 kJ/mol) and hydrogen bonds (4-42
kJ/mol)51. Therefore, MW alone is not capable of inducing
bond breakages within the bulk MCC structure. On the
contrary, when MW and GO were combined, glucose was
generated in the system with about 2.5 to 4.5 wt% yield at 200
to 800 W (Fig. 3b). When compared with GO-hydrothermal
treatment, the yield was also about 4.5 wt% in 12 min of
heating up time, signifying that MW+GO can significantly
accelerate the reaction process up to 25 times faster. Cello-
oligomers could be trapped in the hole defect sites and the –
COOH in GO can activate the β-1,4 glycosidic bonds towards
hydrolysis52,53. Then, MW could have induced the necessary
conformational changes to eventually cleave adjoined glucose
monomers41,52. Hence, even at a lower heating rate, i.e. 1 K/s
(200 W) with MW+GO resulted to glucose generation
compared to 1.7 K/s (400 W) in the absence of GO. Glucose
was also not detected in the hydrolysates using Gr and CNT,
indicating the importance of the oxygen functionalities in the
depolymerization process. The high glucose yield observed at
800 W could be likely due to the sudden uneven energy
distribution and high temperature gradient known as
“hotspots” or thermal runaway54. Thus, low MW power was
utilized in the succeeding experiments to minimize this
occurrence.
Microwave power
level
MW
MW+GO
MCC*
200 W
5.3 ± 0.3
-
4.5 ± 0.6
4.7 ± 0.3
5.0 ± 0.3
4.7 ± 0.2
400 W
600 W
800 W
5.4± 0.3
5.5± 0.4
5.4± 0.3
*Unreacted microcrystalline cellulose
O-sites on the MCC reducing ends, such as proposed by
Suganuma et al.48.
This has potentially weakened its connection from the bulk
crystalline structure, resulting to amorphous domains. Thus,
the combined MW and GO can effect in situ decrystallization
on the crystalline domain nearest the surface. The gradual
reduction in crystallite size (Table 1), i.e. from 5.3 to 4.5 nm at
200 W, can be associated with the disruption of crystalline
regions via surface attrition49. That is, MCC degrades from its
surface to the innermost part repeated layer-by-layer, until the
bulk crystalline structure is totally disrupted50.
Next, we determined the extent of depolymerization
reaction via glucose and TOC analysis. Glucose was not
detected in the hydrolysate of uncatalyzed reactions indicating
an incomplete depolymerization process. Thus, TOC analysis
was employed to determine the fate of the amorphous
cellulose removed from the MCC surface. TOC results revealed
the presence of small amounts of water-soluble organics
generally known as cello-oligomers (Fig. 3a). The relatively
small TOC has confirmed that only the amorphous region
Henceforth, it was established that the synergy of MW and
GO can affect depolymerization of MCC comprising of in situ
decrystallization, dissolution and hydrolysis steps while
heating up was in progress, and which considerably accelerate
glucose generation.
This journal is © The Royal Society of Chemistry 20xx
J. Name., 2017, 00, 1-13 | 5
Please do not adjust margins

Please do not adjust margins
Green Chemistry
Page 6 of 13
ARTICLE
Journal Name
View Article Online
DOI: 10.103[9B/C]7GC01691C
[A]
800
800
600
400
200
600
400
200
0
20
Total organic carbon (mg/L)
[A] Total organic carbon (TOC) results for depolymerization reactions using microwave (MW) alone; [B] Glucose yield for depolymerization
40
60
80
0
2
4
6
8
10
Glucose yield (%)
Fig. 3
reactions with the synergy of MW and graphene oxide (GO) during ramping from room temperature to 473 K at various MW power levels. (Reaction
conditions: 0.1 g MCC, 0.1 g GO, 10 ml distilled water, zero holding time. Error bars indicate ± standard deviation, n=2)
60
50
40
30
20
10
0
Table 2
Properties and catalytic activities of selected solid
acid catalysts. Heating rates obtained using 0.1 g MCC, 0.1 g
catalyst, 10 ml distilled water and 200 W while ramping from
room temperature to 473 K (zero holding time)
Entry
Catalyst
Functional
group
Heating
rate
(K/s)
Acid
density
(mmol/g)
BET
surface
area
(m2/g)
35
1
2
3
4
5
6
GO
A15
OH, COOH
SO3H
3.7
0.9
1.1
2.3
1.5
0.8
3-5a
4.756
1.3657
0.4758
1-3 a
35
44
2-
SZr
SO4
0
0
0
H3PW12O40
H2CO3-GO OH, COOH
NH2-GO NH2
--
8.1
31
(nd)
11
a) Boehme titration; (Ref. 56-58) NH3-TPD; (nd) Data currently not available
3.3 Performance evaluation of selected solid acids
The exclusivity of the synergism between MW and GO was
evaluated by comparing with Gr and CNT and commercially
available solid catalysts such as Amberlyst 15 (A15), sulphated
zirconia (SZr) and phosphotungstic acid (H3PW12O40), together
with carbonic acid functionalized (H2CO3-GO) and amine-
functionalized (NH2-GO) graphene oxide. Control experiments
Fig. 4 Comparison of catalytic performance of GO with control
experiments, graphite, carbon nanotube (CNT) and selected solid acids in
synergy with microwave irradiation in terms of glucose yield. (Reaction
conditions: 0.1 g MCC, 0.1 g catalyst, 10ml water, 473 K and 200 W, and
60 min holding time, for Gr and CNT: 453 K); * - GO hydrothermal,
procedure **See ref. 37
without MW and GO respectively, were also shown for associated with high acidity and high surface area55, it was
comparison. Fig. 4 shows the glucose yields while Table 2 expected that the ion exchange resin A15, the metal oxide SZr
summarizes the properties of the catalysts tested. TRS analysis or the superacid H3PW12O40 would outperform GO which bears
was used to validate the glucose yield obtained by HPLC. Yields weakly acidic functionalities (pKa >3). On the contrary, GO
obtained in both TRS and HPLC analyses are highly correlated generated the highest glucose yield under the stipulated
(Fig. S3).
