Oxygen-Deficient Tungsten Oxide as Versatile and Efficient Hydrogenation Catalyst

Article abstract of DOI:10.1021/acscatal.5b01522

Heterogeneous hydrogenation is one of the most important industrial operations, and reduced metals (mostly noble metals and a few inexpensive metals) generally serve as the catalyst to activate molecular H₂. Herein we report oxygen-deficient tungsten oxide, such as WO_2.72, is a versatile and efficient catalyst for the hydrogenation of linear olefins, cyclic olefins, and aryl nitro groups, with obvious advantages compared with non-noble metal nickel catalyst from the aspect of activity and selectivity. Density functional theory calculations prove the oxygen-deficient surface activates H₂ very easily in both kinetics and thermodynamics. Testing on several oxygen-deficient tungsten oxides shows a linear dependence between the hydrogenation activity and oxygen vacancy concentration. Tungsten is earth-abundant, and WO_2.72 can be synthesized in large scale using a low-cost procedure, which provides an ideal catalyst for industrial application. Because oxygen vacancy is a common characteristic of many metal oxides, the findings in this work may be extended to other metal oxides and thus provide the possibility for exploring a new type of hydrogenation catalyst.

Full text of DOI:10.1021/acscatal.5b01522

Letter
pubs.acs.org/acscatalysis
Oxygen-Deficient Tungsten Oxide as Versatile and Efficient
Hydrogenation Catalyst
∥
∥
Jiajia Song, Zhen-Feng Huang, Lun Pan, Ji-Jun Zou,* Xiangwen Zhang, and Li Wang
Key Laboratory for Green Chemical Technology of the Ministry of Education, School of Chemical Engineering and Technology,
Tianjin University; Collaborative Innovative Center of Chemical Science and Engineering (Tianjin), Tianjin 300072, China
*
S Supporting Information
ABSTRACT: Heterogeneous hydrogenation is one of the most important
industrial operations, and reduced metals (mostly noble metals and a few
inexpensive metals) generally serve as the catalyst to activate molecular H2.
Herein we report oxygen-deficient tungsten oxide, such as WO2.72, is a versatile
and efficient catalyst for the hydrogenation of linear olefins, cyclic olefins, and
aryl nitro groups, with obvious advantages compared with non-noble metal
nickel catalyst from the aspect of activity and selectivity. Density functional
theory calculations prove the oxygen-deficient surface activates H very easily in
2
both kinetics and thermodynamics. Testing on several oxygen-deficient
tungsten oxides shows a linear dependence between the hydrogenation activity
and oxygen vacancy concentration. Tungsten is earth-abundant, and WO2.72
can be synthesized in large scale using a low-cost procedure, which provides an
ideal catalyst for industrial application. Because oxygen vacancy is a common characteristic of many metal oxides, the findings in
this work may be extended to other metal oxides and thus provide the possibility for exploring a new type of hydrogenation
catalyst.
KEYWORDS: heterogeneous catalysis, hydrogenation, oxygen vacancy, tungsten oxide, molecular H activatation
2
eterogeneous catalytic hydrogenation is one of the most
valuable synthetic transformations in chemical industry,
have stable crystal phase, not just surface reduced WO , which
3
inspires interesting properties for new applications.
18,19,21
H
including in petroleum refinery and processing, as well as in the
Specifically, WO2.72 is a well-studied phase consisting of a
1
−4
manufacture of fine and bulk chemicals.
In particular,
distorted ReO unit where cornersharing distorted and tilt WO
octahedra are connected in the a-, b-, and c-direction.
3
6
1
6,19
hydrogenation of olefins to alkanes is a crucial step for the
upgrading of petroleum streams; meanwhile, transforming
substituted nitroarenes to anilines is of great importance
because anilines are among the most important intermediates
Oxygen vacancies in metal oxide serving as shallow donors can
improve the electrical conductivity and donor density and thus
enhance the surface species (such as H
and CO
adsorption, suggesting great potential in H activation and
2
)
2
2
5
−9
16
necessary for dyestuffs and pharmaceuticals.
