Article

Electrocatalytic Oxidation of Cellulose to Gluconate

on Carbon Aerogel Supported Gold Nanoparticles
Anode in Alkaline Medium
Hanshuang Xiao, Meifen Wu and Guohua Zhao *
Received: 25 November 2015; Accepted: 25 December 2015; Published: 30 December 2015
Academic Editor: Rafael Luque

Department of Chemistry, Tongji University, Shanghai 200092, China; xhs@tongji.edu.cn (H.X.);

wumf@tongji.edu.cn (M.W.)
* Correspondence: g.zhao@tongji.edu.cn; Tel.: +86-21-6598-1180; Fax: +86-21-6598-2287

Abstract: The development of high efficient and low energy consumption approaches for the
transformation of cellulose is of high significance for a sustainable production of high value-added
feedstocks. Herein, electrocatalytic oxidation technique was employed for the selective conversion of
cellulose to gluconate in alkaline medium by using concentrated HNO3 pretreated carbon aerogel
(CA) supported Au nanoparticles as anode. Results show that a high gluconate yield of 67.8% and
sum salts yield of 88.9% can be obtained after 18 h of electrolysis. The high conversion of cellulose and
high selectivity to gluconate could be attributed to the good dissolution of cellulose in NaOH solution
which promotes its hydrolysis, the surface oxidized CA support and Au nanoparticles catalyst which
possesses high amount of active sites. Moreover, the bubbled air also plays important role in the
enhancement of cellulose electrocatalytic conversion efficiency. Lastly, a probable mechanism for
electrocatalytic oxidation of cellulose to gluconate in alkaline medium was also proposed.

Keywords: cellulose; alkaline condition; electrocatalytic oxidation; carbon aerogel supported Au

nanoparticles; gluconate

1. Introduction
Cellulose, the most abundant source of biomass, is considered the most attractive alternative
carbon source for the sustainable production of functionalized feedstocks since it is cheap, inedible
and available on a large scale [1,2]. However, the transformation of cellulose into the specific chemicals
is difficult because it owns low reactivity due to its crystalline fiber structure of β-1-4-glycosidic
linkages in its long chains. Moreover, the network of intra- and intermolecular hydrogen bonds formed
between the hydroxyl groups in adjacent cellulose makes it insoluble in water and in most organic
solvents under mild conditions, which further aggravates the restriction for its utilization. In previous
studies, various approaches including mineral acids hydrolysis, hydrothermal oxidation, enzymes
hydrolysis, heterogeneous acids catalysis, metal and metal oxide catalysis, etc., were attempted to
transform cellulose [3–8]. High value-added chemicals and fuels, such as glucose, polyols and organic
acids, have been obtained. Among them, gluconic acid is an important industrial product, which
can be usually used as water-soluble cleaner or as the additive in food and beverages. Wang et al. [9]
performed the selective oxidation of cellobiose on Au nanoparticles supported on carbon nanotube
(CNT) at 145 ˝ C and 0.5 MPa O2 and obtained an 80% yield to gluconic acid. In their study, the surface
functional groups on CNT surfaces generated by pretreatment with HNO3 is considered a key factor
for the enhancement of the hydrolysis of cellobiose into glucose, while Au nanoparticles are attributed
mainly to the oxidation of glucose into gluconic acid. An et al. [4] detected the oxidation of cellobiose
to gluconic acid over polyoxometalate-supported Au nanoparticles. A 97% yield to gluconic acid was

Catalysts 2016, 6, 5; doi:10.3390/catal6010005 www.mdpi.com/journal/catalysts

Catalysts 2016, 6, 5 2 of 10

achieved over Au/CS1.7 H1.3 PW12 O40 at 145 ˝ C and 0.5 MPa O2 , which was attributed to the high
acidity of the catalyst. Moreover, the same catalyst afforded a 70% conversion of a ball-milled cellulose
by oxidation at 145 ˝ C under 1 MPa of O2 , resulting in a 60% yield to gluconic acid after 11 h. By using
Au/TiO2 as catalyst, Borgna et al. [7] achieved a 93% conversion of cellobiose and a 73.7% selectivity of
goluconic acid. The high catalytic activity of Au/TiO2 is considered to depend on the catalyst as a
whole, especially the strong metal-support interaction between gold nanoparticles and titania support,
rather than the sole size range of Au nanoparticles.
Electrocatalytic oxidation is another potential technique for biomass resource utilization due to its
high reaction activity, low energy consumption and environmental compatibility. Many carbohydrates
could be effectively transformed in an electrochemical system under mild conditions [10–15].
For example, the conversion of glucose can be performed through electrocatalytic oxidation obtaining
a high selectivity to gluconic acid [16,17]. As we known, the glucose is a compulsory intermediate
for the transformation of cellulose, thus it makes possible for the selective conversion of cellulose to
gluconic acid in an electrocatalytic oxidation process. However, the investigations of cellulose directly
transforming via electrocatalytic oxidation were still quite limited, which might be ascribed to the
water insolubility of cellulose. Recently, alkaline aqueous solutions, such as NaOH, have been reported
the novel green solvents of cellulose, which can dissolve cellulose by a freezing-thawing process [18].
Moreover, the dissolved cellulose has been revealed more readily to be hydrolyzed because of the
broken of the intra- and inter-hydrogen bonding of cellulose [19]. Hence, we attempted to degrade
the cellulose pretreated by concentrated NaOH solution through electrocatalytic oxidation in this
study. Au nanoparticles supported on carbon aerogel (Au/CA) was designed as the electrocatalytic
anode. Many studies indicated that Au nanoparticles, especially those with the size less than 10 nm,
show superior catalytic performance and high selectivity in the conversion of cellobiose to gluconic
acid [9,20]. Moreover, Au can be applied in an alkaline medium with a superior stability. The relatively
high PH condition can even promote its catalytic activity [20,21]. While the CA support is selected
not only by its good electrical conductivity but also by the excellent adsorption properties due to its
high BET surface area and large pore volume [22]. Moreover, as a carbon support, it is expected to
promote the cellulose hydrolyzing into glucose via the activation of β-1-4-glycosidic bond of cellulose,
especially when the support was pretreated by concentrated mineral acid [9,23].
In this work, the properties of Au/CA electrocatalytic anode and the reaction condition were
carefully investigated for exploring the high selective conversion efficiency of cellulose. Moreover,
the main intermediates were also monitored with the variation of reaction time, which helps to
understand the probable mechanism for the electrocatalytic oxidation of cellulose in alkaline condition.

2. Results and Discussion

2.1. Dissolution of Cellulose

Since cellulose is not dissolved in most solutions including water, which greatly inhibits its
hydrolysis, the concentrated NaOH solution was used for the pretreatment of cellulose. Obviously,
via a frozen-thawing process, the cellulose-NaOH solution presented a nearly homogeneous aqueous
state (shown in Figure 1a). More dissolution imformation is included in supplementary information
(Figure S1). Even Comparing with the XRD patterns for the dissolved and undissolved cellulose,
Sugano, et al. [19] indicated that the intra- and inter-hydrogen bonding of cellulose are broken with
the pretreatment of NaOH, which implies the cellulose molecular chain can migrate more easily in
the solution and fully contact the catalytic electrode. Moreover, many studies also demonstrated
that the hydrolysis of cellulose can be greatly promoted by the good dissolution of cellulose [19,24].
Deng et al. pointed out that the active sites of the glycosidic bond in dissolved cellulose can be more
readily attacked and lead to the cellulose hydrolysis to glucose [25]. In an electrochemical oxidation
system, sufficient ¨ O2 , ¨ OOH and chemisorbed active oxygen species are generated, which are proved
Catalysts 2016, 6, 5 3 of 10

to be Catalysts
active 2016,
radicals
6, 4 towards the hydrolysis of cellulose [11,14], suggesting efficient glucose3/10
might be
produced in the cellulose-NaOH solution.
Catalysts 2016, 6, 4 3/10

Figure 1. Photographs of cellulose solution: (a) Dissolved in NaOH solution; the products after
Figure 1. Photographs electrolysisof
on; cellulose
(b) Au/carbon solution: (a)
aerogel (CA); (c) Dissolved
50 nm‐Au/CA; in NaOH solution; the products after
(d) Au/graphite.

electrolysis on;2.2.
(b) Au/carbon aerogel (CA); (c)
Characterization of Electrocatalytic Anode 50 nm-Au/CA; (d) Au/graphite.
Figure 1. Photographs of cellulose solution: (a) Dissolved in NaOH solution; the products after
electrolysis on;Figure 2 shows the SEM images of the CA support and Au/CA electrocatalyst. We can see that
(b) Au/carbon aerogel (CA); (c) 50 nm‐Au/CA; (d) Au/graphite.
the used CA support has a loose interconnected 3D network structure consisting of granular carbon
2.2. Characterization of Electrocatalytic Anode
sphere with the pore size mainly in the range from several to a dozen nanometers. Au nanoparticles
2.2. Characterization of Electrocatalytic
loaded are highly dispersed onAnode
the entire CA surface without aggregation. The further magnified
Figure 2 showsimagethe SEM
showed thatimages of the CA
the Au nanoparticles support
are mainly and Au/CA
sphere structure electrocatalyst.
with the uniform particles sizes We can see that
Figure
the used CA 2 shows
support nm.the
of ~10 has a SEM
The loose images of the
interconnected
Au content determined byCA support
3D network
TG analysis and
is ~2% Au/CA electrocatalyst.
structure
(as shown in Figure consistingWe
S2). can see that
of granular carbon
the used CA support has a loose interconnected 3D network structure
sphere with the pore size mainly in the range from several to a dozen nanometers. Au nanoparticlesconsisting of granular carbon
sphere
loaded with thedispersed
are highly pore size mainly
on theinentire
the rangeCAfromsurfaceseveral to a dozen
without nanometers.
aggregation. The Aufurther
nanoparticles
magnified
loaded are highly dispersed on the entire CA surface without aggregation. The further magnified
image showed that the Au nanoparticles are mainly sphere structure with the uniform particles sizes
image showed that the Au nanoparticles are mainly sphere structure with the uniform particles sizes
of ~10 nm. The Au content determined by TG analysis is ~2% (as shown in Figure S2).
of ~10 nm. The Au content determined by TG analysis is ~2% (as shown in Figure S2).

Figure 2. SEM images of (A) CA; (B) Au/CA; (C) magnification of Au/CA.

