Highly Stable and Recyclable Graphene Layers Protected Nickel–Cobalt Bimetallic Nanoparticles as Tunable Hydrotreating Catalysts for Phenylpropane Linkages in Lignin

Article abstract of DOI:10.1007/s10562-017-2179-1

Abstract: Nickel–cobalt bimetallic nanoparticles coated with several layers of graphene were developed through direct heating treatment of bimetallic oxide precursor prepared by the modified Pechini-type sol–gel method. These nanomaterials were demonstrated to be versatile catalysts for lignin depolymerization. The catalysts showed unexpectedly tunable selectivity that directly depends on the composition of bimetallic nanoparticles. Dimeric lignin model compounds can be converted totally and the hydrogenolysis selectivities above 85% over Ni–Co@C (Ni:Co = 1:3). During the recycling test, the nanocatalyst showed excellent recyclability in the ten-batch investigation. The deposition of graphene layers over bimetal nanoparticles fosters a subtle balance between protecting effects and surface accessibility to catalytic reactions and significantly improves their stability to air and moisture. Ni–Co@C catalysts were readily separated from the liquid mixtures with high recycling ratio due to their magnetic properties. Graphical Abstract: [Figure not available: see fulltext.] Ni–Co bimetallic nanoparticles are coated with graphene layers. Graphene layers over the nanoparticles protect them from deactivation. Ni–Co@C shows tunable selectivity in the hydrogenolysis of dimeric lignin linkage. The non-precious metal catalyst showed excellent recyclability and can be reused ten times without significant loss of activity.

Full text of DOI:10.1007/s10562-017-2179-1

Catal Lett
DOI 10.1007/s10562-017-2179-1
Highly Stable and Recyclable Graphene Layers Protected
Nickel–Cobalt Bimetallic Nanoparticles as Tunable Hydrotreating
Catalysts for Phenylpropane Linkages in Lignin
1
1
1
Bingfeng Chen · Fengbo Li · Guoqing Yuan
Received: 1 April 2017 / Accepted: 25 August 2017
©
Springer Science+Business Media, LLC 2017
Abstract Nickel–cobalt bimetallic nanoparticles coated
with several layers of graphene were developed through
direct heating treatment of bimetallic oxide precursor pre-
pared by the modified Pechini-type sol–gel method. These
nanomaterials were demonstrated to be versatile catalysts
for lignin depolymerization. The catalysts showed unexpect-
edly tunable selectivity that directly depends on the com-
position of bimetallic nanoparticles. Dimeric lignin model
compounds can be converted totally and the hydrogenolysis
selectivities above 85% over Ni–Co@C (Ni:Co=1:3). Dur-
ing the recycling test, the nanocatalyst showed excellent
recyclability in the ten-batch investigation. The deposition
of graphene layers over bimetal nanoparticles fosters a subtle
balance between protecting effects and surface accessibility
to catalytic reactions and significantly improves their sta-
bility to air and moisture. Ni–Co@C catalysts were readily
separated from the liquid mixtures with high recycling ratio
due to their magnetic properties.
Graphical Abstract
Ni–Co bimetallic nanoparticles are coated with graphene
layers. Graphene layers over the nanoparticles protect them
from deactivation. Ni–Co@C shows tunable selectivity in
the hydrogenolysis of dimeric lignin linkage. The non-pre-
cious metal catalyst showed excellent recyclability and can
be reused ten times without significant loss of activity.
Keywords Nanoparticles · Bimetallic catalysts ·
Graphene layers · Hydrogenolysis · Lignin
1
Introduction
*
*
Fengbo Li
Biomass represents an enormous store of renewable carbon
resources, which can be used as the feedstocks for the sus-
tainable productions of chemicals and fuels through biore-
finery [1, 2]. Lignocellulose materials consist of three pri-
mary fractions: hemicellulose (pentose, hexoses), cellulose
(glucose-polymer), and lignin (phenol-polymer) [3]. Lignin
is plant polymer made from phenylpropanoid building unites
lifb@iccas.ac.cn
Guoqing Yuan
yuangq@iccas.ac.cn
1
Beijing National Laboratory of Molecular Science, Key
Laboratory of Green Printing, Institute of Chemistry,
Chinese Academy of Sciences, Beijing 100190,
People’s Republic of China
Vol.:(0
1
123456789)
3