Molecular
hydrogen activation/dissociation is the first and often rate-
subsequent hydrogenation. Herein, we report oxygen-deficient
tungsten oxide, especially WO2.72, is a versatile and efficient
catalyst for the saturation of olefins and selective transform of
nitroarenes to anilines. Our first-principles computations and
experiments unambiguously show the unique hydrogenation
activity is governed by oxygen vacancies that can easily activate
10−13
limiting step for the hydrogenation reaction.
Heteroge-
neous hydrogenation catalysts often come from noble metals or
their alloys because they are effective in activating H at modest
2
2
,3
temperature and pressure, but the high-cost and limited
availability stimulates numerous efforts in seeking a cheap
7
−9,14
molecular H . Also these earth-abundant oxides are synthesized
catalyst.
Nickel-based catalysts are cheap and widely used;
2
using a low-cost method and thus specifically suitable for large-
scale application.
First, using density functional theory (DFT) calculations, we
however, the hydrogenation activity needs improvement, and
3
caution is necessary when handling this flammable metal. In
addition, Pt, Pd, and Ni are not so selective in the
5
,15
demonstrated that the oxygen-deficient surface of tungsten
hydrogenation of substituted nitroarenes.
Therefore, there
oxide can activate H very easily in kinetics and thermody-
has been an ongoing effort to find an earth-abundant, low-cost,
highly active, highly selective, and stable-in-air hydrogenation
catalyst.
2
namics. We computed the adsorption, activation, and
dissociation of molecular H2 on three different surfaces:
stoichiometric WO surface, WO surface with oxygen vacancy
Tungsten oxides have attracted great attention for organic
3
3
16−21
synthesis, photocatalysis, and electrochemistry.
A unique
advantage of tungsten oxides is the nonstoichiometric property
as the lattice can withstand a considerable amount of oxygen
Received: April 16, 2015
Revised: August 28, 2015
1
6
vacancies. Moreover, many oxygen-deficient oxides (WOx<3
)
©
XXXX American Chemical Society
6594
DOI: 10.1021/acscatal.5b01522
ACS Catal. 2015, 5, 6594−6599

ACS Catalysis
Letter
Figure 1. (a) Formation energy of WO −V surface under different O chemical potential (T = 0 K). (Inset: Top view of WO surface with oxygen
3
O
3
vacancy. The dotted vertical lines indicate the allowed range of O chemical potential.) (b) Differential charge density of H adsorption. (The yellow
2
3
and blue colors represent charge depletion and accumulation, respectively, with iso-surface value of 0.002 eů . Number of electrons (N) represents
the electron transferred from H to catalyst, and plus (+) represents electron transferring from surface to H while minus (−) represents the reverse).
2
2
(
c) Free energies of the initial (IS), transition (TS), and final states (FS) for the dissociation of H .
2
(
(
WO −V ), and WO surface abundant with oxygen vacancy
bond cleavage become slightly lower (0.428 and 1.294 eV,
3
O
2.72
Figure S1 and Figure 1). The WO surface is built from
respectively), indicating a tendency for easier H activation.
3
2
cleaving the (001) surface on the O plane from a γ-monoclinic
Notably, the presence of oxygen vacancy changes the manner of
WO supercell with 24 W atoms and 72 O atoms (Figure S1a,
H dissociation, with one dissociated H atom binding with a
2
3
b), with the stoichiometric composition preserved by removing
surface W atom and the other with a bridge O atom. Actually,
22,23
one-half of the O atoms on the top and bottom O plane.
WO −V is created by removing terminal O atoms from
semiconductor-to-metal transition of WO has been reported
3−x
2
6−29
recently,
suggesting H may be activated on them in a
2
3
O
23,24
stoichiometric WO (Figure S1c).
The formation energy of
similar way as the metal catalyst. On WO2.72(001) surface H is
3
2
−
0.44 eV (T = 0 K) under oxygen-poor conditions indicates a
activated in the same way, but the adsorption of H is very easy
2
modest propensity to form oxygen vacancies (Figure 1a and
Figure S2). However, VO will be easy to form under
hydrogenation reaction conditions because the oxygen chemical
potential significantly decreases at high temperature and
(with adsorption free energy of −0.346 eV). Importantly, the
WO (001) surface has a strong interaction with H by
2.72
2
transferring 0.419 e¯ to H , and the H−H bond is elongated to
2
1.392 Å. As a result, the energy barrier for H−H bond
25
extremely low oxygen partial pressure, which is verified by
dissociation is only 0.902 eV with considerable exothermic
the experiment below. The WO2.72 surface is built from cleaving
energy of 0.683 eV, indicating the H activation is favorable
2
the (001) surface on the O plane from a γ-monoclinic WO2
from the aspect of both kinetics and thermodynamics. This
result suggests that the presence of oxygen vacancies helps to
.72
1
6
supercell with 22 W atoms and 59 O atoms (Figure S1d, e),
4+
5+
6+
wherein W , W , and W species coexist.
activate molecular H , and therefore, tuning the oxygen
2
As shown in Figure 1b,c, on the stoichiometric WO (001)
vacancies may make inert oxides become an active hydro-
genation catalyst.
3
surface, the adsorption of H is unfavorable (with adsorption
2
free energy of 0.436 eV), and H is almost unactivated (with
We then tested the activity of stoichiometric WO for
2
3
only 0.030 e¯ transferred from H to WO (001) and H−H
hydrogenation of 1-hexene in a stirred batch reactor. WO
2
3
3
2
−1
bond elongated from 0.749 to 0.759 Å). Subsequently, the
breaking of the H−H bond is kinetically unfeasible (with high
barrier of 1.643 eV), with two dissociated H atoms binding with
terminal and bridge O atoms, respectively. On WO −V (001),
(BET surface area of 169 m ·g ) was prepared by alcoholysis
of WCl in ethanol and calcining at 420 °C for 30 min in
6
18,19
air.
Its monoclinic phase (JCPDS No. 43-1035) and urchin
mophology are evidenced by the XRD pattern, SEM, and high-
resolution TEM images (Figure 2a,e,f). It is surprising that this
3
O
the H adsorption free energy and energy barrier for H−H
2
6
595
DOI: 10.1021/acscatal.5b01522
ACS Catal. 2015, 5, 6594−6599

ACS Catalysis
Letter
Figure 2. (a) XRD patterns, (b) diffuse reflectance spectra, (c, e) SEM
images, (d, f) TEM images and high-resolution TEM image (inset) of
WO2.72 and WO3.
Figure 3. (a−d) Fitted W 4f XPS spectra of WO , WO -R60, WO -
3
3
3
R350, and WO2 , respectively. (e) Kinetic data of 1-hexene
.72
oxide shows considerable activity (Figure S3), although the
above DFT calculation suggests the stoichiometric surface can
hydrogenation at 40 °C. (f) Linear correlation between initial reaction
rate and oxygen vacancy concentration (surface O/W ratio). Reaction
conditions: 20 g of 1-hexene, 0.8 g of catalyst, 180 g of cyclohexane
solvent, 30 bar of hydrogen.
hardly activate H . It is noted that, however, after the reaction
2
the yellowish WO becomes intense greenish and shows optical
3
absorption in the 480−700 nm region (Figure 2b), which is
16,20
characteristic of reduced WO .
The absence of a hydroxyl
3
(
O−H) peak or broader shoulder around 532 eV in the XPS
ions (Figure 3a−c), and the surface O/W ratio decreases to
spectra (Figure S4) excludes the formation of tungsten bronze
2.87 and 2.95 (Table S1), respectively. WO (BET surface
2
.72
2
4,30
2
−1
(
H WO ).
Although the bulk crystal strucutre and
area of 178 m ·g ) was prepared by alcoholysis of WCl in
x
3
6
mophology of used WO is not changed appreciably (Figure
ethanol. XRD and high-resolution TEM image (Figure 2a,d)
confirm the monoclinic phase (JCPDS No. 36-0101), and the
SEM image (Figure 2c) shows urchin-like morphology similar
to WO , WO -R60, and WO -R350. The strong absorption tail
3
2
a and Figure S5), the W 4f XPS spectra (Figure 3a and Figure
6
+
S4b) shows that some surface W ions have been reduced to
W
5+
4+
and W , and the surface atomic O/W ratio is dramatically
3
3
3
lower than the stoichiometric ratio (Table S1), so the surface of
in the UV−vis diffuse reflectance spectrum (Figure 2b) and
blue emission band in photoluminescence spectrum (PL,
Figure S7) undoubtedly confirmed the presence of many
oxygen vacancies, in agreement with reported characteristic of
used WO becomes oxygen-deficient, via the following process
3
of 2W6
+/5+
+ O + H → 2W
2−
5+/4+
+ V + H O, wherein O is
2−
2
O 2
lattice oxygen and V is oxygen vacancy. In fact, temperature-
O
19,29
programmed reduction shows the WO surface can be reduced
WO2.72
.
