Enhancing Metal-Support Interactions by Molybdenum Carbide: An Efficient Strategy toward the Chemoselective Hydrogenation of α,β-Unsaturated Aldehydes

Article abstract of DOI:10.1002/chem.201600323

Metal-support interactions are desired to optimize the catalytic turnover on metals. Herein, the enhanced interactions by using a Mo₂C nanowires support were utilized to modify the charge density of an Ir surface, accomplishing the selective hydrogenation of α,β-unsaturated aldehydes on negatively charged Ir^δ- species. The combined experimental and theoretical investigations showed that the Ir^δ- species derive from the higher work function of Ir (vs. Mo₂C) and the consequently electron transfer. In crotonaldehyde hydrogenation, Ir/Mo₂C delivered a crotyl alcohol selectivity as high as 80 %, outperforming those of counterparts (<30 %) on silica. Moreover, such electronic metal-support interactions were also confirmed for Pt and Au, as compared with which, Ir/Mo₂C was highlighted by its higher selectivity as well as the better activity. Additionally, the efficacy for various substrates further verified our Ir/Mo₂C system to be competitive for chemoselective hydrogenation.

Full text of DOI:10.1002/chem.201600323

DOI: 10.1002/chem.201600323
Full Paper
&
Chemoselective Hydrogenation
Enhancing Metal–Support Interactions by Molybdenum Carbide:
An Efficient Strategy toward the Chemoselective Hydrogenation
of a,b-Unsaturated Aldehydes
Sina He,[a] Zheng-Jiang Shao,[b] Yijin Shu,[a] Zhangping Shi,[c] Xiao-Ming Cao,*[b]
Qingsheng Gao,*[a] Peijun Hu,[b] and Yi Tang[c]
Abstract: Metal–support interactions are desired to optimize
the catalytic turnover on metals. Herein, the enhanced inter-
actions by using a Mo2C nanowires support were utilized to
modify the charge density of an Ir surface, accomplishing
the selective hydrogenation of a,b-unsaturated aldehydes
on negatively charged IrdÀ species. The combined experi-
mental and theoretical investigations showed that the IrdÀ
species derive from the higher work function of Ir (vs. Mo2C)
and the consequently electron transfer. In crotonaldehyde
hydrogenation, Ir/Mo2C delivered a crotyl alcohol selectivity
as high as 80%, outperforming those of counterparts
(<30%) on silica. Moreover, such electronic metal–support
interactions were also confirmed for Pt and Au, as compared
with which, Ir/Mo2C was highlighted by its higher selectivity
as well as the better activity. Additionally, the efficacy for
various substrates further verified our Ir/Mo2C system to be
competitive for chemoselective hydrogenation.
Introduction
et al. for Cu/TiC and Au/TiC catalysts in hydrodesulfurization re-
actions, CO2 conversion to methanol, O2 dissociation, and
water gas shift.[12–18]
Metal–support interactions play essential roles in the catalytic
applications of supported nanocatalysts, which affect the size,
morphology, and valence state of active species through mass
transport and interfacial charge redistribution.[1–3] The com-
bined effect of the above-mentioned factors would facilitate
the achievement of high catalytic conversion with satisfied se-
lectivity.[3,4] Due to the possession of noble-metal-like electron-
ic property because of d-band contraction in metal–carbon
alloys,[5,6] transition-metal carbides are recently discovered as
promising supports.[7] They serve as excellent supports for the
dispersion of metal catalysts.[8–11] And more importantly, they
can also modify the reactivity of a supported metal through
electronic perturbations, which has been proven by Rodriguez
Chemoselective hydrogenation of a,b-unsaturated aldehydes
to unsaturated alcohols is an important step toward value-
added chemicals.[19–23] Because saturated aldehydes instead of
the desired unsaturated alcohols are the thermodynamically fa-
vored products,[20,24] selective catalysts are highly demanded to
hydrogenate the carbonyl group from the conjugated ethylen-
ic (C=C) and carbonyl (C=O) groups. Although supported iridi-
um (Ir) with a large d band is considered as one of the efficient
catalysts according to the semi-empirical extended Hückel cal-
culation,[21] the selectivity is unsatisfied in experimental re-
ports.[25–27] Commonly, good activity but poor selectivity is ob-
served in liquid-phase hydrogenation,[25–27] owing to the high-
density unoccupied d orbitals adsorbing/activating both the
C=C and the C=O group. In this regard, electron regulation on
the Ir is required. For example, the excessive electrons in the Ir
species would enhance the repulsive force with the C=C
group,[28] and also promote an electron feedback to the p* or-
bital in the polar C=O group,[21] resulting in the selective hy-
drogenation of the C=O moiety. Such regulation is expected to
be achieved on metal–carbide supports through electronic
metal–support interactions, which drive the charge transfer be-
tween the carbides and the Ir due to their variation in work
functions.[1,29,30]
[a] S. He, Y. Shu, Prof. Q. Gao
Department of Chemistry, Jinan University
Guangzhou 510632 (P. R. China)
E-mail: tqsgao@jnu.edu.cn
[b] Z.-J. Shao, Dr. X.-M. Cao, Prof. P. Hu
Key Laboratory for Advanced Materials
Center for Computational Chemistry and Research Institute of
Industrial Catalysis, East China University of Science & Technology
Shanghai 200237 (P. R. China)
E-mail: xmcao@ecust.edu.cn
[c] Z. Shi, Prof. Y. Tang
Department of Chemistry
Shanghai Key Laboratory of Molecular Catalysis and Innovative Materials
Laboratory of Advanced Materials and
Collaborative Innovation Center of Chemistry for Energy Materials
Fudan University, Shanghai 200433 (P. R. China)
Hydrogenating crotonaldehyde (CRAL) to crotyl alcohol
(CROL) remains the main challenge of chemoselective hydro-
genation,[21,31] which is limited by the low selectivity associated
with the negligible steric hindrance and electronic effects of
a small methyl group at the g-sites. Herein, we develop Mo2C-
Supporting information for this article is available on the WWW under
http://dx.doi.org/10.1002/chem.201600323.
