Sulfonated graphene oxide as effective catalyst for conversion of 5-(hydroxymethyl)-2-furfural into biofuels

Article abstract of DOI:10.1002/cssc.201301149

The acid-catalyzed reaction of 5-(hydroxymethyl)-2-furfural with ethanol is a promising route to produce biofuels or fuel additives within the carbohydrate platform; specifically, this reaction may give 5-ethoxymethylfurfural, 5-(ethoxymethyl)furfural diethylacetal, and/or ethyl levulinate (bioEs). It is shown that sulfonated, partially reduced graphene oxide (S-RGO) exhibits a more superior catalytic performance for the production of bioEs than several other acid catalysts, which include sulfonated carbons and the commercial acid resin Amberlyst-15, which has a much higher sulfonic acid content and stronger acidity. This was attributed to the cooperative effects of the sulfonic acid groups and other types of acid sites (e.g., carboxylic acids), and to the enhanced accessibility to the active sites as a result of the 2D structure. Moreover, the acidic functionalities bonded to the S-RGO surface were more stable under the catalytic reaction conditions than those of the other solids tested, which allowed its efficient reuse. Graphene on the scene: Sulfonated, partially reduced graphene oxide (S-RGO) exhibits a superior catalytic performance than other carbocatalysts and Amberlyst-15 in the acid-catalyzed conversion of 5-(hydroxymethyl)-2-furfural to products for biofuels. The beneficial effects are associated with the 2D structure of S-RGO and its acidic surface enriched with sulfur and oxygen functionalities.

Full text of DOI:10.1002/cssc.201301149

CHEMSUSCHEM
FULL PAPERS
DOI: 10.1002/cssc.201301149
Sulfonated Graphene Oxide as Effective Catalyst for
Conversion of 5-(Hydroxymethyl)-2-furfural into Biofuels
Margarida M. Antunes,[a] Patrꢀcia A. Russo,[a] Paul V. Wiper,[a] Jacinto M. Veiga,[a]
Martyn Pillinger,[a] Luꢀs Mafra,[a] Dmitry V. Evtuguin,[a] Nicola Pinna,[b] and
Anabela A. Valente*[a]
The acid-catalyzed reaction of 5-(hydroxymethyl)-2-furfural
with ethanol is a promising route to produce biofuels or fuel
additives within the carbohydrate platform; specifically, this re-
action may give 5-ethoxymethylfurfural, 5-(ethoxymethyl)furfu-
ral diethylacetal, and/or ethyl levulinate (bioEs). It is shown
that sulfonated, partially reduced graphene oxide (S-RGO) ex-
hibits a more superior catalytic performance for the production
of bioEs than several other acid catalysts, which include sulfo-
nated carbons and the commercial acid resin Amberlyst-15,
which has a much higher sulfonic acid content and stronger
acidity. This was attributed to the cooperative effects of the
sulfonic acid groups and other types of acid sites (e.g., carbox-
ylic acids), and to the enhanced accessibility to the active sites
as a result of the 2D structure. Moreover, the acidic functionali-
ties bonded to the S-RGO surface were more stable under the
catalytic reaction conditions than those of the other solids
tested, which allowed its efficient reuse.
Introduction
The portfolio of plant-biomass-derived useful chemicals is in-
creasing with the aim to reduce the global socioeconomic de-
pendence on fossil fuels, reduce carbon emissions and waste,
and enhance energy security.[1–5] In addition to bioethanol and
fatty acid alkyl esters, which are commercialized biofuel addi-
tives for gasoline and diesel, respectively, there is increasing in-
terest to produce biofuels or fuel additives with benefits in
terms of cost, type of feedstock used, and properties, for ex-
ample, energy density, stability, toxicity, and flow properties.
Promising routes involve carbohydrates as feedstocks (which
contemplate the most abundant biogenic polymers) with the
intermediate formation of furanic compounds. In particular,
carbohydrates can be converted by a series of acid-catalyzed
reactions to furfural and 5-(hydroxymethyl)-2-furfural (HMF),
which have applications in different sectors of the chemical in-
dustry.[6–11] The acid-catalyzed conversion of HMF gives 5-(al-
koxymethyl)furfural (AMF) and alkyl levulinate (AL)[11–23]
(Scheme 1). In particular, 5-ethoxymethylfurfural (EMF) is con-
sidered an attractive fuel additive[24] with an energy density of
Scheme 1. Conversion of HMF into products for biofuels applications.
(9.7 kWhLꢀ1).[11,25] EMF has a relatively low toxicity[14] and has
been tested in blends with diesel fuel.[26] Moreover, ethyl levuli-
nate (EL) may find use as a petroleum diesel fuel oxygenate
additive.[27–30] ALs may find additional applications in the flavor-
ing industry as solvents, fragrances, plasticizers, etc.[1,11,15,30–32]
As a result of the high value of the compounds mentioned
above, several catalysts have been investigated for the reac-
tions of hexose-based carbohydrates or HMF to AMF and/or
ALs. The homogeneous catalysts tested include strong mineral
acids,[12,14] inorganic salts,[33] ionic liquids,[16,20,34] heteropoly
acids,[12] and organic acids.[14,18,19] Insoluble solid acids are,
however, very attractive catalysts as they are environmentally
friendly. Therefore, various heterogeneous catalysts have been
investigated such as zeolites,[19,22,35] clays,[36] sulfated metal
8.7 kWhLꢀ1, which is greater than that of ethanol (6.1 kWhLꢀ1
)
and similar to that of gasoline (8.8 kWhLꢀ1) and diesel
[a] Dr. M. M. Antunes, Dr. P. A. Russo, Dr. P. V. Wiper, J. M. Veiga,
Dr. M. Pillinger, Dr. L. Mafra, Prof. D. V. Evtuguin, Dr. A. A. Valente
Department of Chemistry, CICECO
University of Aveiro
Campus de Santiago, 3810-193 Aveiro (Portugal)
E-mail: atav@ua.pt
[b] Prof. N. Pinna
Institut fꢀr Chemie
Humboldt Universitꢁt zu Berlin
Brook-Taylor-Straße 2, 12489 Berlin (Germany)
Supporting information for this article is available on the WWW under
http://dx.doi.org/10.1002/cssc.201301149.