B. Chen et al.
through a series of characteristics linkages and end groups
4, 5]. These building blocks are connected with each other
Ru, Rh, and Pd) bimetallic nanoparticles catalysts have been
reported in the hydrogenolysis of lignin model compounds
(β-O-4 linkage) and organosolv lignin [14, 15]. Some cata-
lysts using base metals (Ni, Co, Cu) have been reported in
the hydrogenolysis of lignin model compounds. The crack-
ing of β-O-4 lignin-type dimers was catalyzed by Cu/Al O
[
to produce the complex architecture of lignin. There are sev-
eral linkages of the monolignol units that form different C–O
and C–C intermonomeric bonds in the polymer, as shown
in Scheme 1.
2
3
Native lignins and separated lignins are the only renew-
able resource for producing industrial aromatics. Through
a well-designed depolymerization process, lignin matrix
may be broken into a family of high-value phenolic struc-
tures, such as cresols, catechols, resorcinols, quinones, or
guaiacols [6]. However, pathways for producing high-value
molecules from lignin are in their very early state and the
selective transformation of lignin remains a great chal-
lenge [7]. There are several potential strategies for lignin
depolimerization. Thermochemical processes (combustion,
gasification, thermolysis, and hydrothermal hydrolysis) are
generally nonselective [8]. Although hydrotreating of lignin
leads to the increasing demand for hydrogen, it is one of
the most promising methods for depolymerization of lignin,
which can produce simple molecule mixtures with high
yields. Catalytic cracking of C–O intermonomeric bonds
in the presence of hydrogen, generally termed hydrogen-
olysis, has attracted much attention for lignin valorization.
or Cu/MgO–Al O [9]. Supported nickel catalysts, such as
2 3
Ni/SiO and Ni/TiN, have been used in the hydrogenolysis
2
of aryl ethers as lignin model compounds [16, 17]. Biddy
and Beckham also reported that Ni/hydrotalcite was quite
active (>99% conversion) in the cleavage of 2-phenoxy-
1-phenethanol in methyl isobutyl ketone at 270 °C for 1 h
[18]. Recently, NiAlO catalyst has been reported in the
x
hydrogenolysis of aromatic ethers to cyclohexanol deriva-
tives in the presence of Lewis acid [19].
Hydrotreating catalysts based on precious metals exhibit
efficient catalytic performances. However, their high prices
limit the practical application and lead to substantial costs in
the required recovery and recycle processes [20–23]. There
are great efforts put into developing base metal catalysts
(Ni, Co, Cu, Fe) with improved selectivity and recyclability
[24, 25]. The quantum size effect of nanoscale materials
offers many solutions to develop supported metal nano-
particles and metal species dispersed over nanomaterials.
The main drawbacks of base metals (Ni, Co, Cu, Fe) nano-
particles are their instability in air and moisture and fast
deactivation under the practical operation [26–28]. In our
recent work, carbon-coated Cu–Co bimetallic nanoparticles
show excellent catalytic performances in the hydrogenoly-
sis of 5-(hydromethyl)furfural for production of biofuel (2,
5-dimethylfuran) [29].
2
-Phenoxy-1-phenylethanol is always serviced as lignin
model compound for the β-O-4 linkage, which accounts
for approximately 50% of all the linkages in lignin [9]. Ru/
CNT was used in the hydrodeoxygenation of lignin-derived
phenols in a biphasic H O/n-dodecane system [10]. Pd/C
2
was employed as a catalyst for the cleavage of β-O-4 ether
bonds in lignin model compounds and moderate depolymeri-
zation of native lignins [11, 12]. Supported Pd–Fe bimetallic
catalyst has been reported in the catalytic hydrogenolysis of
phenethyl phenyl ether into aromatics [13]. Ni–M (M=Au,
In this work, a well-designed strategy was developed to
prepare nickel, cobalt, and nickel–cobalt mixed nanopar-
ticles, which are coated with several layers of graphene.
There is a subtle balance between protecting effects and
surface accessibility to catalytic reactions. Carbon-coated
nickel–cobalt mixed nanoparticles have unexpectedly tun-
able selectivity in catalytic hydrotreating of model molecule
of phenylpropane linkages in lignin, as shown in Scheme 1.
The deposition of graphene layers over the metal nanopar-
ticles markedly improves their stability to air and moisture
and leads to excellent recyclability.
2
Experimental
2
.1 Catalyst Preparation
Carbon-coated Ni–Co bimetallic nanoparticles were
prepared by the modified Pechini-type sol–gel method.
Co(NO ) ·6H O and Ni(NO ) ·6H O was added to water
3
2
2
3 2
2
(
20 ml, total metal amount: 6.25 mmol). The molar ratio
Scheme 1 Illustration of lignin fragment with the notation of phenyl-
propane linkage (carbon–oxygen bond)
of Ni to Co of metal precursors can be adjusted from 3:1
1
3

Highly Stable and Recyclable Graphene Layers Protected Nickel–Cobalt Bimetallic Nanoparticles…
to 1:5. Then tartaric acid (TA, 12.5 mmol) was added to
the mixture. Glycerol/water mixture (v/v=4:1, 60 ml) and
polyethene glycol (average MW: 6000, 5.0 g) were added.
Nanographite materials (0.5 g) were added and dispersed
under sonication for 1.0 h. The mixture was transferred to
a hydrothermal reaction vessel, and then heated to 150 °C
for 13 h. The solid materials were collected by centrifugal
separation, and washed with ethanol for three times, and
dried at 100 °C overnight. The carbon-coated Ni–Co bime-
tallic nanoparticle catalysts were obtained by thermal treat-
ment of the precursors at 800 °C in argon flow for 2 h. The
carbon-coated Ni or Co catalyst was prepared by the similar
method, and Ni(NO ) ·6H O or Co(NO ) ·6H O was used
products mixtures were analyzed by GC and chlorobenzene
was used as an external standard. Identification of main
products was based on GC-MS as well as by comparison
with authentic samples. The product distribution was shown
on the mole basis. In consecutive batch tests, the catalyst
was recovered by magnets and washed with EtOH for three
times. The catalyst was recycled into the autoclave with the
feedstream.
3 Results and Discussions
3.1 Development of Nanocatalysts
3
2
2
3 2
2
as metal precursor.
Through the modified Pechini-type sol–gel process, the
mixed metal oxides are formed and coated with polyeth-
ylene glycol. Further heating treatment gradually carbon-
izes the polymer layers, which serve as the carbon sources
for graphene layers. Meanwhile, the carbonized polymers
are the reductant for the formation of bimetallic nanoparti-
cles. Transmission electron microscopy (TEM) micrograph
showed formation of well-dispersed carbon-coated nickel
nanoparticles in the range of 3–10 nm (Fig. 1a–c),and the
average particle size is about 6 nm. Carbon-coated Ni–Co
bimetallic nanoparticles have larger particle size and their
average particle size is about 10 nm (Fig. 1d). The graphene
layers over the bimetallic nanoparticles are clearly revealed
by HRTEM images (Fig. 1e, f). The thickness of carbon
shell is about 2–6 nm. These carbon depositions provide
direct protection against oxidation and deactivation and
improve their recyclability and recoverability. The energy-
dispersive spectroscopy (EDS) measurement is carried out to
investigate the elemental composition and the distribution of
the element in the bimetallic nanocatalysts. Figure 2a shows
the selected particles in the marked areas for elemental map-
ping analysis. Every nanoparticle contains both Ni and Co,
and Ni and Co are uniformly distributed in the these par-
ticles (Fig. 2b, c). Elemental mapping of carbon overlaps
the noise signals from carbon membrane of TEM sample
support (Fig. 2d).
2
.2 Catalyst Characterization
Powder X-ray diffraction (XRD) patterns were measured on
a Rigaku Rotaflex diffractometer equipped with a rotating
anode and a Cu-Kα radiation source (40 kV, 200 MA; λ=1.
5
4,056 Å). Inductive coupling plasma emission spectrometer
(
ICP-OES) analysis was conducted on the Varian 710 ICP-
OES with ICP Expert II software. XPS data were obtained
with an ESCALab220i-XL electron spectrometer from VG
Scientific using 300 W Al-Kα radiations. The base pressure
−9
was approximately 3×10 mbar. The binding energies were
referenced to the C1s line at 284.8 eV from adventitious
carbon. The Eclipse V2.1 data analysis software supplied
by the VG ESCA-Lab200I-XL instrument manufacturer
was used to manipulate the acquired spectra. Transmission
electron microscopy (TEM) was performed on a JEOL 2010
TEM equipped with an attachment for local energy disper-
sion analysis (EDX). The accelerating voltage was 200 kV,
and the spot size was 1 nm. High-angle annular dark field
scanning transmission microscopy (HAADF-STEM) was
performed on the carbon-coated Ni–Co bimetallic nano-
particles catalysts with JEOL JEM-2100F microscope in a
scanning transmission electron microscopy (STEM) mode
operated at 200 kV.
2
.3 Catalyst testing
X-ray photoelectron spectroscopy (XPS) measurement
is carried out to investigate the chemical composition of
carbon-coated Ni–Co bimetallic nanoparticles. Ni 2p
Lignin model molecule (2-phenoxy-1-phenylethanol) was
synthesized according to the method reported by Rothenberg
3
/2
and 2p XPS peaks are at the binding energy values of
1
/2
[
9]. Catalytic hydrotreating of 2-phenoxy-1-phenylethanol
852.3 and 869.7 eV, respectively (Fig. 3A-a). The single
metal sample (Ni@Carbon) prepared by the same method
shows similar XPS peaks of Ni 2p. These results indicate
that nickel is almost in the zero-valent states. Co 2p XPS
peaks show more complex patterns (Fig. 3B-c). There are
three typical patterns: zero-valent metal species (Co 2p
was carried out in a 100 ml autoclave with Teflon liner.
The catalyst (50 mg) was added to a solution of 2-phenoxy-
1
-phenylethanol (150 mg) in ethanol (2 ml) and H O (8 ml).
2
The autoclave was sealed and purged with H several times.
2
The reaction was performed at 170 °C with 2 MPa hydrogen
pressure for 6 h. Then, the reactors were cooled down to
room temperature using an ice bath, and the organic products
were extracted with ethyl acetate (5 ml × 2). The organic
1
/2
777.5 eV), cobalt oxide species (Co 2p 781.6 eV), and
1
/2
the shakeup satellite signals (785.2 and 801.0 eV). The
single metal sample (Co@Carbon) shows the same trend
1
3

B. Chen et al.
Fig. 1 TEM image (a) and
HRTEM images (b–c) of
Ni@C; TEM image (d) and
HRTEM images (e–f) of Ni–
Co@Carbon (Ni:Co=1:3)
(
Fig. 3B-d). However, cobalt oxide species have a higher
through development of catalytic surface with tunable
selectivity. Carbon-coated Ni–Co mixed nanoparticles
shows unexpected versatility and flexibility in the selec-
tively hydrotreating the lignin linkage unit. Nine products
are detected in the final reaction mixtures. Product (2) is
formed by direct hydrodeoxygenation of the side hydroxyl
group. Aryl ether is kept constant and this conversion cannot
lead to depolymerization of lignin matrix. Other products
are from ether bond cleavage and further hydrodeoxygena-
tion and hydrogenation, which are summed as hydrogen-
olysis products. Generally, the catalytic cleavage of β-O-4
ether bonds in lignin model compounds is easier than the
hydrogeneolysis of aryl ether. Benzene has not been detected
in the product mixture. The reaction network is shown in
Scheme 2.
atomic concentration in the Co@Carbon than that in the
Ni–Co@C.
The XRD patterns of monometallic (Ni@C and Co@C)
and bimetallic (Ni–Co@C) samples, thermal treated at
8
00 °C for 2 h under argon atmosphere, are shown in Fig. 4.
For Ni@C catalyst, the diffraction peaks located at 2θ=44.4,
1.8 and 76.2° indicated the existence of the characteristics
of Ni metal phase (JCPDS 04-0850), while the peaks at
5
2
6.4° and 54.5° were assigned to the diffraction of graphite
phase (JCPDS 41-1487). For the Co@C catalyst, the diffrac-
tion peaks located at 44.2°, 51.5° and 75.9° were ascribed
to the characteristics of Co metal phase (JCPDS 15-0806).
3
.2 Catalytic Performances in Hydrotreating of Lignin
Model Linkages
Table 1 summarizes the results of screening experiments.