Chem. Eur. J. 2016, 22, 5698 – 5704
5698
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
nanowire-supported Ir (Ir/Mo2C) for the selective hydrogena-
tion of CRAL in an aqueous system. The combined experimen-
tal and theoretical investigations show that the negatively
charged IrdÀ species stems from the higher work function of Ir
(5.22 vs. 5.00 eV of Mo2C) and the consequent electron transfer,
150 nm are well observed in the scanning electron microscopy
(SEM) and transmission electron microscopy (TEM) images (Fig-
ures 1c and d). And the attached energy dispersive spectrum
(EDS, Figure S2 in the Supporting Information) identifies the
successful loading of Ir on Mo2C, whose size is determined as
which promotes the adsorption and activation of the polar C= approximately 2–3 nm by a high-resolution TEM (HRTEM) in-
O group towards the enhanced production of unsaturated al-
cohol. As expected, the selectivity as high as 80% is superior
to that of SiO2-supported Ir (23%), and comparable to the best
of the current Pt-group catalysts.[32–37] Similar metal–support in-
teractions affecting the selectivity are also confirmed for Mo2C-
supported Pt and Au, illustrating their efficient regulation over
reaction pathways. Noticeably, the high selectivity as well as
the good activity for a wide range of substrates verifies our Ir/
Mo2C system as a promising catalyst for the chemoselective
hydrogenation of a,b-unsaturated aldehydes.
vestigation (Figure 1c).
In order to elucidate the interactions between the Ir and the
Mo2C support, hydrogen temperature-programmed reduction
(H2-TPR) and X-ray photoelectron spectroscopy (XPS) analysis
are conducted. As shown in the H2-TPR results (Figure 2a), Ir/
Mo2C presents a promoted reducibility of Ir3+ species at lower
Results and Discussion
Hierarchical nanowires composed of Mo2C nanoparticles
(ꢀ15–20 nm in size, Figure S1 in the Supporting Information)
are used as the catalyst support. They possess a rich nano-
porosity and a large surface (ꢀ40 m2 gÀ1) free from carbon
deposition.[38] After an impregnation procedure and a following
reduction by using 5 vol% H2/Ar at 3008C, a series of Ir/Mo2C
with different Ir loading was harvested (Figure 1a and Table S1
in the Supporting Information). With the increased Ir loading,
the diffraction peaks assigned to the cubic phase Ir (JCPDS NO.
87-0715) emerge in the X-ray diffraction (XRD) patterns (Fig-
ure 1a), along with the characteristic peaks of a-Mo2C (JCPDS
No. 31-0871). Taking 3.7% Ir/Mo2C as the model sample, nano-
wires with a length of several micrometers and a width of 80–
Figure 2. a) H2-TPR and b) XPS results of the Ir/Mo2C and Ir/SiO2 systems.
The dash lines in b) indicate the theoretical values of standard Ir 4f7/2 and
4f5/2. Charge distribution of Ir on c) Mo2C and d) SiO2 surfaces determined by
the Bader charge analysis. The blue and yellow isosurfaces represent positive
and negative charges, respectively, and only the absolute isovalues greater
than 0.008 ebohrÀ3 are shown. The purple, green, gray, red, white, and blue
balls represent Ir, Mo, C, O, H, and Si atoms, respectively.
temperature (2008C) than Ir/SiO2 (2608C). Given the similar size
of Ir on both Mo2C and SiO2 (see Figures 1b and S3 in the Sup-
porting Information), the strong metal–support interactions
between Ir and Mo2C are reasonably suggested. As further vali-
dated by XPS (Figure 2b), the Ir 4f7/2 and 4f5/2 peaks in Ir/Mo2C
are red shifted to 60.3 and 63.2 eV, respectively, in comparison
with those of metallic Ir (Ir 4f7/2 =60.9 eV, Ir 4f5/2 =63.9 eV)[39,40]
on inert SiO2. This observation well indicates the negatively
charged Ir (IrdÀ) on the Mo2C surface.[41]
Generally, the variation in the work functions of metals and
supports usually contributes to the electronic metal–support
interactions in heterogeneous catalysts.[29,30] Owing to the
lower work function of Mo2C (5.00 eV) than that of Ir
(5.22 eV),[42,43] the formation of IrdÀ on Mo2C should be ascribed
Figure 1. a) XRD patterns of Mo2C and Ir/Mo2C with various Ir loading, and
b) typical SEM and c) typical TEM images of 3.7% Ir/Mo2C.
Chem. Eur. J. 2016, 22, 5698 – 5704
www.chemeurj.org
5699
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
to the charge transfer through the interfaces. We further exam-
ined the charge redistribution of Ir on Mo2C and SiO2 supports
by using periodic spin-polarized calculations in the framework
of density functional theory (DFT) implemented within the
Vienna ab initio simulation program (VASP).[44,45] The Ir/Mo2C
and Ir/SiO2 systems were modeled by adding a three-layer
strip of metal Ir on the Mo2C(121) and the SiO2(001) surfaces
(Figure S4 in the Supporting Information), respectively. The
most stable facet of (121) in a-Mo2C is mainly observed in
Mo2C nanowires (Figure 1c), which provides a rational mod-
eled surface for active Ir. It is clearly showed that electronic
transfer occurs at the interfaces between Ir and Mo2C, which is
however negligible in Ir/SiO2. The average charges per Ir atom
achieved from Mo2C is 0.21 e, as shown in the isosurfaces of
charge density (Figure 2c). This well indicates the enhanced
electronic metal–support interactions between Ir and Mo2C
due to the difference of their work functions, in which the
higher Fermi energy level in Mo2C promotes the electron trans-
fer to Ir. The similar electronic interactions on carbide surface
was ever evidenced on TiC-supported Cu and Au by DFT calcu-
lation, which showed the significant electronic perturbations
of metals due to the strong CuÀC and AuÀC interac-
tions.[12,13,17] In comparison, only 0.02 e per Ir atom are ob-
served on SiO2 (Figure 2d), which is attributed to the lower
Fermi energy level of the nanoparticles compared to the
bulk.[46] The enhanced electronic density in Ir would increase
the repulsive four-electron interactions with the C=C moiety,
and favor a d-electron feedback into the p* orbital of the polar
C=O group.[21,28] We anticipate that such electronic metal–sup-
port interactions in Ir/Mo2C would benefit the selective hydro-
genation of a,b-unsaturated aldehydes to unsaturated alco-
hols.
Figure 3. a) Reaction pathways for the hydrogenation of CRAL, and time-de-
pendent profiles over b) 3.7% Ir/Mo2C, c) Mo2C, and (d) 3.5% Ir/SiO2.
Table 1. Hydrogenation results of CRAL over supported Ir catalysts.[a]
Entry Catalysts
t [min] Conv. [%] Select. [%]
Initial rate[c]
rC rC
CROL BUAL
=
O
=
C
1
2
3
4
5
6
7
3.7% Ir/Mo2C
3.5% Ir/SiO2
3.8% Ir/Mo2C(c) 150
150[b] >99
80
23
60
69
75
73
69
2
60
32
17
7
502
77
150
88
51
134 419
175 105
589 316
544 141
327 119
242 153
0.9% Ir/Mo2C
2.4% Ir/Mo2C
5.0% Ir/Mo2C
6.3% Ir/Mo2C
240[b]
83
240[b] >99
210[b] >99
240[b] >99
6
8
In this work, the selective hydrogenation of CRAL to CROL is
investigated, whose reaction pathways are shown in Figure 3a.
The desired CROL can be produced through the direct hydro-
genation of the C=O double bond, namely, a 1,2-addition.[47]
The undesired butanal (BUAL) generates from either the direct
hydrogenation of the C=C double bond undergoing a 3,4-addi-
tion mechanism, or the indirect 1,4-addition to but-1-en-1-ol
(ENOL) followed by a swift tautomerization.[47] The above-de-
scribed partial hydrogenation products, including CROL, BUAL,
and even ENOL, can be further hydrogenated to saturated bu-
tanol (BUOL). As expected, the catalyst comprising 3.7% Ir/
Mo2C presents the prominent performance. In the time courses
of the reaction (Figure 3b), the desired CROL is the main prod-
uct, and the byproducts BUAL and BUOL are limited to a low
level. As CRAL is almost completely converted (>120 min), the
further hydrogenation of BUAL to BUOL is obvious, but that of
CROL is negligible, suggesting the favored hydrogenation of
the C=O double bond. Such performance is confirmed by
three catalytic tests showing good degrees of reproducibility
(Figure S5 in the Supporting Information). Considering the
poor CRAL conversion and the CROL yield over bare Mo2C
under the same condition (Figure 3c), highly dispersed Ir is be-
lieved as the main active species in Ir/Mo2C for CRAL hydroge-
nation. By contrast, Ir/SiO2 with a similar Ir loading (3.5%) dis-
plays the undesired hydrogenation routes to BUAL (Figure 3d).
[a] Reaction condition: catalyst (25 mg), CRAL (2 mmol), H2O (50 mL), tem-
perature (373 K), H2 (2 MPa), stirring rate (600 rpm). [b] Values correspond-
À1
ing to the maximum yield of CROL. [c] In [mmols
g
Ir
À1].
The detailed hydrogenation results over the supported cata-
lysts of Ir are summarized in Table 1. At a reaction time of
150 min, when the maximum yield of CROL is achieved, 3.7%
Ir/Mo2C shows a CROL selectivity as high as 80% with an
almost complete CRAL conversion (entry 1 in Table 1). The
CROL selectivity is obviously higher than that observed for Ir/
SiO2 (23%) with a similar Ir loading (entry 2 in Table 1), and
comparable to the best of the current Pt-group catalysts.[32–37]
This situation is also confirmed by the initial formation rates of
CROL (rC O) and BUAL (rC ). Remarkably, 3.7% Ir/Mo C presents
=
=
C
2
À1
À1
a
higher rC
value of 502 mmols
g
than rC
value
C
=
=
O
Ir
À1
(77 mmols
g
À1), whereas 3.5% Ir/SiO2 only displays a lower
Ir
rC value than rC value. TEM (Figures 1b and S3 in the Sup-
=
O
=
C
porting Information)) and the CO-uptake (Table S2 in the Sup-
porting Information) analyses were conducted to identify the
particle size and the active-site amounts of Ir on Mo2C and
SiO2, in which the difference is negligible. It is reasonable to as-
cribe the obviously improved CROL selectivity on Ir/Mo2C to
the enhanced electronic metal–support interactions. The nega-
Chem. Eur. J. 2016, 22, 5698 – 5704
www.chemeurj.org
5700
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
tively charged IrdÀ on Mo2C, distinguished from the metallic Ir0
on inert SiO2, would prefer the adsorption and activation of
the polar Cd+=OdÀ bond due to the enriched d electrons in Ir,
which is commonly accepted in the chemoselective hydroge-
nation by using Pt-group metals.[21,28,48]
CO-uptake analyses (Figure S8 and Table S2 in the Supporting
Information), respectively. In XPS measurements (Figure 4a),
the peaks of Pt 4f7/2 and Pt 4f5/2 are red shifted to 70.3 and
73.8 eV, respectively, on Mo2C, in comparison with those of
metallic Pt on SiO2 (71.4 and 74.6 eV).[50] This indicates the for-
mation of negatively charged PtdÀ owing to the electron trans-
fer from Mo2C to Pt, which possesses a higher work function
of 5.66 eV than Mo2C.[42,43] In the same way, Au with a higher
work function of 5.10 eV[42] accumulates electrons from Mo2C,
as verified by the obviously red shifted Au 3d7/2 and 3d5/2
peaks in the XPS (Figure 4b).[51] Meanwhile, these metal–sup-
port interactions are consistent with the results of the H2-TPR
analysis (Figure S9 in the Supporting Information). Further-
more, the Bader charge analysis by DFT calculations clearly dis-
plays the different charge density of the metals on Mo2C and
SiO2. As for Pt/Mo2C and Au/Mo2C, the average charges per
metal atom achieved from Mo2C are 0.23 and 0.16 e (Fig-
ures 4c and e), respectively. However, the charges of the
metals in the Pt/SiO2 and Au/SiO2 systems are negligible (Pt:
0.05 e and Au: 0.03 e, Figures 4d and f). These are in good
agreement with the trend suggested by the difference of the
work function between bulk metals and supports, in which the
lower Fermi energy level the metal has, the more electrons are
transferred from Mo2C with the high Fermi energy level.
Furthermore, the selective hydrogenation on Ir/Mo2C is influ-
enced by the nature of the Mo2C and the Ir loading, as a result
of the Ir–Mo2C interactions. When commercial Mo2C is em-
ployed as supports [denoted as Ir/Mo2C(c)], the CRAL conver-
sion (51%) and the CROL selectivity (60%) are lower than
those on Ir/Mo2C under the same condition (entry 3 in Table 1),
which can be ascribed to the small surface of bulky Mo2C (Fig-
ure S6 in the Supporting Information). And the evolving cata-
lytic performance associated with the Ir loading from 0.9 to
6.3% (entries 4–7 in Table 1) suggests the optimized loading
around 3.7%. The slightly low selectivity at high and low Ir
loading should be ascribed to the prohibited metal–support
interactions by excessive Ir and the unselective hydrogenation
on Mo2C, respectively. Additional tests show the satisfied cycle-
ability of Ir/Mo2C (Figure S7 in the Supporting Information),
and its high selectivity and activity in aqueous systems superi-
or to those in organic solvents (Table S3 in the Supporting In-
formation), pointing out a promising candidate for the envi-
ronmentally clean reduction of unsaturated aldehydes.[49]
In Mo2C-supported Pt and Au catalysts, the metal–support
interactions are also observed. The similar preparation was em-
ployed for the metals on Mo2C and SiO2, in which the similari-
ties in the particle size and the active-site amounts of the
metal on the different supports were confirmed by TEM and
Accordingly, the CRAL hydrogenation is affected by the elec-
tronic metal–support interactions in the above-described cata-
lysts. Focusing on the performance at a reaction time of
120 min, the CRAL conversion and the CROL selectivity are
summarized in Figure 5a. The activity of the Mo2C-supported
Figure 4. XPS results of a) Pt and b) Au supported by Mo2C and SiO2. Charge distribution of c,d) Pt and e,f) Au on Mo2C (c and e) and SiO2 (d and f) surfaces
determined by the Bader charge analysis. The blue and yellow isosurfaces are positive and negative charges, respectively, and only the absolute isovalues
greater than 0.008 ebohrÀ3 are shown. The cyan, golden, green, gray, red, white, and blue balls represent Pt, Au, Mo, C, O, H, and Si atoms, respectively.
Chem. Eur. J. 2016, 22, 5698 – 5704
www.chemeurj.org
5701
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
metals outperforms that of their counterparts on SiO2. More
importantly, the varied CROL selectivity for each metal de-
pends on the charge distribution on the interfaces. Pt/Mo2C
with detectable PtdÀ species presents a remarkably higher
CROL selectivity of 66% than Pt/SiO2 (32%), which is consistent
with the previously reported PtdÀ that benefits the C=O hydro-
genation.[28,48] Correspondingly, Au/Mo2C delivers a slightly
higher selectivity of CROL (53%) than Au/SiO2 (49%).[52]
metals without electronic metal–support interactions,[53] as il-
lustrated by the initial performance on SiO2-supported Ir, Pt,
and Au (Figure 5b). Ir and Pt with large unoccupied d orbitals
and thus, a high hydrogenation activity, usually give a low C=
O selectivity, whereas Au catalysts, presenting a good selectivi-
ty, are limited by their low activity. Employing the electronic
metal–support interactions from Mo2C, the C=O selectivity of Ir
and Pt can be remarkably improved (Figure 5b), and both the
activity and the selectivity of Au are slightly increased. Signifi-
cantly, our Ir/Mo2C system successfully fulfils the high selectivi-
ty as well as a good activity among these catalysts, as con-
firmed by the outstanding performance in the initial reaction
(Figure 5b) and after a reaction time of 120 min (Figure 5a).
Our Ir/Mo2C catalyst can be considered in a sense as a univer-
sal catalyst for the selective hydrogenation of a,b-unsaturated
aldehydes. The performance for various substrates is summar-
ized in Table 2, in which the activity and selectivity are dis-
cussed at the beginning of the reaction (10 min) and after the
time required for the maximum yield of products. Beside the
prominent performance for CRAL (entry 1 in Table 2), the selec-
tive routes for 3-methyl-2-butenal and 2-pentenal are also ach-
ieved (entries 2 and 3 in Table 2, respectively). For 3-methyl-2-
butenal, the C=O selectivity is maintained around 82–83%.
And a high activity (conversion of ꢀ80% after only 40 min)
with a C=O selectivity of approximately 55% is observed for 2-
pentenal. Remarkably, Ir/Mo2C shows a highly selective hydro-
genation for cinnamaldehyde and citral (entries 4 and 5 in
Table 2, respectively), and the higher selectivity (ꢀ90%) in
comparison with CRAL should be ascribed to the steric hin-
drance and electronic effects of the substituents at the g-
carbon atom.[21] For cinnamaldehyde, the selectivity of approxi-
mately 90% over Ir/Mo2C is superior to those of previously re-
ported Ir catalysts (Ir/SiO2: 57%, Ir/CNT: 68%, Au–Ir/TiO2: 83%)
and Pt-on-Au nanostructures (<80%),[25,54–56] and comparable
In the hydrogenation of CRAL to CROL, a high activity and
selectivity are usually difficult to achieve at the same time on
Figure 5. Comparison of a) the CRAL conversion and the CROL selectivity
after a reaction time of 120 min and b) the initial reaction rates in the CRAL
hydrogenation on a series catalysts of 3.7% Ir/Mo2C, 3.5% Ir/SiO2, 3.0% Pt/
Mo2C, 3.4% Pt/SiO2, 3.8% Au/Mo2C, and 3.8% Au/SiO2.
Table 2. Hydrogenation of a,b-unsaturated aldehydes over 3.7% Ir/Mo2C.
Entry
1[a]
Substrate
Product
t [min]
Conv. [%]
Select. [%]
10
150[c]
44
>99
77
87
10
180[c]
11
>99
83
82
2[a]
10
40[c]
44
80
55
55
3[a]
10
150[c]
35
>99
88
90
4[b]
10
240[c]
19
95
91
94
5[b]
[a] H2O (50 mL). [b] EtOH (20 mL) and H2O (30 mL). [c] Value corresponding to the maximum yield of unsaturated alcohols.
Chem. Eur. J. 2016, 22, 5698 – 5704
www.chemeurj.org
5702
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
ing at 3008C. The H2-TPR and CO chemisorption measurements
were both conducted on a XianQuan instrument TP 5076.
to metal–organic-framework-confined Pt (91%) and ligand-
capped PtFe (94%).[57,58] And in citral hydrogenation, the C=O
selectivity of 94% outperforms those of Ir (Ir/SiO2: 47–68%),[59]
and other metals (Pt/Ti-doped mesoporous silica): ꢀ35%, Zn-
Pt/mesoporous silica: ꢀ75%, Pt/TiO2-H2: ꢀ80%, Au/CNT
(CNT=carbon nanotube): ꢀ50%, Au/Nb2O5: ꢀ70%, Ag/SiO2
and Ag-In/SiO2: ꢀ80%).[48,60–65]
Catalytic performance measurement: CRAL hydrogenation was
carried out in a 100 mL stainless steel autoclave (Parr 4598), in
which the catalyst (25 mg), CRAL (2 mmol), and H2O (50 mL) were
loaded. The reactor was sealed and purged with H2 to remove the
air for five times, and then the reactor was heated to the desired
temperature. H2 (2 MPa) was purged into the reactor after the de-
sired temperature was reached and the stirrer was started. The
products were analyzed by using a Shizumadu GC-2014C with
a FID detector.
Conclusion
Density functional theory calculations: The periodic spin-polar-
ized calculations in the framework of the density functional theory
implemented within the Vienna ab initio simulation program
(VASP)[44,45] were performed to model the supported metal catalysts
for understanding the charge transfer between the metal and the
supports. The projector-augmented-wave (PAW) pseudopoten-
tials[68] were utilized to describe the core–electron interactions. The
generalized gradient approximation (GGA) exchange-correlation
functional of Perdew–Burke–Ernzerhof (PBE)[69] form was utilized to
describe the exchange–correlation interaction. The cut-off energy
of the plane wave basis set was set to 400 eV.
We have developed Ir/Mo2C catalysts with enhanced metal–
support interactions for the chemoselective hydrogenation of
a,b-unsaturated aldehydes, in which the IrdÀ active species re-
sulting from an electron transfer at the interfaces contributes
to the selective routes. Showing the prevailing influences on
various metals by Mo2C supports, this work opens up new op-
portunities for the design of selective catalysts through inter-
face engineering. It is envisioned that the future work with
a better control over the electron and surface properties of
carbide supports[66,67] would further boost the relevant catalytic
applications.
The most stable hydroxylated cleaved stoichiometric (001) surface
of a-quartz, which is the most stable phase of silica at the reaction
temperature,[70] was modeled by a p(3”2) unit cell of 14.85”
2
9.90 Š with a 12 Š vacuum between the slabs in the z direction.
The slab contains 18-layer SiO2, corresponding to 36 units of SiO2
per slab. The Brillouin zone was sampled with 2”3”1 Monkhorst–
Pack k-point mesh. For Mo2C, the a-phase according to the experi-
mental result was researched. The crystal structure of this phase
has a hexagonal structure with the space group Pbcn. The most
stable (121) surface of this phase was modeled by a p(3”2) unit
Experimental Section
Catalyst preparation: The synthesis of the Mo2C nanowires was
conducted according to our previous report.[38] Ammonium hepta-
molybdate [(NH4)6Mo7O24·4H2O] (2.48 g) was dissolved in distilled
water (40 mL) and aniline (3.34 g) was added to this solution. After-
wards, a 1m aqueous solution of HCl was dropwise added with
magnetic stirring at room temperature until a white precipitate ap-
peared (pH 4–5). After stirring at 508C in an oil bath for approxi-
mately 2–6 h, the precursor of Mo3O10(C6H8N)2·2H2O was received.
Finally, the Mo2C nanowires were harvested after calcining
Mo3O10(C6H8N)2·2H2O at 7758C for 5 h under an Ar flow.
2
cell of 18.29”11.41 Š with a 12 Š vacuum between the slabs in
the z direction. The slab contains 9-layer Mo2C, corresponding to
36 units of Mo2C per slab. The Brillouin zone was sampled with 2”
3”1 Monkhorst–Pack k-point mesh. The metal/SiO2 and metal/
Mo2C systems were modeled by adding a three-layer strip of metal
Ir, Pt, or Au on the SiO2(001) or Mo2C(121) surface, respectively.
During all the optimization processes, the bottom nine layers of
SiO2 and three layers of Mo2C were fixed in the slab, whereas the
upmost layers of the supports and metal strips were relaxed. For
each metal–support system, the five lowest energy structures from
the results of Nose thermostat molecular dynamics simulation (T=
1508C, 1 fs per step, 2000 steps) were selected to be further opti-
mized by conjugate gradient methods. Geometry optimization was
performed until all the remaining forces on each relaxed atom
were lower than 0.02 eVŠÀ1. The structure with the lowest energy
was utilized for the charge population analysis. The charge differ-
ence in the scheme of the Bader charge analysis was constructed
by subtracting the charges of the metal–support systems from the
charges of the metal and the support with each in the same struc-
ture. The isosurfaces of the charge density differences were also
constructed by subtracting the charge densities of the metal–sup-
port systems from the charge densities of the metal and the sup-
porter with each in the same structure.
An impregnation procedure was introduced to prepare the sup-
ported catalysts of Ir and Pt. Typically, the catalyst supports (Mo2C
or SiO2) were impregnated with an aqueous solution containing
the relevant metal ions (i.e., HIrCl4 or H2PtCl6), and then the solu-
tions were stirred for 4 h at 808C. The solids were dried at 508C
overnight, followed by a careful reduction with a stream of 5 vol%
H2/Ar at 3008C for 2 h. Meanwhile, a typical deposition–precipita-
tion procedure was employed to prepare the Au/Mo2C and Au/
SiO2 catalysts. Briefly, Mo2C or SiO2 was dispersed with an aqueous
solution of HAuCl4, and the pH was adjusted to 9.0 by dropwise
addition of a 0.25m aqueous solution of NH3·H2O. After stirring for
6 h and aging for another 2 h, the catalysts were washed with de-
ionized water and then dried at 508C overnight, followed by a re-
duction with 5 vol% H2/Ar at 3008C for 2 h.
Catalyst characterization: XRD analysis was performed on
a Bruker D8 diffractometer by using CuKa radiation (l=1.54056 Š).
SEM and TEM investigations were taken on a ZEISS ULTRA55 and
a JEOL JEM 2100F, respectively. EDS attached on TEM was carried
out on a JEOL JEM 2100F. XPS was processed on a Perkin–Elmer
PHI X-tool, by using C 1s (B.E.=284.6 eV) as a reference. The metal
loading was determined by using inductively coupled plasma-
atomic emission spectroscopy (ICP-AES). The Brunauer–Emmett–
Teller (BET) specific surface areas were determined by adsorption–
desorption measurements of nitrogen at the liquid nitrogen tem-
perature, by using a Micromeritics TriStar 3000 equipment, degass-
Acknowledgements
This work is financially supported by the National Basic Re-
search Program of China (2013CB934101), the National Natural
Science Foundation of China (21373102, 21433002, 21333003,
and 21303051), and the Fundamental Research Funds for the
Chem. Eur. J. 2016, 22, 5698 – 5704
www.chemeurj.org
5703
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Full Paper
[34] G. Kennedy, L. R. Baker, G. A. Somorjai, Angew. Chem. Int. Ed. 2014, 53,
3405; Angew. Chem. 2014, 126, 3473.
[35] B. Li, G. S. Hu, L. Y. Jin, X. Hong, J. Q. Lu, M. F. Luo, J. Ind. Eng. Chem.
2013, 19, 250.
[36] J. Hidalgo-Carrillo, M. Angeles Aramendia, A. Marinas, J. Maria Marinas,
F. Jose Urbano, Appl. Catal. A 2010, 385, 190.
Central Universities (21615402). Q.S.G. is also thankful for the
support from the Guangdong Natural Science Funds
(2015A030306014 and 2014TQ01N036) and the Guangdong
Higher Education Institute (YQ2013022).
[37] K. Taniya, C. H. Yu, S. C. Tsang, Y. Ichihashi, S. Nishiyama, Catal. Commun.
2011, 14, 6.
[38] Q. Gao, C. Zhang, S. Xie, W. Hua, Y. Zhang, N. Ren, H. Xu, Y. Tang, Chem.
Mater. 2009, 21, 5560.
Keywords: carbides · hydrogenation · iridium · metal–support
interactions · unsaturated aldehydes
[39] A. Infantes-Molina, A. Romero-Perez, E. Finocchio, G. Busca, A. Jimenez-
Lopez, E. Rodriguez-Castellon, J. Catal. 2013, 305, 101.
[40] F. HØroguel, G. Siddiqi, M. D. Detwiler, D. Y. Zemlyanov, O. V. Safonova,
C. Coperet, J. Catal. 2015, 321, 81.
[41] X. Yang, S. He, Y. Shu, Z. Shi, Y. Guo, Q. Gao, Y. Tang, RSC Adv. 2015, 5,
89282.
[42] H. B. Michaelson, J. Appl. Phys. 1977, 48, 4729.
[43] M. Sugihara, K. Ozawa, K. Edamoto, S. Otani, Solid State Commun. 2004,
131, 245.
[44] G. Kresse, J. Hafner, Phys. Rev. B 1994, 49, 14251.
[45] G. Kresse, J. Furthmüller, Comput. Mater. Sci. 1996, 6, 15.
[46] M. D. Scanlon, P. Peljo, M. A. Mendez, E. Smirnov, H. H. Girault, Chem.
Sci. 2015, 6, 2705.
[47] X.-M. Cao, R. Burch, C. Hardacre, P. Hu, J. Phys. Chem. C 2011, 115,
19819.
[48] M. Bidaoui, C. Especel, S. Sabour, L. Benatallah, N. Saib-Bouchenafa, S.
Royer, O. Mohammedi, J. Mol. Catal. A 2015, 399, 97.
[49] M. M. Wang, L. He, Y. M. Liu, Y. Cao, H. Y. He, K. N. Fan, Green Chem.
2011, 13, 602.
[50] R. X. Wang, Y. J. Fan, L. Wang, L. N. Wu, S. N. Sun, S. G. Sun, J. Power
Sources 2015, 287, 341.
[51] R. Liu, Y. Yu, K. Yoshida, G. Li, H. Jiang, M. Zhang, S. J. Fujita, M. Arai, J.
Catal. 2010, 269, 191.
[52] Q. Y. Yang, Y. Zhu, L. Tian, S. H. Xie, Y. Pei, H. Li, H. X. Li, M. H. Qiao, K. N.
Fan, Appl. Catal. A 2009, 369, 67.
[53] X. F. Yang, A. Q. Wang, X. D. Wang, T. Zhang, K. L. Han, J. Li, J. Phys.
Chem. C 2009, 113, 20918.
[54] J. Zhao, J. Ni, J. Xu, J. Xu, J. Cen, X. Li, Catal. Commun. 2014, 54, 72.
[55] H. Rojas, G. Diaz, J. J. Martinez, C. Castaneda, A. Gomez-Cortes, J.
Arenas-Alatorre, J. Mol. Catal. A 2012, 363, 122.
[56] K.-Q. Sun, Y.-C. Hong, G.-R. Zhang, B.-Q. Xu, ACS Catal. 2011, 1, 1336.
[57] Z. Y. Guo, C. X. Xiao, R. V. Maligal-Ganesh, L. Zhou, T. W. Goh, X. L. Li, D.
Tesfagaber, A. Thiel, W. Y. Huang, ACS Catal. 2014, 4, 1340.
[58] K. B. Vu, K. V. Bukhryakov, D. H. Anjum, V. O. Rodinonv, ACS Catal. 2015,
5, 2529.
[59] H. Rojas, G. Borda, P. Reyes, J. J. Martínez, J. Valencia, J. L. G. Fierro,
Catal. Today 2008, 133–135, 699.
[60] E. Bailón-García, F. Carrasco-Marin, A. F. Perez-Cadenas, F. J. Maldonado-
Hodar, J. Catal. 2015, 327, 86.
[61] M. Bidaoui, C. Especel, N. Bouchenafa-Saib, D. Duprez, O. Mohammedi,
S. Royer, Appl. Catal. A 2012, 445–446, 14.
[62] J. Khanderi, R. C. Hoffmann, J. Engstler, J. J. Schneider, J. Arras, P. Claus,
G. Cherkashinin, Chem. Eur. J. 2010, 16, 2300.
[1] G. A. Somorjai, J. Y. Park, Angew. Chem. Int. Ed. 2008, 47, 9212; Angew.
Chem. 2008, 120, 9352.
[2] S. Schauermann, N. Nilius, S. Shaikhutdinov, H. J. Freund, Acc. Chem.
Res. 2013, 46, 1673.
[3] Q. Fu, T. Wagner, Surf. Sci. Rep. 2007, 62, 431.
[4] J. L. Shi, Chem. Rev. 2013, 113, 2139.
[5] A. M. Alexander, J. S. Hargreaves, Chem. Soc. Rev. 2010, 39, 4388.
[6] H. H. Hwu, J. G. Chen, Chem. Rev. 2005, 105, 185.
[7] T. G. Kelly, J. G. Chen, Chem. Soc. Rev. 2012, 41, 8021.
[8] N. M. Schweitzer, J. A. Schaidle, O. K. Ezekoye, X. Pan, S. Linic, L. T.
Thompson, J. Am. Chem. Soc. 2011, 133, 2378.
[9] A. Griboval-Constant, J. M. Giraudon, I. Twagishema, G. Leclercq, M. E.
Rivas, J. Alvarez, M. J. PØrez-Zurita, M. R. Goldwasser, J. Mol. Catal. A
2006, 259, 187.
