Effect of Cu-doping on the structure and performance of molybdenum carbide catalyst for low-temperature hydrogenation of dimethyl oxalate to ethanol

Article abstract of DOI:10.1016/j.apcata.2016.11.009

Several copper-doped molybdenum carbide (Cu–Mo₂C) nanomaterials for the hydrogenation of dimethyl oxalate (DMO) to ethanol at low temperature (e.g., 473 K) have been developed through a facile solid-state pyrolysis method. Characterization techniques including X-ray diffraction, scanning/transmission electron microscopy, N₂-physisorption, energy-dispersive spectroscopy, and X-ray photoelectron spectroscopy were employed to reveal the morphology, structure and properties of the synthesized nanomaterials. The characterization and reaction results suggest that the incorporation of copper species in Mo₂C plays a crucial role in modifying the morphologic structure of Cu–Mo₂C as well as tuning the electronic state of Mo active sites, resulting in an important enhancement in the catalytic performance. Moreover, a strong synergistic effect between Cu and Mo₂C is observed in DMO hydrogenation. Accordingly, the 67.2% yield of ethanol can be attained at a low temperature of 473 K over the Cu–Mo₂C nanomaterials with a suitable atom ratio (e.g., Cu:Mo = 0.03:1), which are higher than those obtained by using a pure Mo₂C (e.g., 13.7%) under the same reaction conditions. The Cu-doped Mo₂C nanomaterials also display excellent catalytic stability during the hydrogenation of DMO to ethanol for longer than 300 h.

Full text of DOI:10.1016/j.apcata.2016.11.009

Accepted Manuscript
Title: Effect of Cu-doping on the structure and performance of
molybdenum carbide catalyst for low-temperature
hydrogenation of dimethyl oxalate to ethanol
Author: Yanting Liu Jian Ding Jicheng Bi Yanping Sun Juan
Zhang Kefeng Liu Fanhua Kong Haicheng Xiao Jiangang
Chen
PII:
S0926-860X(16)30548-8
DOI:
Reference:
http://dx.doi.org/doi:10.1016/j.apcata.2016.11.009
APCATA 16057
To appear in:
Applied Catalysis A: General
Received date:
Revised date:
Accepted date:
28-7-2016
6-11-2016
9-11-2016
Please cite this article as: Yanting Liu, Jian Ding, Jicheng Bi, Yanping Sun, Juan
Zhang, Kefeng Liu, Fanhua Kong, Haicheng Xiao, Jiangang Chen, Effect of Cu-
doping on the structure and performance of molybdenum carbide catalyst for low-
temperature hydrogenation of dimethyl oxalate to ethanol, Applied Catalysis A, General
http://dx.doi.org/10.1016/j.apcata.2016.11.009
This is a PDF file of an unedited manuscript that has been accepted for publication.
As a service to our customers we are providing this early version of the manuscript.
The manuscript will undergo copyediting, typesetting, and review of the resulting proof
before it is published in its final form. Please note that during the production process
errors may be discovered which could affect the content, and all legal disclaimers that
apply to the journal pertain.

Effect of Cu-doping on the structure and performance of molybdenum carbide
catalyst for low-temperature hydrogenation of dimethyl oxalate to ethanol
a,b
c
a
d
a
e
Yanting Liu , Jian Ding , Jicheng Bi , Yanping Sun , Juan Zhang , Kefeng Liu ,
e
e
a,
Fanhua Kong , Haicheng Xiao , Jiangang Chen *
a
State Key Laboratory of Coal Conversion, Institute of Coal Chemistry, Chinese
Academy of Sciences, Taiyuan, 030001, China.
b
University of Chinese Academy of Sciences, Beijing, 100039, China.
c
School of Chemical Engineering, Inner Mongolia University of Science &
Technology, Baotou, 014010, China.
d
Taiyuan University of Technology, Taiyuan, 030001, China.
e
PetroChina Petrochemical Research Institute, Beijing 100195, China.
*
Corresponding author. E-mail address: chenjg@sxicc.ac.cn (J. Chen)
1

Graphical Abstract
2

Highlights





Cu-doped Mo
Cu species are incorporated into Mo
Cu tunes the morphologies and intrinsic properties of carbides.
Synergistic effect between Cu and Mo C is observed in DMO hydrogenation.
Cu‒Mo C can retain the excellent activity for longer than 300 h.
2
C catalyst is synthesized by a solid-state pyrolysis method.
2
C lattice.
2
2
3

A B S T R A C T
2
Several copper-doped molybdenum carbide (Cu‒Mo C) nanomaterials for the
hydrogenation of dimethyl oxalate (DMO) to ethanol at low temperature (e.g., 473 K)
have been developed through a facile solid-state pyrolysis method. Characterization
techniques including X-ray diffraction, scanning/transmission electron microscopy,
2
N -physisorption, energy-dispersive spectroscopy, and X-ray photoelectron
spectroscopy were employed to reveal the morphology, structure and properties of the
synthesized nanomaterials. The characterization and reaction results suggest that the
incorporation of copper species in Mo
morphologic structure of Cu–Mo C as well as tuning the electronic state of Mo active
sites, resulting in an important enhancement in the catalytic performance. Moreover, a
strong synergistic effect between Cu and Mo C is observed in DMO hydrogenation.
Accordingly, the 67.2% yield of ethanol can be attained at a low temperature of 473 K
over the Cu‒Mo C nanomaterials with a suitable atom ratio (e.g., Cu:Mo = 0.03:1),
which are higher than those obtained by using a pure Mo C (e.g., 13.7%) under the
same reaction conditions. The Cu-doped Mo C nanomaterials also display excellent
2
C plays a crucial role in modifying the
2
2
2
2
2
catalytic stability during the hydrogenation of DMO to ethanol for longer than 300 h.
Key words:
Copper, Molybdenum carbide, Hydrogenation, Dimethyl oxalate, Ethanol
4