XRD patterns of the nanoparticles were obtained to investigate their phase and structure, as
displayed in Figure 3a. The peaks at 22.1° and 43.2° were corresponded the typical characteristic
peaks of CA [22]. Those at 38.1°, 44.4°, 64.6° and 77.6° are attributed to the {111},{200},{220} and {311}
of Au, respectively. However, we can clearly see from Figure 3a, that Au/CA with the Au nanoparticle
size of ~10 nm shows relatively diffuse diffractions compared to those in 50 nm‐Au/CA, which are
similar to the 9 nm‐Au/C reported by Wang [26]. This might be ascribed to the presence of low lattice
stabilized gold atoms or clusters of atoms. As we known, the atoms or atoms clusters with low lattice
stability are usually considered active sites or hot spots, which afford good electrocatalytic activity [27].
According to the equation of Debye‐Scherrer, a rough estimate based on the {111} peak indicates that
FigureFigure
the average
2. SEM2. SEM
size of Au images ofof
nanoparticles
images (A)(A) CA;
along
CA; (B){111}
to (B)
the Au/CA; (C)
(C)ismagnification
lattice face
Au/CA; about 7.3 nm, in of
magnification Au/CA.
comparable
of Au/CA. with
the result of the SEM measurement. In addition, XPS was also recorded for the as‐prepared Au/CA
to evaluate the chemical states of the different species. Figure 3b and 3c display the XPS spectra of the
XRD patterns of the nanoparticles were obtained to investigate their phase and structure, as
C1s and Au4f regions. From Figure 3b, we can see the C1s peaks centered at 284.3 eV is corresponded to
XRD patterns
displayed of the
in Figure
elemental 3a.
nanoparticles
The
carbon, peaks
while those atat
were
22.1°
285.5 andobtained
and 288.2 eV43.2° wereto
are assigned
investigate the
tocorresponded
their
the oxygen bound species
phase and structure,
typical
C–O and characteristic
as displayed in Figure 3a.
C=O, respectivelyThe peaks
[28]. at
While the 22.1
peaks at˝ 84.0
and and43.2 ˝ were
87.7 eV corresponded
can be attributed the
to the 4f7/2 and 4f5/2typical
. It is characteristic
peaks of CA [22]. Those at 38.1°, 44.4°, 64.6° and 77.6° are attributed to the {111},{200},{220} and {311}
reported that the 4f7/2˝of Au0 and˝ Au2O3 ˝are located at 83.8
˝ and 86.3 eV [9,28], respectively, suggesting
peaksofofAu,
CA [22]. the
Those
respectively. at 38.1 ,we
Au isHowever,
approximately 44.4
can, clearly
64.6
completely and
reduced. 77.6Figure
see from are attributed
3a, that Au/CA to the with{111},{200},{220} and {311}
the Au nanoparticle
of Au, respectively.
size of ~10 nm showsHowever, we can
relatively clearly
diffuse see fromcompared
diffractions Figure 3a,tothat Au/CA
those with the Au
in 50 nm‐Au/CA, nanoparticle
which are
similar
size of ~10 nmto the 9 nm‐Au/C
shows reported
relatively by Wang
diffuse [26]. Thiscompared
diffractions might be ascribed
to those toin
the50presence
nm-Au/CA, of low lattice
which are
stabilized
similar to the 9gold atoms orreported
nm-Au/C clusters of
byatoms.
WangAs we known,
[26]. the atoms
This might or atoms
be ascribed toclusters with low
the presence of lattice
low lattice
stability are usually considered active sites or hot spots, which afford good electrocatalytic activity [27].
stabilized gold atoms or clusters of atoms. As we known, the atoms or atoms clusters with low lattice
According to the equation of Debye‐Scherrer, a rough estimate based on the {111} peak indicates that
stability are usually considered active sites or hot spots, which afford good electrocatalytic activity [27].
the average size of Au nanoparticles along to the {111} lattice face is about 7.3 nm, in comparable with
According to the equation of Debye-Scherrer, a rough estimate based on the {111} peak indicates that
the result of the SEM measurement. In addition, XPS was also recorded for the as‐prepared Au/CA
the average sizethe
to evaluate of chemical
Au nanoparticles along
states of the to the
different {111}Figure
species. lattice3b
face
andis3c
about 7.3the
display nm,XPSin spectra
comparable
of the with
the result
C1s andofAu
the4f regions.
SEM measurement.
From Figure 3b, Inweaddition, XPS
can see the C1swas
peaksalso recorded
centered for eV
at 284.3 theisas-prepared
correspondedAu/CA to
to evaluate
elementalthecarbon,
chemicalwhile states
thoseofat the
285.5different
and 288.2species. Figure to
eV are assigned 3b,c
thedisplay the XPS
oxygen bound spectra
species C–Oofand the C1s
and AuC=O, respectively
4f regions. [28].Figure
From While 3b,
the peaks
we can at see
84.0 the
andC87.7 eV cancentered
1s peaks be attributed to theeV
at 284.3 4f7/2
is and 4f5/2. It is to
corresponded
reported that the 4f of Au 0 and Au2O3 are located at 83.8 and 86.3 eV [9,28], respectively, suggesting
elemental carbon, while those at 285.5 and 288.2 eV are assigned to the oxygen bound species C–O and
7/2

C=O,the Au is approximately
respectively [28]. While completely
the peaks reduced.
at 84.0 and 87.7 eV can be attributed to the 4f7/2 and 4f5/2 . It is
reported that the 4f7/2 of Au0 and Au2 O3 are located at 83.8 and 86.3 eV [9,28], respectively, suggesting
the Au is approximately completely reduced.
Catalysts 2016, 6, 5 4 of 10
Catalysts 2016, 6, 4 4/10

84.0

Au(111)
284.3

(a) (b) (c)

87.7

CA

Au(200)
CA

Au(311)
Au(220)
285.5
50nm-Au/CA 288.2

10nm-Au/CA

290 288 286 284 282 90 88 86 84 82

20 40 60 80
Binding Energy (eV) Binding Energy (eV)
2θ

(a) (b) (c)

Figure 3. (a) X‐ray diffraction (XRD) patterns of Au/CA and 50nm‐Au/CA; (b) X‐ray photoelectron
Figure 3. (a) X-ray diffraction (XRD) patterns of Au/CA and 50nm-Au/CA; (b) X-ray photoelectron
spectrum analysis (XPS) of C1s; (c) XPS of Au4f.
spectrum analysis (XPS) of C1s ; (c) XPS of Au4f .

2.3. Electrochemical Oxidation of Cellulose

2.3. Electrochemical Oxidation of Cellulose
The pretreated cellulose was subsequently transformed in the electrochemical system. To evaluate
theThe pretreated
efficiency cellulose
of Au/CA, was subsequently
the Au/graphite transformed
and Au/CA in whichinthethesize
electrochemical system.
of Au nanoparticles areTo~50
evaluate
nm
thewere
efficiency
also used as anodes. Obviously, whatever the electrocatalysts selected, the brown solutions cannm
of Au/CA, the Au/graphite and Au/CA in which the size of Au nanoparticles are ~50
were also usedafter
be obtained as anodes. Obviously,
electrolysis with thewhatever
decreasedthe electrocatalysts
viscosity selected,
(Figure 1b–1d). the brown solutions
As neutralized by sulfuriccan
beacid
obtained afterminor
solution, electrolysis with found
solids were the decreased viscosityfrom
in the electrolyte (Figure 1b–d). Asanode,
Au/graphite neutralized by sulfuric
suggesting the
acid solution,conversion
incomplete minor solids were found in the electrolyte from Au/graphite anode, suggesting the
of cellulose.
incomplete conversion
The liquids were of cellulose.
then analyzed by HPLC. The observed oxidized products from electrolysis on
different anodes
The liquids are then
were displayed in Table
analyzed 1. Obviously,
by HPLC. by using
The observed 10 nm‐Au/CA
oxidized productsasfrom
anode, the main on
electrolysis
productanodes
different is gluconate (HPLC curve
are displayed displayed
in Table in Figure S3)
1. Obviously, by with
usingthe10maximum
nm-Au/CA yield
as reaching
anode, the67.8%
main
after 18 h of electrolysis. Except for that, minor oxalate, formate and unidentified
product is gluconate (HPLC curve displayed in Figure S3) with the maximum yield reaching 67.8% product was also
detected,
after 18 h ofand the sum yield
electrolysis. is ~89%.
Except However,
for that, minorby using the
oxalate, Au/graphite
formate or 50 nm‐Au/CA
and unidentified productas anode,
was also
the selectivity
detected, and the forsum
gluconic
yieldacid is evidently
is ~89%. However, decreased,
by using accompanying
the Au/graphitewith the increase
or 50 of otherassmall
nm-Au/CA anode,
molecular organic acid salts such as glycolate and acetate. In addition, the sum yield
the selectivity for gluconic acid is evidently decreased, accompanying with the increase of other small of the produced
organic acid salts was also prominently reduced as using Au/graphite to be the anode, which is only
molecular organic acid salts such as glycolate and acetate. In addition, the sum yield of the produced
52% after 18 h of electrolysis.
organic acid salts was also prominently reduced as using Au/graphite to be the anode, which is only
52% after 18 hTableof electrolysis.
1. Yield of organic acid salt from electrocatalytic oxidation on different anodes.

Yield (%)oxidation on different anodes.

Table 1. Yield of organic acid salt from electrocatalytic
Electrocatalyst
Gluconate Glycolate Acetate Oxalate Formate Sum Yields
Au/graphite 32.7 5.4 4.3Yield (%) 6.5 3.4 52.3
Electrocatalyst
50 nm‐Au/CA 46.9 7.9 6.1 10.5 8.7 80.1
Gluconate Glycolate Acetate Oxalate Formate Sum Yields
Au/CA 67.8 almost no almost no 11.5 9.6 88.9
Au/graphite 32.7 5.4 4.3 6.5 3.4 52.3
50 nm-Au/CA 46.9 7.9 6.1 10.5 8.7 80.1
As we known, the catalytic activity of an electrocatalyst is affected not only by the nature of
Au/CA 67.8 almost no almost no 11.5 9.6 88.9
catalytic material, but also by the support properties. The support possessing high surface area and
large pore volume can enhance the adsorption and mass transport of the reactants/intermediates
during
As we theknown,
reaction,
thewhich can activity
catalytic greatly ofincrease the conversion
an electrocatalyst efficiency.
is affected notHence,
only bybytheusing nature N2 of
adsorption‐desorption, we detected the BET surface area and pore volume of the
catalytic material, but also by the support properties. The support possessing high surface area and catalysts, as shown
in Table
large pore 2. Clearly,
volume can weenhance
can see the
the adsorption
CA supportand hasmass
a high BET surface
transport of the area of 747cm2∙g−1. With
reactants/intermediates
pretreated
during by concentrated
the reaction, which HNO 3, the surface
can greatly area the
increase of CA further increases
conversion and reaches
efficiency. Hence,1004
by cm using2∙g−1,
N2
accompanying with the increase of pore volume from 0.4 to 0.52 m 3∙g−1. Although the loading of Au
adsorption-desorption, we detected the BET surface area and pore volume of the catalysts, as shown
in nanoparticles leadswe
Table 2. Clearly, to acan
decrease
see the of CA
the surface
supportarea
hasand poreBET
a high volume,
surfaceit still
areareaches 499 m
of 747cm ∙g´−11 .and
2 ¨ 2g With
0.45 cm 3∙g−1, respectively, far high than those of Au/graphite. Besides that, the pretreatment
pretreated by concentrated HNO3 , the surface area of CA further increases and reaches 1004 cm ¨ g´1 , of CA 2by
concentrated HNO3 can also introduce the surface functional groups on CA [29,30], which has been
accompanying with the increase of pore volume from 0.4 to 0.52 m3 ¨ g´1 . Although the loading of
proved through the IR detections of CA with and without HNO3 pretreatment (as shown in Figure S4).
Au nanoparticles leads to a decrease−1of the surface area and pore volume, it still reaches 499 m2 ¨ g´1
From Figure S4, a band at ~1631 cm is observed occurring in the spectrum of HNO3 activated CA,
and 0.45 cm3 ¨ g´1 , respectively, far high than those of Au/graphite. Besides that, the pretreatment of
assigned to –COO− group [30]. According to the references [9,31,32], surface oxidized carbons are
CA by concentrated HNO3 can also introduce the surface functional groups on CA [29,30], which has
promising water‐tolerant materials that also demonstrated high catalytic performance for hydrolysis
been proved through the IR detections of CA with and without HNO3 pretreatment (as shown in
of cellulose or cellobiose. This might be attributed to its favorable adsorption to oxygen atoms of the
Figure S4). From Figure S4, a band at ~1631 cm´1 is observed occurring in the spectrum of HNO3
Catalysts 2016, 6, 5 5 of 10

activated CA, assigned to –COO´ group [30]. According to the references [9,31,32], surface oxidized
carbons are promising water-tolerant materials that also demonstrated high catalytic performance for
hydrolysis of cellulose or cellobiose. This might be attributed to its favorable adsorption to oxygen
atoms of the 1-4-glycosidic bonds in cellulose molecules, so as to benefit the catalytic performance.
Moreover, the introduce of surface oxidized functional groups to carbon support can also increase the
interaction between Au nanoparticles and support, and at the same time, promote the selectivity of
gluconic acid [7,9].

Table 2. Structure properties of the catalytic anodes.

Structure Properties CA a CA Au/CA Au/Graphite

BET surface area (cm2 g´1 ) 747 1004 499 5.78
Pore volume (m3 g´1 ) 0.40 0.52 0.45 -
a CA support without the pretreatment by concentrated.

In addition, although Au is one of the most active electrocatalysts for the oxidation of
cellulose [9,33], the high yield and high selectivity to gluconic acid (or gluconate) depends mainly on
the size, dispersion and electronic state of Au. For the 50 nm-Au/CA anode, it obtained a high cellulose
conversion; however, the selectivity of the gluconate is relatively lower, which might be ascribed to
the larger size of Au nanoparticles. Several studies demonstrated that the Au nanoparticles with the
diameter less than 10 nm have high performance of oxidation of glucose into gluconic acid [28,34,35].
In Au/CA electrode, the gold nanoparticles are highly dispersed on the surface of CA, with a uniform
size of ~10 um, approving for its high efficiency.
Considering that oxygen usually plays important roles in the selective oxidation of glucose to
gluconic acid, particularly Au-based catalysts [4,7,28], the effect of oxygen was also studied on cellulose
electrochemical oxidation by whether bubbling air into the reaction system or not. Experiment results
show that the products of cellulose electrolysis are constant regardless of with or without air aeration,
accompanying with a nearly unchanged products distribution. However, the aeration of air can
evidently increase the reaction rate. From Figure 4, we can see, after 18 h electrolysis, the gluconic acid
yield and sum acids yield are 49.6% and 68.2%, respectively, as no air aeration, while those reach 67.8%
and 88.9%, respectively, with bubbling of air. In the degradation of the lignin by PbO2 /Pb anode,
Zhang et al. [11] considered the insufflated air enhances the conversion of O2 to superoxide anion
radical (O2 ´ ) on cathode surface, leading to a higher efficiency for lignin transformation. Burke and
Nugent [36] studied the electrochemistry of gold and concluded that the dissolved oxidant can induce
Au to form AuOx at active sites, which often act as the mediators for oxidation reaction. While the
monomeric Au(III) hydroxyl complexes of the form Au(OH)6 3´ is proposed by Doyle and Lyons [37].
Moreover, the increment of oxygen can also effectively increase the O2 adsorption on Au surfaces by
forming linear O–Au–O structure, contributing to the high catalysis efficiency of gold [38,39]. Hence,
the aeration of air is deduced to increase the active Au-oxide species resulting in a better transformation
of cellulose. More importantly, the aeration of air can accelerate the cathode reaction. It is known that,
in alkaline media, H2 O and dissolved oxygen are the possible reductions which can be reduced at
cathode. As H2 O acts as the reductant, it reacts via the hydrogen evolution reaction:

2H2 O ` 2e´ é H2 ` 2OH´ (1)

As oxygen acting as the reductant, it is generally reduced via two pathways [40,41],
i.e., a four-electron pathway leading directly to OH´ , expressed as:

O2 ` 2H2 O ` 4e´ é 4OH´ (2)

Catalysts 2016, 6, 5 6 of 10

or a two-electron pathway with hydrogen peroxide (HO2 ´ in alkaline media) as the intermediate:

O2 ` H2 O ` 2e´ é HO2 ´ ` OH´ (3)

HO2 ´ ` H2 O ` 2e´ é 3OH´ (4)

Catalysts 2016, 6, 4 6/10
whether the reduction of oxygen to OH´ or to HO2 ´ , it is relatively easier than water to hydrogen
Catalysts 2016, 6, 4 6/10
in
an alkaline condition because it requires a more negative potential [41,42]. Moreover, the reduction
of oxygen is, to some extent, limited by its transport rate to the cathode. With bubbling into air, the
of oxygen
of oxygen is,is,totosome
someextent,
extent, limited byits
limited by itstransport
transport rate
rate to to
thethe cathode.
cathode. WithWith bubbling
bubbling into
into air, theair,
transport of oxygen can be greatly accelerated, which can effectively increase the reduction reaction
thetransport
transportofofoxygen
oxygencancanbebegreatly
greatlyaccelerated,
accelerated,which
whichcancaneffectively
effectively increase
increase the
the reduction
reduction reaction
reaction
rate at cathode, resulting in the promotion of reaction efficiency.
raterate at cathode,
at cathode, resulting
resulting ininthe
thepromotion
promotionofofreaction
reactionefficiency.
efficiency.

100
100
Formate Air
Formate Air
Oxalate
Oxalate
8080 Gluconate
Gluconate Air
Air

6060
Yield (%)
Yield (%)

4040 Air
Air

2020

00 18h
6h
6h 12h
12h 18h
Reaction Time
Reaction Time

Figure
Figure
Figure 4. 4. Yield
4. Yield
Yield ofof
of organic
organic
organic acidsalt
acid
acid saltfrom
salt fromelectrocatalytic
from electrocatalyticoxidation
electrocatalytic oxidation of
oxidation of
ofcellulose
celluloseon
cellulose onAu/CA
on Au/CAanode
Au/CA anode
anodewith
with
with
andand without air bubbled in reaction system.
and without
without air
air bubbled
bubbled in
in reaction
reaction system.
system.

For better understanding the mechanism of cellulose oxidation on Au/CA anode, the
For better
For better understanding
understanding the the mechanism
mechanismof ofcellulosecelluloseoxidation
oxidation on on Au/CAAu/CAanode, the
anode,
intermediates were monitored by HPLC with variation of reaction time. Figure 5 shows the yields of
intermediates
the intermediates werewere
monitored by HPLC with variation of reaction time. Figure 5 shows 5the yieldstheof
gluconate, oxalate andmonitored by HPLC
formate, respectively, with variation
at different of reaction
the reaction time.time.
It canFigure shows
be observed that
gluconate,
yields oxalate
of gluconate, and formate,
oxalaterapidly respectively,
and formate, at different
respectively, at thenthe
different reaction time.
the reaction It can be observed
time. Itreduced that
can be observed
gluconate was produced in the initial 10 h, and it increased at a slightly rate till
gluconate
that was
gluconate produced
was producedrapidly in the
rapidly initial
in the 10 h,
initial and
10 h,then
and it increased
then it
it reached the maximum yield of 67.8% at 18 h. Continuously increasing the reaction time,
at a
increased slightly
at a reducedreduced
slightly rate
the
till
it reached
rate till it the
reachedmaximum
the yield
maximum of 67.8%
yield of at
67.8%18 h.
at Continuously
18 h. Continuously increasing
increasing
concentration of gluconate was decreased, accompanying with the evident concentration increases of the reaction
the time,
reaction the
time,
concentration
the concentration
oxalate, of gluconate
implying of gluconate waswas
the oxalate decreased,
decreased,
might be theaccompanying
accompanying
product withwith
for gluconate thethe
evident
deep evidentconcentration
Thus, aincreases
concentration
oxidation. increases
probable of
oxalate,
of oxalate,
mechanismimplying
implying thethe
oxalate
oxalate
for electrocatalytic might
might bebethe
oxidation theproduct
of product
cellulosefortogluconate
for gluconatedeep
gluconate in anoxidation.
deep oxidation. Thus,
Thus, aa probable
alkaline condition probable
can be
mechanism
mechanism
proposed for for electrocatalytic
electrocatalytic
as Scheme 1. oxidation of cellulose to gluconate in an alkaline
oxidation of cellulose to gluconate in an alkaline condition can condition can be be
proposed as Scheme
proposed as Scheme 1. 1.
100
Oxalate
100
Gluconate
80 Oxalate
Formate
Gluconate
Sum of three salts
80 Formate
60 Sum of three salts
Yield (%)

60
Yield (%)

40

40
20

20
0
0 5 10 15 20 25

0 Reaction Time (h)

0 5 10 15 20 25
Figure 5. Yield of organic acid salt from electrocatalytic oxidation
Reaction Time (h) of cellulose on Au/CA anode varied
with reaction time.
Figure 5.
Figure 5. Yield
Yield of
of organic acid salt
organic acid salt from
from electrocatalytic
electrocatalytic oxidation
oxidationof
of cellulose
celluloseon
onAu/CA
Au/CA anode
anode varied
varied
with reaction time.
with reaction time.
 20

0
0 5 10 15 20 25
Reaction Time (h)

Catalysts Figure
2016, 6, 5.
5 Yield of organic acid salt from electrocatalytic oxidation of cellulose on Au/CA anode varied 7 of 10
with reaction time.

Scheme 1. Mechanism for electrocatalytic oxidation of cellulose to gluconate.

Scheme 1. Mechanism for electrocatalytic oxidation of cellulose to gluconate.

As for formate, it displays an inconspicuous increase with the decomposition of gluconate,

deduced likely to be the direct oxidation of glucose via α-scission [24,43]. In addition, the total yield
of gluconate, oxalate and formate is also calculated and shown in Figure 5. Surprisingly, the sum
yields were very high with the value of ~91% at the reaction time of 22 h, implying the high conversion
efficiency of cellulose on Au/CA anode in alkaline condition.

3. Experimental Section

3.1. Materials
Cellulose and HAuCl4 ¨4H2 O were purchased from Sigma-Aldrich (Shanghai, China). Other
chemicals used were analytical grade. All solutions were prepared with deionized water.

3.2. Pretreatment of Cellulose

Before the electrochemical oxidization process, 1 g cellulose was dissolved in 133.3 mL NaOH
solution(0 ˝ C, 12 wt. %) under stirring at 1000 rpm for 2 h, and then the suspension was stored at
´20 ˝ C for 24 h. The frozen sample was consequently thawed at room temperature and diluted with
deionized water to the volume of 200 mL. Thus, final concentration of cellulose in NaOH solution
was 5 g L´1 .

3.3. Preparation of Au/CA Electrode

CA support is fabricated with the method reported in our previous study [22]. Briefly, resorcinol,
formaldehyde, deionized water and sodium carbonate were mixed with a fixed molar ratio forming
a homogeneous solution, which is then sealed into a cuboid glass reactor and cured at 30, 50 and
90 ˝ C, separately, to obtain organic wet gel. The as-prepared gel was dried via solvent exchange in
acetone under ambient condition and then carbonized in a tubular oven at 950 ˝ C for 4 h in nitrogen
atmosphere to convert to CA. Before loading Au nanoparticles, the CA was pretreated by 68% HNO3
at room temperature for 24 h.
The Au/CA electrode was prepared by using the improved method from Ma, et al. [28]. A solution
was mixed with HAuCl4 (0.39 mg Au mL´1 ) and 2 wt. % PVP with vigorous stirring. To this
solution, a freshly prepared NaBH4 solution was added dropwise, with a NaBH4 /Au molar ratio of 4.
This process is performed in an ice-water condition, which lead to the formation of Au nanoparticles
with a moderate reaction rate. The immobility of Au nanoparticles was achieved by repeatedly
immersing the pretreated CA into the mixed solution for 10 min and then drying at 70 ˝ C for 30 min
for 5 times. Finally, the as-prepared Au/CA catalyst was washed by deionized water fully and dried
at 70 ˝ C for 5 h, then calcined at 200 ˝ C for 3 h. For the purpose of comparison, Au/graphite catalyst
was prepared with the same method mentioned above, and the Au nanoparticles with a size of about
50 nm were also prepared and deposited on CA with a HAuCl4 concentration of 0.59 mg Au mL´1
and the same method.
Catalysts 2016, 6, 5 8 of 10

3.4. Catalyst Characterization

Surface morphology of the catalyst was determined by scanning electron microscopy (FESEM,
Hitach S-4800, Tokyo, Japan) equipped with energy-dispersive X-ray spectrometer (EDS). The surface
area and pore size of the catalysts were carried out by nitrogen adsorption at 77 K using BET
analyzer (Micromeritics Tristar 3000, Norcross, GA, USA). The Au content was determined with
thermogravimetric analysis (TG/DSC, NETZSCH STA 409PC, Bavaria, Germany). The crystallographic
structure was determined by X-ray diffraction (XRD, Bruker D8, saarbrücken, Germany) with Cu Kα
(λ = 0.154056 nm) at a voltage of 40 KV, a current of 40 mA, and a scanning rate of 0.02 step/s.
The surface properties of the catalyst were detected by X-ray photoelectron spectrum analysis
(XPS, Shimadzu Axis-Ultra DLD, Manchester, Britain) with Mg Kα X-rays radiation (Hν = 1253.6 eV).
In addition, cyclic voltammetry (CV) was measured by Electrochemical Workstation (CHI660, Chenhua,
Shanghai, China) for obtaining the electrochemical properites of cellulose solution on Au nanoparticles
electrodes with the results displaying in Figures S5 and S6.

3.5. Electro-Oxidation Reaction

The electrocatalytic oxidation of cellulose was carried out in a cylindrical single compartment cell
system equipped with air-blowing device. The constructed Au/CA electrode with an area of 4 cm2
was used as anode and a Pt sheet with the same area was used as cathode. The gap of electrodes
between electrodes was 1 cm. The current density used was controlled at 10 mA cm´2 with a cell
voltage of ~2.75 V. Air was bubbled into the reaction system at a flow rate of 150 mL min´1 . Each time,
a 50 mL pretreated cellulose solution was used for transformed.

3.6. Products Analysis

After reaction, the alkaline liquid was adjusted by sulfuric acid to pH = 2. Then it was syringe
filtered with 0.45 µm PTFE membrane and analyzed by a high-performance liquid chromatography
(HPLC, Agilent 1260, Santa Clara, CA, USA) on an Agilent Hi-plex H column at a UV detection
wavelength of 210 nm. The mobile phases are using 5 mM sulfuric acid solution with a flow rate of
0.6 mL min´1 and the injection condition is 55 ˝ C.

4. Conclusions
In this work, the electrocatalytic oxidation of cellulose was performed on the concentrated HNO3
pretreated CA supported Au nanoparticles anode in an alkaline medium with the aeration of air.
Results show that cellulose can be effectively degraded on Au/CA anode and a high gluconate
yield of 67.8% can be obtained after 18 h of electrolysis. The high conversion of cellulose and high
selectivity to gluconic acid could be attributed not only to the properties of Au nanoparticles but
also the CA support. Moreover, the dissolution of cellulose in NaOH solution can promote the
hydrolysis of cellulose, which also plays important role in the cellulose conversion. Additionally,
a probable mechanism for electrocatalytic oxidation of cellulose to gluconate in an alkaline condition
was also proposed.

Acknowledgments: This work was supported by the National Natural Science Foundation of China
(NO. 51208367, 21477085).
Author Contributions: The experimental work was conceived and designed by Guohua Zhao and
Hanshuang Xiao, Hanshuang Xiao performed the experiments; Hanshuang Xiao and Meifen Wu analyzed
the data and Hanshuang Xiao drafted the paper. The manuscript was amended through the comments of all
authors. All authors have given approval for the final version of the manuscript.
Conflicts of Interest: The authors declare no conflict of interest.
Catalysts 2016, 6, 5 9 of 10

References
1. Besson, M.; Gallezot, P.; Pinel, C. Conversion of Biomass into Chemicals over Metal Catalysts. Chem. Rev.
2014, 114, 1827–1870. [CrossRef] [PubMed]
2. Wang, L.; Xiao, F.S. Nanoporous catalysts for biomass conversion. Green Chem. 2015, 17, 24–39. [CrossRef]
3. Krochta, J.M.; Hudson, J.S. Alkaline thermochemical degradeation of rice straw to organic acids. Agric. Wastes
1985, 14, 243–254. [CrossRef]
4. An, D.L.; Ye, A.H.; Deng, W.P.; Zhang, Q.H.; Wang, Y. Selective Conversion of Cellobiose and Cellulose
into Gluconic Acid in Water in the Presence of Oxygen, Catalyzed by Polyoxometalate-Supported Gold
Nanoparticles. Chem. Eur. J. 2012, 18, 2938–2947. [CrossRef] [PubMed]
5. Saxena, R.C.; Adhikari, D.K.; Goyal, H.B. Biomass-based energy fuel through biochemical routes: A review.
Renew. Sustain. Energy Rev. 2009, 13, 167–178. [CrossRef]
6. Deng, W.P.; Zhang, Q.H.; Wang, Y. Catalytic transformations of cellulose and cellulose-derived carbohydrates
into organic acids. Catal. Taday 2014, 234, 31–41. [CrossRef]
7. Amaniampong, P.N.; Li, K.; Jia, X.; Wang, B.; Borgna, A.; Yang, Y. Titania-Supported Gold Nanoparticles
as Efficient Catalysts for the Oxidation of Cellobiose to Organic Acids in Aqueous Medium. ChemCatChem
2014, 6, 2105–2114. [CrossRef]
8. Cantero, D.A.; Martinez, C.; Bermejo, M.D.; Cocero, M.J. Simultaneous and selective recovery of cellulose
and hemicellulose fractions from wheat bran by supercritical water hydrolysis. Green Chem. 2015, 17, 610–618.
[CrossRef]
9. Tan, X.S.; Deng, W.P.; Liu, M.; Zhang, Q.H.; Wang, Y. Carbon nanotube-supported gold nanoparticles as
efficient catalysts for selective oxidation of cellobiose into gluconic acid in aqueous medium. Chem. Commun.
2009, 7179–7181. [CrossRef] [PubMed]
10. Yu, J.C.; Bauzer, M.M.; Nobe, K. Electrochemical Generated Acid Catalysis of Cellulose Hydrolysis.
J. Electrochem. Soc. 1986, 135, 83–87. [CrossRef]
11. Zhang, Y.M.; Peng, Y.; Yin, X.L.; Liu, Z.H.; Li, G. Degradation of lignin to BHT by electrochemical catalysis
on Pb/PbO2 anode in alkaline solution. J. Chem. Technol. Biotechnol. 2014, 89, 1954–1960. [CrossRef]
12. Yang, F.; Zhang, Q.; Fan, H.X.; Li, Y.; Li, G. Electrochemical control of the conversion of cellulose
oligosaccharides into glucose. J. Ind. Eng. Chem. 2014, 20, 3487–3492. [CrossRef]
13. Xin, L.; Zhang, Z.Y.; Qi, J.; Chadderdon, D.; Li, W.Z. Electrocatalytic oxidation of ethylene glycol (EG) on
supported Pt and Au catalysts in alkaline media: Reaction pathway investigation in three-electrode cell and
fuel cell reactors. Appl. Catal. B 2012, 125, 85–94. [CrossRef]
14. Kwon, Y.; Kleijn, S.E.F.; Schouten, K.J.P.; Koper, M.T.M. Cellobiose hydrolysis and decomposition by
electrochemical generation of acid and hydroxyl radicals. ChemSusChem 2012, 5, 1935–1943. [CrossRef]
[PubMed]
15. Meng, D.F.; Li, G.; Liu, Z.H.; Yang, F. Study of depolymerization of cotton cellulose by Pb/PbO2 anode
electrochemical catalysis in sulfuric acid solution. Polym. Degrad. Stab. 2011, 96, 1173–1178. [CrossRef]
16. Delidovich, I.V.; Moroz, B.L.; Taran, O.P.; Gromov, N.K.; Pyrjaev, P.A.; Prosvirin, I.P.; Bukhtiyarov, V.I.;
Parmon, V.N. Aerobic selective oxidation of glucose to gluconate catalyzed by Au/Al2O3 and Au/C: Impact
of the mass-transfer processes on the overall kinetics. Chem. Eng. J. 2013, 223, 921–931.
17. Pasta, M.; Mantia, F.L.; Cui, Y. Mechanism of glucose electrochemical oxidation on gold surface.
Electrochem. Acta 2010, 55, 5561–5568. [CrossRef]
18. Duchemin, B.J.C. Mercerisation of cellulose in aqueous NaOH at low concentrations. Green Chem. 2015, 17,
3941–3947. [CrossRef]
19. Sugano, Y.; Vestergaard, M.; Yoshikawa, H.; Saito, M.; Tamiya, E. Direct Electrochemical Oxidation of
Cellulose: A Cellulose-Based Fuel Cell System. Electroanalysis 2010, 22, 1688–1694. [CrossRef]
20. Zhang, J.Z.; Liu, X.; Hedhili, N.J.; Zhu, Y.H.; Han, Y. Highly Selective and Complete Conversion of Cellobiose
to Gluconic Acid over Au/Cs2 HPW12 O40 Nanocomposite Catalyst. ChemCatChem 2011, 3, 1294–1298.
[CrossRef]
21. Biella, S.; Prati, L.; Rossi, M. Selective oxidation of D-glucose on gold catalyst. J. Catal. 2002, 206, 242–247.
[CrossRef]
Catalysts 2016, 6, 5 10 of 10