There is a marked difference in the catalytic performances
between carbon-coated nickel and carbon-coated cobalt
nanoparticles. The conversion over cobalt monometallic
nanoparticles only leads to the hydrodeoxygenation of side
hydroxyl group and no cleavage of ether bond is detected
(Entry 7 in Table 1). In contrast, the products of ether bond
2
-Phenoxy-1-phenylethanol (1) in Scheme 2 is selected as
the model molecule of the linkage unit in Scheme 1. There
are three possible reaction routes: hydrodeoxygenation,
cleavage of ether bond, and hydrogenation. To achieve valu-
able target products, the reaction path should be controlled
1
3

Highly Stable and Recyclable Graphene Layers Protected Nickel–Cobalt Bimetallic Nanoparticles…
Fig. 4 XRD patterns of the Ni@C (a), Ni–Co@C (Ni:Co=1:1) (b),
Ni–Co@C (Ni:Co=1:3) (c), Ni–Co@C (Ni:Co=1:5) (d) and Co@C
e)
(
Fig. 2 TEM analysis of Ni–Co@C catalyst (Ni:Co=1:3): a TEM
dark-field image; b elemental mapping of Co; c elemental mapping of
Ni; d elemental mapping of C; e elemental mapping of O
Scheme 2 Product distribution of 2-phenoxy-1-phenylethanol hydro-
treating over Ni–Co@C catalyst
cleavage are the main resulting compounds over nickel
mono-metallic nanoparticles and no further hydrodeoxy-
genation takes place (Entry 8 in Table 1). Ni–Co bimetallic
nanoparticles show tunable selectivity in the reaction routes
with the change of the proportions of two metal species.
Based on the experimental data in Table 1 and Fig. 5 reveals
the trend about the catalytic selectivity of various Ni–Co
bimetallic nanoparticles. It is noteworthy that the catalytic
selectivity directly depends on the composition of bimetal-
lic nanoparticles. Ni–Co@C (Ni:Co= 1:3) catalyst shows
the best performance with the hydrogenolysis selectivity of
8
3.7% with full conversion of substrate. Nickel species are
active for ether bond cleavage and cobalt species work as
hydordeoxygenation catalysts. These two functions can be
integrated to develop tunable catalytic selectivity for lignin
hydrogenolysis.
Fig. 3 A The Ni 2p XPS spectrum of Ni–Co@C (Ni:Co=1:3)
(
(
a) and Ni@C (b); B the Co 2p XPS spectrum of Ni–Co@C
Ni:Co=1:3) (c) and Co@C catalyst (d)
1
3

B. Chen et al.
Table 1 Screening experiments of hydrotreating 2-phenoxy-1-phenylethanol over various single-metal or bimetal catalysts coated with gra-
phene layers
Entry
Catalysts
Conversion (%)
Selectivity (%)
Hydrogenolysis
selectivity (%)a
2
3
4
5
6
7
8
9
10
1
2
3
4
5
6
Ni–Co@C
89.3
100
100
94.1
90.8
100
35.9
16.6
2.1
11.1
–
–
2.0
18.3
13.5
63.6
77.3
83.7
62.5
49.5
54.4
(
Ni:Co=1:1)
Ni–Co@C
Ni:Co=1:2)
Ni–Co@C
Ni:Co=1:3)
Ni–Co@C
Ni:Co=1:4)
Ni–Co@C
Ni:Co=1:5)
22.1
14.2
32.0
45.8
42.5
8.9
3.8
–
24.4
23.4
22.7
7.3
0.9
–
–
0.7
–
8.5
9.6
6.0
20.3
8.7
30.1
33.2
25.3
4.5
(
17.5
8.5
–
(
–
–
–
–
(
5.0
–
–
12.4
–
–
(
Ni @C +
15.9
1.2
10.1
–
–
18.5
Co@C
(Ni:Co=1:3)
7
8
9
Co@C
Ni @C
100
79.3
100
>99
–
–
–
–
–
–
–
–
–
–
49
–
–
1
–
50
–
–
–
–
–
31.7
5.8
–
18.3
44.2
0
100
100
b
Ni–Co@C
(Ni:Co=1:3)
Reaction condition: 1 (150 mg), catalyst (50 mg), EtOH (2 ml), H O (8 ml), H (2 MPa), 170 °C/6 h
2
2
a
Hydrogenolysis selectivity is the sum of products (3+4+5+6+7+8+9+10)
Reaction condition: 2-phenylethyl phenyl ether 2 (150 mg), catalyst (50 mg), EtOH (2 ml), H2O (8 ml), H (2 MPa), 170 °C/6 h
b
2
summarized in Table 2. The reactions can proceed in a green
solvent (H O/ethanol, 4:1) with high yield under relatively
2
mild conditions. Benzyl alcohol, diphenyl-methanol, and
4
-nitrobenzyl alcohol are totally converted to correspond-
ing methylene compounds. Hydrogenolysis of 3-pyridinyl-
methanol affords the 3-methylpyridine with a yield of 98.2%.
The reaction of 1-phenylethanol and 1-(2-methoxyphenyl)
ethanol give the target products in the yield of 98–99%. Cin-
namyl alcohol is converted to allybenzene (yield: 91.8%) as
the final product. However, the catalytic hydrogenolysis of
cinnamyl alcohol over Pd/C only leads to the formation of
n-propylbenzene [30]. Dehydroxylation of 2-phenylethanol
only gave a moderate yield of ethylbenzene. The hydrogenol-
ysis of 1, 4-phenylenedimethanol to p-tolylmethanol is com-
pleted after 12 h. 2-Phenoxy-1-phenylethanol was converted
to phenylethoxy-benzene in excellent yield (>99%) under
the optimal condition. Although Pd/C has been reported in
the hydrogenolysis of 2-phenoxy-1-phenylethanol under
high hydrogen pressure (5 MPa), the yield of phenylethox-
ylbenzene was only 38% [31].
Fig. 5 The correlation between the reaction selectivity and the com-
position of bimetallic nanoparticles [hydrogenolysis yield is the sum
of products (3+4+5+6+7+8+9+10) and hydrodeoxygenation
yield is based on the product 2]
Hydrodeoxygenation is a very important process to refine
the product distribution during lignin hydrogenolysis. The
catalytic performance of carbon-coated cobalt nanoparticles
is further investigated in hydrodeoxygenation of various sub-
strate molecules. As revealed in Entry 8 of Table 1, Co@C is
a very efficient catalyst for hydrodeoxygenation. Its perfor-
mances in catalytic conversion of other typical molecules are
Ni–Co@C (Ni:Co=1:3) catalyst is also applied to cata-
lyze hydrogenolysis of β-O-4 lignin-type dimmers (Table 3).
These dimeric lignin model compounds are the analogous
linkage units of lignin matrix as illustrated in Scheme 1. The
substitute groups of phenyl rings make the lignin-type dim-
mers more active than 2-phenoxy-1-phenylethanol. Under
the same reaction conditions, these dimeric lignin linkage
1
3

Highly Stable and Recyclable Graphene Layers Protected Nickel–Cobalt Bimetallic Nanoparticles…
Table 2 Dehydroxylation of aromatic alcohols over Co@Ccatalyst
Entry
1
Substrates
Products
t (h)
6
Yield (%)[b]
>99
OH
OH
OH
2
3
6
6
95.6
>99
>99
98.2
46.9
>99
91.8
>99
97.8
O
O
OH
4
6
OH
5
12
6
N
N
OH
6
OH
OH
7
10
10
6
O
2
N
O
2
N
8
OH
9[c]
O
O
Ph
Ph
OH
1
0
12
OH
OH
Reaction condition: alcohol (1 mmol), Co@C catalyst (50 mg, Co: 29.2 wt%), EtOH (2 ml), H2O (8 ml), H2 (2 MPa), 170 °C
a
Yield determined by GC using anisole as an internal reference
b
1
50 °C/6 h
compounds can be converted totally and the hydrogenolysis
selectivity values are all above 85%. These results reveal
that Ni–Co@C (Ni:Co = 1:3) is a highly efficient catalyst
for lignin depolymerization for producing value-added
chemicals.
Table 3 Hydrogenolysis of dimeric lignin model compounds over
Ni–Co@C (Ni:Co=1:3) catalyst
3
.3 Magnetic Recyclability and Activity Stability
Another attractive point is the improved recyclability of
Ni–Co@C (Ni:Co = 1:3) catalyst. Base metal (Ni, Co,
Cu, Fe) hydrotreating catalysts generally show relatively
poorer recyclability than noble metals. The graphene
layers provide the protection against deactivation dur-
ing recycling operation under the exposure to air and
moisture. The magnetic properties of Ni–Co@C cata-
lytic materials are beneficial to their separation from
the reaction mixture. Figure 6 shows the magnetiza-
tion curves of Co@C, Ni–Co@C (Ni:Co = 1:3), and
Ni@C at room temperature. The hysteresis loops in
Entry
Substituent groups
R =OCH , R =OH
Conversion (%)
Hydrogenoly-
sis selectivity
(
%)
1
2
3
>99
>99
>99
90.1
87.3
89.5
1
3
2
R =H, R =OH
1
2
R =H, R =OCH
3
1
2
Reaction condition: substrate (200 mg), catalyst (50 mg), EtOH
2 ml), H O (8 ml), H (2 MPa), 170 °C/6 h
(
2
2
1
3

B. Chen et al.
Fig. 6 Room temperature magnetization curves of a Co@C, b Ni–
Co@C, c Ni@C
these catalysts confirmed their ferromagnetic behaviors.
The corresponding saturation magnetization strengths
(
Ms) are 41.0 emu/g (Co@C), 47.3 emu/g (Ni–Co@C,
Fig. 7 a 12-run recycling test of Ni–Co@C (Ni:Co=1:3) catalyst. b
TEM image of the catalyst after the sixth recycling run. c TEM image
of the catalyst after the tenth recycling run
Ni:Co = 1:3), and 12.9 emu/g (Ni@C). Ms values of these
materials were much lower than those of the bulk cobalt
(
166 emu/g) [32] and nickel (51.3 emu/g) [33]. This is
mainly attributed to size effects of metal nanoparticles
and the presence of carbon layers. However, the Ms value
of Ni–Co@C affords a proper ferromagnetic parameter
for the efficient magnetic separation. After the reaction,
Ni–Co@C catalysts are readily separated from the liq-
uid mixtures by magnet with high recycling ratio. The
catalyst is input into the next batch reaction after simple
washing treatment.
4 Conclusion
In summary, nickel–cobalt mixed nanoparticles, coated with
several layers of graphene, were prepared by the modified
Pechini-type sol–gel method and the following controlled
heating treatment. This nanocatalysts showed unexpectedly
tunable selectivity in catalytic hydrotreating of model mol-
ecule of phenylpropane linkages in lignin. Dimeric lignin
model compounds can be converted totally and the hydrog-
enolysis selectivities were above 85%. The catalytic selectiv-
ity directly depends on the composition of bimetallic nano-
particles. The recyclability and recoverability of Ni–Co@C
(Ni:Co=1:3) was investigated through a 12-run recycling
test. The deposition of graphene layers over the metal nano-
particles markedly improves their stability to air and mois-
ture. The nanocatalyst showed excellent recyclability and
can be reused ten times without significant loss of activity.
As revealed in Fig. 7a, a 12-run recycling test is car-
ried out to investigate the recyclability and recoverabil-
ity of Ni–Co@C (Ni:Co = 1:3) during batch-by-batch
hydrogenolysis of 2-phenoxy-1-phenylethanol. After the
ninth testing run, there is a tendency toward deactiva-
tion. The vanishment of protection of carbon coating has
directly led to oxidation of active metallic species. Dur-
ing the practical operation, it is inevitable that carbon
layers over bimetallic nanoparticles are gradually worn
away by the intergranular friction, the collision with the
stirrer and autoclave wall, and the leaching effect of the
solvent. TEM image of the Ni–Co@C after the sixth run
shows that some abrasion and dislocation of carbon layers
are observed (Fig. 7b). Further recycling test leads to the
gradual loss of carbon layers. They vanish after the tenth
run and the deactivation is observed. However, the loss
of catalytic activity shows little impact on the catalytic
selectivity. It is clearly revealed that graphene protection
layers play a key role in improving catalytic stability of
the bimetallic nanoparticles.
Acknowledgements This work was financially supported by the
National Natural Sciences Foundation of China (Nos. 21403248,
2
1174148, 21101161).
References
1
2
.
.
Li CZ, Zhao XC, Wang AQ, Huber GW, Zhang T (2015) Chem
Rev 115:11559–11624
Zaheer M, Kempe R (2015) ACS Catal 5:1675–1684
1
3

Highly Stable and Recyclable Graphene Layers Protected Nickel–Cobalt Bimetallic Nanoparticles…
3
4
.
.
Sturgeon MR, Kim S, Lawrence K, Paton RS, Chmely SC, Nim-
los M, Foust TD, Beckham GT (2014) ACS Sustain Chem Eng
18. Sturgeon MR, O’Brien MH, Ciesielski PN, Katahira R, Kruger
JS, Chmely SC, Hamlin J, Lawrence K, Hunsinger GB, Foust
TD, Baldwin RM, Biddy MJ, Beckham GT (2014) Green Chem
16:824–835
2
:472–485
Holladay JE, White JF, Bozell JJ, Johnson D (2007) Top Value-
added chemicals from biomass. Vol. II-results of screening for
potential candidates from biorefinery lignin. Pacific Northwest
National Laboratory, Richland, WA, p PNNL-16983
Vanholme R, Demedts B, Morreel K, Ralph J, Boerjan W (2010)
Plant Physiol 153:895–905
19. Cui XJ, Yuan HK, Junge K, Topf C, Beller M, Shi F (2017) Green
Chem 19:305–310
20. Cole-Hamilton DJ (2003) Science 299:1702–1706
21. Corma A, Garcia H, Xamena FXL (2010) Chem Rev
110:4606–4655
5
.
6
7
8
.
.
.
Bjorsvk HR, Liguori L (2002) Org Process Res Dev 6:279–290
Bozell JJ (2014) Top Curr Chem 353:229–255
22. Ogasawara S, Kato S (2010) J Am Chem Soc 132:4608–4613
23. Schoemaker HE, Mink D, Wubbolts MG (2003) Science
299:1694–1697
Pandey A, Bhaskar T, Stocker M, Sukumaran R (2015) Recent
advances in thermochemical conversion of biomass. Elsevier,
Waltham
24. Friedfield MR (2013) Science 342:1076–1080
25. Galyis HMT (2012) Science 335:835–838
26. Westerhaus FA, Jagadeesh RV, Wienhofer G, Pohl MM, Radnik
J, Surkus AE, Rabeah J, Junge K, Junge H, Nielsen M, Bruckner
A, Beller M (2013) Nat Chem 5:537–543
9
.
Strassberger Z, Alberts AH, Louwerse MJ, Tanase S, Rothenberg
G (2013) Green Chem 15:768–774
1
1
1
1
1
0. Chen MY, Huang YB, Pang H, Liu XX, Fu Y (2015) Green Chem
1
7:1710–1717
1. Zhou XY, Mitra J, Rauchfuss TB (2014) ChemSusChem
:1623–1626
2. Galkin MV, Sawadjoon S, Rohde V, Dawange M, Samec JSM
27. Calderone VR, Shiju NR, Curulla-Ferr D, Chambrey S, Khodakov
A, Rose A, Thiessen J, Jess A, Rothenberg G (2013) Angew Chem
Int Ed 52:4397–4401
7
28. Smith RDL, Prévot MS, Fagan RD, Trudel S, Berlinguette CP
(2013) J Am Chem Soc 135:11580–11586
(
2014) ChemCatChem 6:179–184
3. Kim JK, Lee JK, Kang KH, Lee JW, Song IK (2015) J Mol Catal
A 410:184–192
29. Chen B, Li F, Huang Z, Yuan G (2017) Appl Catal B 200:192–199
30. Mirza-Aghayan M, Tavana MM, Boukherroub R (2015) Catal
Commun 69:97–103
4. Zhang JG, Asakura H, van Rijn J, Yang J, Duchesne P, Zhang
B, Chen X, Zhang P, Saeys M, Yan N (2014) Green Chem
31. Galkin MV, Dahlstrand C, Samec JSM (2015) ChemSusChem
8:2187–2192
1
6:2432–2437
1
1
1
5. Zhang JG, Teo J, Chen X, Asakura H, Tanaka T, Teramura K, Yan
N (2014) ACS Catal 4:1574–1583
32. Tan CG, Grass RN (2008) Chem Commun 36:4297–4299
33. Si PZ, Zhang ZD, Geng DY, You CY, Zhao XG, Zhang WS (2003)
Carbon 41:247–251
6. He JY, Zhao C, Lercher JA (2012) J Am Chem Soc
1
34:20768–20775
7. Molinari V, Giordano C, Antonietti M, Esposito D (2014) J Am
Chem Soc 136:1758–1761
1
3