experimental conditions. The other catalysts performed poorly
MW+GO was found to have significantly better presumably because of their insufficient interaction with MW
performance than the control conditions, thus further supported by the heating rates (Table 2, 4th column). This
substantiating the synergism between the two. Both Gr and observation points that MW interaction is crucial for the
CNT did not produced glucose due to the absence of process since the trend in glucose yield highly correlates with
functionalities. As the catalytic effectivity of solid acids is often the heating rates rather than surface properties of the
catalysts. Also, SZr, despite having a higher surface area than
6 | J. Name., 2017, 00, 1-13
This journal is © The Royal Society of Chemistry 20xx
Please do not adjust margins

Page 7 of 13
Green Chemistry
Please do not adjust margins
Journal Name
ARTICLE
View Article Online
DOI: 10.1039/C7GC01691C
GO, generated more oligosaccharides instead. Both A15 and
Therefore, on the basis of glucose yield, the catalyst
SZr completely decomposed under the conditions investigated, performance for the conditions stipulated was ranked as
signifying their thermal instabilities which could have limited follows: GO> H2CO3-GO > A15 > HPA > SZr. At this point, it was
their application at lower temperature range. It can be shown that there exists an exclusive synergy between MW and
inferred that strong acidity alone is inadequate to achieve GO which cannot be demonstrated with other solid catalysts.
glucose formation and that MW interaction is a necessary This behavior arises from the best combination properties of
parameter. H3PW12O40 being an electrolyte59, also acted as a GO, such as structure, MW interaction, variety in
MW sensitizer. However, its acid density is smaller compared functionalities, acid density and surface area.
with the other solid catalysts because of its different structure
resulting to a very low glucose yield. Increasing its weight in 3.4 Process parameters affecting glucose yield
accordance to the acid density of GO would lead to a very high
Effect of reaction time and temperature. The progression of
catalyst amount which is undesirable for the process.
MCC depolymerization was evaluated at various reaction
Moreover, it readily dispersed in water and formed colloidal
times. Once the desired reaction temperature was reached,
solutions that was difficult to recover. Meanwhile,
MW power has no significant influence on the process as it
functionalizing GO with carbonic acid resulted to a changed
becomes temperature-driven. Reactions were carried out from
acid density. While it has retained the properties of the GO
393 to 473 K and 0 to 60 min holding time. As shown in Fig. 5,
carbon backbone, its MW interaction slightly diminished, likely
glucose yield at a particular temperature increased as reaction
attributed to the fewer variety of functionalities on its surface.
time was increased. In a similar way, at a certain time, glucose
Substitution of the acidic functionalities with basic ones such
yield increased as temperature was increased. Consistent with
as the amine group curtailed MW interaction and eliminated
the other findings, the temperature rise was compensated by
any sign of catalytic activity. This implies that the density and
shorter reaction time61,62
.
The elevated temperature
type of functionalities also partake in the MW absorbing
capacity of GO. The functionalities could have acted as
“molecular radiators” that facilitated transfer of MW energy to
their immediate environment53. More importantly, the
presence of phenolic –OH groups in GO acted as a binding site
to draw MCC closer to the catalyst surface48. Without them,
catalytic performance is low, as shown by the data. Other
researchers have also reported that GO functional groups have
contributed to the weakening of the hydrogen bonds and
enhanced binding between MCC and GO63 which allowed the
weakly acidic functionalities of GO (pKa >3) to activate
hydrolysis. Furthermore, the elevated temperature favored
autohydrolysis of water increasing the overall acidity in the
system13. Also, the formation of organic acid intermediates
may have contributed to further accelerate the whole
depolymerization process37.
The rate of additional glucose generation (in mg ml-1 min-1)
was found to gradually decline as the reaction progressed, i.e.
at 473 K, from 5.5 between 0 → 1 min down to 0.48 between
30 →60 min. This slow release of glucose over time has been
linked to surface attrition, a step-by-step and relatively slow
concerted roles that leads to effective catalytic activity41,52
.
The trace amounts of sulfur may have had potential
contributions to the high yield observed, albeit at a much
lower scale. The GO utilized in his study contains only about
0.6% per gram, much lower than reported in the literatures60
to have significant impact on product yields.
50
473 K.TXT
453 K.TXT
433 K.TXT
40
30
20
10
0
0
10
20
30
40
50
60
Reaction time, min
Fig. 5 Temperature and holding time dependency of cellulose
depolymerization reactions under the synergy of microwave and
graphene oxide from 393 – 473 K and 0 to 60 min holding time. There was
no glucose yield at 393 K while 0.4% glucose yield was has been detected
at 413K only after 60min. (Reaction conditions: 0.1 g MCC, 0.1 g GO, 10 ml
distilled water and 200W).
Fig. 6
Thermograms
of
cellulose
residues
obtained
after
depolymerization reactions at 473 K. (Reaction conditions: 0.1 g MCC, 0.1 g
GO, 10 ml distilled water and 200 W).
This journal is © The Royal Society of Chemistry 20xx
J. Name., 2017, 00, 1-13 | 7
Please do not adjust margins

Green Chemistry
Please do not adjust margins
Page 8 of 13
Journal Name
ARTICLE
View Article Online
DOI: 10.1039/C7GC01691C
Table 3
A series of field emission scanning electron microscope (FESEM) images of cellulose residues showing the progression of surface
attrition. Crystallite sizes were estimated from Scherrer equation through X-ray diffraction (XRD) patterns. (Reaction conditions: 0.1 g MCC, 0.1 g
GO, 10 ml distilled water at 473 K, 200 W and 0 to 60 min)
@ 100x
@ 1000x
Crystallite size, nm
5.3 ± 0.3
4.3 ± 0.4
3.8 ± 0.9
3.5 ± 0.8
2.6±0.9
surface peeling process that is progressive over a longer period fresh MCC has rounded edges and a crystallite size of 5.3 nm.
of time. To further probe this observation, FESEM images of As the depolymerization process progressed, the MCC particles
residual glucose at 473 K and various reaction times were become rough and abraded and the crystallite sizes gradually
obtained. Their corresponding crystallite sizes were calculated reduced. After 60 min, the crystallite size measures only 2.3
and shown together on Table 3. Initially, the morphology of nm, or about half of the original. The ~50% crystalline loss was
This journal is © The Royal Society of Chemistry 20xx
J. Name., 2013, 00, 1-3 | 8
Please do not adjust margins

Page 9 of 13
Green Chemistry
Please do not adjust margins
Journal Name
ARTICLE
View Article Online
DOI: 10.1039/C7GC01691C
in agreement with the ~46% glucose yield, thus confirming the process variables that would help elevate product yield were
occurrence of surface attrition. also assessed. We investigated the influence of increasing the
The color of MCC residues ranged from light gray to amount of GO. As the formation of degradation products were
blackish gray as reaction time was prolonged. This physical anticipated at longer reaction times and higher catalyst
observation and the lower reactivity of MCC residues than loading, this set of experiments were conducted at 30 min.
fresh ones may indicate carbonization. The particles may have When the GO dosage was increased 1.5 times, only 2% more
experienced other reaction (i.e. pyrolysis) due to hotspots glucose was obtained (Fig. 7A). Increasing the dosage twice
formation resulting to altered cellulose structure64. In order to increased glucose yield from 32 to 41%, but with minimal
confirm this, we have subjected the residues into TGA analysis. degradation products. Further elevation of the catalyst dosage
Fig. 6 shows the TG curves of MCC before and after to 2.5 times the original yielded only 32% with more
depolymerization reactions. The TG curve of fresh MCC obeys degradation products. The small increase in yield was due to
three inherent thermal stages, in agreement with previous the increase in available active sites for the hydrolysis process.
reports63. The first one below 393 K was attributed to the However, the yield was much lower than expected and may
moisture removal, followed by the second one at ca. 493 K suggest that the initial active sites available has already
until around 753
K
corresponding to the thermal saturated the system, and the excess GO simply accelerated
decomposition of the organic carbon. The third stage at above formation of degradation products. Catalyst productivity also
773 K signifies combustion of the carbonaceous residue. As the decreased with the increase GO loading which indicates that
depolymerization process progressed, thermal stability of the the excess GO has very minimal participation, if any, in the
residues increased, indicating an alteration in their initial reaction. Hence the 1:1 substrate to catalyst loading was taken
decomposition behaviour.
as the best condition. This is much lower than the 400 wt%
utilized by Dreyer et al.26.
Interestingly, degradation products, such as formic acid,
levulinic acid and 5-HMF only started to appear in quantifiable
amounts at 473 K and 60 min (Fig. S4). This suggests that the
reaction duration affect the product yield distribution. Hence,
the termination of the process at the right condition is
necessary to boost glucose selectivity. The depolymerization
route proceeded as follows: MCC → cello-oligomers → glucose
→ degradation products18.
Effect of solvent loading. The effect of solvent loading on the
reaction was evaluated at two different perspectives. The first
one considered varying the actual solvent quantity in the
vessel, and the other, by fixing the solvent quantity similar to
previous experiments but varying the particles dosed into the
solution instead. Since hydrolysis reactions are typically carried
out with excess water, it was anticipated that higher yields will
be obtained at higher water loading. Moreover, the other
Effect of catalyst loading. Since the formation of degradation
products have confined the glucose yield at low range, other
Glucose
Formic acid
Catalyst productivity
5-HMF
Levulinic acid
Glucose
Formic acid
5-HMF
Levulinic acid
100
90
80
70
60
50
40
30
20
10
0
60
50
40
30
20
10
0
0.5
[D]
[B]
[C]
[A]
0.4
0.3
0.2
0.1
0
1:50 1:100 1:200
1:1
1:1.5
1:2
1:2.5
5
10
20
453 K 433 K
Substrate-to-catalyst ratio (g:g)
Reaction volume (ml)
Substrate-to-solvent
ratio (g:g)
Fig. 7
[A] Effect of cellulose (MCC)-to-graphene oxide (GO) ratio on glucose yield. Catalyst productivity is defined as gram glucose released per gram GO
(Reaction conditions: 0.1 g MCC, 0.1-0.25 g GO, 10ml distilled water, 30min); [B] Effect of reaction volume on glucose yield (Reaction conditions: 0.1 g MCC,
0.1 g GO, 5-20 ml distilled water, 60min); [C] Effect of MCC-to-water ratio on glucose yield (Reaction conditions: (1:50): 0.2 g MCC, 0.2 GO, 10ml distilled
water, (1:100): 0.1g MCC, 0.1 g GO, 10ml distilled water, (1:200): 0.05 g MCC, 0.05 g GO, 10ml distilled water, 60min); All experiments were carried out
under 200 W and 473 K [D] Glucose yields at reaction conditions: 0.2 g MCC, 0.2 GO, 10ml distilled water at 453 K and 433 K
This journal is © The Royal Society of Chemistry 20xx
J. Name., 2017, 00, 1-13 | 9
Please do not adjust margins

Green Chemistry
Please do not adjust margins
Page 10 of 13
Journal Name
ARTICLE
View Article Online
DOI: 10.1039/C7GC01691C
molecules can act as MW absorber to maintain the system in
the specified temperature. Contrary to these assumptions, as
shown in Fig. 7B, when the reaction volume was halved to 5
ml, glucose yield increased to 61 wt%. But when the reaction
volume was doubled to 20 ml, only 17 % glucose yield was
obtained. One of the reasons for such behaviour was the
theoretical MW power density (defined as MW energy per
volume of solution) absorbed by the system. When the
effective reaction volume was decreased, the theoretical MW
power density was twice the original reaction volume, thus
increasing glucose yield. The opposite took place when the
reaction volume was doubled. Another potential reason was
the restriction of molecular motion due to smaller space,
thereby allowing a closer interaction between MW and GO.
This increased their potential collision and aiding the
functional groups on the GO surface to connect with the
hydroxyl O sites in the MCC bulk. In contrast, an increased
reaction volume may have interfered with particle collisions.
In order to avoid compromise in comparing results due to a
different reaction volumes, the solvent volume was fixed at 10
ml while the amounts of MCC and GO were either doubled for
1: 50 ratio or halved for 1:200 ratio. Similarly, at 1:50, the
glucose yield sharply increased from 46 to 73% (Fig. 7C). This
suggests that there is such a minimum amount of water that is
consumed in the depolymerization process. Increasing the
water ratio has diluted the quantity of active sites available for
reaction. Furthermore, the excess water molecules compete
with the other particles for MW absorption without necessary
participation in the reaction. Reducing the excess water
molecules could have therefore channelled MW to the actual
reacting molecules instead. Along with the increased glucose
production, 5-HMF, formic acid and levulinic acid were also
obtained suggesting that the decrease in water quantity also
favoured occurrence of dehydration reaction leading to 5-HMF
formation. Thus, the substrate-to-solvent ratio of 1:50 was
taken as the best condition. Performing the experiments at
70
60
50
40
30
20
10
0
First cycle
Second cycle
Fig. 8 Reuse experiments (Conditions: 0.2 g crystalline (MCC) cellulose,
0.2 GO, 10ml distilled water, 453 K, and 60min holding time)
peaks at 1046 cm-1 and 3396 cm-1 assigned to epoxides and
hydroxyls, respectively, on the GO basal plane edge have
weakened.
The TGA plot (Fig. S6a) revealed that
higher
decomposition of spent GO takes place at
a
temperature than that of fresh GO. Altogether, these
characterizations indicate that GO underwent deoxygenation
or partial reduction, wherein some of its functionalities
detached from the surface, as reported elsewhere65,66. The
extent of reduction increases as temperature or time
increases.
Based from these characterizations, we expected that
spent GO will perform poorly than fresh GO. As shown on Fig.
8, glucose yield declined from about 61% to 34% for MCC
which can be attributed to the reduction of functionalities. The
removal of the hydroxyls may have moderated
decrystallization which is necessary to access the glycosidic
bond. Moreover, as GO sheds off hydroxyls, the surface of GO
becomes hydrophobic, as demonstrated by the immiscibility of
spent GO in water (Fig. S6b). This could have inhibited the
insertion of the water molecules. Thus, this has also
contributed to the slowdown of glucose production, as
illustrated and mentioned earlier in Section 3.4. However, the
removal of the oxygen functional groups has increased the
thermal stability of spent GO, suggestive of the trade-off
between stability and catalytic activity. The reduction of GO
and the observed properties of reduced GO (RGO) was also
lower temperature, i.e. 453 K (Fig. 7D), resulted to 61±4%
glucose yield with below detection of degradation products.
3.4 Spent GO characterization and reuse
So far, results indicated that GO exemplifies the characteristics
of a structurally ideal catalyst for cellulose depolymerization.
To address the concerns regarding the recyclability of the
catalyst, spent GO after each reaction was carefully collected
and characterized. Compared with the initial GO properties,
the XRD peaks of spent GO (Fig. S5a) reflects a new broader
peak proximate to 26O. This peak is characteristic of a reduced
GO40, 65. Moreover, the interlayer spacing reduced at 433 K,
453 K and 473 K into 4.4, 3.8 and 3.6 nm, respectively. The
FTIR spectra (Fig. S5b) also gradually changed with respect to
reaction temperature due to the decrease in functionalities
after each cycle. For example, O-H vibration at 1406 cm-1 and
at 1736 cm-1 corresponding to the phenolic hydroxyls (OH) and
carboxyls (-COOH), respectively, along the edge of GO
significantly decreased. Similarly, the
reported previously26,60,67
.
The role of the detached functionalities was further
clarified using CNT as control. CNT, having high MW
interaction as demonstrated in Sec. 3.1 is void of
functionalities and hence, do not lead to glucose generation
under stated conditions. Results of control experiments can be
found in Sec. S7 of the Supplementary Material. The addition
of dilute acids such as sulfuric and formic with CNT (Table S2,
Entries 3,4) resulted to glucose formation signifying the
contribution of dilute acids to the depolymerization process.
The contribution of the dilute acids, however, is quite low,
suggesting that the crystallinity of cellulose is difficult to break
when acidic functionalities are detached. The low contribution
This journal is © The Royal Society of Chemistry 20xx
J. Name., 2013, 00, 1-3 | 10
Please do not adjust margins

Page 11 of 13
Green Chemistry
Please do not adjust margins
Journal Name
ARTICLE
View Article Online
DOI: 10.1039/C7GC01691C
of dilute sulfuric acid indicates that bonded sulfonic groups acid sites. Alternatively, it could also be used as a
remaining on GO surface during preparation is not the major nanoscaffold/support for other particles and compounds such
contributor for the depolymerization process. Apparently, as the NH2-35 and H2CO3- functionalized GO developed in our
detaching the functionalities and making them more mobile in lab. As such, it would help minimize the high cost and
the solution does not significantly improve glucose generation. environmental impacts of utilizing rare precious metal based
This observation coincides with the slowdown in glucose catalysts68. Having graphene like properties, the RGO
production at longer reaction times. Despite the gradual generated has the potential applications for electrodes and
reduction of GO as the process progressed, glucose yield has composite materials69, although it needs further testing from
not considerably improved. When a pre-reduced RGO was this stage. Besides, the regeneration of GO has been recently
added with formic and sulfuric acid, the yield is higher than made possible using UV and ozone70 or conventional
that obtained with CNT under the same conditions (Table S2, method27. Nonetheless, the resulting RGO can be considered
Entry 7). The better performance of RGO than CNT indicates as a by-product of the reaction and can be utilized
that some functionalities still remain on its structure that aids for some other novel applications.
in the depolymerization process. It was also observed that the
The solid residues separated from the hydrolysate is a
selectivity of glucose production was diminished due to the mixture of partially reduced graphene oxide (RGO) and
formation of some other compounds in the solution. This carbonized cellulose. Due to the very small particles of residual
could be attributed to dilute acids that may have had initiated cellulose admixed with RGO, the complete separation of the
random scission of bonds, as opposed to the proposed surface two would be tedious and energy consuming. Because of this,
attrition mechanism. Therefore, GO with its functionalities we carried out further hydrolysis for this remaining fraction.
work synergistically to enable surface attrition that allows for The RGO-cellulose mixture was mixed with water and
high glucose selectivity. Overall, these results have confirmed subjected to MW conditions at elevated temperature without
the proposed mechanism and further affirmed the roleof GO the addition of any catalyst (Table S3). When the temperature
in the conversion of MCC to glucose. Despite being reduced as was elevated to 493 K, we were able to gain an additional 13%
a catalyst, RGO has found applications in catalyzing reactions of glucose, bringing the total yield to ~74%. Further elevation
requiring moderate acidity such as oxidation of benzene to of the temperature to 513 K yielded about 22% additional
phenol and reduction of nitrobenzene27 and oxidation of 5- glucose. The increase in temperature was compensated by
hydroxymethylfufual into 5-diformylfuran65. We hence tested decreasing the holding time to just 5 min. Thus, the 74% at 493
the spent GO from 473 K on ballmilled cellulose (Fig. S7) that K was obtained thru a two-stage process in 80 min combined
has lower crystallinity. The glucose yield obtained was about ramping and heating time. The process can be repeated until
42±4% (473 K and 60 min holding time). This shows that spent such time that only RGO remains with some carbon impurities
GO can still be used to catalyzed other reactions although its derived from cellulose.
catalytic activity is highly dependent upon the presence of the
Scheme 1
Proposed depolymerization pathway of crystalline cellulose by the synergism between microwave irradiation and graphene oxide
This journal is © The Royal Society of Chemistry 20xx
J. Name., 2017, 00, 1-13 | 11
Please do not adjust margins

Please do not adjust margins
Green Chemistry
Page 12 of 13
ARTICLE
Journal Name
View Article Online
DOI: 10.1039/C7GC01691C
3.5 Insights into the cellulose depolymerization mechanism
joint research program in the field of functional materials
(Grant Number: 14528300). We also would like to thank Prof.
Yoichi Shimizu and Prof. Satoko Takase of Kyushu Institute of
Technology for kindly providing the ballmilled cellulose utilized
in the study.
Taking into consideration the experimental and analytical
findings discussed in the previous sections, the MCC
depolymerization into glucose, depicted in Scheme 1, is
proposed to involve several steps occurring successively and
simultaneously under the synergy of MW and GO, as follows.
[1] As the system is irradiated continuously with MW, water
and GO absorbs MW, rapidly heating the bulk solution. As the
temperature increases, collision between GO and MCC
reducing ends take place, coinciding with the weakening of
inter-hydrogen bonds. Phenolic –OH on the GO surface would
form hydrogen bond with the neighboring hydroxyl O-sites,
thus converting crystalline to amorphous regions. The
amorphous domains, would peel off from the bulk MCC layer-
by-layer according to surface attrition mechanism [2]. The
cello-oligomers would get trapped into the GO hole defects
and the β-1,4 glycosidic bonds can be activated to hydrolysis
by -COOH group [3]. MW provides the necessary
conformational changes to eventually cleave the glycosidic
linkage. The resulting anhydroglucose reacts with water
molecules producing glucose monomers that would allow the
–COOH to regenerate afterwards. As the hydrolysis reaction
proceeds, at higher temperature and longer reaction time, side
reactions occur that hinder complete conversion of cellulose to
glucose, such as carbonization of cellulose, formation of
degradation products (levulinic acid, formic acid and 5-
hydroxymethylfurfural) and partial reduction of GO.
References
1
X. Zhao, X. Jinming, W. Aiqin and T. Zhang. Chinese Journal of
Catalysis, 2015, 36, , 1419-1427.
V. da Silva Lacerda, J. Sotelo, A. Guimarães, S. Navarro, M.
Bascones, L. Navas-Gracia, P. Ramos, E. Lebeña, and J. Gil.,
Bioresource Technology, 2015 , 180, 88-96.
9
2
3
4
5
6
7
8
9
J. Ni, N. Teng, H. Chen, J. Wang, J. Zhu, and H. Na.
Bioresource Technology, 2015, 191, 229-233.
M. Sasaki, T. Adschiri, and K. Arai. AIChE Journal, 2004, 50
,
192-202
F. Schüth, R. Rinaldi, N. Meine, M. Käldström, J. Hilgert and
M. Rechulski. Catalysis Today, 204, 234, 24-30.
D. Ciolacu, F. Ciolacu and V. Popa. Cellulose Chemistry and
Technology, 2010, 45, 13-21.
J. Ouyang, B. Liu, M. Zhang, Z. Zheng and H. Yu. Bioresource
Technology, 2013, 146, 288-293.
F.K.Tamaki, E.MAraujo,R.RozenbergandS.R.Marana. Biochem
istry and Biophysics Reports, 2016, 7, 52-55.
M.D. Meinita, B. Marhaeni, T. Winanto, D. Setyaningsih and
Y. Hong. Journal of Industrial and Engineering
Chemistry, 2015, 27, 108-114.
10 R. Chimentão, E. Lorente, F. Gispert-Guirado, F. Medina and
F. López. Carbohydrate Polymers, 2014, 111, 116-124.
11 L. Hu, L. Lin, Z. Wu, S. Zhou and S. Liu. Applied Catalysis B:
Environmental, 2015, 174-175, 225-243.
12 C.M. Martínez, D.A. Cantero, M.D. Bermejo and M.J. Cocero.
Cellulose, 2015, 22(4), 2231-2243.
13 Tolonen, L.K.; Juvonen, M.; Niemelä, K.; Mikkelson, A.;
Conclusions
We have systematically investigated the depolymerization of
microcrystalline cellulose under the synergy of microwave and
graphene oxide. This work has established that collectively,
MW and GO is an effective system for depolymerizing cellulose
into glucose at shorter time through one-pot decrystallization,
dissolution and hydrolysis stages. We were able to directly
Tenkanen, M.; Sixta, H. Carbohydrate Research, 2015, 401
16-23
,
14 Y. Yu and H. Wu. Industrial & Engineering Chemistry
Research, 2010, 49(8), 3919-3925.
observe the occurrence of surface attrition and developed a 15 G. Zhu, Q. Fan and X. Wan. Journal of Analytical and Applied
Pyrolysis, 2011, 90, 182-186.
proposed model that illustrates MCC hydrolysis. GO, when
compared with other solid catalysts such as Amberlyst 15,
sulfated zirconia, phosphotungstic acid and functionalized
graphene oxide, exhibited the best catalytic performance due
16 Onda, T. Och and K. Yanagisawa. Topics in Catalysis, 2009,
52, 801-807.
17 X. Li, Y. Jiang, L. Wang, L. Meng, W. Wang, and X. Mu. RSC
Advances, 2012, 2, 6921-6925.
to its MW interaction, type of functionalities, sufficient acid 18 Y. Huang, and Y. Fu. Green Chemistry, 2013, 15, 1095-1111.
19 T. Komanoya, H. Kobayashi, K. Hara, W. Chun and A.
Fukuoka. Applied Catalysis A: General, 2011, 40, 188-194.
20 T. Su, Z. Fang, F. Zhang, J. Luo and X. Li. Scientific Reports, 5:
17538, 1-14
density and adequate surface area. The results showed that
the combination of MW and GO can be a powerful and
promising technology for biomass conversion. Further study is
needed to increase the thermal stability of GO which can
potentially be achieved by functionalization. Also future
studies can be directed towards the application of reduced GO
to other reactions requiring mild acidity or electrochemical
applications.
21 C. Zhou, X. Xia, C. Lin, D. Tong and J. Beltramini. Chemical
Society Reviews, 2011, 40, 5588-5617.
22 L. Shuai and X. Pan. Energy & Environmental Science 2012, 5,
6889-6894.
23 Y. Wu, Z. Fu, D. Yin, Q. Xu, F. Liu, C. Lu and L. Mao. Green
Chemistry, 2010, 12(4), 696.
24 R. Rinaldi, R. Palkovits, and F. Schüth. Angewandte Chemie,
2008, 120, 8167-8170.
Acknowledgements
We gratefully acknowledge financial support for this research
from the Japan Science and Technology Agency via the e-ASIA
25 M. Yabushita, H. Kobayashi, K. Hara and A. Fukoka. Catal. Sci.
Technol., 2014, 4, 2312-2317
12 | J. Name., 2017, 00, 1-13
This journal is © The Royal Society of Chemistry 20xx
Please do not adjust margins

Page 13 of 13
Green Chemistry
Please do not adjust margins
Journal Name
ARTICLE
View Article Online
DOI: 10.1039/C7GC01691C
26 D.R. Dreyer, H. Jia and C.W. Bielawski. Angewandte Chemie 57 K. Saravanan, B. Tyagi and H. Bajaj. Applied Catalysis B:
International Edition, 2010. doi:10.1002/anie.201002160 Environmental, 2016, 192, 161-170
27 K.B. Dhopte, R.S. Zambare, A.V. Patwardhan and P.R. 58 F. Gomes, F. Mendes and M. Souza. Catalysis Today, 2017,
Nemade. RSC Adv, 2016, 6(10), 8164-8172.
279, 296-304.
28 D. Khalili. Tetrahedron Letters, 2016, 57, 1721-1723.
29 X. Zhao, J. Wang, C. Chen, Y. Huang, A. Wang and T. Zhang.
Chem. Commun., 2014, 50, 3439-3442
59 S. Tsubaki, K. Oono, T. Ueda, A. Onda, K. Yanagisawa, T.
Mitani and J. Azuma. Bioresource Technology, 2013, 144, 67-
73.
30 F. Wei, X. Wang, and X. Zhang. Separation and Purification 60 J. Cheng, Y. Qiu, R. Huang, W. Yang, J. Zhou and K.
Technology, 2014, 133, 421-428.
31 J. Chen, Y. Zhang, M. Zhang, B. Yao, Y. Li, L. Huang, L., … G. 61 C. Zhang, Z. Fu, Y. Liu, B. Dai, Y. Zou, X. Gong, … and D.
Shi. Chem. Sci, 2016, 7(3), 1874-1881. Yin. Green Chemistry, 2012, 14, 1928-1934.
32 Z. Yang, R. Huang, W. Qi, L. Tong, R. Su and Z. He. Chemical 62 N. Ramli and N. Amin. Fuel Processing Technology, 2014,
Engineering Journal, 2015, 280, 90-98. 128, 490-498.
Cen. Bioresource Technology, 2016, 221, 344-349.
33 V. Budarin, P. Shuttleworth, M. De bruyn, T. Farmer, M. 63 S. Radic, N. Geitner, R. Podila, A. Käkinen, P. Chen and F.
Gronnow, L. Pfaltzgraff and J. Clark. Catalysis Today, 2015,
239, 80-89.
