Direct conversion of cellulose into isosorbide over Ni doped NbOPO4catalysts in water

Article abstract of DOI:10.1039/d0nj01403f

Isosorbide is a versatile chemical intermediate for the production of a variety of drugs, chemicals, and polymers, and its efficient production from natural cellulose is of great significance. In this study, bifunctional catalysts based on niobium phosphates were prepared by a facile hydrothermal method and used for the direct conversion of cellulose to isosorbide under aqueous conditions. NH3-TPD analysis showed that a high acid content existed on the catalyst surface, and pyridine infrared spectroscopic analysis confirmed the presence of both Lewis acid and Br?nsted acid sites, both of which played an important role in the process of carbohydrate conversion. XRD and H2-TPR characterization determined the composition and the hydrogenation centers of the catalyst. An isosorbide yield of 47% could be obtained at 200 °C for 24 h under 3 MPa H2 pressure. The Ni/NbOPO4 bifunctional catalyst retains most of its activity after five consecutive runs with slightly decreased isosorbide yield of 44%. In addition, a possible reaction mechanism was proposed that the synergistic effect of surface acid sites and hydrogenation sites was favorable to enhancing the cascade dehydration and hydrogenation reactions during the conversion of cellulose to isosorbide. This study provides as an efficient strategy for the development of novel multifunctional heterogeneous catalysts for the one-pot valorisation of cellulose. This journal is

Full text of DOI:10.1039/d0nj01403f

NJC
View Article Online
View Journal
PAPER
Direct conversion of cellulose into isosorbide over
Ni doped NbOPO4 catalysts in water†
Cite this:DOI: 10.1039/d0nj01403f
ab
Minyao He,abc Jiaxing Guo,ab Xincheng Wang,
*
Yongji Song,ab ShanShan Liu,ab
Hong Wangab and Cuiqing Liab
Isosorbide is a versatile chemical intermediate for the production of a variety of drugs, chemicals, and
polymers, and its efficient production from natural cellulose is of great significance. In this study,
bifunctional catalysts based on niobium phosphates were prepared by a facile hydrothermal method and
used for the direct conversion of cellulose to isosorbide under aqueous conditions. NH3-TPD analysis
showed that a high acid content existed on the catalyst surface, and pyridine infrared spectroscopic
analysis confirmed the presence of both Lewis acid and Brønsted acid sites, both of which played an
important role in the process of carbohydrate conversion. XRD and H2-TPR characterization determined
the composition and the hydrogenation centers of the catalyst. An isosorbide yield of 47% could be
obtained at 200 1C for 24 h under 3 MPa H2 pressure. The Ni/NbOPO4 bifunctional catalyst retains most
of its activity after five consecutive runs with slightly decreased isosorbide yield of 44%. In addition, a
possible reaction mechanism was proposed that the synergistic effect of surface acid sites and
hydrogenation sites was favorable to enhancing the cascade dehydration and hydrogenation reactions
during the conversion of cellulose to isosorbide. This study provides as an efficient strategy for the
development of novel multifunctional heterogeneous catalysts for the one-pot valorisation of cellulose.
Received 21st March 2020,
Accepted 27th May 2020
DOI: 10.1039/d0nj01403f
rsc.li/njc
decomposed into glucose under the action of a hydrolysis catalyst,
then glucose needs to be converted into sorbitol under the action of
Introduction
Cellulose is one of the most abundant renewable biomass a hydrogenation catalyst, and finally, sorbitol is dehydrated in two
resources that can be used for the preparation of fine chemicals steps to form isosorbide. In 2010, Moulijin et al.17 reported the
and fuels. Therefore, the development and utilization of cellulose feasibility of converting cellulose to isosorbide in molten hydrate
has attracted worldwide attention.1,2
media. They proposed that in ZnCl2 hydrate media, cellulose
Generally, cellulose can be catalytically converted into glucose, dissolution and chemical conversion can be performed without
5-HMF, lactic acid, polyols, biofuel, etc.3–8 Among polyhydric intermediate separation steps. Liang et al.18 proposed an effective
alcohols, isosorbide has attracted wide attention because it can method for converting concentrated cellulose to isosorbide. They
be used as a new solvent, pharmaceutical intermediate, and a used a combination of HCl + Ru/C to catalyse the conversion of
variety of polymer material monomers.9 Generally, isosorbide is cellulose for 6 h at 215 1C under 6 MPa H2, and obtained an
prepared by the dehydration of sorbitol under the catalysis of acid isosorbide yield of 49.5%. They believe that the reaction conditions
catalysts. Commonly used catalysts are mainly homogeneous are important factors controlling the reaction rate and product
catalysts (such as sulfuric acid, hydrochloric acid, etc.) and hetero- distribution; especially, the active metal species are the key factors
geneous catalysts (such as Amberlyst 15, modified SiO2, SnPO, influencing the conversion of cellulose to isosorbide. However, due
Nb2O5, etc.).10–15 However, due to the large number of hydrogen to the presence of corrosive liquid acid HCl, it is not suitable for
bonds intra- and inter- molecules of cellulose,16 the direct conver- industrial large-scale production.
sion of cellulose to isosorbide is obviously relatively difficult. Firstly,
Op De Beeck et al.19 reported the use of a combination of
cellulose needs to be decomposed into oligosaccharides and further heteropoly acids H4SiW12O40 and Ru/C for the preparation of
isosorbide from a variety of lignocellulosic raw materials.
a College of Chemical Engineering, Beijing Institute of Petrochemical Technology,
Beijing 102617, China. E-mail: wxcnathan@gmail.com, wangxc@bipt.edu.cn
b Beijing Key Laboratory of Fuels Cleaning and Advanced Catalytic Emission
Reduction Technology, Beijing 102617, China
c Beijing University of Chemical Technology, Beijing 100029, China
† Electronic supplementary information (ESI) available: Cellulose conversion and
results of long reaction times. See DOI: 10.1039/d0nj01403f
Isosorbide yields over 50% were obtained at 210 1C under
5 MPa H2 after a reaction time of 5 h. Moreover, when wheat
straw pulp was used as the raw material, an isosorbide yield of
63% could be obtained after 1 h. Most importantly, heteropoly
acid catalysts are non-toxic, easy to recycle, etc. compared to
liquid acids. Yamaguchi et al.20 reported that the combination
New J. Chem.
This journal is © The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2020

View Article Online
Paper
NJC
of supported Ru/C catalyst and ion exchange resin (Amberlyst 70) and (222) belonging to the lattice plane of Ni(NO3)2 could be
could realize the one-pot conversion of cellulose to isosorbide. observed at 2y = 17.41, 21.01, 23.31, 24.91 and 42.81. With the
The yield of isosorbide was up to 55.8% at 190 1C under 5 MPa increase of Ni loading, the intensity of the characteristic diffraction
H2 after 6 h. Yang et al.21 prepared a Ru2.2/Fe3O4@void@ peaks of Ni(NO3)2 increases gradually, indicating a large number
PMO-SO3H0.43 catalyst by dispersing Ru nanoparticles on the of Ni(NO3)2 on the surface of the support due to aggregation and
structure with Fe3O4 as the core and PMO containing SO3H as increase of particle size of Ni species. The Ni/NbOPO4 catalyst with
the shell, which catalysed the direct conversion of cellulose to 30% Ni loading showed no obvious characteristic diffraction
isosorbide. Under the conditions of 220 1C, 6 MPa H2 an peaks, because Ni(NO3)2 was highly dispersed on the surface of
isosorbide yield of 58.1% can be obtained after 2 h. Xi et al.22 NbOPO4 while the NbOPO4 itself has an irregular crystal form.24
reported a two-step sequential reaction process for the conversion
The FTIR spectra of Ni/NbOPO4 catalysts showed bands
of cellulose to isosorbide. Firstly, cellulose was degraded under centred at approximately 1040 cmÀ1, which could be assigned
the action of Ru/NbOPO4-pH2 catalyst at 170 1C and 4 MPa H2 to to the stretching vibration of Nb–O–P, indicating the presence of
obtain sorbitol and dehydrate sorbitol. Then Ru/NbOPO4-pH2 niobium phosphate skeleton (Fig. 2a).25 The bands at 3384 cmÀ1
,
was filtered off and the reaction was continued at 230 1C under 1631 cmÀ1, 1380 cmÀ1 and 825 cmÀ1 are attributed to nickel nitrate
the action of the NbOPO4-pH2 catalyst. In this way, an isosorbide hexahydrate.26 In detail, the bands at 3384 cmÀ1 and 1631 cmÀ1
yield of 56% was obtained. They found that the acid content of could be attributed to the asymmetric OH stretching and
solid acid catalysts had a significant effect on catalyst stability. bending vibrations of adsorbed water molecules. The band at
Sun et al.23 prepared a Ru@mNbPO catalyst for the conversion of 1380 cmÀ1 is attributed to the symmetrical stretching vibration
cellulose to isosorbide and an isosorbide yield of 52% can be of –NO2; and the band at 825 cmÀ1 is attributed to the stretching
obtained at 220 1C under 6 MPa H2 after 1 h. However, the above vibration of N–O. Raman vibrational spectra revealed the sym-
catalyst systems basically use noble metals as hydrogenation metric stretching mode of NbQO species in highly distorted NbO6
catalysts. Although promising yields of isosorbide were obtained, octahedra with vibrational band in the range of 800–900 cmÀ1 27
the economic costs of the noble metal catalysts were high and not The peaks at 1056 and 1307 cmÀ1 were assigned to the free nitrate,
suitable for large-scale industrial applications. the intensity of which increased dramatically with the increase of
.
In this study, bifunctional supported Ni/NbOPO4 catalysts Ni loading.26 Elemental analysis of the catalysts was conducted
with both acid sites and hydrogenation sites were synthesized
by a simple method and developed for the one-pot conversion
of cellulose into isosorbide. The surface acidity and reducibility
of the catalysts were studied by NH3-TPD, pyridine FTIR and
H2-TPR spectroscopy. The yields were optimized by varying the
process conditions, including Ni loading, reaction temperature,
substrate dosage and catalysts loading. Furthermore, the reaction
mechanism was further inferred based on the characterization
results of the catalyst.
Results and discussion
X-ray diffraction patterns, FTIR and Raman spectra
The XRD results of Ni/NbOPO4 catalysts with different Ni loading
are shown in Fig. 1. The appearance of (220), (111), (321), (400)
Fig. 2 (a) FTIR and (b) Raman spectra of Ni/NbOPO4 catalysts: (i) 10Ni/
NbOPO4, (ii) 30Ni/NbOPO4, (iii) 50Ni/NbOPO4, (iv) 70Ni/NbOPO4, (v) 90Ni/
Fig. 1 X-ray diffraction patterns of Ni/NbOPO4 materials.
NbOPO4.
New J. Chem.
This journal is © The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2020

View Article Online
NJC
Paper
Table 1 Catalytic conversion of cellulose to isosorbide by Ni/NbOPO4 catalysts (cellulose = 0.5 g, Ni/NbOPO4 = 0.5 g, 80 mL H2O, 3 MPa H2, 200 1C, 24 h)
Ni loadinga
(wt%)
Ni/Nb
Specific surface
Total acid
B/L
Cellulose
conv. (%)
Isosorbide
(%)
Entry
Catalyst
(molar ratio)a
area (m2 gÀ1
)
sitesb (mmol gÀ1
)
ratioc
1
2
3
30Ni/NbOPO4
70Ni/NbOPO4
90Ni/NbOPO4
31.2
72.5
88.4
1.08
2.52
3.07
60.5
42.4
40.6
0.64
0.72
0.81
0.09
0.42
0.31
88.97
100
100
30.73
47.19
42.97
a
b
c
Determined by ICP-OES analysis. Determined by NH3-TPD analysis. The Brønsted/Lewis site (BAS/LAS) ratio was determined by monitoring
pyridine desorption at 200 1C with FTIR. Note: the abbreviations do not suggest any stoichiometric composition.
through ICP-OES analysis after dissolution of the samples in Acidity evaluation by NH3-TPD and FTIR spectroscopy of
concentrated hydrofluoric acid solution. The Ni loading were adsorbed pyridine
consistent with the amount added during the preparation process
(Table 1).
The acidic properties of Ni/NbOPO4 were determined by NH3
temperature-programmed desorption (TPD) (Fig. 4a). The
desorption peaks of NH3 over the ranges of 200 1C to 400 1C were
H2 temperature programmed reduction
generally classified as medium acid sites.31 However, in the current
Fig. 3 shows the H2-TPR spectra of catalysts with varied Ni
loadings. Obviously, two hydrogen consumption peaks were
observed for all the Ni/NbOPO4 samples. The first one appears
at about 320–340 1C could be assigned to the reduction of
large particles of NiO, whose performance is similar to pure
bulk NiO.28–30 The other one is a small and wide hydrogen
consumption peak appearing at 430–450 1C, which is attributed
to the reduction of NiO that has a strong interaction with the
support. In this study, the TPR spectra using nickel nitrate as a
precursor could be divided into two parts. Firstly, nickel nitrate
is decomposed into NiO at high temperatures when NiO inter-
acts with the support NbOPO4, leading to two forms of NiO,
pure bulk NiO and NiO with strong interaction with NbOPO4.
The hydrogen consumption peak at low temperature (340 1C)
indicated the reduction of bulk NiO, while the peak at high
temperature (450 1C) indicated the reduction of NiO with
strong interaction with NbOPO4. In addition, the TPR curves
of the Ni/NbOPO4 catalyst with 30% Ni loading and 90% Ni
loading are basically the same. The hydrogen consumption
peak of 70Ni/NbOPO4 revealed a blue shift of about 20 1C.
Considering that the reduction temperature of the catalyst was
450 1C in this study, the nickel nitrate of the catalyst was
supposed to be completely reduced during the pre-treatment
procedure, achieving the required high hydrogenation activity.
study, there were two distinct peaks in the range, dividing it into
two acid centres of different strengths. Obviously, the acidification
method has promoted the surface acidity of Ni/NbOPO4, getting
rid of weak acid sites over the ranges of 100 1C to 200 1C on pure
NbOPO4. The relatively strong acid sites at ca. 350 1C was attrib-
uted to the desorption of ammonia from both strong Lewis and
Brønsted acid sites.32 Note that the addition of Ni and acidification
resulted in increased ratio of strong acid sites, which is beneficial
to the hydrolysis of cellulose and sorbitol. The total acid sites
increased with the increase of Ni loading (Table 1); however, the
yields of isosorbide did not show the same regularity. This is
because the high amount of total acid sites led to the dehydration
of glucose, which is detrimental to the hydrogenation of glucose to
sorbitol during the production of isosorbide. Therefore, it was
proposed that there was a certain balance between the hydrolysis
of cellulose and the hydrogenation of glucose, overcoming the
dehydration of glucose.
The pyridine adsorption by in situ IR spectroscopy was used
to evaluate the surface Lewis/Brønsted character of Ni/NbOPO4
(Fig. 4b–d). The bands at 1450, 1575 and 1608 cmÀ1 were
attributed to Lewis acid sites,33 while the bands at 1540 cmÀ1 and
1635 cmÀ1 to Brønsted acid sites.34 The characteristic vibration
peaks of Lewis and Brønsted acids are overlapped at 1490 cmÀ1 35
.
The IR analysis showed that the intensities decreased as the
desorption temperature was increased from 25 to 200 1C. The
relative Brønsted-to-Lewis acidity ratios (B/L ratios) could be
deduced by integrating the areas under the characteristic bands
of 1540 cmÀ1 and 1450 cmÀ1 while considering the extinction
coefficients.36 The B/L ratio of 30Ni/NbOPO4 was 0.09, indicating
that most of the acid sites were Lewis acid sites as can be seen from
Fig. 4b. The Ni/NbOPO4 with 70% Ni loading obtained the highest
B/L ratio. Indeed, strong Brønsted acidity was favourable for hydro-
lysis of cellulose and the synergy between Brønsted and Lewis
acidity favoured the conversion of glucose to sorbitol.
Catalysis
Optimization of the process conditions for the conversion of
cellulose to isosorbide catalysed by Ni/NbOPO4. To gain insight into
the effects of the process conditions during the transformation of
Fig. 3 H2-TPR spectra of the Ni/NbOPO4 samples.
New J. Chem.
This journal is © The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2020

View Article Online
Paper
NJC
Fig. 4 NH3-TPD profiles of the Ni/NbOPO4 samples (a) and FTIR spectra of pyridine dosed onto Ni/NbOPO4 samples (pure powder pressed disks) and
then desorbed. (b) 30Ni/NbOPO4, (c) 70Ni/NbOPO4, (d) 90Ni/NbOPO4. Evacuation temperature: (i) 25 1C, (ii) 50 1C, (iii) 100 1C, (iv) 150 1C, (v) 200 1C.
Absorbance peaks characteristic to strong Lewis (SL), weak Lewis (WL), Brønsted (B) and Lewis (L) sites are indicated.
cellulose to isosorbide, the liquid products obtained using cellulose was so promising that only the sample with 10% Ni
Ni/NbOPO4 were identified and quantified by HPLC analysis. loading gave a conversion lower than 70%. Taking the isosorbide
Obviously, the yield of isosorbide depended strongly on the Ni yield into consideration, high cellulose conversion tends to
loading of the catalysts (Fig. 5a). Only 5.1% of isosorbide was leading to high isosorbide yields. The total acid sites of the most
obtained over 10Ni/NbOPO4. Fortunately, the yield increased promising catalyst, 70Ni/NbOPO4, were among the highest of the
gradually with the increase of Ni loadings on NbOPO4. The catalysts prepared, together with the largest BAS to LAS ratio. The
hydrogenation effect of the catalysts with Ni loading below 30% NH3-TPD results indicated a wide acid strength distribution of
is obviously inferior to which above 30%. In the meantime, 70Ni/NbOPO4, which is possibly the reason for the good perfor-
there is little difference in the catalytic effect of catalysts with Ni mance of the catalyst. Therefore, it is believed that the high total
loadings of 30%, 40%, 50%, and 60%, with the yields of acid sites ensure the fast hydrolysis of cellulose and the further
isosorbide all being about 33%. The isosorbide yield reached hydrolysis of sorbitol. Moreover, the BAS/LAS ratio did affect the
47.19% using the catalyst with 70% Ni loading. However, formation of isosorbide. The synergistic effect of BAS and LAS is
higher Ni loading did not necessarily lead to higher isosorbide reflected during the catalytic conversion of cellulose to isosor-
yield (isosorbide yields being 42.97%, 42.63%, respectively, for bide. Appropriate total acid sites and high BAS/LAS ratio are
Ni loadings of 80% and 90%). Indeed, the increase of nickel important factors for the direct catalytic conversion of cellulose
content might provide more active sites for hydrogenation, to isosorbide. From the economic point of view, nickel was
which is conducive to the formation of sorbitol by hydrogenation competitive, although the hydrogenation effect of nickel was not
of glucose hydrolysed from cellulose. Then sorbitol was converted as high as the ruthenium catalyst. Therefore, 70Ni/NbOPO4 was
to isosorbide via dehydration over acid sites of Ni/NbOPO4.
selected for further optimization of reaction conditions.
Fig. 5b shows the effect of reaction temperature on the
A reaction time of 24 h was used in order to enhance the
experimental efficiency because the long-term reaction of cel- conversion of cellulose to isosorbide. At 160 1C, cellulose could
lulose tend to lead to decreased isosorbide yield (Fig. S1, ESI†). not react substantially, while at 180 1C, only a small amount of
The cellulose conversion and isosorbide yields corresponding isosorbide (11.64%) was formed after 24 h. At 200 1C, the yield
to different supported catalysts at the end of the reaction time of isosorbide increased dramatically to 47.19% after 24 h.
of 24 h are showed in Table S1 (ESI†). The conversion of However, at temperatures above 200 1C, isosorbide yields
New J. Chem.
This journal is © The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2020

View Article Online
NJC
Paper
Fig. 5 Optimization of the reaction conditions: (a) Ni loading, (b) reaction temperature, (c) substrate dosage, and (d) catalyst loading (conditions:
cellulose = 0.5 g, Ni/NbOPO4 = 0.5 g, 80 mL H2O, 3 MPa H2, 200 1C).
showed a trend of increasing and then decreasing, possibly due that the highest yield occurs when the catalyst dosage is 0.5 g,
to intense side reactions that led to the decrease of isosorbide while the yield of isosorbide didn’t increase with the catalyst
selectivity. Therefore, the reaction temperature of 200 1C was dosage, which may because that a large number of acid sites in
employed as the reaction condition hereafter.
the catalyst first hydrolyse cellulose to glucose, and then further
Fig. 5c shows the effect of the substrate dosage on the dehydrate it to smaller molecules of sugar, Therefore, the
preparation of isosorbide from cellulose. An optimum yield of decrease of glucose hydrolysed from cellulose results in the
isosorbide (45.27%) was obtained when 0.25 g cellulose was decrease of sorbitol and further the decrease of isosorbide yield.
used, indicating that the amount of catalyst was excessive and
To verify the reusability of the catalyst, five consecutive
isosorbide was further converted to side-products afterwards. reactions were conducted with 70Ni/NbOPO4 under the same
The use of 0.5 g cellulose gave the highest isosorbide yield, while reaction conditions after catalyst recovery (Fig. 6). Between
excessive amount of cellulose, such as 0.75 g and 1 g, led to runs, the catalyst was washed by deionized water and dried at
drastically lower isosorbide yield. This might be due to the 110 1C overnight. For each run, the catalyst was weighed to
increase of cellulose dosage amplified their limited contact with maintain a constant substrate-to-catalyst ratio. Complete con-
catalyst during the heterogeneous catalytic reaction process.
version of cellulose was observed without a significant loss of
The effects of the 70Ni/NbOPO4 dosage on the conversion of the isosorbide yield after five consecutive runs, demonstrating
cellulose and the yield of isosorbide are presented in Fig. 5d. The that the catalyst was recyclable.
highest yield was obtained when the catalyst dosage is 0.5 g, while
Mechanism investigation
the yield of isosorbide didn’t increase with further increase of
catalyst amount. Indeed, a large amount of acid sites in the catalyst Generally, the conversion of cellulose to isosorbide involves
was beneficial for the first stage hydrolyzation of cellulose to four consecutive reactions.23 Firstly, the hydrolysis of cellulose
glucose, and the further dehydration glucose to side-products such to glucose is catalysed by acid sites. Subsequently, glucose is
as furfural. In this way, the decrease of glucose resulted in the hydrogenated to sorbitol over nickel sites under a hydrogen
decrease of the second stage hydrogenation of glucose to sorbitol atmosphere. Finally, two consecutive dehydration reactions of
and the further consecutive dehydration of sorbitol to isosorbide. sorbitol lead to isosorbide over acid sites with 1,4-sorbitan as the
The effect of catalyst dosage on the reaction of cellulose to intermediate. The further dehydration of 1,4-sorbitan requires
isosorbide is showed in Fig. 5d. It can be seen from the figure strong acidity. Therefore, hydrothermally stable bifunctional
New J. Chem.
This journal is © The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2020

View Article Online
Paper
NJC
Conclusions
This work presented a preliminary study of the use of a
bifunctional catalyst based on niobium phosphates for the
one-pot conversion of cellulose to isosorbide. The surface
acidity was investigated by NH3-TPD and pyridine infrared
spectroscopic analysis, revealing the high total acid sites and
high B/L ratios. The H2-TPR characterization determined the
hydrogenation centres of the catalyst. The product distribution
depended closely on the Ni loading, reaction temperature,
substrate dosage, and catalyst loading. An optimal isosorbide
yield of 47% was obtained at 200 1C after 24 h. A possible
reaction mechanism regarding the synergy of acid sites and
hydrogenation sites for cascade conversion of cellulose to
isosorbide was proposed. Currently, further efforts to improve
the efficiency of the bifunctional niobium-based catalysts for
the direct conversion of cellulose are ongoing in our group.
Fig. 6 Results of the reusability study (conditions: cellulose = 0.5 g, 70Ni/
NbOPO4 = 0.5 g, 80 mL H2O, 200 1C, 24 h)
solid acid catalyst with both high total acid sites and high
hydrogenation activity is efficient for the conversion of cellulose
into isosorbide in one pot.
Experimental
According to the characterization results, the hydrogenation
sites derived from the reduced Ni on the surface of the catalyst,
while the Brønsted acidity originates from the surface P–OH or
Nb–OH groups and the Lewis acidity originates from the
partially hydrolysed framework Nb species or coordinatively
unsaturated Nb(V) sites. In this study, the niobium sites were
responsible for the hydrolysis of cellulose and the dehydration
of sorbitol and 1,4-sorbitan. Indeed, glucose and 1,4-sorbitan
were detected in the filtrate after the reaction (Table S1, ESI†).
Mannitol was also observed due to the hydrogenation of
fructose, which derived from the isomerization of glucose.
However, no certain trends were obtained considering the
production of these side products with different Ni loadings.
The 70Ni/NbOPO4 with the largest B/L ratio gave the best
catalytic performance. Because of the water-tolerate Lewis
acidity of the NbOPO4, the Lewis acid sites was supposed to
adhere the substrates due to the electrostatic attraction between
the negatively charged O atoms and the positively charged metal
sites of Nb. The reaction mechanism is depicted in Scheme 1.
In a typical synthesis process, a certain amount of Nb2O5
powder and KOH were mixed uniformly at a molar ratio of
1 : 10, the mixture was calcined at 400 1C for 3 hours to obtain a
molten mixture, and then dissolved to obtain potassium niobate
solution. Potassium niobate solution (100 mL) was stirred in a
water bath at 40 1C and then adjusted to pH = 4 using 2 M
phosphoric acid. After stirring for 1 hour, this mixture was
stirred for additional 2 hours at 80 1C, and the solution was then
aged in a Teflon-lined autoclave for 24 h at 160 1C. After cooled
down, the solid was filtered, washed with distilled water and
then dried at 110 1C overnight. Finally, NbOPO4 sample was
obtained by calcination at 400 1C for 3 h in static air.37 The
niobium phosphate was further acidified in 2 M nitric acid at
60 1C for 1 hour, followed by centrifugation, washing with
distilled water and drying. Nickel–niobium phosphate catalysts
were prepared by a facile impregnation method. The niobium
phosphates were impregnated in different concentrations of
aqueous nickel nitrate solution for 8 h at 40 1C. Then the solid
was filtrated and dried at 110 1C overnight. Before use, the
catalysts were reduced at 450 1C for 3 h under H2 atmosphere.
The resulting materials were designated xNi/NbOPO4, where x
refers to the weight percent of metallic Ni with respect to the
weight of niobium phosphates.
Structural information about the Ni/NbOPO4 samples with-
out reduction was obtained by powder X-ray diffraction (PXRD)
(Rigaku D/Max-2600/PC), The Brunauer–Emmett–Teller (BET)
method was used to calculate the specific surface area (ASAP
2020). Temperature-programmed reduction by H2 (H2-TPR) was
recorded on Hiden QIC-20. Initially, a catalyst sample about
100 mg was degassed at 500 1C for 1.0 h under a nitrogen flow
of 30 mL minÀ1. Then, the sample was cooled to 40 1C and
heated linearly from 40 to 800 1C under a constant H2 (10% H2
in Ar) flow of 40 mL minÀ1. The actual Ni loadings and the
molar ratio of metal to Nb was determined by ICP-OES analysis.
The total acid sites were determined by ammonia temperature-
programmed desorption (NH3-TPD) (Hiden QIC-20) with a thermal
Scheme 1 Proposed reaction mechanism for the one-pot conversion of
cellulose to isosorbide.
New J. Chem.
This journal is © The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2020

View Article Online
NJC
Paper
7 N. Ji, T. Zhang, M. Y. Zheng, A. Q. Wang, H. Wang, X. D.
Wang, Y. Y. Shu, A. L. Stottlemyer and J. G. Chen, Catal.
Today, 2009, 147(2), 77–85.
8 G. W. Huber, J. N. Chheda, C. J. Barrett and J. A. Dumesic,
Science, 2005, 308, 1446–1450.
conductivity detector. Typically, a sample of approximately 100 mg
was initially reduced at 450 1C for 40 min and then cooled down to
100 1C. Ammonia was adsorbed at 100 1C for 60 min to reach
saturation. Then, the ammonia supply line was shut off, and N2
was purged at 15 mL minÀ1 for 2 h to remove physically adsorbed
NH3. The sample was then heated linearly from 100 1C to 800 1C at
a rate of 10 1C minÀ1. The Lewis and Brønsted acid sites were
investigated by FTIR using pyridine as the probe molecule on a
Bruker Vertex 70 FTIR spectrometer with an MCT/A detector,
according to previously reported method.38
9 J. J. Xia, D. H. Yu, Y. Hu, B. Zou, P. Sun, H. Li and H. Huang,
Catal. Commun., 2011, 12(6), 544–547.
10 N. A. Khan, D. K. Mishra, S.-S. Hwang, Y.-W. Kwak and
S. H. Jhung, Res. Chem. Intermed., 2011, 37, 1231–1238.
11 R. Barker, J. Org. Chem., 1970, 35(2), 461–464.
12 N. A. Khan, D. K. Mishra, J.-S. Hwang, Y.-W. Kwak and
S. H. Jhung, Res. Chem. Intermed., 2011, 37, 1231–1238.
13 J. Shi, Y. H. Shan, Y. Tian, Y. Wan, Y. T. Zheng and
Y. Y. Feng, RSC Adv., 2016, 6, 13514–13521.
Conversion of cellulose was carried out in a 100 mL stainless
steel autoclave reactor. Typically, 0.5 g cellulose (ball milled
with a 250 mL ZrO2 pot and 6 ZrO2 balls (+ 20 mm) for 60 h at
a speed of 600 rpm before use),39,40 0.5 g catalyst and 80 mL
water were added into the reactor, then flushed with nitrogen to
remove the air and purged with 3 MPa of hydrogen gas at room
temperature. Afterwards, the mixture was heated to 200 1C at
101 minÀ1. After the reaction, the solution was filtered, and the
filtrate was analysed by high performance liquid chromatography.
Cellulose conversion [%] was calculated by using the weight
difference of cellulose before and after reaction with an uncertainty
of Æ3%.22 The concentrations of isosorbide were analysed by HPLC
(Agilent 1200, USA) using a Biorad Aminex 87-H column at 45 1C
14 M. Y. Gu, D. H. Yu, H. M. Zhang, P. Sun and H. Huang,
Catal. Lett., 2009, 133, 214–220.
15 X. G. Zhang, A. I. M. Rabee, M. Isaacs, A. F. Lee and K. Wilson,
ACS Sustainable Chem. Eng., 2018, 6(11), 14704–14712.
16 M. Kitano, D. Yamaguchi, S. Suganuma, K. Nakajima, H. Kato,
S. Hayashi and M. Hara, Langmuir, 2009, 25(9), 5068–5075.
17 R. M. de Almeida, J. Li, C. Nederlof, P. O’Connor, M. Makkee
and J. A. Moulijn, ChemSusChem, 2010, 3(3), 325–328.
18 G. F. Liang, C. Y. Wu, L. M. He, J. Ming, H. Y. Cheng,
L. H. Zhuo and F. Y. Zhao, Green Chem., 2011, 13, 839–842.
19 B. Op de Beeck, J. Geboers, S. Van De Vyver, J. Van Lishout,
J. Shelders, W. J. Huijgen, C. M. Courtin, P. A. Jacobs and
B. F. Sels, ChemSusChem, 2013, 6(1), 199–208.
with 0.005 M H2SO4 as the eluent at a flow rate of 0.5 mL minÀ1
.
A refraction index detector was used. The experiments were
replicated at least three times and were reported as the mean
values.
20 A. Yamaguchi, O. Sato, N. Mimura and M. Shirai, Catal.
Commun., 2015, 67, 59–63.
Yield of isosorbide (%) = (moles of isosorbide/
21 Y. Yang, W. Zhang, F. Yang, B. Zhou, D. H. Zeng, N. Zhang,
G. M. Zhao, S. J. Hao and X. Zhang, Nanoscale, 2018, 10,
2199–2206.
22 J. X. Xi, Y. Zhang, D. Q. Ding, Q. N. Xia, J. J. Wang, X. H. Liu,
G. Z. Lu and Y. Q. Wang, Appl. Catal., A, 2014, 469, 108–115.
23 P. Sun, X. D. Long, H. He, C. G. Xia and F. W. Li, Chem-
SusChem, 2013, 6(11), 2190–2197.
moles of glucose unit in cellulose) Â 100%
Conflicts of interest
There are no conflicts to declare.
24 X. Q. Wang, Y. Liu and Z. B. Wu, Chem. Eng. J., 2020,
382, 122941.
25 A. Dutta, A. K. Patra, S. Dutta, B. Saha and A. Bhaumik,
J. Mater. Chem., 2012, 22, 14094–14100.
26 T. J. Taylor, D. Dollimore, G. A. Gamlen, A. J. Barnes and
M. A. Stuckey, Thermochim. Acta, 1986, 101, 291–304.
27 K. Nakajima, Y. Baba, R. Noma, M. Kitano, J. N. Kondo,
S. Hayashi and M. Hara, J. Am. Chem. Soc., 2011, 133,
4224–4227.
Acknowledgements
This project was supported by National Natural Science Foundation
of China (21908009), Beijing Natural Science Foundation (2184101
and 2202016), and Beijing Education Committee Science and
Technology Project (KM201810017001).
Notes and references
28 A. M. Diskin, R. H. Cunningham and R. M. Ormerod, Catal.
Today, 1998, 46(2–3), 147–154.
1 T. Y. Deng, J. Y. Sun and H. C. Liu, Sci. China: Chem., 2010,
53(7), 1476–1480.
29 C. Louis, Z. X. Cheng and M. Che, J. Phys. Chem., 1993,
97(21), 5703–5712.
30 D. W. Southmayd, C. Contescu and J. A. Schwarz, J. Chem.
Soc., Faraday Trans., 1993, 89, 2075–2083.
2 P. Gallezot, Chem. Soc. Rev., 2012, 41(4), 1538–1558.
3 Y. B. Huang and Y. Fu, Green Chem., 2013, 15, 1095–1111.
4 L. S. Yan, R. S. Ma, H. X. Wei, L. Z. Li, B. Zou and Y. W. Bin,
Bioresour. Technol., 2019, 279, 84–91.
31 X. C. Wang, Y. J. Song, C. P. Huang and B. Wang, Sustainable
Energy Fuels, 2018, 2, 1530–1541.
32 D. Y. Xi, Q. M. Sun, J. Xu, M. Cho, H. S. Cho, S. Asahina,
Y. Li, F. Deng, O. Terasaki and J. H. Yu, J. Mater. Chem. A,
2014, 2(42), 17994–18004.
5 Y. L. Wang, W. P. Deng, B. J. Wang, Q. H. Zhang, X. Y. Wan,
Z. C. Tang, Y. Wang, C. Zhu, Z. X. Cao, G. C. Wang and
H. L. Wan, Nat. Commun., 2013, 4, 2141.
6 W. W. Zhu, H. M. Yang, J. Z. Chen, C. Chen, L. Guo, H. M. Gan,
X. G. Zhao and Z. S. Hou, Green Chem., 2014, 16(3), 1534–1542.
New J. Chem.
This journal is © The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2020

View Article Online
Paper
NJC
33 F. de Clippel, M. Dusselier, R. Van Rompaey, P. Vanelderen, 36 C. A. Emeis, J. Catal., 1993, 141(2), 347–354.
J. Dijkmans, E. Makshina, L. Giebeler, S. Oswald, G. V. 37 X. C. Wang, Y. J. Song, C. P. Huang and B. Wang, Sustainable
Baron, J. F. M. Denayer, P. P. Pescarmona, P. A. Jacobs and
B. F. Sels, J. Am. Chem. Soc., 2012, 134(24), 10089–10101.
34 R. M. West, M. S. Holm, S. Saravanamurugan, J. M. Xiong,
Energy Fuels, 2018, 2, 1530–1541.
38 X. C. Wang, Y. J. Song, L. Huang, H. Wang, C. P. Huang and
C. Q. Li, Catal. Sci. Technol., 2019, 9, 1669–1679.
Z. Beversdorf, E. Taarning and C. H. Christensen, J. Catal., 39 H. Kobayashi, M. Yabushita, T. Komanoya, K. Hara, I. Fujita
2010, 269(1), 122–130. and A. Fukuoka, ACS Catal., 2013, 3, 581–587.
35 M. G. Mazzotta, D. Gupta, B. Saha, A. K. Patra, A. Bhaumik 40 L. S. Ribeiro, J. J. M. Orfao and M. F. R. Pereira, Green Chem.,
and M. M. Abu-Omar, ChemSusChem, 2014, 7(8), 2342–2350. 2015, 17, 2973–2980.
´
˜
New J. Chem.
This journal is © The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2020