Moreover, its blue color (Figure 2b) originated
3
6
+/5+
by H even at a temperature as low as 40 °C (Figure S6), in
from the intervalence charge-transfer transition between W
2
and W5
+/4+
is quite stable in air and can remain unchanged for
good agreement with the DFT calculation shown in Figure 1a.
So we inferred that it is the in situ formed oxygen-deficient
surface under hydrogenation conditions, instead of the
stoichiometric WO3 surface, that accounts for the hydro-
genation activity of WO3.
several months. Its surface O/W ratio determined by XPS is
2.74 (Figure 3d and Table S1), very close to the bulk
compostion.
Using the above-mentioned samples, we illustrated the
dependence of hydrogenation activity on the concentration of
oxygen vacancies. The hydrogenation of 1-hexene was
condcuted at very low temperature of 40 °C, and the kinetic
profiles were fitted by the first-order decay model to calculate
the initial reaction rate (Figure 3e). In this case, the in situ
reduction of the catalyst surface is avoided, and so the inherent
catalytic activity can be reasonably illustrated. The hydro-
We also prepared three tungsten oxides with a different
amount of oxygen vacancies. WO was treated using H flow at
3
2
2
3
50 °C (denoted as WO -R350, BET surface area of 171 m ·
3
1
−
g ) and 60 °C (denoted as WO -R60, BET surface area of 173
3
2
−1
m ·g ) for 1 h, respectively. Their bulk crystal structure is not
affected as evidenced by XRD (Figure 2a), but the surface 4i+s
partly reduced as indicated by the formation of W5+ and W
6
596
DOI: 10.1021/acscatal.5b01522
ACS Catal. 2015, 5, 6594−6599

ACS Catalysis
Letter
genation over WO is negligible under this condition. With the
the yield of n-hexane about >90% when prolonging the reaction
time or increasing the reaction temperature (Figure 4b).
Furthermore, WO2.72 performs stably in a recycling experiment,
with activity loss <10% in five runs (Figure 4c), probably
because of the cracked morphology (Figure S8) and decreased
3
presence of oxygen vacancies, WO -R and WO
show
3
2.72
obvious activity, and the more oxygen vacancies, the higher
the activity. More importantly, the initial reaction rate is linearly
dependent on the surface O/W ratio with a very good
correlation coefficient of 0.997 (Figure 3f), the lower the ratio,
the quicker the reaction proceeds. This result clearly shows the
hydrogenation activity is governed by the unique oxygen-
deficient surface.
In particular, WO2.72 is very active in the hydrogenation of 1-
hexene. The reaction occurs smoothly with 2-hexene as the
intermediate (Figure 4a). After a 3 h reaction at 100 °C, 40% n-
hexane is produced. The hydrogenation can be finished with
2
−1
BET surface area (160 m ·g ). However, the crystal structure
and surface chemical state of used WO2.72 do not change
(
Figure 2a,b, Table S1 and Figure S9−S10) even though the
reaction temperature reaches 150 °C.
Finally, we showed the versatility of WO2.72 as hydrogenation
catalyst using three types of typical reactants including linear
alkenes, cyclic alkenes and nitroarenes (Table 1). Except the
aromatic rings, all the CC bonds and -NO group are
2
quantitatively converted to C−C bonds and -NH group,
2
respectively. We compared WO2 with the commonly used
.72
non-noble nickel catalyst. WO2.72 still shows comparable or
higher conversion than commercial amorphous nickel (SRNA-
2
−1
4
, BET surface area of 357 m ·g ) that is believed to be more
3
1−33
active than commonly used Raney nickel.
