Effective conversion of heteroatomic model compounds in microalgae-based bio-oils to hydrocarbons over β-Mo2C/CNTs catalyst

Article abstract of DOI:10.1016/j.molcata.2015.10.018

Hydrotreatment of heteroatomic model compounds in microalgae-based bio-oils into diesel-like hydrocarbons was carried out over carbon nanotubes (CNTs)-supported β-Mo₂C catalyst with superior activity and selectivity under a mild condition (≤200 °C). The results show that stearic acid and hexadecanamide can be completely converted into n-C15-C18 alkanes over β-Mo2C/CNTs catalyst. The β-Mo2CC/CNTs favors the pathway of hydrogenation-dehydration-hydrogenation to produce n-octadecane with an optimal yield of 91.24% at a lower temperature of 180 °C during hydrotreating process of stearic acid. The recycle tests demonstrate that the β-Mo2CC/CNTs exhibits excellent stability, and can be reused for seven times consecutively without reduction of catalytic stability. Based on the determined products, a brief reaction pathway is proposed. Therefore, a novel approach to produce diesel-like hydrocarbons via catalytic hydrotreatment of microalgae-based bio-oils over β-Mo₂C/CNTs is introduced, which provides a basic research as well as technical parameters for its further industrialization.

Full text of DOI:10.1016/j.molcata.2015.10.018

Journal of Molecular Catalysis A: Chemical 411 (2016) 95–102
Contents lists available at ScienceDirect
Journal of Molecular Catalysis A: Chemical
journal homepage: www.elsevier.com/locate/molcata
Effective conversion of heteroatomic model compounds in
microalgae-based bio-oils to hydrocarbons over ␤-Mo2C/CNTs
catalyst
Junmei Lianga,b, Ranran Dingb, Yulong Wub,∗, Yu Chenb, Kejing Wub, Yongqiang Menga,
Mingde Yangb, Yaowu Wangb
a College of Materials Science and Engineering, Hebei University of Science and Technology, Shijiazhuang 050018, PR China
b Institute of Nuclear and New Energy Technology, Tsinghua University, Beijing 100084, PR China
a r t i c l e i n f o
a b s t r a c t
Article history:
Received 14 August 2015
Received in revised form 13 October 2015
Accepted 13 October 2015
Available online 17 October 2015
Hydrotreatment of heteroatomic model compounds in microalgae-based bio-oils into diesel-like hydro-
carbons was carried out over carbon nanotubes (CNTs)-supported ␤-Mo2C catalyst with superior activity
and selectivity under a mild condition (≤200 ◦C). The results show that stearic acid and hexadecanamide
can be completely converted into n-C15–C18 alkanes over ␤-Mo2C/CNTs catalyst. The ␤-Mo2C/CNTs favors
the pathway of hydrogenation–dehydration–hydrogenation to produce n-octadecane with an optimal
yield of 91.24% at a lower temperature of 180 ◦C during hydrotreating process of stearic acid. The recycle
tests demonstrate that the ␤-Mo2C/CNTs exhibits excellent stability, and can be reused for seven times
consecutively without reduction of catalytic stability. Based on the determined products, a brief reac-
tion pathway is proposed. Therefore, a novel approach to produce diesel-like hydrocarbons via catalytic
hydrotreatment of microalgae-based bio-oils over ␤-Mo2C/CNTs is introduced, which provides a basic
research as well as technical parameters for its further industrialization.
Keywords:
␤-Mo2C/CNTs
Diesel-like hydrocarbons
Hydrotreatment
Stearic acid
Hexadecanamide
© 2015 Elsevier B.V. All rights reserved.
1. Introduction
the heteroatoms in the bio-oils to achieve high quality transport
fuel.
The increasing concerns over the depletion of non-renewable
fossil resources and global climate change have attracted significant
attention to explore renewable and environmental friendly sources
of energy [1–5]. Of all the alternatives, biomass is regarded as the
only material for the production of liquid fuel, which is considered
to be one of the most sustainable alternatives to petroleum because
of viable means for environmental and economic sustainability [6].
Among the biomass, microalgae is considered as one of the most
promising energy sources for fuel production due to superior pho-
tosynthetic efficiency, higher growth rate, area-specific yield, and
higher carbon dioxide utilization capabilities compared to terres-
trial plants [7–10]. However, microalgae-based bio-oils obtained
from thermochemical conversion contain appreciable quantities
of oxygen and nitrogen heteroatoms, which result in undesirable
property including corrosiveness, thermal instability, high viscosity
and low heating value [11–14]. Therefore, it is desirable to remove
As reported in previous works, microalgae-based bio-oils
contain a large amount of oxygen-containing compounds and
nitrogen-containing compounds, especially C16–C18 long-chain
fatty acids and amide compounds [13,15]. To this end, the neces-
sary upgrading processes must be carried out via deoxygenation
and denitrification prior to actual application. Among the upgrad-
ing technologies, catalytic hydrotreatment used for processing
heavy crude petroleum and shale oils, is one of the most efficient
approaches for removing heteroatoms in the microalgae-based bio-
oils [16,17]. However, the conventional hydrotreating catalysts,
such as sulfided NiMo and CoMo [18,19], may easily lead to the
generation of sulfur compounds and sulfur leaching, which result
in the deactivation of catalysts. As hydrotreating catalysts, sup-
ported noble metals (Pt, Pd, Rh, etc.,) also show superior activity
and selectivity [20–25]. However, the high price greatly limits
their large-scale practical applications [26]. Therefore, it is urgent
to develop more sulfur-free and cost-effective catalyst with high
activity and stability under mild conditions (≤200 ◦C) for the
upgrading of microalgae-based bio-oils.
