Investigation on the cleavage of Β-O-4 linkage in dimeric lignin model compound over nickel catalysts supported on ZnO-Al2O3 composite oxides with varying Zn/Al ratios

Article abstract of DOI:10.1016/j.cattod.2017.05.048

Catalytic depolymerization of lignin is still a challenge due to its low conversion and repolymerization of the reactive intermediates. Reductive depolymerization over supported nickel catalysts with probable surface acidic and basic properties is a very promising process. It is therefore very important to investigate the effect of acidity and basicity of the supports on catalytic reactivity. In this paper, we synthesized a series of nickel based catalysts supported on ZnO-Al₂O₃ composites with varying Zn/Al atom ratios (Zn/Al = 2, 3, 5, ∞) and tested their catalytic performances over a model compound 2-phenoxy-1-phenylethanone containing β-O-4 bond. All these catalysts showed 100% conversion by reacting at 250 °C for 2 h under 2 MPa of H₂. Higher selectivity towards ethylcyclohexane could be obtained over the catalyst Ni/ZnO-Al₂O₃-5. The possible cleavage pathways of selective oxidized β-O-4 ether linkage have been proposed.

Full text of DOI:10.1016/j.cattod.2017.05.048

CatalysisTodayxxx(xxxx)xxx–xxx
Contents lists available at ScienceDirect
Catalysis Today
journal homepage: www.elsevier.com/locate/cattod
Investigation on the cleavage of β-O-4 linkage in dimeric lignin model
compound over nickel catalysts supported on ZnO-Al2O3 composite oxides
with varying Zn/Al ratios
Chen Xua,b,1, Si-Fu Tanga,1, Xianyong Sunc, Yuanyuan Suna, Guangci Lia, Jingbo Qia,b
,
Xiaoyu Lia,b, Xuebing Lia,⁎
a
Key Laboratory of Biofuels, Qingdao Institute of Bioenergy and Bioprocess Technology, Chinese Academy of Sciences, Qingdao, 266101, PR China
University of Chinese Academy of Sciences, Beijing, 100049, PR China
b
c
National Institute of Clean-and-Low-Carbon Energy (NICE), Beijing, 102211, PR China
A R T I C L E I N F O
A B S T R A C T
Keywords:
Lignin
HDO
β-O-4 linkage
Catalytic depolymerization
ZnO-Al2O3 composite
Catalytic depolymerization of lignin is still a challenge due to its low conversion and repolymerization of the
reactive intermediates. Reductive depolymerization over supported nickel catalysts with probable surface acidic
and basic properties is a very promising process. It is therefore very important to investigate the effect of acidity
and basicity of the supports on catalytic reactivity. In this paper, we synthesized a series of nickel based catalysts
supported on ZnO-Al2O3 composites with varying Zn/Al atom ratios (Zn/Al = 2, 3, 5, ∞) and tested their
catalytic performances over a model compound 2-phenoxy-1-phenylethanone containing β-O-4 bond. All these
catalysts showed 100% conversion by reacting at 250 °C for 2 h under 2 MPa of H2. Higher selectivity towards
ethylcyclohexane could be obtained over the catalyst Ni/ZnO-Al2O3-5. The possible cleavage pathways of
selective oxidized β-O-4 ether linkage have been proposed.
1. Introduction
CeOeC bonds are the heart aspects of these conversions.
Some technologies (thermochemical, hydrolytic, reductive, and
The increasing energy demand and fossil fuel depletion have
inspired great interest in developing alternative energy sources, espe-
cially in direct transformation of biomass into hydrocarbon fuels.
Lignin, the second most abundant lignocellulose resource, is rich in
aromatic rings and energy density that makes it attractive as a
promising renewable bio-resource to produce fuels and chemicals [1].
Although lignin constitutes 15–30% by weight [1,2] and ca. 40% of the
energy [1–3] of lignocellulosic biomass, it is currently of low utilization
and especially burned as a low value fuel in industrial applications [2].
Lignin is a three-dimensional, highly branched, polyphenolic substance
containing plenty of hydroxy- and methoxysubstituted phenylpropane
units [4] through several types of CeOeC (β-O-4, α-O-4, 4-O-5, and so
on) and CeC (5-5′, β-1, β-5, β-β) linkages [2]. Theoretical calculations
have revealed that the ether bonds are more fragile than the CeC
bonds. Therefore, it is currently feasible to break down the former ones
while retaining the latter ones which can be utilized to optimize the
composition of traffic fuel via subsequent deoxygenation and hydro-
genation at the same time. So, selective cleavage and deoxygenation of
oxidative approaches) have been proposed to depolymerize the com-
plex lignin into phenolic mixtures which are called bio-oil. Among these
technologies, reductive depolymerization has been realized a very
promising strategy for the producing of fuel additives and aromatic
chemicals. Comparing with other technologies, it combines the depo-
lymerization and hydrodeoxygenation (HDO) reactions in a single step
[5]. In the presence of hydrogen, not only the complex structure of
lignin can be broken apart but also the oxygen content of the product
can be greatly reduced to increase the energy density which is usually
denounced in the fast pyrolysis process [6]. For example, Kou and co-
workers have reported that lignin can be catalytically depolymerized to
yield 46 wt% monomeric phenols over noble metal catalysts (Pt, Ru, Pd,
and Rh supported on activated carbon) under 4 MPa of H2 [7]. Raw
woody biomass can be hydrogenated to diols and monophenols (yield:
46.5%, based on lignin) at 235 °C under 6 MPa of H2 over Ni-W2C
catalyst supported on carbon [8]. Abu-Omar and co-workers reported
the selective cleavage and HDO of the β-O-4 ether linkages in wood
lignin and model compounds over a bimetallic Pd/C and ZnII catalyst
⁎
Corresponding author.
E-mail address: lixb@qibebt.ac.cn (X. Li).
