Discovery of Superior Cu-GaOx-HoOy Catalysts for the Reduction of Carbon Dioxide to Methanol at Atmospheric Pressure

Article abstract of DOI:10.1002/cctc.201600020

Catalytic conversion of carbon dioxide to liquid fuels and basic chemicals by using solar-derived hydrogen at, or near, ambient pressure is a highly desirable goal in heterogeneous catalysis. If realized, this technology could lead to a more sustainable society together with decentralized power generation. A novel class of holmium-containing multi-metal oxide Cu catalysts, discovered through the application of high-throughput methods, is reported. In particular, ternary systems of Cu-GaO_x-HoO_y>Cu-CeO_x-HoO_y>Cu-LaO_x-HoO_y supported on γ-Al₂O₃ exhibited superior methanol production (10×) with less CO formation than previously reported catalysts at 1 atm pressure. Holmium was shown to be highly dispersed as few-atom clusters, suggesting that the formation of trimetallic sites could be the key for the promotion of methanol synthesis by Ho.

Full text of DOI:10.1002/cctc.201600020

DOI: 10.1002/cctc.201600020
Communications
Discovery of Superior Cu-GaOx-HoOy Catalysts for the
Reduction of Carbon Dioxide to Methanol at Atmospheric
Pressure
Bahman Zohour,[a] Iskender Yilgor,[b] Mehmet A. Gulgun,[c] Ozgur Birer,[b] Ugur Unal,[b]
Craig Leidholm,[d] and Selim Senkan*[a]
Catalytic conversion of carbon dioxide to liquid fuels and basic
chemicals by using solar-derived hydrogen at, or near, ambient
pressure is a highly desirable goal in heterogeneous catalysis.
If realized, this technology could lead to a more sustainable so-
ciety together with decentralized power generation. A novel
class of holmium-containing multi-metal oxide Cu catalysts,
discovered through the application of high-throughput meth-
ods, is reported. In particular, ternary systems of Cu-GaOx-
HoOy >Cu-CeOx-HoOy ~Cu-LaOx-HoOy supported on g-Al2O3 ex-
hibited superior methanol production (10”) with less CO for-
mation than previously reported catalysts at 1 atm pressure.
Holmium was shown to be highly dispersed as few-atom clus-
ters, suggesting that the formation of trimetallic sites could be
the key for the promotion of methanol synthesis by Ho.
combustion engines and fuel cells, thereby allowing the con-
tinuous generation of electricity and enabling energy sustaina-
bility.
Currently, methanol is produced exclusively from syngas
(CO/CO2/H2) obtained from the reforming of fossil fuels in cen-
tralized facilities, over Cu-ZnO catalysts promoted by Al2O3 at
50–100 bar (1 bar=0.1 MPa) and 200–3008C.[9–15] This catalyst
is also active for the water–gas-shift reaction.[16–19] Owing to
the practical importance of this technology, catalysts for meth-
anol synthesis have been studied extensively and optimized
for use with syngas at high pressures. Most catalysts investigat-
ed to date for the title reaction are based on the Cu-ZnO-Al2O3
system modified by metals such as Zr, Ga, Si, Al, B, Cr, V, Ti, and
others.[20] Although high methanol yields have been reported
at a pressure of 360 bar,[21] Cu-ZnO-Al2O3 system could not be
used efficiently for neat CO2 reduction at low pressures.[22] This
finding demonstrated the need for the discovery and optimiza-
tion of new and more active catalysts for methanol synthesis.
Recently, a Ni-Ga/SiO2 system prepared by the standard im-
pregnation technique was reported to give methanol yields
that were comparable to the traditional Cu-ZnO-Al2O3 catalysts
prepared by co-precipitation at atmospheric pressure.[22] In an-
other recent report, also at 1 atm, CeOx deposited on Cu(111)
and Cu-CeOx co-deposited on TiO2(110) surfaces (by chemical
vapor deposition) produced significantly more CH3OH than the
Cu-ZnO(0001) surfaces.[23] A hybrid oxide catalyst comprising
MnOx nanoparticles supported on mesoporous Co3O4 was also
recently reported to exhibit significant CH3OH production at
a higher pressure of 4 bar, together with CO and hydrocarbon
products.[24]
Efficient catalytic reduction of carbon dioxide, the primary
product of the combustion of fossil fuels and an increasing at-
mospheric greenhouse gas, to methanol represents an impor-
tant step towards the creation of a sustainable society. Metha-
nol has been proposed both as a chemical feedstock as well as
a convenient liquid medium of energy storage.[1–4] Although
the CO2 reduction process requires molecular hydrogen, it can
easily be obtained by the electrolysis of water by using elec-
tricity obtained from photovoltaic cells or wind turbines.[5–8]
Once produced, methanol can be transformed into a wide
range of useful chemicals, such as dimethyl ether (DME), ethyl-
ene, gasoline, diesel, and others by using established tech-
nologies.[1–3] In addition, methanol can also be used directly in
Methanol formation from carbon dioxide and hydrogen pro-
ceeds via the following reaction:
[a] B. Zohour, Prof. Dr. S. Senkan
Department of Chemical Engineering
University of California
CO2 þ 3 H2 $ CH3OH þ H2O DHr ¼ À52 kJ molÀ1
ð1Þ
Los Angeles, CA (USA)
E-mail: ssenkan@gmail.com
This reaction is often accompanied by the reverse water–
gas-shift reaction (RWGS):
[b] Prof. Dr. I. Yilgor, Prof. Dr. O. Birer, Prof. Dr. U. Unal
Department of Chemistry, KUYTAM
Koc University
CO2 þ H2 $ CO þ H2O DHr ¼ 42 kJ molÀ1
ð2Þ
Sariyer, Istanbul (Turkey)
In the presence of acidic surfaces, for example, H-ZSM5 or g-
Al2O3, the CH3OH produced is readily converted to DME, which
is also a desirable product, by dehydration:[9,10,25]
[c] Prof. Dr. M. A. Gulgun
Department of Material Science and Engineering
Sabanci University
Tuzla, Istanbul (Turkey)
2-CH3OH $ CH3OCH3 þ H2O DHr ¼ À15 kJ molÀ1
ð3Þ
[d] C. Leidholm
Laboratory Catalyst Systems, LLC
Los Angeles, CA (USA)
Herein, we report the discovery of a series of superior and
novel holmium-containing catalytic materials for the low-pres-
sure reduction of CO2 to CH3OH and DME. By using impregna-
Supporting Information for this article can be found under http://
dx.doi.org/10.1002/cctc.201600020.
ChemCatChem 2016, 8, 1464 – 1469
1464
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Communications
tion and high-throughput (HT) catalyst-screening technologies
developed in our laboratories,[26,27] we systematically investigat-
ed the oxides of single, binary, and ternary combinations of 27
metals (total metal atom loading of 20 wt%) with several sup-
port materials in over 3000 experiments. Our studies led to the
discovery of g-Al2O3-supported Cu-GaOx-HoOy as well as Cu-
CeOx-HoO, and Cu-LaOx-HoOy systems, which exhibit superior
methanol production and less CO formation than other materi-
als reported in the literature. The observed higher activity and
selectivity of the Cu-Ga-Ho system could to be related to the
formation of trimetallic active sites.
For example, at 2608C (Figure 1a), the Cu-GaOx-HoOy catalyst
(4) produced CH3OH at 1.14”10À4 %, which is about a factor of
10 higher than the Cu-Zn/Al2O3 system (1) of 0.112”10À4 %,
while producing similar levels of CO. It is also important to
note that catalyst 4 also produced significant levels of DME. In
fact, if we were to combine the yields for DME (2”0.42”
10À4 =0.84”10À4) and CH3OH (1.14”10À4) (4), at 2608C, the
performance of catalyst 4 would be a factor of 17 higher than
that of catalyst 1 and 13. These results correspond to
a CH3OH+DME selectivity of 48% for the Cu-GaOx-HoOy (4)
catalyst at 2608C. The turnover frequency (TOF) of catalyst 4
was estimated to be approximately 1.6”10À4 sÀ1 at 2608C (Fig-
ure 1a) for the combined production of CH3OH and DME; this
was calculated by assuming ꢀ7 nm diameter spherical Cu
metal clusters (ꢀ20000 Cu atoms) and ꢀ3000 surface atoms
exposed for reaction and 10% reactant gas utilization.[27] Simi-
lar considerations for the Cu-Zn/Al2O3 catalyst (1) result in
a TOF value of 0.45”10À5 sÀ1, which is in agreement with the
values reported in literature.[22,24]
Initial screening experiments led to the determination of
a number of binary systems that exhibited catalytic activity for
CH3OH synthesis mostly over the g-Al2O3 support. These binary
systems, in decreasing order of CH3OH production at 2608C
were: Cu-Ga[14,28] >Cu-La[29] ~Cu-Ce[23] >Cu-Ho~Cu-Zr[15] >Ga-
Ni/SiO2[22] ~Cu-Zn[10,12,13] ~Ga-Ho~Cu-Mg>Zn-Ir, which are con-
sistent with the literature. The validity of our experimental ap-
proach is supported by the observation that the relative per-
formances of our as-prepared Cu-Zn/Al2O3 and Ga-Ni/SiO2 cata-
lysts are similar to one another, a finding that is identical to re-
sults reported by Studt et al. who used co-precipitation to syn-
thesize the traditional Cu-Zn-Al2O3 catalysts.[22] The higher-
performing binary systems were then used as the basis to ex-
plore the ternary systems at different loadings and tempera-
tures. In Figure 1, the reactor exit mole percentages for CH3OH
(”104), DME (”104), and CO (”0.5”103) are presented for the
Ho-containing ternary catalysts with the best performance to-
gether with selected binary systems for comparison. The
values presented in Figure 1 correspond to the average of
three different sets of experiments that were within 10% of
each other. It should be noted that CO2 conversions, thus
product mole fractions, were small because of the high gas ve-
locities used (GHSVꢀ200000 hÀ1). The high gas velocities al-
lowed the catalysts to remain isothermal, which enabled the
undertaking of rigorous comparisons of their intrinsic activities.
The results reported herein must be studied in greater detail
to better understand the catalyst structures, activities, selectivi-
ties, reaction mechanisms, and optimization of their perform-
ances.
Holmium also had a dramatic promotional effect on some of
the reported binary CH3OH catalyst systems. For example, both
CH3OH and DME production increased significantly by the Ho
doping of catalyst 1 by more than a factor of two (2) at 3008C.
For the case of the Cu-CeO2 (11), Ho doping was also influen-
tial, increasing CH3OH levels by approximately a factor of two
(12) at 2608C. However, Ho did not promote CH3OH formation
in the Ga-Ni/Al2O3 system (15), although it significantly in-
creased CO and CH4 (not included in Figure 1) production.
Increasing the temperature from 260 to 2808C significantly
increased CH3OH production for the Cu-GaOx-HoOy (4) catalyst.
However, increasing the temperature further from 280 to
3008C resulted in a smaller increase in CH3OH formation. This
result is not surprising in view of the equilibrium considera-
tions of this exothermic reaction [Eq (1)].[31] On the other hand,
increasing the temperature increased the CO production sub-
stantially, clearly demonstrating the need to develop low-tem-
perature catalysts for the synthesis of CH3OH from CO2.
Rapid decreases in CH3OH production were observed within
few hours with the Cu-Zn/Al2O3 (1) catalyst. None of the g-
Al2O3-supported Cu-GaOx-HoOy (4), Cu-LaOx-HoOy (10), or Cu-
CeOx-HoOy (12) exhibited any significant deactivation or
change in methanol selectivity during ꢀ10 h of continuous
runs or after repeated reduction–reaction cycles. The time-on-
stream performance of the Cu-GaOx-HoOy (4) catalyst present-
ed in Figure 2 at 2608C shows that the combined selectivities
for CH3OH and DME remained at approximately 48% for the
entire 10 h testing period.
From Figure 1a it can be seen that our HT experiments pro-
duced the following order for CH3OH production for some of
the previously reported catalysts at 2608C: Cu-Ga2O3 (3)[14,28]
Cu-La2O3 (9)[29] >Cu-CeO2 (11)[23] >Cu-Zn-(Zr-Al2O3) (7)[15,30]
>
>
Cu-ZrO2 (6)[15] >Ga-Ni/SiO2 (13)[22] ~MnOx/m-Co3O4 hybrid
(16)[24] ~Cu-ZnO2/Al2O3 (1).[10,12,13] Unlike the reported high per-
formance at 4 bar, the methanol production of the hybrid
MnOx/m-Co3O4 catalyst (16)[24] was surprisingly poor at 1 atm,
only on par with the as-prepared Cu-Zn/Al2O3 (1)~Ga-Ni/SiO2
(13) systems (see Figure 1a); nevertheless, 16 was a very active
catalyst producing very high levels of CO and CH4 along with
some C2H4 and higher hydrocarbons (C3+).
In the Cu-Zn/Al2O3 system (1), the metallic copper clusters
are accepted to be the sites for methanol synthesis, whereas
ZnO has been proposed to act both as a physical promoter
(i.e., to assist in the formation of a larger number of surface Cu
sites) and as a promoter for the Cu-ZnO synergy.[11] The same
Cu-ZnO sites are also believed to be catalysts for the RWGS re-
action. It is also widely accepted that CH3OH production from
CO2 over Cu-ZnO catalysts occurs via the formation of surface
formates HCOO-M!HCOOH-M!CH3O2-M!CH2O-M!CH3O-
M!CH3OH-M.[31]
As evident from Figure 1, our g-Al2O3-supported ternary Cu-
GaOx-HoOy catalyst (4, Cu-Ga-Ho at 8-8-4 metal wt%) together
with the Cu-LaOx-HoOy (10, Cu-La-Ho at 10-5-5 metal wt%)
and Cu-CeOx-HoOy (12, Cu-Ce-Ho at 10-5-5 metal wt%) sys-
tems significantly outperform the previously reported systems.
ChemCatChem 2016, 8, 1464 – 1469
www.chemcatchem.org
1465
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Communications
Figure 1. Reactor exit CH3OH, DME, and CO mol% levels for the high-performing Ho-containing catalysts together with some previously reported catalytic
systems at a) 2608C, b) 2808C, and c) 3008C. Significant amounts of CH4 were also produced with the Ni-containing catalysts (13, 14, 15) and for 16, but are
not presented for clarity. [n]* refers to catalytic materials reported recently in the literature at low pressures.
For the Ni-Ga/SiO2 system (13), the Ga-rich sites have been
reported to facilitate methanol synthesis, whereas the Ni-rich
sites have been suggested to be responsible for the RWGS and
methanation reactions.[22] Consequently, the superior per-
formance of the Cu-GaOx-HoOy system (4), which contains
both Cu and Ga sites, can be due, in part, to their mutual phys-
ical and chemical promotion (3). The remarkable effect of Ho
in the promotion of methanol synthesis could be attributed to
the formation of very small clusters that are highly dispersed.
As discussed in the characterization section below, STEM
images of catalyst 4 showed the presence of few-atom (1–
3 atom) Ho clusters. Some of these clusters also appear to be
positioned along the Cu and Ga cluster interfaces or on the
surfaces of their alloys, creating trimetallic sites that could be
the key for the promotion of methanol synthesis by Ho. This
representation is also supported by the experimentally ob-
served order of activity of the CH3OH catalysts, that is, Cu-Ga-
Ho (4)>Cu-Ga (3)>Cu-Ho>Ga-Ho.
The reduced Cu-GaOx-HoOy system was characterized (4, Cu-
Ga-Ho at 8-8-4 metal wt% loading and 5.3/4.8/1 atom ratio) by
using BET analysis (Micromeritics ASAP 2020 using N2), scan-
ning transmission electron microscopy (STEM, JEOL, JEM-
ARM200CFEG UHR, with EDS), X-ray diffraction (XRD) (Bruker
Xflash 5010), and X-ray photoelectron (XPS) (ThermoScientific
ChemCatChem 2016, 8, 1464 – 1469
www.chemcatchem.org
1466
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Communications
Ex situ XPS studies of the reduced catalyst 4 indicated the
presence of both Cu0 and CuO, whereas only Ga2O3 and Ho2O3
were observed at the surface (see the Supporting Information).
Some oxidation of Cu was observed because of exposure to
the atmosphere during sample transfer to the XPS system.
Consequently, Cu-Ga2O3-Ho2O3 could be the correct represen-
tation of catalyst 4. It is interesting to note that the surface
concentration of Ga was higher, whereas those for Cu and Ho
were lower, than their nominal/bulk values for the Cu-Ga2O3-
Ho2O3 (4) catalyst. Further characterizations, as well as detailed
kinetic and quantum modeling studies, are underway to better
understand the nature of the active site(s) and to explain the
mechanism of action of the novel Cu-Ga2O3-Ho2O3 system in
the promotion of CH3OH synthesis from CO2 and H2.
Figure 2. Time-on-stream behavior of the Cu-GaOx-HoOy (4) catalyst shows
no significant performance change over a 10 h testing period.
In summary, high-throughput impregnation synthesis and
reaction screening of the binary and ternary combinations of
27 metals and 5 supports for the hydrogenation of CO2 to
CH3OH under atmospheric pressure led to the discovery of
a novel class of superior ternary catalytic materials containing
holmium, while reproducing the trends for the relative per-
formances of the already established binary catalytic systems
from the literature. The Cu-Ga2O3-Ho2O3/g-Al2O3 system exhibit-
ed the highest CH3OH production together with significant
levels of DME formation, and maintained its activity and selec-
tivity over a long period of time. Considering the significant
practical interest of these processes for a sustainable chemical
industry and society in general, the newly discovered catalytic
materials reported herein require further investigations to
better understand the nature of the active sites, optimize their
synthesis and operating conditions, including studies at higher
pressures, to increase CH3OH and DME yields.
K-Alpha) spectroscopy to develop insights on its structure and
surface chemistry. BET analysis indicated a surface area of
142 m2 gÀ1 for the catalyst, consistent with the alumina sup-
port. The XRD spectra (see the Supporting Information) only
showed the features of the Al2O3 support, suggesting the pres-
ence of amorphous Cu, Ga, and Ho moieties or their nanosize
crystallites. This finding was also confirmed by the high-resolu-
tion TEM studies (Figure 3); HR-TEM images of the reduced cat-
alyst 4 (Cu-Ga-Ho at 8-8-4 wt% metal loading and atom ratios
of 5.2/4.8/1) is provided in Figures 3a and 3b. Element map-
ping images obtained from STEM/EDS indicate the presence of
5–7 nm Ga (Figure 3e) and Cu (Figure 3 f) clusters in close
proximity to one another, which also suggests the possibility
of some alloy formation. On the other hand the Ho map did
not show the presence of larger particles (Figure 3c). The high
angular annual dark field (HAADF) Z-contrast (~z2) image indi-
cated that the heavy Ho (z=165) is highly dispersed and exists
only as few-atom (1–3 atom) clusters (Figure 3d).
Figure 3. a,b) HR-TEM images of the reduced catalyst 4 (Cu-Ga-Ho at 8-8-4 wt% metal loading and atom ratios of 5.2/4.8/1). c,d) STEM-EDS images of the ele-
ment maps of Ho, e) Ga, and f) Cu over g-Al2O3.
ChemCatChem 2016, 8, 1464 – 1469
www.chemcatchem.org
1467
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Communications
99.9% purity) and 75 vol% H2 (Matheson, 99.99%). Gas sampling
was accomplished by withdrawing reactor exit gases by using
a passivated 200 micron ID capillary sampling probe that was se-
quentially positioned into each reactor channel, followed by on-
line gas analysis either by mass spectrometry (MS, Stanford Re-
search Systems, RGA-200) or by gas chromatography (Micro-GC
Varian, CP-4900). The GC had dual Porapak U (10 m) and molecular
sieve 13X (10 m) modules each equipped with individual thermal
conductivity detectors (TCD). Using MS analysis the screening of
the entire 80-catalyst library typically took 30 min or less depend-
ing on the sampling capillary dwell time inside and outside the re-
actor channels and the mass range and scan rate of the MS.[26,27]
Consequently, MS analysis was used for initial screening to rapidly
identify promising leads. The leads were then studied in greater
detail by gas chromatography in smaller sets of 10 to 20 catalysts
to better compare them under similar time-on-stream conditions.
In the present work, each GC analysis took approximately 2.5 min
for completion including a 30 s sampling time. The following prod-
ucts were detected and quantified: CH3OH, CH3OCH3 (DME), CO,
CH4, C2s, and C3s. However, only CH3OH, CH3OCH3 (DME), and CO
were reported in Figure 1 for clarity.
Experimental Section
The catalysts were prepared by impregnation of powders of g-
Al2O3, CeO2, SiO2, TiO2, and Y-ZrO2 with single, binary, and ternary
mixtures of aqueous (nitrate) salt solutions of 27 metals: Li, Na, Rb,
Cs, Mg, Ca and Sr, Ga, V, Cr, Mn, Co, Ni, Cu, Zn, Zr, Ru, Ir, Ag, Au, La,
Ce, Pr, Dy, Ho, Er, and Yb. Systematic consideration of different
metal ratios and loadings necessitated the preparation of approxi-
mately 3000 distinct catalytic materials. For example, the best-per-
forming trimetallic (Cu-GaOx-HoOy/g-Al2O3, 4) catalysts were pre-
pared as follows: A predetermined amount of g-Al2O3 support (Alfa
Aesar, surface area of ꢀ150 m2 gÀ1) was soaked in an aqueous solu-
tion of Cu(NO3)2·2.5H2O, Ga(NO3)3·6H2O, and Ho(NO3)3·5H2O (Alfa
Aesar) at concentrations previously determined to yield the desired
metal loadings. The mixture was then dried at 1208C while stirring,
followed by calcining in air at 4508C for 6 h. In addition, the re-
cently reported hybrid MnOx/m-Co3O4 catalyst (16) was acquired
and tested.[24]
Catalyst screenings were performed by using a high-throughput
(HT) array channel microreactor system, which is shown in
Figure 4, details of which have been described previously.[26,27] The
system allows parallel screening of up to 80 catalytic materials. In
array microreactors, reactant gases flow over the flat surfaces of
compacted powders (20 mg of g-Al2O3-based and 6 mg of SiO2-
based) of catalytic materials that are placed into the wells in each
reactor channel (see Figure 4 top right inset). Consequently, the
majority of the gases exit the reactor while only a small fraction
(ꢀ10%) participate in the catalytic reaction process.[27] This ar-
rangement results in the establishment of identical flow rates or
contact times in every channel, which enables the rapid compari-
son of the catalytic performances of up to 80 catalysts in a single
experiment. The experiments were performed in the following
manner: Firstly, the catalysts were reduced under H2/He (50/50)
flow at a temperature in the range of 230–3508C for 2 h. Catalysts
were then cooled to the desired reaction temperature while still
under H2/He flow, and the gas flow was switched to the reactants.
The experiments were performed at 260, 280, and 3008C, at 1 atm
pressure and at gas hourly space velocity (GHSV) of approximately
200000 hÀ1. The feed gas consisted of 25 vol% CO2 (Matheson,
Acknowledgements
We thank Laboratory Catalyst Systems (LCS), LLC for funding and
access to their high-throughput equipment and catalyst data
base. We also thank Dr. Selim Alayoglu and Dr. Kwangjin An of
the Lawrence Berkeley National Laboratory for providing the
hybrid MgOx/CoOy catalyst. Author contributions: BZ, CL, and SS
initiated and performed the original catalyst discovery work at
LCS. IY, OB, and UU performed and interpreted the XRD and XPS
analysis. MAG performed and interpreted the HRTEM analysis.
Keywords: carbon dioxide · dimethyl ether · heterogeneous
catalysis · high-throughput testing · methanol synthesis
[1] G. A. Olah, Angew. Chem. Int. Ed. 2005, 44, 2636–2639; Angew. Chem.
2005, 117, 2692–2696.
[2] G. A. Olah, A. Goeppert, G. S. Prakash, J. Org. Chem. 2009, 74, 487–498.
[3] G. A. Olah, A. Goeppert, G. S. Prakash, Beyond Oil and Gas: The Methanol
Economy, Wiley, Weinheim, 2009.
[4] G. A. Olah, Angew. Chem. Int. Ed. 2013, 52, 104–107; Angew. Chem.
2013, 125, 112–116.
[5] N. S. Lewis, D. G. Nocera, Proc. Natl. Acad. Sci. USA 2006, 103, 15729–
15735.
[6] W. El-Khattam, M. M. A. Salama, Electr. Power Syst. Res. 2004, 71, 119–
128.
[7] G. Crabtree, J. Sarrao, Phys. World 2009, 22, 24–30.
[8] M. Z. Jacobson, M. A. Delucchi, Energy Policy 2011, 39, 1154–1169.
[9] G. Bozga, I. T. Apan, R. E. Bozga, Recent Pat. Catal. 2013, 2, 68–81.
[10] F. Pontzena, W. Liebner, V. Gronemann, M. Rothaemel, B. Ahlers, Catal.
Today 2011, 171, 242–250.
[11] M. Behrens, F. Studt, I. Kasatkin, S. Kühl, M. Hävecker, F. Abild-Pedersen,
S. Zander, F. Girgsdies, P. Kurr, B. L. Kniep, M. Tovar, R. W. Fischer, J. K.
Nørskov, R. Schlçgl, Science 2012, 336, 893–897.
[12] M. M. Günter, T. Ressler, B. Bems, C. Büscher, T. Genger, O. Hinrichsen, M.
Muhler, R. Schlçgl, Catal. Lett. 2001, 71, 37–44.
[13] M. Behrens, S. Zander, P. Kurr, N. Jacobsen, J. Senker, G. Koch, T. Ressler,
R. W. Fischer, R. Schlçgl, J. Am. Chem. Soc. 2013, 135, 6061–6068.
[14] R. Ladera, F. J. PØrez-Alonso, J. M. Gonzµlez-Carballo, M. Ojeda, S. Rojas,
J. L. G. Fierro, Appl. Catal. B 2013, 142, 241–248.
[15] S. G. Jadhav„ P. D. Vaidya, B. M. Bhanage, J. B. Joshi, Chem. Eng. Res. Des.
2014, 92, 2557–2567.
Figure 4. High-throughput microreactor catalyst screening system. A) Reac-
tor bank with 20 parallel channels (4 banks, total 80 channels) micro-ma-
chined over a ceramic plate; B) Capillary gas-sampling probe; C) Gas analysis
performed by mass spectrometer (MS) or gas chromatograph (GC); D) Wells
for the placement of catalyst pellets or powders; E) Temperature-controlled
heating block with a preheat zone (not shown); F) Insulation. The top right
insert shows the schematic of the side view of a channel reactor: G) Reac-
tant and product gases flow through the channel, which is embedded into
the lower ceramic plate (H) and isolated by the cover plate (I). Gases flow
over the catalyst bed (D) resulting in minimal pressure drop. The capillary
probe (B) is also shown inserted into the channel for sampling.
[16] F. S. Stone, D. Waller, Top. Catal. 2003, 22, 305–318.
ChemCatChem 2016, 8, 1464 – 1469
www.chemcatchem.org
1468
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim

Communications
[17] I. Nakamura, T. Fujitani, T. Uchijima, J. Nakamura, Surf. Sci. 1998, 400,
387–400.
[26] S. Senkan, Angew. Chem. Int. Ed. 2001, 40, 312–329; Angew. Chem.
2001, 113, 322–341.
[18] M. GinØs, N. Amadeo, M. Laborde, C. Apesteguía, Appl. Catal. A 1995,
131, 283–296.
[27] S. Senkan, K. Kevin, O. Sukru, Angew. Chem. Int. Ed. 1999, 38, 2794–
2799; Angew. Chem. 1999, 111, 2965–2971.
[19] M. Spencer, Top. Catal. 1999, 8, 259–266.
[28] J. Toyir, P. R. de La Piscina, J. L. G. Fierro, N. Homs, Appl. Catal. B 2001,
34, 255–266.
[29] D. Andriamasinoro, R. Kieffer, A. Kiennemann, Appl. Catal. A 1993, 106,
201–212.
[30] C. Li, X. Yuan, K. Fujimoto, Appl. Catal. A 2014, 469, 306–311.
[31] L. C. Grabow, M. Mavrikakis, ACS Catal. 2011, 1, 365–384.
[20] Y. Hartadi, D. Widmann, R. J. Behm, ChemSusChem 2015, 8, 456–465.
[21] A. Bansode, A. Urakawa, J. Catal. 2014, 309, 66–70.
[22] F. Studt, I. Sharafutdinov, F. Abild-Pedersen, C. F. Elkjær, J. S. Hummel-
shøj, S. Dahl, I. Chorkendorff, J. K. Nørskov, Nat. Chem. 2014, 6, 320–
324.
[23] J. Graciani, K. Mudiyanselage, F. Xu, A. E. Baber, J. Evans, S. D. Sena-
nayake, D. J. Stacchiola, P. Liu, J. Hrbek, J. F. Sanz, J. A. Rodriguez, Sci-
ence 2014, 345, 546–550.
[24] C. S. Li, G. Melaet, W. T. Ralston, K. An, C. Brooks, Y. Ye, Y.-S. Liu, J. Zhu, J.
Guo, S. Alayoglu, G. A. Somorjai, Nat. Commun. 2015, 6, 6538.
[25] W. H. Chen, B. J. Lin, H. M. Lee, M. H. Huang, Appl. Energy 2012, 98, 92–
101.
Received: January 8, 2016
Revised: February 25, 2016
Published online on March 16, 2016
ChemCatChem 2016, 8, 1464 – 1469
www.chemcatchem.org
1469
ꢀ 2016 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim