The effect of physical morphology and the chemical state of Ru on the catalytic properties of Ru-carbon for cellulose hydrolytic hydrogenation

Article abstract of DOI:10.1039/d0nj03014g

Ru-carbon catalysts with different physical morphologies and chemical states of Ru were prepared by different methods and used to catalyze the hydrolytic hydrogenation of cellulose at high temperatures. The physical morphology of Ru particles and the chemical state of Ru significantly influenced the catalytic performance. The Ru nanoparticles in Ru@MC prepared by thein situcarbothermal reduction method exhibited a special chemical state due to the strong interaction with carbon. The special structure could not only prevent the growth of Ru particles but also enhance the hydrogen spillover effect and improve the hydrogenation efficiency. Among the Ru-carbon catalysts, Ru@MC showed the best catalytic performance with a 72.4% yield of sorbitol. Furthermore, the embedded structure of Ru@MC stabilized the Ru nanoparticles, and the catalyst could be reused at least 6 times.

Full text of DOI:10.1039/d0nj03014g

NJC
View Article Online
View Journal
PAPER
The effect of physical morphology and the
chemical state of Ru on the catalytic properties of
Ru–carbon for cellulose hydrolytic
hydrogenation†
Cite this:DOI: 10.1039/d0nj03014g
Gang Zhang,ab Tong Chen, *a Yi Zhang,ab Tao Liuab and Gongying Wanga
Ru–carbon catalysts with different physical morphologies and chemical states of Ru were prepared by
different methods and used to catalyze the hydrolytic hydrogenation of cellulose at high temperatures.
The physical morphology of Ru particles and the chemical state of Ru significantly influenced the
catalytic performance. The Ru nanoparticles in Ru@MC prepared by the in situ carbothermal reduction
method exhibited a special chemical state due to the strong interaction with carbon. The special
structure could not only prevent the growth of Ru particles but also enhance the hydrogen spillover
effect and improve the hydrogenation efficiency. Among the Ru–carbon catalysts, Ru@MC showed the
best catalytic performance with a 72.4% yield of sorbitol. Furthermore, the embedded structure of
Ru@MC stabilized the Ru nanoparticles, and the catalyst could be reused at least 6 times.
Received 15th June 2020,
Accepted 6th August 2020
DOI: 10.1039/d0nj03014g
rsc.li/njc
products, thus making sure to obtain high selectivity to high-
value chemicals.
1. Introduction
The conversion of renewable biomass and its derivatives to high-
The synthesis of sorbitol from cellulose is performed via a
value chemicals and fuels is of significant importance for sustain- cascade reaction: hydrolysis of cellulose to glucose via acid
able social development.1 Sorbitol, one of the promising platform catalysts and then hydrogenation of glucose into sorbitol over
compounds, has not only been widely used in industrial produc- metal catalysts.7,8 Hot compressed water can dissociate an H+
tion such as pharmaceuticals and cosmetics, but also used as a ion reversibly and promote the hydrolysis of cellulose when the
raw material for the preparation of other chemical products such temperature is above 190 1C.9 A sorbitol yield of 22% was
as ethylene glycol and isosorbide.2,3 Generally, the industrial obtained by using commercial ruthenium carbon (Ru/C) in
preparation of sorbitol involves the hydrogenation of glucose, hot compressed water at 245 1C.10 However, high temperature
which mainly comes from the hydrolysis of starch.4 However, easily caused polymerization of glucose and hydrogenolysis of
starch is mainly derived from edible crops, and its use will sorbitol, thus the yield of sorbitol was low.11 Subsequent
compete with the human food supply. Unlike corn and starch, studies have found that the Ru supported on acidic carriers
cellulose is one type of renewable inedible biomass and its use such as sulfonated carbon,12–14 phosphate,2 molecular sieve,15,16
will not compete with food supplies. Nevertheless, the primary etc. could directly promote the conversion of cellulose to sorbitol
obstacle for producing sorbitol from cellulose is the high cost of at low temperatures, reaching a high yield of 71%. However, it
pre-treatment of raw lignocellulose, especially cellulose hydrolysis should be highlighted that the process was usually time-
and purification of glucose from by-products.5 Therefore, many consuming because the low acidity restrained the hydrolysis
research studies have been conducted to achieve the direct rate. The acid–Ru binary catalyst can achieve efficient conversion
preparation of sorbitol from cellulose avoiding the degradation of cellulose in a short time due to the easy adjustment of the
of metastable glucose6 and separation process of intermediate ratio of acidity to reducibility. For example, Geboers et al. found
that heteropoly acids coupled with Ru/C could effectively
a Chengdu Institute of Organic Chemistry, Chinese Academy of Sciences,
Chengdu 610041, Sichuan, P. R. China. E-mail: chentongw@sina.com.cn;
Fax: +86 028 85220713; Tel: +86 028 85250005
b National Engineering Laboratory for VOCs Pollution Control Material &
Technology, University of Chinese Academy of Sciences, Beijing 101408,
P. R. China
improve cellulose conversion, a C6 alcohol (sorbitol + mannitol)
yield of 68% was obtained in only 1 h.17 However, the poor water
solubility of heteropoly acids makes them difficult to recycle,
which hampered their industrial-scale application. To improve
the recoverability and reusability of the solid acid catalyst, Liao
et al. used zirconium phosphate (ZrP) instead of heteropoly acid
as an acid catalyst, and the yield of C6 alcohol reached 85%.18
† Electronic supplementary information (ESI) available. See DOI: 10.1039/
d0nj03014g
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
The Ru–carbon catalyst is crucial for the whole process as it was heated to 750 1C and kept for 3 h with the protection of N2.
influences the product distribution, promotes cellulose conver- Finally, the obtained black powder was washed with hot
sion, and reduces the polymerization of hydrolysates. Generally, sodium hydroxide solution to remove silica. The obtained black
the Ru physical morphology and chemical state in the catalyst powder was denoted as MC. Ru/MC was prepared by the wet
could affect its catalytic performance and stability. For example, impregnation method. The obtained MC was used as the
modifying carbon materials with hydrophilic groups19 or intro- support and added to the 15 mL aqueous solution containing
ducing a dopant (N, P) into carbon materials6 could decrease 0.07 g RuCl3. Then the samples were dried at 120 1C overnight.
the metal-nanoparticle size and improve the catalytic activity. The sample was reduced at 300 1C for 2 h under a H2 flow
However, the conventional supported Ru catalysts exhibit a weak before use, and denoted as Ru/MC.
interaction between Ru and the support to immobilize them, so
Ruthenium embedded in mesoporous carbon (Ru@MC).
Ru catalysts easily deactivate under hydrothermal conditions, The Ru@MC was prepared according to the literature30 with a
due to the leaching, sintering, and partial oxidation of Ru little modification. 5.0 g SiO2 was added into 10 mL of an
nanoparticles.20 We recently found that the combination of aqueous solution containing 0.07 g of RuCl3Á3H2O, and allowed
Ru/MC and ZrP was effective in the conversion of cellulose to to stand for 12 h. Then the mixture was dried at 110 1C for 12 h
sorbitol.21 However, the Ru/MC deactivated severely, and the to obtain RuCl3/SiO2. RuCl3/SiO2 was used as a hard template
yield of sorbitol was only 30.9% for the fourth use. Therefore, it and ground with sucrose and citric acid for 1 h. The mixture
is necessary to develop a new preparation strategy, which would was then placed in an oven at 100 1C for 6 h and heated to
improve the stability of Ru nanoparticles. Several attempts have 160 1C, and kept for another 6 h to obtain the catalyst precursor
been made to provide Ru with more stability, such as intensify- (Ru/SiO2@C). Then, the precursor was heated to 750 1C and
ing the bonding between the support and metal,22 or encapsu- kept for 3 h with the protection of N2. Finally, the black powder
lating metal particles in a support,23 but this is not sufficient was washed with hot sodium hydroxide solution to remove
under harsh reaction conditions. Recently, metals embedded in silica. The obtained black powder was denoted as Ru@MC.
supports have received much attention as a methodology to Note that oxidized Ru was reduced during carbonization, thus,
obtain stable catalysts.24–26 Researchers found that Ru semi- the reduction of catalysts with hydrogen or other reductants
embedded in the support could facilitate the strong metal– was unnecessary. The obtained catalyst was calcined at 450 1C
support interaction (SMSI) between Ru and the support to and 550 1C under a N2 atmosphere, and denoted as Ru@MC-450
inhibit agglomeration and leaching of Ru nanoparticles, and and Ru@MC-550
.
exhibited excellent stability in the reaction of ammonia Coated ruthenium carbon (Ru@C). RuCl3, sucrose, and
synthesis,27 hydrogenation of levulinic acid24 and benzoic citric acid were ground together for 1 h. The mixture was then
acid.28 Besides, the existence of SMSI between Ru and carbon placed in an oven at 100 1C for 6 h and heated to 160 1C, and
could also change the chemical state of Ru and improve the kept for another 6 h. Then, the black powder was heated to
hydrogenation activity.29 While such research has not been 750 1C and kept for 3 h with the protection of N2, the obtained
carried out for the direct synthesis of sorbitol from cellulose.
In this study, the use of Ru embedded in a carbon catalyst
black powder was denoted as Ru@C.
Ruthenium reference. Commercial ruthenium carbon (Ru/C,
was intended to increase the stability of the catalyst, but we Aladdin, 5 wt% Ru) was used as a reference sample.
unexpectedly found that it also improved the hydrogenation
2.2 Catalyst characterization
activity of the catalyst in the hydrolytic hydrogenation of
cellulose. Therefore, to investigate the influence of the physical Thermogravimetric analysis (TGA) was conducted on a TA Q50
morphology of Ru and chemical state of Ru@MC on the TG instrument, the sample was heated from 30 1C to 900 1C
hydrolytic hydrogenation of cellulose, Ru carbon catalysts with with a temperature ramp rate of 10 1C minÀ1 under a N2 flow.
different structures were prepared by the in situ carbothermal The nitrogen sorption isotherms of the catalyst were determined
reduction method and the impregnation method. The results by a Micromeritics ASAP 2020 apparatus with N2 (at À196 1C) as
showed that the embedded structure of Ru@MC significantly the sorbate. XRD patterns were obtained on a Thermo ESCALAB
changed the physical morphology and chemical state of Ru, 250XI diffractometer using Cu Ka (1.54 Å) with 2y between
enhancing the interaction between Ru and carbon. Ru@MC 101 and 801 with a step size of 0.021. CO chemisorption was
exhibited terrific catalytic activity and reliable reusability in the conducted on an AutoChem1 II 2920 chemisorption instrument.
hydrolytic hydrogenation of cellulose in hot compressed water. First of all, the samples were loaded into a quartz tube, and
heated to 300 1C for 2 h in H2. Then cooled to room temperature
and flushed with nitrogen for 1 h. Finally, the static adsorption
of the CO molecule was performed at 40 1C. H2-TPD was also
performed on an AutoChem1 II 2920 chemisorption instrument.
Before measuring, the sample was heated to 150 1C and kept for
2. Experimental
2.1 Catalyst preparation
Ruthenium supported on carbon (Ru/MC). 5.0 g SiO2, 10.5 g 30 min under a He atmosphere, and cooled to room tempera-
sucrose, and 1.5 g citric acid were ground together for 1 hour. ture. The sample was reduced in a 10% H2/N2 flow at 300 1C.
The mixture was placed in an oven at 100 1C for 6 h and heated Then the sample was cooled to room temperature and exposed
to 160 1C, and kept for another 6 h. Then, the yellow powder to H2 for 30 min. After being flushed with He for 2 h, the
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
temperature-programmed desorption was performed in a He was rapidly increased under the low P/P0 (o0.2), and the pore
stream at a flow rate of 30 mL minÀ1 with a temperature ramp size distribution was centered at o2 nm and 3.5 nm, suggest-
rate of 10 1C minÀ1. Transmission electron microscopy (TEM) ing that Ru/C was a hierarchically structured porous material
and high-resolution TEM (HRTEM) micrographs were recorded with micropores and mesopores. The isotherm of Ru@C was
on a JEOL 2100F instrument. X-ray photoelectron spectroscopy type-II without hysteresis loops, indicating the absence of the
(XPS) was recorded on an ESCALAB 250 X-ray photoelectron porous structure. The BET surface area, average pore diameter,
spectrometer equipped with a monochromated micro-focused and mesopore surface area of the samples are listed in Table 2.
Al X-ray source.
The specific areas of Ru@MC and Ru/MC were 866 m2 gÀ1 and
858 m2 gÀ1, and the mesopore surface areas reached 689 m2 gÀ1
and 667 m2 À1, respectively, meaning that the Ru@MC and
g
2.3 Catalyst evaluation
Ru/MC mainly had a mesoporous structure. However, the
specific area of Ru@C was very small, and mainly the external
surface. Besides, even though the specific surface area of Ru/C
The testing experiments of the catalytic activity of the catalysts
for glucose hydrogenation or cellulose conversion was per-
formed in a 100 mL stainless steel autoclave reactor agitated
with a magnetic stirring bar. 50 mL of glucose aqueous solution
(0.1 mol LÀ1) or 0.5 g of cellulose and a certain amount of catalyst
were loaded in the reactor. Then, the reactor was flushed with N2
to evacuate air and charged with H2. When the test finished, the
reactor was quickly placed into an ice bath. The liquid products
were separated by filtration. Qualitative and quantitative analysis
of liquid products were conducted on a high-performance liquid
chromatography (Waters e2695) system. The solid was collected
after reaction by filtration and washed with water and ethanol.
The catalyst was then dried for 2 h at 65 1C in a vacuum oven.
Finally, the obtained catalyst was reused to conduct new tests
under the aforementioned reaction conditions.
was up to 1027 m2
485 m2
g
À1, the mesopore surface area was only
g
À1, indicating that the dominant pore structure was
micropores. The physical structure of the catalysts had an
important effect on their catalytic performance. Ru@MC and
Ru/MC, prepared with the hard template, both exhibited
high specific surface areas and large pore sizes, which would
facilitate the large molecule transport (Table 1).
The XRD patterns of the samples are given in Fig. 2. The
existence of two weak and broad diffraction peaks at around 201
and 451 revealed that all samples have an amorphous carbon
structure.33 The XRD patterns of Ru@MC and MC were similar,
indicating that the introduction of Ru species didn’t change the
structure of MC. For Ru@MC and Ru/MC, the absence of Ru
diffraction peaks indicated the Ru nanoparticles were highly
dispersed in the two samples. It should be noted that no signal
of Ru was detected after the thermal treatment of Ru@MC at
450 1C, indicating the high stability of Ru@MC. Sucrose directly
interacted with cationic Ru during the preparation of Ru@MC, so
the abundant hydroxyl groups of sucrose could stabilize the
cationic Ru.30 Nevertheless, a weak peak was observed at 441 for
Ru/MC-450, illustrating that the Ru nanoparticles were easy to sinter
at high temperatures due to the weak interaction between Ru and
the carbon support. Ru@C was prepared by direct carbonization of
the RuCl3–sucrose mixture without SiO2 and the XRD result showed
no peak of Ru, further confirming that abundant hydroxyl groups
could stabilize Ru, inhibiting the sintering of Ru particles during
the carbonization. The above result suggested that preparation
methods had important influence on the dispersion and stability of
Ru catalysts. Ru@MC prepared by an in situ carbothermal
reduction method not only made Ru highly dispersed, but also
stabilized the Ru nanoparticles. While the Ru nanoparticles in
Ru/MC were easy to sinter at high temperatures.
3 Results and discussion
3.1 Catalyst characterization
The porosity of the samples was determined by N2 absorption–
desorption characterization. The isotherms and pore size dis-
tribution of four samples prepared by different methods are
shown in Fig. 1. Ru@MC and Ru/MC exhibited typical type-IV
isotherms with an H2 hysteresis loop, which was a typical
characteristic of mesoporous materials.31 The shape of the hysteresis
loop indicated the existence of large uniform mesopores,32 and
the pore size was centered at 10 nm. The isotherm of Ru/C was
type-IV with an H4 hysteresis loop. The amount of adsorbed N2
The TEM images of Ru catalysts prepared by different
methods and Ru@MC-450 are shown in Fig. 3. The average
particle sizes of Ru@MC (Fig. 3(a)) and Ru@MC-450 (Fig. 3(c))
Table 1 Pore features of the Ru catalysts
Samples
SBET (m2 gÀ1
)
Smeso (m2 gÀ1
)
D (nm)
Ru@MC
Ru/MC
Ru@C
Ru/C
866
858
20
689
667
—
6.5
6.8
—
Fig. 1 N2 sorption isotherms and pore size distribution of four samples.
1027
485
1.8
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. 2 XRD patterns of various Ru carbon catalysts and MC.
were 3.5 nm and 3.3 nm, respectively, indicating that Ru
nanoparticles were homogeneously dispersed on the MC and
exhibited high stability. As shown in the HRTEM image
(Fig. 3(b)), the characteristic spacing of 0.20 nm agreed well with
the Ru (101) crystallographic planes (PDF number 89-4903),
indicating that the oxidized Ru was reduced to metallic Ru0 Fig. 3 TEM images of Ru–carbon catalysts. (a and b) Ru@MC, (c) Ru@MC-450
,
(d) Ru@C, and (e) Ru/MC, (inset in (a), (c), (d), and (e) are the particle size
distribution estimated by TEM).
during the carbonization. The TEM image of Ru@C is shown in
Fig. 3(d). The average particle size of Ru@C was 2.8 nm, which
was smaller than that of Ru@MC, because sucrose stabilized
Ru more efficiently without the presence of the template.30
This result was identical to the results of XRD. The TEM image
of Ru/MC is given in Fig. 3(f). The average particle size was
2.6 nm, and the existence of large particles indicated that Ru
nanoparticles were heterogeneously dispersed on the support.
The dispersion of Ru in the samples was conducted by CO
chemisorption, and the results are listed in Table 2. The
amount of CO chemisorbed on Ru@C, Ru/C Ru@MC, and
Ru/MC is 12.5 mmol gÀ1 cat, 45.7 mmol gÀ1 cat, 25.0 mmol gÀ1
cat and 59.8 mmol gÀ1 cat, respectively. The dispersion of Ru in
Ru@C was the lowest among the four catalysts, as Ru nano-
Table 2 Ru dispersion in Ru@MC, Ru/MC, Ru/C, and Ru@C determined
by CO chemisorption
Particle size (nm)
Ru
CO uptake
Dispersion By CO
By
Samples loading (mmol gÀ1 cat) (%)
chemisorption TEM
Ru@MC 0.97
25.0
59.8
224.3
12.5
26.1
55.4
45.7
13.4
5.2
2.4
2.9
9.9
3.5
2.6
—
Ru/MC
Ru/C
1.09
4.96
0.94
Ru@C
2.8
particles were completely coated with carbon, the CO molecules Ru particle sizes of Ru@MC and Ru@C calculated by CO
could not come into contact with the Ru nanoparticles. It should chemisorption were much larger than that determined by TEM
be noted that even though Ru/C exhibited a high specific surface analysis. This was because in Ru@MC and Ru@C, Ru particles
area, the dispersion of Ru was only 45.7%, which was lower than were coated with carbon, which would inhibit the contact of
that of Ru/MC. The reason was that the Ru loading of Ru/C the CO molecule with Ru. Therefore, the dispersion and particle
was 4.96%, while the Ru loading of Ru/MC was only 1.09%. size were overestimated.28 According to a previous study,31 the
According to the nitrogen absorption–desorption and ICP-OES, embedding degree of the catalyst was obtained from the formula
the specific area and the Ru loading were almost the same, Dembedding = (1 À Ns/Nc) Â 100%, where Ns and Nc refer to
but the dispersion Ru in Ru@MC was much smaller than that in the number of metal atoms on the surface and the average
Ru/MC. Besides, according to the particle size of Ru calculated crystallite size, respectively. The Dembedding of Ru@MC was
from CO chemisorption and TEM analysis, Ru particle sizes 32.7%, indicating that the in situ carbothermal reduction
measured by two methods were almost the same, while method could successfully semi-embed Ru nanoparticles in the
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. 4 XPS spectra of Ru 3p core levels of (a) Ru/SiO2@C, Ru@MC
carbonized at 550 1C (b) and 750 1C (c), Ru@MC calcined at 450 1C (d),
and Ru/MC (e).
Fig. 5 The H2-TPD profile of Ru@MC.
30.1%, which was similar to that based on CO chemisorption.
The H2 desorption peak at 400 1C was ascribed to the Ru sites
covered by spillover hydrogen, illustrating that the SMSI between
Ru and carbon promoted the hydrogen spillover effect.37
carbon matrix. The CO chemisorption results suggested that
different preparation methods could affect the physical morphology
of the exposed Ru active site of catalysts. The Ru particles of Ru@MC
were semi-embedded in the MC. Ru nanoparticles of the three
catalysts were either semi-embedded in the carbon, or supported on
the surface of carbon, or completely coated with carbon.
3.2 Glucose hydrogenation
To study the effect of physical morphology, chemical state, and
XPS was performed to obtain information on the chemical pore size of Ru–carbon catalysts on the glucose hydrogenation,
state Ru in the samples, the analytical results are shown in glucose hydrogenation was performed separately, and the
Fig. 4. Before carbonization, the Ru 3p3/2 binding energy (BE) of results are presented in Table 3. The hydrogenation performance
the Ru/SiO2@C was deconvoluted into two peaks, which corre- of the three catalysts prepared by different methods was quite
spond to Ru4+ (463.4 eV) and Ru6+ (465.2 eV), indicating the different. Ru@MC showed the best catalytic activity, and the
oxidized Ru was the major component. After being carbonized sorbitol yield was 99.3%. Ru@C gave the lowest sorbitol yield.
at 550 1C, the BE of Ru 3p3/2 was deconvoluted into four peaks. Although the Ru nanoparticles in Ru@C were homogeneously
The peak at ca. 461.8 eV was attributed to Ru0,34 revealing that and highly dispersed on carbon, most of Ru nanoparticles were
oxidized Ru was reduced to Ru0 during the carbonization. The completely coated with carbon, and they could not be exposed to
peaks at ca. 463.4 eV and 465.2 eV were attributed to the oxidized reactants during the reaction. Ru loading and pore structure
Ru, indicating that the Ru could not be fully reduced at 550 1C. of Ru@MC and Ru/MC were almost the same, but Ru@MC
Note that a new peak at around 462.7 eV, between metallic and exhibited better hydrogenation activity than Ru/MC, which was
oxidized Ru, was observed, which is possibly because the p orbital ascribed to the special chemical state of Ru promoting the
of carbon formed a hybrid orbital with the d orbital of Ru during transport of adsorbed and dissociated hydrogen from Ru to the
the carbonization process and the BE of Ru exhibited a positive support for hydrogen spillover and hydrogenation reactions.38
shift.35 Upon being carbonized at 750 1C, the dominant peak Thus, the TOF of Ru@MC was up to 1128 hÀ1, which was almost
appeared at around 462.5 eV and the peak at B461.8 eV two times higher than that of Ru/MC. Furthermore, the hydro-
disappeared, demonstrating that the oxidized Ru was almost fully genation activity of Ru/C was as good as that of Ru@MC, but the
reduced and SMSI existed between Ru and carbon, which further selectivity to sorbitol was inferior to that of Ru@MC. This was
influenced the BE of Ru0. Therefore, it is obvious that the in situ because the microporous structure of Ru/C could easily cause side
carbothermal reduction method could affect the chemical state of reactions. The results of glucose hydrogenation demonstrated
Ru. The XPS pattern of Ru@MC-450 was similar to that of Ru@MC, that the excellent hydrogenation performance of Ru@MC was
illustrating that the chemical state of Ru didn’t changed after ascribed to its special structure, which enhanced the hydrogena-
being calcined at high temperature under a N2 flow. For Ru/MC, tion efficiency. Besides, when the catalyst was calcined at 450 1C
the peak at 461.8 eV was ascribed to Ru0, indicating that Ru3+ was and 550 1C under a N2 flow for 2 h, the catalytic activity did
completely reduced.
not significantly decline, indicating that the Ru@MC had high
The H2-TPD profile of Ru@MC is shown in Fig. 5. The peak thermal stability.
at around 200 1C was ascribed to the stoichiometric chemi-
Compared with previous works,12,15,39,40 Ru@MC enhanced the
sorption of hydrogen on Ru sites36 and the dispersion of Ru was hydrogenation efficiency and exhibited excellent hydrogenation
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
Table 3 Catalytic performance of various catalysts for the conversion of glucose to sorbitol
Yield/%
Catalysts
Ru loading/wt%
t/h
P/MPa
Conv./%
Sorbitol
Mannitol
TOF/hÀ1
Ref.
Ru@MC
Ru/MC
Ru/C
0.97
1.09
4.97
0.94
—
—
5
5
1
1
2
2
2
2
2
2
1.5
1.5
2
2
2
2
2
2
2
3
3
100
89.2
100
22.8
100
100
99.7
100
94.9
100
99.3
88.4
86.3
22.2
99.1
99.0
98.6
83.1
92.2
98.7
0.2
0.6
12.6
—
0.8
0.8
1.0
—
1128
503
210
480
—
—
—
—
—
—
This work
This work
This work
This work
This work
This work
12
15
39
40
Ru@C
Ru@MC-450
Ru@MC-550
Ru/CCD
Ru/MCM-41
Ru/NiO_TiO2
Ru/HYZ
5.5
5.5
—
0.7
2
Reaction temperature: 120 1C.
activity, thus a high yield of sorbitol could be obtained with the for cellulose hydrolysis due to its abundant acid sites, water
complete conversion of glucose under the relatively low H2 pressure. tolerance, easy preparation, and good hydrothermal stability.41
The effect of the carbonization temperature of Ru@MC The performance of Ru@MC combined with ZrP in hydrolytic
catalytic activity on glucose hydrogenation was investigated, hydrogenation of cellulose was determined, and the results are
and the results are shown in Fig. S2 (ESI†). It was obvious that given in Table 4. The hydrolytic hydrogenation of cellulose was
the carbonization temperature significantly influenced the performed at 170 1C with an H2 pressure of 5.0 MPa. For Ru/MC,
hydrogenation of glucose. When the carbonization temperature a yield of 46.4% was obtained with a 90.6% conversion of
was 550 1C, the glucose conversion and sorbitol yield were only cellulose. Although the Ru/C gave 100% conversion of cellulose,
26.1% and 23.0%, respectively. With the increase in the carbo- the yield of sorbitol was only 40.1%. Ru@MC gave the highest
nization temperature, the glucose conversion and sorbitol yield sorbitol yield with the complete conversion of cellulose, indicating
were escalated significantly. The yield of sorbitol was up to the high hydrogenation efficiency could improve the cellulose
99.3% with the complete conversion of glucose when the carbo- conversion and sorbitol yield simultaneously. As reported in
nization temperature was 750 1C. Upon further increasing the previous works,42,43 if the catalyst and cellulose were pretreated
carbonization temperature, the glucose conversion and sorbitol by mix-milling, both the crystallinity and the polymerization
yield remained constant. The above-mentioned results revealed degree of cellulose were decreased, facilitating the hydrolysis of
that under low carbonization temperature, the catalytic activity cellulose.7 Hence, when cellulose and ZrP were mix-milled
of Ru@MC was poor because Ru3+ was partially reduced, which together, the cellulose conversion and sorbitol yield reached
was confirmed by the XPS result. When carbonization tempera- 100% and 72.4% at 170 1C in just 1.5 h, respectively.
ture was higher than 750 1C, most of Ru3+ was reduced to Ru0
The results of previous studies are also listed in Table 4.2,10,14,17,18
and exhibited outstanding hydrogenation activity. Therefore, the To date, a high yield of 71.1% was obtained with the Ru/AC–SO3H
optimal carbonization temperature was 750 1C.
catalyst, but the process was time-consuming. The combination
of Ru/C with the acid catalyst was effective in the conversion of
cellulose, while the selectivity to sorbitol was poor and the
3.3 Hydrolytic hydrogenation of cellulose
In the synthesis of sorbitol from cellulose, the hydrogenolysis of products were usually the mixture of sorbitol and mannitol.
sorbitol would be dominated and the yield of sorbitol reduced In this work, Ru@MC exhibited high selectivity to sorbitol, thus
if the hydrolysis rate of cellulose was too slow.19 Thus, to a high sorbitol yield was obtained at low temperatures with a
short reaction time.
promote the hydrolysis rate, ZrP was chosen as the acid catalyst
Table 4 Catalytic performance of various catalysts for the conversion of cellulose to sorbitol
Yield/%
Sorbitol
Catalysts
Ru loading/wt%
T/1C
P/MPa
t/h
3.5
3.5
3.5
1.5
24
0.5
36
Conv./%
Mannitol
Ref.
Ru/MC
Ru/C
Ru@MC
Ru@MCa
Ru/NbOPO4
Ru/C
1.09
4.97
0.97
0.97
5
5
10
170
170
170
170
170
245
165
180
190
5
5
5
5
4
6
5
5
6
90.6
100
100
100
100
85.5
100
99
46.4
40.1
61.5
72.4
69.1
22.2
71.1
68c
2.4
13.1
3.6
2.5
1.3
9.7
—
This work
This work
This work
This work
2
10
14
17
18
Ru/AC-SO3H
Ru/Cb
5
5
2
2.5
Ru/Cd
94.7
85.5c
a
b
c
d
Cellulose and ZrP were mix-milled. H4SiW12O40 were used as the acid catalyst. C6 (sorbitol and mannitol) yield. ZrP were used as the acid catalyst.
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
Ru@MC exhibited excellent stability and there was no obvious
loss of activity even after being reused 5 times, and the selectivity
of sorbitol remained above 99% each time, indicating that the
embedded structure could enhance the stability of the catalyst.
4. Conclusion
In this study, Ru–carbon catalysts with different structures were
prepared to investigate the effect of the physical morphology
and chemical state of Ru on the conversion of cellulose. Ru
embedded in the mesoporous carbon catalyst was used for the
first time in the synthesis of sorbitol from cellulose. Compared
with the traditional Ru supported catalysts, the Ru@MC catalyst
changed the physic morphology and chemical state of Ru simulta-
neously and intensified the interaction between Ru and carbon,
improving the hydrogen spillover effect and hydrogenation effi-
ciency. Hence, the TOF of Ru@MC on the glucose hydrogenation
was much higher than that of Ru/MC. Ru@MC exhibited out-
standing activity and stability in the synthesis of sorbitol from
cellulose. The yield of sorbitol was over 70% in just 1.5 h of
reaction. Furthermore, Ru@MC could be reused at least 6 times.
Fig. 6 The recyclability of Ru@MC for cellulose conversion. Reaction
conditions: Cellulose 0.5 g, ZrP 0.9 g, Ru@MC 60 mg, H2O 50 mL, 5 MPa H2,
170 1C, 3.5 h.
3.4 Recyclability of the catalyst
The recycling experiments of Ru@MC combined with ZrP for the
hydrolytic hydrogenation of cellulose were performed under the
optimized reaction conditions and the results are shown in
Fig. 6. The cellulose conversion and sorbitol yield declined to
90.6% and 52.5% in the second run, respectively. The P leaching
of fresh ZrP under the hydrothermal reaction conditions was
responsible for the decreased sorbitol yield, which has been
confirmed in the previous studies.18 Then, the catalysts of
Ru@MC combined with ZrP remained catalytic activity in the
subsequent operations and could be used at least 6 times.
To further confirm the stability of Ru@MC, the recycling
experiments of Ru@MC in glucose hydrogenation without ZrP
were also performed and the results are shown in Fig. 7.
Conflicts of interest
There are no conflicts to declare.
Acknowledgements
The authors gratefully acknowledge the financial support from
the Sichuan Province Science and Technology Program (Grant
No. 2018JY0615).
References
1 S. Van de Vyver, J. Geboers, P. A. Jacobs and B. F. Sels,
ChemCatChem, 2011, 3, 82–94.
2 J. Xi, Y. Zhang, Q. Xia, X. Liu, J. Ren, G. Lu and Y. Wang,
Appl. Catal., A, 2013, 459, 52–58.
3 M. Dolores Adsuar-Garcia, J. Xavier Flores-Lasluisa, F. Zahra
Azar and M. Carmen Roman-Martinez, Catalysts, 2018, 8.
4 S. Esposito, B. Silvestri, V. Russo, B. Bonelli, M. Manzoli,
F. A. Deorsola, A. Vergara, A. Aronne and M. Di Serio, ACS
Catal., 2019, 9, 3426–3436.
5 N. Rey-Raap, L. S. Ribeiro, J. J. de Melo Orfao, J. L. Figueiredo
and M. F. Ribeiro Pereira, Appl. Catal., B, 2019, 256.
6 X. Zhao, J. Xu, A. Wang and T. Zhang, Chin. J. Catal., 2015,
36, 1419–1427.
7 L. S. Ribeiro, J. J. Delgado, J. J. M. Orfao and M. F. R. Pereira,
Appl. Catal., B, 2017, 217, 265–274.
8 W. Deng, X. Tan, W. Fang, Q. Zhang and Y. Wang, Catal.
Lett., 2009, 133, 167–174.
9 H. Kobayashi, Y. Ito, T. Komanoya, Y. Hosaka, P. L. Dhepe,
K. Kasai, K. Hara and A. Fukuoka, Green Chem., 2011, 13,
326–333.
Fig. 7 The recyclability of Ru@MC for glucose hydrogenation. Reaction
conditions: glucose 50 g (0.1 mol LÀ1), Ru@MC 80 mg, 120 1C, 2 MPa H2, 2 h.
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
10 C. Luo, S. Wang and H. Liu, Angew. Chem., Int. Ed., 2007, 46, 26 F. Su, L. Lv, F. Y. Lee, T. Liu, A. I. Cooper and X. S. Zhao,
7636–7639. J. Am. Chem. Soc., 2007, 129, 14213–14223.
11 G. Liang, L. He, H. Cheng, W. Li, X. Li, C. Zhang, Y. Yu and 27 Y. Zhou, Y. Ma, G. Lan, H. Tang, W. Han, H. Liu and Y. Li,
F. Zhao, J. Catal., 2014, 309, 468–476. Chin. J. Catal., 2019, 40, 114–123.
12 Z. Li, Y. Liu and S. Wu, BioResources, 2018, 13, 28 Z. Jiang, G. Lan, X. Liu, H. Tang and Y. Li, Catal. Sci.
1278–1288. Technol., 2016, 6, 7259–7266.
13 A. Onda, T. Ochi and K. Yanagisawa, Top. Catal., 2009, 52, 29 X. Liu, G. Lan, P. Su, L. Qian, T. Ramirez Reina, L. Wang,
801–807.
Y. Li and J. Liu, Catal. Today, 2020, 351, 75–82.
30 Y. Li, G. Lan, H. Wang, H. Tang, X. Yan and H. Liu, Catal.
Commun., 2012, 20, 29–35.
31 Y. Zhang, T. Chen, G. Zhang, G. Wang and H. Zhang, Appl.
Catal., A, 2019, 575, 38–47.
14 J. W. Han and H. Lee, Catal. Commun., 2012, 19, 115–118.
15 J. Zhang, L. Lin, J. Zhang and J. Shi, Carbohydr. Res., 2011,
346, 1327–1332.
16 A. Romero, A. Nieto-Marquez and E. Alonso, Appl. Catal., A,
2017, 529, 49–59.
32 S. K. Das, M. K. Bhunia, A. K. Sinha and A. Bhaumik, ACS
Catal., 2011, 1, 493–501.
17 J. Geboers, S. Van de Vyver, K. Carpentier, K. de Blochouse,
P. Jacobs and B. Sels, Chem. Commun., 2010, 46, 3577–3579. 33 K. Nakajima and M. Hara, ACS Catal., 2012, 2, 1296–1304.
18 Y. Liao, Q. Liu, T. Wang, J. Long, L. Ma and Q. Zhang, Green 34 M. Guo, G. Lan, J. Peng, M. Li, Q. Yang and C. Li, J. Mater.
Chem., 2014, 16, 3305–3312.
Chem. A, 2016, 4, 10956–10963.
19 B. F. Machado, M. Oubenali, M. Rosa Axet, T. Trang Nguyen, 35 Y.-X. Xiao, J. Ying, G. Tian, Y. Tao, H. Wei, S.-Y. Fan,
M. Tunckol, M. Girleanu, O. Ersen, I. C. Gerber and P. Serp,
Z.-H. Sun, W.-J. Zou, J. Hu, G.-G. Chang, W. Li, X.-Y. Yang
J. Catal., 2014, 309, 185–198.
and C. Janiak, Appl. Catal., B, 2019, 259.
20 Y. Zhang, M. Zhao, H. Wang, H. Hu, R. Liu, Z. Huang, 36 B. Zhang, Q. Wu, C. Zhang, X. Su, R. Shi, W. Lin, Y. Li and
C. Chen, D. Chen and Z. Feng, Bioresour. Technol., 2019,
288, 121532.
F. Zhao, ChemCatChem, 2017, 9, 3646–3654.
37 R. Prins, Chem. Rev., 2012, 112, 2714–2738.
21 G. Zhang, T. Chen, Y. Zhang, T. Liu and G. Wang, Catal. 38 S. D. Lin and M. A. Vannice, J. Catal., 1993, 143, 539–553.
Lett., 2020, 150(8), 2294–2303.
22 J. Ftouni, A. Munoz-Murillo, A. Goryachev, J. P. Hofmann,
39 D. K. Mishra, J.-M. Lee, J.-S. Chang and J.-S. Hwang, Catal.
Today, 2012, 185, 104–108.
E. J. M. Hensen, L. Lu, C. J. Kiely, P. C. A. Bruijnincx and 40 D. K. Mishra, A. A. Dabbawala, J. J. Park, S. H. Jhung and
B. M. Weckhuysen, ACS Catal., 2016, 6, 5462–5472. J.-S. Hwang, Catal. Today, 2014, 232, 99–107.
23 Z. Wei, Y. Chen, J. Wang, D. Su, M. Tang, S. Mao and 41 G. Gliozzi, A. Innorta, A. Mancini, R. Bortolo, C. Perego,
Y. Wang, ACS Catal., 2016, 6, 5816–5822. M. Ricci and F. Cavani, Appl. Catal., B, 2014, 145, 24–33.
24 Z. Wei, J. Lou, C. Su, D. Guo, Y. Liu and S. Deng, 42 H. Kobayashi, M. Yabushita, T. Komanoya, K. Hara, I. Fujita
ChemSusChem, 2017, 10, 1720–1732. and A. Fukuoka, ACS Catal., 2013, 3, 581–587.
25 H. Wang, C. Chen, H. Zhang, G. Wang and H. Zhao, Chin. 43 M. Yabushita, H. Kobayashi, K. Hara and A. Fukuoka, Catal.
J. Catal., 2018, 39, 1599–1607.
Sci. Technol., 2014, 4, 2312–2317.
New J. Chem.
This journal is © The Royal Society of Chemistry and the Centre National de la Recherche Scientifique 2020