Moreover, the
turnover number (TON, based on metal atom) of WO2.72 is
significantly higher than that of nickel catalyst. In particular,
WO2.72 gives 100% selectivity for substituted anilines by
selectively reducing the NO group to NH and keeping
2
2
other functional groups such as chloride, formyl, and carboxyl
groups unaffected, whereas nickel leads to much lower
selectivity because the substituted groups or aromatic rings
are hydrogenated. Because high temperature may affect the
selectivity of nickel catalyst in nitroarenes hydrogenation, we
also conducted the reaction at lower conversion level by
lowering the reaction temperature (Table 1). It is found that
WO2.72 still exhibits higher TON and selectivity than the nickel
catalyst, although the nickel catalyst exhibits minimally
improved selectivity under this condition.
The perfect selectivity of WO2.72 in hydrogenation of
nitroarenes is very interesting. It has been reported that Ni,
Pt, and Pd catalysts do not show good selectivity in
hydrogenation of functional groups such as C−X, CO, and
5
,15
−
NO .
WO
may capture the oxygen atom of −NO
34−37
2
2.72
2
preferentially to fill the oxygen vacancy,
and thus, only the
nitro group is hydrogenated to −NH . DFT calculation on
2
adsorption of p-nitrochlorobenzene further confirms this
speculation. Initially, p-nitrochlorobenzene is placed parallel
to the WO2.72 surface, but it automatically transfers into
perpendicular adsorption after optimization, indicating that
perpendicular adsorption is much more favored than parallel
adsorption. In the perpendicular adsorption, the nitro group is
vertically adsorbed onto the WO2.72(001) surface with two O
atoms taking the place of one Vo site and interacting with two
neighborhood W atoms (Figure 5a). In contrast, nitroarenes
are parallel absorbed on metal surface such as Ni and Pd
3
8,39
(
Figure S11).
The parallel absorption typically leads to
40
equal possibility for hydrogenation of all functional groups,
whereas the vertical absorption prevents other substituted
groups being activated and hydrogenated. Moreover, WO2.72
shows very stable conversion and selectivity in the recycling
experiment (Figure 5b), indicating the occupied oxygen
vacancies can be fully recovered in the catalytic circle.
In conclusion, we reported oxygen-deficient tungsten oxide is
a very active catalyst for hydrogenation of linear olefins, cyclic
olefins, and aryl nitro groups. The hydrogenation rate is linearly
dependent on the oxygen vacancy concentration because the
Figure 4. (a, c) Time-dependent concentration plots and yield in five
recycling of 1-hexene hydrogenation. (b) Hydrogenation yield in a
long-term test at different temperature. Reaction conditions: 20 g of 1-
hexene, 0.8 g of WO2.72, 180 g of cyclohexane solvent, 30 bar of
hydrogen, 100−150 °C. The recycling in (c) was conducted under the
conditions of (a).
oxygen-deficient surface activates molecular H easily from the
2
aspects of kinetics and thermodynamics. Specifically, WO2.72
6
597
DOI: 10.1021/acscatal.5b01522
ACS Catal. 2015, 5, 6594−6599

ACS Catalysis
Letter
Table 1. Activity of WO2.72 and Amorphous Nickel (SRNA-4) in the Hydrogenation of Linear Alkenes, Cyclic Alkenes, and
Nitroarenes
Reaction conditions: 20 g of substrate, 0.8 g of catalyst, 180 g of ethanol (for nitroarenes) or cyclohexane solvent (for linear and cyclic alkenes), 30
bar of hydrogen, 150 °C/15 h for nitroarenes, 100 °C/3 h + 120 °C/3 h + 150 °C/3 h for linear alkenes, 100 °C/3 h + 120 °C/3 h + 150 °C/3 h +
a
1
70 °C/3 h for cyclic alkenes. Represents low conversion reaction: 120 °C/5 h for nitroarenes, 100 °C/3 h for linear and cyclic alkenes.
which provides an earth-abundant, easily synthesized, and low-
cost catalyst for industrial application. This work emphasizes
the role of oxygen vacancy in H activation and suggests many
2
cheap metal oxides may be made as hydrogenation catalyst
through defect engineering.
ASSOCIATED CONTENT
Supporting Information
The Supporting Information is available free of charge on the
ACS Publications website at DOI: 10.1021/acscatal.5b01522.
■
*
S
Figure 5. (a) Optimized adsorption structre of p-nitrochlorobenzene
over WO2.72(001) surface (T = 0 K). (b) Performance of WO2.72 in
five recycling of p-nitrochlorobenzene hydrogenation. Reaction
conditions: 20 g of p-nitrochlorobenzene, 0.8 g of catalyst, 180 g of
ethanol, 30 bar of hydrogen, 150 °C/15 h.