22. Wu, M.F.; Jin, Y.N.; Zhao, G.H.; Li, M.F.; Li, D.M. Electrosorption-promoted Photodegradation of Opaque
Wastewater on A Novel TiO2 /Carbon Aerogel Electrode. Environ. Sci. Technol. 2010, 44, 1780–1785.
[CrossRef] [PubMed]
23. Komanoya, T.; Kobayashi, H.; Hara, K.; Chun, W.J.; Fukuoka, A. Catalysis and characterization of
carbon-supported ruthenium for cellulose hydrolysis. Appl. Catal. A 2011, 407, 188–194. [CrossRef]
24. Deng, W.P.; Zhang, Q.H.; Wang, Y. Catalytic transformations of cellulose and its derived carbohydrates into
5-hydroxymethylfurfural, levulinic acid, and lactic acid. Sci. China Chem. 2015, 58, 29–46. [CrossRef]
25. Yan, L.F.; Qi, X.Y. Degradation of Cellulose to Organic Acids in its Homogeneous Alkaline Aqueous Solution.
ACS Sustain. Chem. Eng. 2014, 2, 897–901. [CrossRef]
26. Wang, S.; Zhao, Q.; Wei, H.; Wang, J.Q.; Cho, M.; Cho, H.; Terasaki, O.; Wan, Y. Aggregation-Free Gold
Nanoparticles in Ordered Mesoporous Carbons: Toward Highly Active and Stable Heterogeneous Catalysts.
J. Am. Chem. Soc. 2013, 135, 11849–11860. [CrossRef] [PubMed]
27. Plowman, B.J.; O’Mullane, A.P.; Bhargava, S.K. The active site behaviour of electrochemically synthesised
gold nanomaterials. Faraday Discuss. 2011, 152, 43–62. [CrossRef] [PubMed]
28. Ma, C.Y.; Xue, W.J.; Li, J.J.; Xing, W.; Hao, Z.P. Mesoporous carbon-confined Au catalysts with superior
activity for selective oxidation of glucose to gluconic acid. Green Chem. 2013, 15, 1035–1041. [CrossRef]
29. Ruffo, M.P.R.; Falletta, E.; Mari, C.M.; Pina, C.D. Alkaline glucose oxidation on nanostructured gold
electrodes. Gold Bullet. 2010, 43, 57–64.
30. Wu, C.; Yan, P.T.; Zhang, R.J.; Jin, J.L.; Zhang, X.H.; Kang, H.M. Comparative study of HNO3 activation effect
on porous carbons having different porous characteristics. J. Appl. Electrochem. 2015, 45, 849–856. [CrossRef]
31. Zhao, X.; Wang, J.; Chen, C.; Huang, Y.; Wang, A.; Zhang, T. Graphene oxide for cellulose hydrolysis: How it
works as a highly active catalyst? Chem. Commun. 2014, 50, 3439–3442. [CrossRef] [PubMed]
32. Delidovich, I.; Palkovits, R. Impacts of acidity and textural properties of oxidized carbon materials on their
catalytic activity for hydrolysis of cellobiose. Microporous Mesoporouws Mater. 2016, 219, 317–321. [CrossRef]
33. Sugano, Y.; Latonen, R.M.; Akieh-Pirkanniemi, M.; Bobacka, J.; Ivaska, A. Electrocatalytic Oxidation of
Cellulose at a Gold Electrode. ChemSusChem 2014, 7, 2240–2247. [CrossRef] [PubMed]
34. Qi, P.Y.; Chen, S.S.; Chen, J.; Zheng, J.W.; Zheng, X.L.; Yuan, Y.Z. Catalysis and Reactivation of Ordered
Mesoporous Carbon-Supported Gold Nanoparticles for the Base-Free Oxidation of Glucose to Gluconic Acid.
ACS Catal. 2015, 5, 2659–2670. [CrossRef]
35. Ishimotoa, T.; Hamatakea, Y.; Azunob, H.; Kishidab, T.; Koyamaa, M. Theoretical study of support effect of
Au catalyst for glucose oxidation of alkaline fuel cell anode. Appl. Surf. Sci. 2015, 324, 76–81. [CrossRef]
36. Burke, L.D.; Nugent, P.F. The electrochemistry of gold: II The electrocatalytic behavior of the metal in
aqueous media. Gold Bullet. 1998, 31, 39–50. [CrossRef]
37. Doyle, R.; Lyons, M.E. The mechanism of oxygen evolution at superactivated gold electrodes in aqueous
alkaline solution. J. Solid State Electrochem. 2014, 18, 3271–3286. [CrossRef]
38. Sun, K.; Kohyama, M.; Tanaka, S.; Takeda, S. Direct O2 Activation on Gold/Metal Oxide Catalysts through a
Unique Double Linear O–Au–O Structure. ChemCatChem 2013, 5, 2217–2222. [CrossRef]
39. Sun, K.; Kohyama, M.; Tanaka, S.; Takeda, S. Theoretical Study of Atomic Oxygen on Gold Surface by Hückel
Theory and DFT Calculations. J. Phys. Chem. A 2012, 116, 9568–9573. [CrossRef] [PubMed]
40. Genies, L.; Bultel, Y.; Faure, R.; Durand, R. Impedance study of the oxygen reduction reaction on platinum
nanoparticles in alkaline media. Electrochimica Acta 2003, 48, 3879–3890. [CrossRef]
41. Ge, X.M.; Sumboja, A.; Wuu, D.; An, T.; Li, B.; Goh, F.W.T.; Hor, T.S.A.; Zong, Y.; Liu, Z.L. Oxygen Reduction
in Alkaline Media: From Mechanisms to Recent Advances of Catalysts. ACS Catal. 2015, 5, 4643–4667.
[CrossRef]
42. Shinagawa, T.; Garcia-Esparza, A.T.; Takanabe, K. Mechanistic Switching by Hydronium Ion Activity for
Hydrogen Evolution and Oxidation over Polycrystalline Platinum Disk and Platinum/Carbon Electrodes.
ChemElectroChem 2014, 1, 1497–1507. [CrossRef]
43. Jin, F.M.; Yun, J.; Li, G.M.; Kishita, A.; Tohji, K.; Enomoto, H. Hydrothermal conversion of carbohydrate
biomass into formic acid at mild temperatures. Green Chem. 2008, 10, 612–615. [CrossRef]

© 2015 by the authors; licensee MDPI, Basel, Switzerland. This article is an open access
article distributed under the terms and conditions of the Creative Commons by Attribution
(CC-BY) license (http://creativecommons.org/licenses/by/4.0/).