Transition-metal carbides, especially nanosized molybdenum
carbides (Mo2C), have attracted considerable attention as promis-
ing alternative catalytic materials in hydrogen-involved reactions
∗
Corresponding author at: Tsinghua University, Insitute of Nuclear and New
Energy Technology, 205, PR China. Fax: +86 10 69771464.
E-mail address: wylong@tsinghua.edu.cn (Y. Wu).
http://dx.doi.org/10.1016/j.molcata.2015.10.018
1381-1169/© 2015 Elsevier B.V. All rights reserved.

96
J. Liang et al. / Journal of Molecular Catalysis A: Chemical 411 (2016) 95–102
Scheme 1. Proposed reaction pathways for stearic acid conversion over ␤-Mo2C/CNTs.
since the discovery of their “platinum-like” behavior, preferable
resistance to poisoning of carbon monoxide and better anti-
sintering characteristics [27–29]. Ren et al. [30] demonstrated that
the Mo2C exhibited high hydrodeoxygenation (HDO) activity with
propanal as a model compound. Han et al. [31] reported that car-
bon nanotubes (CNTs)-supported Mo2C could efficiently transform
esters into diesel-like hydrocarbons in the presence of hydrogen.
These papers clearly indicate the potential application of Mo2C as
a selective catalyst for hydrotreating of oxygen-containing com-
pounds. As mentioned above, long-chain fatty acids and amide
compounds are main components of the microalgae-based bio-oil
[13], and it is necessary to convert these heteroatomic compounds
into chain hydrocarbons. However, few papers have reported the
conversion of long-chain fatty acids and amide compounds into
hydrocarbons over Mo2C catalyst. Although Stellwagen and Bitter
[32] has researched the catalytic conversion of stearic acid over
pre-treated carbon nanofibers (CNFs)-supported Mo2C, the yield
of alkanes is relatively low of 80% at 300 ◦C. In addition, the cat-
alytic conversion of a mixture of long-chain fatty acid and amide
compound in microalgae-based bio-oils over Mo2C catalyst has not
been reported. Herein we describe for the first time that Mo2C/CNTs
can efficiently catalyze the conversion of stearic acid and a mixture
of stearic acid and nitrogen-containing compound to diesel-like
hydrocarbons at a lower temperature (≤ 200 ◦C).
will develop a foundation on industrial production of liquid hydro-
carbons via hydrotreating process over inexpensive transition
metal catalysts with the economic value and strategic significance.
2. Experimental
2.1. Materials
Stearic acid (98%), octadecane (≥99.5%), heptadecane (≥99.5%),
hexadecane (≥99%), pentadecane (≥99%) and octadecanol (≥99.5%)
were purchased from Aladdin Chemistry Co., Ltd. (Shanghai, China).
Hexadecanamide was provided from Tokyo Chemical Industry Co.,
Ltd., Decane (>99%) and ammonium molybdate (99.0%) were pur-
chased by Guangfu Technology Development Co., Ltd. (Tianjin,
China). Multi-walled CNTs (>95%) were generously provided from
XFNANO Materials Tech., Co., Ltd. (Nanjing, China). All reagents
were used as-received without further purification.
2.2. Catalyst preparation
The ␤-Mo2C supported CNTs (BET specific surface area:
178 m2 g−1) catalyst was synthesized by the CHR method described
as follows: in a typical run, ␤-Mo2C/CNTs catalyst was prepared by
first dissolving ammonium molybdate (1.36 g) in deionized water
(15 mL) and then slowly dropping this solution onto CNTs (2 g) with
continuously stirring in a 100 mL beaker. After allowing the metal
to permeate into the support for 12 h at ambient temperature, the
catalyst was first dried at 105 ◦C for 3 h. The dried sample was then
calcined in air at 500 ◦C with a heating rate of 10 ◦C min−1, and
maintained at that temperature for 3 h. Thereafter, it was disposed
by reduction in a H2 flow of 80 mL min−1 from room temperature to
700 ◦C with 10 ◦C min−1 and reduced at 700 ◦C for 4 h. The theoret-
ical ␤-Mo2C loading was approximately 5% based on CNTs support.
In this work, catalytic hydrotreatment of stearic acid and hex-
adecanamide as the oxygen-containing, nitrogen-containing model
compound in microalgae-based bio-oils, respectively, was carried
out over the ␤-Mo2C/CNTs catalyst, which was synthesized by the
carbothermal hydrogen reduction (CHR) method. The aim of the
present work is to evaluate catalytic performance of ␤-Mo2C/CNTs
catalyst in multi-aspect including HDO and hydrodenitrogener-
ation (HDN) process under mild conditions (≤200 ◦C). Reaction
parameters optimization and a brief pathway based on the experi-
mental results had been investigated in the HDO process. This work
Fig. 1. XRD patterns of the catalyst: (a) ␤-Mo2C/CNTs; (b) CNTs.