1
. Introduction
The production of ethanol (EtOH) has drawn extensive attention because of its
relevance to energy and environmental implications [1-3]. The indirect
synthesis of EtOH from syngas is one of the acknowledged powerful routes
[
4-7], which includes syngas production from biomass, organic wastes and
coal, the coupling of CO with nitrite esters to form dimethyl oxalate (DMO)
shown in eq. 1) [8], and subsequent hydrogenation to EtOH through eq. 2. In
(
fact, the first two steps have been successfully industrialized [9-13]. Because of
the
2
CO + 2CH
3
NO
2
→ CH
3
OOCCOOCH
3
+ 2NO
(1)
(2)
CH OOCCOOCH
3
3
+ 5H
2
→ CH CH OH + 2CH OH + H O
3
2
3 2
extremely high activity for the hydrogenation of DMO to EtOH, copper is
currently adopted as the main DMO hydrogenation catalyst [7, 14-16].
However, a poor lifespan due to the agglomeration of copper particles under
reaction conditions is the major drawback for Cu-based catalysts in
industrialized applications [17-19]. Furthermore, to facilitate EtOH formation,
the hydrogenation of DMO is generally catalyzed by Cu-based catalysts at high
reaction temperatures (543–553 K) [4, 19, 20], which is far beyond the Hüttig
temperature of Cu (407 K) [21] and is a high energy consumption process.
Therefore, exploration of alternative catalytic materials with low reaction
temperatures and high catalytic activity is essential for developing a high
5

efficiency DMO hydrogenation catalyst toward industrial-scale EtOH
production.
2 2
More recently, we demonstrated silica-supported molybdenum carbide (Mo C/SiO )
exhibits high catalytic performance in the hydrogenation of DMO to EtOH [22], and
the reaction temperature can be greatly reduced from 553 K to 473 K. So far as we
know, it is the lowest reaction temperature for vapor phase hydrogenation of DMO to
EtOH. More specifically, this novel catalyst can achieve 100% DMO conversation
with ca. 70% EtOH selectivity at a lower temperature of 473 K. In the real world
2
practice, however, the utilization of silica as a support material to disperse Mo C
particles suffers a fatal flaw ascribe to its severe leaching caused by the reaction of
silica with methanol in this reaction system [23], which leads to their weak
mechanical strength. As such, research on the synthesis and catalytic activities of
unsupported molybdenum carbide catalysts will be of considerable value. It is
2
well-known that the activity of Mo C can be improved by constructing various
nanostructures, such as nanotubes [24], nanowires [25], and nanospheres [26], etc.,
which may enhance exposure of surface active sites. In addition, the addition of other
transition metals (e.g. Fe, Co, and Ni, etc.) is also an effective way to improve the
catalytic performance of Mo
and/or the tuned electronic state of Mo active sites [27, 28]. For instance, Co-modified
Mo C has been studied for its excellent performance in hydrogenation of CO [29, 30],
reduction of CO and conversion of aromatic nitro compounds [31, 32]. Furthermore,
2
C owing to the promoted formation of metal carbides
2
2
2
Kelly et al. claimed that Ni or Cu modified Mo C could control C–O, C–C and C–H
6

bond scission for deoxygenation, reforming, and dehydrogenation of EtOH [33]. Wan
et al. reported the possibility of doping Fe or Ni into the lattice of Mo C [34].
Similarly, the interaction between Ni and Mo C was also studied as an enhanced
catalyst for hydrogenation reactions [35]. Recently, Posada-Pérez et al. found that the
deposition of Au and Cu on δ-MoC and β-Mo C surfaces can effectively improve the
performance of the catalyst for the hydrogenation of CO [36, 37]. However, to the
best of our knowledge, the application of transition-metal-doped Mo C nanomaterials
within the DMO hydrogenation has not yet been reported. Among the transition
metals, Cu has high activity for H dissociation and the selective hydrogenation of
2
2
2
2
2
2
carbon–oxygen bonds, while remaining relatively inactive for the hydrogenolysis of
carbon–carbon bonds [15]. Thus, evaluating catalytic activities toward DMO
hydrogenation of Cu-doped Mo
2
C nanomaterials has merit.
In this study, a series of copper-doped molybdenum carbide (Cu–Mo
2
C)
nanomaterials were synthesized by a facile solid-state thermolysis route. In this
method, hexamethylenetetramine (HMTA)-related molybdate with uniform
2
+
copper ion (Cu ) doping was pre-prepared in ammonia aqueous solution and
used as the precursor of solid-state thermolysis. With the aid of X-ray
diffraction (XRD), scanning electron microscopy (SEM), transmission electron
microscopy (TEM), energy-dispersive spectroscopy (EDS), and X-ray
2
+
photoelectron spectroscopy (XPS), we found that the introduction of Cu in the
precursor solution played a crucial role in tailoring the morphology and tuning
the intrinsic properties of carbides, hence effectively affected their catalytic
7