Detailed experimental and computation procedures,
results of characterizations (PDF)
AUTHOR INFORMATION
Corresponding Author
*E-mail: jj_zou@tju.edu.cn (J.-J. Zou). Fax: +86-22-27892340.
Tel: +86-22-27892340.
■
shows a higher TON than the nickel catalyst and gives 100%
selectivity in the transformation of nitroarenes to anilines,
6
598
DOI: 10.1021/acscatal.5b01522
ACS Catal. 2015, 5, 6594−6599

ACS Catalysis
Letter
Author Contributions
These authors contributed equally (J.S. and Z.-F.H.).
(27) Di Valentin, C.; Pacchioni, G. Acc. Chem. Res. 2014, 47, 3233−
∥
3
241.
(
28) Le, H. M.; Vu, N. H.; Phan, B.-T. Comput. Mater. Sci. 2014, 90,
71−176.
29) Wu, R.; Zhang, J.; Shi, Y.; Liu, D.; Zhang, B. J. Am. Chem. Soc.
015, 137, 6983−6986.
(30) Bathe, S. R.; Patil, P. S. Solid State Ionics 2008, 179, 314−323.
31) Pei, Y.; Zhou, G.; Luan, N.; Zong, B.; Qiao, M.; Tao, F. Chem.
Soc. Rev. 2012, 41, 8140−8162.
32) Meng, X.; Mu, X.; Zong, B.; Min, E.; Zhu, Z.; Fu, S.; Luo, Y.
Catal. Today 2003, 79−80, 21−27.
33) Zou, J.-J.; Zhang, X.; Kong, J.; Wang, L. Fuel 2008, 87, 3655−
3659.
Notes
1
(
2
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
(
The authors appreciate the support from the National Natural
Science Foundation of China (21222607, U1462119,
(
2
1476168), the Tianjin Municipal Natural Science Foundation
(
15JCZDJC37300), and the Program for New Century
(
Excellent Talents in University (NCET-09-0594).
(34) Su, Y.; Lang, J.; Li, L.; Guan, K.; Du, C.; Peng, L.; Han, D.;
Wang, X. J. Am. Chem. Soc. 2013, 135, 11433−11436.
REFERENCES
■
(
35) Aditya, T.; Pal, A.; Pal, T. Chem. Commun. 2015, 51, 9410−
(
1) Kyriakou, G.; Boucher, M. B.; Jewell, A. D.; Lewis, E. A.; Lawton,
T. J.; Baber, A. E.; Tierney, H. L.; Flytzani-Stephanopoulos, M.; Sykes,
E. C. Science 2012, 335, 1209−1212.
2) Studt, F.; Abild-Pedersen, F.; Bligaard, T.; Sorensen, R. Z.;
Christensen, C. H.; Norskov, J. K. Science 2008, 320, 1320−1322.
3) Alexander, A. M.; Hargreaves, J. S. Chem. Soc. Rev. 2010, 39,
388−4401.
4) Yang, Z.; Guo, S.; Pan, X.; Wang, J.; Bao, X. Energy Environ. Sci.
9
431.
36) Wu, Q. P.; van de Krol, R. J. Am. Chem. Soc. 2012, 134, 9369−
375.
37) Schaub, R.; Wahlstro
Stensgaard, I.; Besenbacher, F. Science 2003, 299, 377−379.
38) Lyu, J.; Wang, J.; Lu, C.; Ma, L.; Zhang, Q.; He, X.; Li, X. J. Phys.
Chem. C 2014, 118, 2594−2601.
39) Mahata, A.; Rai, R. K.; Choudhuri, I.; Singh, S. K.; Pathak, B.
Phys. Chem. Chem. Phys. 2014, 16, 26365−26374.
40) Blaser, H.-U.; Steiner, H.; Studer, M. ChemCatChem 2009, 1,
10−221.
(
9
(
(
̈
m, E.; Rønnau, A.; Lægsgaard, E.;
(
4
(
2
(
(
5
(
(
011, 4, 4500−4503.
5) Corma, A.; Serna, P. Science 2006, 313, 332−334.
(
2
6) Dai, Y.; Liu, S.; Zheng, N. J. Am. Chem. Soc. 2014, 136, 5583−
586.
(
7) Vile, G.; Bridier, B.; Wichert, J.; Perez-Ramirez, J. Angew. Chem.,
Int. Ed. 2012, 51, 8620−8623.
8) Jagadeesh, R. V.; Surkus, A.-E.; Junge, H.; Pohl, M.-M.; Radnik, J.;
Rabeah, J.; Huan, H.; Schunemann, V.; Bruckner, A.; Beller, M. Science
013, 342, 1073−1076.
9) Westerhaus, F. A.; Jagadeesh, R. V.; Wienhofer, G.; Pohl, M. M.;
(
̈
̈
2
(
Radnik, J.; Surkus, A. E.; Rabeah, J.; Junge, K.; Junge, H.; Nielsen, M.;
Bruckner, A.; Beller, M. Nat. Chem. 2013, 5, 537−543.
(
10) Chopra, I. S.; Chaudhuri, S.; Veyan, J. F.; Chabal, Y. J. Nat.
Mater. 2011, 10, 884−889.
11) Goodman, D. W.; Kelley, R. D.; Madey, T. E.; Yates, J. T., Jr J.
Catal. 1980, 63, 226−234.
12) Rodriguez, J. A.; Hanson, J. C.; Frenkel, A. I.; Kim, J. Y.; Perez,
M. J. Am. Chem. Soc. 2002, 124, 346−354.
13) Andersson, M. P.; Abild-Pedersen, F.; Remediakis, I. N.;
(
(
(
Bligaard, T.; Jones, G.; Engbæk, J.; Lytken, O.; Horch, S.; Nielsen, J.
H.; Sehested, J.; Rostrup-Nielsen, J. R.; Nørskov, J. K.; Chorkendorff,
I. J. Catal. 2008, 255, 6−19.
(14) Bullock, R. M. Science 2013, 342, 1054−1055.
(15) Upadhya, T. T.; Katdare, S. P.; Sabde, D. P.; Ramaswamy, V.;
Sudalai, A. Chem. Commun. 1997, 1119−1120.
16) Huang, Z.-F.; Song, J.; Pan, L.; Zhang, X.; Wang, L.; Zou, J.-J.
Adv. Mater. 2015, 27, 5309−5327.
17) Gosselink, R. W.; Stellwagen, D. R.; Bitter, J. H. Angew. Chem.,
Int. Ed. 2013, 52, 5089−5092.
18) Huang, Z.-F.; Song, J.; Pan, L.; Lv, F.; Wang, Q.; Zou, J.-J.;
Zhang, X.; Wang, L. Chem. Commun. 2014, 50, 10959−10962.
19) Xi, G.; Ouyang, S.; Li, P.; Ye, J.; Ma, Q.; Su, N.; Bai, H.; Wang,
C. Angew. Chem., Int. Ed. 2012, 51, 2395−2399.
20) Wang, G.; Ling, Y.; Wang, H.; Yang, X.; Wang, C.; Zhang, J. Z.;
Li, Y. Energy Environ. Sci. 2012, 5, 6180−6187.
21) Lu, Y.; Jiang, Y.; Gao, X.; Wang, X.; Chen, W. J. Am. Chem. Soc.
(
(
(
(
(
(
2
(
(
014, 136, 11687−11697.
22) Cheng, Z.; Lo, C. S. ACS Catal. 2015, 5, 59−72.
23) Di Valentin, C.; Wang, F.; Pacchioni, G. Top. Catal. 2013, 56,
1
(
404−1419.
24) Li, W.; Da, P.; Zhang, Y.; Wang, Y.; Lin, X.; Gong, X.; Zheng, G.
ACS Nano 2014, 8, 11770−11777.
25) Reuter, K.; Scheffler, M. Phys. Rev. B: Condens. Matter Mater.
Phys. 2001, 65, 03540601−03540611.
26) Wang, F.; Di Valentin, C.; Pacchioni, G. Phys. Rev. B: Condens.
Matter Mater. Phys. 2011, 84, 0731031−0731035.
(
(
6
599
DOI: 10.1021/acscatal.5b01522
ACS Catal. 2015, 5, 6594−6599