J. Liang et al. / Journal of Molecular Catalysis A: Chemical 411 (2016) 95–102
97
2.3. Catalyst characterization
X-ray powder diffraction (XRD) patterns were recorded on
a D8 ADVANCE diffractometer using copper radiation (Cu K␣
ꢀ = 0.1541 nm) at 36 kV and 2 mA, and the samples were ana-
lyzed by scanning at an angle ranging from 0◦ to 80◦ with a
scan rate of 2◦ min−1. Nitrogen adsorption–desorption isotherms
and pore size distribution plot were measured in a Micromerit-
ics Tristar analyzer at −196 ◦C. Pretreatment of samples was done
by flowing N2 for 3 h at 300 ◦C. The surface area was calculated
by the Brunauer–Emmett–Teller (BET) method, and the pore size
distribution was calculated by the Barrett–Joyner–Halenda (BJH)
method. X-ray photoelectron spectroscopy (XPS) measurements
were performed on a PHI5300 X-ray photoelectron spectrometer
with a monochromatic X-ray source (Al K␣). Transmission elec-
tron microscope (TEM) images were obtained by a FEI Tecnai F20
Field Emission Electron Microscope, and operated at an accelerat-
ing voltage of 200 kV. At least 100 particles were counted to obtain
the average diameter of ␤-Mo2C particles.
Fig. 2. N2 absorption–desorption isotherms (a) and pore size distribution (b) of
␤-Mo2C/CNTs.
2.4. Catalytic activity measurements
The catalytic hydrogenation experiments of stearic acid and a
mixture of stearic acid and hexadecanamide (5:1) were carried
out in a batch autoclave. The autoclave was heated with an exter-
nal electrical furnace, and the temperature was measured with
a thermocouple and controlled at 1 ◦C. In a typical experiment,
reactant (0.5 g), decane (50 mL), and catalyst (0.25 g) were injected
into the reactor, and then the autoclave was purged for three times
with 0.2 MPa H2 to remove air and pressurized with additional H2
to 4 MPa at ambient temperature. The reaction temperature was
increased to 180 ◦C followed by a soak at 180 ◦C for 4 h at a stir-
ring speed of 300 rpm. The liquid samples were manually collected
for analysis during the run and end of reaction. The catalyst was
deposited at the bottom of the autoclave when the reaction was
complete, and recovered for catalyst recycling by filtration and
drying without further activation treatment.
The diffraction peaks at 26.0◦ and 44.0◦ are contributed to CNTs
with hexagonal structure. No other competing phases are detected,
which indicates that the successful synthesis of ␤-Mo2C/CNTs
catalyst with well-crystallized structure proved by narrow and
intensive diffraction peaks.
3.1.2. N2 adsorption–desorption analysis
The typical N2 adsorption–desorption isotherms for the ␤-
Mo2C/CNTs catalyst and corresponding BJH pore size distribution
plot are shown in Fig. 2. The shape of the sample isotherm exhibites
representative type IV curves according to IUPAC recommenda-
tions [35], indicating that the ␤-Mo2C/CNTs catalyst is a typical
mesoporous material with the presence of an H1-type hystere-
sis loop at relatively high pressures of 0.80–0.99 [Fig. 2(a)], which
is usually associated with the open-ended capillary condensation
that occurs in the mesopores [36]. As shown in Fig. 2(b), the sam-
ple possesses a large number of pores in the mesopore range, and
larger pore size distribution centers at 25 nm. The pore volume of
␤-Mo2C-supported catalysts is 0.5 cm3 g−1, which may be related
to the surface reconstruction of catalyst during preparation. More-
2.5. Product analysis
The final products of model compounds reaction were ana-
lyzed using a gas chromatograph (Agilent GC6820) equipped with
flame ionization detector (FID) and a GsBP-Inowax capillary col-
umn (30 m × 0.32 mm × 0.25 ␮m). The GC oven temperature was
programmed as follows: 2 min hold at 120 ◦C, and then ramped to
260 ◦C soaking for 10 min at a rate of 10 ◦C min−1. 2 ␮L samples were
injected into the column with a split ratio of 10:1, and the flow rate
of carrier gas (helium) was 10 mL min−1. The injector and detector
temperatures were maintained at 320 ◦C and 280 ◦C, respectively.
Conversion was defined as mole% and calculated as moles of con-
verted reactants (such as stearic acid, hexadecanamide) per moles
of starting reactants times 100%. Yields were defined as mole% and
calculated as moles of each product per mole of total product times
100%. Yield was calculated as conversion multiplied by the selec-
tivity.
over, the typical BET surface area of ␤-Mo2C/CNTs is 116 m2 g−1
.
3.1.3. XPS analysis
The XPS measurement was employed to identify the surface
electronic state and chemical composition of the ␤-Mo2C/CNTs cat-
alyst. Fig. 3(a) shows a typical survey XPS spectrum of the catalyst.
The peaks of Mo (Mo 3d, 3p3/2, 3p1/2) and C (C 1s) in the survey
XPS spectrum can be assigned to the characteristic of molybdenum
carbides and CNTs. Meanwhile, the peak of O1s is attributed to the
molybdenum oxides. Fig. 3(b) shows the high-resolution XPS spec-
trum for Mo 3d. The Mo 3d3/2 peak is centered at 232.1 eV, whereas
the Mo 3d5/2 peak is found at 228.9 eV, with a spin energy separa-
tion of 3.2 eV. This characteristic doublet of core-level Mo 3d5/2
and Mo 3d3/2 indicates the Mo (IV) oxidation state of MoO2. The
characteristic peaks at 231.9 and 235.1 eV can be attributed to Mo
(VI) 3d5/2, 3d3/2, probably due to the slight oxidation of metastable
MoO2 in air [37–39]. In addition, the peaks at 227.9 eV and 231.3 eV
could be ascribed to Mo (II) 3d5/2 and 3d3/2 in ␤-Mo2C, respectively
[32,40–42]. The content ratio of the Mo2+/(Mo4+ + Mo6+) on the sur-
face of catalyst is 1.58, indicating the ␤-Mo2C is the main species on
the surface of the catalyst, which is consistent with the XRD result.
The high-resolution C 1s spectrum is deconvolved into four
peaks corresponding to carbon atoms in different oxygen-
3. Results and discussion
3.1. Catalysts characterization
3.1.1. XRD analysis
The XRD pattern (shown in Fig. 1) of the prepared catalyst
shows that the diffraction peaks appear at 2ꢁ of 34.3◦, 38.0◦, 39.4◦,
52.1◦, 61.5◦, 69.6◦,74.6◦ and 75.5◦, which are assigned to the planes
of (1 0 0), (0 0 2), (1 0 1), (1 0 2), (1 1 0), (1 0 3), (1 1 2) and (2 0 1)
of the hexagonal close-packed ␤-Mo2C crystalline phase [33,34].

98
J. Liang et al. / Journal of Molecular Catalysis A: Chemical 411 (2016) 95–102
Fig. 3. XPS spectra of the ␤-Mo2C/CNTs: (a) survey XPS spectrum; high-resolution XPS spectra of (b) Mo 3d, (c) C 1s and (d) O 1s.
containing functional groups. As can been seen from Fig. 3(c). The
peak at 283.9 eV is eligible for ␤-Mo2C, 284.8 eV for C C, 285.8 eV
for C H, and 288.9 eV for O
C O, respectively [43–49]. Among
them, the strongest peak of 283.9 eV is assigned to ␤-Mo2C, and the
others are relatively weak, suggesting the main phase of ␤-Mo2C
on the surface of catalyst. The O 1s XPS signals of ␤-Mo2C/CNTs
catalyst can be fitted by three kinds of oxygen atoms locating at
530.3 eV, 531.6 eV and 532.7 eV [Fig. 3(d)]. The peak at 530.3 eV
can be indexed as oxygen-containing group in the molybdenum
oxides. The peaks at 531.6 eV and 532.7 are assigned to OH and
C
O bonds, respectively [43,50].
3.2. Deoxygenation of stearic acid over ˇ-Mo2C/CNTs
3.2.1. The effect of reaction temperatures
To study the influence of the reaction temperatures on the cat-
alytic activity of the used catalyst and product distribution, the
catalytic hydrogenation of stearic acid over ␤-Mo2C/CNTs at dif-
ferent temperatures was carried out. The results show that the
conversion of stearic acid is only 68.39% at 160 ◦C, but dramatically
increases to 100% at 180 ◦C. The product distribution results reveal
that the highest yield for n-octodecane is up to 91.24% at the tem-
perature of 180 ◦C, as shown in Fig. 4. There is no obvious difference
on the yield of n-octadecane at the reaction temperature of 180 ◦C
and 190 ◦C. When the temperature increases to 200 ◦C, the yield of
Fig. 4. The product distribution of catalytic hydrogenation of stearic acid under
various reaction temperatures over ␤-Mo2C/CNTs. Experimental conditions: stearic
acid (0.5 g), decane (50 mL), catalyst (0.25 g), 4 h, 4 MPa H2.

J. Liang et al. / Journal of Molecular Catalysis A: Chemical 411 (2016) 95–102
99
Fig. 6. The conversion of octadecanol and the yield of products as a function of reac-
tion times over ␤-Mo2C/CNTs. Experimental conditions: octadecanol (0.5 g), decane
(50 mL), catalyst (0.25 g), 4 MPa H2, 180 ◦C.
Fig. 5. The conversion of stearic acid and the yield of products as a function of reac-
tion times over ␤-Mo2C/CNTs. Experimental conditions: stearic acid (0.5 g), decane
(50 mL), catalyst (0.25 g), 4 MPa H2, 180 ◦C.
n-octadecane decreases to 86.28% because of the enhancement of
heptadecane yield via decarbonylation.
stearic acid, because of the fast hydrogenation of the double bond
to n-octadecane over ␤-Mo2C/CNTs catalyst. No other products
are found in the deoxygenation process, such as esters or ketones,
which indicates the good hydrodeoxygenation performance of ␤-
Mo2C/CNTs. Gas products obtained over the ␤-Mo2C/CNTs were
also analyzed, and the main oxygen-containing product is water.
Trace amounts of CO2 and CH4 were also detected in the reaction
system.
Large amounts of octadecanol can be achieved at a lower tem-
perature (<180 ◦C) during the process of catalytic hydrogenation,
which indicates that the dehydration reaction of octadecanol to n-
octodecane requires a higher temperature in the overall reaction
when ␤-Mo2C/CNTs used as the catalyst. With the increase of reac-
tion temperatures, the yield of n-heptadecane gradually increases,
and the highest n-heptadecane yield of 13.72% is achieved at 200 ◦C.
This result suggests that higher temperatures favor the decar-
bonylation reaction of octadecanal to produce n-hepadecane over
␤-Mo2C/CNTs. By tracing the conversion course of stearic acid,
intermediate octadecanal was confirmed. Because only traces of
octadecanal were detected in the overall reaction, the octadecanal
reduction rate to the octadecanol was proved to be faster than its
production rate over ␤-Mo2C/CNTs. In addition, a blank test with-
out catalyst was carried out to evaluate the contributions of the
used catalyst at a reaction temperature of 200 ◦C, and a very lower
stearic acid conversion can be obtained. That is to say the presence
of the ␤-Mo2C/CNTs catalyst plays a crucial role in the hydrogena-
tion reaction of stearic acid. On the basic of the results mentioned
above, 180 ◦C is selected as the favorable reaction temperature in
this paper.
To understand the deoxygenation step of stearic acid to alka-
nes better, the representative intermediate, octadecanol, was
employed in
a catalytic hydrogenation experiment over ␤-
Mo2C/CNTs with the same reaction conditions mentioned in
conversion of stearic acid. As shown in Fig. 6, the yield of n-
octadecane and the conversion of octadecanol increase obviously
with reaction times, and the highest n-octadecane yield of 88.26%
is obtained with the octadecanol conversion of 100% at 2.5 h, which
also supports the view that hydrogenation–dehydration of octade-
canol is the major reaction path of deoxygenation of stearic acid. In
addition, the equilibrium between the octadecanol and octadecanal
exists in the fact that yields of individually produced n-heptadecane
and n-octadecene increase linearly with reaction times.
3.2.3. The effect of hydrogen pressures
3.2.2. The effect of reaction times
The catalytic hydrogenation of stearic acid under various reac-
tion pressures were investigated over ␤-Mo2C/CNTs catalyst. The
results indicate that initial hydrogen pressures have a significant
effect on the conversion of stearic acid, and the conversion of
stearic acid is proportional to the initial hydrogen pressure in
the range of 0–1 MPa. As shown in Fig. 7, a complete conver-
sion of stearic acid is obtained when the initial hydrogen pressure
increases to 1 MPa. With further increase of the hydrogen pres-
sure to 4 MPa, the yield of n-octadecane enhances obviously to
91.24%, whereas the n-heptadecane yield reduces from 36.77 to
8.76%. This is related to the fact that the higher hydrogen pres-
sure shifts the equilibrium from octadecanal to octadecanol, and
benefits the hydrogenation–dehydration route of octadecanol to
n-octadecane, and thus, the yield of n-octadecane without car-
bon atom loss is enhanced, whereas the yield of n-heptadecane
obtained from decarbonylation of octadecanal is suppressed [14].
This suggested that higher hydrogen pressures heavily favor the
hydrogenation–dehydration pathway over ␤-Mo2C/CNTs catalyst.
In order to investigate the influence of reaction times on deoxy-
genation of stearic acid, the experiments of catalytic hydrogenation
as the increase of reaction times were carried out, and the results
are shown in Fig. 5. The conversion of stearic acid and the yields
of liquid hydrocarbons almost linearly increase with the reaction
times. The major liquid product is n-octadecane with the high-
est yield of 91.24% and the stearic acid conversion of 100% after
3 h, whereas the yield of n-heptadecane enhances slightly from
5.12 to 8.76%, which indicates that the hydrogenation–dehydration
of octadecanol is the major reaction route during the deoxygena-
tion process of stearic acid. The content of octadecanol is relatively
stable in the overall reaction, and the yield declines sharply with
almost complete conversion of stearic acid after 2.5 h, which can
been explained that the initial deoxygenation reaction proceeds
firstly by a fast hydrogenation of the carboxylic group to octade-
canol, and then the obtained octadecanol undergoes dehydration
to produce the major product of n-octadecane over ␤-Mo2C/CNTs.
Only traces of n-octadecene are detected during transformation of

100
J. Liang et al. / Journal of Molecular Catalysis A: Chemical 411 (2016) 95–102
Fig. 7. The conversion of stearic acid and the yield of products as a function of hydro-
gen pressures over ␤-Mo2C/CNTs. Experimental conditions: stearic acid (0.5 g),
decane (50 mL), catalyst (0.25 g), 4 h, 180 ◦C.
Fig. 8. Catalytic recycling tests for stearic acid conversion over ␤-Mo2C/CNTs. Exper-
imental conditions: stearic acid (0.5 g), decane (50 mL), catalyst (0.25 g), 4 h, 4 MPa
H2, 180 ◦C.
3.2.4. Deoxygenation pathway of stearic acid
Based on the results above, the reaction pathway of stearic acid
over ␤-Mo2C/CNTs is proposed (see Scheme 1). The reaction path-
way proceeds through an initial hydrogenation of the carboxylic
group on ␤-Mo2C surface to form octadecanal as an intermediate,
and then the octadecanal either decomposes by decarbonylation
to produce hydrocarbon with one carbon atom loss (minor route)
or converts into octadecanol by hydrogenation, subsequently, the
produced octadecanol undergoes sequential hydrogenation and
dehydration to obtain the final product of n-octadecane with-
out carbon atom loss (major route). In addition, the equilibrium
between the aldehyde and alcohol exists in the overall reaction,
which can be deeply influenced by the hydrogen pressure, and fur-
ther affects the yields of n-heptadecane and n-octadecane. Further
investigation of the mechanism and conditions of ␤-Mo2C/CNTs for
deoxygenation of fatty acid is ongoing in our group.
from Fig. 8. The catalytic stability was gradually reduced with the
increase of corresponding octadecanol after the seventh runs of cat-
alyst, and no significant variation in the conversion of stearic acid
was observed. It is related to the fact that the dehydration reaction
of aliphatic alcohol requires a higher energy in the overall reac-
tion [51]. On the whole, the ␤-Mo2C/CNTs catalyst is quite stable
and recyclable in the HDO of stearic acid under ambient reaction
conditions.
In order to investigate the reasons of ␤-Mo2C/CNTs catalyst
for gradually reduced stability, TEM analysis was carried out to
investigate the surface morphology and particle size of fresh ␤-
Mo2C/CNTs and that after the twelfth cycle (Fig. 9a and b). As shown
in Fig. 9a and b, agglomerations of a few of ␤-Mo2C particles are
observed after twelfth recycling. This phenomenon reveals that
long-term exposure of catalyst in reaction substrate of stearic acid
may cause mutual adhesion of CNTs, which leads to change of the
interaction between CNTs and ␤-Mo2C, and further influences the
distribution and agglomeration of ␤-Mo2C particles. The average
diameter of ␤-Mo2C nanoparticles approximately increases from
∼10.99 (fresh) to ∼16.17 nm (after the twelfth cycle) according to
statistical analysis. To further elaborate reduction of catalytic sta-
3.2.5. Stability tests of ˇ-Mo2C/CNTs catalyst
The stability of the ␤-Mo2C/CNTs catalyst was evaluated in this
work, and the results are shown in Fig. 8. It is interesting to note
that the catalyst can be recyclable up to seven runs without con-
siderable variation in the catalytic performance. As can be seen
Fig. 9. TEM images of (a) freshly prepared and (b) used after the twelfth circulation of ␤-Mo2C/CNTs.

J. Liang et al. / Journal of Molecular Catalysis A: Chemical 411 (2016) 95–102
101
bility, the catalyst after the twelfth cycle was characterized by XRD.
As shown in Fig. 10, the XRD patterns of ␤-Mo2C/CNTs show that
the position and intensity of ␤-Mo2C peaks are nearly the identical
to the fresh Mo2C/CNTs, but the diffraction peak of CNTs at 2ꢁ of 26◦
widens after the catalytic recycling tests, these results suggest that
the structural destruction of CNTs is the main reason for gradually
reduced stability of ␤-Mo2C/CNTs catalyst.
3.3. Hydrotreatment of stearic acid and hexadecanamide over
ˇ-Mo2C/CNTs catalyst
As mentioned above, the ␤-Mo2C/CNTs show the superior cat-
alytic activity and stability in converting of stearic acid to produce
liquid hydrocarbons. Whereas hexadecanamide is another impor-
tant compound existed in microalgae-based bio-oils [13]. In order
to investigate the catalytic hydrotreating of microalgae-based bio-
oils over ␤-Mo2C/CNTs, the hexadecanamide was selected as the
nitrogen-containing model compound in this study, mixed with
stearic acid for catalytic hydrotreating at the same reaction con-
ditions used in catalytic hydrogenation of pure stearic acid. The
difference of carbon chain length between stearic acid and hexade-
canamide is beneficial to analysis and calculation of products yields
based on specific product ownership of two different models. The
contents ratio of stearic acid and hexadecanamide are selected at
5:1 according to chemical compositions of real microalgae-based
bio-oils prepared by our group, of which fatty acid has highest con-
tent of 57.72%, followed by hexadecanamide of 7.08% [13], and the
results of catalytic hydrogenation of a mixture of stearic acid and
hexadecanamide over ␤-Mo2C/CNTs are listed in Fig. 11. The prod-
uct compositions reveal that the yield of alkanes are very low at
temperature of 160 ◦C, but dramatically increase with temperature
to 53.47% at 180 ◦C. Further increasing the reaction temperature to
200 ◦C, the highest yield for alkanes are almost up to 100%. This
results show that stearic acid and hexadecanamide can be con-
verted completely into n-C15–C18 alkanes over ␤-Mo2C/CNTs.
Large amounts of alcohol can be observed in the liquid-phase
products when performing the reaction at a lower temperature
(≤180 ◦C). This result indicates that dehydration of the alcohol
requires a higher temperature in the overall reaction when ␤-
Mo2C/CNTs used as the catalyst, which is consistent with the results
of deoxygenation of stearic acid. Compared to the complete con-
version of pure stearic acid to alkanes at 180 ◦C, the n-octadecane
yield decreases from 91.24% to 56.73% with n-heptadecane yield
increasing from 8.76% to 19.85% in the co-hydroprocessing of hex-
Fig. 11. The composition of products of stearic acid (
octadecanol ( ), hexadecanol ( ), octadecane (
cane ( ), pentadecane ( ) and others ( ) over ␤-Mo2C/CNTs. Experimental
conditions: stearic acid (0.5 g), hexadecanamide (0.1 g), decane (50 mL), catalyst
(0.25 g), 4 h, 4 MPa H2.
), hexadecanamide (
), heptadecane ( ), hexade-
),
adecanamide and stearic acid at 200 ◦C, the remarkable variation
on yields of n-octadecane and n-heptadecane may be related to
the enhanced decarbonylation of octadecanal intermediate in pres-
ence of hexadecanamide over ␤-Mo2C/CNTs. In addition, the sum
mass fraction of n-octadecane and n-heptadecane is five times less
than the sum mass fraction of n-hexadecane and n-pentadecane
corresponding to the contents of 5:1 for two models by quan-
titative transformation, which may be related to the cracking of
stearic acid. However, no cracking existed in hydrodeoxygena-
tion of pure stearic acid, which confirms that hexadecanamide can
promote cracking of partial stearic acid. Finally, the fact that no
any nitrogen-containing compounds in liquid-phase products is
detected suggests the superior high catalytic denitrification perfor-
mance of ␤-Mo2C/CNTs, and the nitrogen in bio-oil was removed
as the form of ammonia in gas phase.
To evaluate the contributions of ␤-Mo2C/CNTs on hydrotreat-
ment of stearic acid and hexadecanamide, a blank experiment
without catalyst was conducted at the temperature of 200 ◦C. It
is found that large amounts of stearic acid and hexadecanamide
remain unreacted with no alkanes production after 4 h. This result
suggests that the presence of the ␤-Mo2C/CNTs plays a crucial role
in the hydrogenation reaction of stearic acid and hexadecanamide.
The yield ratio of hexadecanol and octadecanol intermediate
increases obviously in blank experiment at 200 ◦C compared with
that at lower temperature (≤180 ◦C), which indicates that hex-
adecanamide is easier to be converted than stearic acid at 200 ◦C
without catalyst. Further investigation of the catalytic performance
of ␤-Mo2C/CNTs for hydrotreatment of microalgae-based bio-oils
is on going in our group.