ꢁ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemSusChem 2014, 7, 804 – 812 804

CHEMSUSCHEM
FULL PAPERS
www.chemsuschem.org
oxides,[22,35,37] modified mesoporous silicas,[22] and ion-exchange
resins, which include Amberlyst-15.[12,14,17,38] Although the use
of organic catalysts such as ion-exchange resins may avoid the
problematic metal leaching phenomena that can occur for
metal-containing catalysts, their low stability is a significant
drawback, which leads to restrictions to the operation temper-
ature.
of several micrometers in size is shown in Figure 1a. Energy
dispersive spectroscopy (EDS) confirmed the incorporation of
sulfur upon H2SO4 treatment (Figure 1c). Moreover, the carbon
and sulfur distribution maps (Figure 1b and c) show that the
S-RGO sheets were densely and homogeneously functionalized
with sulfur, which reflects the uniformity of the modification
treatment. Two other acid-treated carbon materials were also
investigated, namely, carbon black, which is made of particles
Carbon materials are attractive metal-free, stable, cheap, and
recyclable catalysts to achieve green and sustainable catalytic
transformations. Among carbons, graphene oxide (GO) and re-
lated materials are an emergent new type of promising carbo-
catalysts because of their unique properties, which include
their 2D structure, high stability, and high surface areas.[39]
Moreover, their catalytic performance can be engineered by
the attachment of large quantities of functionalities to both
sides of the carbon sheets to create readily accessible catalyti-
cally active sites. GO is produced by the chemical oxidation of
graphite. According to the most accepted structural models,
the basal planes of GO contain mostly epoxide and hydroxyl
groups, whereas carboxylic acids and small amounts of other
groups (ketones, etc) are present at the edges and hole de-
fects.[40] These oxygen-containing functional groups give rise to
the mild acidic and oxidative properties of GO, which were
first exploited by Bielawski and co-workers.[41] The authors
found that GO was able to catalyze the hydration of alkynes,
although high catalyst loadings (200 wt% relative to the sub-
strate) were required to achieve high conversions. Stronger
acid sites can be introduced on these carbons by further func-
tionalization. Very recently, GO and graphene with anchored
aryl sulfonic acid functionalities, prepared by the reaction of
the carbon with aryl diazonium salts of sulfanilic acid, were
successfully applied as catalysts for the hydrolysis of ethyl ace-
tate[42] and xylose dehydration.[43] Sulfated graphene prepared
by hydrothermal sulfation with fuming sulfuric acid at 1808C
was found to catalyze the esterification of acetic acid, the
Pechmann condensation, and the hydration of propylene
oxide.[44]
Figure 1. (a) SEM image of S-RGO; (b) carbon and (c) sulfur distribution maps
of the S-RGO region shown in image (a); SEM images of (d) S-CB and (e) S-
CNTs.
with an approximately spherical shape and 30 nm in size (Fig-
ure 1d), and stacked-cup CNTs with an external diameter of ap-
proximately 100 nm (Figure 1e).
The carbons have distinct surface textural properties
(Table 1, Figure S1). S-CB is microporous and has a high appar-
ent BET surface area (SBET), whereas S-CNTs and S-RGO exhibit
N2 adsorption isotherms typical of nonporous solids and have
low surface areas. The surface area of S-RGO is much lower
than the theoretical value for completely exfoliated and isolat-
ed graphene sheets (~2600–2700 m2 gꢀ1).[45,46] However, the
textural properties of GO in the wet/dispersed state differ sig-
nificantly from those in the dried state, and the restacking of
Herein, we show that partially reduced graphene oxide
(RGO) modified by treatment with sulfuric acid (S-RGO) is
a highly active, selective, and stable catalyst for the production
of EMF and EL from HMF. The catalytic performance of S-RGO
was found to be superior to that of other carbon and organic
catalysts, namely, amorphous carbon black (S-CB) and carbon
nanotubes (S-CNTs) modified by the same procedure, organic
lignosulfonate (LSF) derived from the spent liquors of acidic
Mg-based sulfite pulping of wood, and the well-known com-
mercial sulfonic acid resin Amberlyst-15, which has previously
shown exceptional activity and selectivity for the etherification
of HMF.[14]
Table 1. Chemical and textural characteristics of the carbon catalysts.
Sample
C
S
Acid sites
[mmolgꢀ1
SBET
[m2 gꢀ1
Vmp
[cm3 gꢀ1
]
[b]
[c]
[f]
[wt%][a] [mmolgꢀ1
]
]
]
S-RGO
S-CNTs
S-CB
64.79
89.70
96.31
1.08
0.14
0.25
2.2 (0.3[d]
0.7 (0.6[d]
0.6 (0.2[d]
)
)
)
12
44
–
–
0.04
228 (129[e]
)
Results and Discussion
[a] Carbon and [b] sulfur content obtained by elemental analysis. [c] Acid
groups determined by acid–base titration. [d] Acid sites of the respective
parental carbons. [e] Area of the surface external to the micropores calcu-
lated by the as method. [f] Micropore volume calculated by the as
method.
Characterization of the carbon materials
Partially reduced GO was modified by treatment with sulfuric
acid to introduce sulfur-containing acid groups onto the
carbon surface. An SEM image of a few stacked S-RGO sheets
ꢁ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemSusChem 2014, 7, 804 – 812 805

CHEMSUSCHEM
FULL PAPERS
www.chemsuschem.org
the sheets upon drying leads to a strong decrease in the spe-
cific surface area.[45]
spectroscopy (Figure 2b). All of the spectra show a band at
around 1735 cmꢀ1 assigned to the C=O stretch in carboxylic
acids. A band at approximately 1630 cmꢀ1, attributed to the
C=O stretch in quinones, is found in the spectra of CB, S-CB,
and CNTs, but not in that of S-CNTs,[48,50,51] which suggests that
the sulfuric acid treatment modified these groups on the CNTs
surface. The band at approximately 1380 cmꢀ1 corresponds to
the stretching vibration of OꢀH in alcohols.[50] Specific bands
of RGO and CNTs that correspond to the vibration of the C=C
bonds of the graphene sheets and CNTs backbone appear at
1570 and 1547 cmꢀ1, respectively.[52] The region of the spectra
between 1300 and 1000 cmꢀ1 contains several bands that are
associated with the vibration of CꢀO bonds that belong to the
functional groups, such as carboxylic acids, alcohols, ketones,
and ethers (RGO).[48,50] Additionally, the spectra of the acid-
treated materials have bands at approximately 1170, 1112, and
1030 cmꢀ1 attributed to the vibration of SꢀO and SꢀC bonds
in sulfonic acid groups,[50,51,53] which are particularly evident in
the spectrum of S-CB because of its lower oxygen content.
The total acid sites content was measured by acid–base ti-
tration and was found to exceed the sulfur content for all
H2SO4-treated carbons (Table 1). These results are consistent
with the presence of acid sites other than sulfonic acids on the
surface of the materials, specifically carboxylic acids and hy-
droxyl groups (denoted generically as O-acid sites, i.e., those
that are not sulfonic groups), as determined by FTIR spectros-
copy. The amount of O-acid sites on S-RGO and S-CB are
higher than on the respective parent carbons, which indicates
that the H2SO4 treatment introduced O-acid sites in addition to
sulfonic acids. For the CNTs, however, the content of these
sites did not change significantly upon H2SO4 treatment.
The acid treatment introduced a much higher amount of
sulfur on RGO than CB and CNTs (Table 1), which suggests that
the surface of RGO is more readily modified. Notably, sulfur
was not detected on any of the carbon materials prior to
H2SO4 treatment. The higher degree of surface functionaliza-
tion achieved for S-RGO may be associated with the higher re-
activity of its surface towards the functionalizing agent and/or
the surface area available for modification (upon dispersion, as
discussed above).[45] The fact that RGO was only partially re-
duced may facilitate the subsequent reaction with sulfuric
acid. Indeed, the carbon content of RGO is lower than that of
S-CB and S-CNTs obtained after sulfuric acid treatment
(Table 1).
The Raman spectra of the H2SO4-treated and untreated
carbon materials are shown in Figure 2a. The spectra show the
G band at approximately 1580–1590 cmꢀ1, which originates
from the graphene or graphite-like regions of the carbon, and
the D band at around 1340–1350 cmꢀ1, which arises from vi-
bration modes induced by the presence of defects such as
amorphous carbon (sp3-hybridized carbon), vacancies, or addi-
tional atoms.[47–49] The ratio between the area of the D and G
bands (ID/IG) increases from 1.3, 1.1, and 1.4 for the RGO, CNTs,
and CB to 1.6, 1.3, and 1.8 for S-RGO, S-CNTs, and S-CB, respec-
tively, which indicates an increase of the structural disorder
consistent with the modification of the carbon surface by the
introduction of sulfur groups.
The type of functional groups present on the carbons before
and after modification by sulfuric acid was evaluated by FTIR
The acid strength of S-RGO was investigated by 31P magic-
angle spinning (MAS) NMR spectroscopy with adsorbed trie-
thylphospine oxide (TEPO) as a base probe molecule; for com-
parison, spectra were also recorded for Amberlyst-15 and S-
CNT (Figure 3). This method has been widely used to deter-
mine the acid strength of mesoporous, microporous (zeolites),
and amorphous supports. The interaction of TEPO with acid
sites of increasing strength causes the 31P NMR resonance to
shift downfield, that is, stronger acid sites lead to higher chem-
ical shifts.[54] The 31P MAS NMR spectrum of TEPO adsorbed on
S-RGO reflects the existence of acid sites with a wide range of
strengths on the surface of this solid, especially those incorpo-
rated upon H2SO4 treatment. Moreover, it indicates the pres-
ence of acid sites in a variety of different chemical environ-
ments (specifically, different amounts and types of surface
functionalities). If we compare the 31P NMR spectra of RGO and
S-RGO, we observe that both materials show a resonance at
the same chemical shift (~55 ppm). The two materials contain
non-sulfur acid groups (as discussed above), which interact
with TEPO to produce the resonance at ~55 ppm. For S-RGO,
additional 31P NMR resonances arise from S-containing acid
sites, which yield a 31P NMR resonance at higher chemical
shifts (stronger acidity) from 71–80 ppm. The presence of more
acidic chemical functionalities corroborates the relatively high
sulfur content in S-RGO (Table 1). The 31P NMR resonance asso-
ciated with S-CNT shows a shoulder on the left attributed to
Figure 2. (a) Raman spectra of CB (1), S-CB (2), RGO (3), S-RGO (4), CNTs (5),
and S-CNTs (6); (b) FTIR spectra of CNTs (1), S-CNTs (2), CB (3), S-CB (4), RGO
(5), and S-RGO (6).
ꢁ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemSusChem 2014, 7, 804 – 812 806

CHEMSUSCHEM
FULL PAPERS
www.chemsuschem.org
Table 2. Reaction of HMF to BioEs in the presence of the carbon catalysts
at 1408C.[a]
Catalyst
Conversion
[%][b]
Product yield [%]
EL
EMF
EMFda
bioEs
none
CNTs
S-CNTs
S-CNTs-run2
S-CNTs-140ct
CB
22
49
99
65
71
32
99
30
42
51
100
99
99
100
0
3
14
3
2
3
20
2
1
0
36
52
39
44
23
55
15
10
33
42
73
71
58
0
0
20
5
0
39
86
47
49
26
85
18
12
41
94
93
96
100
3
<1
10
1
1
4
9
3
3
12
S-CB
S-CB-run2
S-CB-140ct
RGO
4
S-RGO
43
17
22
30
S-RGO-run2
S-RGO-run3
S-RGO-140ct
[a] Reaction conditions: 0.33m [HMF]0 in ethanol, catalyst loading
10 gcat. dmꢀ3, 24 h, 1408C. [b] HMF conversion.
Figure 3. 31P MAS NMR spectra of TEPO adsorbed onto carbocatalysts and
Amberlyst-15.
96% conversion reached after 30 min (Figure 4). These results
correlate with the amount of acid sites in the catalysts
(Table 1). Nevertheless, the superior performance of S-RGO was
also observed in comparison to S-CNTs and S-CB based on an
approximately equivalent amount of acid sites (Figures 4 and
5). Under these conditions, the initial reaction rate calculated
on the basis of the amount of acid sites [molas] was at least
three times higher for S-RGO (78 molmolasꢀ1 hꢀ1) than for the
other carbon catalysts (25 molmolasꢀ1 hꢀ1). If we used a lower
loading of S-RGO (3 gcat. dmꢀ3) and 2.5 times higher initial con-
centration of HMF (0.83m), we obtained a 49% BioEs yield at
67% conversion after 30 min with yields of 4, 41, and 4% EL,
EMF, and EMFda, respectively. Under these (more demanding)
reaction conditions for S-RGO, the catalytic results were still
better than those for S-CNTs and S-CB used with a higher mass
stronger acid sites that result from the H2SO4 treatment. The
reduced intensity of the shoulder in the spectrum of S-CNT is
in good agreement with the much lower concentration of
sulfur acid sites compared to that of S-RGO (Table 1). The spec-
trum of the benchmark acid catalyst Amberlyst-15 displays
a single resonance at 90.5 ppm, which indicates its stronger
acidity than the carbocatalysts.
The thermal stability of the carbon catalysts was investigat-
ed by thermogravimetric analysis (Figure S2). The weight loss
observed on all curves below approximately 1108C is from the
desorption of physisorbed water. For S-CB, a second weight
loss is observed between approximately 230–3208C, which
most likely corresponds to the decomposition of the sulfur-
and oxygen-containing groups. For the S-CNTs and S-RGO, this
second weight loss starts at approximately 2308C and gradual-
ly continues as the temperature increases; the weight of the
remaining solid at 6008C corresponds approximately to the
carbon content of the material. These results suggest that S-
CNTs and S-RGO have some surface groups that are more
stable those of S-CB. Furthermore, the surface groups of S-RGO
are more stable than those of GO, which completely decom-
pose at approximately 2008C.[55]
Catalytic performance of the carbon materials
The reaction of HMF with ethanol was investigated in the pres-
ence of the carbon catalysts at 1408C. The H2SO4-treated cata-
lysts led to the formation of EL, EMF, and EMFda (bioEs de-
notes EL, EMF, and EMFda). These products were not formed
without the addition of a catalyst, and the HMF conversion
was 22% after 24 h (Table 2). Of the tested carbon catalysts, S-
RGO exhibited superior performance, which led to the highest
bioEs yield (94%) at 100% conversion of HMF (Table 2, Fig-
ures 4 and 5). The reaction was much faster for S-RGO with
Figure 4. Dependence of HMF conversion on the reaction time in the pres-
~
&
*
ence of S-RGO ( ), S-CNTs ( ), S-CB ( ), and LSF (ꢂ) using a catalyst loading
of 10 gcat. dmꢀ3. S-RGO (3 gcat. dmꢀ3) used in approximately equivalent total
amount of acid sites as the other carbon catalysts (+). The inset shows the
results over 2 h of reaction. Reaction conditions: [HMF]0 =0.33m in ethanol,
1408C.
ꢁ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemSusChem 2014, 7, 804 – 812 807

CHEMSUSCHEM
FULL PAPERS
www.chemsuschem.org
yield reaches a maximum and
then decreases with a concomi-
tant increase in EL yield
(Figure 5).[12,14,21] It has been pro-
posed that the reaction mecha-
nism of the conversion of HMF
to EL involves the interconver-
sion of EMF and EMFda as inter-
mediates.[14,19,21,55] In general, the
bioEs total selectivity tended to
increase as the conversion ap-
proached 100%, which may be
partly because of the existence
of relatively slow intermediate
elementary steps. The distribu-
tion of bioEs products changed
with time, and the EL yield
tended to increase even after
100% HMF conversion was
reached; the reaction time can
be varied to obtain different
composition mixtures of bioEs
products. Small amounts of
a few byproducts were detected,
which included angelic lactone
and levulinic acid, which have
been identified previously in the
reaction of HMF with etha-
nol.[33,56] Ethyl formate was also
detected, which is a co-product
of the conversion of HMF to EL.
Diethyl ether (DEE) was formed
in small amounts (moles of DEE/
initial moles of HMF <0.5), pos-
sibly by the intermolecular dehy-
dration reaction of ethanol.[37]
~
&
*
Figure 5. Dependence of the yields of the BioEs products ( EL, EMF, EMFda) on the reaction time of HMF
with ethanol in the presence of (a) S-RGO, (c) S-CB, or (d) S-CNTs using a catalyst loading of 10 gcat. dmꢀ3. (b) S-
RGO (3 gcat. dmꢀ3) used in approximately equivalent total amount of acid sites as the other carbon catalysts. The
insets show the results over 2 h of reaction. Reaction conditions: [HMF]0 =0.33m in ethanol, 1408C.
of catalyst and lower initial amount HMF. These results may be
Catalytic stability of the carbon materials
partly because of the presence of relatively strong acid sites
on S-RGO (discussed above).
The catalyst stability was evaluated by performing consecutive
24 h batch runs. Although a considerable drop in catalytic ac-
tivity and bioEs yield was observed after two runs for S-CNTs
and S-CB, this did not occur for S-RGO (only a slight change in
the product distribution with an increased EMF yield; Table 2).
The leaching of sulfonic acid groups under the reaction con-
ditions is a frequent cause of deactivation of sulfonated car-
bons.[57–59] Additionally, it has recently been shown that the sul-
fonic and carboxylic acid groups of carbons prepared by hy-
drothermal carbonization of glucose may react with alcohols
to form sulfonic and carboxylic alkyl esters and ethers, which
leads to catalyst deactivation.[57] To understand the differences
in stability observed for our catalysts, the carbons were treated
with ethanol under the typical reaction conditions, in the ab-
sence of HMF. The resultant solids (S-CB-140ct, S-CNTs-140ct,
and S-RGO-140ct from S-CB, S-CNT, and S-RGO, respectively)
were subsequently tested in the HMF reaction (Table 2) and
characterized. The catalytic performances of S-CB-140ct and S-
To investigate the sole contribution of the O-acid sites to
the catalytic activity, the parent carbons RGO, CNT, and CB
(prior to H2SO4 treatment) were also tested for the reaction of
HMF. The reaction was much slower in the presence of these
catalysts, and low bioEs yields were reached (Table 2). Howev-
er, the results were still much better than those for the reac-
tion without a catalyst. Hence, the O-acid sites play a role in
the catalytic conversion of HMF to bioEs, but the sulfonic acid
groups are essential to achieve high conversions and bioEs
yields. The poorer catalytic performances of the parent carbons
in comparison to the respective H2SO4-treated materials may
be because of the smaller amount of acid sites (Table 1). How-
ever, the strength of the acid sites may influence the product
yields. S-RGO, which possesses stronger acid sites, led to
a higher bioEs yield than RGO (Table 2).
The kinetic profiles are consistent with the formation of EMF
as an intermediate in the conversion of HMF to EL; the EMF
ꢁ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemSusChem 2014, 7, 804 – 812 808

CHEMSUSCHEM
FULL PAPERS
www.chemsuschem.org
CNTs-140ct were poor and similar to those of the correspond-
ing solids used in a second batch run (Table 2). These results
can be attributed to the loss of sulfonic acid groups and de-
crease in the total amount of acid sites during the contact test
and catalytic reaction, as determined by elemental analysis and
acid–base titration (Table S1).
