Selective two-step synthesis of 2,5-diformylfuran from monosaccharide, disaccharide, and polysaccharide using H-Beta and octahedral MnO2 molecular sieves

Article abstract of DOI:10.1016/j.mcat.2018.11.001

This study deals with an economical, efficient, and sustainable catalytic process with excellent EcoScore and E-factor values for the selective, two-step synthesis of 2,5-diformylfuran (with a yield higher than 90%) from carbohydrates, especially disaccharide and polysaccharide. This catalytic process involves the abundant and low cost carbohydrate as reactant and economical and sustainable catalysts, H-Beta and octahedral MnO₂ molecular sieve. In this two step approach, H-Beta is used in the first step for the conversion of carbohydrate to 5-hydroxymethylfurfural (HMF). Br?nsted acidity of H-Beta catalyzes the hydrolysis and dehydration reactions whereas Lewis acidity catalyzes the isomerization to produce HMF. Thus, H-Beta is suitable for the conversion of various carbohydrate molecules (starch, sucrose, glucose and fructose) to HMF. After the first step, H-Beta is withdrawn and then the reaction mixture is subjected to oxidation catalyst, octahedral MnO₂ molecular sieve, in the presence of O₂ atmosphere to achieve the high 2,5-diformylfuran (DFF) yield. Molecular oxygen (1 atmosphere, O₂ filled balloon) is used as an eco-friendly and economical oxidant. Furthermore, no over oxidation product of DFF is observed. HMF and DFF are recovered from the reaction mixture by the extraction with methyl isobutyl ketone (MIBK) using water and brine solutions. DFF yields of 97.1%, 95.3%, 93.4% and 91.5% are obtained when the reactions are carried out using fructose, sucrose, glucose, and starch, respectively. Stepwise addition of catalysts improves DFF yield (>90%) from fructose, glucose, sucrose, and especially starch. Moreover, the H-Beta and octahedral MnO₂ molecular sieve are easily separable and recyclable.

Full text of DOI:10.1016/j.mcat.2018.11.001

MolecularCatalysis462(2019)92–103
Contents lists available at ScienceDirect
Molecular Catalysis
journal homepage: www.elsevier.com/locate/mcat
Selective two-step synthesis of 2,5-diformylfuran from monosaccharide,
disaccharide, and polysaccharide using H-Beta and octahedral MnO2
molecular sieves
Bhaskar Sarmah, Rajendra Srivastava⁎
Department of Chemistry, Indian Institute of Technology Ropar, Rupnagar, 140001, India
A R T I C L E I N F O
A B S T R A C T
Keywords:
This study deals with an economical, efficient, and sustainable catalytic process with excellent EcoScore and E-
factor values for the selective, two-step synthesis of 2,5-diformylfuran (with a yield higher than 90%) from
carbohydrates, especially disaccharide and polysaccharide. This catalytic process involves the abundant and low
cost carbohydrate as reactant and economical and sustainable catalysts, H-Beta and octahedral MnO2 molecular
sieve. In this two step approach, H-Beta is used in the first step for the conversion of carbohydrate to 5-hy-
droxymethylfurfural (HMF). Brönsted acidity of H-Beta catalyzes the hydrolysis and dehydration reactions
whereas Lewis acidity catalyzes the isomerization to produce HMF. Thus, H-Beta is suitable for the conversion of
various carbohydrate molecules (starch, sucrose, glucose and fructose) to HMF. After the first step, H-Beta is
withdrawn and then the reaction mixture is subjected to oxidation catalyst, octahedral MnO2 molecular sieve, in
the presence of O2 atmosphere to achieve the high 2,5-diformylfuran (DFF) yield. Molecular oxygen (1 atmo-
sphere, O2 filled balloon) is used as an eco-friendly and economical oxidant. Furthermore, no over oxidation
product of DFF is observed. HMF and DFF are recovered from the reaction mixture by the extraction with methyl
isobutyl ketone (MIBK) using water and brine solutions. DFF yields of 97.1%, 95.3%, 93.4% and 91.5% are
obtained when the reactions are carried out using fructose, sucrose, glucose, and starch, respectively. Stepwise
addition of catalysts improves DFF yield (> 90%) from fructose, glucose, sucrose, and especially starch.
Moreover, the H-Beta and octahedral MnO2 molecular sieve are easily separable and recyclable.
2,5-Diformylfuran
5-Hydroxymethylfurfural
Biomass conversion
H-Beta
Octahedral MnO2 molecular sieve
1. Introduction
Moreover, they are corrosive and non-recyclable. Sustainable produc-
tion of HMF requires recyclable heterogeneous catalysts. Zeolites, silica,
The sole renewable and naturally abundant source of carbon-rich
compounds is biomass [1–3]. Biomass affords a better alternative to our
dependence on crude oil, obtained from fossil fuels, for their applica-
tions in the transportation fuels and commodity chemicals [4–8]. De-
partment of Energy (United States) has given emphasis on a list of most
valuable chemicals produced from carbohydrate biomass useful to
mankind [9]. 5-Hydroxymethylfurfural (HMF) and its oxidized deri-
vatives are amongst the most valuable products. HMF can be prepared
by the acid hydrolysis of monosaccharide, disaccharide, and poly-
saccharide [10–12]. Functional sites present in HMF provide an op-
portunity to synthesize various furan-based chemicals and fuel ad-
ditives just by oxidation, reduction, and condensation reactions
[13,14]. The easiest route to synthesize HMF is the triple dehydration of
fructose. Mineral acids have the ability to convert fructose to HMF [15].
However, they produce side products along with the desired HMF.
metal oxide, carbon-based sustainable catalysts, and ionic liquids have
been developed for the catalytic conversion of monosaccharide, dis-
accharide, and polysaccharide to HMF [16–24]. Recent literature re-
ports several solid acid catalysts including H-Beta and FePW12O40 for
fructose to HMF conversion [25]. These acid catalysts produce 97%
HMF yield when the reaction is carried out by evacuation at 0.97 × 105
Pa. Any pressure other than 1 atmosphere is responsible for inferior
EcoScore value [26] and is, therefore, discouraged by the industries.
Further, the reported study is only limited to the fructose to HMF
conversion. Moreover, such catalysts are amicable for the fructose to
HMF conversion, while only a limited number of reports are available
for the transformation of other carbohydrates such as sucrose, glucose,
or polysaccharide to HMF [27–30]. The difficulties that often arise in
the transformation of carbohydrate to HMF are the simultaneous gen-
eration of formic acid (HMF degraded product), condensation products
⁎ Corresponding author.
E-mail address: rajendra@iitrpr.ac.in (R. Srivastava).
https://doi.org/10.1016/j.mcat.2018.11.001
Received 5 September 2018; Received in revised form 31 October 2018; Accepted 1 November 2018
2468-8231/©2018ElsevierB.V.Allrightsreserved.

