Highly stable Ru nanoparticles incorporated in mesoporous carbon catalysts for production of γ-valerolactone

Article abstract of DOI:10.1016/j.cattod.2018.12.039

The hydrogenation of levulinic acid to γ-valerolactone with water as solvent is a crucial reaction for producing fine chemicals. However, the development of highly stable catalysts is still a major challenge. Here, we prepared a Ru nanoparticles incorporated in mesoporous-carbon (Ru-MC) catalyst to achieve high stability in acidic aqueous medium. The Ru-MC showed excellent catalytic performance (12024 h^?1 turnover frequency) in the hydrogenation of LA-to-GVL. Compared with Ru supported on mesoporous carbon catalyst (Ru/MC) prepared by conventional wet impregnation method, the Ru-MC showed excellent reusability (more than 6 times) and thermal stability (up to 600 °C). Based on H₂-TPR-MS characterization, it was proposed that the incorporated structure significantly increased the interaction between Ru nanoparticles and carbon support, which effectively prevent the leaching and sintering of Ru nanoparticles and contributed to increased high reusability and thermal stability of the Ru-MC.

Full text of DOI:10.1016/j.cattod.2018.12.039

Accepted Manuscript
Title: Highly stable Ru nanoparticles incorporated in
mesoporous carbon catalysts for production of ␥-valerolactone
Authors: Xiaoyan Liu, Guojun Lan, Panpan Su, Lihua Qian,
Tomas Ramirez Reina, Liang Wang, Ying Li, Jian Liu
PII:
S0920-5861(18)31675-4
DOI:
Reference:
https://doi.org/10.1016/j.cattod.2018.12.039
CATTOD 11856
To appear in:
Catalysis Today
Received date:
Revised date:
Accepted date:
30 October 2018
13 December 2018
17 December 2018
Please cite this article as: Liu X, Lan G, Su P, Qian L, Ramirez Reina T,
Wang L, Li Y, Liu J, Highly stable Ru nanoparticles incorporated in mesoporous
carbon catalysts for production of ␥-valerolactone, Catalysis Today (2018),
https://doi.org/10.1016/j.cattod.2018.12.039
This is a PDF file of an unedited manuscript that has been accepted for publication.
As a service to our customers we are providing this early version of the manuscript.
The manuscript will undergo copyediting, typesetting, and review of the resulting proof
before it is published in its final form. Please note that during the production process
errors may be discovered which could affect the content, and all legal disclaimers that
apply to the journal pertain.

Highly stable Ru nanoparticles incorporated in
mesoporous carbon catalysts for production of γ-
valerolactone
Xiaoyan Liu,a,b Guojun Lan,a Panpan Su,b Lihua Qian,a Tomas Ramirez Reina,c Liang Wang,d
Ying Lia* and Jian Liub,c*
a Institute of Industrial Catalysis, Zhejiang University of Technology, Hangzhou Chaowang
Road 18, 310014, PR China
b State Key Laboratory of Catalysis, Dalian Institute of Chemical Physics, Chinese Academy of
Sciences, Dalian Zhongshan Road 457, 116023, PR China
c Department of Chemical and Process Engineering, University of Surrey, Guildford, Surrey,
GU2 7XH, UK
d College of Chemical Engineering, Zhejiang University of Technology, Hangzhou Chaowang
Road 18, 310014, PR China
*Corresponding author. Tel: +81 571 88320766. E-mail address: liying@zjut.edu.cn (Y. Li);
jianliu@dicp.ac.cn; jian.liu@surrey.ac.uk (J. Liu)
1

Graphical abstract
TOF: 12024 h-1
100
50
0
1
2
3
4
5
6
Cycle
Highlights
 Ruthenium nanoparticles incorporated in mesoporous carbon catalysts for the production of
γ-valerolactone from levulinic acid in water are reported.
 Ruthenium nanoparticles incorporated in mesoporous carbon catalysts showed excellent
reusability and thermal stability.
 Leaching could be minimised by changing to incorporated structure.
 The mechanism of high reusability and thermal stability for incorporated ruthenium catalysts
are studied.
Abstract
The hydrogenation of levulinic acid to γ-valerolactone with water as solvent is a crucial
reaction for producing fine chemicals. However, the development of highly stable catalysts is
still a major challenge. Here, we prepared a Ru nanoparticles incorporated in mesoporous-carbon
(Ru-MC) catalyst to achieve high stability in acidic aqueous medium. The Ru-MC showed
excellent catalytic performance (12024h-1 turnover frequency) in the hydrogenation of LA-to-
2

GVL. Compared with Ru supported on mesoporous carbon catalyst (Ru/MC) prepared by
conventional wet impregnation method, the Ru-MC showed excellent reusability (more than 6
o
times) and thermal stability (up to 600 C). Based on H2-TPR-MS characterization, it was
proposed that the incorporated structure significantly increased the interaction between Ru
nanoparticles and carbon support, which effectively prevent the leaching and sintering of Ru
nanoparticles and contributed to increased high reusability and thermal stability of the Ru-MC.
Keywords: Mesoporous carbon, Ru nanoparticles, hydrogenation, biomass, stability
1. Introduction
With the depletion of fossil resources and concern for environment, the conversion of
biomass to chemicals and liquid fuels has attracted interest worldwide. γ-valerolactone (GVL), a
naturally occurring chemical in fruits, is a safe and value-added product that can be widely used
as liquid fuel, renewable solvent, food additives, and intermediate in the synthesis of fine
chemicals[1-3]. In the past few years, attention was highly focused on GVL obtained by
hydrogenation of levulinic acid (LA). LA also has been of interest for many years because it can
be produced cost effectively and in high yield from renewable feedstock in a new industrial
process, via tandem dehydration and hydrolysis reactions of C5 and C6carbohydrates [4-
9].However, the production of GVL from LA still use traditional and corrosive homogeneous
acid catalysis that are limited by a low yield of GVL, high energy consumption, and severe
pollution. To overcome these problems and due to heterogeneous catalysts' easy recycling and
eco-friendly properties, numerous heterogeneous catalysts have been explored and designed for
the conversion of LA. Ruthenium appears as the most active and selective metal for the
conversion of LA to GVL compared with the other metals (Ir, Rh, Pd, Pt…) [10-15]. From a
3

very detailed kinetic investigation of solvent effects in hydrogenation of LA on Ru/C catalyst,
Wang [16] pointed out the highest hydrogenation activity was observed in protic solvents.
Therefore water is the best 'green' option for its polar protic properties and vast reserves around
the world. Actually, it is well recognized that the nature of supports have an obvious influence
on the catalytic performance of the catalysts. Carbon materials, including active carbon, graphite,
etc., are excellent catalyst supports due to their inertness, non-toxicity, good water resistant,
textural and mechanical properties [17, 18].
While most studies focused on the activity and selectivity of the catalyst, the third critical
parameter of the catalytic performance, i.e. stability, has been much less investigated [19]. LA is
generally produced from lignocellulose, with a mineral acid employed as catalyst [20]. The
presence of mineral acid in LA product streams creates additional challenges to the catalyst
stability. Although supported Ru on carbon (Ru/C) catalysts have excellent initial catalytic
performance, their activity decreases sharply in a batch hydrogenation process. Wei [21] pointed
out that at least four factors were reported to be responsible for Ru/C catalysts deactivation in
water: (I) Ru leaching into the reaction bulk; (II) Ru aggregation on the carbon surface; (III)
carbon deposition on Ru nanoparticles and (IV) the phase transformation of Ru nanoparticles and
support. To improve stability, many different strategies are applied in recent studies. Jamal [22]
and Wen [20] used metallic oxide supports to increase the interaction between Ru NPs and
supports to gain higher stability. Wei [21] used embedding method to decelerate the deactivation
of Ru catalyst in methanol system. The embedded structure could prevent the Ru NPs from
migration, aggregation, and leaching during reaction process. Similarly, Wang’s group [23]
encapsulated metal nanoparticles in porous graphene layers, enhancing the stability of the
catalyst. Tan [24] prepared a stable catalyst by modifying the supports with 3-
4

aminopropyltriethoxysilane and hence enhancing the Ru-support interaction. As a result, the
ruthenium metal was bound to support's surface via coordination with the amino ligands and
valid gain stability. Li [25] and Liu [26] also found that bimetal catalyst obtained much higher
stability, mainly due to the second metal change the geometric and electronic structures thus
preventing the agglomeration and sintering of metal particles. From the above-mentioned studies
it can be concluded that there are at least three methods to enhance the catalysts’ stability: (i) by
modifying the supports with organic groups or metal oxide; (ii) by changing the catalysts’
structure and (iii) by introducing other metals to form alloy.
In recent reports, we have developed a Ru-Carbon catalyst with uniform Ru NPs
incorporated on the mesoporous carbon supports (Ru-MC)[27, 28]. And the Ru-MC catalyst
exhibited excellent catalytic performance and stability for the hydrogenation of benzoic acid with
water as solvent. The first key point of high stability of Ru-MC is that the special incorporated
structure can prevent the Ru NPs aggregation due to the confinement of the carbon matrix, on the
other hand, Ru NPs incorporated in the carbon can significantly enhance the interaction between
Ru NPs and carbon supports. Herein, we demonstrate that the Ru-MC catalyst has a significant,
positive impact on stability in the hydrogenation of LA to GVL in acidic aqueous medium. The
catalytic performance of the incorporated Ru-MC catalyst was compared to the supported
Ru/MC catalyst prepared by traditional wet impregnation method. We then propose further
insights into the enhanced stability of Ru-MC compared with Ru/MC via a series of
characterization techniques.
2. Experimental
2.1. Materials
5

Ruthenium chloride hydrate (RuCl3•xH2O) was purchased from Sino-Platinum Metals Co.
Ltd. A commercialized Nano-silica with particle size of 15 ± 5 nm and surface area of 250 ± 30
m2 g-1 was provided by Hangzhou Wanjing New Material Co. Ltd. Levulinic acid was purchased
from Aladdin Co.Ltd. Other reagents were obtained from Shanghai Chemical Reagent Inc. of
Chinese Medicine Group. All materials were analytical grade and used without any further
purification.
2.2. Preparation of incorporated Ru-MC
A typical process for preparation of the Ru-MC catalyst was described as in the reference
[27]: Firstly, 6.00 g of SiO2 was loaded with 0.18 g of RuCl3•xH2O aqueous solution by an
impregnation method for 12h. Then, 12.54 g of sucrose, 0.99 g of oxalic acid and SiO2@
RuCl3•xH2O powder were added into an agate pot with several different sizes iron balls, and
ground for 60 minutes at 150 rpm, until the solid powder turned gray. The mixed composite was
o
o
dried at 100 C for 6 h and at 160 C for 6 h, respectively. After that, the composite was
o
carbonized at 850 C for 3 h under N2 flow (50 mL min-1). The Ru-carbon-silica composite was
o
then washed with sodium hydroxide solution at 70 C twice to remove the silica template
completely. The template-free product was thus obtained (denominated as Ru-MC). It should be
noted that the Ru3+ ions can be reduced into Ru NPs during the carbonization process. For
clarity, the preparation methods for the above samples are summarized in Scheme 1.
2.3. Preparation of supported Ru/MC
The supported ruthenium catalyst on mesoporous carbon (MC) (denominated as Ru/MC)
was prepared by a wet impregnation method. The process for preparation of the MC was similar
with the Ru-MC except without addition of the RuCl3•xH2O at the first step. Then Ru/MC was
6

reduced at 400 oC for 2 h under high purityH2 flow (30 mL min-1) before being tested in levulinic
acid hydrogenation. The synthesized MC was used as support without any additional treatment.
2.4. Characterization of Catalysts
N2physisorption isotherms were recorded with a Quantachrome Autosorb-IQ apparatus
o
setup operating at -196 C. Prior to performing the measurements, the samples were outgassed
o
for 8 h at 300 C. Surface areas were determined using the Brunauer−Emmett−Teller (BET)
theory, while total pore volumes were determined from the aggregation of N2 vapor adsorbed at a
relative pressure of 0.99. The pore size distribution was acquired from the desorption branches of
the isotherms using the Barrett-Joyner-Halenda (BJH) model. Powder X-ray diffraction (XRD)
patterns were measured with a Rigaku D/Max-2500/pc powder X-ray diffractometer using Cu Kα
radiation (λ= 0.154 nm) in a 2θrange from 10 to 80o and a scan speed of 4o min-1. TEM images
were recorded on a JEOL JEM-1200EX instrument and particle size distribution of Ru
nanoparticles was performed from the TEM images using ImageJ software.
The metal loading of fresh and used Ru catalysts was analyzed by ultraviolet
spectrophotometry with an ultraviolet spectrophotometer (model 7600) from Shanghai Jinghua
Instruments Co. Ltd. A typical process was described: First, 0.10 g catalyst was calcined with 2.5
g Na2O2 under 680 oC for 20 min in a muffle furnace to oxidize the Ru0 completely. The product
was then dissolved with a mixed 40 mL concentrated HCl-Ethanol solution (volume ratio=1: 1)
o
then 10 mL thiourea aqueous solution (0.1 g mL-1) was added at 80 C for 10 min to get a blue,
clear solution. Last, the absorbance of as-prepared solution was measured with 620 nm excitation
wavelength. With external standard method, the Ru content of catalyst could be calculated.
CO chemisorption was carried out on a Quantachrome Autosorb-IQ Chemisorb apparatus.
Prior to measurements, the pre-reduced catalysts were further reduced in situ for 2 h at 400 oC in
7

o
high purtyH2. Then vacuum 2 hours, and static adsorption of CO molecule at 40 C. The metal
dispersion was estimated based on the assumption of a spherical geometry of the particles and an
adsorption stoichiometry of one CO molecule on one Ru surface atom [27].The dispersion (D) of
Ru NPs was calculated based on following equations:
푉푁푇푃
푀
퐷 =
∗ 100
푉푀푚퐶푃
Here, VNTP refers to the adsorption volume of CO at normal temperature and pressure and
VM= 22.414 L mol-1. M is the atomic mass of Ru (101.07 g mol-1). And m is the weight of
catalyst. C is the stoichiometry between Ru and CO chemisorption (CO:Ru=1:1). P is the amount
of Ru loading.
The incorporating degree of the Ru nanoparticles was estimated via an equation based on
the particle size measured via CO chemisorption and TEM characterization. Dincorporating = 100 ×
(1-Ns/Nc)% in which Ns is the number of metal atoms present on the surface measured by using
the CO chemisorption technique and Nc is the number of metal atoms present on the surface
based on the average crystallite size determined by using TEM characterization.
Hydrogen-temperature-programmed-reduction (H2-TPR-MS) of Ru catalysts were carried
out with a self-made TPD/TPR instrument. The sample (50 mg) was heated in a fixed bed U-
shaped quartz tubular reactor located inside an electrical furnace under 5 % H2/Ar flow (30 mL
min-1) and the mass spectra were collected by an on-line Hiden gas analyzer (QIC 20). The
o
temperature was raised at a rate of 10 C min−1 from 100 to 850 oC. The following mass signals
of H2, CH3 and H2O were monitored simultaneously by a quadrupole mass spectrometer: m/e = 2
(H2), 15 (CH3) and 18 (H2O) amu.
Argon-Temperature-programmed-desorption (Ar-TPD-MS) of Ru catalysts were also
carried out with the same instrument but under high purity argon. The temperature program was
8

o
o
o
raised from 100 C up to 850 C at a rate of 10 C min−1 under a flow of high purity Ar (30 mL
min−1) during the whole test program. The following mass signals of H2O and CO2 were
monitored simultaneously by the quadrupole mass spectrometer: m/e = 18 (H2O) and 44 (CO2)
amu.
2.5. Evaluation of Catalytic Activity and Reusability
The evaluation of the catalytic performance of the Ru catalysts for LA hydrogenation was
carried out in a 50 mL stainless-steel stirred pressure reactor. 18 mg of Ru catalysts and17.8
mmol of LA, were placed in the sealed reactor loaded with 20 mL of deionized water. Prior to
the liquid-phase reaction, the whole reactor was purged with H2 three times to drain air. Next, the
reaction mixture was heated to the desired temperature, and then charged with H2. The stirring
speed was 1000 rpm to remove the effects of mass transfer resistance. At the end of the reaction,
the reactor was cooled rapidly to room temperature in an ice bath, after which the remaining H2
was released. The product mixtures were identified and quantified by a SHIMADZU GC-2014
gas chromatography apparatus equipped with a AT-FFAP chromatographic column (30 m×0.32
mm×0.5 μm) and a flame ionizing detector, using authentic samples for calibration.
The reusability of the catalysts under the applied batch conditions was assessed by multiple
recycling tests. Considering that the catalysts will be lost during the recovery process, three-fold
mass (54 mg) of catalyst and levulinic acid (53.4 mmol) were placed in the reactor. The solid
o
was recovered after reaction by centrifugation followed by drying for 6 h at 80 C in vacuum
oven. Finally, the catalyst was transferred back to the reactor and a new catalytic run was
performed with the addition of a new batch of substrate under standard reaction conditions.
The turnover frequency (TOF) was calculated based on the following equations:
푛퐿퐴푐표푛푣.퐿퐴
푇푂퐹 =
푛푅푢퐷푡
9

Here, nLA and nRu refer to the LA and Ru moles. Conv.LA is the conversion of LA after
reaction. D is the dispersion of Ru from CO chemisorption and t is the reaction time in hour.
3. Results and discussion
3.1. Catalytic performance, reusability and Thermal stability
As reported previously, the conversion of LA to GVL can be conducted at moderate reaction
o
temperature in the range of 20 to 230 C. With the above concern, reaction temperature was in
o
the range of 30 to 110 C andH2 pressure was set at 4 MPa. Yield of GVL as a function of
o
various catalysts are shown in Table 1. There is nearly no LA conversion at 110 C after 1 h
without catalyst (entry 1). Similar results were obtained when the support MC was used as
catalyst (entry 2). Compared Ru/MC with Ru-MC, Ru-MC clearly displayed higher catalytic
activity than Ru/MC under identical conditions (entries 3 and 4), with a LA conversion of 64%
o
and a GVL yield of 64% at 70 C after 1 h (12024h-1 turnover frequency). Conversely, the
turnover frequency of Ru/MC was 3348h-1, which was one third of Ru-MC. When the reaction
o
temperature changed from 30 to 110 C, the LA conversion and GVL yield quickly improved
from 15% and 15% to 100% and 99%, respectively (entries 5 to 8). Clearly, the Ru-MC
significantly enhanced the catalytic activity.
The reusability of Ru-MC and Ru/MC were examined by performing multiple recycling
runs. As shown in Figure 1, a clear, continuous deactivation was observed with Ru/MC catalyst,
with a decrease in LA conversion from initial 73% to 22% after five consecutive runs, with a
reduction level of 70%. Under the same conditions, the Ru-MC showed only a small drop in
GVL yield from 98 to 90% after six cycles, with a reduction level of only 8% indicating the
superiority of the new type of Ru-MC catalyst for multiple runs.
10

o
o
To study the thermal stability of Ru-MC, the catalyst was treated at 400 C and 600 C
o
under high purity nitrogen for 2 h before evaluation (10 C min-1, 50 mL min-1). As shown in
Table 1, for Ru-MC-400 and Ru-MC-600, the catalytic performances did not decrease after
calcination (entries 9-10). These results demonstrate the high thermal stability of Ru-MC
catalyst. The above results revealed the obvious advantages of Ru-MC over a conventional
Ru/MC catalyst in terms of catalytic performance and stability.
3.2. Physicochemical properties of catalysts
The porous structures of Ru-MC and Ru/MC catalysts are analyzed with N2 sorption
techniques. For both Ru-MC and Ru/MC, the isotherms are typical type IV with an H1 hysteresis
loop (Fig. 2a), which is typical for mesoporous materials. From Figure 2b, the pore size of Ru-
MC and Ru/MC are 9.6 nm (calculated from desorption branch of the isotherm). The specific
surfaces area of Ru-MC is 961 m2 g-1 and pore volume is 1.72 cm3 g-1, while Ru/MC is 956 m2 g-
1 and the pore volume is 1.78 cm3 g-1, both catalysts present high specific surfaces areas and pore
volumes (Table 1). The similarity in textural properties among both catalysts indicates that their
different catalytic performance is not directly associated to their porous structure.
The XRD patterns of MC, Ru-MC, Ru/MC, Ru-MC-400, Ru-MC-600, Ru/MC-400 and
Ru/MC-600 catalysts are given in Figure 3 (the black lines). Two broad peaks appeared at 22o
and 43o for all catalysts, corresponding to (002) and (001) crystal planes from the graphitic
structure of the carbon supports [29, 30], which indicates that MC, Ru-MC and Ru/MC have
amorphous carbon framework. One weak peak appeared at 44.0o for Ru/MC catalyst which
correspond to (101) diffractions of bulk hexagonal Ru (Ru PDF#65-1863), because during the
o
high temperature reduction process (400 C for 2 h), the Ru nanoparticles are easily sintering.
Using Scherrer equation, we calculated the Ru particle size in Ru/MC to be 9.2 nm. No peak
11

appeared for Ru-MC of bulk hexagonal, indicating the high dispersion and small particle size of
Ru NPs in this sample. In fact these observations are consistent with our previous study [27].It
o
o
should be pointed out that despite the thermal treatment at 400 C and 600 C, respectively; no
signs of sintering were detected in the Ru-MC catalyst. But for Ru/MC-600, an obvious peak at
44.0o for Ru could be observed which indicates that the loaded Ru NPs were sintered during heat
o
treatment at 600 C. These results clearly show the Ru-MC has the successful incorporation of
Ru particles in the MC matrix and has excellent thermal stability.
The status of Ru nanoparticles was characterized by CO chemisorption and TEM, the data
and figures are given in Table 2 and Figure 4, respectively. For Ru-MC, the level of CO
chemisorption and dispersion of Ru nanoparticles were 52.0 μmol gcat-1 and 26.6%, respectively.
The dispersion of Ru nanoparticles was lower than that for Ru/MC (59.2%). Such a level of Ru
dispersion for Ru-MC is relatively low. From Table 2, the amount of CO chemisorbed on the Ru-
MC catalyst was lower by one-fold compared with Ru/MC catalyst but the particle size obtained
from TEM shows was 2.7 nm (Fig. 4a). Very likely the underestimation in the dispersion value
for Ru-MC compared with Ru/MC is due to the shielding effect of the support; i.e. part of the
ruthenium particles are buried in the framework of carbon for Ru-MC and cannot be reached by
CO molecules [27]. The estimated incorporating degree was 58.7% for Ru-MC. This indicates
that the synthesis method for preparing Ru-MC could successfully incorporate part of the Ru
nanoparticles within the framework of the carbon. As shown in Figure 4, the high dispersion and
uniform size of Ru NPs can be observed. This result is consistent with the XRD characterization
(Fig. 3).
During hard-template SiO2 etching by NaOH-water-ethanol solution process, -COOH
groups could be introduced to the surface of catalyst. Argon-temperature-programmed-
12

desorption detected by mass spectroscopy (Ar-TPD-MS) was used to investigate the surface
environment of Ru-MC and Ru/MC. The results are shown in Figure 5. Three main CO2
o
(m/e=44) desorption peaks were observed at 265, 325 and 490 C for Ru-MC (Figure 5a) and
two CO2 desorption peaks are observed at 388 and 515 oC for Ru/MC (Figure 5b). The first peak
o
at 265 C for Ru-MC could be assigned to the decomposition of –COOH, because a H2O
(m/e=18) peak could be observed at the same temperature [31]. The –COOH of Ru/MC
o
decomposed during the reduction process at 400 C. The other two CO2 peaks of Ru-MC and
Ru/MC were both ascribed to the decomposition of lactone groups and anhydride groups.
Combining the catalytic performance of Ru-MC, Ru-MC-400, Ru-MC-600 and Ru/MC with
TPD-MS, there was no direct relationship between surface groups and catalytic performance. As
a result, the high activity of Ru-MC was not caused by the organic groups present on the surface
of catalysts.
To identify the metal-support interaction between Ru NPs and the carbon framework,
hydrogen-temperature-programmed-reduction detected by mass spectroscopy (H2-TPR-MS) of
Ru-MC and Ru/MC catalysts were performed under 5% H2/Ar. The results are shown in Figure
o
6. There are two H2 (m/e=2) consumption peaks observed at 290 and 411 C for Ru-MC from
Fig. 6a. These two peaks corresponded to methane since CH3 (m/e=15) peaks were observed at
similar temperatures while no H2O (m/e=18) signal was observed within the range investigated.
Therefore, both peaks could be assigned to the methanation of carbon species with hydrogen in
presence of metallic Ru. For Ru/MC (Figure 6b), three H2 consumption peaks were observed at
185, 284 and 528 oC. Combined with the signal of H2O (m/e=18) and CH3 (m/e=15) (magnified
5 times), the first peak at 185 oC was assigned to the reduction of Ru3+ due to the signal of water
o
appearing at the same temperature. The peaks at 284 and 528 C were attributed to the
13

o
methanation of carbon. One H2O peak was observed at 528 C, which could be assigned to the
reaction between H2 and lactone group. Compared to the methanation temperature and signal
strength of Ru-MC and Ru/MC catalysts, the Ru-MC have relatively lower temperature for
occurring methanation, and higher methanation strength. This observation can be explained by
an easier methanation of the carbon support for incorporated Ru-MC catalyst than for supported
Ru/MC catalyst owing to the strong interaction between Ru NPs on the carbon support.
The strong interaction between Ru NPs and carbon matrix for Ru-MC could enhance the
catalyst's activity during reaction process. The H2moleculescan be adsorbed dissociatively on the
Ru surface and spilled over from Ru NPs onto the support [28], in addition, LA molecules are
often found to be adsorbed on the surface of the support next to the metal-support interface [32].
It has been reported that the hydrogenation of LA molecules occurs partly through the co-
adsorption of dissociative H atoms and LA molecules on the carbon support sites [33].The strong
interaction between Ru NPs and carbon support acts as a bridge which can facilitate the
migration of dissociative H atoms from the active metal to the support, which could effective
increase the concentration of dissociative H atoms on carbon support [28], thus greatly improves
the specific activity of the Ru-MC catalysts.
Asa result, Ru-MC catalyst possessed useful characteristics such as large surface area,
highly dispersed Ru nanoparticles, strong interaction between Ru NPs and carbon support and
enhanced hydrogenation capacity. Overall, the excellent catalytic performance and stability of
Ru-MC was due to its engineered structure, which promotes metal-support interaction and results
in small and well dispersed Ru clusters with outstanding hydrogenation capacity.
3.3. Insight into stability and deactivation mechanism
14

To get insight into the deactivation mechanism, the fresh and used Ru-MC and Ru/MC
catalysts were characterized by N2physisorption, TEM, XRD and TPD-MS. The N2physisorption
measurements showed a slight (Table 3), drop in surface area and pore volume from 961 to 839
m2 g-1 and 1.72 to 1.54 cm3 g-1, respectively for Ru-MC after recycling. On the other hand, the
surface area and pore volume from Ru/MC dropped from 956 to 850 m2 g-1 and 1.78 to 1.65 cm3
g-1, respectively. But the pore diameter of the used Ru-MC and Ru/MC catalysts were both 9.6
nm. That means the decrease in surface area was due to the organic molecules be absorbed to the
catalysts. The reduction in surface area was 12.7% and 11.1%, respectively for Ru-MC and
Ru/MC (calculated from [(Sfresh-Sused)/Sfresh]·100). In light of these results we can point out that
the discrepancy on the reusability is not correlated to the decreased in surface area of the studied
materials.
To study the surface properties of used catalysts, an Ar-TPD-MS experiment was
performed, as shown in Figure 7. The surface oxygenic functional groups of used Ru-MC did not
change significantly (Fig. 5a and Fig. 7a). However, comparing Figure 5b and Figure 7b, it was
o
found that a peak of CO2 appeared at 218 C for used Ru/MC. This indicates that carboxyl
groups present on the catalyst surface after reaction. However, as we demonstrated that changes
in surface carboxyl groups did not affect the activity and stability of the catalysts. The results
suggest that the deactivation of the catalysts was not caused by the change of organic groups on
surface of catalyst.
The distribution of Ru nanoparticles on fresh and used Ru-MC (recycled five times) was
characterized by high resolution TEM techniques, as shown in Figure 8. It can be observed that
after recycling, no obvious particle aggregation occurred (Figure 8a). The size distribution of
used Ru-MC (recycled five times) was still around 3.0 nm. Therefore, TEM measurements
15

performed on both fresh and used Ru-MC exclude sintering and corroborates the robustness of
Ru NPs of incorporated in the porous carbon framework.
The XRD patterns of used Ru-MC and Ru/MC are shown in Figure 3 (red lines). No
obvious diffraction peak assigned to Ru is observed for used Ru-MC, indicating absence of large
Ru NPs after reaction. This result suggests that the incorporated structure can prevent the
sintering of Ru NPs, which was also confirmed by the high resolution TEM results as shown in
Fig 8. However, the peak intensity at 44.0o for bulk Ru NPs decreased for used Ru/MC
compared to the fresh sample. This might be due to the leaching of bigger Ru NPs from the
support. This is discussed further in the next section.
The leaching of Ru catalysts can be monitored by determining either the Ru concentration
in the reaction bulk after hydrogenation or the Ru content difference between the fresh and used
Ru catalysts. In our work, the leaching of Ru was determined from the Ru content between the
fresh and used catalysts (been used one time) by spectrophotometric method. Results on Table 3
reveal that the maximum discrepancies of fresh and used catalysts, with the incorporated Ru-MC
catalyst Ru content dropping from 1.96 to 1.76% after been used, the ratio of loss Ru was 10.2%.
This loss could be attributed to Ru leaching from NPs that were not properly incorporated in the
framework of the carbon supports. Despite of the loss of Ru, the catalyst still displayed
remarkable reusability levels. The Ru content of supported Ru/MC catalyst dropped from 2.24 to
1.40% after one-time use representing an overall Ru loss of 37.5%. As shown in Figure 1, the
decline in LA conversion for Ru/MC is apparent after one-time use. Due to the large decrease in
conversion for Ru/MC after recycling five times, the reasonable conjecture can be made that the
Ru content in Ru/MC decreased from 2.24% to barely any. Hence, the deactivation of the
Ru/MC catalysts were caused by leaching of Ru, and the incorporated Ru-MC could effectively
16

prevent the leaching of Ru NPs due to the confinement of Ru NPs by carbon supports and the
strong interaction between Ru NPs and carbon supports.
4. Conclusion
A new generation of Ru stabilized on porous carbon structures with promising applications
in bio-resources hydrogenation reactions has been developed in this work. The comparison of the
catalytic performance of the incorporated Ru catalysts (Ru-MC) vs traditional Ru supported
catalysts (Ru/MC) in LA hydrogenation reveals remarkable differences. The incorporated Ru-
MC catalyst performed much better than supported Ru/MC catalysts. The Ru-MC catalyst
displayed high reusability upon recycling for six times with the LA conversion above 90%. In
addition, the Ru-MC also exhibited excellent thermal stability, maintaining high dispersion of Ru
o
NPs after calcination at 600 C for 2 h in nitrogen, no sintering and deactivation occurred.
However, the Ru/MC catalyst showed clear deactivation after recycling for five times with the
LA conversion dropping from 73% to 22%. Extensive characterization of the catalysts showed
that the deactivation of the Ru/MC catalyst was mainly due to the weak interaction between Ru
and carbon support, resulting in significant leaching of Ru NPs during recycling. As a result, the
incorporated structure of Ru-MC catalyst showed high stability for LA hydrogenation in water
system. The excellent catalytic performance achieved with the Ru-MC material is due to its
incorporated structure, which enhanced the interaction between Ru nanoparticles and carbon
support and promotes metal-support interaction resulting in small and stable Ru NPs with
outstanding hydrogenation capacity. Overall this work provides a novel strategy to design highly
efficient hydrogenation catalysts with potential impact in biomass upgrading processes.
‘Declarations of interest:
17

none’.
Acknowledgement
The financial support from Natural Science Foundation of Zhejiang Province
(LY17B030010) is gratefully acknowledged. J. L. gratefully acknowledges the support of
Chinese Government 1000 young talent plan.
References:
[1] A. Osatiashtiani, A. F. Lee, K. Wilson, J. Chem. Technol. Biotechnol., 92 (2017) 1125-
1135.
[2] X. Tang, Z. Li, X. Zeng, Y. Jiang, S. Liu, T. Lei, Y. Sun, L. Lin, ChemSusChem, 8 (2015)
1601-1607.
[3] C. B. Chen, M. Y. Chen, B. Zada, Y. J. Ma, L. Yan, Q. Xu, W. Z. Li, Q. X. Guo, Y. Fu,
RSC Adv., 6 (2016) 112477-112485.
[4] H. Li, Z. Fang, S. Yang, ACS Sustain. Chem. Eng., 4 (2016) 236-246.
[5] P. Balla, V. Perupogu, P. K. Vanama, V. R. C. Komandur, J. Chem. Technol. Biotechnol.,
91 (2016) 769-776.
[6] G. Amenuvor, B. C. E. Makhubela, J. Darkwa, ACS Sustain. Chem. Eng., 4 (2016) 6010-
6018.
[7] Z. Wei, X. Li, J. Deng, J. Wang, H. Li, Y. Wang, Mol. Catal., 448 (2018) 100-107.
18

[8] B. Banerjee, R. Singuru, S. K. Kundu, K. Dhanalaxmi, L. Bai, Y. Zhao, B. M. Reddy, A.
Bhaumik, J. Mondal, Catal. Sci. Technol., 6 (2016) 5102-5115.
[9]J. J. Bozell, L. Moens, D. C. Elliott, Y. Wang, G. G. Neuenscwander, S. W. Fitzpatrick, R.
J. Bilski, J. L. Jarnefeld, Resour. Conserv. Recy., 28 (2000) 227-239.
[10] A. S. Amarasekara, M. A. Hasan, Catal. Commun., 60 (2015) 5-7.
[11] C. Michel, P. Gallezot, ACS Catal., 5 (2015) 4130-4132.
[12] Z. P. Yan, L. Lin, S. Liu, Energy Fuel, 23 (2009) 3853-3858.
[13] T. Pan, J. Deng, Q. Xu, Y. Xu, Q. X. Guo, Y. Fu, Green Chem., 15 (2013) 2967-2974.
[14] C. Ortiz-Cervantes, J. J. García, Inorg. Chim. Acta., 397(2013) 124-128.
[15] L. E.Manzer, Appl. Catal. A: Gen., 272(2004) 249-256.
[16] L.Wang, F.S. Xiao, ChemCatChem, 6(2014) 3048-30527.
[17] Y. Boyjoo, Y. Cheng, H. Zhong, H. Tian, J. Pan, V. K. Pareek, S. P. Jiang, J.-F.
Lamonier, M. Jaroniec, J. Liu, Carbon, 116 (2017) 490-499.
[18] T. Yang, H. Ling, J.-F.Lamonier, M. Jaroniec, J. Huang, M. J. Monteiro, J. Liu, NPG Asia
Materials, 8 (2016) e240-e246.
[19] J. P. Lange, Angew. Chem. Int. Edit., 54 (2015) 13186-13197.
[20] W. Cao, W. Luo, H. Ge, Y. Su, A. Wang, T. Zhang, Green Chem., 19 (2017) 2201-2211.
[21] Z. Wei, J. Lou, C. Su, D. Guo, Y. Liu, S. Deng, ChemSusChem, 10 (2017) 1720-1732.
19

[22] J.Ftouni, A. Muñoz-Murillo, A. Goryachev, J. P. Hofmann, E. J. M. Hensen, L. Lu, C. J.
Kiely, P. C. A. Bruijnincx B. M., Weckhuysen, ACS Catal., 6 (2016) 5462-5472.
[23] Z. Wei, Y. Chen, J. Wang, D. Su, M. Tang, S. Mao, Y. Wang, ACS Catal., 6 (2016) 5816-
5822.
[24] J. Tan, J. Cui, G. Ding, T. Deng, Y. Zhu, Y.-w. Li, Catal. Sci. Technol., 6 (2016) 1469-
1475.
[25] Z. Li, M. Zuo, Y. T. Jiang, X. Tang, X. H. Zeng, Y. Sun, T. Z. Lei, L. Lin, Fuel, 175
(2016) 232-239.
[26] J. J. Liu, L. L. Zhang, J. T. Zhang, T. Liu, X. S. Zhao, Nanoscale, 5 (2013) 11044-11050.
[27] Z. Jiang, G. Lan, X. Liu, H. Tang, Y. Li, Catal. Sci. Technol., 6 (2016) 7259-7266.
[28] G. Lan, H. Tang, Y. Zhou, W. Han, H. Liu, X. Li, Y. Li, ChemCatChem, 6 (2014) 353-
360.
[29] Y. Li, G. Lan, H. Wang, H. Tang, X. Yan, H. Liu, Catal. Commun., 20 (2012) 29-35.
[30] W. Jiang, Y. Li, W. Han, Y. Zhou, H. Tang, H. Liu, J. Energy Chem., 23 (2014) 443-452.
[31] Y. Li, G. Lan, G. Feng, W. Jiang, W. Han, H. Tang, H. Liu, ChemCatChem, 6 (2014)
572-579.
[32] A. M. Ruppert, M. Jędrzejczyk, O. Sneka-Płatek, N. Keller, A. S. Dumon, C. Michel, P.
Sautet, J. Grams, Green Chem., 18 (2016) 2014-2028.
20

[33] A. S. Piskun, H. H. van de Bovenkamp, C. B. Rasrendra, J. G. M. Winkelman, H. J.
Heeres, Appl. Catal. A: Gen., 525 (2016) 158-167.
Figure Captions:
Scheme 1. Schematic illustration of the preparation of incorporated Ru-MC catalyst.
Fig. 1. Reusability results of Ru-MC and Ru/MC. Reaction conditions: LA (53.4 mmol), water
(20 mL), catalyst dosage 5000 with respect to LA (molar ratio), 110 oC, 4 MPa, 1000 rpm, and 1
h.
Fig. 2. (a) N2 adsorption-desorption isotherm and (b) pore size distribution of Ru-MC and
Ru/MC catalysts.
Fig. 3. XRD patterns of MC, fresh Ru-MC, fresh Ru/MC, fresh Ru-MC-400, fresh Ru-MC-600,
fresh Ru/MC-400, fresh Ru/MC-600, used Ru-MC and used Ru/MC.
Fig. 4. The TEM (a) (inset is particle size distribution of Ru NPs) and HRTEM (b) images for
fresh Ru-MC catalysts. White circles points Ru NPs.
Fig. 5. Ar-TPD-MS profiles of fresh Ru-MC (a) and Ru/MC (b) catalysts.
Fig. 6. H2-TPR-MS profiles of fresh Ru-MC (a) and Ru/MC (b) catalysts.
Fig. 7. Ar-TPD-MS profiles of used Ru-MC (a) and used Ru/MC (b) catalyst.
Fig. 8. The TEM (a) (inset is particle size distribution of Ru NPs) and HRTEM images (b) for
used Ru-MC catalysts. White circles points Ru NPs.
21

Ru-MC
Ru/MC
100
80
60
40
20
0
1
2
3
4
5
6
Cycles
Fig. 1.
a
b
9.6 nm
Ru-MC
Ru-MC
Ru/MC
Ru/MC
0.0 0.2 0.4 0.6 0.8 1.0
P/P0
0
20 40 60 80
Pore Diameter (nm)
Fig. 2.
22

Ru (101)

MC
Ru-MC
Ru-MC-used
Ru/MC
Ru/MC-used
Ru-MC-400
Ru-MC-600
Ru/MC-400
Ru/MC-600
10 20 30 40 50 60 70 80
2 Theta (degree)
Fig. 3.
a
b
40
20
0
20 nm
5 nm
0
1
2
3
4
5
Particle size (nm)
Fig. 4.
23

a
b
m/e=18
m/e=18
m/e=44
265
388
325
515
m/e=44
490
200 400 600 800
Temperature (oC)
200 400 600 800
Temperature (oC)
Fig. 5.
a
b
m/e=2
m/e=2
284
185
528
m/e=18
411
m/e=18
m/e=15
290
m/e=15
5
200
400
600
800
200
400
600
800
Temperature (oC)
Temperature (oC)
Fig. 6.
24

a
b
m/e=18
m/e=18
380
210
507
m/e=44
218
342
400
536
m/e=44
200 400 600 800
200
600
800
o
o
Temperature ( C)
Temperature ( C)
Fig. 7.
a
b
40
20
0
20 nm
5 nm
5
0
1
2
3
4
Particle size (nm)
Fig. 8.
25

Ru3+
Ru3+
Ru3+
Ru3+
Ru3+
Ru3+
Ru3+
Ru3+
Ru3+
RuCl3
Ru3+
Ru3+
Ru3+
Ru3+ Ru3+
Ru3+
Ru3+
Ru3+
Ru3+
Impregnation
Ru3+
Ru3+
Ru3+/SiO2
Ru3+/SiO2@sucrose
SiO2 nanosphere
Carbonization
at850 oC, N2
SiO2 nanosphere
Sucrose
Ru nanoparticle
Carbon
Reduction with C
SiO2 etching
Ru0/SiO2@C
Ru-MC
Scheme 1.
Table 1. Texture and catalytic properties of Ru-MC and Ru/MC catalysts in the hydrogenation
of LA.[a]
Yield.
GVL
(%)
1
TOF
S.A.
P.D. Ru
P.V.
T
Conv.
LA (%)
content
Entry catalysts
(cm3
g-1) [c]
[e]
(m2 g-
(nm)[
(oC)
1
d]
[b]
)
(wt %)
(h-1)
1
2
3
4
5
6
7
8
9
No catalyst
MC
-
-
-
-
110
110
70
1
-
935
956
961
1.74
1.72
1.78
9.6
9.6
9.6
-
3
3
-
Ru/MC
Ru-MC
2.24
1.96
39
64
15
37
93
100
64
63
39
3348
70
63
12024
30
15
-
50
37
-
90
92
-
110
70
99
-
Ru-MC-400 977
1.83
1.80
9.6
9.6
1.96
1.96
63
11988
11916
10 Ru-MC-600 964
70
62
[a] Reaction conditions: LA (17.8 mmol), H2O (20 mL), 1000 rpm, PH2 4MPa, reaction time 1
h, the catalyst dosage was defined as the molar ratio (5000) between LA and Ru.
[b] BET surface area.
26

[c] Total pore volume.
[d] Pore diameter calculated from desorption branch of the isotherm.
[e] The turnover frequency in our work was calculated based on the surface Ru determined by
CO chemisorption.
Table 2. Dispersion of ruthenium NPs determined by CO chemisorption of Ru-MC and Ru/MC.
CO-chemisorption
TEM
Catalysts
CO
Ru
Dincorpor
ating (%)
Dispers
ion
Dispersion
(%)
Particle
Uptake
S. A.
Size (nm)a
(%)
(µmolg-1) (m2g-1)
Ru-MC
Ru/MC
52
97
26.6
59.2
49.3
-
2.7
-
58.7
-
116
215
Table 3. Texture properties and Ru content of used Ru-MC and used Ru/MC catalysts.
Used
S.A.
P.V.
P.D
Ru
Ru loss ratio
(%)
Ru
(m2 g- (cm3 g-1) (nm)
(wt.%
)
Used
Catalysts
1
[a]
[b]
[c]
)
(wt.%)
1.76
1.40
Ru-MC
Ru/MC
839
850
1.54
1.65
9.6
9.6
1.96
2.24
10.2
37.5
[a] BET surface area.
[b] Total pore volume.
[c] Pore diameter calculated from desorption branch of the isotherm.
27