Catalytic hydrodeoxygenation of anisole as lignin model compound over supported nickel catalysts

Article abstract of DOI:10.1016/j.cattod.2014.02.014

Catalytic hydrodeoxygenation (HDO) of anisole, a methoxy-rich lignin model compound, has been investigated over a series of Ni-containing (10 wt% loading) catalysts with activated carbon, SBA-15, SiO₂, and γ-Al ₂O₃ supports, in order to understand their ability for removal of OCH₃ from anisole. This catalytic reaction had been carried out in an autoclave at 180-220 °C and 0.5-3.0 MPa H₂ pressure. Nickel-catalyzed aromatic ring-hydrogenation compared with the subsequent demethylation and deoxygenation is the fastest step in HDO of anisole under the present reaction conditions. Among these catalysts, the aromatic ring-saturated cyclohexyl methyl ether is mainly obtained over Ni/AC, while the Ni/SiO₂ displayed the highest activity in HDO of anisole (selectivity to deoxygenated products >95%). Differences in HDO efficiency among the catalysts are attributed not only to variations in the dispersion of the active phase, but also to the acid sites which may contribute to the cleavage of CO bonds. It has also been shown that the activity toward oxygen-removal strongly depended on reaction temperature and the conversion of anisole favors the production of oxygen-free aromatics by the direct demethoxylation pathway at the relatively low H₂ pressure.

Full text of DOI:10.1016/j.cattod.2014.02.014

G Model
CATTOD-8905; No. of Pages8
ARTICLE IN PRESS
Catalysis Today xxx (2014) xxx–xxx
Contents lists available at ScienceDirect
Catalysis Today
journal homepage: www.elsevier.com/locate/cattod
Catalytic hydrodeoxygenation of anisole as lignin model compound
over supported nickel catalysts
Shaohua Jina, Zihui Xiaoa, Chuang Lia, Xiao Chena, Lei Wanga, Jiacheng Xinga,
Wenzhen Lib, Changhai Lianga,∗,1
a Laboratory of Advanced Materials and Catalytic Engineering, School of Chemical Engineering, Dalian University of Technology, Dalian 116023, China
b Department of Chemical Engineering, Michigan Technological University, Houghton, USA
a r t i c l e i n f o
a b s t r a c t
Article history:
Catalytic hydrodeoxygenation (HDO) of anisole, a methoxy-rich lignin model compound, has been inves-
tigated over a series of Ni-containing (10 wt% loading) catalysts with activated carbon, SBA-15, SiO2, and
␥-Al2O3 supports, in order to understand their ability for removal of OCH3 from anisole. This catalytic
reaction had been carried out in an autoclave at 180–220 ◦C and 0.5–3.0 MPa H2 pressure. Nickel-catalyzed
aromatic ring-hydrogenation compared with the subsequent demethylation and deoxygenation is the
fastest step in HDO of anisole under the present reaction conditions. Among these catalysts, the aromatic
ring-saturated cyclohexyl methyl ether is mainly obtained over Ni/AC, while the Ni/SiO2 displayed the
highest activity in HDO of anisole (selectivity to deoxygenated products >95%). Differences in HDO effi-
ciency among the catalysts are attributed not only to variations in the dispersion of the active phase, but
also to the acid sites which may contribute to the cleavage of C O bonds. It has also been shown that
the activity toward oxygen-removal strongly depended on reaction temperature and the conversion of
anisole favors the production of oxygen-free aromatics by the direct demethoxylation pathway at the
relatively low H2 pressure.
Received 6 November 2013
Received in revised form 1 January 2014
Accepted 8 February 2014
Available online xxx
Keywords:
Hydrodeoxygenation
Anisole
Methoxy group
Lignin
Supported-nickel catalysts
© 2014 Elsevier B.V. All rights reserved.
1. Introduction
Catalytic hydrodeoxygenation is considered to be one of the
most effective routes for lignin-derived bio-oils upgrading [6].
Faced with the exhaustion of fossil fuel, great attention has been
paid to develop a new-generation sustainable alternative energy.
The utilization of biomass to produce bio-fuels and its derivative
chemicals is an effective and sustainable way to relieve the energy
crisis [1]. Numerous studies have been investigated to convert cel-
lulose and hemicellulose to liquid fuels, such as alcohols or furan
compounds [2,3]. However, only approximately two-thirds of the
lignocellulosic biomass was utilized, as lignin occupies remaining
20–30% of total mass in the plants [4]. Lignin, as the most abundant
renewable resources composed of aromatic units in nature, and
the largest contributor to soil organic matter, can be broken down
to phenolic monomers by catalytic and thermochemical processes
[5]. However, these phenolic compounds cannot be used as a liq-
uid fuel directly owing to their high oxygen content, poor chemical
stability, corrosivity to the equipment [1].