99% conversion, but this was only reached at 48 h reaction.
The catalytic reaction in the presence of LSF at 1108C was
slower than that at 1408C to give 33% bioEs yield at 45% con-
version (48 h reaction) with EMF as the main product (29%
yield).
In comparison to S-RGO, AmberlystTM-15 led to a lower initial
reaction rate at 1108C. After 30 min of reaction, the kinetic
curves tend to similar conversions for the two catalysts
(Figure 6). However, S-RGO led to higher BioEs yields than Am-
The performance of S-RGO-140ct, however, was identical to
that of the fresh catalyst, which is consistent with the charac-
terization data of this sample. Indeed, no significant loss of
sulfur or change in the total acid sites content was detected
for S-RGO-140ct (Table S1). Moreover, no changes in the FTIR
spectrum of S-RGO-140ct with respect to that of S-RGO were
observed (Figure S3), which reflects the absence of chemical
modifications of the sulfur and oxygen functionalities. The re-
sults indicate that the surface functional groups of S-RGO are
very stable under the catalytic reaction conditions, which sug-
gests that the bonds established between the surface groups
and the carbon framework of RGO are relatively strong. The
higher stability of the S-RGO surface groups is responsible for
the high conversions and bioEs yields observed after consecu-
tive batch runs (Table 2).
Comparison of the catalytic performance of S-RGO with or-
ganic catalysts
Figure 6. Dependence of HMF conversion on reaction time in the presence
~
of S-RGO ( ) and Amberlyst-15 (ꢀ) at 1108C. Reaction conditions:
[HMF]0 =0.33m in ethanol; catalyst loading of 10 gcat. dmꢀ3
.
For comparative purposes, the reaction of HMF was performed
in the presence of two organic catalysts: LSF and Amberlyst-
15.
LSF
is
a
lignosulfonate
with
approximately
ꢀ1 [60,61]
1.4 mmol
g
,
and, in contrast to the other solids, it was
berlyst-15, therefore, it is a more selective catalyst for bioEs
production than the acid resin (Figure 7). The regenerated Am-
berlyst-15 catalyst was used in a second 24 h batch run, which
led to similar conversions of HMF for the two runs, but with
a considerable decrease in BioEs yield from run 1 to run 2 (92
to 59%). Furthermore, in contrast to what happens with S-
RGO, the treatment of the acid resin with ethanol under the re-
action conditions caused a decrease of the acid sites content
(Table S1), which may be at least partly a result of leaching as
reported previously for this catalyst tested in the reaction of
furfuryl alcohol with ethanol under similar reaction condi-
tions.[62]
SO3H
completely soluble in the reaction medium, thus it acts as a ho-
mogeneous catalyst. Amberlyst-15 is a well-known commercial
polymer-based strong solid acid with a high content of sulfon-
ic acid groups (4.3 mmolgꢀ1), which has shown very high activ-
ity for a wide variety of acid-catalyzed reactions. It has also ex-
hibited an excellent performance in the conversion of HMF to
bioEs.[14] Comparisons with Amberlyst-15 were made by per-
forming the reactions at 1108C as its recommended maximum
operating temperature is 1208C because of its low stability.
The decrease of the reaction temperature from 140 to 1108C
for S-RGO led to a similar trend in the catalytic results in recy-
cling runs as that seen at 1408C
(Figure S4). The lower reaction
temperature promoted the for-
mation of EMF and EMFda over
EL (Figure S4). Similar behavior
has been reported for other cat-
alysts in the same reaction if the
temperature was reduced (e.g.,
H2SO4,[14] ion-exchange resins,[14]
NKG-9,[18] and p-toluenesulfonic
acid[12,18]).
LSF was less active than S-
RGO even if a lower amount of
S-RGO was used (with fewer
equivalents of acid sites) at
Figure 7. Dependence of yields of the BioEs products on the reaction time of HMF with ethanol in the presence
1408C (Figure 4). LSF led to
a high BioEs yield of 94% at catalyst loading of 10 gcat. dmꢀ3
~
&
*
of (a) S-RGO or (b) Amberlyst-15 at 1108C ( EL, EMF, EMFda). Reaction conditions: [HMF]0 =0.33m in ethanol;
.
ꢁ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemSusChem 2014, 7, 804 – 812 809

CHEMSUSCHEM
FULL PAPERS
www.chemsuschem.org
The superior catalytic performance of S-RGO for the conver-
sion of HMF to bioEs was also observed and it was compared
with Amberlyst-15 based on approximately equivalent total
amount of acid sites (5 and 10 gcat. dmꢀ3 S-RGO and Amberlyst-
15 loading, respectively). Although the acid resin catalyst led
to 80% bioEs yield at 100% conversion (Figure S5), S-RGO led
to 97% bioEs yield.
Experimental Section
Preparation of the catalysts
GO was prepared from graphite by
a modified Hummers
method[63] as described in a previous paper.[64] RGO was obtained
by the treatment of GO with benzyl alcohol under microwave irra-
diation at 1908C for 20 min. The solid was then collected by centri-
fugation, washed with ethanol, and dried at 658C. Sulfuric acid
treatment was performed as follows:[65] RGO (1.5 g) was heated in
H2SO4 (30 mL, 97 wt%) at 1608C for 5 h under N2. The suspension
was then centrifuged, and the solid was washed thoroughly with
hot distilled water (808C) and finally dried at 1008C. This process
was repeated twice more. Carbon black Vulcan XC72 (CB, Cabot)
and stacked-cups carbon nanotubes (CNTs, PR24-PS, Pyrograf Prod-
ucts) were also treated in the same way. Prior to the sulfuric acid
treatment, the CNTs were oxidized with concentrated nitric acid ac-
cording to the following procedure: CNTs (5 g) were added to
nitric acid (250 mL, 70%) and heated at 1008C for 10 h; the solid
was recovered by filtration, washed with distilled water until neu-
tral pH, and dried at 1008C.