B. Sarmah, R. Srivastava
MolecularCatalysis462(2019)92–103
Scheme 1. Various possible products formed during the conversion of carbohydrate to DFF.
(polymer or humins) and levulinic acid etc.
summarized in Table S1. Moreover, these reactions have been carried
out using most reactive carbohydrate fructose. Only a limited efforts
have been made in which difficult carbohydrates such as starch and
glucose have been converted to DFF via tandem methodology. Hence, a
sustainable route for the transformation of difficult carbohydrates to
DFF is a challenging and important research area.
Another possibility is to carry out this reaction in a two-step process
using two different catalysts. In this process, high DFF yield can be
achieved. In this approach, former step is carried out using an efficient
acid catalyst. After the removal of acid catalyst, the latter step is carried
out with an oxidation catalyst in the presence of O2 to achieve high DFF
yield. Using this approach, oxidative decomposition of carbohydrate
can be significantly minimized, and, as a result, the DFF yield can be
maximized. Further, the catalyst regeneration process (if required)
becomes simpler.
Further, the interest lies in the conversion of HMF to 2,5-di-
formylfuran (DFF). During the oxidation of HMF, partial oxidized and
complete oxidized furan derivative are formed in addition to the de-
sired product DFF (Scheme 1). Therefore, selective synthesis of DFF is a
challenging task. Several hazardous catalysts as well as sustainable
catalysts have been explored for this selective oxidation process
[31–35]. A wide range of metal oxides including various structural
forms of manganese oxide have been investigated for this oxidation
reaction [36]. Among different structural forms of manganese oxide,
OMS-2 has produced the best activity with 97% yield of DFF at 5 bar O2
pressure. A penalty of 3 is given to any operating pressure other than
1 atm in the EcoScore calculation [36]. Such a process is, therefore, not
encouraged by industries because a huge investment is required for the
installation of process equipment.
More challenge lies in the synthesis of DFF directly from mono-
saccharide or disaccharide or polysaccharide [37–44]. DFF can be di-
rectly synthesized from carbohydrates with the aid of multi-functional
catalysts. These multi-functional catalysts should contain acidic and
oxidation capabilities. The role of a multi-functional catalyst is to
convert carbohydrate molecules to HMF using acidic sites and, then,
convert HMF to DFF employing its oxidation sites. Two methodologies
can be adopted for the direct conversion of carbohydrate to DFF. In the
first methodology, carbohydrate can be reacted with a multi-functional
catalyst in O2 atmosphere to form DFF. However, using this metho-
dology, an inferior DFF selectivity is expected due to the degradation of
carbohydrate in the presence of multi-functional catalyst. Second
methodology is the tandem process. In this methodology, first carbo-
hydrate can be reacted with multi-functional catalyst in N2 atmosphere
to form HMF and then in the successive step it is reacted in O2 atmo-
sphere to form DFF. In the recent time, a wide range of multi-functional
catalysts based on metal oxides (especially magnetic nanoparticles,
vanadium based systems, mixed metal oxide & heteropolyacid), meso-
porous silica, zeolite, ion-exchange resins, ionic liquids, MOF, poly-
aniline, and carbon (C3N4 and graphene oxide) based systems have
been explored [38–62]. However, these multi-functional catalysts have
also provided somewhat inferior DFF selectivity. Inhibition of glucose
isomerization and oxidative decomposition of carbohydrate over these
catalysts are prime reasons for such inferior DFF selectivity. The details
of some selected catalysts and reasons for inferior DFF selectivity are
This study provides an economical, user-friendly and sustainable
route with excellent EcoScore & E-factor for the selective production of
DFF from polysaccharide starch, disaccharide sucrose and mono-
saccharide (fructose and glucose) at 1 atmosphere O2. Herein, two ro-
bust and economical catalysts have been employed in the two-step
methodology for the selective formation of DFF from carbohydrates. In
the first step, H-Beta is used as an acid catalyst whereas, in the second
step, octahedral MnO2 molecular sieve (OMS) is used as an oxidation
catalyst. H-Beta possesses Lewis acidity and Brönsted acidity.
Therefore, hydrolysis, dehydration, and isomerization can be accom-
plished using H-Beta and thus can be applicable to a wide range of
carbohydrates. We have reported that OMS prepared using urea as an
additive has exhibited better oxidation ability [63] than OMS-2 pre-
pared with conventional method reported earlier for the conversion of
HMF to DFF at 5 bar O2 pressure [36]. Therefore, we anticipated that
OMS prepared with urea can provide better oxidation ability and
transform HMF to DFF at 1 atmosphere O2. The catalytic process de-
monstrated here offered greater than 90% DFF yield from various
carbohydrates mentioned above at 1 atmosphere O2 (oxygen filled
balloon) with higher EcoScore & lower E-factor values, hitherto re-
ported in the literature.
93

B. Sarmah, R. Srivastava
MolecularCatalysis462(2019)92–103
Fig. 1. SEM images of (a) OMS and (b) H-Beta and HRTEM images of (c,d) H-Beta, and (e,f) OMS.
2. Experimental
2.1. Catalyst synthesis
H-Beta and H–Y (Si/Al = 2.7) were obtained from Süd-Chemie
India Pvt. Ltd. Urea assisted synthesis route was adopted to prepare
OMS [63]. KMnO4 (8 mmol) was added to 20 ml of deionized water.
Urea (28 mmol) was added to the above solution, followed by stirring at
ambient temperature for 30 min. The resultant solution was added
dropwise to a vigorously stirred mixture of MnSO4.H2O (10.5 mmol in
6 ml water) and 0.6 ml conc. HNO3. The resulting mixture was trans-
ferred to a round-bottomed flask fitted with a condenser and refluxed
for 24 h. Synthesis mixture was cooled to ambient temperature, the
solid product was filtered, washed with deionized water, and dried at
393 K for 10 h to obtain OMS with 95% yield. Conventional ZSM-5 and
nanocrystalline ZSM-5 were prepared by following the reported pro-
cedure [64,65]. Details of textural properties of conventional ZSM-5
and nanocrystalline ZSM-5 are provided in Table S2.
2.2. Procedure of the catalytic reactions
Fig. 2. High resolution XPS spectrum of Mn 2p present in OMS.
Procedures for the independent first step using H-Beta catalyst
94

B. Sarmah, R. Srivastava
MolecularCatalysis462(2019)92–103
Fig. 3. Influence of (a) temperature, (b) time, (c) catalyst amount, and (d) solvent in the transformation of sucrose to HMF. Reaction condition: For (a) sucrose
(1 mmol) and DMSO (5 mL), H-Beta (120 mg), time (3 h), and N2 flow rate (10 mL/min); (b) sucrose (1 mmol) and DMSO (5 mL), H-Beta (120 mg), temp. (393 K) and
N
2 flow rate (10 mL/min); (c) sucrose (1 mmol) and DMSO (5 mL), H-Beta (catalyst), temp. (393 K), time (3 h), and N2 flow rate (10 mL/min); (d) sucrose (1 mmol)
and solvent (5 mL), H-Beta (catalyst), temp. (393 K), time (3 h), and N2 flow rate (10 mL/min).
(HMF synthesis) & second step using OMS catalyst (DFF synthesis), and
two-step catalytic process for the synthesis of DFF (from carbohy-
drates), and NMR data for HMF and DFF are provided in SI. Details of
isolation of HMF and DFF from the reaction mixture using methyl
isobutyl ketone (MIBK) are provided in SI. Analytic methods utilized for
the determination of yield using 1H NMR [66,67] and gas chromato-
graphy are also described in SI.
autoclave (23 mL) at 423 K for 12 h. Progress of the reaction was
monitored using 1H NMR spectroscopy. H-Beta catalyst was withdrawn
from the reaction mixture using centrifuge machine after the reaction
vessel was cooled to ambient temperature. Then the reaction mixture
was transferred to a double-neck round-bottom flask fitted with a
condenser, and the oxidation step was carried out using OMS (100 mg).
The resulting mixture was heated at 393 K for 10 h under the O2 en-
vironment (O2 balloon). The OMS catalyst was withdrawn from the
reaction mixture using centrifuge machine, and the reaction mixture
was analyzed by 1H NMR. DFF with an isolated yield of 91.5% was
obtained using MIBK as extracting medium.
2.2.1. Two-step direct conversion of polysaccharide (starch) to DFF
Starch (340 mg), DMSO (5 mL), and H-Beta (150 mg) (pre-activated
at 393 K for 4 h) were charged into a Teflon-lined stainless steel
95

B. Sarmah, R. Srivastava
MolecularCatalysis462(2019)92–103
Fig. 4. The plot for lnk vs 1/T for the
determination of Ea for the (a) dehy-
dration of sucrose to HMF and (b) oxi-
dation HMF to DFF. Reaction condi-
tion: (a) Sucrose (1 mmol) and DMSO
(5 mL), H-Beta (120 mg), time (3 h),
and N2 flow rate (10 mL/min); (b) HMF
(1 mmol) and DMSO (5 mL), OMS
(60 mg), time (10 h), and O2 (1 atm,
balloon). Plots were constructed by
taking the average values from Figs. 3
and 5.
3. Results and discussion
nanocrystalline morphology with the particle sizes of 20–25 nm in the
TEM images (Fig. 1c and d). Whereas the HRTEM images of OMS
confirm the nanowire morphology (Fig. 1e and f). The nanowire
thickness is in the range of 10–20 nm (Fig. 1f).
3.1. Physico-chemical characterization
Only a brief description of the physicochemical properties of the
investigated materials is provided. The diffraction peaks and corre-
sponding planes obtained from the powder XRD of H-Beta shown in Fig.
S1 match well with the diffraction peaks of BEA framework structure
[68]. OMS exhibits highly crystalline XRD pattern. The diffraction
peaks and corresponding planes obtained from the powder XRD of OMS
shown in Fig. S1 match well with the diffraction peaks of octahedral
MnO2 (tetragonal structure of cryptomelane-type octahedral molecular
sieves) (ICDD No. 00-29-1020).
The presence of Lewis and Brönsted acidity is well documented in
the literature for the zeolite Beta [69]. Lewis and Brönsted acid sites
present in the H-Beta were evaluated from FT-IR measurement using
pyridine as a probe molecule. Fig. S3a shows the pyridine FT-IR spec-
trum of H-Beta zeolite. The peak at 1545 cm−1 corresponds to Brönsted
acid site (Si–O(H)–Al groups), and the peak at 1445 cm−1 corresponds
to Lewis acid site (Al3+). Furthermore, the peak at 1490 cm−1 can be
attributed to the interaction between pyridine and both Brönsted and
Lewis acidic sites. Acidity of H-Beta was determined from ammonia-
temperature programmed desorption (NH3-TPD) measurement. Deso-
rption peak at higher temperature represents stronger acid sites. H-Beta
demonstrates three types of acid sites (Fig. S3b). First desorption in
323 K–553 K range corresponds to weak acid sites, second desorption in
553 K–643 K range corresponds to medium strength acid sites, and third
desorption in 643 K–823 K range corresponds to strong acid sites. In this
material weak acid sites are predominant over strong acid sites. Total
acidity of H-Beta is found to be 0.72 mmol/g.
Oxidation state of Mn present in OMS was confirmed from the XPS
measurement. The high resolution Mn 2p XPS spectrum is shown in
Fig. 2. Gaussian–Lorentzian curve fitting of XPS spectrum shows the
presence of Mn 2p3/2 and Mn 2p1/2. Mn 2p3/2 can be deconvoluted into
two peaks located at 641.5 and 642.8 eV, which can be assigned to Mn
(III) and Mn(IV) oxidation states [70]. Similarly, Mn 2p1/2 can also be
deconvoluted into two peaks located at 653.1 and 654.2 eV, which can
be assigned to Mn(III) and Mn(IV) oxidation states, respectively.
Surface area and porosity of H-Beta and OMS were calculated using
N2-volumetric analyzer (Fig. S2, SI). NLDFT and BJH methods were
used to determine the micropore and mesopore size distribution (inter-
crystalline porosity), respectively (Fig. S2, SI). H-Beta exhibits type-II
isotherm and shows higher adsorption at lower P/P0 that correlates
with the micropores filling in the material. In the intermediate pressure
range (P/P0 = 0.2–0.8), a moderate growth in the adsorbed volume is
perceived. This is followed by swift increase in the adsorption volume
at P/P0 higher than 0.9. This swift increase in the adsorbed volume can
be correlated to the inter-particle porosity present in the material which
is consistent with BJH pore size distribution in the range of 4–10 nm
obtained for this material. Similarly, OMS also exhibits type-II isotherm
and a moderate growth in the adsorbed volume is perceived in the
intermediate pressure range (P/P0 = 0.2-0.85). This is followed by the
vertiginous increase in the adsorbed volume at P/P0 > 0.85 (Fig. S2,
SI). Micropores (approximately, 0.5 nm) is noticed during the pore size
determination. Further, bimodal mesopore size distribution (first:
2–10 nm (mesopores) and second: 10–120 nm (mesoporous-macro-
porous overlapping domain) with a peak maximum of 33 nm) is noticed
in BJH analysis. Obtained numerical values of surface area and porosity
derived from N2-adsoption studies are summarized in Table S2 (SI).
Scanning electron microscope (SEM) provides information with re-
spect to microstructure and morphology of the synthesized materials.
OMS shows highly cross-linked nanowire morphology (Fig. 1a). H-Beta
shows aggregated crystal morphology (Fig. 1b). The high-resolution
transmission electron microscope (HRTEM) provides information with
respect to nanostructure. H-Beta exhibits highly aggregated
3.2. Catalytic activity
A two-step methodology is reported here for the synthesis of DFF
from biomass-derived carbohydrates (monosaccharide, disaccharide,
and polysaccharide) via HMF as an intermediate. In this methodology,
two catalysts were utilized in a step-wise manner to achieve DFF. The
first step was realized with a solid acid catalyst H-Beta and the second
step was accomplished with a solid oxidation catalyst OMS. To achieve
the best DFF yield, it is important to optimize both the steps, in-
dependently, involving H-Beta and OMS as catalysts.
96

B. Sarmah, R. Srivastava
MolecularCatalysis462(2019)92–103
Scheme 2. Plausible mechanism for the two-step synthesis of DFF from sucrose involving sucrose to HMF using H-Beta catalyst, followed by HMF oxidation using
OMS catalyst.
3.2.1. Optimization of the first step (HMF synthesis)
103.4 kJ/mol for the dehydration of fructose has been reported using
3 M H2SO4 and ion-exchanged resin as catalysts, respectively [71,72].
However, the present study using H-Beta, Ea of 67.6 kJ/mol was cal-
culated for difficult substrate sucrose when compared to fructose re-
ported above, signifying the better activity of H-Beta at the optimum
condition. During the initial period, the reaction rate was high (Yield of
HMF after 1 h = 51%; product selectivity: HMF (52%), fructose (22%),
and glucose (26%)) (Fig. S4, SI). Almost quantitative yield of HMF was
obtained after 3 h of reaction (Fig. 3b). 1H NMR recorded at 0 h, 1 h,
2 h, and 3 h for the reaction using sucrose as substrate is shown in Fig.
S4. Influence of catalyst amount was investigated for the reactions
carried out for 3 h (Fig. 3c). Based on the results obtained, an optimum
amount of 120 mg H-Beta was optimized to get the maximum yield of
HMF. The influence of solvents having different polarity was in-
vestigated in the dehydration of sucrose to HMF (Fig. 3d). Result shows
that aprotic polar solvent DMSO was excellent to achieve the highest
The first step was optimized using sucrose as a model substrate. The
reaction was carried out in N2 atmosphere (10 mL/min) with H-Beta
catalyst in DMSO solvent medium. 1H NMR was used as an analytical
tool to determine the yield of the product in the optimization reactions.
Parameters influencing the dehydration of sucrose to HMF were in-
vestigated. Temperature played significant role to enhance the product
(HMF) yield. The influence of temperature was investigated in
363 K–403 K (Fig. 3a). With an increase in the temperature from 363 K
to 393 K, the HMF yield was increased from 45% to 99% (MIBK ex-
tracted yield 95.4%). Considering the highest yield at 393 K, this tem-
perature was optimized for rest of the studies. Rate of reaction, calcu-
lated at different temperature, was used to determine the activation
energy (Ea) (Fig. 4a). Ea (67.6 kJ/mol) was calculated from lnK vs 1/T
plot for this reaction using H-Beta catalyst. Catalytic path having lower
activation energy (Ea) is suitable for a reaction. Ea of 99 kJ/mol and
97

B. Sarmah, R. Srivastava
MolecularCatalysis462(2019)92–103
Fig. 5. Influence of (a) temperature, (b) catalyst amount, and (c) reaction time, in the transformation of HMF to DFF. Reaction condition: for (a) HMF (1 mmol),
DMSO (5 mL), OMS (60 mg), time (10 h), and O2 (1 atm, balloon); (b) HMF (1 mmol), DMSO (5 mL), time (10 h), temperature (393 K), and O2 (1 atm, balloon); and
(c) HMF (1 mmol), DMSO (5 mL), temperature (393 K), OMS (60 mg), and O2 (1 atm, balloon).
yield of HMF. DMSO plays an important role in the dehydration of
carbohydrates (discussed below in Scheme 2). In a strong polar aprotic
solvent, especially DMSO, the electronic pair of O, N, or S can attract
and disperse the positive charge of DMSO, and accelerates the dehy-
dration of carbohydrate to HMF. DMSO is known to form sulfuric and
methanesulfonic acids at high temperature which are suitable for the
dehydration of carbohydrate (fructose) [73,74]. Moreover, DMSO has
the ability to suppress the formation of undesired side product such as
humus due to the degradation of HMF [75–77]. In another study,
Caroline Laugel et al. [73] reported that HMF could degrade in organic
solvents other than DMSO. Furthermore, a wide range of carbohydrates
are soluble in DMSO when compared to other solvents makes DMSO an
excellent solvent for this study. After the successful transformation of
sucrose to HMF, fructose and glucose were converted to HMF.
An excellent HMF yield (MIBK extracted yield 96.2%) was obtained
in the absence of catalyst with fructose as a reactant in just 1 h. 1H NMR
spectra recorded at 0 h, 0.5 h, and 1 h for the dehydration of fructose
using H-Beta catalyst is shown in Fig. S5. Lone pairs of oxygen present
in DMSO helps in the step-wise dehydration to form HMF (Scheme 2).
Result shows that glucose required more time for the conversion to
HMF using H-Beta catalyst. HMF and fructose are observed in the 1H
NMR spectrum recorded after 2 h in addition to the un-reacted glucose
(Fig. S6, SI). In this case, more than 98% HMF (MIBK extracted yield
93.2%) was observed in the time-dependent 1H NMR spectrum ob-
tained after 5 h (Fig. S6, SI).
The optimum condition required for monosaccharide and
98

B. Sarmah, R. Srivastava
MolecularCatalysis462(2019)92–103
Fig. 6. Time-dependent 1H NMR spectra recorded during two-step conversion of sucrose to DFF.
disaccharide was not suitable for the conversion of polysaccharide
starch to HMF. Therefore, the reaction was carried out at 423 K for 12 h
using H-Beta catalyst (150 mg) to obtain the quantitative yield of HMF.
1H NMR spectra recorded at 0 h, 5 h, and 12 h of the reaction are pre-
sented in Fig. S7 (SI). After 5 h, 1H NMR spectrum shows 99% starch
conversion to its hydrolyzed products: HMF (50%), glucose (24%), and
fructose (26%). Whereas, after 12 h, no by-product was observed and
98% yield (MIBK extracted yield 92.1%) of HMF was obtained (Fig. S7,
SI).
ZSM-5, H–Y, nanocrystalline ZSM-5, and OMS were investigated for
the sucrose to HMF transformation (Table S3, SI). Conventional ZSM-5
and OMS were found to be inactive whereas H–Y and nanocrystalline
ZSM-5 were active but required longer time to produce lower yield of
HMF. Since OMS contains only Fe and V sites, therefore OMS was in-
active during the conversion of sucrose to HMF. Acid sites are located in
the micropores of conventional ZSM-5 with 0.55 nm pore size therefore
conventional ZSM-5 was found to be inactive to catalyze this reaction.
Nanocrystalline ZSM-5 has large external surface area and only the acid
sites present on the pore mouth are accessible to sucrose molecules
therefore nanocrystalline ZSM-5 exhibited inferior activity. H–Y zeolite
with lower Si/Al ratio and inferior Brönsted acidity produced lower
yield of HMF. In this case, an additional side product, dimer of HMF
[5,5-oxy-(bismethylene)-2-furaldehyde] was obtained. H-Beta with
pore size of 0.65 nm and significant defect sites makes H-Beta suitable
for this reaction. Products selectivity obtained using H-Beta at different
time interval during the conversion of sucrose to HMF is provided in
Table S3 (SI).
3.2.2. Selective oxidation of HMF to DFF
Since the aim was to develop two-step reaction protocol, therefore
one has to optimize the oxidation of HMF to DFF in the same solvent as
it was optimized in the first step (carbohydrate to HMF). Considering
the high yield of HMF in DMSO, reactions were carried out in DMSO
solvent using OMS as a catalyst. In this step, the product yield was
easily determined using GC. Reaction parameters were optimized using
oxygen filled balloon (at 1 atm). HMF oxidation was carried out in
DMSO under the optimized reaction condition without any catalyst.
Only 1.9% DFF yield was obtained as product in the absence of catalyst.
Only a negligible amount of DFF (2.1% DFF) was formed when H-Beta
was used as a catalyst. These control experiments suggest that DMSO
did assist in the oxidation reaction but without suitable catalyst its
activity was negligible. With increase in the temperature in the range of
353 K – 393 K using OMS catalyst, the DFF yield (or HMF conversion)
was increased from 40% to 99% (since DFF was observed as the only
product) (Fig. 5a). From the plot of lnk vs 1/T, Ea was determined
(Fig. 4b). The Ea for this reaction using OMS was determined to be
61.5 kJ/mol (Reaction condition: HMF (1 mmol), OMS (60 mg), DMSO
(5 mL), time (10 h), and temperature (353–393 K). Furthermore, the
activation energy calculated for HMF to DFF conversion is lower
(61.5 kJ/mol) using OMS catalyst when compared to other catalysts
(VOx/TiO2 (67 kJ/mol) [78] and multilayered V2O5/TiO2 (77 kJ/mol)
[79]) suggesting the superiority of the present catalyst at the optimum
reaction condition. With an increase in the catalyst amount, from 20 mg
to 60 mg, the DFF yield was increased and the highest DFF yield was
obtained using 60 mg of catalyst (Fig. 5b). With the increase in reaction
time from 2 h to 10 h, DFF yield was increased gradually from 29% to
99

B. Sarmah, R. Srivastava
MolecularCatalysis462(2019)92–103
Fig. 7. Time-dependent 1H NMR spectra recorded during two-step conversion of fructose to DFF.
99% (Fig. 5c). Therefore, the optimum condition for the oxidation of
HMF to DFF was chosen as: Reaction time – 10 h; temperature – 393 K;
and the catalyst amount – 60 mg.
The aim of this study was to maximize the yield in both steps so that
the separation can be avoided, especially after the first step of the
process (carbohydrate to HMF). Further, any lower yield has penalty
points during the determination of EcoScore of the process [26].
Therefore, the reaction parameters were optimized to maximize the
HMF and DFF yields.
isolated from the reaction mixture by extracting with MIBK using water
and brine solutions and DFF was isolated with 95.3% yield. The Eco-
Score and E-factor were calculated to be 85 and 0, respectively (SI).
The bi-functional catalyst, OMS(30%)/H-Beta, was also prepared by
following the reported procedure [80]. This bi-functional catalyst was
investigated in the two-step reaction. Sucrose (1 mmol) was dissolved in
DMSO (5 mL) and then heated to 393 K in the presence of OMS (30%)/
H-Beta (120 mg). The first reaction was performed for 3 h in N2 flow
(10 mL/min) and then next step was carried out in the presence of O2
balloon. In this two-step reaction, sucrose conversion (72%) and DFF
yield (37%) were obtained.
3.3. Two-step conversion of carbohydrate to DFF
The mechanism for the formation of HMF from sucrose is presented
in Scheme 2. The first step of this conversion process involves hydro-
lysis of disaccharide sucrose to monosaccharide, fructose and glucose.
Small amount (0.5 wt %) of water present in the DMSO facilitates this
step. Mild Brönsted acidity accelerates the hydrolysis step by the clea-
vage of glycosidic bond present in the sucrose to form equimolar
fructose and cyclic oxonium ion (Scheme 2). Then water reacts with the
cyclic oxonium ion and forms the glucose. Later Lewis acid sites present
in H-Beta facilitate the isomerization of glucose to fructose. Brönsted
acid sites present in H-Beta protonate the hydroxyl group (most basic)
attached to the anomeric carbon of the fructose and facilitate the de-
hydration process via enol-keto tautomerization to form HMF. The
oxidation step is favored in the presence of OMS. Initially, the lattice
oxygen present in OMS oxidizes HMF to DFF and simultaneously Mn4+
reduces to Mn3+ [36]. Then the refilling of oxygen vacancies (i.e.
consumed lattice oxygen atom) by molecular O2 and re-oxidation of
Mn3+ to Mn4+ [36] takes place.
Having optimized the reaction condition independently for the first
and second steps, the stage was set to perform reactions for the trans-
formation of carbohydrates to DFF in the two-step process.
In the two-step process, sucrose (1 mmol) was dissolved in DMSO
(5 mL) and then heated at 393 K under N2 flow (10 mL/min) in the
presence of H-Beta (120 mg). The transformation of sucrose to HMF was
evaluated using 1H NMR (Fig. 6). After achieved the maximum HMF
yield in 3 h, the N2-flow was stopped, and the reaction vessel was
cooled to room temperature. The catalyst was withdrawn from the re-
action mixture with the help of centrifuge machine. Then oxidation
catalyst OMS (60 mg) was added and the reaction mixture was heated
to 393 K. The reaction was continued for another 10 h using O2 filled
balloon (1 atm). After 13 h of the reaction, the catalyst was withdrawn
and reaction mixture was analyzed using 1H NMR. Time-dependent 1H
NMR spectra for the two-step transformation of sucrose to DFF using H-
Beta and OMS catalysts are shown in Fig. 6. After 13 h of the reaction,
100% sucrose conversion and 99% DFF yield were obtained. DFF was
The catalysts were easily withdrawn and reused for multiple cycles.
100

B. Sarmah, R. Srivastava
MolecularCatalysis462(2019)92–103
Fig. 8. 1H NMR spectra recorded during two-step conversion of glucose to DFF at different time.
H-Beta and OMS were recovered from the reaction mixture using cen-
trifuge machine after each step. The catalysts were washed with DMSO
and deionized water, and, then, dried in an oven at 373 K for 10 h. H-
Beta was activated at 773 K for 4 h in a furnace whereas OMS was ac-
tivated at 473 K in an oven for 10 h before next cycle. Catalytic activity
data obtained after each cycle confirmed that no significant decrease in
the DFF yield was observed after five cycles (Fig. S8, SI). XRD (Fig. S9,
SI) and N2-adsorption investigations (Table S2, SI) of the recycled
catalysts confirm that the catalyst was stable after recycle. Mn, Al, and
Si species were not detected in the reaction mixture which confirmed
that these elements were not leached into the reaction mixture, and H-
Beta & OMS catalyzed the reaction as heterogeneous catalysts to pro-
duce DFF. Recyclability study of H-Beta zeolite was carried out at lower
conversion. Therefore, H-Beta was recycled for three times after per-
forming the reaction for 0.5 h for the conversion of sucrose to HMF.
After each cycle, H-Beta was withdrawn from the reaction mixture
using centrifuge machine. The catalyst was washed with DMSO and
deionized water, and, then, dried in an oven at 373 K for 10 h and ac-
tivated at 773 K for 4 h before next cycle. Table S3 (SI) shows that H-
Beta was successfully recycled with no considerable loss in the activity
after three recycles.
water and brine solutions. The EcoScore and E-factor were calculated to
be 86 and 0, respectively (SI).
Next, two-step transformation of glucose to DFF was carried out. In
the case of glucose, after 2 h, only 60% glucose conversion and 70%
HMF yield were obtained. 1H NMR spectrum shows that after 5 h, 99%
glucose conversion with quantitative yield for HMF was obtained
(Fig. 8). After 5 h, H-Beta was removed and the next step was carried
out in the presence of OMS (60 mg) in O2 balloon for 10 h. 1H NMR
(Fig. 8) shows the quantitative yield for DFF (98%) after 15 h of the
reaction. DFF was recovered from the reaction mixture by extracting
with MIBK using water and brine solutions and DFF was isolated with
93.4% yield. The EcoScore and E-factor were calculated to be 84 and
0.016, respectively (SI).
Finally, the reaction with polysaccharide starch was carried out. For
starch, the first step was carried out at 423 K for 12 h using H-Beta
catalyst (150 mg). 1H NMR spectra recorded after 5 h and 12 h are
shown in Fig. 9. After 12 h of reaction, a quantitative yield of HMF was
obtained (Fig. 9). Subsequently, the second step was performed using
OMS catalyst (100 mg) for 10 h using an O2 balloon. After, 17 h of the
reaction, approximately 50% DFF yield was obtained. 1H NMR in-
vestigation shows the 98% yield of DFF was obtained after 22 h of total
reaction time (Fig. 9). DFF was recovered from the reaction mixture by
extracting with MIBK using water and brine solutions and DFF was
isolated with 91.5% yield. The EcoScore and E-factor were calculated to
be 83 and 0.016, respectively (SI).
Similarly, two-step transformation of fructose to DFF was accom-
plished. In this case, the first step was carried out in the absence of H-
Beta for 1 h to achieve the quantitative yield of HMF (Fig. 7). The
second step was carried out in the presence of OMS (60 mg) for 10 h in
O2 (balloon). Time-dependent 1H NMR study shows the 100% DFF yield
after 11 h of the reaction in two-step process (Fig. 7). DFF (yield 97.1%)
was recovered from the reaction mixture by extracting with MIBK using
The comparative catalytic efficiency of this catalyst system (H-Beta
& OMS) in the carbohydrates to DFF conversion with other reported
catalysts are summarized in Table S4-S7 (SI). Literature shows that only
101

B. Sarmah, R. Srivastava
MolecularCatalysis462(2019)92–103
Fig. 9. Time-dependent 1H NMR spectra recorded during two-step conversion of starch to DFF.
a limited success has been obtained in the transformation of glucose to
DFF [37–43]. Reported heterogeneous catalysts could able to produce a
maximum of 51% DFF yield from glucose (Table S5, SI). However, the
present catalytic process is able to afford 93.4% DFF yield from glucose.
Literature also shows that only a meager success has been made using
sucrose and starch as reactants. Only a few reports are available for the
conversion of starch to DFF and the reported catalysts have afforded
DFF yield < 40% (Table S6, SI) [39,42]. However, the present catalytic
process is able to afford 95.3% and 91.5% yields of DFF from sucrose
and starch, respectively, under mild reaction condition. EcoScore and E-
factor values are superior to various reported catalysts (Table S8-S11).
transformation of carbohydrates (fructose, glucose, sucrose, and espe-
cially starch) to DFF was investigated using O2 (1 atm, O2 filled bal-
loon). After the two-step process, DFF was recovered by the solvent
extraction process and the DFF yields of 97.1%, 95.3%, 93.4% and
91.5% were obtained for the reactions carried out using fructose, su-
crose, glucose, and starch. The Beta and OMS catalysts demonstrated no
loss in the activity in the sucrose to DFF conversion even after five
cycles. Isolated yield, EcoScore, and E-factors were better than hitherto
reported catalysts in literature. Such an eco-friendly two-step trans-
formation of carbohydrates to DFF with economical reusable hetero-
geneous catalysts will be suitable for commercial execution.
Acknowledgements
4. Conclusion
Authors are grateful to Directo,r IIT, Ropar, India for financial as-
sistance (9-35/2009/IITRPR/43 dated 09/01/2018). BS thanks UGC,
New Delhi, for SRF fellowship. We are grateful to SAIF, IIT Bombay for
TEM analysis.
In summary, a facile, economical, and sustainable two-step trans-
formation of carbohydrates (including monosaccharide, disaccharide,
and polysaccharide) to DFF was demonstrated under mild reaction
condition. Each step was independently optimized to maximize the
carbohydrate conversion and to achieve the excellent yield of DFF.
Moreover, the activation energy for each step, sucrose to HMF over H-
Beta catalyst was determined to be 67.5 kJ/mol and HMF to DFF over
OMS catalyst was determined to be 61.5 kJ/mol. The activation en-
ergies were lower than various reported catalysts demonstrating the
effectiveness of the present catalytic process. Brönsted acid sites and
Lewis acid sites (Al-centre) present in H-Beta facilitated the hydrolysis,
isomerization, and dehydration steps. Furthermore, the oxidation cat-
alyst, OMS which contains the redox Mn4+/Mn3+ sites facilitated the
oxidation of the HMF to DFF. Having received the maximum yields of
HMF from carbohydrates and DFF from HMF oxidation, a two-step
References
[1] G. Koçar, N. Civaş, Renew. Sustain. Energy Rev. 28 (2013) 900–916.
[2] H. Kobayashi, H. Kaiki, A. Shrotri, K. Techikawara, A. Fukuoka, Chem. Sci. 7 (2016)
692–696.
[3] C. Li, X. Zhao, A. Wang, G.W. Huber, T. Zhang, Chem. Rev. 115 (2015)
11559–11624.
[4] G.W. Huber, S. Iborra, A. Corma, Chem. Rev. 106 (2006) 4044–4098.
[5] A.J. Ragauskas, C.K. Williams, B.H. Davison, G. Britovsek, J. Cairney, C.A. Eckert,
W.J. Frederick, J.P. Hallett, D.J. Leak, C.L. Liotta, J.R. Mielenz, R. Murphy,
R. Templer, T. Tschaplinski, Science 311 (2006) 484–489.
[6] P. Zhou, Z. Zhang, Catal. Sci. Technol. 6 (2016) 3694–3712.
102

B. Sarmah, R. Srivastava
MolecularCatalysis462(2019)92–103
[7] P. Gallezot, Chem. Soc. Rev. 41 (2012) 1538–1558.
[8] Y.-T. Cheng, J. Jae, J. Shi, W. Fan, G.W. Huber, Angew. Chem. Int. Ed. 124 (2012)
1416–1419.
[41] Z.-Z. Yang, J. Deng, T. Pan, Q.-X. Guo, Y. Fu, Green Chem. 14 (2012) 2986–2989.
[42] W. Zhang, W. Hou, T. Meng, W. Zhuang, J. Xie, Y. Zhou, J. Wang, Catal. Sci.
Technol. 7 (2017) 6050–6058.
[9] T. Werpy, G. Petersen, A. Aden, J. Bozell, J. Holladay, J. White, A. Manheim,
D. Eliot, L. Lasure, S. Jones, Top Value Added Chemicals From Biomass. Volume 1 –
Results of Screening for Potential Candidates From Sugars and Synthesis Gas, U.S.
Department of Energy Report; Pacific Northwest National Laboratory/U.S.
Department of Energy, Oak Ridge, TN, 2004, pp. 1–77.
[43] B. Sarmah, R. Srivastava, Catal. Sci. Technol. 8 (2018) 2870–2882.
[44] Z. Zhang, G.W. Huber, Chem. Soc. Rev. 47 (2018) 1351–1390.
[45] G.A. Halliday, R.J. Young, V.V. Grushin, Org. Lett. 5 (2003) 2003–2005.
[46] Z. Zhang, Z. Yuan, D. Tang, Y. Ren, K. Lv, B. Liu, ChemSusChem 7 (2014)
3496–3504.
[10] J.B. Binder, R.T. Raines, J. Am. Chem. Soc. 131 (2009) 1979–1985.
[11] G. Yong, Y. Zhang, J.Y. Ying, Angew. Chem., Int. Ed. 47 (2008) 9345–9348.
[12] H. Zhao, J.E. Holladay, H. Brown, Z.C. Zhang, Science 316 (2007) 1597–1600.
[13] A. Corma, S. Iborra, A. Velty, Chem. Rev. 107 (2007) 2411–2502.
[14] J.N. Chheda, G.W. Huber, J.A. Dumesic, Angew. Chem. Int. Ed. 46 (2007)
7164–7183.
[15] F.S. Asghari, H. Yoshida, Ind. Eng. Chem. Res. 45 (2006) 2163–2173.
[16] E. Nikolla, Y. Román-Leshkov, M. Moliner, M.E. Davis, ACS Catal. 1 (2011)
408–410.
[47] F. Xu, Z. Zhang, ChemCatChem 7 (2015) 1470–1477.
[48] R. Liu, J. Chen, L. Chen, Y. Guo, J. Zhong, ChemPlusChem 79 (2014) 1448–1454.
[49] J. Chen, Y. Guo, J. Chen, L. Song, L. Chen, ChemCatChem 6 (2014) 3174–3181.
[50] G. Lv, H. Wang, Y. Yang, T. Deng, C. Chen, Y. Zhud, X. Hou, Green Chem. 18 (2016)
2302–2307.
[51] J. Zhao, A. Jayakumar, J.-M. Lee, ACS Sustain. Chem. Eng. 6 (2018) 2976–2982.
[52] J. Zhao, A. Jayakumar, Z.-T. Hu, Y. Yan, Y. Yang, J.-M. Lee, ACS Sustain. Chem.
Eng. 6 (2018) 284–291.
[53] R. Fang, R. Luque, Y. Li, Green Chem. 19 (2017) 647–655.
[54] M. Cui, R. Huang, W. Qi, R. Su, Z. He, RSC Adv. 7 (2017) 7560–7566.
[55] N. Mittal, G.M. Nisola, L.B. Malihan, J.G. Seo, H. Kim, S.-P. Lee, W.-J. Chung, RSC
Adv. 6 (2016) 25678–25688.
[56] J. Zhao, J. Anjali, Y. Yan, J.-M. Lee, ChemCatChem 9 (2017) 1187–1191.
[57] F. Wang, L. Jiang, J. Wang, Z. Zhang, Energy Fuels 30 (2016) 5885–5892.
[58] W. Zhang, T. Meng, J. Tang, W. Zhuang, Y. Zhou, J. Wang, ACS Sustain. Chem. Eng.
5 (2017) 10029–10037.
[17] W.-H. Peng, Y.-Y. Lee, C. Wu, K.C.-W. Wu, J. Mater. Chem. 22 (2012)
23181–23185.
[18] R. Noma, K. Nakajima, K. Kamata, M. Kitano, S. Hayashi, M. Hara, J. Phys. Chem. C
119 (2015) 17117–17125.
[19] X. Qi, H. Guo, L. Li, S.R.L. Jr, ChemSusChem 5 (2012) 2215–2220.
[20] L. Cottier, G. Descotes, J. Lewkowski, R. Skowroñski, E. Violett, J. Heterocyclic
Chem. 32 (1995) 927–929.
[21] N. Candu, M.E. Fergani, M. Verziu, B. Cojocaru, B. Jurca, N. Apostol, C. Teodorescu,
V.I. Parvulescu, S.M. Coman, Catal. Today (2018), https://doi.org/10.1016/j.
cattod.2018.08.004.
[59] R. Fang, R. Luque, Y. Li, Green Chem. 18 (2016) 3152–3157.
[60] Z. Yang, W. Qi, R. Su, Z. He, Energy Fuels 31 (2017) 533–541.
[61] Z. Yang, W. Qi, R. Su, Z. He, ACS Sustain. Chem. Eng. 5 (2017) 4179–4187.
[62] W. Hou, Q. Wang, Z. Guo, J. Li, Y. Zhou, J. Wang, Catal. Sci. Technol. 7 (2017)
1006–1016.
[22] H. Li, Q. Zhang, X. Liu, F. Chang, Y. Zhang, W. Xue, S. Yang, Bioresour. Technol.
144 (2013) 21–27.
[23] A.H. Jadhav, H. Kim, I.T. Hwang, Bioresour. Technol. 132 (2013) 342–350.
[24] Y. Yang, X. Xiang, D. Tong, C. Hu, M.M. Abu-Omar, Bioresour. Technol. 116 (2012)
302–306.
[25] K.-I. Shimizu, R. Uozumi, A. Satsuma, Catal. Commun. 10 (2009) 1849–1853.
[26] K.V. Aken, L. Strekowski, L. Patiny, Beilstein J. Org. Chem. 2 (2006) 3–10.
[27] C. Wang, L. Fu, X. Tong, Q. Yang, W. Zhang, Carbohydr. Res. 347 (2012) 182–185.
[28] C. Yue, G. Li, E.A. Pidko, J.J. Wiesfeld, M. Rigutto, E.J.M. Hensen, ChemSusChem 9
(2016) 1–10.
[63] B. Sarmah, R. Srivastava, P. Manjunathan, G.V. Shanbhag, ACS Sustain. Chem. Eng
3 (2015) 2933–2943.
[64] R. Srivastava, N. Iwasa, S.I. Fujita, M. Arai, Chem. Eur. J. 14 (2008) 9507–9511.
[65] B. Sarmah, R. Srivastava, Asian J. Org. Chem. 6 (2017) 873–889.
[66] A.S. Amarasekara, L.T.D. Williams, C.C. Ebede, Carbohydr. Res. 343 (2008)
3021–3024.
[67] A.A. Rosatella, L.C. Branco, C.A.M. Afonso, Green Chem. 11 (2009) 1406–1413.
[68] J.M. Newsam, M.M.J. Treacy, W.T. Koetsier, C.B. deGruyter, Proc. R. Soc. Lond. A
420 (1988) 375–405.
[29] S.R. Goswami, M.-J. Dumont, V. Raghavan, Ind. Eng. Chem. Res. 55 (2016)
4473–4481.
[69] J.X. Liu, N. He, C.Y. Liu, G.R. Wang, Q. Xin, H.C. Guo, RSC Adv. 8 (2018)
9731–9740.
[30] A. Teimouri, M. Mazaheri, A.N. Chermahini, H. Salavati, F. Momenbeik, M. Fazel-
Najafabadi, J. Taiwan Inst. Chem. Eng. 000 (2014) 1–11.
[31] W. Partenheimer, V.V. Grushin, Adv. Synth. Catal. 343 (2001) 102–111.
[32] C.A. Antonyraj, B. Kim, Y. Kim, S. Shin, K.-Y. Lee, I. Kim, J.K. Cho, Catal. Commun.
57 (2014) 64–68.
[33] J. Artz, S. Mallmann, R. Palkovits, ChemSusChem 8 (2015) 672–679.
[34] B. Karimi, H.M. Mirzaei, E. Farhangi, ChemCatChem 6 (2014) 758–762.
[35] K. Ghosh, R.A. Molla, M.A. Iqubal, S.S. Islam, S.M. Islam, Appl. Catal. A Gen. 520
(2016) 44–52.
[70] T. Zhang, X. Yan, D.D. Sun, J. Hazard. Mater. 243 (2012) 302–310.
[71] M. Bicker, J. Hirth, H. Vogel, Green Chem. 5 (2003) 280–284.
[72] X. Qi, M. Watanabe, T.M. Aida, R.L.J. Smith, Green Chem. 10 (2008) 799–805.
[73] C. Laugel, B. Estrine, J.L. Bras, N. Hoffmann, S. Marinkovic, J. Muzart,
ChemCatChem 6 (2014) 1195–1198.
[74] A. Ranoux, K. Djanashvili, I.W.C.E. Arends, U. Hanefeld, ACS Catal. 3 (2013)
760–763.
[75] Y. Román-Leshkov, J.N. Chheda, J.A. Dumesic, Science 312 (2006) 1933–1937.
[76] H. Wang, Q. Kong, Y. Wang, T. Deng, C. Chen, X. Hou, Y. Zhu, ChemCatChem 6
(2014) 728–732.
[36] J. Nie, H. Liu, J. Catal. 316 (2014) 57–66.
[37] X. Xiang, L. He, Y. Yang, B. Guo, D. Tong, C. Hu, Catal Lett. 141 (2011) 735–741.
[38] Y. Liu, L. Zhu, J. Tang, M. Liu, R. Cheng, C. Hu, ChemSusChem 7 (2014)
3541–3547.
[39] P.V. Rathod, S.D. Nale, V.H. Jadhav, ACS Sustain. Chem. Eng. 5 (2017) 701–707.
[40] A. Takagaki, M. Takahashi, S. Nishimura, K. Ebitani, ACS Catal. 1 (2011)
1562–1565.
[77] B. Saha, M.M. Abu-Omar, Green Chem. 16 (2014) 24–38.
[78] J. Nie, H. Liu, Pure Appl. Chem. 84 (2012) 765–777.
[79] C. Moreau, R. Durand, C. Pourcheron, D. Tichit, Stud. Surf. Sci. Catal. 108 (1997)
399–406.
[80] B. Sarmah, R. Srivastava, Ind. Eng. Chem. Res. 56 (2017) 15017–15029.
103