Among the characteristic groups of lignin, methoxy group which
associated with phenylpropane, syringyl and guaiacyl units (as
shown in Scheme 1), is one of the major functional groups
and has different reactivities in chemical reactions [7]. Although
lots of works have been studied on the catalytic conversion of
lignin model compounds, the reports on the hydrodeoxygena-
tion of methoxy-rich compounds were limited. Consequently, it
is particularly important to find an effective catalyst for removal
of aromatic carbon–oxygen (C O) bonds. Conventional sulfide
NiMo and CoMo/␥-Al2O3 catalysts have been studied for cat-
alytic hydrotreatment of bio-oils and their model compounds
[8–10]. Noble metal catalysts, such as Pt/C, Pt-Sn/CNFs, Pd, Ru,
Rh/SiO2-Al2O3 [11–13], also exhibited excellent activities for HDO
of lignin-derived chemicals. However, sulfided catalysts are poten-
tially problematic due to sulfur contamination to the final products,
water-induced catalyst deactivation and coke accumulation [14].
In addition, the high cost of noble metal catalysts also limits their
wide application. Nickel-based catalysts, including nickel phos-
phide [15–18], seem to be a promising noble-metal free catalytic
system that can satisfy the HDO process requirements, such as low
cost, non-sulfided nature and high activity [19].
∗
Corresponding author. Fax: +86 411 84986353.
E-mail address: changhai@dlut.edu.cn (C. Liang).
1
http://finechem.dlut.edu.cn/liangchanghai.
http://dx.doi.org/10.1016/j.cattod.2014.02.014
0920-5861/© 2014 Elsevier B.V. All rights reserved.
Please cite this article in press as: S. Jin, et al., Catalytic hydrodeoxygenation of anisole as lignin model compound over supported nickel
catalysts, Catal. Today (2014), http://dx.doi.org/10.1016/j.cattod.2014.02.014

G Model
CATTOD-8905; No. of Pages8
ARTICLE IN PRESS
2
S. Jin et al. / Catalysis Today xxx (2014) xxx–xxx
˚
diffractometer using a Cu K␣ radiation (ꢀ = 1.5418 A), operated at
40 kV and 100 mA. Temperature-programmed reduction of hydro-
gen (H2-TPR) was performed in a stream of 10% H2 in Ar with a
flow rate of 50 cm3/min. The samples were heated up to a final
temperature of 900 ◦C at 10 ◦C/min and H2 consumption was mon-
itored by a thermal conductivity detector. In the case of NH3-TPD
experiments, the reduced catalysts were outgassed in He at 350 ◦C
for 0.5 h, and finally saturated at 100 ◦C in a 10% NH3/He stream
(50 ml/min) for 1 h. After removing most weakly physisorbed NH3
by flowing He (50 ml/min) for 30 min, the chemisorbed ammonia
was determined by using TCD by heating at 10 ◦C/min up to 700 ◦C
under the same flow of He. BET surface areas, pore volumes and
pore size distributions of the catalysts (approximately 0.1 g sample)
were determined by nitrogen physisorption at liquid N2 temper-
ature with a autosorb iQ automated gas sorption analyzer. The
amounts of metal active sites were estimated from irreversible CO
adsorption measurements performed at room temperature. Typi-
cally, the catalyst samples were reduced in a 10% H2/He at 400 ◦C,
followed by exposure to flowing He for 2 h. CO adsorption was con-
ducted when the sample was cooled to room temperature. Then,
the samples were purged by the flowing He for 1 h, and CO adsorp-
tion was repeated. The particle size and dispersion of the samples
were analyzed by transmission electron microscopy (TEM). Pow-
der samples were ultrasonicated in ethanol and dispersed on holey
carbon films on copper grids.
Scheme 1. Schematic representation of a typical lignin fragment.
Various catalyst supports appear to play a significant role in
deoxygenation of the lignin monomers. Ohta et al. found carbon-
supported Pt showed high activity and durability in the HDO of
phenols [11]. ␥-Al2O3 effectively promoted dehydration reaction
of the alcohol produced from hydrogenation of the aromatic ring
[20]. Mesoporous silica materials, in particular SBA-15, have gener-
ated interests as catalyst supports in the HDO of a biomass-derived
compounds due to their high structure regularity, low-cost and
non-toxicity [21]. SiO2 supported Fe also showed good activity
and selectivity toward conversion of guaiacol into aromatic hydro-
carbons [22]. However, a limited number of studies have been
performed on the effects of catalyst supports for the hydrotreating
of methoxy-rich lignin model compounds.
For those reasons, herein we report a screening study on the
catalytic HDO of methoxy-rich lignin model compound anisole over
Ni nanoparticles dispersed on different supports (active carbon, ␥-
Al2O3, SiO2 and SBA-15) under mild conditions (210 ◦C, 3.0 MPa),
in order to understand the effect of supports on the activity toward
the removal of oxygen from anisole.
