A biomass-derived metal-free catalyst doped with phosphorus for highly efficient and selective oxidation of furfural into maleic acid

Article abstract of DOI:10.1039/d0gc04205f

The present work introduces an extremely simple and eco-friendly strategy for the highly selective synthesis of maleic acid (MA)viaoxidation of renewable furfural using the abundant biomass-derived P-C-Tcatalyst. The P-C-Tcarbon catalyst was metal-free and prepared by the pyrolyzation of phytic acid, which is a ubiquitous natural molecule containing phosphorus. Extensive characterization was carried out to reveal the morphological and elemental properties of the synthesized P-C-Tseries. The effect of the annealing temperature of pyrolyzing phytic acid on the properties of the yielded P-C-Tand subsequent catalytic performance was also explored. The catalytic oxidation of furfural to MA was carried out in the presence of H₂O₂within an aqueous system. The reaction conditions including the catalyst loading, H₂O₂concentration, reaction temperature and duration were further optimized. It was found that P-C-600 exhibited a remarkable catalytic activity for MA synthesis from furfural oxidation with a maximum yield of 76.3% achieved in water. The excellent catalytic performance of P-C-600 was attributed to its unique atomic layered structure and suitable acidity. P-C-600 could also be re-used for at least six runs without any obvious decrease in its catalytic performance. The intrinsic advantages of green synthesis, low cost, and excellent catalytic performance in the catalytic oxidation of furfural to MA suggested that P-C-600 would be a promising catalyst in future industrial applications for MA synthesis in a green manner.

Full text of DOI:10.1039/d0gc04205f

Green Chemistry
View Article Online
View Journal | View Issue
PAPER
A biomass-derived metal-free catalyst doped with
phosphorus for highly efficient and selective
oxidation of furfural into maleic acid†
Cite this: Green Chem., 2021, 23,
370
1
a
b
c
d
Huifa Zhang, Shaolin Wang, Huixian Zhang, James H. Clark
Fahai Cao
and
a
*
The present work introduces an extremely simple and eco-friendly strategy for the highly selective syn-
thesis of maleic acid (MA) via oxidation of renewable furfural using the abundant biomass-derived P–C-T
catalyst. The P–C-T carbon catalyst was metal-free and prepared by the pyrolyzation of phytic acid,
which is a ubiquitous natural molecule containing phosphorus. Extensive characterization was carried out
to reveal the morphological and elemental properties of the synthesized P–C-T series. The effect of the
annealing temperature of pyrolyzing phytic acid on the properties of the yielded P–C-T and subsequent
catalytic performance was also explored. The catalytic oxidation of furfural to MA was carried out in the
2 2 2 2
presence of H O within an aqueous system. The reaction conditions including the catalyst loading, H O
concentration, reaction temperature and duration were further optimized. It was found that P–C-600
exhibited a remarkable catalytic activity for MA synthesis from furfural oxidation with a maximum yield of
7
6.3% achieved in water. The excellent catalytic performance of P–C-600 was attributed to its unique
Received 12th December 2020,
Accepted 18th January 2021
atomic layered structure and suitable acidity. P–C-600 could also be re-used for at least six runs without
any obvious decrease in its catalytic performance. The intrinsic advantages of green synthesis, low cost,
and excellent catalytic performance in the catalytic oxidation of furfural to MA suggested that P–C-600
would be a promising catalyst in future industrial applications for MA synthesis in a green manner.
DOI: 10.1039/d0gc04205f
rsc.li/greenchem
7
cals derivable from biomass. It is an important raw material
in the chemical industry and has been extensively applied in
Introduction
The exploration of renewable resources for sustainable pro- multiple fields, such as medicine, plasticisers, copolymers,
7
duction of chemicals and fuels is of urgent necessity and has and agrochemicals. Currently, MA is mainly produced from
attracted growing interest over the past decade due to the fossil fuels by the hydrolysis of maleic anhydride, and has
rapid depletion of fossil and other virgin resources. Biomass is high environmental burden and is limited more and more by
1
a promising renewable carbon source, and has the potential fossil fuel depletion. Hence, the efficient synthesis of MA from
to be converted into chemicals and liquid transportation fuels renewable biomass based resources is of paramount
at a large scale. It has been converted to many platform mole- importance.
2
3
cules, such as formic acid, levulinic acid, 2-furfuraldehyde
(
In fact, many attempts have been devoted to the production
furfural), 5-hydroxymethyl-2-furfural, and lactic acid that of MA using biomass as the starting material. There are two
4
5
6
8
–13
14,15
could be further used to synthesize a multitude of value-added main routes: the furfural
route and the HMF route.
products. Maleic acid (MA) is one of the value-added chemi- Both routes are based on selective oxidation over various cata-
lysts, either homogeneous or heterogeneous. Due to the pro-
blems of easy degradation and cross-polymerization, the syn-
a
Engineering Research Centre of Large Scale Reactor Engineering and Technology of
thesis of HMF from biomass at a large scale has not yet been
Ministry of Education, East China University of Science and Technology, Shanghai
1
6
successful. Compared with HMF, furfural has been success-
fully produced from lignocellulose commercially, which paves
the way for further MA synthesis.
2
b
00237, China. E-mail: fhcao@ecust.edu.cn
Shanghai Key Laboratory of Green Chemistry and Chemical Processes, School of
Chemistry and Molecular Engineering, East China Normal University, Shanghai
2
00062, China
For MA synthesis from furfural, both gas and liquid reac-
tion systems have been reported. The liquid phase reaction
system was preferred, since the reaction in the vapor phase
can suffer from the drawbacks of low substrate concentration
c
Sinopec North China E&P Company, Zhengzhou 450006, China
d
Green Chemistry Centre of Excellence, University of York, York YO105DD, UK
†
Electronic supplementary information (ESI) available. See DOI: 10.1039/
d0gc04205f
1370 | Green Chem., 2021, 23, 1370–1381
This journal is © The Royal Society of Chemistry 2021

View Article Online
Green Chemistry
Paper
and low MA yield because of the elevated reaction temperature. more, most P doping methods are accompanied by the for-
Various homogeneous and heterogeneous catalytic systems mation of various P and O containing functional groups, such
3
1,32
were then extensively explored for carrying out the reaction as phosphonate and phosphate groups,
under aerobic conditions or in the presence of hydrogen per- lysts acidic.
making the cata-
3
3,34
Hence, it was anticipated that P doped carbon
oxide in the liquid phase. Homogeneous systems are usually materials could facilitate the reaction to selectively produce
performed under mild conditions, but can suffer from con- more MA from furfural. Considering the advantages of P
siderable pollution. The first study of homogeneous systems doped carbon materials mentioned above and the fact that
can be traced back to the report by Yin et al. in 2011. In their phytic acid is extremely abundant and easily available from
work, the reaction was carried out in liquid media using phos- legumes and cereals, the current work was designed and
phomolybdic acid as the catalyst and dioxygen as the carried out to investigate the use of phytic acid as a starting
1
7
oxidant. A 49.2% yield of MA was achieved under the opti- material for the P–C catalyst for the catalytic oxidation of fur-
mized reaction conditions with a 51.7% selectivity toward MA. fural to MA.
In order to enhance the selectivity of MA, organic solvents
Herein, we report an extremely simple and eco-friendly
were further introduced into the reaction systems by means of strategy for the highly selective production of MA from furfural
1
8
the reaction extraction method. Later, other kinds of in- using a P-doped carbon catalyst derived from phytic acid. The
organic protic acids, such as sulfuric acid and hydrochloric catalysts were prepared by the pyrolysis of phytic acid at
19
acid, were investigated for the catalytic oxidation of furfural different annealing temperatures under an Ar atmosphere (P–
to MA with hydrogen peroxide as the oxidant. Due to the extre- C-T). Extensive characterization was carried out to reveal the
mely unstable character of furfural under acidic conditions, morphological and elemental properties of the synthesized P–
these homogeneous inorganic acid catalysts usually gave low C-T series. The prepared P–C-T metal-free carbon catalysts
MA yields. It was later demonstrated that C1–C4 organic acids were then used for MA synthesis in the presence of hydrogen
1
2
a
can be catalytic systems as well as solvents. The pK of peroxide at 333 K with pure water as the solvent. The reaction
organic acids was also shown to play a crucial role in the cata- conditions of the catalytic oxidation of furfural to MA was
lytic oxidation of furfural to MA with a 95% MA yield achieved further optimized when P–C-600, the best catalyst among the
in formic acid media. Adding an organic solvent to the reac- tested ones, was used. The reusability of P–C-600 was also eval-
tion system is well acknowledged to enhance the yield of MA, uated to substantiate its practical application potential. This
but this is not consistent with the principles of green chem- cheap, readily available, and environmentally friendly carbon
istry. What’s more, the tedious procedures and high cost catalyst is believed to be a promising catalyst for industrial
cannot be ignored for the recovery of homogeneous catalysts applications of MA synthesis from a renewable resource in a
from the products after the reaction. Alternatively, hetero- green manner.
geneous catalysts were then developed to circumvent the above
problems. Although metal catalysts, such as vanadium oxide-
1
7,20
8,11,21
22
based,
titanium silicate-1 (TS-1),
copper phosphate
Materials and methods
Materials
2
3
and Mo–V–O catalysts, can give relatively high MA yields in
the liquid phase oxidation of furfural, metal catalysts are also
problematic due to potential leaching and deactivation during
the reaction processes. In comparison, carbon-based materials
are more promising as they have several distinct advantages,
including large surface area, corrosion resistance, and the
ability to tailor physicochemical and electronic properties.
Furfural (99%) and MA were of analytical grade and purchased
from Aladdin Chemicals Co., Ltd (Shanghai, China). Fumaric
acid was purchased from Sinopharm Chemical Reagent Co.,
Ltd (Shanghai, China). 50% (w/w) Phytic acid aqueous solution
was purchased from Tianjin Guangfu Chemical Industry
24
(
(
Tianjin, China). Hydrogen peroxide aqueous solution 30%
w/w) and other main reagents used in the study were pur-
Sulfonated carbon materials, such as Amberlyst-15, Nafion
2
4
24
25,26
NR50, Nafion SAC13 and SO3-GO,
were then developed
chased from Shanghai Titan Scientific Co., Ltd (Shanghai,
China) and they were analytically pure. All chemicals except
furfural were used as received without further purification.
Furfural was purified by distillation before use.
to convert furfural into MA and SA, but such types of carbon
catalysts containing sulfonic acid groups usually afforded a
higher yield of SA over MA. Carbon materials doped with
heteroatoms, such as nitrogen (N), boron (B), or phosphorus
(
P), are currently under intense investigation. A few pioneering
Catalyst preparation
trials were reported using heteroatom doped carbon catalysts
to catalytically convert furfural to MA in the presence of hydro-
gen peroxide.
Preparation of the P doped carbon catalyst. The P–C-T
9
,27
However, these N doped carbon catalysts are carbon catalyst was synthesized according to the following pro-
far from industrialization due to their performance.
cedures. 10 g of 50% phytic acid solution was first subjected to
A P doped carbon catalyst was proven to be an excellent hydrothermal treatment in an oven at 120 °C for 24 h. The
2
8–31
catalyst in the oxidation of hydrocarbons and alcohols.
semi-solidified mixture was then collected and calcined in a
The lower electronegativity and larger diameter of P atoms muffle furnace at different temperatures (500, 600, 700 and
endow P doped carbon materials with distinct features which 800 °C) for 4 h under an argon atmosphere with a heating rate
could probably bring out distinct catalytic activities. What’s of 3 °C min− . The obtained black granular materials were
1
This journal is © The Royal Society of Chemistry 2021
Green Chem., 2021, 23, 1370–1381 | 1371

View Article Online
Paper
Green Chemistry
3
5
ground to powder and termed P–C-T, where T represented the mated by using integrated molar extinction coefficients. The
3
1
annealing temperature.
solid state P MAS NMR experiments were carried out on a
Catalyst characterization. The overall morphologies of P–C-T Bruker Avance AV 300 spectrometer at a Larmor frequency of
carbon catalysts were analyzed using a Hitachi S-4800 scan- 600 MHz using a single pulse with a high power decoupling
ning electron microscope (SEM) instrument operated at an sequence. Inductively coupled plasma mass spectrometry
accelerating voltage of 3.0 kV. The microstructures of the P–C- (ICP-MS) measurement was performed using an Agilent
T carbon catalysts were tested on a Tecnai G2 F30 transmission ICP-MS 4500 instrument to detect the leaching of P to the reac-
electron microscope (TEM) instrument operating at an acceler- tion solution.
ating voltage of 200 kV. TEM samples were prepared by dissol-
Catalytic reaction. The catalytic oxidation of furfural was
ving the P–C-T catalyst in ethanol and were transferred onto a carried out in a 70 ml Teflon-lined stainless steel reactor. The
copper TEM grid coated with a holey support film. The solvent reactor was integrated with an oil bath and a magnetic stirrer,
of the dropped sample was evaporated in air at room tempera- which could be used to control the temperature and stirring
ture before testing. The phase composition of the synthesized rates of the reactor precisely. In a typical reaction, the pro-
samples was determined using a Bruker D8 Advanced XRD cedure for furfural oxidation was as follows: 2.5 mmol furfural,
(X-ray diffractometer) in the 2θ range between 10 and 80° with 1.96 g 30 wt% H
2 2 2 2
O (H O , 20 mmol, determined by iodo-
Cu-kα (λ = 1.541 Å) operating at room temperature at a scan- metric titration), 0.15 g P–C-T carbon catalyst and 5 ml water
−1
ning rate of 0.016° s . To analyze the physical surface charac- were successively added into the above reactor. After sealing
teristics of the synthesized carbon catalysts, N adsorption–de- the valve, the sealed reactor was heated using the oil bath to
sorption measurements were carried out using an ASAP 2020 the specified temperature within a certain time duration. The
Micromeritics). The samples were degassed in a vacuum at a reaction was carried out under vigorous stirring and main-
2
(
temperature of 350 °C for 6 h to remove impurities prior to the tained at the reaction temperature for a specific duration. After
measurements. Surface areas were estimated using the the reaction, the mixture was cooled to room temperature with
Brunauer–Emmett–Teller (BET) model. Pore size distribution iced water. The products were collected and separated using a
of carbon samples was calculated via the Barrett–Joyner– filter. Quantitative analysis of the product was performed
Halenda (BJH) method. Raman spectra of the samples were using HPLC equipped with an XDB-C18 reversed phase
recorded using a laser with an excitation wavelength of 785 nm column and a UV detector (mobile phase 5 : 95 ratio of aceto-
−
1
at room temperature using the Horiba Jobin Yvon LabRAM nitrile: 0.01% phosphoric acid, flow rate 0.6 mL min
HR800 Raman spectrometer. The elemental composition and 240 nm).
bonding information of carbon catalysts was determined by
,
X-ray photoelectron spectroscopy (XPS, PHI-5702). Generally,
the charge-up shift of all samples was corrected by the stan-
dard binding energy peak of C 1s (284.6 eV). Elemental
Results and discussion
mapping was carried out using STEM-EDS. In order to study Characterization of carbon catalysts
the acidic properties of the prepared carbon catalysts, tempera-
In this study, a renewable biomass-derived molecule, phytic
acid, a unique natural substance containing phosphorus ubi-
quitously found in legumes and grains, was chosen as the
starting material of P–C-T carbon catalysts. Phytic acid is a six-
fold dihydrogen phosphate ester of inositol containing the
same level of C and P in one molecule, thus guaranteeing a
uniform P doping in the fabricated carbon catalysts. The syn-
thesis process of P–C-T is illustrated in Scheme 1. The as-
received phytic acid solution (50 wt%) was used directly
without further purification. After being heated in an oven at
120 °C for 24 h, the semi-solidified mixture was annealed at
different temperatures for 4 h under an argon atmosphere.
The obtained black carbon residues were then ground to
powder and termed P–C-T catalyst.
ture programmed desorption of ammonia (NH -TPD) of
3
samples was carried out on a Micromeritics ChemiSorb 2920
equipped with a TC detector (TCD). Before carrying out the
TPD, samples of 100 mg were first thermally treated in situ at
3
00 °C for 2 h in flowing helium to remove the gases physi-
sorbed on the sample surface. NH was chemisorbed on the
catalyst at 50 °C for 1 h and then desorption of ammonia was
3
−
1
carried out under a flow rate of 40 ml min of helium from 50
−
1
to 700 °C at a linear heating rate of 10 °C min . Desorbed
NH3 was measured quantitatively using the TCD detector.
FT-IR pyridine adsorption technique was used to investigate
the Lewis and Brønsted acidic sites of the catalysts. Pyridine
infrared spectroscopy (Py-IR) was recorded on a Bruker Vertex
7
0. Self-supporting disks (6.5 mm diameter, 10 mg) of the
samples were prepared applying 6 ton pressure. Prior to the
measurements, the samples were dehydrated by heating at
4
00 °C for 3 h under vacuum. After cooling down, pyridine
adsorption was conducted in a saturated pyridine atmosphere
at room temperature for 5 h. Subsequent pyridine desorption
was carried out at different temperatures for 30 min and the
infrared spectrum was recorded. The concentration of
Brønsted and Lewis acidic sites in the carbon samples was esti- Scheme 1 Conversion route of phytic acid to P–C-T.
1372 | Green Chem., 2021, 23, 1370–1381
This journal is © The Royal Society of Chemistry 2021

View Article Online
Green Chemistry
Paper
All of the P–C-T carbon samples prepared from phytic acid
precursors exhibited an aggregated morphology with particles
of a size of several micrometers according to the images
shown in Fig. 1 (left), while the microporous structure of them
was shown by the TEM images (right column of Fig. 1). The as-
prepared P–C-T catalysts possessed a lacunaris graphene-like
structure with an obvious thickness sheet. It seemed that the
morphology of the P–C-T catalysts was affected by the anneal-
ing temperature. More pores could be developed with the
increase in the pyrolysis temperature, resulting in more defects
evolved in the carbon materials. In addition, it was also clear
that from the TEM images that the thickness of the materials
became thin obviously with the increase of pyrolysis tempera-
ture. High-resolution TEM analysis was further carried out to
reveal the texture structure of the P–C-600 catalyst, and abun-
dant irregular pores could thus be clearly observed (Fig. 2). Fig. 2 High-resolution TEM image and elemental mapping of P–
C-600.
The corresponding STEM-EDS elemental mapping images of
the P–C-600 catalyst are shown in Fig. 2, and it could then be
confirmed that P element was uniformly incorporated into the
graphene-like carbon framework.
Since it is well known that specific surface area and pore
size distribution are two important features of catalysts,
especially the carbon materials for catalytic performance, BET
measurement of the as-prepared P–C-T was performed. As
shown in Fig. 3a, a typical type-IV isotherm curve with a
H4 hysteresis loop at low-relative and medium pressure was
observed, indicating the coexistence of both mesoporous and
microporous structures in all as-synthesized P–C-T catalysts.
These results were consistent with the TEM images. The
respective surface areas were calculated via the (BET) method,
2
Fig. 3 (a) N adsorption/desorption isotherms and (b) BJH pore size
distribution curves of P–C-T.
as shown in Table 1. P–C-500, P–C-600, P–C-700 and
2
−1
P–C-800 had the specific surface areas of 33.7 m
1
g ,
2
−1
2
−1
2
−1
038.6 m g , 1521.6 m g and 1637.4 m g , respectively
(
Table 1). The surface areas of the P–C-T catalysts were greatly
affected by the pyrolysis temperature. The higher the anneal-
ing temperature, the higher the specific surface area of the
catalyst sample was. It was interesting to notice that the pore
size of all samples could be divided into two types: micropores
with diameters of 0.02–0.8 nm and mesopores with diameters
of 0.2–3.5 nm. Both the microporous and mesoporous pore
size also greatly increased with an increase of the pyrolysis
temperature. A significant porosity enhancement was obtained
when the annealing temperature increased to 500 °C and
higher. It could then be ascertained that the morphological
features of the P–C-T carbon catalysts largely depended on the
treatment temperature based on the above results.
Table 1 Structural properties of P–C-T
Surface area
Pore size
(nm) meso/
micro
Pore volume
(ml g ) meso/
micro
2
−1
−1
(
m g ) meso/
Entry
micro
P–C-500
P–C-600
P–C-700
P–C-800
33.7
1038.6
1521.6
1637.4
32.5
387.6
774
0.2
0.02
18.8
3.5
3.5
0.86
1.02
0.48
0.41
0.92
0.95
3.5
0.17
0.35
0.84
Fig. 1 SEM (left) and TEM (right) images of the synthesized P–C-T.
898
1.09
This journal is © The Royal Society of Chemistry 2021
Green Chem., 2021, 23, 1370–1381 | 1373

View Article Online
Paper
Green Chemistry
The X-ray diffraction patterns of the prepared carbon instead of graphitization might be evolved in the carbon struc-
samples are shown in Fig. 4. Two peaks at around 24.5 degree tures under a high pyrolysis temperature, thus resulting in an
and 44 degree could be evidently observed in the XRD patterns increase in I
of P–C-600, P–C-700, and P–C-800, whereas only one peak at be grafted into the carbon lattice as more phosphorus atoms
4.5 degree could be distinguished pertaining to P–C-500. The would be peeled off from the carbon lattice. Accounting to the
D G
/I . At the same time, less P heteroatoms could
2
dominant and broad peak centered at 24.5 was corresponding fact that the defect level of carbon materials and the degree of
to the (002) reflection of graphite, suggesting that the syn- P heteroatoms doped on the carbon catalyst both determined
thesized samples were highly amorphous with a typical turbos- the catalytic activity, the balance of the defect level and P
3
6,37
tratic graphite structure.
The peak centered at 44 degree doping degree on the current carbon catalyst should be opti-
reflected the size of the aromatic layers. The intensity of the mized to favor the catalysis.
peak at around 44 degree enhanced with the annealing temp-
In order to quantitatively determine the elemental compo-
erature, suggesting the formation of intralayer condensation in sition and valence states of elements in the P–C-T, XPS was
the graphite layers in the catalysts synthesized under a higher carried out and the results are shown in Fig. 6. It was noted
pyrolysis temperature.
that four predominant peaks at 284.6 eV, 532 eV, 193 eV, and
Raman spectroscopy was further used to investigate the 133 eV corresponding to C 1s, O 1s, P 2s, and P 2p, respect-
doping effect of graphene. The Raman spectra of all samples ively, appeared in the XPS spectra of all P–C-T carbon
(
Fig. 5) displayed a typical D band and a G band peak appeared materials, confirming the successful synthesis of carbon
−
1
at 1330 and 1600 cm , respectively. The relative strength of materials doped with P by annealing the biomass derived
the D-band is a reflection of the degree of the sp defect sites phytic acid containing phosphorus under an elevated tempera-
3
in the graphite structure and the G-band is associated with the ture. This was consistent with the above EDS mapping results.
2
sp carbon networks. The relative intensity ratio of I
/I
D G
ratio The relative percentages of various atoms of the P–C-T catalysts
is typically used to evaluate the level of defects and heteroatom calculated from the XPS spectra are summarized in Table 2. It
doping. The ID/IG ratios of P–C-500, P–C-600, P–C-700 and should be noted that the atomic relative composition of P
P–C-800 were 2.55, 2.73, 1.93 and 1.91, respectively. The results decreased significantly with the increase in the annealing
indicate that pyrolysis was a complicated process. More defects temperature, but with a concomitant increase in the percen-
tage of C atoms. A similar trend was observed for the element O.
The decrease of P and O atomic concentration with the
increase in the pyrolysis temperature could be attributed to the
fact that the heteroatoms could be desquamated from the
Fig. 4 X-ray diffraction patterns of P–C-T.
Fig. 6 XPS scan spectra of P–C-T.
Table 2 XPS analysis on the element contents of P–C-T
Total (%)
Calculated (%)
Entry Catalyst C (at%) P (at%) O (at%) P–C (at%) P–O (at%)
1
2
3
4
P–C-500 38.17
P–C-600 72.82
P–C-700 82.06
P–C-800 82.86
13.59
6.37
3.69
3.58
48.24
20.8
14.25
13.56
0.38
0.19
0.15
0.2
13.21
6.18
3.54
3.38
Fig. 5 Raman spectra of P–C-T.
1374 | Green Chem., 2021, 23, 1370–1381
This journal is © The Royal Society of Chemistry 2021

View Article Online
Green Chemistry
Paper
doped carbon materials under a high pyrolysis temperature.
Considering that the acidity strength of the catalyst is directly
3
4,38
related to the phosphorus content,
a high treating temp-
erature may lead to the decrease of the acidity of the catalyst.
The C 1s peak in the high resolution XPS was deconvoluted
into three components attributed to O–CvO (289.5 eV), C–C
(284.6 eV), and C–P (285.9 eV) bonds, respectively (Fig. S1†).
The deconvolution of the high-resolution P 2p band described
two types of P species at 132.6 eV and 134.2 eV, corresponding
to the P–C and P–O groups, respectively (Fig. 7). The formation
of the P–C bond under an elevated temperature indicated that
P atoms were incorporated into the carbon lattice at a higher
carbonization. Furthermore, the relative peak area ratio of P
species was greatly affected by the pyrolysis temperature. The
relative peak area ratio of P–C gradually increased with the
increase in the pyrolysis temperature, whereas the one of P–O
decreased in turn. A similar phenomenon has been reported
3
Fig. 8 NH -TPD profiles of P–C-T.
TPD method and the results are shown in Fig. 8. All of the syn-
thesized samples exhibited two broad NH desorption peaks at
around 100 °C and 300 °C, respectively. The former was attrib-
uted to weak acidic sites, while the latter was usually con-
sidered to be the medium acid sites. It could be seen from
Fig. 8 that P–C-500 and P–C-600 owned more medium-strength
acid sites than P–C-700 and P–C-800. In addition, it is worth
noting that the total acid amounts introduced to carbon
decreased when the pyrolysis temperature increased from 500
to 800 °C. This might be attributed to the fact that the hetero-
atoms grafted on carbon could be destroyed under an elevated
temperature, thus resulting in less phosphonate and phos-
phate groups in the P–C-T catalysts.
3
9,40
3
by many research studies.
This may be due to the fact that
the dissociation energy of the P–C bond is higher than that of
4
1
the P–O bond. The polarity of the P–C bond is opposite to
4
5
3
1
that of the N–C bond. The lower electronegativity of P atoms
than the C atoms makes the P atoms in the P–C bond with
positively charge, which may result in its Lewis acidity in
4
2
the catalyst.
from the surface P–O groups (phosphonate and phosphate
The Brønsted acidity might be originated
3
9,43,44
groups).
In summary, it was reasonable to suggest that
the treating temperature not only greatly affected the acidity
strength of the catalyst, but also the types of acidic sites. The
pronounced O 1s spectrum was fitted into three sub-peaks,
C/PvO (531.4 ± 0.1 eV), C/P–O–C (532.5 ± 0.1 eV) and C/P-OH
To further investigate the types of acidic sites, the surface
acidity was studied by pyridine adsorption experiments. The
IR spectra of adsorbed pyridine over P–C-600 at different temp-
eratures are shown in Fig. 9. Three adsorption peaks at about
(533.5 ± 0.1 eV), respectively (Fig. 7). Considering all the above
results, the successful synthesis of P–C catalysts can be
ascertained.
As mentioned in the introduction, the acid of this P doped
carbon material might play a determining role in the catalytic
oxidation of furfural to MA. The acid amount and strength of
540 cm− , 1490 cm and 1450 cm , corresponding to the
1
−1
−1
1
adsorption of pyridine at the Brønsted acidic sites (B),
Brønsted and Lewis acidic sites (B + L) and Lewis acidic sites
(
L), respectively, were clearly observed in all the samples.
3
the as-synthesized P–C-T catalysts was then analyzed by a NH -
However, as compared to P–C-500 (Fig. S2†), the major acid
sites were Brønsted acid sites, while most of the acid sites of
Fig. 9 Pyridine-FTIR spectra of P–C-600 obtained after evacuation at
different temperatures.
Fig. 7 XPS high-resolution P 2p (left) and O 1s (right) spectra of P–C-T.
This journal is © The Royal Society of Chemistry 2021
Green Chem., 2021, 23, 1370–1381 | 1375

View Article Online
Paper
Green Chemistry
4
6–48
31
the other three carbon catalysts were Lewis acid sites. This four different P–O moieties.
The P NMR spectra of other
could be attributed to the fact that more relative P–C bonds three carbon catalysts exhibited different behaviours with the
were evolved under a higher treating temperature leading to P–C-500 catalyst. Almost no obvious peaks could be observed
3
1
the generation of relatively higher Lewis acidity sites. What’s in the P NMR spectra of the three catalysts. Furthermore, the
more, the band of the three obvious peaks of the four carbon peak at around −11 ppm gradually decreased with increasing
catalysts decreased dramatically with the increase in the de- pyrolysis temperature and vanished at 800 °C. All these results
sorption temperature, suggesting that almost all kinds of acid indicated that the P–O bond gradually decreased with the
sites on these four catalysts are relatively weak. These results increase of the catalyst calcination temperature, which was in
3
1
3
were coincident with the NH -TPD results. The above results agreement with the XPS results. P NMR spectra of the P–C
revealed that the increase in the pyrolysis temperature can species were barely detected. This might be attributed to two
reduce the strength of the strong acid sites of the obtained cat- reasons. Small amount of P atoms were grafted onto the
alysts. The quantitative results of the acid sites obtained by the carbon lattice leading the signals lower than the noise. On the
3
1
integral peak areas are summarized in Table S1.† other hand, P NMR spectra of the P–C bond are usually rela-
Quantification of the medium and strong acid sites was per- tively broad peaks, which are difficult to distinguish.
formed using the data obtained from the spectra of the
3
5
adsorbed pyridine according to the method reported. As Catalytic performance
shown, the medium acidic sites of P–C-500 and P–C-600 were
Catalytic oxidation of furfural into MA over various catalysts.
The catalytic activity of the as-synthesized P–C-T catalysts was
evaluated by the oxidation of furfural in the presence of H O
2 2
more than their weak acidic sites and strong acidic sites,
respectively. Total acid amounts decreased gradually with
increasing the carbonization temperature from 500 to 800 °C.
and the results are presented in Table 3. For comparison,
carbon derived from glucose was also investigated for this reac-
tion. As shown in Table 3, when the oxidation reaction was
conducted without any catalysts, only 30.1% of furfural could
be converted with a trace yield of maleic acid (Table 3, entry 1),
which confirmed the necessity of catalysts for this reaction.
The low yield of MA could be attributed to the fact that furfural
was highly prone to non-selective polymerization, forming
furanic resins. The conversion of furfural and the selectivity of
MA over the C-600 catalyst was 51.5% and 26.0%, respectively
3
This was consistent with the NH -TPD results. It is worth
noting that the total Lewis acid sites of the P–C-T reached a
maximum at around 600 °C. This might be due to the fact that
more P atoms were incorporated into the carbon lattice under
a higher treating temperature above 600 °C. Further increasing
the pyrolysis temperature above 600 °C to 800 °C would
destroy the P atoms in the carbon structure, resulting in a
decrease in the total Lewis acid sites. Considering that the
acidity of the catalysts might be significantly beneficial for the
oxidation of furfural to MA, P–C-500 and P–C-600 obtained
under a lower annealing temperature might have more active
sites.
(
Table 3, entry 2). It was interesting to notice that the furfural
conversion and MA selection was tremendously enhanced
upon the application of current P–C-T as the catalysts. As
shown in Table 3 (entries 3–6), the highest furfural conversion
was obtained when P–C-500 was used, while the highest MA
yield (62%) was achieved with P–C-600. The furfural conver-
sion decreased in an order of P–C-500, P–C-600, P–C-700 and
P–C-800 with the same loading of catalysts, whereas the selecti-
vity of MA exhibited a different phenomenon, suggesting that
the catalytic activity of P–C-T catalysts was pronouncedly
affected by the annealing temperature. It was worth noting
that the conversion of furfural was positively related to the
total number of acid sites, while the selectivity of MA was posi-
tively related to the amount of Lewis acid sites. P–C-600 could
The chemical environment of the P atoms in the syn-
thesized material had been investigated with solid state NMR
3
1
and the results are shown in Fig. 10. P NMR spectra of
P–C-500 showed four dominant signals centered at 31 ppm,
−11 ppm, −25 ppm, and at −42 ppm which corresponded to
Table 3 Oxidation of furfural to MA over various catalysts in the pres-
ence of H O
2 2
Catalyst
loading (mg)
Furfural
Con. (%)
MA yield
(%)
MA Sel.
(%)
Entry
Catalyst
1
2
3
4
5
6
7
Blank
C-600
P–C-500
P–C-600
P–C-700
P–C-800
P–C-1200
0
30.1
51.5
96.6
89.9
72.5
69.3
60.6
3.3
13.4
42.9
62
42.8
39.5
18.5
11
26
44.4
69
59
57
30.6
100
100
100
100
100
100
3
1
Fig. 10
P-NMR spectra of P–C-T.
1376 | Green Chem., 2021, 23, 1370–1381
This journal is © The Royal Society of Chemistry 2021

View Article Online
Green Chemistry
Paper
offer the best catalytic performance for the selective oxidation lowing experiments were fixed at this dosage if not otherwise
of furfural to MA (entry 4) among all synthesized P–C-T cata- indicated.
lysts with an 89.9% conversion of furfural and a 69.0% selecti-
Effects of the H
2
O
2
amount on the catalytic oxidation of fur-
concentration
vity of MA. According to the NH -TPD and Py-IR results, it was fural into MA over P–C-600. The effect of H O
3
2 2
noted that the absolute amount and relative proportion of within the reaction system on the catalytic oxidation of furfural
Lewis acid sites of P–C-600 were the highest among all the to MA was also examined over P–C-600 at 60 °C for 6 h. The
carbon catalysts synthesized. Therefore, it was speculated that correlation of furfural conversion and MA selectivity with the
the Lewis acid sites could be the active sites for the oxidation H O concentration from 10–30 mmol is thereby shown in
2
2
2 2 2 2
of furfural into MA in the presence of H O , rather than the Fig. 12. It was shown that the H O amount was also crucial
total amount of acid sites or Brønsted acid sites. For compari- for the reaction to proceed. Remarkably, within the current
son, P–C-1200 catalyst doped with less P was also tested in par- concentration range, the larger the H O concentration, the
2
2
allel under the same reaction conditions. A 18.5% yield of MA higher the furfural conversion. Of note, when the reaction was
was obtained over the P–C-1200 catalyst, which was slightly performed without H addition, the conversion of furfural
higher than the yield obtained over the C-600 catalyst. could be neglected. The MA yield could be increased from
Moreover, only 1.78% P (at%) has been detected by XPS in the 41.9% to 69.1% when the H concentration was increased
P–C-1200 catalyst, much lower than the P content in the other from 10 to 20 mmol. However, further increase in H con-
four P carbon materials. These results further confirmed that centration to 30 mm only showed a negligible improvement on
the acid sites were the key active sites in the catalytic oxidation the MA yield. Hence, it could be confirmed that H played
of furfural into maleic acid. Hence, P–C-600 was chosen as the an important role in the oxidation of furfural to MA and the
2 2
O
2 2
O
2 2
O
2 2
O
catalyst in the following experiments for further optimization optimum H O /furfural mol ratio was approximately
8
2
2
of this oxidation reaction if not otherwise indicated.
Effects of catalyst dosage on the catalytic oxidation of fur- studies.
fural into MA over P–C-600. To further optimize the reaction Effects of the reaction temperature on the catalytic oxidation
(20 mmol H
2
O
2
), which was then fixed in the following
conditions, the influence of catalyst dosage on the catalytic of furfural into MA over P–C-600. The effect of the reaction
performance with varying amounts from 0.05 to 0.3 g was temperature on the catalytic oxidation of furfural into MA over
investigated within the same reaction system. The correlation P–C-600 was also explored and the results are shown in
between catalyst dosage and furfural conversion as well as MA Fig. 13. It could be seen that both the conversion of furfural
selectivity/yield is shown in Fig. 11. As expected, the furfural and the selectivity of MA were sensitive to the reaction temp-
conversion exhibited a positive correlation with the loading of erature. Although the conversion of furfural increased with the
P–C-600 within the tested amount range, whereas the selecti- increase in the reaction temperature, the MA selectivity and
vity to MA increased first with the increase in the catalyst yield reached the maximum (74.5% and 69.1%) at 60 °C. The
amount and then decreased, suggesting that the active sites of further increase in the reaction temperature from 60 to 80 °C
catalysts could promote the formation of MA only to a certain did not result in any improvement in the MA yield. This could
extent. When the loading of the catalyst increased from 0.05 to be ascribed to two aspects. Furfural was very unstable under a
0
.15 g, the selectivity to MA increased from 59.0 to 74.5%. high temperature, which could result in the polymerization of
With the further increase from 0.15 to 0.3 g, the selectivity of furfural. On the other hand, a high reaction temperature could
MA decreased from 74.5% to 66.0%. This might be partly facilitate the decomposition of peroxides, which were crucial
attributed to the unstable characteristics of furfural, especially for the oxidation of furfural to MA. Accordingly, the optimal
in the presence of an acid catalyst. Since the maximum yield of
MA (69.1%) was achieved with 0.15 g P–C-600 catalyst, the fol-
Fig. 11 Effects of the P–C-600 loading amount on the oxidation of fur-
fural into MA. Reaction conditions: 2.5 mmol of furfural, 5 ml H O,
0 mmol H , 60 °C, 6 h.
2 2
Fig. 12 Effects of the H O amount on the oxidation of furfural into
2
MA. Reaction conditions: 2.5 mmol of furfural, 5 ml H
60 °C, 6 h.
2
O, 0.15 g catalyst,
2
2 2
O
This journal is © The Royal Society of Chemistry 2021
Green Chem., 2021, 23, 1370–1381 | 1377

View Article Online
Paper
Green Chemistry
intermediate product during the reaction, which increased at
the beginning and reached a maximum at 4 h, then decreased
rapidly beyond 4 h, while the selectivity of MA increased
accordingly.
To date, only several metal-free heterogeneous catalysts
have been investigated for this reaction. A comparison of
P–C-600 catalyst vs. metal-free heterogeneous catalysts reported
previously for the synthesis of MA from furfural is shown in
Table 4. It was evident that the P–C-600 catalysts afforded the
highest MA yield. Sulfonated carbon catalysts usually afforded
a higher yield of SA with a low yield of MA. Compared with the
sulfonated carbon catalysts, the NC-900 catalyst could afford a
relatively high MA yield. However, the preparation process of
the NC-900 catalyst was quite complex. Thus, the P–C-600 cata-
lyst exhibited a remarkable catalytic performance with a 76.3%
MA yield, which indicated good efficiency of the catalyst for
the synthesis of MA.
Fig. 13 Effects of the reaction temperature on the oxidation of furfural
into MA. Reaction conditions: 2.5 mmol of furfural, 5 ml H
lyst, 20 mmol H , 60 °C, 6 h.
2
O, 0.1 g cata-
2
O
2
Reusability of P–C-600. In addition to the catalytic perform-
ance of P–C-600 for the oxidation of furfural to MA, the re-
usability of catalysts was also an important issue in practical
applications. Therefore, the reusability tests of P–C-600 were
performed by successive reuses at 60 °C for 10 h. After each
run, the catalyst was recovered simply by centrifugation and
thoroughly washed with 75% ethyl alcohol and dried overnight
at 120 °C in an air oven. In order to avoid the impact of catalyst
loss, each reuse was scaled down. The test results are pre-
sented in Fig. 15. It could be seen that the yield of MA
remained almost constant (76%) even after four successive
runs. A minor drop was noticed in the sixth cycle. However,
when the cycles exceeded 7 runs, the catalyst showed a con-
tinuous decrease of activity.
The decrease of activity could be ascribed to the deposition
of humin formed via the polymerization of furfural on the
surface of P–C-600 during the reaction.
To further investigate the reasons for the decrease in the
activity, P in the reaction solution had been detected by ICP.
No P had been detected in the reaction solution, indicating
that the acid sites were not leached into the solution.
Moreover, the spent P–C-600 catalyst (after 8 runs) was charac-
terized by TEM, BET (Fig. 16) and XPS (Fig. S4 and S5†). It was
reaction temperature for P–C-600 catalytic oxidation of furfural
to MA was 60 °C based on the above compromised effects.
Effects of the reaction duration on the catalytic oxidation of
furfural into MA over P–C-600. The effect of the reaction dur-
ation was also studied in order to obtain the maximum MA
yield in the current reaction system. The corresponding corre-
lation of the time courses with the furfural conversion and MA
yield over P–C-600 is shown in Fig. 14. It was shown that the
conversion of furfural had a positive relation to the reaction
time. The oxidation of furfural was quite fast at the beginning
of the reaction, and it took only around 2 h to reach the fur-
fural conversion of 82.5%. The further increase in the reaction
duration to 6 h resulted in the furfural conversion of 92.8%. It
seemed that the catalyst was highly active for this reaction.
Unlike the trend of furfural conversion, the yield of MA
increased gradually and reached a maximum at 10 h (76.3%)
with a concomitant yield of furoic acid smaller than 2% at all
reaction durations. Further prolonging the reaction duration
did not result in any increase in the yield of MA. It was indi-
cated from these results that MA was stable in this reaction
system. This might be attributed to the formation of some
Table 4 The comparison of catalytic activity and selectivity of different
metal-free heterogeneous catalysts for the catalytic oxidation of furfural
to MA
MA
MA
Reaction
conditions (%)
yield Sel.
Entry Catalyst
(%)
Ref.
1
2
3
4
5
6
7
8
9
NC-900
Amberlyst-15
80 °C, 5 h
80 °C, 5 h
100 °C, 3 h 16.8
70 °C, 24 h 11.3
61.0
16.0
61.0
16.3
16.82
11.3
9.5
34.3
1.5
27
27
9
24
24
26
25
g-C N
3 4
Nafion NR50
Nafion SAC13
70 °C, 24 h
9.5
34.3
1.2
SO
SO
3
H-carbocatalysts 80 °C, 1 h
Fig. 14 Effects of the reaction time on the oxidation of furfural into
3
H/GO
70 °C, 24 h
60 °C, 6 h
60 °C, 10 h 76.3
MA. Reaction conditions: 2.5 mmol of furfural, 5 ml H
0 mmol H , 60 °C.
2
O, 0.15 g catalyst,
P–C-600
P–C-600
69.1
74.5
76.3
This work
This work
2
2
O
2
1378 | Green Chem., 2021, 23, 1370–1381
This journal is © The Royal Society of Chemistry 2021

View Article Online
Green Chemistry
Paper
Fig. 17 Oxidation of various substituted furan compounds under opti-
mized reaction conditions: substrate (1 mmol), formic acid (4 ml), 31%
2 2
H O (1 ml), 100 °C, 4 h.
Fig. 15 Cycle usage of P–C-600.
1
2,19,24
4.
2(5H)-furanone is an important intermediate associ-
ated with the formation of MA from furfural in path 4, while
Z-4-oxopent-2-enedial is reported to be one of the intermedi-
ates in pathway 3. To determine the specific reaction pathway
for the acid-catalyzed H O oxidation of furfural to MA over
2
2
the current P–C-600, three plausible intermediates, furoic acid,
furan and 2(5H)-furanone were tried under the optimized oxi-
dation conditions. As shown in Fig. 17, the yield of MA
obtained from 2(5H)-furanone was trace and only a 5% conver-
Fig. 16 High-resolution TEM image (left) and N
2
adsorption/desorption sion of 2(5H)-furanone was converted. However, with furan
isotherms (right) of the spent P–C-600. Reaction conditions: 2.5 mmol and furoic acid, the yield of MA was 89% and 13%, respect-
of furfural, 5 ml H O, 0.15 g catalyst, 20 mmol H , 6 h.
2
2 2
O
ively. In spite of the fact that furan can offer an 89% yield of
MA, furoic acid was the up-stream intermediate in route 1 and
route 2, indicating that route 1 and route 2 were unlikely to be
the oxidation route. In addition, the yield of 2-furoic acid was
always less than 2% during all oxidation processes in our reac-
tion system. Since MA was either determined to be in a very
trace amount or was even undetectable using 2(5H)-furanone
as the starting material, route 4 could be further excluded.
Taking the above results into consideration, it was speculated
that the oxidation reaction of furfural to MA over the
P–C-600 catalyst proceeded via the path 3 oxidation route.
Moreover, owing to their undersupplies and difficulty in the
synthesis, Z-4-oxopent-2-enedial was not investigated in our
study.
clear that there was no obvious difference between the fresh
P–C-600 and the spent P–C-600 catalyst from the TEM and XPS
results. However, the BET analysis results indicated that the
2
surface area of the spent P–C-600 carbon catalyst (382.6 m
−
1
g
(
) was much lower than that of the fresh P–C-600 catalyst
2
−1
1038.6 m g ). This might be attributed to humin deposited
on the active sites by the polymerization of furfural on the
surface of the catalyst during the reaction process. This depo-
sition decreased the exposure of the acid active sites, thus
resulting in a decrease in the activity of the catalyst. Although
the mechanism of humin deposition is still unknown, some
attempts have been made to remove the deposition of humins
4
9
on the catalyst. For example, low molecular weight humins
formed at a low temperature can be removed by washing with
an organic solvent. Moreover, humins can also be removed by
Conclusions
5
0
hydrogenation. It was thus speculated that the humins gen- In the present study, an acid carbon catalyst doped with P was
erated in this process might be partially or completely successfully synthesized using phytic acid, a naturally abun-
removed by hydrogenation or washing with an organic solvent. dant biobased molecule, as the starting material. The syn-
Reaction pathway for MA from furfural. Although the reac- thesized P–C-600 catalyst demonstrated a superior catalytic
tion pathway of the acid-catalyzed oxidation of furfural to MA activity in the catalytic oxidation of furfural to MA in an
in the presence of H
2 2
O was still under debate, four types of aqueous reaction system. A maximum MA yield of 76.3% was
reaction paths have been proposed, as shown in Scheme S1.† achieved at 60 °C for 10 h over P–C-600 in the presence of
1
0,51
52
In route 1
and route 2, furoic acid and furan are the two 8 equivalents of H O /furfural in the aqueous solution. The
2 2
oxidation intermediates during MA formation from furfural. A excellent catalytic performance of the P–C-600 catalyst should
1
0
Baeyer–Villiger oxidation is accompanied in path 3 and path be ascribed to the high amount of Lewis acidic sites on the
This journal is © The Royal Society of Chemistry 2021
Green Chem., 2021, 23, 1370–1381 | 1379

View Article Online
Paper
Green Chemistry
catalyst surface. The as-synthesized P–C-600 catalyst exhibited 16 F. Menegazzo, E. Ghedini and M. Signoretto, Molecules,
a similar catalytic activity when being reused for up to four 2018, 23, 2201.
cycles. The P–C-600 catalyst could be a promising candidate in 17 S. Shi, H. Guo and G. Yin, Catal. Commun., 2011, 12, 731–
industrial-level applications for MA production in a green
manner.
733.
18 H. Guo and G. Yin, J. Phys. Chem. C, 2011, 115, 17516–
17522.
1
2
9 H. Choudhary, S. Nishimura and K. Ebitani, Appl. Catal., A,
2013, 458, 55–62.
0 N. Alonso-Fagúndez, M. Ojeda, R. Mariscal, J. L. G. Fierro
and M. López Granados, J. Catal., 2017, 348, 265–
Conflicts of interest
There are no conflicts to declare.
2
75.
1 J. Su, L. Yang, R. N. Liu and H. Lin, Chin. J. Catal., 2014,
5, 622–630.
2 T. Soták, M. Hronec, M. Gál, E. Dobročka and
J. Škriniarová, Catal. Lett., 2017, 147, 2714–2723.
2
2
2
2
2
2
Acknowledgements
3
This research did not receive any specific grant from funding
agencies in the public, commercial, or not-for-profit sectors.
The authors thank the Research Center of Analysis and Test of
East China University of Science and Technology for the help
on characterization.
3 X. Li, B. Ho and Y. Zhang, Ill. Nat. Hist. Surv., Biol. Notes,
2
016, 18, 2976–2980.
4 H. Choudhary, S. Nishimura and K. Ebitani, Chem. Lett.,
012, 23, 409–411.
2
5 W. Zhu, F. Tao, S. Chen, M. Li, Y. Yang and G. Lv, ACS
Sustainable Chem. Eng., 2019, 7, 296–305.
6 U. Thubsuang, S. Chotirut, K. Nuithitikul, A. Payaka,
N. Manmuanpom, T. Chaisuwan and S. Wongkasemjit,
J. Colloid Interface Sci., 2020, 565, 96–109.
References
1
H. Mandova, S. Leduc, C. Wang, E. Wetterlund, P. Patrizio,
W. Gale and F. Kraxner, Biomass Bioenergy, 2018, 115, 231–
2
43.
27 C. Van Nguyen, J. R. Boo, C.-H. Liu, T. Ahamad,
S. M. Alshehri, B. M. Matsagar and K. C. W. Wu, Catal. Sci.
Technol., 2020, 10, 1498–1506.
2
3
4
5
6
V. P. Muiuane, M. Ferreira, P. Bignet, A. P. Bettencourt and
P. Parpot, J. Environ. Chem. Eng., 2013, 1, 1237–1244.
Y. W. Tiong, C. L. Yap, S. Gan and W. S. P. Yap, Ind. Eng. 28 X. Hu, M. Fan, Y. Zhu, Q. Zhu, Q. Song and Z. Dong, Green
Chem. Res., 2018, 57, 4749–4766. Chem., 2019, 21, 5274–5283.
Y. Luo, Z. Li, X. Li, X. Liu, J. Fan, J. H. Clark and C. Hu, 29 C. Li, M. Yang, R. Liu, F. Zhao, H. Huang, Y. Liu and
Catal. Today, 2019, 319, 14–24. Z. Kang, RSC Adv., 2014, 4, 22419–22424.
T. Deng, X. Cui, Y. Qi, Y. Wang, X. Hou and Y. Zhu, Chem. 30 Z. Long, L. Sun, W. Zhu, G. Chen, X. Wang and W. Sun,
Commun., 2012, 48, 5494–5496. Chem. Commun., 2018, 54, 8991–8994.
S. Yamaguchi, M. Yabushita, M. Kim, J. Hirayama, 31 M. A. Patel, F. Luo, M. R. Khoshi, E. Rabie, Q. Zhang,
K. Motokura, A. Fukuoka and K. Nakajima, ACS Sustainable
Chem. Eng., 2018, 6, 8113–8117.
C. R. Flach, R. Mendelsohn, E. Garfunkel, M. Szostak and
H. He, ACS Nano, 2016, 10, 2305–2315.
7
8
9
R. Wojcieszak, F. Santarelli, S. Paul, F. Dumeignil, F. Cavani 32 J. Albero, A. Vidal, A. Migani, P. Concepcion, L. Blancafort
and R. V. Gonçalves, Sustainable Chem. Processes, 2015, 3, 9.
Y. Lou, S. Marinkovic, B. Estrine, W. Qiang and
G. Enderlin, ACS Omega, 2020, 5, 2561–2568.
and H. Garcia, ACS Sustainable Chem. Eng., 2019, 7, 838–
846.
33 A. Villa, M. Schiavoni, P. F. Fulvio, S. M. Mahurin, S. Dai,
R. T. Mayes, G. M. Veith and L. Prati, J. Energy Chem., 2013,
22, 305–311.
T. Yang, W. Li, Q. Liu, M. Su, T. Zhang and J. Ma,
BioResources, 2019, 14, 5025–5044.
1
1
0 N. Alonso-Fagúndez, I. Agirrezabal-Telleria, P. L. Arias, 34 S. Campisi, F. Sanchez, D. Motta, T. Davies, N. Dimitratos
J. L. G. Fierro, R. Mariscal and M. L. Granados, RSC Adv.,
014, 4, 54960–54972.
1 Y. Rodenas, R. Mariscal, J. L. G. Fierro, D. Martín Alonso,
and A. Villa, C, 2018, 4, 9.
35 J. Datka, A. M. Turek, J. M. Jehng and I. E. Wachs, J. Catal.,
1992, 135, 186–199.
2
J. A. Dumesic and M. López Granados, Green Chem., 2018, 36 E. Billeter, D. McGlamery, M. Aebli, L. Piveteau,
2
0, 2845–2856.
2 X. Li, B. Ho, D. S. W. Lim and Y. Zhang, Green Chem., 2017,
9, 914–918.
3 Y. Ni, Z. Bi, H. Su and L. Yan, Green Chem., 2019, 21, 1075–
079.
4 X. Li and Y. Zhang, Green Chem., 2016, 18, 643–647.
M. V. Kovalenko and N. P. Stadie, Chem. Mater., 2018, 30,
4580–4589.
37 L. Zhou, Y. Sun, B. Li, Z. Li, Z. Zhang, J. Yun and R. Liu,
Catal. Commun., 2019, 126, 44–49.
38 A. M. Puziy, O. I. Poddubnaya, M. Sobiesiak and
B. Gawdzik, Adsorption, 2013, 19, 717–722.
1
1
1
1
1
1
5 L. Chai, X. Hou, X. Cui, H. Li, N. Zhang, H. Zhang, 39 W. Li, S. Jin, R. Zhang, Y. Wei, J. Wang, S. Yang, H. Wang,
C. Chen, Y. Wang and T. Deng, Chem. Eng. J., 2020, 388,
24187.
M. Yang, Y. Liu, W. Qiao, L. Ling and M. Jin, RSC Adv.,
2020, 10, 12908–12919.
1
1380 | Green Chem., 2021, 23, 1370–1381
This journal is © The Royal Society of Chemistry 2021

View Article Online
Green Chemistry
Paper
4
0 Y. Wang, S. Zuo, J. Yang and S.-H. Yoon, Langmuir, 2017, 47 H. Nakayama, T. Eguchi, N. Nakamura, S. Yamaguchi,
3
3, 3112–3122.
M. Danjyo and M. Tsuhako, J. Mater. Chem., 1997, 7, 1063–1066.
48 A. MacIntosh, G. Jiang, P. Zamani, Z. Song, A. Riese,
K. Harris, X. Fu, Z. Chen, X. Sun and G. Goward, J. Phys.
Chem. C, 2018, 122, 6593–6601.
4
4
1 X. Wu and L. R. Radovic, Carbon, 2006, 44, 141–151.
2 L.-P. Yuan, Z.-Y. Wu, W.-J. Jiang, T. Tang, S. Niu and
J.-S. Hu, Nano Res., 2020, 13, 1376–1382.
4
3 A. S. Singh, J. H. Advani and A. V. Biradar, Dalton Trans., 49 L. Filiciotto, A. M. Balu, J. C. Van der Waal and R. Luque,
020, 49, 7210–7217. Catal. Today, 2018, 302, 2–15.
4 S. Kumar, M. B. Gawande, J. Kopp, S. Kment, R. S. Varma 50 C. I. Meyer, A. J. Marchi, A. Monzon and T. F. Garetto, Appl.
2
4
and R. Zbořil, ChemSusChem, 2020, 13, 5231–5238.
5 H. Sato, Catal. Rev., 1997, 39, 395–424.
6 T. Mariappan, Y. Zhou, J. Hao and C. A. Wilkie, Eur. Polym.
J., 2013, 49, 3171–3180.
Catal., A, 2009, 367, 122–129.
4
4
51 Y. Tachibana, T. Masuda, M. Funabashi, K.-I. Kasuya and
M. Kunioka, ACS Symp. Ser., 2012, 1105, 91–110.
52 P. Kumar and R. K. Pandey, Green Chem., 2000, 2, 29–32.
This journal is © The Royal Society of Chemistry 2021
Green Chem., 2021, 23, 1370–1381 | 1381