performances. Furthermore, the synergistic effect between Cu and Mo
the catalyst to exhibit unexpected catalytic performance in DMO
hydrogenation. As a result, the optimized Cu–Mo C catalyst showed 100%
2
C allows
2
conversion of DMO with 67.2% selectivity of EtOH, which is comparable to
that of silica-supported molybdenum carbide catalyst [22]. More importantly,
the SiO
the reported Si-leaching problem during the hydrogenation of DMO [23, 38].
. Experimental
2 2
-free characteristics of our Cu-doped Mo C nanomaterials can avoid
2
2.1 Catalyst preparation
In this work, a solid-state thermolysis route with ammonium molybdate
tetrahydrate ((NH
sources, respectively, was used to synthesize nanostructure Mo
according to literature procedures [39, 40]. However, it was found that the
obtained Mo C nanoparticles prepared though this method could easily
aggregate to form a block-like structure, which is detrimental for Mo C-based
4
)
6
Mo
7
O
24·4H
2
O) (AMT) and HMTA as the Mo and C
2
C materials
2
2
catalysts because aggregation results in low specific surface area and
subsequently low density of catalytic active sites. To modify the morphologies
2
and intrinsic properties of Mo C nanomaterials, the synthetic strategy was
2
+
improved by introducing Cu into the precursor solution. The addition amount
of AMT and HMTA remained unchanged with a molar ratio of 1: 8. Typically,
3.0 g of AMT, 2.8 g of HMTA and 0.08 g of copper nitrate trihydrate
(Cu(NO
3
)
2
·3H
2
O) were dissolved in 25% ammonia aqueous solution in a
8

beaker under mechanical stirring. After stirring for 24 h at 298 K, the mixture
dried in vacuo at 353 K, yielding a blue solid, which was used as the precursor.
-1
Subsequently, the precursor was heated to 1023 K at a rate of 15 K min under
-
1
an argon flow of 20 mL min , and then maintained for 2 h to produce a black
powder. Finally, the sample was cooled down to room temperature naturally.
Four Cu-doped Mo
.04:1, and 0.05:1 were synthesized through this process by varying the amount
of added Cu(NO ·3H O. The obtained products were denoted as sample xCu‒
Mo C, where x represented the preset atomic ratio of Cu/Mo. For comparison,
Cu-mixed Mo C catalyst was also prepared by a synthesis method of
post-impregnation. In a typical procedure, a certain quantity of aqueous
Cu(NO ·3H O solution was added into a 100 mL round-bottom flask
containing pure Mo C under flowing argon gas. The mixture was stirred for 12
2
C nanomaterials with Cu/Mo atomic ratios of 0.02:1, 0.03:1,
0
3
)
2
2
2
2
3
)
2
2
2
h at 298 K to form a homogenous ink, and dried at 393 K for 12 h. Finally,
calcined for 4h at 623 K in argon, yielding the catalyst precursor denoted as
xCu‒Mo
.2 Catalyst characterization
The XRD patterns for each sample were obtained on a Bruker AXS-D8
2
C-i.
2
diffractometer with monochromatic Cu Kα radiation source (λ = 0.15418 nm)
operated at 30kV and 15mA. Nitrogen adsorption-desorption analyses of the
samples were carried out with Micromeritics Tristar II 3020. Specific surface
areas of the samples were calculated from single point nitrogen desorption
9

peaks. Pore volume and pore size distributions were estimated by the Barrett–
Joyner–Halenda (BJH) method. TEM and dark-field scanning TEM (STEM)
measurements were carried out on a JEM-2100F high-resolution transmission
operated at 200 keV. Samples for TEM were prepared by placing a drop of a
dispersed solution onto a standard nickel grid. SEM images were collected on a
Zeiss Supra 55. EDS maps were gathered on SEM. Characterizations of XPS
were conducted on a VG MiltiLab 2000 spectrometer equipped with an Al Kα
radiation (1486.6 eV). Prior to the measurements, the precursor of each sample
was pressed into a thin disk and calcinated under an argon flow at 1023 K for 2
h in an auxiliary pretreatment chamber. After the calcination, the obtained
sample was directly introduced into the XPS chamber to avoid exposure to air.
The XPS spectra were recorded at room temperature.
2 2
H -temperature-programmed desorption (H -TPD) of catalyst was carried out
using a quadrupole mass spectrometer (HIDEN, Hpr-20; Pfeiffer Vacuum
Technology AG) equipped by the following procedure. The catalyst was first
activated in diluted hydrogen (10% H
maintained for 1 h under Ar flow at 673 K to sweep the surface of activated
catalyst. After that, the sample was cooled down to 303 K and exposed to H
for 1 h. Finally, the sample was heated from 323 to 973 K under Ar flow with a
2 2
in N ) at 673 K for 3 h, and then
2
-
1
ramping rate of 10 K min .
2.3 Catalytic activity test
10

Catalytic activity tests for the hydrogenation of DMO were carried out using
a fixed-bed reactor. Typically, 0.5 g of sample (40–60 meshes) were packed
into a stainless steel tubular reactor (i.d. = 8 mm). Prior to the measurement, all
2 2
samples were activated by a stream of diluted hydrogen (10% H in N ) at 673
-
1
K under atmospheric pressure for 3 h with a flow rate of 30 mL min . After
cooling to the reaction temperature, the system pressure was precisely
controlled at 2.5 MPa. 12.5 wt% DMO in methanol was injected by
double-plunger pump and pure H
BROOKS) was fed into the reactor. The obtained products were condensed
and analyzed by a gas chromatograph (Shimadzu Trace GC ultra) fitted with a
2
which controlled by mass flowmeter
(
3
3
3
0 m HP-5 capillary column and a flame ionization detector (FID).
. Results and discussion
.1 Characteristics of xCu–Mo
Table 1 illustrates the physical characterizations of the as-prepared Mo
Cu-doped Mo C nanomatreials. Fig. 1 shows the nitrogen physisorption
isotherms and the corresponding pore size distributions of all samples. If no
2
C
2
C and
2
2
+
Cu is introduced into the synthesis, Mo
2
C sample exhibits a typical type II
isotherm according to IUPAC classification (Fig. 1A) [41], suggesting that the
2
−1
material is almost nonporous with a low surface area of 7.9 m g . After
doping of Cu, however, the isotherms of samples are intermediate between type
II and type IV with an H3 hysteresis loop when the Cu/Mo atomic ratio is lower
0.04:1, suggesting that these samples hold mesoporous structures, as shown in
11