2.3. Anisole HDO and product analysis
Anisole HDO was carried out in a 50 mL stainless steel auto-
clave equipped with a magnetic stirrer and an electric temperature
controller. Prior to reaction, the obtained Ni-based catalysts were
reduced by H2 at 400 ◦C for 2 h and then passivated in Ar overnight.
8 wt% anisole (1.2 g, 0.0108 mol) dissolved in 20 mL n-decane with
2 wt% n-dodecane as an internal standard for quantitative GC anal-
ysis and 0.1 g reduced catalysts were rapidly introduced into the
autoclave to prevent from contacting with air for long. After-
wards, the autoclave was sealed and purged repeatedly with H2
to eliminate air, pressurized to the desired hydrogen pressure
and then heated to the desired temperature at 700 rpm stir-
ring speed. The zero time point was defined when the autoclave
was heated. It takes about 30 min to reach reaction temperature.
The autoclave was cooled to room temperature and brought to
ambient pressure after the reaction was finished. The products
were analyzed by gas chromatograph (GC-7890F, FID, FFAP col-
umn 30 m × 0.32 mm × 0.5 ␮m) and identified by a GC–MS (Agilent
6890, HP-5 MS capillary column, 30 m × 0.25 mm × 0.25 ␮m).
2. Experimental
2.1. Catalyst preparation
SBA-15 mesoporous silica was synthesized following a reported
procedure [23]. The other three supports active carbon (AC), ␥-
Al2O3, SiO2 were purchased commercially. Before use, the AC
supports were washed to remove the impurities in a 0.5 mol/L
aqueous nitric acid solution at room temperature for 24 h. The sup-
port powders were all sieved to obtain particles smaller than 100
meshes before use and dried in air at 140 ◦C overnight. The 10 wt%
Ni catalysts supported on AC, SiO2, ␥-Al2O3 and SBA-15 were pre-
pared by an incipient impregnation method using the aqueous
solution of the metal salt Ni(NO3)2·6H2O. The precursor solutions
were added dropwise to the supports. After the impregnation, the
catalysts were dried at 110 ◦C in air for 12 h and then calcined at
400 ◦C in Ar for 3 h.
3. Results and discussion
3.1. Catalyst properties
For all catalysts, the chemical analysis reveals that the actual
metal loading is almost close to the theoretical value of 10%, as
shown in Table 1. N2 sorption analyses were performed to deter-
mine the differences in support morphology of all catalysts. The N2
adsorption–desorption isotherms and pore size distributions (PSD)
of all catalysts after calcination are shown in Fig. 1. AC and SiO2 sup-
ported Ni catalysts both showed a typical Type I isotherm according
to the IUPAC classification, suggesting a large number of micropores
existed, which also can be obtained from the PSD support. In addi-
tion, PSD curves of Ni/AC and SiO2 showed the presence of small
mesopores (smaller than 4 nm). Owing to the presence of abundant
micropores, Ni/AC and Ni/SiO2 had a high surface area of 1293 and
549 m2/g. Compare with Ni/SiO2, Ni/AC owned much more micro-
pores and larger pore volume of 0.6 cm3/g. As-synthesized SBA-15
2.2. Catalyst characterization
The content of Ni was measured by inductively coupled plasma-
optical emission spectroscopy (ICP-OES). After dissolved with
nitromurlatic acid solution, the samples were filtered and ana-
lyzed by ICP-OES. The phase structure of the samples was obtained
by powder X-ray diffraction (XRD) analysis in a D/MAX-2400
Please cite this article in press as: S. Jin, et al., Catalytic hydrodeoxygenation of anisole as lignin model compound over supported nickel
catalysts, Catal. Today (2014), http://dx.doi.org/10.1016/j.cattod.2014.02.014

G Model
CATTOD-8905; No. of Pages8
ARTICLE IN PRESS
S. Jin et al. / Catalysis Today xxx (2014) xxx–xxx
3
Table 1
Chemical and structure properties of the supported Ni catalysts.
b
b
b
Catalyst
ICP-OESa (wt%)
SBET (m2/g)
d (nm)b
Vmeso (cm3/g)
Vtotal (cm3/g)
Ni/AC
10.2
9.3
10.3
9.0
1293
229
662
549
1.9
7.0
6.0
2.2
0.22
0.34
0.85
0.16
0.60
0.40
1.00
0.31
Ni/␥-Al2O3
Ni/SBA-15
Ni/SiO2
a
As determined by the ICP-OES technique.
b
BET surface area (SBET), average pore diameter (d) and total pore volume (Vtotal) as determined by N2 adsorption–desorption isotherms at −196 ◦C.