These authors contributed equally to this paper.
1
http://dx.doi.org/10.1016/j.cattod.2017.05.048
Received 14 November 2016; Received in revised form 20 March 2017; Accepted 11 May 2017
0920-5861/©2017ElsevierB.V.Allrightsreserved.
Pleasecitethisarticleas:Xu,C.,CatalysisToday(2017),http://dx.doi.org/10.1016/j.cattod.2017.05.048

C. Xu et al.
CatalysisTodayxxx(xxxx)xxx–xxx
system under moderate reaction conditions (150–225 °C, 300–500 psi
H2) [9]. Xu and co-workers reported high yields and selectivity of 4-
ethylguaiacol and 4-propylguaiacol via the catalytic conversion of
lignosulfonate over heterogeneous nickel catalysts [10].
thus facilitates the depolymerization of lignin [24,25]. Therefore, it is
quite typical to select it as model compound for the study of lignin
depolymerization. The effect of the Zn/Al atom ratio in the support on
the cleavage of β-O-4 linkage was investigated and the possible reaction
pathways proposed.
To have a better understanding of the depolymerization mechanism
and simplify the analysis of the products, monomeric and dimeric
substituted ethers and phenols have been selected as model compounds
to mimic the depolymerization and HDO of lignin and lignin-derived
bio-oils over supported metal catalysts. The commercial sulfided CoMo
and NiMo catalysts, which are widely used in the hydrodesulfurization
(HDS) of petroleum, can also be used for the HDO of bio-oil [11].
However this kind of sulfided catalysts can cause serious sulfur
contamination and have poor stability. Noble metals (Ru, Pd, Pt, Re,
and Rh), base metals (Ni and Cu), and metal phosphides and carbides
supported on activated carbon, alumina or silica have also been
reported in the hydrocracking/upgrading of lignin model compounds
to produce aromatic compounds and alkanes [12–15]. The noble metals
show superior catalytic performance than the transition metals while
the limited reserves and high cost inhibit their application in a large
scale. In recent years, nickel-based catalysts have attracted more and
more attentions and shown excellent chemoselectivity for aromatic
products or high activity for CeO bond cleavage [16,17]. The work
reported by Xu et al. proved that nickel not only had high activity for
CeO bond cleavage but also had a propensity towards the protection of
the benzene ring [10]. Lercher and coworkers reported that quantita-
tive yields of C5–C9 gasoline-range hydrocarbons could be obtained
from hexane-extracted pyrolysis oil over Ni/HZSM-5 under mild
conditions (250 °C and 5 MPa H2) in water [18]. Supported nickel
catalyst (Ni/SiO2) could selectively and quantitatively cleave C−O
bonds in aromatic ethers to generate smaller aromatic molecules,
cycloalkanes, and cyclohexanol under very mild conditions in water
[19].
It should be noted that the surface acidic and basic properties of a
catalyst may have a remarkable impact on its catalytic reactivity
because of their different adsorption and desorption behaviors towards
reactants and products [20]. Metal catalyst supported on acidic
supports has been considered as the most effective catalyst for the
HDO of lignin and lignin-derived bio-oil. Previous studies have revealed
that the hydroprocessing of lignin and lignin-derived bio-oil involves
cascade of reactions including hydrogenolysis, hydrolysis, hydrogena-
tion and dehydration [21]. The active metal phase can catalyze the
cleavage of CaryleOH, CaryleOCH3, and CalkyleOR bonds via hydro-
genolysis whereas the acid sites can catalyze the cleavage of CaryleOR
and CalkyleOR bonds through hydrolysis. However, this kind of bifunc-
tional catalyst system suffers from remarkable carbon deposition and
moderate hydrothermal stability. In contrast, catalysts supported on
solid base have long been regarded as robust catalysts and show the
resistance to carbon deposition but have moderate HDO catalytic
activity [22]. It is therefore highly desirable to combine both the
advantages of these two catalytic processes. Additionally, solid base
substrates are mesoporous materials which are benefit to the transfer
and diffusion of depolymerization monomer. The oxygen vacancy on
the solid base surface is similar to that on the sulfide type catalyst
surface and is advantageous to the removal of oxygen atoms. In our
previous work, it was found suitable amount of basicity on the support
was favorable for oleic acid deoxygenation [23]. Nickel catalyst
supported on ZnO-Al2O3 composites exhibited excellent catalytic
activity and stability without obvious coke formation.
2. Experimental
2.1. Catalysts and model compound preparation
The supports were prepared through hydrothermal synthesis pro-
cess and followed by thermal treatment at high temperature [23].
.
.
Briefly, a certain molar ratio of Zn(NO3)2 6H2O, Al(NO3)3 9H2O and
urea were dissolved in 40 mL of deionized water to form a colorless
solution, where the molar ratio of [urea]/[NO3−] was set at 3.0. Then
the solution was transferred into a 100 mL Teflon-lined stainless steel
autoclave and heated at 180 °C for 3 h. The obtained precipitate was
collected by filtration, washed with deionized water to get neutral pH,
and then dried overnight in an oven at 80 °C. Finally, the dry substance
was calcined in air from room temperature to 500 °C at a heating rate of
2 °C/min and maintained at that temperature for 4 h, leading to the
formation of ZnO-Al2O3 support. In order to discuss the effect of surface
acidic and basic properties of support on the hydroprocessing of
dimeric lignin model compound, a series of supports with varying
Zn/Al atom ratios (2, 3, 5 and ∞) were prepared using the same
procedure as above.
Nickel based catalysts supported on the above ZnO-Al2O3 supports
were prepared by an incipient wetness impregnation method and the
procedure was carried out as follows: An aqueous of solution of Ni
(NO3)2∙6H2O (3.4 mmol) corresponding to a 10.0 wt% metal loading
was added dropwise onto 1.8 g of the as-prepared ZnO-Al2O3 support
with continuously agitation at room temperature for 2 h. Then, the
obtained substance was dried overnight at 80 °C and calcined in a flow
of N2 at 400 °C for 4 h (flow rate: 100 mL min−1), followed by
reduction at 500 °C in a flow of H2 for 4 h (flow rate: 100 mL min−1).
The heating rate for calcination and reduction was 2 °C/min. A series of
nickel-based catalysts were then obtained and labeled as Ni/ZnO and
Ni/ZnO-Al2O3-n (n = 2, 3 and 5), where n indicates the Zn/Al atom
ratio.
The dimeric model compound (2-phenoxy-1-phenylethanone) was
synthesized from the reaction of phenol with 2-bromoacetophenone
according to literature procedure [26]. The detailed synthesis proce-
dure and characterization data can be found in the supporting
information (Fig. S1).
2.2. Characterization
The textural properties of supports were determined by nitrogen
adsorption/desorption isotherms that were acquired at −196 °C on
Micromeritics ASAP 2020 adsorption analyzer. Prior to adsorption-
desorption experiments, each sample was degassed in vacuum at 200 °C
for 200 min. The specific surface area was calculated by Brunauer-
Emmett-Teller (BET) method. The pore volumes and pore size were
determined by the Barrett–Joyner–Halenda (BJH) method from deso-
rption branch of the isotherms.
The crystalline structures of supports were characterized by powder
X-ray diffraction (XRD) on a Bruker AXS-D8 Advance diffractometer
using Cu Kα radiation (λ = 1.5406 Å) generated at 40 kV and 40 mA.
The patterns were recorded at a scanning rate of 4°/min from 5° to 80°
(2θ).
In the present work, a series of nickel catalysts supported on ZnO-
Al2O3 composites with varying Zn/Al atom ratios (Zn/Al = 2, 3, 5, ∞)
have been prepared and tested for the hydroprocessing of lignin model
compound (2-phenoxy-1-phenylethanone). 2-Phenoxy-1-phenyletha-
none contains a β-O-4 bond which is the most abundant one among
the ether linkages of lignin. Additionally, it has been reported that the
selective oxidation of the β-O-4-alcohol to β-O-4-ketone can decrease
the bond dissociation energy (BDE) from 274.0 to 227.8 kJ mol−1 and
The surface basic properties of the catalysts were determined using
carbon dioxide temperature programmed desorption (CO2-TPD) on a
Micromeritics AutoChem II 2920 instrument. Each sample was loaded
in a U-shaped quartz cell and pretreated in Ar (flow rate: 20 mL min−1
)
at 150 °C for 2 h (ramp rate: 10 °C min−1), then cooled to 100 °C
followed by changing the gas flow to a mixture of 10% vol. CO2/He
2

C. Xu et al.
CatalysisTodayxxx(xxxx)xxx–xxx
(flow rate: 20 mL min−1) for 150 min. The sample was then cleaned up
with He (20 mL min−1) at 100 °C for 1 h to eliminate free adsorbed
CO2. Subsequently, the sample was heated up to 900 °C at a rate of
5 °C min−1 and maintained for 30 min. The desorbed CO2 was mon-
itored with a thermal conductivity detector (TCD) during this process.
The acid sites in the supports were determined from the IR spectra
of adsorbed pyridine (Py-IR) which were recorded at room temperature
on a Thermo Nicolet 6700 FTIR Spectrometer with a resolution of
0.5 cm−1. The sample was pressed into a thin self-supporting wafer and
activated in a vacuum (p = 10−7 mbar) at 350 °C for 1 h (heating
rate = 5 °C min−1). After cooling to ambient temperature, the acti-
vated sample was exposed to 0.1 mbar of pyridine vapor for 1 h,
followed by outgassing at 150 °C for 2 h.
The reduction process of Ni species of the catalysts was investigated
by means of hydrogen temperature programmed reduction (H2-TPR)
measurements on a Micromeritics AutoChem II 2920 chemisorption
analyzer. Each sample was charged in a U-shaped quartz cell and
pretreated in Ar (flow rate: 20 mL min−1) at 300 °C for 1 h (ramp rate:
10 °C min−1), then cooled to 100 °C. Afterwards, a mixture of 10 vol.%
Fig. 1. XRD patterns of different supports: ZnO (a), ZnO-Al2O3-5 (b), ZnO-Al2O3-3 (c),
ZnO-Al2O3-2 (d).
H2/Ar was introduced into the sample tube at
a flow rate of
20 mL min−1. After blowing for 20 min, the sample was heated to
600 °C at a rate of 10 °C min−1 and maintained at that temperature for
20 min, during which the hydrogen consumption was measured with a
TCD detector.
corresponding to the characteristic reflections of ZnO phase (JCPDS
#65-3411). For the ZnO-Al2O3 composite oxides, their XRD patterns are
similar and mainly consisted of ZnO phase and ZnAl2O4 phase (JCPDS
#55-0669). With the decrease of Zn/Al ratio, the diffraction peaks of
ZnO phase get visibly weaker (Fig. 1b–d) while the peaks of ZnAl2O4
phase become evident (31.2, 36.8, 44.8, 58.5, 65.4°), indicating the
increasing content of spinel phase. The diffraction peaks of ZnO-Al2O3
composite oxides are much weaker and broader than those of ZnO,
suggesting their modrate crystallinity and highly dispersion of ZnAl2O4
phase in ZnO phase.
The pore structure parameters of the supports, including specific
surface area (SBET), pore volume and average pore size are shown in
Table 1. A clear decreasing trend of SBET with the increase of Zn/Al
atom ratio is observed. As we know that particles with smaller size may
possess larger SBET. On the contrary, support comprises of larger
secondary particles will have a lower SBET, such as ZnO in the present
work. Additionally, the change of particle size affects the pore volume
of the supports as well. For a certain amount of solid powder, larger
particles will occupy more inner space, which leads to a decrease in
pore volume.
The morphologies of the supports were observed with a Hitachi S-
4800 field emission scanning electron microscopy (FE-SEM) with an
accelerating voltage of 1.5 kV. The reduced catalysts were ultrasoni-
cally dispersed in ethanol and dropped onto a carbon-coated copper
grid. Then it was tested by transmission electron microscopy (TEM)
analysis which was carried out on a JEM-2100HR electron microscopy
with an accelerating voltage of 200 kV. At least 200 Ni particles were
measured in order to determine Ni nanoparticle size distribution.
The electronic states of surface active metal of the nickel catalysts
were analyzed by means of X-ray photoelectron spectroscopy (XPS) on
an Escalab 250Xi Thermo Spectrometer equipped with a monochro-
matic X-ray source of Al Kα under ultra-high vacuum (2–3 × 10−6 Pa)
and a hemispherical analyzer. The XPS data related to the Ni2p core
level were recorded for the samples. The obtained spectra were fitted by
mixed Gaussian-Lorentzian functions. The respective binding energies
(BE) were calibrated against the C1 s peak of carbon (284.8 eV),
respectively.
To investigate the influence of Zn/Al atom ratio on the morphology
and microstructure of the supports, all the supports were characterized
by means of SEM, and the images are shown in Fig. 2. It can be found
that the ZnO-Al2O3 composite oxides (Fig. 2d to b) mainly exhibit
lamellar and particle morphology and the proportion of lamellar
structure as well as their size increased gradually with the increase of
Zn/Al atom ratio. At the same time, the shape of these lamellas becomes
more regular and smoother (Fig. 2d to a). Combining with the results in
Fig. 1, it is proposed that the crystallites of ZnAl2O4 phase tend to
aggregate into larger irregular particles while that of ZnO phase
preferentially grow into regular micro-sheets, and this difference
further affects the textural properties of the supports.
2.3. Catalytic reactions
All the experiments were carried out in a 300 mL batch stainless
autoclave equipped with a mechanical stirrer. In a typical reaction,
0.25 g of fresh catalyst, 1.0 g of lignin model compound and 50 mL of n-
hexane were added into the reactor. Prior to the reaction, the reactor
was purged with H2 for three times to exclude the residue air and then
pressurized with H2 to 2 MPa at ambient temperature. After that, the
reaction system was heated to reaction temperature and maintained for
a designated time with a stirring rate of 500 rpm. Finally, the reaction
system was cooled to ambient temperature, and the liquid products
were collected for subsequent analysis. The products were analyzed
using GC–MS (Agilent 7890A-5975C) equipped with a flame ionization
detector (FID) and a HP-5MS column.
To probe the acidities of the supports, IR spectra of adsorbed
pyridine (Py-IR) were recorded (Fig. 3). As expected, there was nearly
Table 1
3. Results and discussion
Textural properties of the supports with different Zn/Al ratios.
SBET (m2/g)
Pore volume (cm3/g)
Average pore size (nm)
3.1. Characterization of the supports and catalysts
Sample
ZnO-Al2O3-2
ZnO-Al2O3-3
ZnO-Al2O3-5
ZnO
121.5
121.3
119.2
21.05
0.21
0.20
0.18
0.01
6.91
6.60
6.04
1.90
3.1.1. Characterization of the supports
The powder X-ray diffraction patterns of the supports are shown in
Fig. 1. After heat treatment, all precursors were transformed to ZnO or
ZnO-Al2O3 composite oxides. The ZnO sample exhibits sharp peaks at
2θ ca. 31.8, 34.5, 36.3, 47.6, 56.6, 62.9, 66.4, 68.0, 69.1, 72.6, 76.9°,
The average pore size data were calculated by this formula (4 × Pore volume/SBET).
3

C. Xu et al.
CatalysisTodayxxx(xxxx)xxx–xxx
Fig. 2. SEM images of the supports: ZnO (a), ZnO-Al2O3-5 (b), ZnO-Al2O3-3 (c), ZnO-Al2O3-2 (d).
Fig. 3. The infrared spectra of adsorbed pyridine of the suports: ZnO (a), ZnO-Al2O3-5 (b),
Fig. 4. XRD patterns of the nickel catalysts: Ni/ZnO (a), Ni/ZnO-Al2O3-5 (b), Ni/ZnO-
ZnO-Al2O3-3 (c), ZnO-Al2O3-2 (d).
Al2O3-3 (c), Ni/ZnO-Al2O3-2 (d).
no acid in ZnO because of its alkaline character. For the zinc aluminum
composite oxide supports, the peaks around 1450 cm−1 clearly indicate
the existence of Lewis acid sites, whereas the peaks around 1540 cm−1
are absent, suggesting the composite oxide supports have only Lewis
acid sites. The area at 1450 cm−1 for Lewis acid sites increased with the
increasing of Al content which suggusted the acidity increased in the
support.
diffraction peaks and these peaks are obvious narrower than those of
the corresponding supports, indicating better crystallinity or larger size
after calcination and reduction. Besides the ZnO and ZnAl2O4 peaks, the
diffraction peaks of NiO phase also appear on the catalyst samples
(2θ= 43.6 and 62.9°, JCPDS #44-1159) which can be assigned to the
(012) and (110) planes. The NiO phase could be formed during the
passivation process using 10 vol.% O2/He. Because of the low content
and small particle size, no evident diffraction peak of nickel metal can
be detected. But the color change from grey to black after reduction
adequately indicated the Ni species on the surface have been reduced.
3.1.2. Characterization of the catalysts
Fig. 4 shows the diffraction patterns of the catalyst samples. As
expected, these catalyst samples exhibit characteristic ZnO and ZnAl2O4
4

C. Xu et al.
CatalysisTodayxxx(xxxx)xxx–xxx
characterized using TEM. As shown in Fig. 6, the Ni particles in the
used catalysts are still well dispersed. The average particle sizes are
around 35.1, 22.3, 18.2, 13.1 nm, respectively, which are very similar
to those of fresh catalysts. This indicates that these catalysts are very
stable under current reaction conditions.
Table 2
Textural properties of the Ni-based catalysts with different Zn/Al ratios.
Sample
SBET (m2/g)
Pore volume (cm3/g)
Average pore size (nm)
Ni/ZnO-Al2O3-2
Ni/ZnO-Al2O3-3
Ni/ZnO-Al2O3-5
Ni/ZnO
71.2
64.5
53.0
3.425
0.14
0.18
0.15
0.003
7.8
11.2
11.2
3.5
To gain the basicity information of the catalysts, the catalysts were
tested by CO2-TPD as shown in Fig. 7. The strength and amount of basic
sites are reflected by the desorption temperature and the peak area in
the TPD plots, respectively. In this study, the catalysts supported on
ZnO-Al2O3 composite oxides exhibit similar profiles showing three
desorptioni peaks at about 150, 250 and 350 °C, corresponding to weak,
moderate and strong basic sites, respectively[27,28]. For the catatlyst
supported on ZnO, a desorption peak is observed at about 650 °C,
suggesting stronger basicity. It can be seen that the total basicity of
these catalysts increases with the content of Al, which is consistent to
the order of surface area, demonstating that the surface area has a
significant influence on amount of basic sites. Among these catalysts,
Ni/ZnO-Al2O3-5 possesses larger specific surface area and moderate
basicity which are advantageous to the reaction.
In order to further explore the force between the support and nickel
species, unreduced nickel-based catalysts were tested by H2-TPR as
shown in Fig. 8. It was found that the H2-TPR curves of the unreduced
catalysts supported on composite oxides displayed one reduction peak
at about 260 °C which could be attributed to the reduction of oxidized
nickel nanoparticles by air in the atmospheric environment [29].
Comparing with bulk NiO (reduced at 362 °C) in nature [30], the
reduction temperature of “re-oxidized” NiO phase is lower. The sample
supported on ZnO displayed a reduction peak at about 400 °C, indicat-
ing stronger interaction between the support and NiO. Additionaly, the
peak appearing above 500 °C, could be ascribed to the dehydration of
Al-OH groups (see Fig. S2) and more difficult reducible NiO because
nickel aluminum spinel can be formed from the reaction of Ni species
with surface Al-OH groups of the supports. Through integration the
area of the reduction peaks, the hydrogen consumptions per gram
catalyst before 500 °C are calculated to be: 1.177, 0.471, 0.451,
The average pore size data were calculated by this formula (4 × Pore volume/SBET).
The textural properties of the catalysts are shown in Table 2. It is
found that the surface area values of the nickel catalysts are obvious
less than those of the supports but showing the same trend which
decreases with the increasing Zn/Al atom ratio. The introduction of
nickel could block the pores of the supports, resulting in smaller BET
surface area. The average pore sizes of the catalysts are in the range of
7.8 nm to 11.2 nm, demonstrating their mesoporous characters.
The morphology and metal dispersion of the catalysts were char-
acterized by TEM as shown in Fig. 5. It can be seen that the Ni particles
in the four catalysts all display regular shapes and are well dispersed on
the composite oxide supports. These Ni particles can steadily attach to
the surface of the supports even though the catalysts were prepared at
500 °C, demonstrating the good anti-sintering ability of this kind of
composite oxide supports. The Ni particle size decreases from 36.2 nm
to 12.1 nm with the Zn/Al atom ratio varies from ∞ to 2. One reason
could be related to the specific surface area of the catalysts, because
large surface area is advantageous to the dispersion of Ni particles and
results in smaller particles. However, the BET surface areas of the
supports are very close (see Table 1). The underlying reason could be
the Ni-support interaction. Ni species preferably interact with the acidic
alumina support to form NieOeAl bonds which can potentially inhibit
the migration and aggregation of Ni species and improve the dispersion.
With the increase of Zn content, the amount of acid sites is decreased
and the metal-support interaction is weakened.
The spent catalysts (reaction temperature: 250 °C) were also
Fig. 5. TEM of the fresh catalysts: Ni/ZnO (a), Ni/ZnO-Al2O3-5 (b), Ni/ZnO-Al2O3-3 (c), Ni/ZnO-Al2O3-2 (d).
5

C. Xu et al.
CatalysisTodayxxx(xxxx)xxx–xxx
Fig. 6. TEM of the spent catalysts: Ni/ZnO (a), Ni/ZnO-Al2O3-5 (b), Ni/ZnO-Al2O3-3 (c), Ni/ZnO-Al2O3-2 (d). Reaction condition: catalyst 0.25 g, 2-phenoxy-1-phenylethanone (1.0 g), n-
hexane (50 mL), H2 (2.0 MPa), 250 °C, 2 h.
Fig. 8. H2-TPR curves of the unreduced catalysts: Ni/ZnO (a), Ni/ZnO-Al2O3-5 (b), Ni/
ZnO-Al2O3-3 (c), Ni/ZnO-Al2O3-2 (d).
Fig. 7. The CO2-TPD curves of the catalysts: Ni/ZnO (a), Ni/ZnO-Al2O3-5 (b), Ni/ZnO-
Al2O3-3 (c), Ni/ZnO-Al2O3-2 (d).
electron spectroscopy (XPS) as shown in Fig. 9. The two peaks
appearing at 879.8 and 873.4 eV can be ascribed to Ni2p1/2, whereas
the peaks of Ni2p3/2 can be found in this range of 851.9–860 eV and can
be deconvoluted into three peaks. The peak at 852.2 eV can be ascribed
to the nickel metal phase, while the peaks at 854.6 eV and 856.1 eV
belong to the nickel oxide and nickel aluminate phase, respectively
[31–33]. The integration of the three peaks reveals the content of nickel
0.439 mmol g−1 for the unreduced samples supported on ZnO and Zn-
Al2O3-n (n = 5, 3 and 2), respectively. It should be noted that the
theoretical hydrogen consumption is 1.7 mmol g−1 if NiO is totally
reduced for all the catalysts, which explains less NiO was reduced in the
reduction condition.
In order to determine the electronic states of the surface nickel
species, the reduced nickel catalysts were analyzed by X-ray photon-
6

C. Xu et al.
CatalysisTodayxxx(xxxx)xxx–xxx
Fig. 9. XPS spectra of Ni2p for the catalysts: Ni/ZnO (a), Ni/ZnO-Al2O3-5 (b), Ni/ZnO-Al2O3-3 (c), Ni/ZnO-Al2O3-2 (d).
aluminate is the most. It is found the percentage of surface metallic
nickel decreases with the Zn/Al ratio reaching a minimum at Zn/Al = 5
and increasing again for Ni/ZnO (see Fig. S3).
the catalysts could be improved by increasing the Zn/Al atom ratio of
the support. On the contrary, aromatic ring can be retained and results
in high selectivity towards ethylbenzene. From the above results, it may
be concluded that bond cleavage (hydrogenolysis) and hydrogenation
of the dimeric model compound are favored over the highly basic
support. However for the catalyst supported on pure ZnO (Ni/ZnO), the
selectivity of ethylcyclohexane (15.0%) was even lower than that of Ni/
ZnO-Al2O3-2 (20.4%). Probably, it could be ascribed to the smaller
specific surface area of ZnO support which can lead to poor dispersion
of the active nickel phase (see Table 2). Considering the high selectivity
of ethylcyclohexane obtained on Ni/ZnO-Al2O3-5, it was selected as the
catalyst for the subsequent studies.
XPS measurements were also performed on the used catalysts to
explore the change of oxidation state of nickel after reaction (reaction
temperature: 250 °C) (Fig. 10). It was found the percentages of surface
metallic nickel in the used catalysts except Ni/ZnO-Al2O3-2 increased
because the catalysts were further reduced during the reactions. It was
also found that the increasing extent reduced with the increasing of Al
content (see Fig. S3). The reason should be that nickel aluminum spinel
phase in the catalyst was difficult to reduce whereas the fraction of
spinel phase in the catalysts increased with the decreasing of Zn/Al
ratio. Additionally, with the increase of surface area of the catalysts, the
nickel particles became well dispersed and smaller which were difficult
to reduce due to size effect.
3.3. Effect of reaction temperature
The effect of reaction temperature on the conversion of the dimeric
model compound was investigated in the range of 120–250 °C to
evaluate the catalytic performance of Ni/ZnO-Al2O3-5. As can be seen
from Table 3, the conversion of the dimeric model compound at 120 °C
was only 20% and 2-phenoxy-1-phenylethanol was the dominant
product. This indicates this temperature was favor to the hydrogenation
of the carbonyl group but not sufficient for the cleavage of β-O-4 bond.
As the temperature was increased from 120 to 190 °C, the conversion
increased remarkably from 20 to 100%, demonstrating the significant
role of reaction temperature. The content of the three main products
(ethylcyclohexane, ethylbenzene and cyclohexanol) also increased
gradually while the content of 2-phenoxy-1-phenylethanol reduced
from 93.2% to 4.6%. With the further increase of temperature, the
selectivity of ethylcyclohexane can be enhanced greatly accompanied
3.2. Role of the support
In order to investigate the role of acidic and basic properties of the
support on the catalyst activity, the hydrotreating experiments of 2-
phenoxy-1-phenylethanone over different nickel catalysts with varying
Zn/Al atom ratios were performed, and the results are summarized in
Fig. 11. From the reaction results, it was found that complete conver-
sions of 2-phenoxy-1-phenylethanone could be achieved for all catalysts
at 250 °C with ethylcyclohexane, ethylbenzene and cyclohexanol being
the main products. With the increase of Zn/Al atom ratio from 2 to 5,
the selectivity of ethylcyclohexane increased from 20.4% to 45.2%
while the selectivity of ethylbenzene decreased from 29.9% to 4.5%.
This indicates the hydrogenation and deoxygenation performance of
7

C. Xu et al.
CatalysisTodayxxx(xxxx)xxx–xxx
Fig. 10. XPS spectra of Ni 2p for various catalysts after reaction: Ni/ZnO (a), Ni/ZnO-Al2O3-5 (b), Ni/ZnO-Al2O3-3 (c), Ni/ZnO-Al2O3-2 (d). Reaction condition: catalyst 0.25 g, 2-
phenoxy-1-phenylethanone (1.0 g), n-hexane (50 mL), H2 (2.0 MPa), 250 °C, 2 h.
At relatively lower temperature, the reaction mainly proceeds through
the hydrogenation of the keto group in β-O-4-ketone to produce β-O-4-
alcohol instead of hydrogenolysis of the β-O-4 ether bond. Therefore,
the conversion at 120 °C was quite low and 2-phenoxy-1-phenylethan-
1-ol was observed as the dominant product. With the increase of
temperature, 2-phenoxy-1-phenylethan-1-ol can be either converted to
phenethoxybenzene or cleaved into phenol and 1-phenylethan-1-ol via
hydrogenolysis. These intermediates can be further converted to
cyclohexanol and ethylcyclohexane via cascade reactions of hydroge-
nolysis-hydrogenation (see Scheme 1(a)). The controlled experiments at
190 °C also confirmed this (see Table S1). The selectivity of 2-phenoxy-
1-phenylethan-1-ol diminished with the reaction time while the selec-
tivity of phenethoxybenzene increased with reaction time reaching a
maximum at 2 h and decreased again thereafter. There were 10% of
phenethoxybenzene and 3.7% of 2-phenoxy-1-phenylethan-1-ol in the
products even after three hours of reaction, suggesting that hydro-
genolysis could happen at lower temperature but still very slow. When
the reaction temperature was further increased above 200 °C, 2-
phenoxy-1-phenylethan-1-ol and phenethoxybenzene disappeared in
the products. With the extension of reaction time, the selectivity of
ethylcyclohexane increased while the selectivity of ethylbenzene di-
minished constantly (see Table S2). The selectivity of cyclohexanol and
1-cyclohexylethan-1-ol remains about 45% and 1.1%, respectively. It
could be concluded that the conversion at higher temperature occurred
in a different way as illustrated in Scheme 1(b). Because the bond
dissociation energy of β-O-4-ketone (227.8 kJ mol−1) is lower than that
of β-O-4-alcohol (274.0 kJ mol−1), the dimeric model compound can
be first cleaved into acetophenone and phenol via hydrogenolysis at
relatively higher temperature. The intermediate acetophenone is then
Fig. 11. Product distribution over the catalysts: Ni/ZnO (a), Ni/ZnO-Al2O3-5 (b), Ni/ZnO-
Al2O3-3 (c), Ni/ZnO-Al2O3-2 (d). Reaction condition: catalyst 0.25 g, 2-phenoxy-1-
phenylethanone (1.0 g), n-hexane (50 mL), H2 (2.0 MPa), 250 °C, 2 h.
by the reduction of ethylbenzene.
3.4. Reaction routes
Based on the above results, the possible reaction pathways of the
dimeric model compound (2-phenoxy-1-phenylethan-1-one) can be
proposed as shown in Scheme 1. The conversion of the dimeric model
compound mainly proceeds through two parallel competitive pathways.
8

C. Xu et al.
CatalysisTodayxxx(xxxx)xxx–xxx
Table 3
Product distribution of 2-phenoxy-1-phenylethanone over Ni/ZnO-Al2O3-5 as a function of temperature.
Reaction Temperature (°C)
Conversion (%)
Selectivity of the products (%)
others
120
160
180
190
200
250
20.0
87.0
93.9
100
100
100
0
0.9
9.1
14.4
25.8
26.8
4.5
5.9
0
0
0
0.3
0
0
0.02
0
0
0
0
0
0
1.3
2.8
4.8
2.0
0
93.2
50.4
34.0
4.6
0
2.1
6.4
9.3
13.7
45.2
19.3
23.1
39.7
48.9
47.0
1.5
0.7
0
0
0
3.7
6.5
0
0
0
1.0
1.0
2.1
4.6
1.3
3.5
2.8
1.6
0
5.8
5.9
0
0
0
3.2
4.1
14.2
1.3
0
0
0
Typical reaction conditions: Ni/ZnO-Al2O3-5 (0.25 g), 2-phenoxy-1-phenylethanone (1 g), n-hexane (50 mL), H2 (2 MPa), 2 h, stirred at 500 rpm.
hydrogenated either at the benzene group to produce 1-cyclohexy-
lethan-1-one or at the carbonyl group to produce 1-phenylethan-1-ol,
which can be further converted to cyclohexanol and ethylcyclohexane
via steps of hydrogenation-hydrogenolysis.
indicates the catalyst has outstanding adhesion of the Ni metal on the
support and maintained a high catalytic activity during the recycling
tests, which have been confirmed by the results of XPS and TEM
obtained on the spent catalysts.
3.5. Recycling tests
4. Conclusion
The recyclability of Ni/ZnO-Al2O3-5 was tested in n-hexane at
190 °C under 2 MPa H2 for 2 h (Fig. 12 and Table S3). After each
reaction, the catalyst was collected and washed with n-hexane and
dried at 70 °C for the next run. From the results, it can be seen that the
conversion of lignin model compound (2-phenoxy-1-phenylethanone)
slightly decreased from 100% to 95% after three runs of reactions. The
selectivities of the main products (ethylcyclohexane, ethylbenzene,
cyclohexanol) and the two intermediates (phenethoxybenzene and 2-
phenoxy-1-phenylethanol) are very similar in the three runs. This result
In summary, the catalytic hydroprocessing of 2-phenoxy-1-pheny-
lethan-1-one, a β-O-4-ketone dimeric lignin model compound, over a
series of nickel catalysts supported on ZnO-Al2O3 composite oxides was
studied. It was found that the Zn/Al atom ratio played an important role
in the selectivity of the products. Higher selectivity towards ethylcy-
clohexane could be obtained by increasing the Zn/Al atom ratio in the
catalyst, on the contrary higher selectivity of ethylbenzene could be
obtained instead. The reaction temperature had a major impact on the
catalytic hydroprocessing of β-O-4-ketone dimeric lignin model com-
Scheme 1. Proposed reaction pathways of 2-phenoxy-1-phenylethanone conversion on Ni/ZnO-Al2O3-5.
9

C. Xu et al.
CatalysisTodayxxx(xxxx)xxx–xxx
References
[1] J. Zakzeski, P.C.A. Bruijnincx, A.L. Jongerius, B.M. Weckhuysen, Chem. Rev. 110
(2010) 3552–3599.
[2] A. Rahimi, A. Ulbrich, J.J. Coon, S.S. Stahl, Nature 515 (2014) 249–252.
[3] D.M. Alonso, S.G. Wettstein, J.A. Dumesic, Green Chem. 15 (2013) 584–595.
[4] M. Stocker, Angew. Chem. Int. Ed. 47 (2008) 9200–9211.
[5] X. Huang, T.I. Korányi, M.D. Boot, E.J.M. Hensen, ChemSusChem 7 (2014)
2276–2288.
[6] T. Parsell, B.C. Owen, I. Klein, T.M. Jarell, C.L. Marcum, L.J. Haupert,
L.M. Amundson, H.I. Kenttämaa, F. Ribeiro, J.T. Miller, M.M. Abu-Omar, Chem. Sci.
4 (2013) 806–813.
[7] N. Yan, C. Zhao, P.J. Dyson, C. Wang, L.T. Liu, Y. Kou, ChemSusChem 1 (2008)
626–629.
[8] C.Z. Li, M.Y. Zheng, A.Q. Wang, T. Zhang, Energy Environ. Sci. 5 (2012)
6383–6390.
[9] T. Parsell, S. Yohe, J. Degenstein, T. Jarrell, I. Klein, E. Gencer, B. Hewetson,
M. Hurt, J. Kim, H. Choudhari, B. Saha, R. Meilan, N. Mosier, F. Ribeiro,
W.N. Delgass, C. Chapple, H.I. Kenttämaa, M. Abu-Omar, Green Chem. 17 (2015)
1492–1499.
[10] Q. Song, F. Wang, J. Xu, Chem. Commun. 48 (2012) 7019–7021.
[11] A.L. Jongerius, R. Jastrzebski, P.C.A. Bruijnincx, B.M. Weckhuysen, J. Catal. 285
(2012) 315–323.
Fig. 12. Recycling tests of Ni/ZnO-Al2O3-5. Reaction conditions: 1.0 g lignin model
compound, 0.25 g catalyst, n-hexane 50 mL, 2 MPa H2, 190 °C, 2 h, 500 rpm.
[12] H. Thomas, E. Maria, J. Brian, L. Andy, Catalysis of Organic Reactions, CRC Press,
2005, pp. 135–144.
[13] P.T. Patil, U. Armbruster, M. Richter, A. Martin, Energy Fuels 25 (2011)
4713–4722.
pound. With the increase of temperature, both the conversion and
selectivity increased greatly. It was also revealed that the conversion of
the dimeric model compound mainly proceeded through two parallel
competitive pathways. At relatively lower temperature, the reaction
mainly proceeded through the hydrogenation of the keto group in β-O-
4-ketone to produce β-O-4-alcohol instead of hydrogenolysis of the β-O-
4 ether bond. At relatively higher temperature (above 200 °C), the
dimeric model compound can be first cleaved into acetophenone and
phenol via hydrogenolysis. The intermediates can be further converted
to cyclohexanol and ethylcyclohexane via cascade reactions of hydro-
genolysis-hydrogenation. Further work will be undertaken to apply this
strategy to the depolymerization of real lignin.
[14] R.C. Runnebaum, T. Nimmanwudipong, D.E. Block, B.C. Gates, Catal. Sci. Technol.
2 (2012) 113–118.
[15] W. Xu, S.J. Miller, P.K. Agrawal, C.W. Jones, ChemSusChem 5 (2012) 667–675.
[16] A.G. Sergeev, J.F. Hartwig, Science 332 (2011) 439–443.
[17] X. Wang, R. Rinaldi, Energy Environ. Sci. 5 (2012) 8244–8260.
[18] W. Song, Y. Liu, E. Baráth, C. Zhao, J.A. Lercher, Green Chem. 17 (2015)
1204–1218.
[19] J. He, C. Zhao, J.A. Lercher, J. Am. Chem. Soc. 134 (2012) 20768–20775.
[20] S. Hu, M. Xue, H. Chen, J. Shen, Chem. Eng. J. 162 (2010) 371–379.
[21] C. Zhao, J.A. Lercher, Angew. Chem. Int. Ed. 51 (2012) 5935–5940.
[22] J. Long, Q. Zhang, T. Wang, X. Zhang, Y. Xu, L. Ma, Bioresour. Technol. 154 (2014)
10–17.
[23] G. Li, F. Zhang, L. Chen, C. Zhang, H. Huang, X. Li, ChemCatChem 7 (2015)
2646–2653.
[24] A. Rahimi, A. Ulbrich, J.J. Coon, S.S. Stahl, Nature 515 (2014) 249–252.
[25] C. Zhang, J. Lu, X. Zhang, K. MacArthur, M. Heggen, H. Li, F. Wang, Green Chem.
18 (2016) 6545–6555.
[26] A. Rahimi, A. Azarpira, H. Kim, J. Ralph, S.S. Stahl, J. Am. Chem. Soc. 135 (17)
(2013) 6415–6418.
Acknowledgment
[27] J.A. Cecilia, A. Infantes-Molina, E. Rodríguez-Castellón, A. Jiménez-López,
S.T. Oyama, Appl. Catal. B Environ. 136 (2013) 140–149.
[28] H. Zuo, Q. Liu, T. Wang, L. Ma, Q. Zhang, Q. Zhang, Energy Fuels 26 (2012)
3747–3755.
[29] W. Song, C. Zhao, J.A. Lercher, Chem. Eur. J. 19 (2013) 9833–9842.
[30] K. Hadjiivanov, M. Mihaylov, D. Klissurski, P. Stefanov, N. Abadjieva, E. Vassileva,
L. Mintchev, J. Catal. 185 (1999) 314–323.
This work is financially supported by National Natural Science
Foundation of China (Grant No. 21276263, 21761132006, U1162203
and 21676287).
[31] Q. Liu, F. Gu, X. Lu, Y. Liu, H. Li, Z. Zhong, G. Xu, F. Su, App. Catal. A Gen. 488
(2014) 37–47.
Appendix A. Supplementary data
[32] K.V. Manukyan, A.J. Cross, A.V. Yeghishyan, S. Rouvimov, J.J. Miller,
A.S. Mukasyan, E.E. Wolf, App. Catal. A Gen. 508 (2015) 37–44.
[33] Y. Bang, S. Park, S.J. Han, J. Yoo, J.H. Song, J.H. Choi, K.H. Kang, I.K. Song, Appl.
Catal. B Environ. 180 (2016) 179–188.
Supplementary data associated with this article can be found, in the
online version, at http://dx.doi.org/10.1016/j.cattod.2017.05.048.
10