Fig. 1B, pore-size distributions (2−60 nm) were observed for the Cu-doped
Mo C nanomaterials. Furthermore, the specific surface area and mesopore
2
structures first increases and then decreases as Cu content increasing. With a
further increasing of Cu/Mo atomic ratio (0.05:1), the isotherm restores to the
type II shape, indicating the disappearance of the mesopore structure, which
leads to a considerable drop in the specific surface area. The results clearly
indicate that the pore structures of carbides are easily changed after introducing
Cu species into molybdenum carbide. The change of the pore structure may
result from the formation of different morphology of MoO
during the pyrolysis process.
x
and/or Mo-Cu alloy
To further support the above observation and to clarify the microstructures of
Mo C and Cu-doped Mo C nanomaterials, the morphologies of all the samples
were investigated by SEM. As displayed in Fig. 2A, if no Cu is introduced
into the precursor solution, Mo C particles are aggregated from a block-like
structure and are almost nonporous, in good agreement with previous reports
39, 40]. However, the morphologic structure of Mo C is changed greatly after
the introduction of Cu species. For a Cu‒Mo C sample with a very low Cu
content (Cu/Mo = 0.02, Fig. 2B), for instance, many block particles in the
original Mo C seems to be broken into numerous small nanorods about 80‒120
nm in length. Besides, a lot of mesopores can be observed in the sample, as
indicated in high-magnification SEM image of 0.02Cu‒Mo C in Fig. 2B, which
has decreased in apparent density by comparison with the pure Mo C sample.
2
2
2
+
2
[
2
2
2
2
2
12

When the Cu/Mo atomic ratio is increased to 0.03:1, we notice the resulting
products have a loose and mesoporous structure that is assembled from
2
nanorods about 60 nm in length (Fig. 2C). The porous nature of 0.03Cu‒Mo C
sample should significantly benefit the catalytic DMO hydrogenation because
porous structures could provide channels that allow efficient transport of
reactants and products [21]. For a higher Cu/Mo atomic ratio of 0.04:1,
however, the large nanosheets becomes dominant and only a few small
nanorods exist in the sample (Fig. 2D). This results in a considerable loss to
mesoporous structure and specific surface area. When atomic ratio of Cu:Mo is
further increased to 0.05:1, the sample appears to lose the porosity and become
a more compact solid structure (Fig. 2E), which significantly decreases the
density of exposed edge structures. These observations are in good agreement
with the results obtained from nitrogen physisorption. The representative TEM
images of all samples shown in Fig. 3 also agree with the findings that, the
2
+
introduction of Cu during the synthesis leads to the formation of rod-like or
sheet-like Cu‒Mo C nanomaterials with higher specific surface area, which
enhances exposure of surface active sites. Obviously, the addition of Cu
during the synthesis plays a crucial role in regulating the development of Cu‒
2
2
+
2
+
Mo
2 2
C nanostructures. The introduction of Cu may suppress the Mo C crystal
growth along the basal planes, thus lowering the probability of formation of
2
+
agglomerates among the nanoparticles [42]. However, excessive Cu will
13

result in the morphology of the Cu-doped Mo
Mo C. Therefore, the optimized Cu/Mo atomic ratio in our synthesis is 0.03:1.
The EDS measurements were performed to study the chemical composition
of pure/Cu-doped Mo C samples (Fig. S1). The spectrum of Mo C indicates
that the atomic ratio of Mo/C is close to 2:1, which is consistent with the
stoichiometric ratio of Mo C. Moreover, the Cu/Mo molar ratios in 0.02Cu‒
Mo C, 0.03Cu‒Mo C, 0.04Cu‒Mo C and 0.05Cu‒Mo C are listed in Table 1.
These data are consistent with the atomic ratio of Cu/Mo in the starting
precursor salts, EDS mapping of large areas for these four Cu-doped Mo
2
C returned back to that of pure
2
2
2
2
2
2
2
2
2
C
nanomaterials proves both Mo and Cu are uniformly distributed and overlapped
in the particles (Fig. S2). To further confirm the elemental composition of the
2
xCu‒Mo C samples, XPS measurements were carried out. As shown in Fig. S3,
the spectrum recorded from 0 to 1100 eV shows the presence of Mo, Cu, and C
elements. It is noteworthy, however, that the Cu/Mo atomic ratios measured by
XPS are remarkably lower than the corresponding bulk Cu/Mo atomic ratios
determined by EDS (Table 1), which may be attributed to the photoelectron
attenuation lengths ranging from approximately 2 nm to 5 nm [43]. On the basis
of the results, we believe that the main constituents in all the Cu-doped Mo
nanomaterials should be Cu-incorporated Mo C. EDS mapping images of small
areas were collected in TEM to further determine the surface distributions of
Cu and Mo species on the Cu-doped Mo C nanomaterials, as shown in Fig. 4.
One can see that both Mo and Cu in all the Cu-doped Mo C samples are
2
C
2
2
2
14

uniformly distributed in the randomly selected area. Meanwhile, the
distribution of incorporated copper species on the structure is very
homogeneous, although its concentration is slightly higher near the defect
regions.
Both XRD patterns of Mo
2
C and Cu-doped Mo
2
C nanomaterials show
well-resolved peaks corresponding to the β-Mo
2
C (JCPDS No. 35-0787) with a
hexagonal close packed structure in the 30–85° range (Fig. 5A). The diffraction
peaks around 43.3°correspond to the (111) lattice planes of Cu (JCPDS No.
0
4-0836) (Fig. 5C). Furthermore, the diffraction peak intensity of Cu gradually
decreases with decreasing Cu contents in the Cu-doped Mo C samples. For
2
Cu/Mo atomic ratios lower than 0.04:1, the diffraction peak of Cu cannot be
observed, possibly because the Cu has been dispersed well on the surface of
β-Mo
comparing the XRD patterns of Mo
a slight shift to higher diffraction angle in comparison with pure Mo
observed, which indicates that lattice compression takes place after doping Cu
into the β-Mo C. The shrink of Mo C lattice may be attributed to the formation
of Mo-Cu alloy or some copper is getting into the β-Mo C lattice and the
replacement of some Mo atoms with smaller Cu in lattice [44]. This
observation is consistent with the Fe, Co, and Ni-doped Mo C [28, 34, 44].
2
C or is incorporated into the β-Mo
2
C lattice. Through carefully
C samples (Fig. 5B),
C can be
2
C and Cu-doped Mo
2
2
2
2
2
2
Thus it allows the existence of a strong interaction between Cu and carbide,
which could be also benefit for its catalytic activity.
15

Fig. 6 shows representative high-resolution TEM (HRTEM) images of Mo
and 0.03Cu‒Mo C, revealing the structural changing caused by incorporation
of Cu species into Mo C. Pure Mo C hold high crystallinity and well-defined
atomic planes (Fig. 6A, inset). The planar d (0.229 nm and 0.239 nm) of Mo
particles is consistent with the (101) and (002) planes of β-Mo C, respectively.
However, the crystalline quality of 0.03Cu‒Mo C is importantly reduced after
doping Cu to Mo C, as shown in Fig. 6B, a large number of twin boundaries
and stacking faults can be observed in the surface of 0.03Cu‒Mo C particles,
which can be recognized as highly energetic sites in the sample [45]. The Cu is
incorporated into a Mo C matrix can be justified from the clear lattice fringe of
.203 nm, as indicated by red circles and arrows in Fig 6B. It is worth
mentioning that the (002) plane spacing of 0.03Cu‒Mo C are reduced to 0.235
nm from 0.239 nm observed in bare Mo C, which again confirms that lattice
compression takes place after the incorporation of Cu into the β-Mo C.
2
C
2
2
2
2
C
2
2
2
2
2
0
2
2
2
To further understand the elemental chemical states of all samples, XPS
measurements were carried out. Overlapping peaks were deconvoluted, and the
peaks were identified as different chemical states. In Mo 3d region, there are
0
3+
two oxidation states (Mo and Mo ) on the surface of xCu‒Mo
2
C particles
0
3+
(
Fig. 7). The dominant Mo peaks along with small peaks of Mo can be
ascribed to the Mo sites coordinated to Mo or C neighbouring atoms in
0
2
molybdenum carbide [46]. Compared with pure Mo C, the peaks for Mo
around 228.95 and 232.12 eV assigned to the Mo 3d5/2 and 3d3/2 in the
16

2
Cu-doped Mo C samples have lower binding energy values. The downshift of
the Mo binding energy may be attributed to the strong interaction between Mo
and Cu species [47]. Furthermore, Mo binding energy decreases with increasing
Cu/Mo atomic ratios. Larger Cu/Mo atomic ratios result in lower Mo binding
energy, implying that the electronic structure of Mo is modified when Cu is
doped into Mo
increasing Cu/Mo atomic ratios in the Cu-doped Mo
Additionally, Fig. 8A shows the characteristic peak of Cu 2p in the Cu-doped
Mo C. The peak positions around 952.8 and 932.9 eV for Cu 2p1/2 and Cu 2p3/2
may be assigned to metallic copper, respectively. No additional peaks for Cu 2p
2
C. The electronic effect between Cu and Mo
2
C increases with
2
C nanomaterials.
2
2
+
satellite at a binding energy of 942–944 eV corresponding to Cu species are
0
observed. The Cu 2p3/2 BE corresponds to Cu for 0.02Cu‒Mo
2
C, 0.03Cu‒
Mo
and 933.3 eV, respectively. Compared with bulk Cu (932.4 eV) [48], the Cu in
the xCu‒Mo C samples has higher binding energy values, which is the
indicative of the charge transfer from Cu metal to the matrix of Mo C,
indicating a strong interaction between the Cu and Mo C. Given that the Cu
3/2 and Cu 2p1/2 binding energies for Cu and Cu are nearly the same, Cu
2 2 2
C, 0.04Cu‒Mo C and 0.05Cu‒Mo C sample located at 932.8, 932.9, 933.1,
2
2
2
+
0
2
p
LMM XAES is usually adopted to discriminate these peaks [49]. As shown in
Fig. 8B, a single symmetric peak at a binding energy of 918.1 eV is observed.
2
The above results confirm that the Cu species in the xCu‒Mo C samples were
0
predominantly present as Cu form.
17

3
.2 Catalytic performance of xCu‒Mo
It is known that the hydrogenation of DMO comprises several continuous
reactions [20]. Our previous study indicated that the reaction pathway for the
formation of EtOH from DMO over Mo C-based catalysts seems to include
methyl acetate (MA) intermediate [22]. As shown in eqs 3‒5, DMO reacted
with H to form methyl glycolate (MG), first, and then hydrogenated to MA
2
C
2
2
instead of ethylene glycol (EG) forming over the Cu catalyst. At last EtOH
would be obtained by deep hydrogenation of MA. Moreover, by-products of
methanol and methyl formate were also produced due to the C–C bonds
scission (eqs 6‒9).
CH
HOCH
CH COOCH
CH OOCCOOCH
HOCH
CH COOCH
HCOOCH
3
OOCCOOCH
COOCH + H
+ 2H → CH
3
+ 2H
→ CH
CH
→ 2HCOOCH
2
→ HOCH
COOCH
OH + CH
2
COOCH
+ H
OH
3
+ CH
3
OH
(3)
(4)
(5)
(6)
(7)
(8)
(9)
2
3
2
3
3
2
O
3
3
2
3
2
3
3
3
+ H
2
3
2
COOCH
3
+ H
2
3 3
→ CH OH + HCOOCH
3
3
+ H
2
→ CH
4
+ HCOOCH
3
3
+ 2H
2
→ 2CH
3
OH
In the present study, an evaluation of the catalytic performance of all
nanomaterials in DMO hydrogenation was conducted under identical reaction
-
1
-1
2
conditions: 473 K, 2.5 MPa, H /DMO = 200 (mol mol ), and liquid hourly
−
1
space velocity (LHSV) = 0.2 h . The test results are shown in Table 2.
Remarkably, most of the Cu-doped Mo C nanomaterials show high catalytic
2
18

activity, with a DMO conversion above 79.8%. In contrast, the pure Mo
sample demonstrates only 74.2% conversion of DMO, indicating that the
introduction of Cu species into Mo C can greatly enhance the hydrogenation
activity, as compared with pure Mo C. Meanwhile, the Cu/Mo atomic ratio has
2
C
2
2
an important impact on tuning the selectivity of EtOH during DMO
hydrogenation. Introducing a very small amount of Cu (e.g., Cu/Mo = 0.02:1)
in the Mo
times as high as that of pure Mo
.03:1, the catalyst possesses the maximum EtOH yield of 67.2%. However,
2
C catalyst increases the EtOH yield to 30.8%, which is almost 2.3
2
C sample. When the Cu/Mo atomic ratio is
0
further increases in the Cu/Mo atomic ratio will deteriorate the yield of EtOH.
It should be noted that increasing Cu/Mo atomic ratio concurrently suppressed
the methanol formation during the hydrogenation of DMO (Table 2). Previous
studies indicated that Cu sites account for the selective hydrogenation of C−O
bonds and are relatively inactive for the hydrogenolysis of C−C bonds [4, 7, 15].
Thus, the hydrogenolysis of C−C bonds can be hindered by the doping of
copper with a larger amount.
To rigorously compare the catalytic performances between Mo
2
C and
Cu-doped Mo C nanomaterials, the DMO conversion was kept below 20% by
2
raising the LHSV. The DMO conversion data were then used to calculate the
turnover frequencies (TOF), as listed in Table 2. The TOF exhibits a volcano
type of dependence on Cu content, which reaches its maximum value of 0.58
-
1
h when the Cu/Mo atomic ratio is 0.03:1. It is highly consistent with the trend
19

of DMO hydrogenation activity presented above. Such a tendency of catalytic
performance of Cu-doped Mo C nanomaterials is also actually in agreement
with their physical properties as indicated in Fig. 1–4 and Table 1. Especially,
.03Cu–Mo C has a loose and mesoporous structure with the largest specific
2
0
2
surface area among the as prepared nanomaterials. In general, higher specific
surface area can provide more active sites where the catalytic reaction takes
2
place. Meanwhile, the presence of large mesopores in 0.03Cu–Mo C sample
facilitates diffusion of the reactants and products during DMO hydrogenation.
An important feature of the catalytic hydrogenation of DMO process on
metal catalysts is that the hydrogenation capacity of catalysts can be increased
by elevating the reaction temperature. We conducted the reaction over the pure
Mo
under a constant LHSV, as shown in Fig. 9A and B. For the pure Mo
DMO conversion is only 50.7% at 453 K, while the 0.03Cu‒Mo
exhibits an excellent catalytic activity at the same temperature, up to 100%
conversion of DMO. Similarly, the selectivity for EtOH over 0.03Cu‒Mo
attained 67.2% at a lower temperature of 473 K. However, it is only 18.3% at
73 K and exceeds 51.1% at higher temperature of 493 K over pure Mo
sample. Obviously, the catalytic capability of DMO hydrogenation over the
Cu-doped Mo C catalyst is significantly enhanced, especially the activity at low
temperatures. That is to say, the use of the Cu-doped Mo C nanomaterials will
effectively decrease the reaction temperature of DMO hydrogenation. One
2
C and optimized 0.03Cu‒Mo
2
C catalyst at different reaction temperatures
C sample,
C catalyst
2
2
2
C
4
2
C
2
2
20

2
interesting feature is that formation of MA from DMO over the 0.03Cu‒Mo C
catalyst is favorable at temperatures lower than 473 K, whereas elevated
temperatures (e.g., >473 K) aggravate the hydrogenolysis of C−C bonds and
induce the dehydration of EtOH to form diethyl ether, which diminish the
selectivity for EtOH. To the best of our knowledge, Cu-based catalysts in
previous literature reports generally attained a high level of selectivity for EtOH
only at high temperature of 553 K (Table S1) [4, 7]. The EtOH selectivity of
2
the Cu-doped Mo C catalyst is 67.2% at low temperature of 473 K in the
current study, which is higher than the reported Cu-based catalysts at relatively
low temperatures.
To investigate the long-term stability of the Cu-doped Mo
2
C nanomaterials,
the vapor phase hydrogenation of DMO was performed over the 0.03Cu‒Mo
2
C
catalyst for 300 h. As shown in Fig. 9C, the catalyst give a 100% DMO
conversion and more than 67% selectivity to EtOH after 37 h on stream and
then remained relatively constant for at least 300 h, indicative of a good
stability for this type of catalyst. The spent catalyst after 300 h on stream shows
the same phases as the fresh catalyst in the XRD patterns (Fig. 10A), which is
quite consistent with the stable catalytic activity of the 0.03Cu‒Mo
SEM and TEM analyses indicate that the 0.03Cu‒Mo C catalyst have no
noticeable change in morphologic structure after 300 h on stream (Fig. 10B−D),
which again shows the superior stability of the Cu-doped Mo C nanomaterials.
2
C catalyst.
2
2
21