Ding. Scientific Reports, 2013, 3:2273, 1-8.
64 G. Leal, L. Ramos, D. Barrett, A. Curvelo and C.
Rodella. Thermochimica Acta, 2015, 616, 9-13.
34 R. Y. Krishnan, M. Chandran, V. Vadivel and K. Rajan.
Separation and Purification Technology, 2016, 170, 224-233
35 F. Baldovino, A. Quitain, N. Dugos, S. Roces, M. Koinuma, M.
65 G. Lv, H. Wang, Y. Yang, T. Deng, C. Chen, Y. Zhu and X. Hou.
ACS Catal, 2015, 5, 5636-5646.
Yuasa and T. Kida. RSC Adv, 2017,
36 A. Allahbakhsh, F. Sharif, S. Mazinani, M. Kalaee, Int. J. Nano
Dimens. 2014, , 11-20
7
, 3852-3852
66 C. Xu, X. Shi, A. Ji, L. Shi, C. Zhou and Y. Cui. PLOS ONE, 2015,
10(12), e0144842.
5
67 C. Chen, X. Guo, G. Lu, C.M. Pedersen, Y. Qiao, X. Hou and Y.
Wang. New Carbon Materials, 2017. 32(2), 160-167.
68 J. Yu, Z. Liu, L. Zhai, T. Huang and J. Han. International Journal
of Hydrogen Energy, 2016, 41(5), 3436-3445.
37 A. Quitain, T. Kai, M. Sasaki and M. Goto. Industrial &
Engineering Chemistry Research, 2013, 52, 7940-7946.
38 G.Miller. Anal Chem,1959, 31, 426-428
39 L. Gurgel, K. Marabezia, L. Ramos and A. Curvelo. Industrial 69 T. Kuila, A.K. Mishra, P. Khanra, N.H. Kim and J.H. Lee.
Crops and Products, 2012, 36, 560-571. Nanoscale, 2013, 5(1), 52-71.
40 S.N. Alam, N. Sharma and L. Kumar.Graphene, 2017, 06(01), 70 F. Güneş, G.H. Han, H. Shin, S.Y. Lee, M. Jin, D.L. Duong …
1-18.
Y.H. Lee. Nano, 2011, 06(05), 409-418.
41 J. Bermudez, J. Angel Menendez, A. Arenillas, R. Martínez-
Palou, A.
Romero
and
R.
Luque.
Catalysis
Communications, 2015, 72, 133-137.
42 S. Tsubaki, M. Hiraoka, S. Hadano, K. Okamura, T. Ueda, H.
Nishimura, … T. Mitani. Carbohydrate Polymers, 2015, 115
,
78-87.
43 G. Khurana, N. Kumar, S. Kooriyattil and R. Katiyar. Journal of
Applied Physics, 2015, 117, 17E106.
44 J. Menéndez, A. Arenillas, B. Fidalgo, Y. Fernández, L.
Zubizarreta, E. Calvo, and J.M. Bermúdez. Fuel Processing
Technology , 2010, 91, 1-8.
45 M. Hotta, M. Hayashi, M.T. Lanagan, D.K. Agrawal, D. K. and
K. Nagata. ISIJ International, 2011, 51(11), 1766-1772.
46 H. Zhao, J. Kwak, Z. Zhang, H. Brown, B. Arey and J. Holladay.
Carbohydrate Polymers, 2007, 68, 235-241
47 Deguchi, S., Tsujii, K., & Horikoshi, K. Green Chem, 2008, 10
,
191-196.
48 S. Suganuma, K. Nakajima, M. Kitano, H. Kato, A. Tamura, H.
Kondo, … and M. Hara. Microporous and Mesoporous
Materials, 2011,143, 443-450.
49 E. Jasiukaityte-Grojzdek, M. Kunaver and I. Polijansek.
BioResources, 2012, 7, 3008-3027
50 B. Sun, L. Duan, G. Peng, X. Li and A. Xu. Bioresource
Technology, 2015, 192, 253-256.
51 M. Nüchter, B. Ondruschka, W. Bonrath, and A. Gum. Green
Chem, 2004, 6, 128-141.
52 C. Su., M. Acik, K. Takai, J. Lu, S. Hao, Y. Zheng, … and K. Ping
Loh. Nature Communications, 2012, 3:1298, 1-9.
53 J. Fan, M. De bruyn, V. Budarin, M. Gronnow, P. Shutleworth,
S. Breeden, D. Macquarrie and J. Clark. Journal of American
Chemical Society, 2013, 135, 11728 – 11731
54 A. Arshanitsa, Y. Akishin, E. Zile, T. Dizhbite, V. Solodovnik
and G. Telysheva. Renewable Energy, 2016, 91, 386-396.
55 Rinaldi and F. Schüth. Energy
Science, 2009, , 610-626.
&
Environmental
2
56 S. Lanke and B. Bhanage. Catalysis Communications, 2013,
41, 29-33.
This journal is © The Royal Society of Chemistry 20xx
J. Name., 2017, 00, 1-13 | 13
Please do not adjust margins