4. Conclusions
In summary, a highly selective HDO and HDN approach of
heteroatomic compounds in microalgae-based bio-oils to pro-
duce diesel-like hydrocarbons is provided over ␤-Mo2C/CNTs in
the work. The catalyst was synthesized by CHR method, and
tested in catalytic hydrogenation reaction of stearic acid and
hexadecanamide. It is found that ␤-Mo2C/CNTs is a potential alter-
native catalyst to convert heteroatomic model compounds into
n-C15–C18 alkanes under a mild condition (≤200 ◦C) effectively.
In the hydrotreating process of stearic acid, the ␤-Mo2C/CNTs
Fig. 10. XRD patterns of (1) freshly prepared and (2) used ␤-Mo2C/CNTs

102
J. Liang et al. / Journal of Molecular Catalysis A: Chemical 411 (2016) 95–102
favors the pathway of hydrogenation–dehydration–hydrogenation
to produce n-octadecane with an optimal yield of 91.24%. The recy-
cle tests demonstrate that the ␤-Mo2C/CNTs exhibited excellent
stability, and can be reused for seven times consecutively with-
out reduction of catalytic stability. When the cycle increases to
the twelfth, the catalytic stability of ␤-Mo2C/CNTs is obviously
decreased. The agglomerations of active component ␤-Mo2C due
to structural destruction of CNTs under acidic condition, is the rea-
son for reduction of catalytic stability. Based on the determined
products, a brief reaction pathway is proposed. The obtained results
above open a new possibility to upgrade microalgae-based bio-oils
into high-grade transportation bio-fuels.
[17] X.B. Chen, B.X. Shen, J.P. Sun, C.X. Wang, H.H. Shan, C.H. Yang, C.Y. Li, Energy
Fuels 26 (2012) 1707–1714.
[18] N. Azizi, S. Ahmed Ali, K. Alhooshani, T. Kim, Y.J. Lee, J.I. Park, J. Miyawaki, S.H.
Yoon, I. Mochida, Fuel Process. Technol. 109 (2013) 172–178.
[19] S. Gong, A. Shinozaki, E.W. Qian, Ind. Eng. Chem. Res. 51 (2012) 13953–13960.
[20] S. Lestari, I. Simakova, A. Tokarev, P. Mäki-Arvela, K. Eränen, D.Y. Murzin,
Catal. Lett. 122 (2008) 247–251.
[21] I. Simakova, B. Rozmysłowicz, O. Simakova, P. Mäki-Arvela, A. Simakov, D.Y.
Murzin, Top Catal. 54 (2011) 460–466.
[22] S. Lestari, P. Mäki-Arvela, K. Eränen, J. Beltramini, G.Q. Max Lu, D.Y. Murzin,
Catal. Lett. 134 (2010) 250–257.
[23] S. Lestari, P. Mäki-Arvela, H. Bernas, O. Simakova, R. Sjöholm, J. Beltramini,
G.Q. Max Lu, J. Myllyoja, I. Simakova, D.Y. Murzin, Energy Fuels 23 (2009)
3842–3845.
[24] J. Fu, X.Y. Lu, P.E. Savage, Energy Environ Sci. 3 (2010) 311–317.
[25] E. Sari, C. DiMaggio, M. Kim, S.O. Salley, K.Y. Simon Ng, Ind. Eng. Chem. Res. 52
(2013) 11527–11536.
[26] E. Santillan-Jimenez, M. Perdu, R. Pace, T. Morgan, M. Crocker, Catalysts 5
(2015) 424–441.
Acknowledgments
[27] S.A.W. Hollak, R.W. Gosselink, D.S. van Es, J.H. Bitter, ACS Catal. 3 (2013)
2837–2844.
[28] M.K. Neylon, S. Choi, H. Kwon, K.E. Curry, L.T. Thompson, Appl. Catal. A 183
(1999) 253–263.
[29] X.W. Chen, T. Zhang, M.Y. Zheng, L.G. Xia, T. Li, W.C. Wu, X.D. Wang, C. Li, Ind.
Eng. Chem. Res. 43 (2004) 6040–6047.
[30] H. Ren, W.T. Yu, M. Salciccioli, Y. Chen, Y.L. Huang, K. Xiong, D.G. Vlachos, J.G.
Chen, ChemSusChem 6 (2013) 798–801.
This work was supported by the National Natural Science Foun-
dation of China (Nos. 21376140 and 21176142), Program for New
Century Excellent Talents in University (No. NCET-12-0308), Pro-
gram for Changjiang Scholars and Innovative Research Team in
University (No. IRT13026), and the Independent Research Programs
of Tsinghua University (No. 20111081067).
[31] J.X. Han, J.Z. Duan, P. Chen, H. Lou, X.M. Zheng, H.P. Hong, Green Chem. 13
(2011) 2561–2568.
References
[32] D.R. Stellwagen, J.H. Bitter, Green Chem. 17 (2015) 582–593.
[33] C.C. Liu, M.G. Lin, D. Jiang, K.G. Fang, Y.H. Sun, Catal. Lett. 144 (2014) 567–573.
[34] Q.L. Zhu, B. Zhang, J. Yang, J.X. Wang, J. Zhao, S.F. Ji, H.Q. Wang, New J. Chem.
27 (2003) 1633–1638.
