Enhancing the catalytic activity of Ru NPs deposited with carbon species in yolk-shell nanostructures

Article abstract of DOI:10.1039/c6ta03576k

The synthesis of metal NPs with a well-defined size, shape and composition provides opportunities for tuning the catalytic performance of metal NPs. However, the presence of a stabilizer on the metal surface always blocks the active sites of metal NPs. Herein, we report an efficient method to remove the stabilizer on the metal surface via H₂ pyrolysis with Ru-poly(amindoamine) encapsulated in silica-based yolk-shell nanostructures as an example. The CO uptake amount of Ru NPs increases sharply after H₂ pyrolysis, indicating that the exposure degree of Ru NPs is increased. No aggregation of the colloidal Ru NPs occurs after H₂ pyrolysis, which could be mainly assigned to the protection effect of C and N species formed on Ru NPs. The overall activity of Ru NPs in the yolk-shell nanostructure after the pyrolysis could reach as high as 20 300 mmol per mmol Ru per h in the hydrogenation of toluene, which is much higher than that of most reported Ru-based solid catalysts. It was found that the yolk-shell nanostructure could efficiently prevent the leaching of Ru NPs during the catalytic process. Ru NPs in the yolk-shell nanostructure could also catalyze the hydrogenation of benzoic acid and Levulinic acid with high activity and selectivity.

Full text of DOI:10.1039/c6ta03576k

Journal of
Materials Chemistry A
View Article Online
View Journal
PAPER
Enhancing the catalytic activity of Ru NPs
deposited with carbon species in yolk–shell
nanostructures†
Cite this: DOI: 10.1039/c6ta03576k
ab
c
ab
a
a
a
Miao Guo, Guojun Lan, Juan Peng, Mingrun Li, Qihua Yang* and Can Li*
The synthesis of metal NPs with a well-defined size, shape and composition provides opportunities for
tuning the catalytic performance of metal NPs. However, the presence of a stabilizer on the metal
surface always blocks the active sites of metal NPs. Herein, we report an efficient method to remove the
stabilizer on the metal surface via H
yolk–shell nanostructures as an example. The CO uptake amount of Ru NPs increases sharply after H
pyrolysis, indicating that the exposure degree of Ru NPs is increased. No aggregation of the colloidal Ru
NPs occurs after H pyrolysis, which could be mainly assigned to the protection effect of C and N
2
pyrolysis with Ru–poly(amindoamine) encapsulated in silica-based
2
2
species formed on Ru NPs. The overall activity of Ru NPs in the yolk–shell nanostructure after the
pyrolysis could reach as high as 20 300 mmol per mmol Ru per h in the hydrogenation of toluene,
which is much higher than that of most reported Ru-based solid catalysts. It was found that the yolk–
shell nanostructure could efficiently prevent the leaching of Ru NPs during the catalytic process. Ru NPs
in the yolk–shell nanostructure could also catalyze the hydrogenation of benzoic acid and Levulinic acid
with high activity and selectivity.
Received 29th April 2016
Accepted 14th June 2016
DOI: 10.1039/c6ta03576k
www.rsc.org/MaterialsA
Compared with the traditional impregnation method, the
synthesis of metal NPs based on a rational design may provide
Introduction
Inspired by the sustainability, environmental legislation and more opportunities for tuning the catalytic properties of metal
6,7
energy saving, it is highly desirable for chemical reactions to NPs. Many strategies have been developed for the synthesis of
proceed with high selectivity and conversion under mild reac- discrete metal clusters with precise size control. Metal NPs have
tion conditions. Metal nanoparticles (NPs) with a controllable been supported and stabilized by mesoporous shells to improve
size have been widely used in heterogeneous catalysis due to the catalytic activity since mesoporous shells can allow reac-
their high catalytic activity and selectivity derived from high tants to diffuse through pore channels while preventing the
1–3
8–19
surface energy and large surface-to-volume ratio.
leaching of metal NPs.
But the size and morphology control
It is well known that downsizing the particle size of metal of metal NPs is still not easy. Another alternative approach is to
NPs could dramatically enhance their catalytic activity in most use a discrete macromolecular ligand such as poly(amindo-
cases owing to the increased concentration of low coordination amine) dendrimer (PAMAM) for the size-controlled synthesis of
4
,5
20
sites and surface vacancies. However, decreasing the particle NPs at the nanometre scale. Inevitably, weak or strong stabi-
size down to several nanometers not only makes the surface lizers are always employed for avoiding the aggregation of metal
atoms chemically active, but also makes the metal NPs NPs. However, the stabilizer on the metal surface could cap the
21
unstable. Thermodynamically, “naked” NPs easily tend to active sites. To solve this problem, metal NPs with a stabilizer
aggregate into larger particles, which results in a remarkable are generally supported on a solid support or encapsulated with
decrease in catalytic activity. The synthesis and stabilization of a metal oxide shell, followed by removing the stabilizer via
22
23
24
25
NPs is of extreme importance for their application in catalysis. thermal, chemical, UV-ozone, and plasma treatments.
However, the strong interaction of the metal NPs and support
a
could not be avoided; otherwise, aggregation of metal NPs may
State Key Laboratory of Catalysis, Dalian Institute of Chemical Physics, Chinese
Academy of Sciences, 457 Zhongshan Road, Dalian 116023, China. E-mail: occur.
yangqh@dicp.ac.cn; canli@dicp.ac.cn; Fax: +86-411-84694447; Tel: +86-411-
2
Based on the previous understanding that H pyrolysis could
decompose organic stabilizers, we tried to remove the stabilizer
on metal NPs via the H2 pyrolysis method with Ru–poly-
8
b
4379552; +86-41184379070
University of Chinese Academy of Sciences, Beijing 100039, China
c
Institute of Industrial Catalysis, Zhejiang University of Technology, Hangzhou,
(amindoamine) as an example. Our strategy involves the
3
†
1
10014, China
encapsulation of dendrimer PAMAM (G4-OH) stabilized Ru NPs
Electronic supplementary information (ESI) available. See DOI:
0.1039/c6ta03576k
(ꢀ2 nm) in the hollow interiors of silica-based yolk–shell
This journal is © The Royal Society of Chemistry 2016
J. Mater. Chem. A

View Article Online
Journal of Materials Chemistry A
Paper
nanostructures via a sol–gel coating method. The yolk–shell
nanostructure was chosen in order to prevent the leaching of Ru rate of 30 mL min at the desired temperature for 2 h with
NPs during the catalytic process. The subsequent heat treat- a heating rate of 1 C min . The sample aer pyrolysis was
ment in the hydrogen atmosphere could remove most of the denoted as Ru@C@YSN-T, where T (150, 200, and 500) refers to
PAMAM on Ru NPs. The characterization results suggest that no the pyrolysis temperature.
aggregation of Ru NPs occurs aer pyrolysis, which is mainly
PA–Ru@YSN was pyrolysed in a H
2
atmosphere with a ow
ꢁ
1
ꢂ
ꢁ1
attributed to the protection effect of C and N residues formed Catalytic tests
2
during the H pyrolysis process on the Ru surface.
Hydrogenation of toluene: in a typical process, a desired
ꢁ
3
amount of solid catalyst (3 ꢃ 10 mmol Ru) was added into an
ampoule tube, followed by the addition of toluene (30 mmol)
Experimental section
Chemicals and reagents
(
S/C ¼ 10 000). The ampoule tube was loaded into a stainless
steel autoclave (300 mL) with a thermocouple probed detector.
All reagents were of analytical grade and used as received Aer purging with hydrogen six times, the nal pressure was
ꢂ
without any further purication. Tetraethoxysilane (TEOS), adjusted to 2 MPa, and the reactor was heated to 323 K (50 C)
cetyltrimethylammonium chloride (CTAC) and triethanolamine with vigorous stirring. Aer reaction, the solid catalyst was
were obtained from Sinopharm Chemical Reagent Co., Ltd separated by centrifugation and the ltrate was collected,
(Shanghai, China). Cetyltrimethylammonium bromide (CTAB), diluted with ethanol and analyzed using an Agilent 7890 GC
1
,2-bis-(trimethoxysilyl)ethane (BTME), and RuCl $3H O were equipped with an Agilent J&W GC HP-5 capillary column (30 m
3
2
purchased from Sigma-Aldrich Company Ltd (USA). A uoro- ꢃ 0.32 mm ꢃ 0.25 mm).
carbon surfactant, FC-4, was bought from Yick Vic Chemicals
For recycling the catalyst, the liquid was decanted aer
(
(
Hong Kong). Poly(amindoamine) (PAMAM) dendrimer G4-OH centrifugation, and the residual catalyst was dried under
fourth-generation, hydroxyl-terminated) in methanol was vacuum at room temperature and used directly for the next run.
purchased from Weihai ChenYuan Molecule of New Materials
Co., Ltd. Other reagents were used as received from Shanghai Stability test using a xed-bed reactor
Chemical Reagent, Inc. of the Chinese Medicine Group. An
In the xed-bed experiment, 260 mg of Ru@C@YSN-200 catalyst
was packed into a commercial tubular glass mini-reactor
aqueous solution of dendrimer (PAMAM-G4-OH) stabilized Ru
nanoparticles (PA–Ru NPs) was prepared according to the
(
5.5 mm diameter ꢃ 10 mm length). The system was degassed
26
literature method. Dendritic SiO
2
nanospheres with a particle
ꢁ1
with H
2
at 5 mL min ow rate for 1 h and then a degassed
size of 130 nm were synthesized based on the literature
solution of toluene in EtOH was allowed to ow through the
27
method.
ꢁ1
catalytic bed at a constant 0.614 mL min rate. Aer 1 h and
2
together with the substrate solution, H was allowed to ow
ꢁ1
Synthesis of the solid catalyst with Ru@C NPs encapsulated in
through the catalyst at a constant ow rate of 5 mL min at
yolk–shell nanospheres (Ru@C@YSN)
room temperature. And then, the column was introduced into
an oven at 62.6 C. Aer the system reaching steady state, the
reaction products were collected at regular intervals and ana-
lysed using a precalibrated gas chromatograph.
ꢂ
In the initial step, PA–Ru was encapsulated in the hollow inte-
riors of silica-based yolk–shell nanospheres via a sol–gel
28,29
method.
Typically, 80 mg of dendritic SiO nanospheres with
2
a particle size of 130 nm was added to 16 mL of PA–Ru aqueous
ꢁ
1
Characterization
solution (2 mmol L of Ru) and the mixture was sonicated for
one hour for the full adsorption of the PA–Ru onto SiO nano-
spheres. Then, 8 mL of deionized water, 0.15 g of CTAB and ASAP2020 volumetric adsorption analyzer. Samples prepared in
2
2
N sorption isotherms were recorded on a Micromeritics
ꢂ
ꢂ
0
.5 mL of NH $H O (25 wt%) were added successively at 38 C. this work were degassed at 120 C for 6 h prior to the
3 2
Aer stirring for 10 min, an aqueous solution (40 mL) and measurements. The Ru content was determined by PLASAM-
ethanol (20 mL) were added and the mixture was stirred for SPEC-II inductively coupled plasma atomic emission spec-
another 30 min. Then, TEOS (250 mL) was added and the trometry (ICP-AES). The thermogravimetric analysis (TGA) was
mixture was stirred for 2 h, followed by the addition of an performed using a NETZSCH STA 449F3 analyzer from 30 to
ꢂ
ꢂ
ꢁ1
aqueous solution (1.5 mL) containing FC4 (0.02 g), CTAB 900 C with a heating rate of 10 C min under air atmosphere.
ꢂ
(0.04 g), and NH
3
$H
2
O (100 mL, 25 wt%) at 45 C. Finally, 180 mL X-ray powder diffraction (XRD) patterns were recorded on
of BTME in ethanol solution (1 mL) was added under vigorous a Rigaku RINT D/Max-2500 powder diffraction system using Cu
stirring. Aer stirring for 1.5 h, the temperature was increased Ka radiation (l ¼ 0.1541 nm). X-ray photoelectron spectroscopy
ꢂ
to 80 C and maintained for 1 h, and the powder product was (XPS) was performed on VG ESCALAB MK2 apparatus using Al
collected by ltration and dried at room temperature. To Ka (h
remove the surfactant, the as-synthesized materials (1 g) were the spectra were tted by using the shareware program XPS-
l
¼ 1486.6 eV) as the excitation light source. The peaks in
dispersed in a solution of 80 mg of ammonium nitrate and PEAK with Gaussian–Lorentzian peak shapes and a Shirley
ꢂ
1
3
20 mL of ethanol (95%). The mixture was heated at 60 C for background. Transmission electron microscopy (TEM) was
0 min and the process was repeated three times. The powdered performed using a HITACHI HT7700 at an acceleration voltage
product was denoted as PA–Ru@YSN.
of 100 kV. SEM was performed by using a FEI Quanta 200F
J. Mater. Chem. A
This journal is © The Royal Society of Chemistry 2016

View Article Online
Paper
Journal of Materials Chemistry A
scanning electron microscope operating at an acceleration
voltage of 1–30 kV. The dispersion of Ru was obtained by using
ꢂ
the CO chemisorption technique, which was performed at 40 C
on a Quantachrome Autosorb-1/C chemisorb apparatus. Before
measurements, pre-reduced catalysts were reduced in situ for
ꢂ ꢂ ꢂ
2
h at 150 C, 200 C, and 500 C under a H atmosphere for
2
Ru@C@YSN-150, Ru@C@YSN-200, and Ru@C@YSN-500,
respectively. The metal dispersion and particle size were esti-
mated on the basis of the assumption of a spherical geometry of
Ru particles and an adsorption stoichiometry of one CO mole-
cule on one Ru surface atom. The average Ru particle size was
calculated by using the equation dD ¼ 6Mrsite/(rmetal/N), where
d is the Ru particle size, D is dispersion, M is the molecular mass
Fig. 1 (A) TEM images (scale bar: 50 nm) and (B) HAADF-STEM images
of (a) PA–Ru@YSN, (b) Ru@C@YSN-150, (c) Ru@C@YSN-200 and (d)
Ru@C@YSN-500 (inset in (B) is the particle size distribution estimated
by STEM).
ꢁ
1
(
for Ru: 101.07 g mol ), rsite is the Ru surface site density (for
ꢁ
2
Ru: 16.3 atoms nm ), rmetal is the metal density (for Ru: 12.3
ꢁ
3
23
ꢁ1
g cm ), and N is the Avogadro number (6.022 ꢃ 10 mol ),
30
giving d ¼ 1.33/D (nm). FT-IR spectra were recorded on
a Nicolet Nexus 470 IR spectrometer (KBr pellets were
prepared).
interesting to mention that no obvious aggregation of Ru NPs
was observed and the size distribution of Ru NPs is centralized
at 2.1 nm for all samples although the percentage of particle
size larger than 2.1 nm increases as the pyrolysis temperature
increases. No diffraction peaks could be observed in the wide
angle XRD patterns of PA–Ru@YSN and Ru@C@YSN-T
(Fig. S3†), indicating that the particle size of Ru NPs is less than
Results and discussion
Preparation of solid catalysts
31
4
nm. This further conrms the HAADF-STEM results.
The synthesis procedure for the Ru@C@YSN catalyst is out-
lined in Scheme 1. First, PAMAM-G4-OH stabilized Ru NPs with
a particle size of 2.1 nm (Fig. S1†) were supported on dendritic
silica nanospheres with a pore diameter of 7.3 nm. Then a silica
layer was coated around silica nanospheres by the sol–gel
coating method for the formation of yolk–shell nano-
The FT-IR spectra of PA–Ru@YSN and Ru@C@YSN-T
samples are displayed in Fig. 2. PA–Ru@YSN clearly shows
ꢁ
1
ꢁ1
vibration peaks at 1653 cm and 1261 cm from PAMAM and
2
ꢁ
1
2 2
925–2852 cm assigned to –CH CH –, showing the successful
encapsulation of Ru NPs and the formation of an ethane–silica
shell. As the pyrolysis temperature increases, the intensity of the
vibration peaks decreases due to the decomposition of the
PAMAM dendrimer. It should be mentioned that Ru@C@YSN-
28,29
spheres.
The subsequent pyrolysis in a H2 atmosphere
results in the formation of the Ru@C@YSN-T catalyst (T refers
to the pyrolysis temperature). The monodisperse nanospheres
with a particle size of 220 nm could be clearly observed in the
SEM images of PA-Ru@YSN (the sample before pyrolysis) and
all Ru@C@YSN-T catalysts (Fig. S2†). The TEM images of PA–
Ru@YSN and Ru@C@YSN-T catalysts clearly show the yolk–
shell nanostructure with a particle size of 200–250 nm, shell
thickness of 25 nm and core size of 130 nm (Fig. 1). Based on
HAADF-STEM images, Ru NPs with a particle size of ca. 2.1 nm
are uniformly distributed in the core region for all samples
500 still affords vibration peaks assigned to PAMAM and
–
CH CH –, suggesting that the surface of Ru NPs has PAMAM
2
2
residues even aer pyrolysis at high temperatures. A control
sample, Ru@YSN-500, was also prepared by pyrolysis of
ꢂ
PA–Ru@YSN at 500 C in air (Fig. 2e). No vibration peaks of
(Fig. 1).
Ru@C@YSN-T catalysts have an almost identical yolk–shell
nanostructure to the parent material, indicating that the yolk–
shell nanospheres are stable during the pyrolysis process. It is
Scheme
1 General procedure for the preparation of yolk–shell
nanoreactors via a direct encapsulation method using PAMAM–Ru as Fig. 2 FT-IR spectra of (a) PA–Ru@YSN, (b) Ru@C@YSN-150, (c)
metal NPs.
Ru@C@YSN-200, (d) Ru@C@YSN-500 and (e) PA–Ru@YSN-500 (air).
This journal is © The Royal Society of Chemistry 2016
J. Mater. Chem. A

View Article Online
Journal of Materials Chemistry A
Paper
3
5,36
organic moieties could be found, showing that the organic respectively.
But Ru@C@YSN-500 only gives one obvious
ꢂ
groups of dendrimers and the shell are removed completely. weight loss at 450–700 C, showing that Ru@C@YSN-500 is very
The above results are consistent with a previous report that stable at high temperatures probably due to the removal of
ꢂ
PAMAM could be removed by heat treatment in O at 630 C but unstable species during the high temperature pyrolysis process.
2
some organic residues will be formed on the metal surface in Also, the weight loss of Ru@C@YSN-500 is much less than that
32
a H
The BET surface area of all solid catalysts was characterized dues is le in Ru@C@YSN-500.
using N sorption isotherms (Table 1, Fig. S4†). All Ru@C@YSN- XPS analysis was used to probe the chemical valence state of
2
atmosphere.
of other samples, suggesting that only a small amount of resi-
2
T samples show a higher BET surface area and pore volume Ru and surface carbon species on Ru NPs. Considering the fact
than PA–Ru@YSN, which is due to the decomposition of PA by that PA–Ru@YSN and Ru@C@YSN-T samples have a shell
pyrolysis in H . Ru@C@YSN-200 and Ru@C@YSN-500 with thickness of 25 nm, it is impossible to detect the Ru NPs
2
comparable textural parameters have a higher BET surface area directly. Thus, a control sample, PA–Ru/SiO with a Ru content
2
and larger pore volume than Ru@C@YSN-150, probably due to of 1.2 wt%, was prepared by impregnating PA–Ru onto dendritic
the higher decomposition degree of dendrimers at higher SiO
temperatures. All samples have two sets of mesopores at 1.8 nm yolk–shell nanospheres). PA–Ru/SiO
and 3.4 nm, assigned respectively to the pore in the shell and in tical conditions to PA–Ru@YSN for affording Ru@C/SiO
the core. The shell has a smaller pore diameter than the particle (T refers to treatment temperature). With Ru@C/SiO -500 as the
2
nanospheres (the same material used for the synthesis of
was treated under iden-
-T
2
2
2
size of Ru, indicating that the yolk–shell nanostructure could representative sample, the Ru particle size of Ru@C/SiO -T was
2
efficiently prevent the leaching of Ru NPs during the catalytic analyzed by the TEM technique. The TEM result suggests that
process.
Ru@C/SiO -500 almost has a similar Ru particle size to
2
The Ru dispersion was measured by CO chemisorption Ru@C@YSN-500 (Fig. S6†). As the Ru 3d core level overlaps with
Table 1). No CO could be adsorbed on PA–Ru@YSN due to the the C 1s core level region, the Ru 3p core level is used. Ru 3p
(
coverage of PAMAM on the surface of Ru NPs. The CO mono- core-level spectra of PA–Ru/SiO
2
before and aer treatment are
samples before and aer
mmol g along with the pyrolysis temperature from 150 to pyrolysis, the Ru 3p energies are similar and could be decon-
layer uptake of Ru@C@YSN increases from 33.5 to 107.9 shown in Fig. 3. For all PA–Ru/SiO
2
ꢁ
1
ꢂ ꢂ
00 C. The exposed Ru surface area increases sharply from voluted to Ru, RuO and RuO . PA–Ru/SiO treated at 150 C has
2 3 2
5
Ru@C@YSN-150 to Ru@C@YSN-500. The Ru dispersity of exactly the same Ru 3p BE as the parent sample. A slight
Ru@C@YSN-200 and Ru@C@YSN-500 could reach 70–75%, downward shi in BE for Ru 3p BE levels by 0.1 eV is observed
demonstrating that Ru NPs on these two catalysts have a rela-
tively clean surface. Ru@C@YSN-150 has lower Ru dispersity,
probably due to the fact that most of the Ru is covered by
PAMAM residues due to the low pyrolysis temperature. The CO
chemisorption results clearly show that pyrolysis in H
temperatures above 200 C could efficiently increase the expo-
2
at
ꢂ
sure degree of Ru NPs. Similar to CO chemisorptions, H uptake
2
amounts increase along with pyrolysis temperature (Table S1†).
The Ru dispersion degree measured by H
tendency to that measured by CO uptake although a relatively
lower dispersion degree was obtained based on
2
uptake is in a similar
H
2
33,34
chemisorption.
The thermal stability of catalysts was investigated by TG
analysis (Fig. S5†). PA–Ru@YSN, Ru@C@YSN-150 and
Ru@C@YSN-200 exhibit similar TG curves with two weight loss
steps at 280–450 C and 450–700 C due to the decomposition of
PAMAM (dendrimer) and ethane fragments of the outer shell, SiO -500.
Fig. 3 XPS spectra of (A) Ru 3p, (B) C 1s and (C) N 1s core levels of (a)
ꢂ
ꢂ
2 2 2
PA–Ru/SiO , (b) Ru@C/SiO -150, (c) Ru@C/SiO -200 and (d) Ru@C/
2
Table 1 Physical parameters and CO chemisorption capacity of PA–Ru@YSN and Ru@C@YSN-T
CO chemisorption
BET surface area
Pore diameter
(nm)
Pore volume
Dispersion
(%)
2
ꢁ1
3
ꢁ1
ꢁ1
Samples
(m g
)
(cm g
)
Uptake (mmol g
)
PA–Ru@YSN
581
709
854
807
1.8, 3.3
1.8, 3.4
1.8, 3.4
1.8, 3.4
0.74
0.89
1.01
0.92
—
—
Ru@C@YSN-150
Ru@C@YSN-200
Ru@C@YSN-500
33.5
96.3
107.9
27.3
70.6
75.8
J. Mater. Chem. A
This journal is © The Royal Society of Chemistry 2016

View Article Online
Paper
Journal of Materials Chemistry A
ꢂ
for samples pyrolyzed at temperature $200 C, which is indicate that most of the dendrimers could be removed by high
possibly due to the increase in the crystallinity of Ru. The energy temperature pyrolysis in H . But the surface of Ru is still covered
2
0
d+
position of Ru 3p could be used to calculate the Ru /Ru ratio by some C and N species even at high pyrolysis temperature.
in the samples. Aer peak deconvolution, Ru 3p spectra were These C and N species may prevent Ru NP aggregation during
0
4+

tted using three peaks, corresponding to Ru (461.6 eV), Ru
the pyrolysis process.
6
+
0
d+
(
463.4 eV) and Ru (466.5 eV). The Ru /Ru ratio increases
from 0.71 to 1.18 along with the pyrolysis temperature, showing
that the reduction degree of Ru increases at high pyrolysis
temperature (Table 2).
Hydrogenation reaction
The catalytic performance of solid catalysts was tested in the
As the dendrimer itself contains a number of chemically hydrogenation reactions using toluene as the model substrate
different carbons, the C 1s region of the XPS spectra is (Table 3). No product could be detected with PA–Ru and PA–
complicated. Small peaks are observed in the 280–282 eV region Ru@YSN as catalysts. Similarly, no product was observed for the
from the Ru (3d) signal. To simplify, the carbon species are hydrogenation of cyclohexene and nitrobenzene. The CO
classied into three groups: the peak with a binding energy chemical sorption results show that no CO could be adsorbed
37
peak at 288.3 eV (I) is assigned to carbonyl carbons, the peak at on PA–Ru@YSN. This suggests that the Ru surface is completely
86.2 eV (II) corresponds to C–OH, C–N (amine) and C–N covered with the dendrimer, which cannot be accessed by
2
38
(amide) species, and the peak at 284.4 eV (III) should be toluene. All Ru@C@YSN-T catalysts obtained by pyrolysis of
39,40
attributed to C–C and CHx.
All the above mentioned carbon PA-Ru@YSN in H2 are active for toluene hydrogenation.
species could be clearly observed in the C 1s XPS spectra of all Ru@C@YSN-150 obtained at low pyrolysis temperature still
Ru@C/SiO -T catalysts, which is consistent with the FT-IR could afford 60.8% conversion in 2 h, suggesting that the
2
results that carbon residues are formed aer pyrolysis. As the pyrolysis makes the active sites of Ru accessible to toluene.
pyrolysis temperature increases, the relative intensity of peak II Under similar conditions, 100% conversion was obtained for
to peak III increases obviously, suggesting the transformation Ru@C@YSN-200 and Ru@C@YSN-500.
from C]O, C–OH and C–N species into C–C species during
pyrolysis in a H atmosphere.
The kinetic curves for toluene hydrogenation on
2
Ru@C@YSN-T are displayed in Fig. 4. Ru@C@YSN-200 and
46
The XPS spectra of N 1s core levels of all samples show three Ru@C@YSN-500 afford zero-order reaction curves (Fig. S7†),
main peaks at 399.6 eV (assigned to surface species resulting indicating a strong adsorption of the toluene on the surface of
from C–N bond scission in the decomposed dendrimer), Ru. It should be noted that Ru@C@SiO -150 obviously shows
2
4
00.4 eV (assigned to amine), and 402 eV (assigned to amide) for an induction time during the catalytic process, probably due to
41,42
PA–Ru/SiO
2
.
Based on the XPS quantitative data (Table 2), the loss of dendrimer or residuals loosely capped on Ru during
the atomic ratio of O/Si for all samples is around 1.9–2, which is the catalytic process.
consistent with the chemical composition of SiO . As the The TOF of Ru@C@YSN-T increases sharply from 2000 to
pyrolysis temperature increases, the surface relative atomic 5270 mol per mol Rh per h along with the treatment tempera-
2
ꢂ
ꢂ
ratio of C (may be contaminated) and N decreases, showing the ture from 150 C to 500 C and a similar tendency was also
decrease in the content of C and N species on the Ru surface at observed taking the surface Ru atom into consideration (ob-
high pyrolysis temperature. Ru@C/SiO -150 gives all the peaks tained by CO chemisorption). This may be attributed to the C
2
related to three types of N species although the relative peak and N species deposited on the Ru surface, slightly enhanced
intensity changes a lot. Only one broad peak centered at 400 eV crystallinity and increased uptake capacity of Ru at high pyrol-
was observed in N 1s XPS spectra of Ru@C/SiO
2
-200 and Ru@C/ ysis temperature.
With Ru@C@YSN-500 as a model catalyst, the toluene hydro-
ꢂ
SiO
2
-500. With pyrolysis temperature $200 C, the intensities of
N species decrease sharply. For Ru@C/SiO
2
-200, the integrated genation was performed under different conditions (Table 3). The
nitrogen signal decreases by 50%, and the decomposition TOF of Ru@C@YSN-500 increases from 8320 to 20 300 mol per
species (399.6 eV) becomes dominant. The nitrogen signal mol Ru per h with the reaction temperature increasing from 50
ꢂ
ꢂ
continues to decrease to 15% at 500 C. In the literature, the N to 100 C at 3 MPa H2 (Fig. S8†). Ru@C@YSN-500 affords
ꢂ
45
1
s signal drops almost to zero aer reduction at 500 C. This a much higher TOF than most recently reported solid cata-
difference may be due to the support effect. The XPS results lysts.
4
4,47–50
The apparent activation energy for toluene hydro-
ꢁ
1
ꢁ1
genation is 16.8 kJ mol and 19.9 kJ mol calculated with
Arrhenius plots of Ru@C@YSN-500 and Ru@C@YSN-200,
respectively (Fig. S9†). This is much lower than the activation
energy reported for the transition-metal catalyzed hydrogena-
tion of toluene (E
literature.
Ru particle size and the positive effect of C and N species in
facilitating H activation.
In addition to toluene, Ru@C@YSN-500 could also effi-
ciently catalyze the hydrogenation of aromatic rings with
different substituents to afford 100% selectivity to the product
Table 2 The XPS quantitative data and the binding energy (BE) of PA–
Ru/SiO
2
pyrolyzed at different temperatures in H
2
ꢁ
1
a
¼ 30–50 kJ mol ) that we found in the
BE
0
d+
50–54
Samples
O/Si
Si/C
Si/N
50.4
48.4
99.9
Ru /Ru
(eV, Ru 3p3/2
)
The possible reason may be related to the small
PA–Ru/SiO
2
2
4.2
6.9
8.6
9.3
0.71
0.74
0.90
1.18
461.6
461.6
461.5
461.5
55,56
2
Ru@C/SiO
Ru@C/SiO
Ru@C/SiO
2
2
2
-150
-200
-500
1.9
1.9
1.9
206.6
This journal is © The Royal Society of Chemistry 2016
J. Mater. Chem. A

View Article Online
Journal of Materials Chemistry A
Paper
a
b
Table 3 The catalytic performance of Ru@C@YSN-T catalysts for the hydrogenation of toluene and benzene and comparison of the results of
this work with those reported in the literature
d
e
TOF
TOF
(
mol per mol
c
ꢂ
Samples
Substrate
Ru (wt%)
T ( C)
Conv. (%)
(mol per mol Ru per h)
RuCO per h)
PA–Ru
Toluene
Toluene
Toluene
Toluene
Toluene
—
50
50
50
50
50
100
50
100
100
ꢀ0
ꢀ0
ꢀ0
PA–Ru@YSN
Ru@C@YSN-150
Ru@C@YSN-200
Ru@C@YSN-500
1.19
1.24
1.38
1.44
1.44
1.44
5
ꢀ0
ꢀ0
ꢀ0
60.8
>99.9
>99.9
>99.9
99.2
>99
2000
4840
5270
20 300
9960
5000
14 200
7315
6860
6957
26 800
13 150
—
f
Toluene
Benzene
Benzene
Ru/SiO
Ru/CNF–P (ref. 44)
2
(ref. 43)
f
Toluene
1.6–1.7
>99
—
a
b
Reaction conditions for toluene hydrogenation: P ¼ 2 MPa H
2
, S/C ¼ 10 000, time ¼ 2 h. Reaction conditions for benzene hydrogenation: P ¼
ꢂ
ꢂ
c
2
MPa H
2
, S/C ¼ 10 000, T ¼ 50 C, time ¼ 1.5 h for Ru@C@YSN-500; P ¼ 2 MPa H
2
, S/C ¼ 2500, T ¼ 100 C, time ¼ 0.5 h for Ru/SiO
2
. The Ru
d
content was determined by means of ICP-AES. Apparent TOF is calculated as moles of converted toluene per mol of Ru per hour based on the slope
e
of the conversion with the conversion. TOF (mol per mol RuCO per h) is calculated based on surface Ru atoms measured via CO chemisorption.
f
P ¼ 3 MPa H , S/C ¼ 35 800.
2
respectively, to naphthenic acid and g-valerolactone under mild
conditions with high selectivity. These results demonstrate the
excellent catalytic activity of Ru@C@YSN-500.
The recyclability of Ru@C@YSN-500 was tested with toluene
hydrogenation as the model reaction. Aer the rst reaction
cycle, the catalyst could be recovered by centrifugation. Aer
being washed with ethanol thoroughly and dried under
vacuum, the recovered catalyst was directly used for the next
reaction cycle. Ru@C@YSN-500 could be stably recycled for
4
cycles without obvious loss in catalytic activity. Aer the
fourth cycle, the catalyst was characterized by TEM as shown in
Fig. 5B, the catalyst still retain its original york-shell nano-
structure with uniformly distributed Ru NPs. No aggregation of
Ru NPs could be observed. But the decrease in activity was
observed for 5 and 6 cycles. The TEM image clearly shows the
broken yolk–shell nanostructure and the aggregation of Ru NPs
without the shell protection. Thus, the decrease in the activity is
due to the low mechanical stability of Ru@C@YSN-500. This
further suggests the importance of the shell in preventing
aggregation of Ru NPs. We also tested the stability of the
Ru@C@YSN-200 catalyst under continuous xed-bed ow
conditions (Fig. S10†). The toluene hydrogenation reaction
under xed-bed ow conditions was typically monitored for
ꢂ
Fig. 4 Catalytic hydrogenation of toluene by Ru@C@YSN-T, 50 C,
MPa H
, S/C ¼ 10 000.
2
2
with only the aromatic ring hydrogenated (Table 4). Compared
with toluene, Ru@C@YSN-500 affords a lower TOF for the
hydrogenation of ethylbenzene and o-xylene under similar
reaction conditions. The activity follows the order of toluene >
ethylbenzene > o-xylene, which is consistent with the steric
57
effect of the substituent group. Ru@C@YSN-500 could also
efficiently convert the benzoic acid and Levulinic acid,
12 h time-on-stream each day; no obvious decay of conversion
Table 4 Catalytic performance of catalysts in the hydrogenation of other substrates
Apparent TOF
ꢁ
1
Entry
Substrate
Product
Conv. (%)
Sel. (%)
(h )
a
1
Ethylbenzene
O-Xylene
Benzoic acid
Levulinic acid
Ethylcyclohexane
Dimethylcyclohexane
Naphthenic acid
g-Valerolactone
>99.9
99.3
88
>99.9
>99.9
>99.9
99.4
1930
1010
26.4
315
a
2
b
3
c
4
94.5
a
ꢂ
Reaction conditions: S/C ¼ 4000, solvent free, temperature (50 C), 2.0 MPa H
2
, reaction time for ethylenebenzene and O-xylene is 2 h and 4 h,
b
respectively. Reaction conditions: benzoic acid (0.06 mmol), Ru (3.3 mol% relative to benzoic acid), S/C ¼ 30, water solvent (3 mL), temperature
ꢂ
c
ꢂ
(55 C), 1.0 MPa H
2
, time 1 h. Reaction conditions: S/C ¼ 1000, water solvent (2 mL), temperature (70 C), 3.0 MPa H
2
, time, 3 h.
J. Mater. Chem. A
This journal is © The Royal Society of Chemistry 2016

View Article Online
Paper
Journal of Materials Chemistry A
Acknowledgements
This work was nancially supported by the NSFC (21325313 and
1321002) and by the Strategic Priority Research Program of the
2
Chinese Academy of Sciences (Grant No. XDB17020200).
Notes and references
1
2
3
4
C. X. Xiao, Z. P. Cai, T. Wang, Y. Kou and N. Yan, Angew.
Chem., Int. Ed., 2008, 47, 746.
Q. L. Zhu, N. Tsumori and Q. Xu, J. Am. Chem. Soc., 2015, 137,
1
1743.
A. Cao, R. Lu and G. Veser, Phys. Chem. Chem. Phys., 2010, 12,
3499.
1
S. H. Joo, J. Y. Park, J. R. Renzas, D. R. Butcher, W. Huang
and G. A. Somorjai, Nano Lett., 2010, 10, 2709.
Fig. 5 (A) Recycling stability of Ru@C@YSN-500 in toluene hydro-
genation, TEM images of Ru@C@YSN-500 after (B) the 4th cycle and
5 J. Kang, S. Zhang, Q. Zhang and Y. Wang, Angew. Chem., Int.
Ed., 2009, 48, 2565.
(
C) 6th cycle.
6
J. Grunes, J. Zhu and G. A. Somorjai, Chem. Commun., 2003,
257.
G. A. Somorjai, F. Tao and J. Y. Park, Top. Catal., 2008, 47, 1.
2
7
and selectivity was observed over 7 days under these conditions.
The recycle test demonstrates that Ru@C@YSN could be stably
recycled.
8 Y. Yin, R. Rioux, C. Erdonmez, S. Hughes, G. Somorjar and
A. Alivisatos, Science, 2004, 304, 711.
9 C. Wang, J. Chen, X. Zhou, W. Li, Y. Liu, Q. Yue, Z. Xue, Y. Li,
A. Elzatahry, Y. Deng and D. Zhao, Nano Res., 2015, 8, 238.
A control sample, Ru@C/SiO
2
-500, was also prepared via
. Ru@C/SiO
2
-500 and 10 Q. Yue, Y. Zhang, C. Wang, X. Wang, Z. Sun, X. Hou, D. Zhao
pyrolysis of PA–Ru supported on SiO
2
Ru@C@YSN-500 were synthesized under similar conditions but
the former sample does not have a yolk–shell nanostructure.
and Y. Deng, J. Mater. Chem. A, 2015, 3, 4586.
11 Y. Li, J. Wei, W. Luo, C. Wang, W. Li, S. Feng, Q. Yue,
M. Wang, A. Elzatahry, Y. Deng and D. Zhao, Chem. Mater.,
2014, 26, 2438.
Ru@C/SiO
2
-500 shows slightly higher activity than
Ru@C@YSN-500 (TOF: 6040 versus 5270 mol per mol Ru per h),
suggesting that the silica shell causes diffusion problems for 12 J. Wei, Q. Yue, Z. Sun, Y. Deng and D. Zhao, Angew. Chem.,
toluene during the catalytic process. However, the ltrate of
2012, 124, 6253.
Ru@C/SiO
-500 aer reaction has brown colour, showing that 13 J. Liu, S. Qiao, B. Hartono and G. Lu, Angew. Chem., 2010,
the leaching of Ru NPs occurs during the catalytic process.
122, 5101.
These results further demonstrate that the yolk–shell nano- 14 L. Zhang, S. Qiao, Y. Jin, Z. Chen, H. Gu and G. Lu, Adv.
2
structure could efficiently prevent the leaching of Ru NPs during
Mater., 2008, 20, 805.
the catalytic process.
15 Q. Yue, J. Li, W. Luo, Y. Zhang, A. Elzatahry, X. Wang,
C. Wang, W. Li, X. Cheng, A. Alghamdi, A. Abdullah,
Y. Deng and D. Zhao, J. Am. Chem. Soc., 2015, 137, 13282.
1
6 W. Li, Q. Yue, Y. Deng and D. Zhao, Adv. Mater., 2013, 25,
Conclusions
5129.
Pyrolysis of Ru–poly(amindoamine) encapsulated in silica- 17 Z. Sun, J. Yang, J. Wang, W. Li, S. Kaliaguine, X. Hou, Y. Deng
based yolk–shell nanospheres results in the formation of a Ru and D. Zhao, J. Mater. Chem. A, 2014, 2, 6071.
NP catalyst. The hydrogenation tests including toluene hydro- 18 Y. Deng, Y. Cai, Z. Sun, J. Liu, C. Liu, J. Wei, W. Li, C. Liu,
genation, benzoic acid hydrogenation and Levulinic acid Y. Wang and D. Zhao, J. Am. Chem. Soc., 2010, 132, 8466.
hydrogenation show their excellent catalytic activity and 19 J. Ge, Q. Zhang, T. Zhang and Y. Yin, Angew. Chem., 2008,
selectivity. The CO and H chemisorption experiments suggest 120, 9056.
that the exposure degree of Ru could be increased signicantly 20 M. Zhao, L. Sun and R. M. Crooks, J. Am. Chem. Soc., 1998,
via H pyrolysis. XPS results show that the Ru surface still has C 120, 4877.
and N species even at high pyrolysis temperature. These C and 21 N. Yan, C. Xiao and Y. Kou, Coord. Chem. Rev., 2010, 254,
N species and yolk–shell nano-structure could efficiently 1179.
prevent the aggregation of Ru NPs and leaching of Ru NPs 22 Z. L. Wang, J. M. Petroski, T. C. Green and M. A. El-Sayed,
during the catalytic process. Ru NPs in yolk–shell nanospheres J. Phys. Chem. B, 1998, 102, 6145.
with a TOF of 20 300 h are among the most active solid 23 J. A. Lopez-Sanchez, N. Dimitratos, C. Hammond, G. L. Brett,
2
2
ꢁ
1
catalysts for toluene hydrogenation and could be stably
recycled.
L. Kesavan, S. White, P. Miedziak, R. Tiruvalam, R. L. Jenkins
and A. F. Carley, Nat. Chem., 2011, 3, 551.
This journal is © The Royal Society of Chemistry 2016
J. Mater. Chem. A

View Article Online
Journal of Materials Chemistry A
Paper
2
2
2
2
2
2
3
3
4 C. Aliaga, J. Y. Park, Y. Yamada, H. S. Lee, C.-K. Tsung, 41 O. Ozturk, T. Black, K. Perrine, K. Pizzolato, C. Williams,
P. Yang and G. A. Somorjai, J. Phys. Chem. C, 2009, 113, 6150.
5 H. Wang, H. Tang, J. He and Q. Wang, Mater. Res. Bull., 2009,
F. Parsons, J. Ratliff, J. Gao, C. Murphy and H. Xie,
Langmuir, 2005, 21, 3998.
4
4, 1676.
6 G. Lafaye, C. T. Williams and M. D. Amiridis, Catal. Lett.,
004, 96, 43.
42 Z. X. Xu, T. Li, Z. M. Zhong, D. S. Zha, S. H. Wu, F. Q. Liu,
W. D. Xiao, X. R. Jiang, X. X. Zhang and J. T. Chen,
Biopolymers, 2011, 95, 682.
2
7 D. Shen, J. Yang, X. Li, L. Zhou, R. Zhang, W. Li, L. Chen, 43 X. Kang, J. Zhang, W. Shang, T. Wu, P. Zhang, B. X. Han,
R. Wang, F. Zhang and D. Zhao, Nano Lett., 2014, 14, 923. Z. Wu, G. Mo and X. Xing, J. Am. Chem. Soc., 2014, 136, 3768.
8 J. Peng, G. Lan, M. Guo, X. Wei, C. Li and Q. Yang, Chem.– 44 M. Takasaki, Y. Motoyama, K. Higashi, S. H. Yoon,
Eur. J., 2015, 21, 10490. I. Mochida and H. Nagashima, Chem.–Asian J., 2007, 2, 1524.
9 Y. Yang, X. Liu, X. Li, J. Zhao, S. Bai, J. Liu and Q. Yang, 45 Y. Gu, G. Wu, X. F. Hu, D. A. Chen, T. Hansen, H. C. zur Loye
Angew. Chem., Int. Ed., 2012, 51, 9164. and H. J. Ploehn, J. Power Sources, 2010, 195, 425.
0 S. F. Yin, Q.-H. Zhang, B. Q. Xu, W. X. Zhu, C. F. Ng and 46 J. W. Thybaut, M. Saeys and G. B. Marin, Chem. Eng. J., 2002,
C. T. Au, J. Catal., 2004, 224, 384. 90, 117.
1 C. Wang, B. Li, H. Lin and Y. Yuan, J. Power Sources, 2012, 47 F. Su, L. Lv, F. Y. Lee, T. Liu, A. I. Cooper and X. S. Zhao,
202, 200.
J. Am. Chem. Soc., 2007, 129, 14213.
3
3
2 L. Sun and R. M. Crooks, Langmuir, 2002, 18, 8231.
3 T. Narita, H. Miura, K. Sugiyama, T. Matsuda and
R. D. Gonzalez, J. Catal., 1987, 103, 492.
48 G. Lan, H. Tang, Y. Zhou, W. Han, H. Liu, X. Li and Y. Li,
ChemCatChem, 2014, 6, 353.
49 A. Nowicki, V. Le Boulaire and A. Roucoux, Adv. Synth. Catal.,
2007, 349, 2326.
3
4 T. Okuhara, T. Kimura, K. Kobayashi, M. Misono and
Y. Yoneda, Bull. Chem. Soc. Jpn., 1984, 57, 938.
5 S. Hamoudi and S. Kaliaguine, Chem. Commun., 2002, 2118.
50 M. H. Prechtl, M. Scariot, J. D. Scholten, G. Machado,
S. R. Teixeira and J. Dupont, Inorg. Chem., 2008, 47, 8995.
3
3
6 J. Reynhardt, Y. Yong, A. Sayari and H. Alper, Adv. Funct. 51 J. Zhang, J. Teo, X. Chen, H. Asakura, T. Tanaka, K. Teramura
Mater., 2005, 15, 1641. and N. Yan, ACS Catal., 2014, 4, 1574.
37 E. Briand, V. Humblot, J. Landoulsi, S. Petronis, 52 D. Cunha and G. Cruz, Appl. Catal., A, 2002, 236, 55.
C. M. Pradier, B. Kasemo and S. Svedhem, Langmuir, 2010, 53 V. V. Pushkarev, K. An, S. Alayoglu, S. K. Beaumont and
2
7, 678.
8 G. Bar, S. Rubin, R. W. Cutts, T. N. Taylor and 54 H. Lang, R. A. May, B. L. Iversen and B. D. Chandler, J. Am.
T. A. Zawodzinski, Langmuir, 1996, 12, 1172. Chem. Soc., 2003, 125, 14832.
G. A. Somorjai, J. Catal., 2012, 292, 64.
3
3
9 B. V. Crist, Handbook of Monochromatic XPS Spectra, the 55 L. F. Wang and R. T. Yang, Energy Environ. Sci., 2008, 1, 268.
Elements of Native Oxides, ed. B. Vincent Crist, Wiley-VCH, 56 C. T. Campbell, J. C. Sharp, Y. X. Yao, E. M. Karp and
October 2000, vol. 1, p. 548, ISBN 0-471-49265-5.
T. L. Silbaugh, Faraday Discuss., 2011, 152, 227.
4
0 S. Biniak, G. Szyma n´ ski, J. Siedlewski and A. Swiatkowski, 57 B. Nicholls and M. Whiting, J. Chem. Soc., 1959, 551.
Carbon, 1997, 35, 1799.
´
J. Mater. Chem. A
This journal is © The Royal Society of Chemistry 2016