supported Ni catalyst displayed a Type IV isotherm and standard
H1 type hysteresis loop by IUPAC classification, showing that the
existence of a highly ordered mesopores. The isotherms for the
Ni/SBA-15 materials showed a sharp inflection in the relative pres-
sure (P/P0) range from 0.7 to 0.8, reflecting the presence of uniform
PSD [24]. Due to well-defined mesoporosity of SBA-15, a high sur-
face area of 662 m2/g and narrow pore size distribution with an
average diameter of 8.2 nm (see Fig. 1c). Porous Ni/␥-Al2O3 catalyst
had a BET specific surface area of 230 m2/g and total pore volume
of 0.4 cm3/g. It contained both micropores and mesopores, with a
narrow micropore size distribution centered at 1 nm and a wide
mesopore size distribution from 3 to 8 nm, as shown in Fig. 1b.
The wide-angel XRD technique was performed to study the
formation of crystalline species for different supports loaded Ni
catalysts after reduction at 400 ◦C for 2 h. Owing to the forma-
tion of metallic nickel, the XRD patterns (Fig. 2) of reduced Ni/AC,
Ni/SBA-15 and Ni/SiO2 catalysts show peaks at 2ꢁ of 45.5◦, 51.9◦,
76.3◦, respectively corresponding to lattice plane (1 1 1), (2 0 0) and
(2 2 0) of metallic nickel (JCPDS 04-0850), along with amorphous
and broad active carbon peaks centered at 2ꢁ of 23◦ and 43◦ and
SiO2 diffraction peaks centered at 2ꢁ of 23◦. While Ni peaks of SBA-
15 catalysts are sharper than that of other catalysts, indicating the
Fig. 2. X-ray diffraction patterns of (a) Ni/AC, (b) Ni/␥-Al2O3, (c) Ni/SBA-15 and (d)
Ni/SiO2 after reduction.
Fig. 1. N2 adsorption/desorption isotherms with DFT pore size distributions ((a) 10% Ni/AC; (b) 10% Ni/␥-Al2O3; (d) 10% Ni/SiO2) and BJH pore size distributions ((c) 10%
Ni/SBA-15) derived from the adsorption branches of the isotherms.
Please cite this article in press as: S. Jin, et al., Catalytic hydrodeoxygenation of anisole as lignin model compound over supported nickel
catalysts, Catal. Today (2014), http://dx.doi.org/10.1016/j.cattod.2014.02.014

G Model
CATTOD-8905; No. of Pages8
ARTICLE IN PRESS
4
S. Jin et al. / Catalysis Today xxx (2014) xxx–xxx
existence of larger clusters. Fig. 2b mainly shows peaks of ␥-Al2O3,
with unapparent peaks of metallic Ni or NiO, suggesting the exis-
tence of very small particles [25], which is also consistent with
its TEM result. On the other hand, because of the strong interac-
tion between nickel particles and ␥-Al2O3 and the formation of an
amorphous surface nickel aluminate [25–27], the complete expo-
sition of activity phase of Ni/␥-Al2O3 may need higher reduction
temperature.
H2-TPR profiles of Ni catalysts on different supports are shown
in Fig. 3. The reduction peak at 365 ◦C for Ni/AC and Ni/SBA-15,
at 390 ◦C for Ni/SiO2 and at 430 ◦C Ni/␥-Al2O3, usually assigned to
the reduction of NiO clusters interacting weakly with the supports
[28]. In the case of Ni/AC, the second peak at 500 ◦C is attributed
to the reduction of highly dispersed NiO interacted with AC, which
is partially overlapped with the peaks of carbon methanation and
CO2 or CO from O-containing functional groups on the surface of AC
supports. For the Ni/␥-Al2O3 catalyst, it has a peak at 570 ◦C, which
indicates the presence of NiO contacted strongly with ␥-Al2O3. The
reduction of Ni/␥-Al2O3 catalyst displays a similar behavior com-
pared to the results in the reference, but is highly dependent on the
TPR conditions employed [25,29]. The reduction profile of Ni/SiO2 is
very similar to Ni/SBA-15, showing a broad peak centered at 624 ◦C,
which is because of the reduction of highly dispersed NiO contacted
strongly with the silica materials or nickel silicates formed during
Fig. 3. H2-TPR profiles of (a) Ni/AC, (b) Ni/␥-Al2O3, (c) Ni/SBA-15 and (d) Ni/SiO2
after calcined in Ar at 400 ◦C.
Fig. 4. TEM images of the reduced (a) Ni/AC, (b) Ni/␥-Al2O3, (c) Ni/SBA-15 and (d) Ni/SiO2.
Please cite this article in press as: S. Jin, et al., Catalytic hydrodeoxygenation of anisole as lignin model compound over supported nickel
catalysts, Catal. Today (2014), http://dx.doi.org/10.1016/j.cattod.2014.02.014

G Model
CATTOD-8905; No. of Pages8
ARTICLE IN PRESS
S. Jin et al. / Catalysis Today xxx (2014) xxx–xxx
5
Table 2
CO chemisorption and metallic particle sizes of the supported Ni catalysts.
a
b
c
Catalyst
CO uptakea (␮mol/g cat.)
D%a
dCO
dXRD
dTEM
Ni/AC
292
163
190
205
18.3
–
12.5
14.4
5.5
–
8.1
6.0
–
–
8.4
5.7
3.9
–
8.0
4.8
Ni/␥-Al2O3
Ni/SBA-15
Ni/SiO2
a
CO uptake, dispersion (D%) and average particle size (dCO) as measured by CO
chemisorption at room temperature.
b
Average particle size (dXRD) calculated by using the Scherrer equation by XRD.
Average particle size (dTEM) obtained from TEM.
c
calcination. Therefore, the H2-TPR results indicate that the strong
interactions between nickel and supports would happen when the
metal salt Ni(NO3)2·6H2O impregnated in supports and then cal-
cined at 400 ◦C in Ar.
TEM technique was conducted to study the distribution of nickel
particles on these four supports. Fig. 4 shows their TEM images
together with the histograms of particle size of the reduced Ni cat-
alysts. The results indicate that Ni particles are well-distributed all
over the AC, SBA-15, SiO2 and ␥-Al2O3. For AC catalyst, owing to
its largest surface area of 1293 cm3/g, the size is smaller than the
rest of the catalysts, with a relatively narrow size distribution. For
␥-Al2O3 catalyst, it is not easy to make a statistics of particle size
distribution, due to a less clear distinction between Ni and support.
However, some irregular Ni particles with mainly globular mor-
phology can be distinguished due to the darkest contrast. SBA-15
catalysts show hexagonal cylinder-like morphologies. A portion of
nickel species entered into its cylinder-like pores, exhibiting a strip-
type distribution. While a small amount of Ni particles inevitably
aggregated outside the SBA-15 when calcined in Ar at 400 ◦C. The
SiO2 catalyst exhibits the similar situation to the AC catalyst, while
its mean particle diameter is slightly larger than that of the former
catalyst, which is likely due to the smaller surface area of 549 cm3/g.
The average size of metal particles, as calculated by statistical anal-
ysis of about 200 particles, is also showed in Fig. 4. The metal
particle size follows the trend Ni/SBA-15 (8.0 3.2 nm) > Ni/SiO2
(4.8 0.9 nm) > Ni/AC (3.9 0.7 nm).
Fig. 5. NH3-TPD profiles of the reduced Ni/AC(a), Ni/␥-Al2O3(b), Ni/SBA-15(c) and
Ni/SiO2(d).
3.2. Catalytic activity
In the present study, we evaluated and compared the catalytic
performance of Ni/AC, Ni/␥-Al2O3, Ni/SBA-15, Ni/SiO2 in HDO of
methoxy-rich compound anisole. Fig. 6 shows the variation of the
conversion of anisole and the yield of products with time. All these
experiments were performed at least in duplicate and the results
were well reproducible. The main products detected by GC–MS
and GC are cyclohexyl methyl ether (CME), cyclohexanol (CHL),
cyclohexane (CHE), cyclohexyl cyclohexyl ether (CCE). In addition,
small amounts of the directly demethoxylated products benzene
(BEN) and methanol can also be detected. Overall, the main anisole
conversion pathways under the current conditions mainly consist
of three steps. The first step is the hydrogenation saturation of
aromatic ring of anisole, followed by the second step of demethy-
lation by the hydrogenolysis of the O CH3 bond of the methoxy
group, resulting in the formation of cyclohexanol. The last step is
the deoxygenation of cyclohexanol to produce cyclohexane by the
intramolecular dehydration reaction of hydroxyl group. Otherwise,
small amounts of CCE were also obtained by the intermolecular
dehydration reaction of two cyclohexanol molecules.
From the results shown in Fig. 6, we can see that the feedstock
anisole were almost completely converted within 90 min for all
four Ni catalysts. Except the Ni/SiO2, the main products obtained
from anisole are aromatic ring-hydrogenated CME, however sub-
sequently the degree of demethylation and deoxygenation with
time going by is quite different depending on the supports. These
results suggest that the fastest step in HDO of anisole is the nickel-
catalyzed aromatic ring-hydrogenation rather than the subsequent
demethylation and deoxygenation [1]. Ni/AC in the current con-
ditions presents a relatively low HDO yield, due to the moderate
capacity for the hydrogenolysis of the C O bonds. After 225 min, the
selectivity of O-free CHN is only 15%, while a selectivity of 71% CME
was accumulated, which shows that Ni/AC has a poor hydrodeoxy-
genation (HDO) efficiency even though its ability for aromatic-ring
hydrogenation is excellent. Although Ni/␥-Al2O3 owns more rela-
tively strong acid sites for the cleavage of the C O bonds, due to the
existence of the hardly-reduced Ni species, the subdued synergetic
catalytic effect between acid sites for hydrogenolysis and metal
sites for hydrogenation prevents the dehydration of CHL to CHN.
As a result, compared with other three catalysts, relatively more
demethylated product CHL with a selectivity of 14.1% was obtained.
Compared with Ni/AC and Ni/␥-Al2O3, Ni catalysts supported on
The amounts of metal active sites were estimated from irre-
versible CO chemisorption measurement and the results are listed
in Table 2, which also summarizes the Ni particle sizes calculated
from XRD, CO chemisorption and TEM. We can see that Ni/AC cat-
alyst has the best metal dispersion and smallest Ni particle size
owing to its large surface area of 1293 m2/g. In addition, the Ni par-
ticle size has a similar tendency (dSBA-15 > dSiO > dAC) calculated
2
from these three methods. However, for ␥-Al2O3, due to the inter-
ference of support peaks and the formation of difficult-reduced
nickel aluminate, it is not easy to calculate the Ni particle size
according to these three methods.
The acidity of all catalysts was determined by temperature-
programmed desorption of ammonia (NH3-TPD) and desorption
profiles are displayed in Fig. 5. For Ni/AC catalyst, it does not exhibit
weak NH3 desorption feature below 500 ◦C and only shows a sharp
peak at around 520 ◦C, which can be attributed to the TCD signals of
CO2 or CO formed by carbon oxidization with O-containing func-
tional groups located on the surface of AC supports, because this
peak also exists when the measure was performed in pure He. In
other cases, Ni/SBA-15 and Ni/SiO2 exhibit the similar TPD profiles
and present three main peaks which are centered at 200 ◦C, 400 ◦C
and 620 ◦C separately. Similarly, Ni/␥-Al2O3 presents two promi-
nent and broad peaks respectively centered at 260 ◦C and 570 ◦C,
which could be associated with weak acidic sites and stronger acidic
sites [30].
Please cite this article in press as: S. Jin, et al., Catalytic hydrodeoxygenation of anisole as lignin model compound over supported nickel
catalysts, Catal. Today (2014), http://dx.doi.org/10.1016/j.cattod.2014.02.014

G Model
CATTOD-8905; No. of Pages8
ARTICLE IN PRESS
6
S. Jin et al. / Catalysis Today xxx (2014) xxx–xxx
Fig. 6. Variation of the conversion of anisole and the yield of products with time for (a) Ni/AC, (b) Ni/␥-Al2O3, (c) Ni/SBA-15 and (d) Ni/SiO2.
Fig. 7. Reduction temperature dependence of conversion and selectivity to main products in HDO of anisole.
silica materials exhibited a better deoxygenation activity with a
HDO% of 70% for Ni/SBA-15 and 95% for Ni/SiO2, which appears to
be related to the favorable synergetic catalytic effect between acid
sites and metal sites [13,20]. On the other hand, the weak acidic
OH sites located on the surface of silica materials favor the adsorp-
tion with O atoms in hydroxyl or methoxy groups, which is also
responsible for excellent ability for the cleavage of C O [22].
To further study the reducibility of all the catalysts, reduction
temperature dependence of conversion of anisole was investigated,
as shown in Fig. 7. According to the activity results, the conversion
of anisole and HDO yields both clearly decrease for Ni/AC, SBA-
15 and SiO2, indicating that the increase of reduction temperature
weakens the hydrogenation capacity of these catalysts, which is
owing to the existence of larger Ni particles by the inevitable aggre-
gation at the high temperature. Nevertheless, Ni/␥-Al2O3 exhibits
even higher anisole conversion and yield of deoxygenated products
CHN when reduced at 500 ◦C. It is because the release of more Ni
particles obtained from hardly-reduced Ni species of this catalyst
results in stronger synergetic catalytic effect between acid sites and
metal sites.
Since it demonstrates the best deoxygenation activity in the
HDO process of anisole [31], Ni/SiO2 catalyst was used to inves-
tigate the effect of temperature on the product distribution and
effectiveness of the overall process. The corresponding results are
shown in Fig. 8. The feedstock anisole was almost completely
transformed at the temperature range of 180–220 ◦C under the
current conditions. As the aromatic ring hydrogenation process
is a fast step, the main products in HDO of anisole catalyzed by
Ni/SiO2 are non-aromatic hydrocarbon, i.e. oxygenous alkane and
oxygen-free alkane. From the results in Fig. 8, we can see that the
Please cite this article in press as: S. Jin, et al., Catalytic hydrodeoxygenation of anisole as lignin model compound over supported nickel
catalysts, Catal. Today (2014), http://dx.doi.org/10.1016/j.cattod.2014.02.014

G Model
CATTOD-8905; No. of Pages8
ARTICLE IN PRESS
S. Jin et al. / Catalysis Today xxx (2014) xxx–xxx
7
Scheme 2. Reaction network of anisole hydrodeoxygenation over supported Ni
catalysts.
Fig. 8. Effect of temperature on the ratio of oxygenous alkane to oxygen-free alkane
(WO-alkane/WO-free alkane) and the deoxygenation degree in HDO of anisole.
two cyclohexanol molecules can also happen over Ni/AC. At the
deficient H2 pressure, the removal of OCH3 from anisole to form
benzene by direct demethoxylation can be achieved. Furthermore,
the direct demethylation process of anisole to produce phenols
(PHE) [15] can also occur, while unfortunately this intermediate
product is not detected by GC–MS, which may be related to faster
phenol hydrogenation rate than its formation rate on the Ni cata-
lysts.
4. Conclusions
Supported nickel catalysts with a broad range of physicochem-
ical properties obtained by an incipient impregnation technique
have been evaluated in hydrodeoxygenation of methoxy-rich lignin
model compound anisole in a batch set-up. The catalysts exhibit
the high activity toward the hydrogenation saturation of aromatic
ring at the relatively low temperature (180–210 ◦C) and moderate
hydrogen pressure (0.5–3.0 MPa). Ni catalysts supported on silica
materials exhibited a better deoxygenation activity with a HDO%
of 70% for Ni/SBA-15 and 95% for Ni/SiO2. Combination of acidic
sites of the supports and good dispersion of the metal leads to a
very good HDO yield over Ni/SiO2, especially at high temperature
(>200 ◦C). At the deficient H2 pressure, even at the atmospheric
pressure, the removal of OCH3 of anisole to form benzene by direct
demethoxylation can also be achieved.
Fig. 9. Pressure dependence of conversion and selectivity toward main products in
HDO of anisole.
deoxygenation activity of Ni/SiO2 strongly depends on reaction
temperature. At the low temperature of 180 ◦C, the products mainly
consist of oxygenous alkane CME. With the increase of temperature,
the ratio of oxygen-free alkane to oxygenous alkane and HDO%
of anisole both increase. At the high temperature of 220 ◦C, the
oxygen-free alkane CHE is the main product in the target process
(HDO) and simultaneously the HDO% of anisole reaches up to 99%.
The pressure dependence of conversion and selectivity toward
main products in HDO of anisole over Ni/SiO2 is shown in Fig. 9.
We can easily get a conclusion that the pressure plays a significant
role on both anisole conversion and main products distribution.
With increasing hydrogen pressure from 0.5 to 3.0 MPa, the con-
version of anisole increases from 22% to 99%. At the high hydrogen
pressure, the hydrogenation saturation of aromatic ring and subse-
quent deoxygenation process can be easily achieved. Nevertheless,
it becomes difficult to carry out the aromatic ring-hydrogenation
process at the low hydrogen pressure. At present, the direct
demethoxylation pathway stands out and the selectivity of BEN
at 0.5 MPa reaches to 53%. According to the reported works [12,32],
we can get the same conclusion that at the low pressure, even at
the atmospheric pressure, it favors the production of oxygen-free
aromatics by the direct demethoxylation process.
Acknowledgments
We gratefully acknowledge the financial support provided by
the National Natural Science Foundation of China (21073023 and
21373038) and the Fundamental Research Funds for the Central
Universities (DUT12YQ03).
References
[1] C. Zhao, Y. Kou, A.A. Lemonidou, X. Li, J.A. Lercher, Angew. Chem. 121 (2009)
4047–4050.
[2] Z. Xiao, S. Jin, M. Pang, C. Liang, Green Chem. 15 (2013) 891–895.
[3] W. Xu, S.J. Miller, P.K. Agrawal, C.W. Jones, ChemSusChem 5 (2012) 667–675.
[4] F.G. Calvo-Flores, J.A. Dobado, ChemSusChem 3 (2010) 1227–1235.
[5] K.L. Deutsch, B.H. Shanks, Appl. Catal. A: Gen. 447/448 (2012) 144–150.
[6] J. Zakzeski, P.C.A. Bruijnincx, A.L. Jongerius, B.M. Weckhuysen, Chem. Rev. 110
(2010) 3552–3599.
[7] H. Li, X.S. Chai, M. Liu, Y.H. Deng, J. Agric. Food Chem. 60 (2012) 5307–5310.
[8] V.N. Bui, G. Toussaint, D. Laurenti, C. Mirodatos, C. Geantet, Catal. Today 143
(2009) 172–178.
[9] E. Laurent, B. Delmon, Appl. Catal. A: Gen. 109 (1994) 77–96.
[10] O.I. Senol, T.R. Viljava, A.O.I. Krause, Catal. Today 100 (2005) 331–335.
[11] H. Ohta, H. Kobayashi, K. Hara, A. Fukuoka, Chem. Commun. 47 (2011)
12209–12211.
[12] X.L. Zhu, L.L. Lobban, R.G. Mallinson, D.E. Resasco, J. Catal. 281 (2011) 21–29.
[13] C.R. Lee, J.S. Yoon, Y.W. Suh, J.W. Choi, J.M. Ha, D.J. Suh, Y.K. Park, Catal. Commun.
17 (2012) 54–58.
According to the previous results, the reaction network for
anisole HDO over Ni catalysts in the current conditions is presented
in Scheme 2. At the sufficient H2 pressure, the hydrogenation sat-
uration of aromatic ring of anisole is easy to conduct for all four
Ni catalysts. However, the supports exhibit significant influence
in subsequent demethylation and deoxygenation to produce O-
free CHN. Moreover, the intermolecular dehydration reaction of
Please cite this article in press as: S. Jin, et al., Catalytic hydrodeoxygenation of anisole as lignin model compound over supported nickel
catalysts, Catal. Today (2014), http://dx.doi.org/10.1016/j.cattod.2014.02.014

G Model
CATTOD-8905; No. of Pages8
ARTICLE IN PRESS
8
S. Jin et al. / Catalysis Today xxx (2014) xxx–xxx
[14] N. Yan, Y. Yuan, R. Dykeman, Y. Kou, P.J. Dyson, Angew. Chem. Int. Ed. 49 (2010)
5549–5553.
[24] O. González, H. Pérez, P. Navarro, L.C. Almeida, J.G. Pacheco, M. Montes, Catal.
Today 148 (2009) 140–147.
[15] K. Li, R. Wang, J. Chen, Energy & Fuels 25 (2011) 854–863.
[16] H.Y. Zhao, D. Li, P. Bui, S.T. Oyama, Appl. Catal. A: Gen. 391 (2011) 305–310.
[17] Y. Yang, J. Chen, H. Shi, Energy & Fuels 27 (2013) 3400–3409.
[18] Y. Yang, C. Ochoa-Hernández, V.A. de la Pen˜a O’Shea, J.M. Coronado, D.P. Ser-
rano, ACS Catal. 2 (2012) 592–598.
[19] X.Y. Wang, R. Rinaldi, Energy Environ. Sci. 5 (2012) 8244–8260.
[20] P.T.M. Do, A.J. Foster, J. Chen, R.F. Lobo, Green Chem. 14 (2012) 1388–1397.
[21] I.T. Ghampson, C. Sepúlveda, R. Garcia, J.L. García Fierro, N. Escalona, W.J.
Desisto, Appl. Catal. A: Gen. 435/436 (2012) 51–60.
[25] A.L. Alberton, M.M.V.M. Souza, M. Schmal, Catal. Today 123 (2007) 257–264.
[26] A.R. Naghash, T.H. Etsell, S. Xu, Chem. Mater. 18 (2006) 2480–2488.
[27] Y. Cesteros, P. Salagre, F. Medina, J.E. Sueiras, Chem. Mater. 12 (2000) 331–335.
[28] E. Heracleous, A.F. Lee, K. Wilson, A.A. Lemonidou, J. Catal. 231 (2005) 159–171.
[29] Y. Qian, S. Liang, T. Wang, Z. Wang, W. Xie, X. Xu, Catal. Commun. 12 (2011)
851–853.
[30] Y.X. Yang, C. Ochoa-Hernández, V.A. de la Pen˜a O’Shea, P. Pizarro, J.M. Coronado,
D.P. Serrano, Appl. Catal. B: Environ. 145 (2013) 91–100.
[31] M.V. Bykova, D.Yu. Ermakov, V.V. Kaichev, O.A. Bulavchenko, A.A. Saraev, M.Yu.
Lebedev, V.A. Yakovlev, Appl. Catal. B: Environ. 113/114 (2012) 296–307.
[32] C.V. Loricera, B. Pawelec, A. Infantes-Molina, M.C. Álvarez-Galván, R. Huirache-
Acun˜a, R. Nava, J.L.G. Fierro, Catal. Today 172 (2011) 103–110.
[22] R.N. Olcese, M. Bettahar, D. Petitjean, B. Malaman, F. Giovanella, A. Dufour, Appl.
Catal. B: Environ. 115/116 (2012) 63–73.
[23] P. Wu, T. Tatsumi, Chem. Mater. 14 (2002) 1657–1664.
Please cite this article in press as: S. Jin, et al., Catalytic hydrodeoxygenation of anisole as lignin model compound over supported nickel
catalysts, Catal. Today (2014), http://dx.doi.org/10.1016/j.cattod.2014.02.014