It has been reported that nanocatalysts with dual components can exhibit high
activity for various reactions largely as a result of the synergistic effect of two
active species [38, 50, 51]. To explore the effects of Cu on the catalytic
performance of Mo
only the Cu source in HMTA (denoted as 0.03Cu−0Mo
.03Cu−0Mo C catalyst exhibits only 4.3% DMO conversion at 473 K, even
though it has high MG selectivity (83.1%) for the hydrogenation of DMO
Table 2), indicating that Mo C species might be the main active center during
the DMO hydrogenation to EtOH. Based on the finding, the Cu species only
promotes the hydrogenation reaction when it is coupled with Mo C.
Characterizations of XPS, XRD, and HRTEM revealed that the strong
interactions occurred between Cu and Mo C species. Indeed, the diffraction
peaks for Mo C in the Cu-doped Mo C samples showed a slight positive shift
in comparison with pure Mo C sample from the XRD (Fig. 5), suggesting that
lattice compression takes place after the introduction of Cu into the Mo C. This
2
C in detail, we prepared reference sample by pyrolysis of
2
C). The
0
2
(
2
2
2
2
2
2
2
conclusion is also supported by the HRTEM images (Fig. 6). Apart from the
possible lattice strain effects, heterometallic bonding interactions, termed the
‘‘electronic effect’’, as indicated in Fig. 7 and Fig. 8, between Mo atoms with
lower band energy and Cu atoms with higher energy and can result in charge
transfer from Cu to Mo, thereby enhancing chemical reactivity of Mo atoms. It
is difficult to separate the strain and electronic effects because they occur
simultaneously [52, 53]. We believe that both these strain and electronic effects
22

play the role of in improving the catalytic performance of xCu−Mo
nanomaterials. Besides, the abundance of highly energetic sites (twin
boundaries and stacking faults) observed on 0.03Cu−Mo C sample (Fig. 6)
might also be the more active than that of single crystal Mo C for the DMO
hydrogenation. To elucidate the synergistic effect of the added copper on the
catalytic performance of the Mo C catalyst for the hydrogenation of DMO, the
reference 0.03Cu−Mo C-i catalyst was also prepared by the post-impregnation
method. The catalyst exhibits superior DMO conversion (81.7%) and EtOH
selectivity (26.2%) compared with the pure Mo C and 0.03Cu-0Mo C sample
Table 2), confirming that the Cu species promote the DMO hydrogenation
2
C
2
2
2
2
2
2
(
activity of the Mo
2
C catalyst when it is coupled with Mo
2
C. Vidal et al.
hydrogenation
dissociation.
Synergistic effects between the metal and carbide have also been reported for
Pt/Mo C catalysts in water−gas shift [47] and Pt–Mo N/SBA-15 catalysts
during the selective hydrogenation of cinnamaldehyde [55]. It is plausible that
Cu species on the 0.03Cu‒Mo C catalysts enhanced the hydrogen surface
concentration, thereby facilitating the hydrogenation of DMO. As shown in Fig.
1, the presence of two strong H desorption peak around 495 and 600 K
correspond to chemisorbed H on the Mo C surface for pure Mo C. After
doping of Cu into Mo C, however, the intensity of H desorption peaks became
stronger and shift to lower temperature, indicating that the introduction of Cu
reported evidence of synergy between Cu or Au and TiC in CO
2
to methanol [54], both Cu and Au are effective sites for H
2
2
2
2
1
2
2
2
2
2
2
23

species into Mo
coverage of the active H-species.
.3 Effects of carbide phase on the catalytic activities of xCu‒Mo
Compared with β-Mo C, γ-MoC is more stable and formed at higher
2 2
C effectively improves H activation and enhances the surface
3
2
C
2
pyrolysis temperature (e.g., above 900 °C) [46, 56]. Previous studies revealed
that γ-MoC is highly efficient to promote the catalytic activity of noble metal
catalysts due to its unique d-band electronic structures [57, 58]. To the best of
our knowledge, however, there are few studies related to the exploration of
γ-MoC used in DMO hydrogenation. Herein, we the introduction of γ-MoC into
2
optimized 0.03Cu‒Mo C catalyst tried to verify the effect of crystalline phases
of molybdenum carbide on the catalytic performances in DMO hydrogenation.
From the XRD patterns, γ-MoC contents are increased with the increase of
pyrolysis temperature in the range 650‒950 °C (Fig. 12A‒D). However, the
DMO conversation drops while the γ-MoC contents increases (Table 3), which
suggests that the contents of γ-MoC as well as Cu species affect the DMO
hydrogenation activity. While the calcination temperature increases to 950 °C,
there is a great deal of γ-MoC formed in the products (Fig. 12D).
Correspondingly, the obtained material (0.03Cu‒Mo
conversion of only 47.6%, indicating an obviously lower DMO hydrogenation
activity than those of these samples contained only β-Mo C phase. Therefore,
the increase of γ-MoC contents in the Cu-doped Mo C nanomaterials is not an
effective strategy to enhance DMO hydrogenation activity. Detailed studies of
2
C-950) shows a DMO
2
2
24

the mechanisms that lead to the drastic activity decreases of these
γ-MoC-modified Cu-doped Mo C catalysts are underway.
. Conclusions
In summary, a series of Cu-doped Mo
low-temperature hydrogenation of DMO to EtOH. Characterization with XRD
and HRTEM show that β-Mo C is the main crystal phase in these Cu-doped
Mo C nanomaterials. Investigation of the nanomaterials by SEM, TEM and
XPS reveal that the incorporated Cu species into Mo C play vital roles in
tailoring the morphology as well as regulating intrinsic properties of carbides.
The optimized 0.03Cu‒Mo C nanomaterials exhibit high catalytic performance
in DMO hydrogenation, such as an EtOH yield of 67.2% at low temperature of
73 K, which is almost 4.9 times higher than that of the one without copper
introduction. Moreover, a strong synergistic effect between Cu and Mo C in the
Cu-doped Mo C catalyst is observed and plays an important role in the
reaction. Finally, we find that the presence of γ-MoC leads to lower the
catalytic activity of the Cu-doped Mo C nanomaterials in DMO hydrogenation.
2
4
2
C nanomaterials were investigated for
2
2
2
2
4
2
2
2
25

(1) Several copper-doped molybdenum carbide catalysts are synthesized by a
solid-state pyrolysis method;
(
2) Cu species are incorporated into Mo
2
C lattice;
(3) Cu tunes the morphologies and intrinsic properties of carbides, resulting in an
important enhancement in the catalytic activity;
4) A strong synergistic effect between Cu and Mo
hydrogenation.
5) Cu-Mo C nanomaterials can retain the excellent activity for longer than 300 h.
(
2
C is observed in DMO
(
2
Acknowledgements
The authors gratefully acknowledge the financial support from the State Key
Laboratory of Coal Conversion, PetroChina (PRIKY15042, 14006, 15038, 15039)
and National Natural Science Foundation of China (Grant No. 21373254 and No.
2
1503256).
26

References:
1]A.E. Farrell, R.J. Plevin, B.T. Turner, A.D. Jones, M. O'Hare, D.M. Kammen,
Science. 311 (2006) 506-508.
[
[
[
[
2]C.J. Cleveland, C.A.S. Hall, R.A. Herendeen, Science. 312 (2006) 1746-1748.
3]J. Goldemberg, Science. 315 (2007) 808-810.
4]J. Gong, H. Yue, Y. Zhao, S. Zhao, L. Zhao, J. Lv, S. Wang, X. Ma, J. Am. Chem.
Soc. 134 (2012) 13922-13925.
[
[
[
5]Y. Zhu, X. Kong, X. Li, G. Ding, Y. Zhu, Y. Li, ACS Catal. 4 (2014) 3612-3620.
6]H. Yue, X. Ma, J. Gong, Accounts Chem. Res. 47 (2014) 1483-1492.
7]S. Zhao, H. Yue, Y. Zhao, B. Wang, Y. Geng, J. Lv, S. Wang, J. Gong, X. Ma, J.
Catal. 297 (2013) 142-150.
8]Q. Li, Z. Zhou, R. Chen, B. Sun, L. Qiao, Y. Yao, K. Wu, Phys. Chem. Chem.
Phys. 17 (2015) 9126-9134.
9]R. Zehner, US4005131, 1977.
[
[
[
[
[
10]M. Mitsunobu, N.K. F. Kouzou, M. Masaoki, S. Hiroyuki, US4005131, 1979.
11]W.J. Bartley, W.V. Charleston, US4677234, 1987.
12]X. Gao, Y. Zhao, S. Wang, Y. Yin, B. Wang, X. Ma, Chem. Eng. Sci. 66 (2011)
3
513-3522.
13]G. Zhenghong, L. Zhongchen, H. Fei, X. Genhui, J. Molec. Catal. A. 235 (2005)
43-149.
14]Y. Wang, Y. Shen, Y. Zhao, J. Lv, S. Wang, X. Ma, ACS Catal. 5 (2015)
[
1
[
27

6
200-6208.
[15]H. Yue, Y. Zhao, X. Ma, J. Gong, Chem. Soc. Rev. 41 (2012) 4218-4244.
[16]J. Ding, J. Zhang, C. Zhang, K. Liu, H. Xiao, F. Kong, J. Chen, Appl. Catal. A.
5
08 (2015) 68-79.
[17]M.V. Twigg, M.S. Spencer, Appl. Catal. A. 212 (2001) 161-174.
[18]J. Zheng, J. Zhou, H. Lin, X. Duan, C.T. Williams, Y. Yuan, The J. Phys. Chem.
C. 119 (2015) 13758-13766.
19]Y. Zhao, S. Zhao, Y. Geng, Y. Shen, H. Yue, J. Lv, S. Wang, X. Ma, Catal.
Today (2016).
20]Y. Song, J. Zhang, J. Lv, Y. Zhao, X. Ma, Ind. Eng. Chem. Res. 54 (2015)
699-9707.
21]H. Yue, Y. Zhao, L. Zhao, J. Lv, S. Wang, J. Gong, X. Ma, AICHE J. 58 (2012)
798-2809.
22]Y. Liu, J. Ding, J. Sun, J. Zhang, J. Bi, K. Liu, F. Kong, H. Xiao, Y. Sun, J. Chen,
Chem. Commun. 52 (2016) 5030-5032.
[
[
9
[
2
[
[
23]C. Wen, Y. Cui, W. Dai, S. Xie, K. Fan, Chem. Commun. 49 (2013) 5195-5197.
24]F. Ma, H.B. Wu, B.Y. Xia, C. Xu, X.W.D. Lou, Angew. Chem. Int. Ed. 54 (2015)
[
1
5395-15399.
[
25]C. Wan, N.A. Knight, B.M. Leonard, Chem. Commun. 49 (2013) 10409-10411.
26]R. Ma, Y. Zhou, Y. Chen, P. Li, Q. Liu, J. Wang, Angew. Chem. Int. Ed. (2015)
[
1
4723-14727.
[27]T. Cotter, B. Frank, W. Zhang, R. Schlögl, A. Trunschke, Chem. Mater. 25 (2013)
28