[10] M. Lewandowski, A. Szymanska-Kolasa, P. Da Costa, C. Sayag, Catal.
Today 2007, 119, 31.
[11] D. V. Esposito, S. T. Hunt, A. L. Stottlemyer, K. D. Dobson, B. E. McCand-
less, R. W. Birkmire, J. G. Chen, Angew. Chem. Int. Ed. 2010, 49, 9859;
Angew. Chem. 2010, 122, 10055.
[12] J. A. Rodriguez, P. Liu, F. Vines, F. Illas, Y. Takahashi, K. Nakamura, Angew.
Chem. Int. Ed. 2008, 47, 6685; Angew. Chem. 2008, 120, 6787.
[13] J. A. Rodriguez, P. Liu, Y. Takahashi, K. Nakamura, F. Vines, F. Illas, J. Am.
Chem. Soc. 2009, 131, 8595.
[14] J. A. Rodríguez, L. Feria, T. Jirsak, Y. Takahashi, K. Nakamura, F. Illas, J.
Am. Chem. Soc. 2010, 132, 3177.
[15] L. Feria, J. A. Rodriguez, T. Jirsak, F. Illas, J. Catal. 2011, 279, 352.
[16] A. B. Vidal, L. Feria, J. Evans, Y. Takahashi, P. Liu, K. Nakamura, F. Illas,
J. A. Rodriguez, J. Phys. Chem. Lett. 2012, 3, 2275.
[17] J. A. Rodriguez, J. Evans, L. Feria, A. B. Vidal, P. Liu, K. Nakamura, F. Illas,
J. Catal. 2013, 307, 162.
[18] J. A. Rodriguez, P. J. Ramirez, G. Giacomo Asara, F. Vines, J. Evans, P. Liu,
J. M. Ricart, F. Illas, Angew. Chem. Int. Ed. 2014, 53, 11270; Angew. Chem.
2014, 126, 11452.
[19] J. W. Medlin, ACS Catal. 2011, 1, 1284.
[20] P. Mäki-Arvela, J. Hajek, T. Salmi, D. Y. Murzin, Appl. Catal. A 2005, 292, 1.
[21] P. Gallezot, D. Richard, Catal. Rev. Sci. Eng. 1998, 40, 81.
[22] A. Tlili, J. Schranck, H. Neumann, M. Beller, Chem. Eur. J. 2012, 18,
15935.
[23] H. Rong, Z. Niu, Y. Zhao, H. Cheng, Z. Li, L. Ma, J. Li, S. Wei, Y. Li, Chem.
Eur. J. 2015, 21, 12034.
[24] P. Barbaro, F. Liguori, Heterogenized Homogeneous Catalysts for Fine
Chemicals Production, Springer, London, 2010.
[25] B. F. Machado, H. T. Gomes, P. Serp, P. Kalck, J. L. Faria, ChemCatChem
2010, 2, 190.
[26] P. Reyes, H. Rojas, G. Pecchi, J. L. G. Fierro, J. Mol. Catal. A 2002, 179,
293.
[27] D. V. Sokolskii, N. V. Anisimova, A. K. Zharmagambetova, S. G. Mukha-
medzhanova, L. N. Edygenova, React. Kinet. Catal. Lett. 1987, 33, 399.
[28] C. Y. Hsu, T. C. Chiu, M. H. Shih, W. J. Tsai, W. Y. Cheu, C. H. Liu, J. Phys.
Chem. C 2010, 114, 4502.
[63] H. Rojas, J. J. Martinez, S. Mancipe, G. Borda, P. Reyes, React. Kinet. Mech.
Catal. 2012, 106, 445.
[64] M. Steffan, A. Jakob, P. Claus, H. Lang, Catal. Commun. 2009, 10, 437.
[65] M. Steffan, M. Lucas, A. Brandner, M. Wollny, N. Oldenburg, P. Claus,
Chem. Eng. Technol. 2007, 30, 481.
[66] Q. S. Gao, N. Liu, S. N. Wang, Y. Tang, Nanoscale 2014, 6, 14106.
[67] J. A. Rodriguez, P. Liu, D. J. Stacchiola, S. D. Senanayake, M. G. White,
J. G. Chen, ACS Catal. 2015, 5, 6696.
[29] C. G. Vayenas, S. Bebelis, S. Ladas, Nature 1990, 343, 625.
[30] Y. Bai, W. H. Zhang, Z. H. Zhang, J. Zhou, X. J. Wang, C. M. Wang, W. X.
Huang, J. Jiang, Y. J. Xiong, J. Am. Chem. Soc. 2014, 136, 14650.
[31] T. Marinelli, S. Nabuurs, V. Ponec, J. Catal. 1995, 151, 431.
[32] C. Li, Y. Chen, S. Zhang, S. Xu, J. Zhou, F. Wang, M. Wei, D. G. Evans, X.
Duan, Chem. Mater. 2013, 25, 3888.
[68] G. Kresse, D. Joubert, Phys. Rev. B 1999, 59, 1758.
[69] J. P. Perdew, K. Burke, M. Ernzerhof, Phys. Rev. Lett. 1996, 77, 3865.
[70] T. P. M. Goumans, A. Wander, W. A. Brown, C. R. A. Catlow, Phys. Chem.
Chem. Phys. 2007, 9, 2146.
[33] J. Shi, M. Zhang, W. Du, W. Ning, Z. Hou, Catal. Sci. Technol. 2015, 5,
3108.
Received: January 23, 2016
Published online on March 2, 2016
Chem. Eur. J. 2016, 22, 5698 – 5704
www.chemeurj.org
5704
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim