This is the pre-peer reviewed version of the following article: Angew. Chem.
Angewandte
Communications
Int. Ed. 2017, 56, 10755 –10760, which has been published in final form at
Chemie
https://onlinelibrary.wiley.com/doi/full/10.1002/anie.201704999. This article
may be used for non-commercial purposes in accordance with Wiley Terms and
International Edition: DOI: 10.1002/anie.201704999
Conditions for Catalysis
Single-Ensemble Use of Self-Archived Versions.
German Edition:

DOI: 10.1002/ange.201704999

Semihydrogenation of Acetylene on Indium Oxide: Proposed
Single-Ensemble Catalysis
Davide Albani+, MarÅal Capdevila-Cortada+, Gianvito Vilÿ, Sharon Mitchell, Oliver Martin,
NÇria LÛpez,* and Javier Pÿrez-RamÌrez*
Abstract: Indium oxide catalyzes acetylene hydrogenation
with high selectivity to ethylene (> 85 %); even with a large
excess of the alkene. In situ characterization reveals the
formation of oxygen vacancies under reaction conditions,
while an in depth theoretical analysis links the surface
reduction with the creation of well-defined vacancies and
surrounding In3O5 ensembles, which are considered responsible for this outstanding catalytic function. This behavior,
which differs from that of other common reducible oxides,
originates from the presence of four crystallographically
inequivalent oxygen sites in the indium oxide surface. These
resulting ensembles are 1) stable against deactivation,
2) homogeneously and densely distributed, and 3) spatially
isolated and confined against transport; thereby broadening
the scope of oxides in hydrogenation catalysis.

Heterogeneously

catalyzed hydrogenations are widely
applied in the production of numerous chemicals (for
example, commodity, fine, specialty, and pharmaceutical
chemicals). Intensive efforts have been devoted to reducing
the ecological footprint of these processes, breaking away
from the reliance on scarce noble metals (Pd, Pt, Ru) and
avoiding modification with toxic selectivity enhancers (for
example, Pb, Bi, and V).[1] Two directions have shown
significant promise: 1) targeting the improved design of
noble-metal-based catalysts by reducing the size of the
metal ensemble through development of ligand-modified
systems, bimetallic compounds, and single-atoms heterogeneous catalysts,[1, 2] and 2) exploiting the untapped potential of
metal oxides.[3] In this context, cerium oxide and supported
iron oxide have emerged as highly selective catalysts for the
semihydrogenation of acetylene, a reaction integral to the
purification of olefin streams for downstream polymerization
processes.[3g, 4] Detailed mechanistic studies of the former have
[*] D. Albani,[+] Dr. G. Vilÿ, Dr. S. Mitchell, Dr. O. Martin,
Prof. J. Pÿrez-RamÌrez
ETH Zurich, Department of Chemistry and Applied Biosciences
Institute for Chemical and Bioengineering
Vladimir-Prelog-Weg 1, 8093 Zurich (Switzerland)
E-mail: jpr@chem.ethz.ch
Dr. M. Capdevila-Cortada,[+] Prof. N. LÛpez
Institute of Chemical Research of Catalonia (ICIQ) and
The Barcelona Institute of Science and Technology
Av. PaÔsos Catalans 16, 43007 Tarragona (Spain)
E-mail: nlopez@iciq.es
[+] These authors contributed equally to this work.
Supporting information and the ORCID identification number(s) for
the author(s) of this article can be found under:
https://doi.org/10.1002/anie.201704999.
Angew. Chem. Int. Ed. 2017, 56, 10755 –10760

unraveled the role of the surface CeˇO pairs in activating
molecular hydrogen (H2) and stabilizing the various reaction
intermediates.[4a] However, progressive surface reduction and
oxygen (O) vacancy formation induced rapid activity loss
under operating conditions, limiting the performance of this
oxide.[4b,c] On the other hand, the fine dispersion of planar
FeII,III oxide nanoparticles on appropriate acidic oxide supports required to attain a high chemo- and stereoselective
character, still presents various synthetic challenges for the
industrial implementation of this oxide. To date, no other
metal oxides have been identified to selectively catalyze this
reaction, leaving ample room for a more extensive investigation on the abilities of oxides in hydrogenation catalysis.
¯
Indium oxide (In2O3, bixbyite structure, space group Ia3)
shows an intrinsic n-type conductivity related to the possibility of forming O vacancies and entrapping hydrogen atoms
(H), which act as carrier donors.[5] We were intrigued by the
rich electronic structure of In2O3, and by the promising
catalytic properties recently identified in the synthesis of
methanol from CO2 and H2.[6] Herein, the outstanding
performance of In2O3 in the semihydrogenation of acetylene
is described. In depth characterization reveals surface reduction by O vacancy formation, while detailed theoretical
assessment points to the creation of a new type of active site
comprising the O vacancy and surrounding In3O5 unit that
behaves as a “single ensemble”.
In2O3 was prepared by the controlled thermal decomposition of In(OH)3 at 673 K (see the Supporting Information
for preparative details). Analysis by X-ray diffraction (XRD)
confirms the transformation (Figure 1 a) of the hydroxide
precursor to the pure oxide, as indicated by the sharp
reflections of its XRD peaks, assigned to a compound with
bixbyite structure with an estimated average crystal size of
about 15 nm. Examination by high-resolution transmission
electron microscopy (HRTEM) reveals the presence of
aggregated nanocrystals of comparable dimensions, which
predominantly display the (111) facet (Figure 1 b; Supporting
Information, Figure S2). The prevalence of this surface is in
line with the lowest energy provided by first principles
calculations.[7a] A type IV isotherm with type II hysteresis
typical of a nanocrystalline material is observed by N2
sorption (Figure S1), evidencing a relatively high surface
area (108 m2 gˇ1) and pore volume (0.38 cm3 gˇ1), and a discrete distribution of intercrystalline pore sizes centered on
10 nm.
The performance of In2O3 was evaluated in the semihydrogenation of acetylene. Light-off curves, obtained by
monitoring the acetylene conversion as the temperature was
ramped from 523 to 673 K, present a distinct S-shape, with the

⌫ 2017 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

10755

Communications

Angewandte

Chemie

Figure 2. a) Acetylene conversion (solid symbol) and ethylene selectivity (open symbol) over In2O3 (blue) and CeO2 (red) as a function of
the reaction temperature at H2 :C2H2 = 30. The temperature windows
for the optimal performance of both catalysts are highlighted. b) Effect
of the feed H2 :C2H2 ratio on the acetylene conversion (c*c) and
c
selectivity to ethylene (a*a), ethane (a~a), and oligomers
a
a
(a&a) over In2O3 at T = 573 K. c) Long-term test in the presence
a
of excess ethylene at T = 558 K; acetylene conversion (*) and selectivity to ethylene (*). d) Conversion (&) and selectivity to ethylene (&),
ethane (&), and oligomers (&) in excess ethylene at T = 623 K. Other
conditions: t = 1 s, P = 1 bar.
Figure 1. a) XRD patterns of In(OH)3 (c), In2O3 (c), and In2O3
c
c
after temperature cycle experiments at 523–673 K (c), and correc
sponding total surface area. b) Transmission electron micrographs of
the In(OH)3 precursor (top left), the fresh In2O3 (top right), and the
In2O3 after the temperature cycle experiments (bottom row). The
vertical lines below the XRD patterns indicate the expected reflection
positions of cubic In2O3.

activity increasing abruptly from a few percent to full
conversion by 600 K. The selectivity toward C2H4 is around
95 % in the low-conversion regime, and converges to about
85 % at full conversion (Figure 2 a). Comparatively, fixing the
temperature at 573 K and increasing the feed ratio of H2 :C2H2
from 5 to 30 enhances the conversion (from 30 to 65 %) and
selectivity to ethylene (from 38 to 85 %), while the selectivity
to oligomers decreases (from 60 to 14 %, Figure 2 b). Almost
no overhydrogenated products (that is, C2H6 formation) are
observed (< 5 %) even at high feed ratios, which is a unique
feature of oxides.[4] No hysteresis in the activity is revealed
upon cycling the temperature of the catalyst bed in the range
523–673 K. A 100 h stability test conducted in the moderate
conversion regime and applying excess ethylene in the feed—
conditions that would accelerate potential deactivation
phenomena and selectivity issues—confirms the stable performance of In2O3 in acetylene semihydrogenation (Figure 2 c). The remarkable selectivity to ethylene was further
demonstrated by a test at full conversion and in the presence
of ethylene (Figure 2 d).

10756

www.angewandte.org

Characterization of the catalyst by XRD after use in the
cyclic experiments at 673 K demonstrates preservation of the
In2O3 phase, which has an increased crystallinity (Figure 1).
This agrees with the lower surface area of the used material
(25 m2 gˇ1) with respect to the fresh catalyst, and the larger
crystal size observed by microscopy analysis. Note that the
crystals retain a similar morphology in the fresh and used
catalysts, and the (111) surface remains dominant (Figures 1 b
and S2). When performing the same cyclic measurement in
Figure 2 a over a wider temperature range (523–873 K),
a drastic irreversible activity loss is observed above 850 K
(Figure S3 a). The XRD pattern of the catalyst recovered
after this test indicates complete reduction of the oxide to
metallic In (Figure S3 b). To confirm the catalytic relevance of
the (111) surface, In2O3 nanocubes mostly exposing (100)
facets were synthesized and evaluated in acetylene semihydrogenation (Figure S4), exhibiting a rate of ethylene
formation (7.5 mmolC2 H4 minˇ1 gˇ1) half that obtained over
the original catalyst (15.6 mmolC2 H4 minˇ1 gˇ1). Analysis of the
used catalyst confirmed the preservation of the cubic
morphology under the testing conditions.
Compared to In2O3, CeO2 exhibits lower temperature
activity (Figure 2 a), reaching a maximum acetylene conversion of 90 % at 523 K with slightly inferior selectivity to
ethylene (80 %). Progressive deactivation is observed at
T > 575 K, resulting from the formation of O vacancies

⌫ 2017 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Angew. Chem. Int. Ed. 2017, 56, 10755 –10760

Communications
under the reaction conditions.[4b,c] The distinct behavior of
In2O3 and CeO2 is striking and calls for a more detailed
mechanistic understanding of the performance of the former
oxide in acetylene hydrogenation.
Hydrogenation reactions require the surface to efficiently
activate H2. As shown by the temperature-programmed
reduction (H2-TPR) profile (Figure 3 a), the In2O3 catalyst
displays a H2 uptake at about 500 K, which is lower than the
acetylene hydrogenation light-off temperature (> 550 K, Figure 2 a). Temperature-programmed desorption of acetylene
(C2H2-TPD), on the other hand, shows that the interaction of
C2H2 with In2O3 depends strongly on the temperature of
adsorption. Upon adsorption at 523 K, the C2H2-TPD profile
exhibits a desorption peak centered at 643 K. In contrast, only
a weak signal at 355 K is observed after adsorption at 323 K
(Figure S5). Density functional theory (DFT) calculations
conducted on the stoichiometric In2O3(111) surface reveal
that H2 adsorption presents a large energy barrier (0.85 eV),
leading to two stable surface hydroxy groups (Eads = ˇ2.54 eV
with respect to gas-phase H2, Figure 3 c). Comparatively,
acetylene adsorption only requires an energy barrier of
0.46 eV and results in an InˇCHCHˇO complex that exhibits
strong chemisorption (ˇ1.51 eV with respect to gas-phase
acetylene). Hence, based on the energy values above, the
adsorption of H2 on In2O3 is hampered in the presence of the
alkyne. To validate this result, the catalyst was analyzed by
H2-TPR after pretreatment with C2H2 at 523 K (Figure 3 a).
Indeed, exposure of the catalyst to the alkyne induces a shift
of the first reduction peak from 500 to 650 K, confirming that
C2H2 chemisorption prevents low-temperature surface reduction. Significantly, the H2 consumption peak starts exactly at
550 K, which corresponds to the onset temperature of
hydrogenation activity (Figure 2 a). In summary, once the
temperature is ca. 535 K, C2H2 starts desorbing from the
surface. Molecular H2 can then adsorb dissociatively, generating two hydroxy species that can recombine to form water
(H2O) with an activation energy of 1.20 eV. These steps have
been experimentally verified by diffuse reflectance infrared
Fourier transform spectroscopy (DRIFTS) under exposure to
H2. This analysis clearly revealed the appearance of features
indicative of OH bending at 1580, 1430, and 1370 cmˇ1,
evidencing the formation of hydroxy species upon hydrogen
activation (Figures 3 b and S6 a).[7b,c] Upon reaching the onset
temperature of surface reduction (Figure 3 a), broad bands
assigned to H2O appear at 1680 and 1210 cmˇ1. The H2O
formed only requires an additional 0.81 eV for its desorption
(see proceeding text), leaving behind a surface O vacancy
(Figure 3 c).
To address the state of the catalyst under reducing
conditions, an ab initio thermodynamics analysis was carried
out, taking into account the chemical potential of H, In, and
H2O (details are given in the Supporting Information). In the
configurational pool, several surface motifs such as O
vacancies, In adatoms, and H coverages were considered.
The most relevant structures and the In adatom distribution
are summarized in Figures S7 and S8, respectively. Under
hydrogenation conditions (that is, atmospheric pressure and
mild temperature), surface reduction is expected to occur by
the loss of H2O and consequent vacancy formation (see
Angew. Chem. Int. Ed. 2017, 56, 10755 –10760

Angewandte

Chemie

Figure 3. a) H2-TPR profiles of In2O3, prior (c) and after (c)
c
c
pretreatment with C2H2 at 523 K, and C2H2-TPD on In2O3 (a).
a
b) DRIFT spectra of In2O3 with varying temperature under H2 flow.
c) Reaction profiles for the chemisorption of C2H2 (c) and the
c
activation, with concomitant H2O formation, of H2 (c) on
c
In2O3(111). Inset: reaction intermediates; In (*), O (*), C (*), and H
(*); atoms at the surface (solid colors), atoms in the subsurface layer
(hatched). d) Top view of the In2O3(111) surface where the 12 surface
O atoms follow the color code according to their vacancy formation
energy (Evac) and the rest of the lattice is shown in light blue (In) and
red (O). The three In atoms belonging to the ensemble are connected
with a black dashed triangle. The vacancy formation energies are
calculated considering both O2 and H2O formation.

proceeding text).[8a,b] Nonetheless, In adatoms were also
considered as potential active sites since they were identified
to form in earlier scanning tunneling microscopy (STM)
experiments upon annealing of In2O3 at 773 K in ultra-high
vacuum (UHV).[8c] The dependence of the surface energy
with respect to the H chemical potential, at fixed In and H2O
chemical potentials (ˇ1.0 and ˇ2.0 eV, respectively) is shown
in Figure 4 a (additional chemical potential scenarios are
shown in Figure S9). From the results of Figures 4 a and S9, it
is shown that In adatoms form at low H and high In chemical
potentials; however, under the relevant hydrogenation con-

⌫ 2017 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

www.angewandte.org

10757

Communications

Angewandte

Chemie

less stable (0.38 and 0.95 eV for a vacancy in the higher and
lower subsurface layers). Diffusion of the vacancy is only
allowed through the three violet positions in Figure 3 d, which
lead to the formation of identical ensembles. Diffusion
outside this confined region is not possible because of the
high energy barriers (Figure 4 b). As a result of the anisotropy
of the lattice, in which O atoms are bonded to differently
coordinated In atoms, the O vacancy is effectively shielded
from aggregation and diffusion, which leads to its confinement and isolation.
The semihydrogenation of acetylene (Figure 5), and its
potential overhydrogenation and oligomerization side paths,
were investigated on the most stable surface configuration
under reaction conditions (that is, a reduced In2O3ˇy(111)
surface). The adsorption of C2H2 occurs at an InˇO pair of the

Figure 4. a) Relative surface energy per (1 î 1) cell of several reduced
configurations, namely the partially reduced surfaces with one In
adatom (In2+xO3(111)), one O vacancy (In2O3ˇy(111)), two In adatoms
(In2+x’O3(111)), or one In adatom and one O vacancy (In2+xO3ˇy(111)),
and the stoichiometric surface (In2O3(111)), as a function of the
hydrogen chemical potential. In (DmIn) and H2O (DmH2 O ) chemical
potentials were fixed to ˇ1.0 and ˇ2.0 eV, respectively. Completely
hydrogenated surfaces were considered for In2O3(111)_11H2,
In2+xO3(111)_11H2, and In2O3ˇy(111)_11H2. b) Single ensembles
formed on In2O3(111) under reaction conditions as a result of an O
vacancy formation; In (*) and O (*) atoms. The three In atoms
belonging to the ensemble are connected with a black dashed triangle.
Diffusion channels of the surface O atoms adjacent to the vacancy are
marked with arrows, the absence of outward diffusion paths indicates
confinement of the vacancy.

ditions (for example, p(H2) ⇡ 0.96 bar), a single isolated O
vacancy per unit cell is the most stable surface configuration.
To further support the formation of vacancies, in situ Raman
spectroscopy measurements in flowing H2 have been carried
out, which reveal a reduction in the intensity of phonon
scattering modes associated with InO6 octahedra (Figure S6 b), in agreement with previous observations.[5c, 8d]
These results corroborate that the surface state under
reaction conditions is O-depleted, and thus the nature and
properties of the associated active site are further explored.
Vacancy formation energies (Evac) of the surface O atoms for
the stoichiometric In2O3(111) surface (Figure 3 d) range
between ˇ0.64 and 0.30 eV with respect to the gas-phase
H2O reference. Differing from the lowest energy surface
terminations of other representative oxides, four types of
inequivalent atoms are distinguished (Table S1 and Figure S10). Those with the lowest abstraction energy are
colored violet in Figure 3 d. For this O vacancy, the excess
charge delocalizes within the ensemble composed by the
vacancy, the first neighboring In atoms, and the second
neighboring O atoms (Figure S11). Once the first vacancy is
formed, the abstraction of a second O at neighboring sites is
less likely (Figure S12) whereas subsurface defects are also

10758

www.angewandte.org

Figure 5. Energy profile for the semihydrogenation of acetylene on the
reduced In2O3ˇy(111) surface, including C2H2 adsorption and activation
(I), H2 coadsorption and activation (II), C2H2 hydrogenation (III), and
C2H4 desorption (IV). Inset: reaction intermediates; color code as in
Figure 3. The three In atoms belonging to the ensemble are connected
with a black dashed triangle.

In3O5 ensemble adjacent to the vacancy, with strong bond
polarization leading to InˇCd-HˇCd+HˇO. In contrast to the
stoichiometric surface, at the ensemble H2 is activated as
HdˇˇHd+ in one of the neighboring InˇO pairs, where the Hdˇ
moiety sits close to the vacant site. To prove that C2H2 and H2
compete for the same site, a kinetic analysis has been
conducted, obtaining reaction orders of 0.5 for both reactants
(Figures S13 a,b). Moreover the apparent activation energy of
the reaction is about 80–90 kJ molˇ1, which corresponds to the
energy for H2 activation once the entropic and zero-point
vibrational energies are taken into account (Figure S13 c).
The hydroxy proton can then be transferred to the adsorbed
alkyne and the Hdˇ is subsequently transferred to the OˇC2H3
intermediate to form ethylene. The activation energies of
both hydrogenation steps, 1.17 and 1.31 eV, respectively, are
attainable at an experimental light-off temperature exceeding
550 K.
The ethylene molecule formed, easily detaches from the
ensemble because of its low desorption energy. The reaction
intermediates of each hydrogenation step are shown in
Figure S14, and the full reaction profile (combining the
steps in Figures 3 c and 5) is presented in Figure S15 of the
Supporting Information. The overhydrogenation of the olefin
is hindered as C2H4 desorption (0.22 eV) requires significantly
lower energy than its hydrogenation (1.54 eV), in line with the

⌫ 2017 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Angew. Chem. Int. Ed. 2017, 56, 10755 –10760

Angewandte

Communications
low selectivity towards ethane (< 5 %, see preceding text).
Similarly, the CˇC coupling of two coadsorbed C2H2 molecules presents rather low activation energy (1.18 eV; Figure S16), explaining the high hydrogen partial pressure
required to achieve the exceptional level of ethylene selectivity observed in Figure 2 b. Furthermore, to prove that the
In3O5 ensemble is the sole active site, the reaction energy
profile was also evaluated on the stoichiometric In2O3(111)
surface (Figure S17), showing that a high activation energy of
2.27 eV is required for the second hydrogenation step.
Similarly, the reaction cannot occur at the In adatom
In2+xO3(111) surface as the heterolytic activation of hydrogen
is thermodynamically unfavorable (Figure S18) and the
homolytic dissociation of H2 is very energy demanding
(Figure S17). In summary, the complementary polarity of
the CdˇˇCd+ and HdˇˇHd+ moieties found on the reduced
In2O3ˇy(111) surface facilitates the reaction and is responsible
for the low activation energy of both hydrogenation steps. As
the heterolytic activation of H2 is only possible at the
ensemble created upon vacancy formation (Figure S18), this
represents the only site where the reaction can proceed with
low energy requirement.
In conclusion, we have described In2O3 as a novel catalyst
for the semihydrogenation of acetylene, which exhibits 85 %
ethylene selectivity at full conversion even in ethylene excess
and at temperatures above 550 K. Theoretical models rationalize this outstanding performance predicting the formation,
under reaction conditions, of a unique In3O5 site consisting of
an O vacancy, an In trimer, and adjacent O atoms, where C2H2
and H2 can coadsorb and react. The precisely defined nature
of these ensembles, which represent a novel variant of
ensemble size control for non-metallic systems in hydrogenations, originates from the low symmetry of the In2O3,
which permits a high ensemble density (ca. 0.08 ML) while
ensuring isolation (as diffusion is hindered) and avoiding
lateral interactions with other centers. Since the active center
contains different types of atoms, this site is denoted as
a “single-ensemble heterogeneous catalyst” (SEHC). In
contrast to oxidation reactions where vacancy sites are
considered active centers, this SEHC requires both the In
and O atoms surrounding the empty site for the reaction to
occur. When compared to single-atom heterogeneous catalysts (SAHC), on the other hand, where atomic dispersion
and high densities are very difficult to attain simultaneously,
the confinement and robustness of the ensemble circumvent
typical pitfalls related to the miniaturization of the ensemble.
Oxides presenting large unit cells with marked anisotropy are
natural scaffolds for single-ensemble sites representing new
selective catalytic materials for hydrogenations and beyond.
Future advances in microscopy techniques will be essential to
gain a deeper understanding of these sites.

Acknowledgements
Financial support from ETH Zurich, ICIQ Foundation, the
Spanish MINECO (CTQ2015-68770-R), and the Swiss
National Science Foundation (Grant No. 200021-169679) is
acknowledged. ScopeM at ETH Zurich is thanked for
Angew. Chem. Int. Ed. 2017, 56, 10755 –10760

Chemie

providing access to its facilities. O. Sambalova and Dr. A.
Borgschulte (EMPA D¸bendorf) are thanked for the in situ
Raman experiments. M.C.-C. also acknowledges MINECO
for a “Juan de la Cierva-FormaciÛn” fellowship (FJCI-2014–
20568). We would like to thank U. Diebold and M. Wagner
(TU Wien) for discussions.

Conflict of interest
The authors declare no conflict of interest.
Keywords: alkyne semihydrogenation ·
density functional theory · ensembles · indium oxide ·
site isolation
How to cite: Angew. Chem. Int. Ed. 2017, 56, 10755 – 10760
Angew. Chem. 2017, 129, 10895 – 10900
[1] a) M. Sankar, N. Dimitratos, P. J. Miedziak, P. P. Wells, C. J. Kiely,
G. J. Hutchings, Chem. Soc. Rev. 2012, 41, 8099 – 8139; b) G. Vilÿ,
D. Albani, N. Almora-Barrios, N. LÛpez, J. Pÿrez-RamÌrez,
ChemCatChem 2016, 8, 21 – 33.
[2] a) F.-M. McKenna, J. A. Anderson, J. Catal. 2011, 281, 231 – 240;
b) A. J. McCue, A. Guerrero-Ruiz, I. RodrÌgez-Ramos, J. A.
Anderson, J. Catal. 2016, 340, 10 – 16; c) Y. Luo, S. A. Villaseca,
M. Friedrich, D. Teschner, A. Knop-Gericke, M. Armbr¸ster, J.
Catal. 2016, 338, 265 – 272; d) G. Kyriakou, M. B. Boucher, A. D.
Jewell, E. A. Lewis, T. J. Lawton, A. E. Baber, H. L. Tierney, M.
Flytzani-Stephanopoulos, E. C. H. Sykes, Science 2012, 335,
1209 – 1212; e) F. R. Lucci, J. Liu, M. D. Marcinkowski, M.
Yang, L. F. Allard, M. Flytzani-Stephanopoulos, E. C. H. Sykes,
Nat. Commun. 2015, 6, 8550; f) H. Wei, X. Liu, A. Wang, L.
Zhang, B. Qiao, X. Yang, Y. Huang, S. Miao, J. Liu, T. Zhang, Nat.
Commun. 2014, 5, 5634; g) G. Xu, H. Wei, Y. Ren, J. Yin, A.
Wang, T. Zhang, Green Chem. 2016, 18, 1332 – 1338; h) P. Liu, Y.
Zhao, R. Qin, S. Mo, G. Chen, L. Gu, D. M. Chevrier, P. Zhang, Q.
Guo, D. Zang, B. Wu, G. Fu, N. Zheng, Science 2016, 352, 797 –
800; i) G. Vilÿ, D. Albani, M. Nachtegaal, Z. Chen, D. Dontsova,
M. Antonietti, N. LÛpez, J. Pÿrez-RamÌrez, Angew. Chem. Int. Ed.
2015, 54, 11265 – 11269; Angew. Chem. 2015, 127, 11417 – 11422.
[3] a) E. J. Grootendorst, R. Pestman, R. M. Koster, V. Ponec, J.
Catal. 1994, 148, 261 – 269; b) Y. Sakata, V. Ponec, Appl. Catal. A
1998, 166, 173 – 184; c) R. V. Jagadeesh, A.-E. Surkus, H. Junge,
M.-M. Pohl, J. Radnik, J. Rabeah, H. Huan, V. Sch¸nemann, A.
Br¸ckner, M. Beller, Science 2013, 342, 1073 – 1076; d) A. Primo,
F. Neatu, M. Florea, V. Parvulescu, H. Garcia, Nat. Commun.
2014, 5, 5291; e) K. V. Kovtunov, D. A. Barskiy, O. G. Salnikov,
A. K. Khudorozhkov, V. I. Bukhtiyarov, I. P. Prosvirin, I. V.
Koptyug, Chem. Commun. 2014, 50, 875 – 878; f) J. Song, Z.-F.
Huang, L. Pan, J. Zou, X. Zhang, L. Wang, ACS Catal. 2015, 5,
6594 – 6599; g) M. Tejada-Serrano, J. R. Cabrero-Antonino, V.
Mainar-Ruiz, M. LÛpez-Haro, J. C. Hernµndez-Garrido, J. J.
Calvino, A. Leyva-Pÿrez, A. Corma, ACS Catal. 2017, 7, 3721 –
3729.
[4] a) G. Vilÿ, B. Bridier, J. Wichert, J. Pÿrez-RamÌrez, Angew. Chem.
Int. Ed. 2012, 51, 8620 – 8623; Angew. Chem. 2012, 124, 8748 –
8751; b) G. Vilÿ, S. Wrabetz, L. Floryan, M. E. Schuster, F.
Girgsdies, D. Teschner, J. Pÿrez-RamÌrez, ChemCatChem 2014, 6,
1928 – 1934; c) G. Vilÿ, S. Colussi, F. Krumeich, A. Trovarelli, J.
Pÿrez-RamÌrez, Angew. Chem. Int. Ed. 2014, 53, 12069 – 12072;
Angew. Chem. 2014, 126, 12265 – 12268.
[5] a) T. Koida, H. Fujiwara, M. Kondo, Jpn. J. Appl. Phys. 2007, 46,
L685 – L687; b) S. Limpijumnong, P. Reunchan, A. Janotti, C.
Van de Walle, Phys. Rev. B 2009, 80, 193202; c) T. Bielz, H.

⌫ 2017 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

www.angewandte.org

10759

Angewandte

Communications
Lorenz, W. Jochum, R. Kaindl, F. Klauser, B. Klçtzer, S. Penner, J.
Phys. Chem. C 2010, 114, 9022 – 9029; d) W. Yin, K. Smithe, P.
Weiser, M. Stavola, W. B. Fowler, L. Boatner, S. J. Pearton, D. C.
Hays, S. G. Koch, Phys. Rev. B 2015, 91, 075208; e) A. Walsh,
Appl. Phys. Lett. 2011, 98, 261910.
[6] O. Martin, A. J. MartÌn, C. Mondelli, S. Mitchell, T. F. Segawa, R.
Hauert, C. Drouilly, D. Curulla-Ferrÿ, J. Pÿrez-RamÌrez, Angew.
Chem. Int. Ed. 2016, 55, 6261 – 6265; Angew. Chem. 2016, 128,
6369 – 6373.
[7] a) A. Walsh, R. A. Catlow, J. Mater. Chem. 2010, 20, 10438 –
10444; b) M. Jothibas, C. Manoharan, S. S. Jeyakumar, P. Praveen,
J. Mater. Sci. Mater. Electron. 2015, 26, 9600 – 9606; c) G. M. Silva,
E. H. de Faria, E. J. Nassar, K. J. Ciuffi, P. S. Calefi, Quim. Nova
2012, 35, 473 – 476.

10760 www.angewandte.org

Chemie

[8] a) K. H. L. Zhang, R. G. Egdell, F. Offi, S. Iacobucci, L. Petaccia,
S. Gorovikov, P. D. C. King, Phys. Rev. Lett. 2013, 110, 056803;
b) F. Lei, Y. Sun, K. Liu, S. Gao, L. Liang, B. Pan, Y. Xie, J. Am.
Chem. Soc. 2014, 136, 6826 – 6829; c) M. Wagner, S. Seiler, B.
Meyer, L. A. Boatner, M. Schmid, U. Diebold, Adv. Mater.
Interfaces 2014, 1, 1400289; d) M. Kumar, V. N. Singh, F. Singh,
K. V. Lakshmi, B. R. Mehta, J. P. Singh, Appl. Phys. Lett. 2008, 92,
171907.

Manuscript received: May 15, 2017
Revised manuscript received: June 24, 2017
Accepted manuscript online: July 11, 2017
Version of record online: July 25, 2017

⌫ 2017 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Angew. Chem. Int. Ed. 2017, 56, 10755 –10760