Synthesis of different ruthenium nickel bimetallic nanostructures and an investigation of the structure-activity relationship for benzene hydrogenation to cyclohexane

Article abstract of DOI:10.1002/cctc.201400096

The catalytic properties of catalysts are generally highly dependent on their nanostructures in most heterogeneous catalytic reactions. Therefore, to acquire targeted catalytic activity, selectivity, and stability, catalysts with a specific nanostructure

Full text of DOI:10.1002/cctc.201400096

CHEMCATCHEM
FULL PAPERS
DOI: 10.1002/cctc.201400096
Synthesis of Different Ruthenium Nickel Bimetallic
Nanostructures and an Investigation of the Structure–
Activity Relationship for Benzene Hydrogenation to
Cyclohexane
[
a]
[a]
[a]
[a]
[a]
[a]
Lihua Zhu, Maohong Cao, Li Li, Hanlei Sun, Yanqing Tang, Nuowei Zhang,
[a]
[a]
[a]
[a]
[b]
Jinbao Zheng,* Hua Zhou, Yunhua Li, Lefu Yang, Chuan-Jian Zhong, and
Bing H. Chen*
[a]
The catalytic properties of catalysts are generally highly depen-
dent on their nanostructures in most heterogeneous catalytic
reactions. Therefore, to acquire targeted catalytic activity, selec-
tivity, and stability, catalysts with a specific nanostructure
should be designed and synthesized. Herein, Ru-Ni bimetallic
nanoparticles with different nanostructures, Ru-Ni alloy, Ru@Ni,
and Ru clusters-on-Ni on carbon, have been synthesized by an-
nealing Ru-Ni/C in flowing N +10% H at different tempera-
particles have been characterized and their catalytic behaviors
were evaluated using benzene hydrogenation to cyclohexane.
The relationship between the Ru-Ni bimetallic nanostructures
and their catalytic performance is presented. It was found that
Ru-Ni alloy/C and Ru clusters-on-Ni/C are much more active
than Ru@Ni/C. This study also provides a simple method to
design and control the nanostructures of the Ru-Ni bimetallic
nanoparticles.
2
2
tures. The various nanostructures of the Ru-Ni bimetallic nano-
Introduction
Both in the past and at present, bimetallic catalysts are widely
in a series of reactions and exhibited distinguishing catalytic
properties. Khashab et al. prepared hollow Au@Pd core–shell
and hollow Au@Pt core–shell nanoparticles (NPs) by a galvanic
[
1]
investigated in fundamental research and used in industrial
[
2]
catalysis processes. It is well known that the bimetallic nano-
structure is one of the key factors to determine the catalytic
[9]
displacement method. These two types of core–shell NPs
showed a higher catalytic activity than those of monometallic
Pt for the ethanol oxidation reaction in an alkaline medium. Au
clusters were loaded onto carbon-supported Pt NPs by a gal-
vanic replacement method by Adzic et al. The as-synthesized
Au nanoparticles-on-Pt nanoparticles/C catalyst showed an un-
expected stability (loss of 4% in electrochemical surface area
after 30000 cycles in a durability test) compared to a Pt/C cata-
[
3]
behavior. Therefore, with the purpose to increase catalytic ac-
tivity, selectivity, and durability, bimetallic catalysts with differ-
ent expected nanostructures such as alloys, core–shell materi-
als, and particles-on-particles have been prepared and utilized
[
3e]
in numerous studies. For instance, Enache synthesized Au-
Pd(alloy)/TiO nanocatalysts to achieve an unprecedented cata-
2
ꢀ
1
lytic activity (turnover frequency (TOF) up to 270000 h ) for
the oxidation of alcohols to aldehydes compared to pure Au/
TiO or Pd/TiO catalysts. Other alloy nanocatalysts, such as Pt-
[10]
lyst for the oxygen reduction reaction.
According to the
above and plenty of other reported studies, it was found, and
is perhaps becoming a generic view, that bimetallic nanocata-
lysts with a particular nanostructure provided a much better
catalytic performance compared to the corresponding mono-
metallic nanocatalysts.
2
2
[
4]
[5]
[6]
[7]
[8]
Co, Pt-Ni, Rh-Ni, Au-Pt, and Pt-Fe, have been exploited
[
a] L. Zhu, M. Cao, L. Li, H. Sun, Y. Tang, Dr. N. Zhang, Dr. J. Zheng,
Dr. H. Zhou, Dr. Y. Li, Dr. L. Yang, Prof. B. H. Chen
Department of Chemical and Biochemical Engineering
National Engineering Laboratory for Green
Productions of Alcohols-Ethers-Esters
College of Chemistry and Chemical Engineering
Xiamen University
Likewise, Ru-Ni bimetallic nanoparticles (BNPs) with different
structures have been obtained in a few previous studies. We
will take the following two reports as examples. The Ru-Ni
core–shell was synthesized by a seeded growth method and it
Xiamen, 361005 (China)
Fax: (+86)592-2184822
E-mail: jbzheng@xmu.edu.cn
[11]
was applied in the ammonia borane hydrolysis reaction. Ru-
Ni alloy NPs were obtained by adopting a colloidal synthetic
approach and acted as a good catalyst for the dehydrogena-
chenbh@xmu.edu.cn
[12]
[b] Prof. C.-J. Zhong
tion of ammonia borane. However, to the best of our knowl-
edge, the synthesis of Ru-Ni alloy, Ru@Ni core–shell, and Ru
clusters-on-Ni particles through the adjustment of the thermal
treatment temperature on same precursor has not been re-
ported to date.
Department of Chemistry
State University of New York at Binghamton
Binghamton, New York, 13902 (United States)
Supporting information for this article is available on the WWW under
http://dx.doi.org/10.1002/cctc.201400096.
ꢀ
2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemCatChem 2014, 6, 2039 – 2046 2039

CHEMCATCHEM
FULL PAPERS
www.chemcatchem.org
In this study, carbon-supported Ni/Ni(OH)2 NPs were ob-
tained at room temperature (RT) by a hydrazine hydrate reduc-
peaks of the Ni(111) (2q=44.48), Ni(200) (2q=51.78), and
Ni(220) (2q=76.38) planes (JCPDS card No. 04-0850) can be
observed in Figure 1a, which indicates face-centered cubic
(fcc) Ni in the 16.84%Ni/C(380) sample. For Ru0.04Ni0.96/C(230)
and Ru0.04Ni0.96/C(280), only three diffraction peaks at 2q=44.2,
51.6, and 76.28 can be seen in Figure 1b and c, located be-
tween the corresponding Ni diffraction peaks in 16.84%Ni/
C(380) (Figure 1a) and Ru diffraction peaks in 2.5%Ru/C(600)
(Figure 1 h). This indicates that Ru atoms have dissolved in fcc
Ni to produce a Ru-Ni alloy, which results in the expansion of
the lattice constant. With the increase of the annealing tem-
perature to 3808C, a weak diffraction peak of Ru(100) appears
tion method, and then Ru was deposited on Ni/Ni(OH) /C by
2
a galvanic replacement reaction to form Ru-Ni/Ni(OH) /C. Addi-
2
tionally, various Ru-Ni nanostructures including Ru-Ni alloy,
Ru@Ni, and Ru clusters-on-Ni particles were obtained success-
fully by simple variation of the annealing temperature in flow-
ing N +10% H . The catalytic activity of the catalysts with dif-
2
2
ferent nanostructures has been evaluated by the reaction of
benzene hydrogenation to cyclohexane. A structure–activity re-
lationship has been established in this work for the benzene
hydrogenation reaction system.
(
JCPDS card No. 06-0663; Figure 1d), which suggests that
some of the Ru atoms begin to segregate from the Ru-Ni alloy
NPs. As the annealing temperature increases to 4808C, the
characteristic diffraction peaks of Ru(100) (2q=38.68) and
Ru(002) (2q=42.38; JCPDS card No. 06-0663) become clear
because of the complete phase segregation between Ru and
Ni. Additionally, if the thermal treatment temperature increases
continuously (e.g., to 580 and 6808C), the XRD peaks of the
pure Ni and Ru crystallite phases are shifted to lower angles,
which indicates clearly that high temperatures improve the lat-
Results and Discussion
Preparation of the catalysts
The Ni/Ni(OH) /C sample was obtained at RT by the hydrazine
2
hydrate reduction of nickel(II) chloride hexahydrate
(
NiCl ·6H O) and using carbon black (BLACK PEARLS 2000 LOT-
2 2
1
1
1
366221) with a very high surface area (approximately
2
ꢀ1
385.3 m g ) as the support with a Ni element loading of
6.84 wt%. The 16.84%Ni/C(380) was obtained after the Ni/
[13]
tice constant expansion. Moreover, the intensity of the peaks
increases gradually, and the full-width at half-maximum is in-
creased with the increase of temperature, which suggests that
the degree of crystallinity and the size of the Ru-Ni BNPs in-
crease. Magnifications of the XRD patterns in the regions 40–
50, 45–65, and 65–858 are presented in Figure S1, in which the
shifts of the characteristic diffraction peaks are clearer.
Ni(OH) /C sample was annealed in N +10% H at 3808C for
2
2
2
3
h. The Ru-Ni/Ni(OH) /C sample was obtained by a galvanic re-
2
placement reaction with 1.25 wt% Ru loading, 15.75 wt% total
Ni (Ni element) loading, and a Ru/Ni atomic ratio of 0.04/0.96.
Ru-Ni/Ni(OH) /C was reduced in N +10% H at different an-
2
2
2
nealing temperatures (230, 280, 380, 480, 580, or 6808C) for
3
1
h, denoted as Ru0.04Ni0.96/C(T). The 2.5%Ru/C(600) and
The lattice parameters of different catalysts are calculated
from the Ni(111) diffraction peak position. The relationship be-
tween the annealing temperature and lattice parameter is dis-
played in Figure 2. Based on this plot, it can be found that the
lattice parameter of Ni for the Ru0.04Ni0.96/C(T) catalysts is larger
than that for 16.84%Ni/C(380) if the treatment temperature is
in the range of 230–3808C. This is because some Ru atoms dis-
solve in fcc Ni to result in the lattice expansion. At 4808C, the
lattice parameter decreases to less than that of the 16.84%Ni/
C(380) catalyst, which indicates the complete phase segrega-
tion of Ru and Ni. If the thermal treatment temperature in-
creases above 4808C (e.g., 580 or 6808C), the increase of the
.25%Ru/C(380) catalysts were prepared by incipient wetness
impregnation for comparison. The synthesis steps are de-
scribed in detail in the Experimental Section.
Catalyst characterization
The powder XRD patterns of 16.84%Ni/C(380), Ru0.04Ni0.96/C(T),
and 2.5%Ru/C(600) are shown in Figure 1. Typical diffraction
Figure 1. Powder XRD patterns of the Ru0.04Ni0.96/C(T) samples. a) 16.84%Ni/
C(380), b) Ru0.04Ni0.96/C(230), c) Ru0.04Ni0.96/C(280), d) Ru0.04Ni0.96/C(380),
e) Ru0.04Ni0.96/C(480), f) Ru0.04Ni0.96/C(580), g) Ru0.04Ni0.96/C(680), and
h) 2.5%Ru/C(600). The blue and black dash-dot lines represent the diffrac-
tion peaks of the Ni and Ru crystallite phases, respectively.
Figure 2. Relationship between the annealing temperature and lattice
parameters of the catalysts.
ꢀ
2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemCatChem 2014, 6, 2039 – 2046 2040

CHEMCATCHEM
FULL PAPERS
www.chemcatchem.org
lattice parameter is a result of
the high temperature, which en-
courages the lattice expan-
[
13]
sion.
Therefore, the evolution of the
Ru-Ni nanostructure with the
thermal treatment temperature
can be demonstrated clearly by
the above XRD results. If the an-
nealing temperature of the
Ru0.04Ni0.96/C(T) catalysts is 230 or
2808C, complete Ru-Ni alloy NPs
were formed. Partial Ru-Ni alloy
and partial Ru-Ni phase segrega-
tion are obtained at 3808C. Fi-
nally, at 480, 580, and 6808C,
complete Ru-Ni phase segrega-
tion is achieved. The accurate
nanostructures of the Ru-Ni
BNPs have also been character-
ized by other techniques.
X-ray photoelectron spectros-
copy (XPS) measurements were
performed to obtain the chemi-
cal states of Ru and Ni in the dif-
ferent catalysts. As displayed in
the Ni2p XPS spectrum (Fig-
ure 3A), it is well known that the
binding energies at 861.8 and
8
80.5 eV are ascribed to multie-
[
14]
lectron excitation.
For these
three samples, 16.84%Ni/C(380),
Ru0.04Ni0.96/C(480), and Ru0.04Ni0.96
C(680), the binding energies of
52.7, 853.9, 855.4, and 857.1 eV
/
8
in the Ni2p3/2 XPS spectrum are
0
caused by Ni , NiO, Ni(OH) , and
2
NiOOH species, respectively. The
other peaks at binding energies
of 870, 871.2, 872.9, and
874.4 eV
in
the
Ni2p1/2 XPS spectrum are as-
0
cribed to Ni , NiO, Ni(OH) , and
2
[
14–15]
NiOOH species, respectively.
On comparing the intensity of
each peak, the main Ni species
in all the samples are found to
be of varying oxidation state as Figure 3. XPS spectra. A), a) Ni2p edge in 16.84%Ni/C(380), b) Ni2p edge in Ru0.04Ni0.96/C(230), c) Ni2p edge in
Ru0.04Ni0.96/C(480), d) Ni2p edge in Ru0.04Ni0.96/C(680). B), a) Ru3p edge in 1.25%Ru/C(380), b) Ru3p edge in
Ru0.04Ni0.96/C(230), c) Ru3p edge in Ru0.04Ni0.96/C(480), d) Ru3p edge in Ru0.04Ni0.96/C(680).
NiꢀNiO, Ni(OH) , and NiOOH. The
2
NiO originates mainly from the
[
15d]
oxidation of Ni NPs in air,
and
[
16]
the formation of Ni(OH) was probably because of the reaction
transfer of an electron from Ru to Ni. This suggests that Ru-
Ni alloy NPs are produced in this catalyst. Although all the cat-
alysts were reduced in N +10% H , a lot of oxidized Ni was
2
of NiO with H O in air. NiOOH was formed as Ni(OH) was fur-
2
2
ther oxidized in air. However, for the Ru0.04Ni0.96/C(230) sample,
2
2
0
the binding energy at 852.5 eV corresponds to Ni with
still detected on the catalyst surface by XPS after exposure to
air, which indicates the existence of many Ni atoms on the sur-
face of these catalysts. XPS binding energies and the relative
a 0.2 eV shift to a lower binding energy relative to the
1
6.84%Ni/C(380) or Ru0.04Ni0.96/C(680) samples because of the
ꢀ
2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemCatChem 2014, 6, 2039 – 2046 2041

CHEMCATCHEM
FULL PAPERS
www.chemcatchem.org
0
composition [mol%] of different Ni species (Ni , NiO, Ni(OH) ,
2
and NiOOH) on the surface of the catalysts are shown in
Table S1.
Generally, it is well known that the Ru3d spectrum is not
[
17]
clear as it is often obscured by the strong C1s signal. There-
fore, the XPS test for Ru was conducted between 458 and
472 eV in which Ru3p peaks are expected to be observed (Fig-
ure 3B). For the 1.25%Ru/C(380) and Ru0.04Ni0.96/C(680) cata-
lysts, the peak positions at binding energies of 462 and
0
4
64.2 eV in the Ru3p3/2 XPS spectra are assigned to Ru and
RuO . The other two peaks at higher energies (484 and
2
4
86.2 eV) in the Ru3p1 XPS spectra are also contributed to by
/2
0
[18]
Ru and oxidized RuO , respectively. The presence of RuO2
2
0
was a result of the surface oxidation of Ru NP after exposure
[
19]
to air prior to characterization,
which demonstrates that
there are Ru atoms on the surface of the Ru-Ni BNPs in
Ru0.04Ni0.96/C(680). However, for the Ru0.04Ni0.96/C(230) sample,
0
the binding energy of Ru in the Ru3p3 XPS spectrum is shift-
/2
ed to a higher energy position (462.3 eV) compared to that of
1
.25%Ru/C(380), which is almost without a RuO signal. This
2
suggests that the Ru-Ni alloy NPs are formed in the Ru0.04Ni0.96
/
C(230) sample, which prevents the oxidation of Ru in air. How-
ever, the XPS signal of the Ru species was very weak in the
Ru0.04Ni0.96/C(480) sample, which confirms that Ru was probably
present as a core inside the Ni particles (Ru@Ni) and was diffi-
cult detect by the XPS method. XPS binding energies and rela-
0
tive compositions [mol%] of the different Ru species (Ru and
RuO ) on the surface of the catalysts are shown in Table S2.
2
TEM images of Ru0.04Ni0.96/C(230), Ru0.04Ni0.96/C(480), and
Ru0.04Ni0.96/C(680) are shown in Figure S2. The mean size of the
Ru-Ni BNPs in the Ru0.04Ni0.96/C(230) sample was controlled at
Figure 4. HRTEM images of the Ru0.04Ni0.96/C(T) catalysts. A) HRTEM image of
Ru0.04Ni0.96/C(230). B) Enlarged view of the selected area in A. The indicated
lattice fringes, 0.209 nm (between 0.203 nm Ni(111) (JCPDS card No. 04-
0
850) and 0.214 nm Ru(002) or 0.234 nm Ru(100) (JCPDS card No. 06-0663),
1
1.56 nm with a relatively narrow size distribution (Figure S2A
which correspond to Ru-Ni alloy. C) HRTEM image of Ru0.04Ni0.96/C(480).
D) Enlarged view of the selected area in C. The indicated lattice fringe,
and B), whereas, if Ru0.04Ni0.96/C(T) was reduced in N +10% H2
2
at 4808C for 3 h, the average size of the Ru-Ni BNPs increased
to 14.06 nm also with a relatively narrow size distribution (Fig-
ure S2C and D). However, for the Ru0.04Ni0.96/C(680) catalyst,
the mean size of the Ru-Ni BNPs is the same as that of the
Ru0.04Ni0.96/C(480) sample but with a wider size distribution
0
.176 nm, is ascribed to Ni(200) planes (JCPDS card No. 04-0850). E) HRTEM
image of Ru0.04Ni0.96/C(680). F) Enlarged view of the selected area in E. The
indicated lattice fringe, 0.203 nm, is attributed to Ni(111) facets (JCPDS card
No. 04-0850). The insets show the respective fast Fourier transform (FFT)
patterns.
(Figure S2E and F). To explore the relationship between the an-
nealing temperature and the Ru-Ni bimetallic nanostructure,
high-resolution transmission electron microscopy (HRTEM)
images were obtained of these three samples. The HRTEM
images of Ru0.04Ni0.96/C(230) reveal lattice fringes with a regular
spacing of 0.209 nm (Figure 4A and B). This spacing is located
between the lattice fringe distances of Ru(100) (0.234 nm) or
Ru(002) (0.214 nm; JCPDS card No. 06-0663) and that of
Ni(111) (0.203 nm; JCPDS card No. 04-0850), which indicates
the formation of a Ru-Ni alloy. However, with the increase of
the annealing temperature to 4808C, the HRTEM image of an
individual Ru-Ni BNP shows lattice fringes with an interplanar
spacing of 0.170 nm (Figure 4C and D) ascribed to Ni(200)
fringes of fcc Ni (JCPDS card No. 04-0850). In the HRTEM image
of the Ru0.04Ni0.96/C(680) sample, the lattice fringe distance of
catalysts further illustrate the Ru-Ni phase segregation in these
two samples.
As it is difficult to distinguish Ni from Ru in HRTEM images
(Figure 4) because of the indistinct contrast between these
two elements, elemental analysis was performed by high-angle
annular dark-field scanning transmission electron microscopy
(HAADF-STEM) combined with energy-dispersive X-ray spec-
troscopy (EDS) to further establish the Ru-Ni nanostructures
and elemental distributions in the Ru0.04Ni0.96/C(230),
Ru0.04Ni0.96/C(480), and Ru0.04Ni0.96/C(680) catalysts. A represen-
tative HAADF-STEM image and its corresponding Ru and Ni el-
emental maps of Ru0.04Ni0.96/C(230) are presented in Figure 5A,
which demonstrate that the distribution ranges of Ni and Ru
could overlap to some extent. Based on the XRD and XPS re-
sults for the Ru0.04Ni0.96/C(230) sample, a conclusion can be
drawn that the as-synthesized Ru0.04Ni0.96/C(230) is definitely
Ru-Ni alloy/C. An HAADF-STEM image and elemental maps of
Ru and Ni for the Ru0.04Ni0.96/C(480) catalyst are shown in Fig-
0.203 nm is observed clearly (Figure 4E and F), which is
a result of Ni(111) facets (JCPDS card No. 04-0850). These
HRTEM images of the Ru0.04Ni0.96/C(480) and Ru0.04Ni0.96/C(680)
ꢀ
2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemCatChem 2014, 6, 2039 – 2046 2042

CHEMCATCHEM
FULL PAPERS
www.chemcatchem.org
ure 5B, and it can be observed clearly that Ru (green) is sur-
rounded by Ni (orange). This probably indicates that a Ru@Ni
core–shell structure was formed in this sample. The HAADF-
STEM image and elemental analysis for Ru0.04Ni0.96/C(680) is dis-
played in Figure 5C, which suggest that the distribution of Ru
and Ni are homogenous. However, the foregoing XRD results
for Ru0.04Ni0.96/C(680) (Figure 1 h) did not reveal the presence
of a Ru-Ni alloy. This indicates that Ru atoms are supported on
the Ni NPs in this sample. Structural models of these three cat-
alysts are also presented in Figure 5, which demonstrate vividly
the different nanostructures in these samples. These conclu-
sions were further verified by a high-sensitivity low-energy ion
scattering (HS-LEIS) study.
To obtain further information on the different structures of
our as-prepared Ru0.04Ni0.96/C(T) catalysts, HS-LEIS measure-
ments were performed because this technique can effectively
provide the atomic composition of the outmost atomic layer
[20]
20
+
of the solid surface.
The 5 keV Ne HS-LEIS spectra of
Ru0.04Ni0.96/C(230), Ru0.04Ni0.96/C(480), and Ru0.04Ni0.96/C(680) are
presented in Figure 6. Ru and Ni atoms coexist on the outmost
surface of the Ru0.04Ni0.96/C(230) catalyst, but the quantity of Ni
atoms is much larger than that of the Ru atoms (Figure 6a),
whereas only the Ni signal appeared on the outer surface of
the Ru0.04Ni0.96/C(480) sample (Figure 6b). As the Ru0.04Ni0.96/C
(
(
uncalcined) catalyst was reduced at a higher temperature
e.g., 6808C), the Ru and Ni signals can be observed at the
same time in the HS-LEIS spectra for the Ru0.04Ni0.96/C(680) cat-
alyst (Figure 6c). Therefore, these HS-LEIS results provide fur-
ther evidence that the evolution of the Ru-Ni bimetallic nano-
structure is highly dependent on the thermal treatment tem-
perature.
Based on the above characterizations, it can be concluded
that Ru-Ni alloy is formed in the Ru0.04Ni0.96/C(230) sample,
Ru@Ni core–shell particles is generated in the Ru0.04Ni0.96
/
C(480) sample, and Ru clusters-on-Ni particles are obtained in
the Ru0.04Ni0.96/C(680) sample.
A schematic illustration of the evolution of the Ru-Ni bimet-
allic nanostructure of the Ru0.04Ni0.96/C(T) catalysts is depicted
in Scheme 1. The evolution process is described as follows.
2
+
Firstly, Ni(OH) is produced by the combination of Ni ions
2
ꢀ
2+
with OH ions; secondly, a small number of Ni ions were
formed gradually from the dissociation of Ni(OH) with a low
2
0
dissociation constant and were reduced to Ni at RT by hydra-
zine hydrate, and the Ni/Ni(OH) was supported on the carbon
2
black. Thirdly, some of the Ni atoms would be displaced with
Ru atoms by a replacement reaction after the addition of aque-
Figure 5. A) HAADF-STEM image and elemental mapping for several Ru-Ni
BNPs in Ru0.04Ni0.96/C(230). a) HAADF-STEM image of Ru0.04Ni0.96/C(230), b)
and c) EDS maps of Ni (red) and Ru (yellow), respectively, d) structural model
of Ru0.04Ni0.96/C(230); B) HAADF-STEM image and elemental mapping for an
individual Ru-Ni BNP in Ru0.04Ni0.96/C(480). a) HAADF-STEM image of
Ru0.04Ni0.96/C(480), b) and c) EDS maps of Ni (orange) and Ru (green), respec-
tively, d) structural model of Ru0.04Ni0.96/C(480); C) HAADF-STEM image and
elemental mapping for an individual Ru-Ni BNP in Ru0.04Ni0.96/C(680).
a) HAADF-STEM image of an individual Ru-Ni BNP, b) and c) EDS maps of Ni
(
orange) and Ru (green), respectively, (d) structural model of Ru0.04Ni0.96
/
C(680).
ꢀ
2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemCatChem 2014, 6, 2039 – 2046 2043

CHEMCATCHEM
FULL PAPERS
www.chemcatchem.org
C(230) (Ru-Ni alloy)ꢁRu0.04Ni0.96/C(680) (Ru clusters-on-Ni parti-
cles)> Ru0 /C(680)> Ru0.04Ni0.96/C(480) (Ru@Ni). Clearly, the
.04
catalytic performance is highly dependent on the Ru-Ni bimet-
allic nanostructure. Such dependence could be explained by
the following principles. It is generally accepted that a mono-
metallic Ru catalyst is more active than a monometallic Ni cata-
lyst for the activation and dissociation of hydrogen molecules
under relatively mild reaction conditions. Furthermore, Ru-
based catalysts are currently considered as highly effective cat-
[19,21]
alysts for the benzene hydrogenation reaction.
Hence, if
Ru atoms are on the surface of the Ru-Ni BNPs in the
Ru0.04Ni0.96/C(230) (Ru-Ni alloy) and Ru0.04Ni0.96/C(680) (Ru clus-
ters-on-Ni particles) catalysts, it could facilitate the activation
and dissociation of hydrogen molecules, which would improve
their catalytic performance for benzene hydrogenation. How-
ever, the Ru0.04Ni0.96/C(480) (Ru@Ni) catalyst exhibited no activi-
ty for this reaction because many of the Ru atoms were in the
2
0
+
Figure 6. HS-LEIS spectra (5 keV Ne ) of a) Ru0.04Ni0.96/C(230), b) Ru0.04Ni0.96
C(480), and c) Ru0.04Ni0.96/C(680). E =Energy of backscattered ion.
/
f
core of the Ru@Ni NPs. The catalytic properties of Ru0.04Ni0.96
/
C(680) were a little poorer than those of Ru0.04Ni0.96/C(230),
which could be attributed to the larger size of the Ru-Ni BNPs
(
Figure S2).
Stability of the Ru0.04Ni0.96/C(T) catalysts for benzene hydro-
genation
Scheme 1. Schematic illustration of the evolution of Ru-Ni alloy, Ru@Ni, and
Ru clusters-on-Ni particles supported on carbon black.
The stability of the Ru0.04Ni0.96/C(T) catalysts was also investigat-
ed in this work. The Ru0.04Ni0.96/C(230), Ru0.04Ni0.96/C(480), and
Ru0.04Ni0.96/C(680) catalysts were reused for another four runs
under the same reaction conditions (reaction pressure, 5.3 MPa
ous RuCl3 solution to form Ru-Ni/Ni(OH) /C. Finally, the
2
Ru0.04Ni0.96/C(T) catalysts with different nanostructures were ob-
tained after they were annealed in flowing N +10% H
H ; reaction temperature, 608C; reaction time, 1 h). No clear
2
deactivation was seen from the results shown in Figure S3, es-
pecially if we take the possibility of tiny catalyst loss during re-
agents–catalyst separation into account. Additionally, the
HRTEM images of the Ru0.04Ni0.96/C(T) (X=230, 480, and 680)
catalysts after five uses are displayed in Figure S4. These
images confirm that the Ru0.04Ni0.96/C(230) catalyst is still in the
form of Ru-Ni alloy NPs, Ru0.04Ni0.96/C(480)-Ru@Ni is still core–
shell, and Ru0.04Ni0.96/C(480) is still Ru clusters-on-Ni particles,
which suggests that the nanostructures of the Ru-Ni BNPs are
very stable during the reaction. The reasons for the excellent
stability of the nanostructures of Ru-Ni BNPs are as follows:
2
2
ꢀ
1
(80 mLmin ) at various temperatures for 3 h.
Catalytic performance of Ru0.04Ni0.96/C(T) for benzene hydro-
genation
The catalytic performance of various catalysts for the reaction
of benzene hydrogenation to cyclohexane was measured
under relatively mild conditions (reaction pressure, 5.3 MPa H2;
reaction temperature, 608C; reaction time, 1 h). All of the as-
prepared catalysts gave 100% selectivity to cyclohexane
(Table 1). For the 16.84%Ni/C(380) sample, no prod-
uct was produced (Table 1, entry 1). However, the
Ru0.04Ni0.96/C(230) catalyst provided a 9.25% yield to
Table 1. Catalytic performances of the Ru0.04Ni0.96/C(T) catalysts under the same reac-
tion conditions for benzene hydrogenation to cyclohexane.
ꢀ
1
cyclohexane with a relatively high TOF of 1684.5 h
Table 1, entry 2), which was approximately 17 times
[a]
(
ꢀ
1
Entry
Catalyst (0.05 g)
T [8C]
TOF [h
]
Yield to cyclohexane [%]
more activity than that of the Ru0.04Ni0.96/C(480)
sample (Table 1, entry 3). This reaction was also con-
ducted over the Ru0.04Ni0.96/C(680) catalyst under the
same reaction conditions, which showed a lower ac-
tivity (Table 1, entry 4; 6.77% yield to cyclohexane
1
2
3
4
5
16.84%Ni/C(380)
Ru0.04Ni0.96/C(230)
Ru0.04Ni0.96/C(480)
Ru0.04Ni0.96/C(680)
60
60
60
60
60
–
<0.1
9.25
0.53
6.77
3.35
1684.5
96.5
1232.8
610.6
[
b]
Ru0.04/C(680)
ꢀ
1
and a TOF of 1232.8 h ) compared to Ru0.04Ni0.96
/
[
a] All reactions were conducted under the following conditions: benzene (10 mL), re-
C(230). Moreover, Ru0.04/C(680) gave a 3.35% yield to
action time (1 h), reaction pressure (5.3 MPa). All of the as-prepared catalysts gave
100% selectivity to cyclohexane. [b] Ru0.04/C(680) was prepared by incipient wetness
ꢀ
1
cyclohexane and a TOF of 610.6 h (Table 1, entry 5)
in the benzene hydrogenation reaction. The catalytic
impregnation using carbon as the support and reduced in N
h.
2 2
+10% H at 6808C for
3
activity of the catalysts was in the order: Ru0.04Ni0.96
/
ꢀ
2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemCatChem 2014, 6, 2039 – 2046 2044

CHEMCATCHEM
FULL PAPERS
www.chemcatchem.org
1
) these catalysts were annealed in N +10% H at high tem-
16.84 wt% (including Ni element in Ni(OH) ) determined by induc-
2
2
2
tively coupled plasma mass spectrometry (ICP-MS). The Ru-Ni/
perature (above 2308C) for 3 h, which is much higher than the
reaction temperature (608C); 2) the annealing atmosphere of
the catalysts is the same as the reaction atmosphere (H2).
Ni(OH) /C catalysts were prepared by galvanic replacement reac-
2
tions by adding a certain amount of aqueous RuCl solution (9.64ꢁ
3
ꢀ3
ꢀ1
1
0
molL ) into Ni/Ni(OH) /C and Ru atoms were anchored onto
2
Ni/Ni(OH) . The catalyst had a Ru/Ni atomic ratio of 4:96, a Ru load-
2
ing of 1.25 wt%, and a Ni loading of 15.57 wt% (determined by
Conclusions
ICP-MS). The Ru-Ni/Ni(OH) /C catalysts were then reduced in flow-
2
ꢀ1
Ru-Ni bimetallic nanoparticles (BNPs) with different nanostruc-
tures, Ru-Ni alloy, Ru@Ni, and Ru clusters-on-Ni particles, sup-
ported on carbon were synthesized successfully by the simple
method of changing the reducing temperature to 230, 480,
and 6808C, correspondingly. The nanostructures of these cata-
lysts were characterized by XRD, X-ray photoelectron spectros-
copy, high-resolution transmission electron microscopy,
energy-dispersive spectroscopy mapping, and high-sensitivity
low-energy ion scattering measurements. The order of the cat-
alytic performance for the hydrogenation of benzene to cyclo-
hexane was Ru0.04Ni0.96/C(230) (Ru-Ni alloy)ꢁRu0.04Ni0.96/C(680)
ing N +10% H2 (80 mLmin ) at various annealing temperatures
2
ꢀ1
for 3 h with a heating rate of 28C min . The products were denot-
ed as Ru0.04Ni0.96/C(T), in which Tannealing =230, 280, 380, 480, 580, or
6
808C. Ru/C catalysts with a Ru loading of 1.25 or 2.5 wt% were
prepared by incipient wetness impregnation using carbon black as
the support and then reduced in flowing N +10% H2
2
ꢀ1
(
80 mLmin ) at 380 or 6008C for 3 h, and the products denoted
as 1.25%Ru/C(380) or 2.5%Ru/C(600), respectively.
Characterization
Powder XRD patterns for the samples were collected by using
a Rigaku Ultima IV X-ray diffractometer equipped with a high-
speed array detection system, and CuKa radiation (35 kV and
20 mA) was used as the X-ray source. Scans were performed over
(Ru clusters-on-Ni particles)>Ru Ni /C(480) (Ru@Ni). In ad-
0.04 0.96
dition, the nanostructures of the Ru-Ni BNPs are very stable in
the reaction. The correlation between the nanostructure and
the catalytic activity of the catalyst for the hydrogenation of
benzene to cyclohexane was established in this work. The ther-
mal treatment method to tune the Ru-Ni bimetallic nanostruc-
ture is likely to be extended to other BNPs, which is expected
to have profound implications in other catalytic reaction sys-
tems.
ꢀ1
the 2q range of 20–908 with a scanning rate of 208min . XPS
measurements were performed by using a PHI Quantum 2000
Scanning ESCA Microprobe equipment with monochromatic AlKa
radiation (1846.6 eV) as the X-ray source. TEM and HRTEM studies
of the catalysts were performed by using a TECNAI F30-HRTEM
with a field-emission source, and the accelerating voltage was
300 kV. HAADF-STEM and STEM-EDS mapping analysis were also
performed by using a TECNAI F20-HRTEM equipped with a field-
emission source, and the accelerating voltage was 200 kV. HS-LEIS
Experimental Section
20
+
measurements were performed by using a Ne beam energy of
keV, a sample current of 1.6 nA. The scattering angle was 1458.
5
Materials
Carbon (BLACK PEARLS 2000 LOT-1366221) was obtained from
Catalytic activity tests
Cabot Corporation. All other reagents (NiCl ·6H O, anhydrous etha-
2
2
nol, NaOH, anhydrous RuCl , polyvinyl pyrrolidone (PVP), and
3
The hydrogenation of benzene to cyclohexane was performed in
a stainless-steel, high-pressure reactor (Parr 4848) with a magnetic
stirrer. A mixture of benzene (10 mL) and catalyst (50 mg) was
added into the reactor, which was then sealed immediately. H2
8
5 wt% hydrazine hydrate solution) were supplied by Sinopharm
Chemical Reagent Co. Ltd. (Shanghai, China). The water used in all
experiments was deionized water produced by a Milli-Q Intergral 3
(
Millipore Corporation). CO (99.5%), H (99.999%), and N +10% H
2 2 2
(4.8 MPa) was introduced into the autoclave after it was purged
were purchased from Linde Gas (Xiamen) Co. Ltd. All reagents
were used as received.
with N for 1 min and then with H for 1 min. The stirring rate was
2
2
approximately 500 rpm. The mixture was heated quickly to the re-
quired temperature (608C) and the H pressure was increased to
2
5
.3 MPa. After the desired reaction time (1 h), the reactor was
Catalyst synthesis
cooled quickly to approximately 58C by using an ice-water bath,
and then the autoclave was evacuated. The catalysts were separat-
ed from the liquid by centrifugation. The products were analyzed
by GC by using a Shimadzu GC 2010 instrument equipped with
a DB-35 60 mꢁ0.32 mm capillary column and a flame ionization
detector (FID), and GC–MS (Shimadzu GC–MS 2010) was used if
necessary. The TOF of the catalysts was calculated from Equa-
tion (1):
Firstly, Ni/Ni(OH) /C was prepared by
a
hydrazine reduction
2
method, using carbon black as the support with a very high sur-
2
ꢀ1 [22]
face area (approximately 1385.3 m g ). Ni/Ni(OH) /C was pre-
2
pared by the following procedure. PVP (0.300 g) and anhydrous
ethanol (12.5 mL) were mixed with a certain amount of aqueous
ꢀ
2
ꢀ1
NiCl ·6H O solution with a concentration of 4.6ꢁ10 molL with
2
2
magnetic stirring at RT for 10 min. Then, aqueous NaOH solution
1.813 g of NaOH dissolved in 12.5 mL of deionized water) was
(
nconversion
nconversion
t ꢂ nNi
added to the above green solution. After 10 min, 85 wt% hydra-
zine hydrate solution (25 mL) was added. Subsequently, carbon
black (1.250 g) was suspended in the as-obtained liquid. The above
liquid was transferred into a Teflon cup and agitated vigorously at
RT for 18 h. The obtained solids were recovered by filtration and
washed repeatedly with deionized water and anhydrous ethanol
TOFRu
¼
;
TOFNi
¼
ð1Þ
-Ni
t ꢂ nRu
in which nconversion is the conversion of benzene [mol], t is the reac-
tion time, nRu is the number of moles of Ru, and nNi is the number
of moles of ruthenium and nickel. TOFNi and TOFRu represent the
TOF of the 16.84%Ni/C(380) and Ru0.04Ni0.96/C(680), Ru0.04/C(680)
catalysts, respectively.
until no chlorine ions were detected. Ni/Ni(OH) /C was obtained
after drying in vacuum at 608C for 6 h. The Ni loading was
2
ꢀ
2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemCatChem 2014, 6, 2039 – 2046 2045

CHEMCATCHEM
FULL PAPERS
www.chemcatchem.org
[
[
8] C. Xu, Q. Li, Y. Liu, J. Wang, H. Geng, Langmuir 2012, 28, 1886–1892.
9] H. M. Song, D. H. Anjum, R. Sougrat, M. N. Hedhili, N. M. Khashab, J.
Mater. Chem. 2012, 22, 25003–25010.
Acknowledgements
The authors would like to thank the National Natural Science
Foundation of China (Grant 20973140, 201106118, and
[
10] J. Zhang, K. Sasaki, E. Sutter, R. R. Adzic, Science 2007, 315, 220–222.
[11] G. Chen, S. Desinan, R. Nechache, R. Rosei, F. Rosei, D. Ma, Chem.
Commun. 2011, 47, 6308–6310.
[
21303140) for financial support.
12] G. Chen, S. Desinan, R. Rosei, F. Rosei, D. Ma, Chem. Eur. J. 2012, 18,
925–7930.
7
[
[
13] C. Solliard, M. Flueli, Surf. Sci. 1985, 156, 487–494.
Keywords: hydrogenation
ruthenium · structure–activity relationships
·
nanoparticles
·
nickel
·
14] K. W. Park, J. H. Choi, B. K. Kwon, S. A. Lee, Y. E. Sung, H. Y. Ha, S. A.
Hong, H. Kim, A. Wieckowski, J. Phys. Chem. B 2002, 106, 1869–1877.
15] a) N. C. Barnard, S. G. R. Brown, F. Devred, J. W. Bakker, B. E. Nieuwen-
huys, N. J. Adkins, J. Catal. 2011, 281, 300–308; b) J. Wang, G. Fan, F. Li,
RSC Adv. 2012, 2, 9976–9985; c) Y. Goto, K. Taniguchi, T. Omata, S.
Otsuka-Yao-Matsuo, N. Ohashi, S. Ueda, H. Yoshikawa, Y. Yamashita, H.
Oohashi, K. Kobayashi, Chem. Mater. 2008, 20, 4156–4160; d) S. Sarkar,
A. K. Sinha, M. Pradhan, M. Basu, Y. Negishi, T. Pal, J. Phys. Chem. C 2011,
[
[
1] a) B. Lim, M. Jiang, P. H. C. Camargo, E. C. Cho, J. Tao, X. Lu, Y. Zhu, Y.
Xia, Science 2009, 324, 1302–1305; b) H. Zhang, T. Watanabe, M. Oku-
mura, M. Haruta, N. Toshima, Nat. Mater. 2012, 11, 49–52; c) T. R. Johns,
J. R. Gaudet, E. J. Peterson, J. T. Miller, E. A. Stach, C. H. Kim, M. P. Balogh,
A. K. Datye, ChemCatChem 2013, 5, 2636–2645; d) L. Delannoy, S. Gior-
gio, J. G. Mattei, C. R. Henry, N. El Kolli, C. Mꢂthivier, C. Louis, Chem-
CatChem 2013, 5, 2707–2716; e) L. Gan, S. Rudi, C. Cui, P. Strasser,
ChemCatChem 2013, 5, 2691–2694.
115, 1659–1673.
[
[
[
16] J. H. Jeong, J. W. Lee, D. J. Seo, Y. Seo, W. L. Yoon, D. K. Lee, D. H. Kim,
Appl. Catal. A 2006, 302, 151–156.
17] C. Ronning, H. Feldermann, R. Merk, H. Hofsꢃss, P. Reinke, J. U. Thiele,
Phys. Rev. B 1998, 58, 2207–2215.
18] a) N. Chakroune, G. Viau, S. Ammar, L. Poul, D. Veautier, M. M. Chehimi,
C. Mangeney, F. Villain, F. Fiꢂvet, Langmuir 2005, 21, 6788–6796; b) Y.
Yamauchi, T. Ohsuna, K. Kuroda, Chem. Mater. 2007, 19, 1335–1342;
c) K. Lasch, L. Jçrissen, J. Garche, J. Power Sources 1999, 84, 225–230.
19] S. Miao, Z. Liu, B. Han, J. Huang, Z. Sun, J. Zhang, T. Jiang, Angew. Chem.
Int. Ed. 2006, 45, 266–269; Angew. Chem. 2006, 118, 272–275.
20] H. H. Brongersma, M. Draxler, M. de Ridder, P. Bauer, Surf. Sci. Rep. 2007,
[2] a) D. M. Alonso, S. G. Wettstein, J. A. Dumesic, Chem. Soc. Rev. 2012, 41,
8
5
075–8098; b) W. Yu, M. D. Porosoff, J. G. Chen, Chem. Rev. 2012, 112,
780–5817.
[3] a) J. Yin, S. Shan, L. Yang, D. Mott, O. Malis, V. Petkov, F. Cai, M. Shan Ng,
J. Luo, B. H. Chen, M. Engelhard, C.-J. Zhong, Chem. Mater. 2012, 24,
4662–4674; b) B. N. Wanjala, J. Luo, R. Loukrakpam, B. Fang, D. Mott,
P. N. Njoki, M. Engelhard, H. R. Naslund, J. K. Wu, L. Wang, O. Malis, C. J.
Zhong, Chem. Mater. 2010, 22, 4282–4294; c) D. Wang, H. L. Xin, R.
Hovden, H. Wang, Y. Yu, D. A. Muller, F. J. DiSalvo, H. D. AbruÇa, Nat.
Mater. 2013, 12, 81–87; d) F. Tao, M. E. Grass, Y. Zhang, D. R. Butcher,
J. R. Renzas, Z. Liu, J. Y. Chung, B. S. Mun, M. Salmeron, G. A. Somorjai,
Science 2008, 322, 932–934; e) D. I. Enache, Science 2006, 311, 362–
[
[
[
6
2, 63–109.
21] a) M. Zahmakıran, Y. ı. Tonbul, S. ꢄzkar, J. Am. Chem. Soc. 2010, 132,
541–6549; b) M. Zahmakiran, S. ꢄzkar, Langmuir 2008, 24, 7065–7067;
6
365; f) F. Besenbacher, I. Chorkendorff, B. S. Clausen, B. Hammer, A. M.
c) G. Sꢅss-Fink, B. Mollwitz, B. Therrien, M. Dadras, G. Laurenczy, A. Meis-
ter, G. Meister, J. Cluster Sci. 2007, 18, 87–95; d) M. Takasaki, Y. Motoya-
ma, K. Higashi, S. H. Yoon, I. Mochida, H. Nagashima, Chem. Asian J.
Molenbroek, J. K. Nørskov, I. Stensgaard, Science 1998, 279, 1913–1915;
g) F. Studt, F. Abild-Pedersen, T. Bligaard, R. Z. Sorensen, C. H. Christen-
sen, J. K. Norskov, Science 2008, 320, 1320–1322.
2
007, 2, 1524–1533.
22] L. Zhu, L. Zheng, K. Du, H. Fu, Y. Li, G. You, B. H. Chen, RSC Adv. 2013, 3,
13–719.
[
4] J. Greeley, I. E. L. Stephens, A. S. Bondarenko, T. P. Johansson, H. A.
Hansen, T. F. Jaramillo, Rossmeisl, Chorkendorff, J. K. Nørskov, Nat.
Chem. 2009, 1, 552–556.
[
7
[5] S. Lu, C. Zhang, Y. Liu, Int. J. Hydrogen Energy 2011, 36, 1939–1948.
[6] H. Duan, D. Wang, Y. Kou, Y. Li, Chem. Commun. 2013, 49, 303–305.
[7] W. Ye, H. Kou, Q. Liu, J. Yan, F. Zhou, C. Wang, Int. J. Hydrogen Energy
Received: January 29, 2014
Revised: March 8, 2014
2012, 37, 4088–4097.
Published online on June 2, 2014
ꢀ
2014 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
ChemCatChem 2014, 6, 2039 – 2046 2046