The comparatively superior catalytic performance of S-RGO
cannot be explained by acid site content or strength. Amber-
lyst-15 has a much higher sulfonic acids content (4.3 mmolgꢀ1
)
than S-RGO. Moreover, the 31P MAS NMR spectrum confirmed
the presence of stronger acid sites in Amberlyst-15 than in S-
RGO (Figure 3). The 2D structure of S-RGO may contribute to
the higher activity as it results in readily accessible acid sites,
which avoids mass-transfer limitations as proposed by other
authors.[44] However, this factor does not fully explain the supe-
rior performance of S-RGO. The 31P MAS NMR spectrum of
TEPO adsorbed on S-RGO reflects the existence of acid sites
with a wide range of strengths on the surface of the solid, es-
pecially the acid sites incorporated upon sulfuric acid treat-
ment. Moreover, it indicates the presence of acid sites in a vari-
ety of different chemical environments (specifically, different
amounts and types of surface functionalities). This is a conse-
quence of the large amount of oxygen functional groups pres-
ent on RGO that was further increased by sulfuric acid treat-
ment, as indicated by the low carbon content of the S-RGO
material (64 wt%). Although the O-acid sites are relatively
weak and do not solely lead to high catalytic activities and se-
lectivities, an excellent catalytic performance is achieved if
they are associated with the presence of sulfonic acid groups.
Therefore, the superior performance of the S-RGO material in
the conversion of HMF to bioEs may be partly attributed to co-
operative effects of the sulfonic acid sites and O-acid sites in
their surroundings.
The lignosulfonate sample tested possesses the empirical formula
(per phenylpropane unit) C9H9.77O3.92S0.10(SO3H)0.36(OCH3)1.51 (denot-
ed LSF); the preparation and characterization of this sample is de-
scribed elsewhere.[60,61] For comparative purposes, catalytic tests
were performed by using the cation-exchange resin Amberlyst-15,
a macroreticular styrene-divinylbenzene copolymer that bears ben-
zenesulfonic acid groups. Prior to use, the commercial Amberlyst-
15 (FlukaChemika) was ground and sieved to give a powder with
particle sizes smaller than 106 mm.
Characterization of the carbon materials
The carbon and sulfur content of the samples was determined by
elemental analysis by using a TruSpec 630 elemental analyzer. SEM
images were recorded by using a Hitachi SU-70 HR microscope op-
erated at 10 or 15 kV coupled with a Bruker Quantax 400 EDS
system. BET specific surface areas were calculated from N2 adsorp-
tion isotherms measured at ꢀ1968C by using a Micromeritics
Gemini 2380 (prior to the measurements, the samples were de-
gassed at 1208C overnight). The external surface area and micro-
pore volumes were calculated by application of the as method by
using standard data for N2 adsorption on nonporous carbon. Ther-
mogravimetric analysis (TGA) was performed from RT to 6008C
with a heating rate of 58Cminꢀ1 under a N2 flow (20 mLminꢀ1) by
using a Shimadzu TGA-50 instrument. The total acid sites content
was determined by acid–base titration. In a typical experiment,
NaCl (20 mL, 0.1m) was added to the sample (0.1 g), the mixture
was stirred for 24 h at RT and then titrated with NaOH (0.01m).
The acid properties of the solids were evaluated by 31P MAS NMR
spectroscopy of chemically adsorbed TEPO. The adsorption of
TEPO was performed as follows: solid (0.1 g) was dehydrated at
110–1208C under vacuum. TEPO (0.015 g) dissolved in anhydrous
n-pentane (5 mL) was added to the solid, and the mixture was
stirred for 30 min under N2 and then dried at 508C under vacuum.
Solid-state NMR spectra were acquired by using a Bruker Avance III
400 spectrometer using a 4 mm double resonance probe operating
at a B0 field of 9.4 T (400 MHz) with a 31P Larmor frequency of
161.9 MHz. 31P {1H} MAS NMR spectra were recorded by using a ro-
tation speed of 12 kHz, a single excitation pulse width of 1.9 ms,
a radio-frequency field strength of 56 kHz (608 flip angle), and 15 s
recycle delay. A two-pulse phase modulation scheme (TPPM-15)
was used for 1H heteronuclear decoupling. FTIR spectra were re-
Conclusions
Partially reduced graphene oxide was modified by treatment
with sulfuric acid to introduce sulfonic acid surface groups.
The resultant material (S-RGO) contained a total amount of
acid sites of 2.2 mmolgꢀ1 (ꢀSO3H, ꢀCOOH, ꢀOH) and exhibited
high activity for the conversion of 5-(hydroxymethyl)-2-furfural
into 5-ethoxymethylfurfural, 5-(ethoxymethyl)furfural diethyla-
cetal, and ethyl levulinate (bioEs), and led to high bioEs yields
in the range 110–1408C. In particular, the catalytic performance
of S-RGO was superior to that of carbon black and carbon
nanotubes modified by the same procedure and of the stron-
ger solid acid Amberlyst-15 (4.3 mmol gꢀ1). This was ex-
SO3H
plained by cooperative effects of the different acid sites and
the 2D structure of the S-RGO material. The acidic surface func-
tionalities of S-RGO were stable, and, consequently, the catalyst
maintained its performance over consecutive catalytic runs, in
contrast to the other catalysts tested.
ꢁ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemSusChem 2014, 7, 804 – 812 810

CHEMSUSCHEM
FULL PAPERS
www.chemsuschem.org
corded by using a Bruker Tensor 27 spectrometer using pellets of
the sample mixed with KBr (400–4000 cmꢀ1, 256 scans, 4 cmꢀ1 res-
olution). Raman spectroscopy was performed by using a JobinYvon
T64000 spectrometer (laser l: 532 nm).
Acknowledgements
We are grateful to the Fundażo para a CiÞncia e a Tecnologia
(FCT) and Fundo Europeu de Desenvolvimento Regional (FEDER)
for general funding (projects POCTI/QUI/56112/2004; PTDC/CTM/
098361/2008), and to the Associate Laboratory CICECO (PEst-C/
CTM/LA0011/2013). The FCT and the European Union are ac-
knowledged for postdoctoral grants to M.M.A. (SFRH/BPD/89068/
2012) and P.A.R. (SFRH/BPD/79910/2011) cofunded by MCTES and
the European Social Fund through the program POPH of QREN.
Catalytic tests
Batch catalytic experiments were performed in tubular glass micro-
reactors with pear-shaped bottoms equipped with an appropriate
PTFE-coated magnetic stirring bar and a valve for gas purging. In
a typical procedure, HMF (0.33m; Aldrich, 99%), powdered catalyst
(loading of 10 gcat. dmꢀ3), and ethanol (Riedel-de Haꢃn, 99.8%,
1 mL) were added to the reactor. Individual experiments were per-
formed for a given reaction time, and the presented results are the
mean values of at least two replicates. The reaction mixtures were
heated with a thermostatically controlled oil bath under continu-
ous magnetic stirring at 1000 rpm. Zero time (when the reaction
began) was taken to be the instant the microreactor was immersed
in the oil bath. After a 24 h batch run, the catalysts were separated
by centrifugation, washed with ethanol and then water, and subse-
quently treated with aq H2SO4 (0.2m) for 4 h at 308C. The catalysts
were subsequently washed with water until the pH was neutral
and dried at 858C overnight. Contact tests (ct) were performed for
the catalysts that consisted of a pretreatment of the materials with
ethanol (without HMF) for 24 h at 110 or 1408C, washing with eth-
anol and drying at 708C. The resultant samples are denoted
(sample name)-Tct, in which T is the treatment temperature [8C].
Keywords: biomass · carbon · graphene · heterogeneous
catalysis · sulfur
[1] G. W. Huber, S. Iborra, A. Corma, Chem. Rev. 2006, 106, 4044–4098.
[2] Y. Romꢄn-Leshkov, C. J. Barrett, Z. Y. Liu, J. A. Dumesic, Nature 2007, 447,
982–985.
[3] S. N. Naik, V. V. Goud, P. K. Rout, A. K. Dalai, Renewable Sustainable
Energy Rev. 2010, 14, 578–597.
[4] J. J. Bozell, G. R. Petersen, Green Chem. 2010, 12, 539–554.
[5] A.-L. Marshall, P. J. Alaimo, Chem. Eur. J. 2010, 16, 4970–4980.
[6] K. J. Zeitsch, The Chemistry and Technology of Furfural and its Many By-
Products, Elsevier Science B. V., Amsterdam, 2000.
[7] J. N. Chheda, G. W. Huber, J. A. Dumesic, Angew. Chem. 2007, 119,
7298–7318; Angew. Chem. Int. Ed. 2007, 46, 7164–7183.
[8] D. M. Alonso, J. Q. Bond, J. A. Dumesic, Green Chem. 2010, 12, 1493–
1513.
[9] S. Dutta, RSC Adv. 2012, 2, 12575–12593.
[10] X. Tong, Y. Ma, Y. Li, Appl. Catal. A 2010, 385, 1–13.
[11] E. D. J. G.-J. Gruber, Biofuels Technol. 2009, 11–17.
[12] P. H. Che, F. Lu, J. J. Zhang, Y. Z. Huang, X. Nie, J. Gao, J. Xu, Bioresour.
Technol. 2012, 119, 433–436.
[13] M. Mascal, E. B. Nikitin, Green Chem. 2010, 12, 370–373.
[14] M. Balakrishnan, E. R. Sacia, A. T. Bell, Green Chem. 2012, 14, 1626–1634.
[15] J. Zhang, S. Wu, B. Li, H. Zhang, ChemCatChem 2012, 4, 1230–1237.
[16] M. Imteyaz Alam, S. De, S. Dutta, B. Saha, RSC Adv. 2012, 2, 6890–6896.
[17] P. Lanzafame, D. M. Temi, S. Perathoner, G. Centi, A. Macario, A. Aloise,
G. Giordano, Catal. Today 2011, 175, 435–441.
[18] L. Bing, Z. Zhang, K. Deng, Ind. Eng. Chem. Res. 2012, 51, 15331–15336.
[19] K. S. Arias, S. I. Al-Resayes, M. J. Climent, A. Corma, S. Iborra, ChemSu-
sChem 2013, 6, 123–131.
[20] S. De, S. Dutta, B. Saha, ChemSusChem 2012, 5, 1826–1833.
[21] X. Hu, R. Gunawan, D. Mourant, C. Lievens, X. Li, S. Zhang, W. Chaiwat,
C.-Z. Li, Fuel 2012, 97, 512–522.
[22] S. Saravanamurugan, A. Riisager, Catal. Commun. 2012, 17, 71–75.
[23] A. A. Rosatella, S. P. Simeonov, R. F. M. Frade, C. A. M. Afonso, Green
Chem. 2011, 13, 754–793.
[24] G. J. M. Gruter, (Furanix Technologies B. V.), 5-substituted 2-(alkoxyme-
thyl)furans, EP2487170A1, The Netherlands, 2012, p. 13.
[25] M. Mascal, E. B. Nikitin, ChemSusChem 2009, 2, 423–426.
[26] M. Mascal, E. B. Nikitin, Angew. Chem. 2008, 120, 8042–8044; Angew.
Chem. Int. Ed. 2008, 47, 7924–7926.
Quantification and identification of the reaction products
The evolution of the catalytic reactions was monitored by GC (for
reaction products) and HPLC (for the substrate). Prior to sampling,
the reactors were cooled to ambient temperature and well mixed
before opening and the work-up procedure, and no pressure-re-
lease was verified. The samples were analyzed as they were pre-
pared. The reagent HMF was quantified by using a Knauer K-1001
HPLC pump and a PL Hi-Plex H 300 mmꢂ7.7 mm (i.d.) ion-ex-
change column (Polymer Laboratories Ltd., UK) coupled to
a Knauer 2600 UV detector (280 nm). The mobile phase was
0.005m H2SO4. Analysis conditions: flow rate 0.6 mLminꢀ1, column
temperature 658C. The reaction products (e.g., EMF, EL) were quan-
tified by using a Varian 3800 GC equipped with a capillary column
(Chrompack, CP-SIL 5CB, 50 mꢂ0.32 mmꢂ0.5 mm) and a flame ion-
ization detector using H2 as the carrier gas. Authentic samples of
HMF and EL were used as standards, and the calibration curve was
measured for quantification. The response factors of the bioEs
products were considered similar to that of EL. The identification
of the reaction products was accomplished by GC–MS by using
a Trace GC 2000 Series (Thermo Quest CE Instruments-DSQ II,
Thermo Scientific), equipped with a capillary column (DB-5 MS,
30 mꢂ0.25 mmꢂ0.25 mm) with He as the carrier gas. The HMF
conversion [%] at reaction time t was calculated using Equation (1):
[27] D. J. Hayes, Catal. Today 2009, 145, 138–151.
[28] B. C. Windom, T. M. Lovestead, M. Mascal, E. B. Nikitin, T. J. Bruno, Energy
Fuels 2011, 25, 1878–1890.
[29] R. L. V. Mao, Q. Zhao, G. Dima, D. Petraccone, Catal. Lett. 2011, 141,
271–276.
[30] E. Christensen, A. Williams, S. Paul, S. Burton, R. L. McCormick, Energy
Fuels 2011, 25, 5422–5428.
HMF conversion ¼ fð½HMFꢁinitialꢀ½HMFꢁat time tÞ=½HMFꢁinitialg ꢂ 100
ð1Þ
[31] A. P. Groves, C. Morkey, J. Smith, A. Stevenson, (Shell Oil Company), Fuel
Compositions, US2005/0144835A1, USA, 2005, p. 8.
The yield of EL [%] at reaction time t was calculated using Equa-
[32] M. Chia, J. A. Dumesic, Chem. Commun. 2011, 47, 12233–12235.
[33] Y. Yang, C. Hu, M. M. Abu-Omar, Bioresour. Technol. 2012, 116, 190–194.
[34] S. Saravanamurugan, O. Nguyen Van Buu, A. Riisager, ChemSusChem
2011, 4, 723–726.
tion (2):
Yield of EL ¼ ½ðconcentration of EL at time tÞ=
ð2Þ
ðinitial concentration of HMFÞꢁ ꢂ 100
[35] L. Peng, L. Lin, H. Li, Q. Yang, Appl. Energy 2011, 88, 4590–4596.
ꢁ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemSusChem 2014, 7, 804 – 812 811

CHEMSUSCHEM
FULL PAPERS
www.chemsuschem.org
[36] G. J. M. Gruter, L. E. Manzer, (Furanix Technologies B. V.), Hydroxymethyl-
furfural ethers from sugars and higher alcohols, US2010/0218415, The
Netherlands, 2010, p. 7.
[52] W. Chen, L. Yan, P. R. Bangal, Carbon 2010, 48, 1146–1152.
[53] H. Yu, Y. Jin, Z. Li, F. Peng, H. Wang, J. Solid State Chem. 2008, 181, 432–
438.
[37] L. Peng, L. Lin, J. Zhang, J. Shi, S. Liu, Appl. Catal. A 2011, 397, 259–265.
[38] H. Zhu, Q. Cao, C. Li, X. Mu, Carbohydr. Res. 2011, 346, 2016–2018.
[39] C. Su, K. P. Loh, Acc. Chem. Res. 2012, 46, 2275–2285.
[40] D. R. Dreyer, S. Park, C. W. Bielawski, R. S. Ruoff, Chem. Soc. Rev. 2010,
39, 228–240.
[41] D. R. Dreyer, H.-P. Jia, C. W. Bielawski, Angew. Chem. 2010, 122, 6965–
6968; Angew. Chem. Int. Ed. 2010, 49, 6813–6816.
[42] J. Li, G. Zhang, H. Chen, S. Wang, G. Zhang, F. Zhang, X. Fan, Chem. Sci.
2011, 2, 484–487.
[43] E. Lam, J. H. Chong, E. Majid, Y. liu, S. Hrapovic, A. C. W. Leung, J. H. T.
Luong, Carbon 2012, 50, 1033–1043.
[44] F. Liu, J. Sun, L. Zhu, X. Meng, C. Qi, F.-S. Xiao, J. Mater. Chem. 2012, 22,
5495–5502.
[45] R. L. D. Whitby, V. M. Gun’ko, A. Korobeinyk, R. Busquets, A. B. Cundy, K.
Lꢄszlꢅ, J. Skubiszewska-Zie˛ba, R. Leboda, E. Tombꢄcz, I. Y. Toth, K.
Kovacs, S. V. Mikhalovsky, ACS Nano 2012, 6, 3967–3973.
[46] S. Stankovich, D. A. Dikin, G. H. B. Dommett, K. M. Kohlhaas, E. J.
Zimney, E. A. Stach, R. D. Piner, S. T. Nguyen, R. S. Ruoff, Nature 2006,
442, 282–286.
[47] S. Eigler, C. Dotzer, A. Hirsch, Carbon 2012, 50, 3666–3673.
[48] L. Stobinski, B. Lesiak, L. Kçvꢆr, J. Tꢅth, S. Biniak, G. Trykowski, J. Judek,
J. Alloys Compd. 2010, 501, 77–84.
[54] A. Zheng, S.-J. Huang, S.-B. Liu, F. Deng, Phys. Chem. Chem. Phys. 2011,
13, 14889–14901.
[55] S. Stankovich, D. A. Dikin, R. D. Piner, K. A. Kohlhaas, A. Kleinhammes, Y.
Jia, Y. Wu, S. T. Nguyen, R. S. Ruoff, Carbon 2007, 45, 1558–1565.
[56] X. Hu, C.-Z. Li, Green Chem. 2011, 13, 1676–1679.
[57] J. M. Fraile, E. Garcꢀa-Bordejꢆ, L. Roldꢄn, J. Catal. 2012, 289, 73–79.
[58] W. Zhang, H. Tao, B. Zhang, J. Ren, G. Lu, Y. Wang, Carbon 2011, 49,
1811–1820.
[59] X. Mo, D. E. Lꢅpez, K. Suwannakarn, Y. Liu, E. Lotero, J. G. Goodwin Jr.,
C. Lu, J. Catal. 2008, 254, 332–338.
[60] A. P. Marques, D. V. Evtuguin, S. Magina, F. M. L. Amado, A. Prates, J.
Wood Chem. Technol. 2009, 29, 322–336.
[61] A. P. Marques, D. V. Evtuguin, S. Magina, F. M. L. Amado, A. Prates, J.
Wood Chem. Technol. 2009, 29, 337–357.
[62] P. Neves, S. Lima, M. Pillinger, S. M. Rocha, J. Rocha, A. A. Valente, Catal.
Today 2013, 218, 76–84.
[63] W. S. Hummers, R. E. Offeman, J. Am. Chem. Soc. 1958, 80, 1339–1339.
[64] P. A. Russo, N. Donato, S. G. Leonardi, S. Baek, D. E. Conte, G. Neri, N.
Pinna, Angew. Chem. 2012, 124, 11215–11219; Angew. Chem. Int. Ed.
2012, 51, 11053–11057.
[65] K. Hou, A. Zhang, L. Gu, M. Liu, X. Guo, J. Colloid Interface Sci. 2012,
377, 18–26.
[49] K. N. Kudin, B. Ozbas, H. C. Schniepp, R. K. Prud’homme, I. A. Aksay, R.
Car, Nano Lett. 2008, 8, 36–41.
[50] N. Lakshmi, N. Rajalakshmi, K. S. Dhathathreyan, J. Phys. D 2006, 39,
2785–2790.
Received: October 25, 2013
[51] Y. Wang, Z. Iqbal, S. Mitra, J. Am. Chem. Soc. 2006, 128, 95–99.
Published online on February 3, 2014
ꢁ 2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemSusChem 2014, 7, 804 – 812 812