[35] J. Rouquerol, D. Avnir, C.W. Fairbridge, D.H. Everett, J.M. Haynes, N. Pernicone,
J.D.F. Ramsay, K.S.W. Sing, K.K. Unger, Pure Appl. Chem. 66 (1994) 1739–1758.
[36] K. Liang, X.Z. Tang, W.C. Hu, J. Mater. Chem. 22 (2012) 11062–11067.
[37] Y.M. Sun, X.L. Hu, W. Luo, Y.H. Huang, J. Mater. Chem. 22 (2012) 425–431.
[38] F.F. Xia, X.L. Hu, Y.M. Sun, W. Luo, Y.H. Huang, Nanoscale 4 (2012) 4707–4711.
[39] D.J. Moon, J.W. Ryu, Catal. Lett. 92 (2004) 17–24.
[40] H.Y. Chen, L. Chen, Y. Lu, Q. Hong, H.C. Chua, S.B. Tang, J. Lin, Catal. Today 96
(2004) 161–164.
[41] Z.H. Yang, P.J. Cai, L. Shi, Y.L. Gu, L.Y. Chen, Y.T. Qian, J. Solid State Chem. 179
(2006) 29–32.
[42] N. Perret, X.D. Wang, L. Delannoy, C. Potvinb, C. Louisb, M.A. Keane, J. Catal.
286 (2012) 172–183.
[43] O.G. Marin Flores, S. Ha, Appl. Catal. A 352 (2009) 124–132.
[44] J. Wang, L.F. Shen, B. Ding, P. Nie, H.F. Deng, H. Dou, X.G. Zhang, RSC Adv. 4
(2014) 7538–7544.
[45] M.L. Xiang, D.B. Lia, W.H. Li, B. Zhong, Y.H. Sun, Catal. Commun. 8 (2007)
513–518.
[46] M.L. Xiang, D.B. Lia, H.C. Xiao, J.L. Zhang, W.H. Li, B. Zhong, Y.H. Sun, Catal.
Today 131 (2008) 489–495.
[47] S. Gao, G.D. Li, Y.P. Liu, H. Chen, L.L. Feng, Y. Wang, M. Yang, D.J. Wang, S.
Wang, X.X. Zou, Nanoscale 7 (2015) 2306–2316.
[48] C.C. Liu, M.G. Lin, D. Jiang, K.G. Fang, Y.H. Sun, Catal. Lett. 144 (2014) 567–573.
[49] Q. Gao, X.Y. Zhao, Y. Xiao, D. Zhao, M.H. Cao, Nanoscale 6 (2014) 6151–6157.
[50] J. Torres, J.E. Alfonso, L.D. López-Carren˜o, Phys. Status Solidi 2 (2005)
3726–3729.
[51] R.R. Ding, Y.L. Wu, Y. Chen, J.M. Liang, J. Liu, M.D. Yang, Chem. Eng. Sci. (2014),
http://dx.doi.org/10.1016/j.ces.2014.10.024.
[1] J.G. Immer, H.H. Lamb, Energy Fuels 24 (2010) 5291–5299.
[2] S.F. Gong, A. Shinozaki, E.W. Qian, Ind. Eng. Chem. Res. 51 (2012)
13953–13960.
[3] M. Arend, T. Nonnen, W.F. Hoelderich, J. Fischer, J. Groos, Appl. Catal. A 399
(2011) 198–204.
[4] Y.N. Duan, Y.L. Wu, Q.H. Zhang, R.R. Ding, Y. Chen, J. Liu, M.D. Yang, J. Mol.
Catal. A 398 (2015) 72–78.
[5] S. Lestari, P. Mäki-Arvela, J. Beltramini, G.Q. Max Lu, D.Y. Murzin,
ChemSusChem 2 (2009) 1109–1119.
[6] A. Demirbas, Prog. Energy Combust. 33 (2007) 1–18.
[7] B.X. Peng, C. Zhao, S. Kasakov, S. Foraita, J.A. Lercher, Chem. Eur. J. 19 (2013)
4732–4741.
[8] B. Rozmysłowicz, P. Mäki-Arvela, A. Tokarev, A.R. Leino, K. Eränen, D.Y.
Murzin, Ind. Eng. Chem. Res. 51 (2012) 8922–8927.
[9] J. Monnier, H. Sulimma, A. Dalai, G. Caravaggio, Appl. Catal. A 382 (2010)
176–180.
[10] J.P. Ford, N. Thapaliya, M.J. Kelly, W.L. Roberts, H.H. Lamb, Energy Fuels 27
(2013) 7489–7496.
[11] D.R. Vardon, B.K. Sharma, J. Scott, G. Yu, Z.C. Wang, L. Schideman, Y.H. Zhang,
T.J. Strathmann, Bioresour. Technol. 102 (2011) 8295–8303.
[12] Y. Chen, Y.L. Wu, R.R. Ding, P. Zhang, J. Liu, M.D. Yang, AIChE J. 61 (2015)
1118–1128.
[13] S.P. Zou, Y.L. Wu, M.D. Yang, C. Li, J.M. Tong, Energy Environ. Sci. 3 (2010)
1073–1078.
[14] B.X. Peng, Y. Yao, C. Zhao, J.A. Lercher, Angew. Chem. 124 (2012) 2114–2117.
[15] S.P. Zou, Y.L. Wu, M.D. Yang, K. Imdad, C. Li, J.M. Tong, Energy 35 (2010)
5406–5411.
[16] J. Ancheyta-Juárez, G. Betancourt-Rivera, G. Marroquín-Sánchez, A.M.
Pérez-Arellano, S.K. Maity, M.T. Cortez, R. del Río-Soto, Energy Fuels 15 (2001)
120–127.