Methanol Synthesis at a Wide Range of H2/CO2 Ratios over a Rh-In Bimetallic Catalyst

Article abstract of DOI:10.1002/anie.202000841

There is increasing interest in capturing H₂ generated from renewables with CO₂ to produce methanol. However, renewable hydrogen production is expensive and in limited quantity compared to CO₂. Excess CO₂ and limited H₂ in the feedstock gas is not favorable for CO₂ hydrogenation to methanol, causing low activity and poor methanol selectivity. Now, a class of Rh-In catalysts with optimal adsorption properties to the intermediates of methanol production is presented. The Rh-In catalyst can effectively catalyze methanol synthesis but inhibit the reverse water-gas shift reaction under H₂-deficient gas flow and shows the best competitive methanol productivity under industrially applicable conditions in comparison with reported values. This work demonstrates a strong potential of Rh-In bimetallic composition, from which a convenient methanol synthesis based on flexible feedstock compositions (such as H₂/CO₂ from biomass derivatives) with lower energy cost can be established.

Full text of DOI:10.1002/anie.202000841

A Journal of the Gesellschaft Deutscher Chemiker
Angewandte
Chemie
International Edition
www.angewandte.org
Accepted Article
Title: Methanol synthesis at a wide range of H2/CO2 ratios over Rh-In
bimetallic catalyst
Authors: Molly Meng-Jung Li, Hanbo Zou, Jianwei Zheng, Tai-Sing
Wu, Ting-Shan Chan, Yun-Liang Soo, Xin-Ping Wu, Xueqing
Gong, Tianyi Chen, Kanak Roy, Georg Held, and Edman Shik
Chi Tsang
This manuscript has been accepted after peer review and appears as an
Accepted Article online prior to editing, proofing, and formal publication
of the final Version of Record (VoR). This work is currently citable by
using the Digital Object Identifier (DOI) given below. The VoR will be
published online in Early View as soon as possible and may be different
to this Accepted Article as a result of editing. Readers should obtain
the VoR from the journal website shown below when it is published
to ensure accuracy of information. The authors are responsible for the
content of this Accepted Article.
To be cited as: Angew. Chem. Int. Ed. 10.1002/anie.202000841
Link to VoR: https://doi.org/10.1002/anie.202000841

10.1002/anie.202000841
Angewandte Chemie International Edition
RESEARCH ARTICLE
1
2
Methanol synthesis at a wide range of H2/CO2 ratios over Rh-In
bimetallic catalyst
3
4
Molly Meng-Jung Li†, Hanbo Zou†, Jianwei Zheng†, Tai-Sing Wu, Ting-Shan Chan, Yun-Liang Soo,
Xin-Ping Wu, Xue-Qing Gong, Tianyi Chen, Kanak Roy, Georg Held, Shik Chi Edman Tsang*
5 Abstract: Recent years have seen an increasing interest in capturing 33 no greenhouse gas emission. Although the current low water
6 hydrogen generated from renewables with CO2 to produce methanol. 34 splitting efficiency does not yet justify for the massive launch of
7 However, renewable hydrogen production is currently expensive and 35 this technology, the progressive improvement in electrolyser,
8 in limited quantity as compared to CO2. Excess CO2 and limited H2 in 36 special locations with particular availability of renewable energy
9 the feedstock gas mixture is not favourable for the CO2 hydrogenation 37 sources, and increasing carbon taxation make this new process
10 to methanol reaction, which causes low activity and poor methanol 38 attractive. For safe and efficient transport of hydrogen energy at
11 selectivity. Here we report a new class of Rh-In catalysts with optimal 39 long distance, suitable organic hydrogen carriers are under
12 adsorption property to the intermediates of methanol production. The 40 extensive investigations. From an economic perspective,
13 Rh-In catalyst can effectively catalyse methanol synthesis but inhibit 41 methanol shows high potential as a hydrogen carrier.[2] Therefore,
14 reverse water-gas shift reaction under H2-deficient gas flow and 42 recent years have seen an increasing interest in storing
15 shows the best competitive methanol productivity under industrially 43 renewable hydrogen by reacting it with CO2 to form green
16 applicable conditions in comparison with the literature reported values. 44 methanol and reverse the process to reobtain the H2.[3] In addition,
17 This work demonstrates a strong potential of Rh-In bimetallic 45 catalytic aqueous-phase reforming (APR) is regarded as a
18 composition, from which a convenient methanol synthesis based on 46 promising technology for production of renewable hydrogen (if the
19 flexible feedstock compositions (e.g. H2/CO2 from biomass 47 reaction is powered by renewable energy sources) and soluble
20 derivatives) with lower energy cost can be established.
48 chemicals from biomass-derived substances in aqueous phase
49 under elevated pressure and temperature. It can result in the
50 formation of hydrogen and CO2 as the main products in the gas
51 phase from fragile biomass-derived substances, providing
52 suitable feedstock mixtures for CO2 hydrogenation to methanol
53 hence creating a good opportunity for the effective valorisation of
54 waste biomass to fuel and chemicals.[4] However, the direct
55 utilisation of biomass for methanol production faces the problem
56 of a large excess CO2 in the reforming gas mixture (H2/CO2 < 3).
57 Therefore, the stoichiometric adjustment has to be applied either
58 by adding hydrogen or removing CO2, which requires
59 burdensome equipment and high costs.[5,6] Similarly, the
60 production of hydrogen from other renewable means is rather
61 expensive and is produced in limited quantity as compared to CO2
62 capture.[7] As a result, an effective catalyst to seize hydrogen in
63 the CO2-excess/H2-deficient conditions to catalyse CO2
64 hydrogenation to methanol would be highly desirable.
21 Introduction
22
Hydrogen always plays a critical role as an energy vector in
23 human activity despite the fact that there is almost no free form of
24 naturally occurring hydrogen on earth. At present, most of the
25 world’s hydrogen is derived from carbon-containing fossil fuels, its
26 utilisation leads to concomitantly a huge surge in carbon emission
27 to the atmosphere. On the other hand, hydrogen can also be
28 produced from electrolysis of water using renewable energy such
29 as solar energy, wind power, hydropower, ground heat, or
30 biomass.[1] When hydrogen is combusted or used in a fuel cell
31 elsewhere, the major by-product is water again, thus completing
32 the truly circular economy with energy storage and dispatch with
65
It is known that bimetallic nanoparticles/alloys with intimate
66 contacts of the two elements can modify the electronic properties
67 of the constituent metals, thus change their adsorption
68 properties.[8,9] A good example can be found in the Cu-Zn system,
69 which shows that the Zn-modified Cu surface gives better
70 methanol production rates than the unmodified Cu surface
71 because the Zn-modified Cu surface has a stronger binding of
72 intermediates and lower energetic barriers to the methanol
73 product.[10–12] Although the adsorption property of the Cu surface
74 can be improved by modifying with Zn species, the Cu surface still
75 possesses the drawbacks on the low activity for hydrogen
76 activation, which leads to low coverage of surface H and slows
77 down the further hydrogenation of the intermediates into
78 methanol.[8] Consequently, for the Cu-based catalysts, high
79 methanol selectivity commonly requires an extreme reaction
80 condition (high pressure of over 10 MPa, high ratios of H2:CO2 ≥
81 3), otherwise CO is favourably produced through the reverse
82 water-gas shift reaction (RWGS) route.[13] In addition, it has been
83 reported that methanol selectivity of the Cu-based catalysts is
84 limited according to thermodynamic calculations,[8] which leads to
[*] Dr. M. M.-J Li[†], Dr. H. Zou[†], Dr. J. Zheng[†], T. Chen, Prof. S. C. E. Tsang*
Department of Chemistry, University of Oxford, OX1 3QR, UK.
E-mail: edman.tsang@chem.ox.ac.uk
Dr. M. M-J Li
Department of Applied Physics, Hong Kong Polytechnic University, Hong Kong
Dr. H. Zou
Department of Chemistry and Chemical Engineering, Guangzhou
University, China.
Dr. T.-S. Wu, Prof. Y.-L. Soo
National Synchrotron Radiation Research Center, Hsinchu, Taiwan and
Department of Physics, National Tsing Hua University, Hsinchu, Taiwan.
Dr. T.-S. Chan
National Synchrotron Radiation Research Center, Hsinchu, Taiwan.
Dr. X.-P. Wu, Prof. X.-Q. Gong
Key Laboratory for Advanced Materials, Centre for Computational
Chemistry and Research Institute of Industrial Catalysis, East China
University of Science and Technology, Shanghai 200237, People’s
Republic of China
Dr. K. Roy, Prof. G. Held
Diamond Light Source, Harwell Campus, Chilton, Oxfordshire OX11 0DE, UK
[†] These authors contributed equally to this work.
Supporting information for this article is given via a link at the end of the
document.
This article is protected by copyright. All rights reserved.

10.1002/anie.202000841
Angewandte Chemie International Edition
RESEARCH ARTICLE
1 significant CO production through the RWGS. Therefore, non-Cu 56 a higher reaction temperature of 330 oC. This suggests that the In
2 based catalysts for effectively utilising renewable hydrogen from 57 modification to Ru sites is not as effective as the Rh sites, which
3 renewables-derived feedstocks (e.g. biomass) to green methanol 58 could be attributed to the ineffective d-band modification of the
4 production are needed to accomplish this development.
59 electron poorer Ru by In to alter their intrinsically strong
5
Here we report a novel Rh-In bimetallic catalyst for the first 60 adsorptive and catalytic properties. Alternatively, In modification
6 time that shows very effective usage of H2 toward methanol 61 to Ru may be extensively carried out using other synthesis
7 production. Rh-In bimetallic catalyst not only shows the best 62 processes to increase the interactions between Ru and In species
8 competitive methanol productivity per gram basis under 63 to improve its effectiveness. This requires more experimental
9 industrially applicable conditions in comparison with the literature- 64 studies.
10 reported values but also maintains high H2 conversion to 65
It is noted that CO2 conversion drops drastically when
11 methanol under H2-deficient gas flow. This new catalyst displays 66 incorporating In into Rh (Fig. 1a), however, a decent CO2
12 its advantages in efficient H2 utilisation with the undesired RWGS 67 conversion (>10%) can be achieved when using In-modified Rh
13 reaction, i.e. CO production, being minimised under flexible 68 catalyst prepared by the co-precipitation method, denoted as co-
14 feedstock compositions.
69 RhIn/(5In5Al)O (see Fig. 1c). The specific surface areas (3.2 ± 0.1
70 m2g-1 for im-RhIn/(5In5Al)O and 279.5 ± 0.2 m2g-1 for co-
71 RhIn/(5In5Al)O) determined by BET as well as the Rh surface
-1
72 areas (17.3 m2g-Rh for im-RhIn/(5In5Al)O and 132.6 m2g-Rh-1 for
15 Results and Discussion
73 co-RhIn/(5In5Al)O) determined by hydrogen/oxygen titration
74 reveal that the co-precipitation method can increase the surface
75 exposure of Rh. TEM images and particle size distribution
76 diagrams (Fig. S3) also suggest that the metal nanoparticles
16
In this study, a series of Rh-containing samples with different
17 In/Al compositions have been synthesised (see Table 1 for
18 sample names and preparation details) and assessed in the CO2
19 hydrogenation reaction. As can be seen from Fig. 1a, at the same
20 reaction temperature of 270 oC, the wet-impregnation samples
21 with In/Al ratios from 0 to 1 give a diverse product composition:
22 The sample contains no In (im-Rh/Al2O3) shows a total conversion
23 of CO2 to methane, while CO is predominant in the product
24 composition when im-RhIn/(1In9Al)O is used. Methanol initially
25 emerges in im-RhIn/(1In9Al)O and becomes substantial (>85%
26 selectivity) in im-RhIn/(5In5Al)O and im-RhIn/In2O3. For the
27 typical partial reduction reaction like CO2 hydrogenation to
28 methanol instead of CH4, methanol selectivity normally has an
29 inverse relationship with CO2 conversion. Therefore the catalysts
30 were also assessed and compared under the same CO2
31 conversion of 1%, where kinetic plug-flow conditions (far from
32 equilibrium) are ensured. The result in Fig. 1b shows the same
33 trend of increasing methanol selectivity as In concentration
34 increases: Methanol selectivity is peaked at the im-RhIn(5In5Al)O
35 catalyst and then starts to decrease. These results suggest that
36 In/Al=1 is the optimal support composition for the Rh catalysts,
37 and the presence of In near Rh can significantly alter the catalytic
38 properties of Rh sites from methanation (In/Al=0) toward RWGS
39 reaction (In/Al=1/9), and finally to methanol production (In/Al>1).
40 Such an effective modification of Rh has not been reported in the
41 literature. Although, when doping with alkaline earth metals[14] or
42 alloying with Co,[15] Rh-based catalysts can have a slight
43 enhancement on CO/CO2 hydrogenation to methanol. However,
44 the yield and selectivity toward methanol production in those early
45 works were much lower than the In-modified Rh catalysts in our
46 study. On the other hand, the In2O3-Al2O3 sample only gives 30%
47 methanol selectivity with a low CO2 conversion under the same
48 testing condition (Fig. 1a&b). This indicates that In2O3 does not
49 contribute to high methanol production due to its limited H2-
77 prepared by the co-precipitation method indeed have
a
78 significantly smaller particle size (1 ~ 5 nm) compared to the wet-
79 impregnation sample (>20 nm). In correlation with the catalytic
80 performance (Fig. 1c), both im-RhIn/(5In5Al)O and co-
81 RhIn/(5In5Al)O catalysts show a good selectivity to methanol.
82 However, the activity of im-RhIn/(5In5Al)O is much lower. The
83 larger particle size of im-RhIn/(5In5Al)O can reduce the number
84 of the exposed active sites, therefore causing a lower CO2
85 conversion. Note that methanol selectivity does not change along
86 with the metal particle size variation, implying that selectivity is not
87 critically size-dependent in this In-modified Rh catalyst system.
88
Fig. 1d shows methanol selectivities of the best-performing
89 catalyst, co-RhIn/(5In5Al)O, assessed in CO2 hydrogenation with
90 different CO2/H2 ratios (from 1/3 to 3). commercial
A
91 Cu/ZnO/Al2O3 catalyst is also evaluated for the comparison. In
92 CO2 hydrogenation reaction to methanol, the use of gas mixture
93 with H2/CO2
<
3
is not thermodynamically favourable.
94 Understandably, the practical methanol selectivities obtained on
95 the Cu/ZnO/Al2O3 catalyst continuously drop when increasing
96 CO2/H2 ratios from 1/3 to 3. Fig. S5a shows that the methanol
97 yields from both co-RhIn/(5In5Al)O and Cu/ZnO/Al2O3 catalysts
98 are lower than the calculated thermodynamic equilibrium values
99 with unconverted reactants when taken both methanol production
100 and RWGS equilibria into account. The data clearly imply that the
101 catalysed reaction is under kinetic control. However, for the co-
102 RhIn/(5In5Al)O catalyst, especially at the excess CO2 conditions,
103 methanol selectivities remain at higher values, which are even
104 higher than the thermodynamic predicted selectivities, assuming
105 both methanol production and RWGS reached equilibria
106 (indicated by the dashed/dotted lines in Fig. 1d). Note that co-
107 RhIn/(5In5Al)O can attain higher methanol yields than that of the
50 splitting ability.[16] An In-modified Ru counterpart, im-Ru/(5In5Al)O, 108 thermodynamic prediction from CO
to methanol and RWGS
2
109 when H2/CO2 is lower than 0.5 (H2-deficient). These results
51 has also been evaluated (Fig. 1c). Although the increases in
52 methanol selectivities can be observed upon adding In into
53 Ru/Al2O3 catalyst, the methanation of CO2 over the im-
54 Ru/(5In5Al)O sample cannot be excluded as a noticeable amount
55 of methane still appears in its product composition, especially at
110 suggest that the RWGS reaction can be effectively suppressed
111 on the In-modified Rh catalyst, particularly under low H2/CO2
112 conditions, hence minimising CO production on this catalyst
113 (details will be discussed later along with the DFT calculation).
114 Considering that H2 is more valuable than CO2, minimising H2
This article is protected by copyright. All rights reserved.

10.1002/anie.202000841
Angewandte Chemie International Edition
RESEARCH ARTICLE
1 consumption in producing unwanted products (i.e. CH4, H2O, etc.)
2 is very important. Fig. 1e presents the H2 conversion (added up
3 by the H2 consumptions in methanol production and the RWGS
4 reaction) of both co-RhIn/(5In5Al)O and Cu/ZnO/Al2O3. Although
5 Cu/ZnO/Al2O3 shows a slightly higher H2 conversion than co-
6 RhIn/(5In5Al)O, its majority of H2 conversion is attributed to
7 RWGS reaction, that produces CO and H2O. On the other hand,
8 co-RhIn/(5In5Al)O shows a very efficient usage of H2 toward
9 methanol production, and it can maintain high H2 conversion to
10 methanol even under H2-deficient conditions (H2/CO2 < 3). This
11 result has proven that co-RhIn/(5In5Al)O catalyst can efficiently
12 seize hydrogen in the H2-deficient environment and selectively
13 catalyse CO2 hydrogenation to methanol without consuming a
14 large quantity of valuable H2 on the RWGS reaction.
-
15
Fig. 1f gives the methanol space-time yields (STY, gMeOH∙gcat
1
16 ∙h-1) and methanol selectivities of co-RhIn/(5In5Al) and
17 Cu/ZnO/Al2O3. Noticeably, the methanol selectivity and STY of
18 Cu/ZnO/Al2O3 are significantly less than those of co-RhIn/(5In5Al).
19 Moreover, methanol selectivity and STY of co-RhIn/(5In5Al) can
20 be further optimised by adjusting the weight hourly space velocity
21 (WHSV) to reach nearly 100% and over 1.0 gMeOH∙gcat-1∙h-1,
22 respectively. As far as we are aware, STY of 1.0 gMeOH∙gcat-1∙h-1
23 obtained by co-RhIn/(5In5Al)O at CO2/H2 = 1/3 condition is among
24 the highest values compared to the state-of-the-art catalysts in
25 the literature.[8,9,16–28] The comparison of co-RhIn/(5In5Al) to the
26 traditional Cu-based catalysts and the state-of-the-art catalysts is
27 presented in Section 4 of the supporting information (SI) and
28 Table S1.
44
29
Time-on-stream (TOS) test was also performed to evaluate
30 the stability of the co-RhIn/(5In5Al) catalyst. Fig.S5b shows that
31 co-RhIn/(5In5Al)O gives consistently high methanol selectivity
32 and CO2 conversion for at least 10 days in our academic
33 laboratory reactor. Although a longer TOS test should be
34 vigorously studied at a large scale to determine its suitability for
35 industrial applications, the Rh-In catalyst appears to be stable
36 during the CO2 hydrogenation test in our laboratory.
37
45 Figure 1. Catalytic performance of CO2 hydrogenation to methanol.
46 Typical testing conditions are the pressure of 45 bar, reactant mixture of
47 CO2/H2=1/3, reaction temperature of 270 oC, WHSV of 18000 mL g-1 h-1 unless
48 otherwise indicated. Error bars in the figures indicate the standard deviation of
49 at least 3 repeated data points taken for each experiment. a) CO2 conversions
50 and the selectivities of CO, CH3OH, and CH4 of Rh catalysts with different In/Al
51 ratios prepared via the wet-impregnation method. b) Methanol selectivities of
52 Rh catalysts prepared by wet-impregnation methods when evaluated under the
53 same CO2 conversion of 1% (with the typical standard deviations lower than
54 0.1%). Reaction temperatures to achieve 1% CO2 conversion are stated for
55 each catalyst. c) CO2 conversions and the selectivities of CO, CH3OH, and CH4
56 of Rh and Ru catalysts with different In/Al ratios prepared via wet-impregnation
57 and co-precipitation methods. d) Methanol selectivities at 250 oC and 270 oC of
58 co-RhIn/(5In5Al)O sample compared with the commercial Cu/ZnO/Al2O3
59 catalyst under different CO2/H2 ratios. The dashed/dotted lines indicate the
60 calculated methanol selectivities by taking both methanol synthesis and RWGS
61 equilibria into account. The decreasing trend for both catalysts is according to
62 the thermodynamic limits when a higher CO2/H2 ratio is employed. e) H2
63 conversion toward methanol production (green) and RWGS reaction (blue) of
64 co-RhIn/(5In5Al)O compared with Cu/ZnO/Al2O3 catalyst under different CO2/H2
65 ratios. f) Methanol space-time yields and selectivities of co-RhIn/(5In5Al)O and
66 Cu/ZnO/Al2O3 under different WHSV.
38
39 Table 1 The details of the synthesis method, metal loading, and In/Al ratio of all
40 catalysts presented in this work.
Rh or Ru loading &
method
In : Al ratio of
In2O3-Al2O3 support
Sample name
im-Rh/Al2O3
im-RhIn/(1In9Al)O
im-RhIn/(3In7Al)O
im-RhIn/(5In5Al)O
im-RhIn/In2O3
5%, wet-impregnation
5%, wet-impregnation
5%, wet-impregnation
5%, wet-impregnation
5%, wet-impregnation
No Rh loading
0 : 10
1 : 9
3 : 7
5 : 5
10 : 0
5 : 5
5 : 5
5 : 5
0 : 10
In2O3-Al2O3
im-Ru/(5In5Al)O
co-RhIn/(5In5Al)O
co-Rh/Al2O3
5%, wet-impregnation
2.5%, co-precipitation
2.5%, co-precipitation
67
To realise the impact of In addition to Rh catalysts, the
41
42
43
68 structural investigation was conducted. From TEM images in Fig.
69 S3, we can see that the reduced co-Rh/Al2O3 has a similar but
70 slightly smaller Rh containing particle size than the reduced co-
71 RhIn/(5In5Al)O. BET analysis of these two samples shows that
72 both have high surface areas (279.5 m²/g for co-RhIn/(5In5Al)O
73 and 327.8 ± 0.5 m²/g for co-Rh/Al2O3). These results imply that In
74 addition does not have a huge impact on the size of Rh
75 nanoparticles but its presence can significantly alter the product
This article is protected by copyright. All rights reserved.

10.1002/anie.202000841
Angewandte Chemie International Edition
RESEARCH ARTICLE
1 composition of CO2 hydrogenation, hence the change of catalytic 60 high-temperature TPR peak may also stem from the strong
2 performance upon varying In/Al ratio is not originated from Rh size 61 interaction of Rh species with the Al2O3 support.[31] As In
3 variation. HR-TEM images in Fig. 2a and Fig. S4a reveal that 62 concentration increases, (from In/Al=0 to In/Al=3/7), the low-
4 small islands of the two most stable crystalline intermediate 63 temperature Rh reduction peak increases with the disappearance
5 phases of the Rh-In system (cubic-RhIn and tetragonal-RhIn3) at 64 of the high-temperature reduction peak, indicating the interaction
6 the material interfaces were observed.[29] TEM of the co- 65 between Rh and Al2O3 support has been diminished, similar to
7 RhIn/(5In5Al)O catalyst retrieved at the end of the long TOS test 66 what was observed in the Rh K-edge EXAFS analysis.
8 also shows that both cubic-RhIn and tetragonal-RhIn3 clusters 67 Interestingly, for the catalysts having a composition of In/Al ≥ 1,
9 embedded with the variable RhInx composition at the metal- 68 the Rh reduction peak flattens and shifts to a higher temperature,
10 support interfaces with no observable particle sintering in the 69 which is indicative of the increased metal-metal interaction of Rh
11 post-reaction catalyst (Fig. S4b), suggesting that Rh-In alloys are 70 and In when In concentration has increased.[31]
12 robust to maintain throughout the CO2 hydrogenation mixture. On 71
13 the other hand, the main light-contrast particles in the vicinity of
14 darker Rh-In alloy clusters in the TEM images are confirmed to be
15 hexagonal In2O3 (Fig. S4 a&b). This observation is in agreement
16 with the XRD patterns in Fig. S4c that the hexagonal In2O3 phase
17 is the predominant structure in the co-RhIn/(5In5Al)O sample
18 either before or after catalytic CO2 hydrogenation testing. We
19 cannot detect any Rh-containing phases from XRD patterns due
20 to either the low Rh metal loading (~2.5 wt.%) or the diffraction-
21 line broadening caused by small particle sizes, and the absence
22 of aluminium oxide phase indicates that it is amorphous in nature.
23
Analysis of the EXAFS at both Rh and In K-edge was
24 employed to provide detailed information on the local atomic
25 structure of Rh and In atoms. Fig. 2b shows Fourier-transform Rh
26 K-edge EXAFS of the reduced Rh-containing catalysts. The k3-
27 weighted EXAFS (K space) with the corresponding fittings are
28 shown in Fig. S6, and the fitting parameters are detailed in Table
29 S2. It can be seen from Fig. 2b that in general, each Rh-
30 containing sample has two main shells, that is, an Rh-O shell at
31 around 2Å and the Rh-metal shell at the longer distance. Since all
32 the samples were pre-reduced in H2 and carefully handled in
33 oxygen-free conditions prior to the XAS measurements, the Rh-O
34 scattering path detected in EXAFS spectra is most likely attributed
35 to the interaction between Rh and the oxygen from the oxide
36 support. For the im-Rh/Al2O3 sample, the observed Rh-O peak is
37 higher than the Rh-Rh bond, indicating that Rh nanoparticles have 72
38 strong interaction with the Al2O3 support. As In concentration 73 Figure 2. Structure and the reduction behavior
39 increases, the coordination number (C.N.) of Rh-O decreases as 74 a) HR-TEM images with the fast-Fourier Transform (FFT) analyses of the
40 well as the distance of the Rh-Rh scattering path at around 2.66- 75 selected Rh-In alloy nanoparticles and the measured d-spacings corresponding
41 2.70 Å increases, suggesting a weakening interaction between 76 to the hexagonal In2O3 phase in the freshly reduced co-RhIn/(5In5Al)O catalyst.
42 Rh atoms and the oxide support while strengthening the 77 b) k3-weighted Rh K-edge EXAFS Fourier transforms of the reduced Rh and In-
43 interaction between Rh and the neighbouring In species. Besides, 78 modified Rh samples. c) H2-TPR profiles of the Rh, In-modified Rh catalysts,
44 in the EXAFS spectra of the catalysts with the optimal support 79 and In2O3-Al2O3 support.
45 composition (In/Al=1), i.e., im-RhIn/(5In5Al)O and co- 80
46 RhIn/(5In5Al)O, there are three distinctive scattering paths 81
The electronic properties of the catalysts were also
47 observed at the distances of 2.64, 2.82 and 3.07 Å, that are 82 investigated using XANES and XPS. Rh K-edge XANES spectra
48 attributed to Rh-In, Rh-In and Rh-Rh bonds, respectively, 83 (Fig. 3a) reveal that the catalysts containing the support
49 evidencing the formation of Rh-In alloy structure. The im- 84 compositions of In/Al ≥ 1 have similar Rh absorption edges that
50 RhIn/In2O3 sample gives the highest Rh-In alloy content among 85 are comparable to the absorption edge of Rh foil. As for the
51 all the Rh-containing samples due to its highest C.N. of Rh-In 86 samples with less In concentrations (In/Al < 1), their Rh
52 scattering paths. Besides, im-RhIn/In2O3 contains no Rh-O, 87 absorption edge positions progressively shift toward high energy
53 showing the fact that the interaction between Rh and the oxygen 88 along with the decrease of In/Al, which is indicative of higher
54 from In2O3 is negligible.
89 oxidation state of Rh. A similar trend of binding energy shift can
55
Fig. 2c gives H2-TPR profiles of the Rh-containing catalysts 90 also be observed from Rh 3d XPS spectra (Fig. 3b). This again
56 and the In2O3-Al2O3 catalyst. For the im-Rh/Al2O3 sample, the 91 reveals that the interaction between Rh and oxygen from the
57 reduction peaks at 135 and 248 oC are attributed to the well- 92 oxide support could be minimised upon In incorporation. Besides,
58 dispersed surface Rh2O3 and the bulk-like/crystalline Rh2O3 93 the alteration of the electronic property of Rh also signifies that
59 particles, respectively,[30] According to the published data, the 94 charge transfer can occur from In to Rh when forming Rh-In alloy,
This article is protected by copyright. All rights reserved.

10.1002/anie.202000841
Angewandte Chemie International Edition
RESEARCH ARTICLE
1 which is in agreement with the thermodynamic observation 54
2 reported in the literature.[29] In contrast, In 3d XPS spectra (Fig.
3 3b) of the reduced samples show no distinguishable shift upon In
4 incorporating into Rh catalysts due to the fact that most In species
5 stay as oxidic and only limited In2O3 reduction could prompt the
6 decoration of In on Rh, as confirmed by XRD (Fig. S4c) and In L3-
7 edge XANES (Fig. S6b) showing that In mainly retains as the
8 In2O3 phase no matter before or after CO2 hydrogenation reaction.
9
Synchrotron-based
near
ambient-pressure
X-ray
10 photoelectron spectroscopy (NAP-XPS) was then employed to
11 study the in-situ change of the chemical states in the co-
12 RhIn/(5In5Al)O under catalytic working conditions. Fig. 3c shows
13 Rh 3d and In 3d NAP-XPS spectra recorded in situ during Ar, H2,
14 and CO2/H2 atmospheres at 290 oC. Our experimental determined
15 binding energies (BE) of Rh 3d5/2 peaks are comparable to the
16 published data (Rh3+ 3d5/2 = 308.4, Rh0 3d5/2 = 307.2).[32] The Rh
17 3d spectra (i) and (ii) provide a clear indication of the rapid and
18 complete Rh3+ reduction to form Rh0. From spectra (ii) to (iv), the
19 Rh 3d5/2 BE keeps decreasing beyond the metallic Rh0 3d5/2 value
20 (307.4 eV) with further exposure to H2 for a longer period of time,
21 which indicates that the Rh species become electron richer. As
22 for the In 3d NAP-XPS spectra, two 3d5/2 and two 3d3/2 peaks can
23 be found in each spectrum taken under different gas conditions.
24 According to the literature, BE of In3+ 3d5/2 is around 444.8 eV,
55
56
57
58
59
60
61
62
63
64
65
66
67
[33,34]
25 whereas BE of In0 3d5/2 is ranging from 442.8 to 443.7 eV.
26 Our experimentally determined In 3d region shows two sets of the 68 Figure 3. Electronic structure of the Rh-In catalyst
27 well-separated 3d5/2 and 3d3/2 orbitals both having the typical 69 a) Normalised Rh K-edge XANES spectra and b) XPS Rh 3d and In 3d spectra
28 energy difference of spin-orbit components (∆=7.6ꢀeV) of In. In 70 of the reduced Rh and In-modified Rh samples. c) NAP-XPS of Rh 3d and In 3d
29 addition, their BEs match the literature-reported values of metallic 71 as a function of different gas atmospheres at 290 oC.
30 In and oxidised In species. As a result, the 3d5/2 and 3d3/2 peaks 72
31 at lower BEs (ca. 443 eV and 450 eV, respectively) can be 73
32 identified as metallic In while the higher-BE In 3d5/2 and 3d3/2 74
It is well-known that the foreign metal atom
33 peaks (ca. 445 eV and 452.6 eV, respectively) are attributed to 75 additives/impurities in metal catalysts can induce the formation of
34 In2O3. From spectrum (i), it can be seen that In shows some 76 bimetallic nanoparticles/alloys, which would significantly modify
35 degree of reduction due to the high-temperature treatment in Ar 77 the electronic configurations hence altering the adsorption
36 atmosphere. When the gas feed is changed to H2, the In0 3d5/2 78 properties of the original metal sites and dictating product
37 signal starts to increase and shifts slightly to higher BE (from 79 specificity. Many examples have been recently reported including
38 442.6 to 442.9 eV), which falls simultaneously in line with the shift 80 the Cu-Zn,[10–12] Pd-Zn,[8,9] Pd-Ga, Pd-In,[35] Ni-Ga[18] bimetallic
39 of Rh0 3d5/2 to lower BE. This offers the evidence of charge 81 catalysts for the enhanced CO/CO2 hydrogenation to methanol.
40 transfer from In to Rh due to the formation of local alloys at the 82 According to theoretical calculations,[18,36–39] catalytic activity of
41 materials‘ interfaces. In the 2nd H2 exposure, we can see from 83 CO2 hydrogenation to methanol is critically dependent on the
42 spectrum (iii) that the intensity of In0 3d5/2 peak further increases 84 overall adsorptivity of catalytic surfaces. In earlier studies on the
43 with its BE shifting to 443.3 eV. This clearly suggests that the 85 reaction pathways of CO2 hydrogenation to methanol, there are
44 prolonged H2 exposure will result in a deeper reduction of support 86 two proposed parallel pathways: One is through a formate
45 so that more In0 species can react with the vicinity Rh0 to form 87 (HCOO*) intermediate without CO formation; the other involves a
46 more extensively In-decorated Rh, leading to two stable Rh-In 88 hydrocarboxyl (COOH*) intermediate, through which CO2 is first
47 alloy phases observed. In the 3rd H2 exposure, see spectrum (iv), 89 converted to CO (RWGS route) and then CO can be further
48 In0 3d5/2 signal stays unchanged compared with spectrum (iii), 90 hydrogenated to methanol.[40,41] The catalytic performance and
49 suggesting the establishment of stable Rh-In alloys at the 91 the intermediates formed during reactions have been
50 interface. When switching the H2 gas to CO2/H2, In0 3d5/2 signal 92 demonstrated to be dependent and sensitive to the composition
51 still exists but the intensity decreases with the reduction of BE, 93 of catalysts. For the industrial Cu/ZnO/Al2O3 catalyst composition,
52 suggesting that CO2 in the reaction mixture can act as an oxidant 94 the formation of electron richer metal, i.e. Zn, tends to stabilise
53 to gently oxidise Rh-In alloys.
95 more to the formate intermediate, however, the role(s) and
96 interactions of each component in this catalyst are still in
97 debate.[10,42]
98
Density functional theory (DFT) calculations were therefore
99 carried out to simulate the change of adsorption property of Rh
100 surface upon incorporating In species from the local In2O3
101 reduction. To be more specific, the competitive routes of HCOO*
This article is protected by copyright. All rights reserved.

10.1002/anie.202000841
Angewandte Chemie International Edition
RESEARCH ARTICLE
1 and COOH* on Rh and bimetallic Rh-In surfaces will be 60 compared to Rh-In. At the H2-deficient conditions, low coverage
2 appreciated and compared with the Cu-Zn surfaces symbolic of 61 of would inhibit the further hydrogenation of HCOO*.
H
3 the commercial Cu/ZnO/Al2O3 catalyst. The effect of In2O3-Al2O3 62 Nevertheless, the formation of CO from COOH* is less dependent
4 support was not considered in our DFT calculations as our 63 on the surface H coverage.[8] Therefore, although the formation of
5 experimental results clearly show that the support does not 64 COOH* is thermodynamically less favourable to take place than
6 contribute to high methanol production. In contrast, the bimetallic 65 that of the parallel route of HCOO*, COOH* will kinetically be
7 nature of the catalyst and the strong synergy between bimetallic 66 rapidly consumed by converting it into CO, giving rise to the higher
8 Rh and In presented in the above X-ray spectroscopic analysis 67 rate of RWGS reaction. In contrast, the Rh-based surfaces show
9 are believed to contribute to the unprecedentedly high methanol 68 superior ability in activating H2 as indicated by their strongly
10 yield in CO2 hydrogenation. From TEM (Fig. 2a), small islands of 69 negative H adsorption energies. In principle, this indicates that Rh
11 two most stable intermediate phases of Rh-In system (RhIn and 70 and Rh-In sites can readily seize H2 and proceed with
12 RhIn3) at the materials‘ interfaces were observed, suggesting that 71 hydrogenation of the preferentially formed HCOO* under H2-
13 limited In reduced from In2O3 and decorated on exposed surfaces 72 deficient conditions. Moreover, a heavily In-doped Rh surface (i.e.,
14 of Rh nanoparticle would result in
a local Rh-In alloy 73 RhIn(211)_b model surface and our high-performing Rh-In
15 composition.[29] Therefore, widely accepted surface bimetallic 74 bimetallic catalyst) with a moderate H adsorption energy provides
16 alloy models were used in our DFT calculations. Considering that 75 active sites for an efficient hydrogen utilisation with enhanced
17 the size of the majority Rh-In nanoparticles in co-RhIn/(5In5Al)O 76 methanol selectivity/yield.
18 catalyst is 1-3 nm, in which the surface defects can have a 77
The difference of the surface adsorption properties of Rh, Rh-
19 profound effect toward the catalytic properties, we performed DFT 78 In, and Cu-Zn surfaces, as demonstrated in the DFT result, were
20 calculations on a representative stepped surface, i.e., Rh(211) 79 then verified by the experimental surface adsorption analysis. The
21 and two In-doped Rh(211) surfaces with 50% or 100% of the step- 80 in-situ FTIR spectra in Fig. 4a indeed show that different forms of
22 edge Rh sites substituted by In, denoted as RhIn(211)_a and 81 adsorbed CO species and formates are preferably formed on co-
23 RhIn(211)_b, respectively, to elucidate the effect of In doping (see 82 Rh/Al2O3 and co-RhIn/(5In5Al)O surfaces, respectively, whereas
24 Computational Details and Fig. S7 in the SI). In addition, to 83 RWGS products (CO and H2O) are found in Cu/ZnO/Al2O3 (further
25 understand the superior catalytic performance of the Rh-In 84 details are shown in Section 7 of the SI). From DFT result, it is
26 bimetallic catalyst to the Cu-Zn bimetallic catalyst, we also 85 important to have appropriate adsorptivity for both CO2 and H2 on
27 performed calculations on Cu(211) and two CuZn(211) surfaces 86 the active sites for efficient conversion of the two molecules to
28 resembling the RhIn(211) counterparts, denoted as CuZn(211)_a 87 methanol: Too strong adsorption for either molecule may cause a
29 and CuZn(211)_b, respectively. In line with the Rh 3d XPS and 88 change in selectivity due to side reactions. According to the pulse
30 Rh K-edge XANES results (Fig. 3), Bader charge analysis (Table 89 experiment result in Fig. 4b, it confirms that co-Rh(5In5Al)O
31 S4), of the surface Rh(211) with different degrees of In 90 contains active adsorption sites for both CO2 and H2, hence giving
32 incorporation confirms that charge transfer from In to Rh will occur 91 rise to excellent methanol production. In contrast, Cu/ZnO/Al2O3
33 upon Rh-In alloy formation. Table 2 summarises the hydrogen 92 and In2O3-Al2O3 show a significantly lower H2 uptake (low
34 adsorption energies, as well as the energy differences of surface 93 hydrogenation activity), therefore, the RWGS reaction becomes
35 adsorbed HCOO* and COOH* (Ediff = EHCOO* – ECOOH*) on the 94 more dominant than the extensive CO2 hydrogenation to
36 selected model surfaces. For the Rh (211) surface, electron- 95 methanol, leading to more CO production. On the other hand,
37 donating In significantly strengthens the relative stability of 96 excellent methanation activity of the co-Rh/Al2O3 catalyst is likely
38 HCOO*, especially for the Rh (211) surface heavily substituted by 97 to stem from its higher intrinsic adsorption capacity of Rh to H2
39 In (denoted as RhIn(211)_b) as anticipated to our bimetallic Rh- 98 over CO2 adsorption. These experimental surface adsorption
40 In catalyst. Interestingly, RhIn(211)_b and CuZn(211)_b surfaces 99 results can explain the alteration of Rh adsorption property by the
41 have similar strongly negative Ediff (-1.20 eV and -1.11 eV, 100 presence of In, and can also demonstrate the reason for the much
42 respectively), indicating the formation of HCOO* is energetically 101 lower methanol selectivity on Cu/ZnO/Al2O3 catalysts than the Rh-
43 more favourable than that of COOH* on both bimetallic alloy 102 In catalyst, under low H2/CO2 conditions in particular.
44 cases. Behrens et al. have reported a similar stabilisation effect 103
45 of adsorbed HCOO* by substituting Cu step sites with the electron 104
46 richer Zn atoms, and this effect is essential for the direct methanol 105 Table 2. Calculated energy differences of surface adsorbed HCOO* and
47 formation.[10] On the other hand, sufficient hydrogen coverage on 106 COOH* species and H adsorption energies at various surfaces.
48 the catalyst surface is an equally important prerequisite to partially 107
49 hydrogenate the HCOO* to methanol or fully to methane in the
[a]
[b]
Model surface
Ediff
Eads
50 CO2 hydrogenation reaction. H adsorption energy (Eads) was then
51 calculated. It corresponds to the difference between the DFT
52 energies of the H adsorbed on the model surface (adsorption
53 complex) and the sum of the clean surface and the gas-phase H2
54 molecule, from which a more negative Eads indicates H species is
55 more favorable to be adsorbed on the surface. The result in Table
56 2 clearly shows that H adsorption on a Cu-Zn surface is
57 energetically favoured but much weaker than that on the Rh-In
58 surface, indicating a lower coverage of H and hence a lower
59 possibility of successful H2 dissociation (activation) on Cu-Zn
Rh(211)
RhIn(211)_a
RhIn(211)_b
Cu(211)
-0.25
-0.37
-1.20
-0.92
-0.90
-1.11
-0.59
-0.56
-0.35
-0.21
-0.21
-0.03
CuZn(211)_a
CuZn(211)_b
108 [a] Ediff (eV): Energy differences of surface adsorbed HCOO* and COOH* species.
109 [b] Eads (eV): H adsorption energies
110
This article is protected by copyright. All rights reserved.

10.1002/anie.202000841
Angewandte Chemie International Edition
RESEARCH ARTICLE
1
43 Keywords: Carbon dioxide •Hydrogenation• Methanol synthesis
44 • various H2/CO2• Rh-In Bimetallic catalyst
2
45
46 [1]
47
J. Turner, G. Sverdrup, M. K. Mann, P.-C. Maness, B. Kroposki, M.
Ghirardi, R. J. Evans, D. Blake, Int. J. Energy Res. 2008, 32, 379–
407.
48
49 [2]
50
M. Niermann, S. Drünert, M. Kaltschmitt, K. Bonhoff, Energy
Environ. Sci. 2019, 12, 290–307.
51 [3]
52 [4]
53
G. A. Olah, Angew. Chemie Int. Ed. 2005, 44, 2636–2639.
R. D. Cortright, R. R. Davda, J. A. Dumesic, Nature 2002, 418, 964–
967.
3
4
5
6
54 [5]
55
M. Specht, A. Bandi, F. Baumgart, C. N. Murray, J. Gretz, Greenh.
Gas Control Technol. 1999, 723, 1–6.
7
8
56 [6]
57
X. Yin, D. Y. C. Leung, J. Chang, J. Wang, Y. Fu, C. Wu, Energy &
Fuels 2005, 19, 305–310.
9 Figure 4. Surface adsorption properties.
10 a) In-situ FTIR spectra of the adsorbed species on the reduced catalysts. A gas
11 flow of 25% CO2 and 75% H2 is passed through the catalyst pellets made by 20
12 mg of samples at various temperatures. b) CO2 (left) and H2 (right) uptakes per
13 gram of catalyst derived from CO2 and H2 pulse experiments at 50 oC. RhInAl:
14 co-RhIn/(5In5Al)O; RhAl: co-Rh/Al2O3; InAl: In2O3-Al2O3; CuZnAl: commercial
15 Cu/ZnO/Al2O3. An unmodified Cu nanoparticles sample, which shows very poor
16 adsorption toward both CO2 and H2, is presented as a comparison. Error bars
17 indicate the standard deviation of 3 repeated data points taken for each
18 experiment.
58 [7]
59
M. Götz, J. Lefebvre, F. Mörs, A. McDaniel Koch, F. Graf, S. Bajohr,
R. Reimert, T. Kolb, Renew. Energy 2016, 85, 1371–1390.
F. Liao, X.-P. Wu, J. Zheng, M. M.-J. Li, A. Kroner, Z. Zeng, X.
Hong, Y. Yuan, X.-Q. Gong, S. C. E. Tsang, Green Chem. 2017, 19,
270–280.
60 [8]
61
62
63 [9]
64
F. Liao, X.-P. Wu, J. Zheng, M.-J. Li, Z. Zeng, X. Hong, A. Kroner,
Y. Yuan, X.-Q. Gong, S. C. E. Tsang, Catal. Sci. Technol. 2016, 6,
7698–7702.
65
66 [10]
67
M. Behrens, F. Studt, I. Kasatkin, S. Kühl, M. Hävecker, F. Abild-
Pedersen, S. Zander, F. Girgsdies, P. Kurr, B.-L. Kniep, et al.,
Science 2012, 336, 893–7.
19 Conclusion
68
20
In summary, the unmodified Rh catalyst is well known to carry
69 [11]
70
S. Kuld, M. Thorhauge, H. Falsig, C. F. Elkjær, S. Helveg, I.
Chorkendorff, J. Sehested, Science 2016, 352, 969–74.
M. M.-J. Li, Z. Zeng, F. Liao, X. Hong, S. C. E. Tsang, J. Catal.
2016, 343, 157–167.
21 out extensive hydrogenation of CO2 to methane. However, this
22 study illustrates that the new Rh-In bimetallic catalyst offers
23 selective sites for capturing hydrogen gas and CO2 to approach
24 stoichiometric methanol formation with unprecedentedly high
71 [12]
72
M. Saito, Catal. Surv. From Japan 1998, 2, 175–184.
A. Gotti, R. Prins, Catal. Letters 1996, 37, 143–151.
H. Kusama, K. Okabe, K. Sayama, H. Arakawa, Appl. Organomet.
Chem. 2000, 14, 836–840.
25 methanol yield (up to
gMeOH∙gcat-1∙h-1) under industrially
73 [13]
1
26 applicable flow conditions. We believe that the usage of such 74 [14]
27 selective hydrogen-to-methanol catalyst may thus open a new
28 avenue for more efficient coupling with the upstream CO2/H2
29 production from renewable biomass feedstocks. Considering the
30 cost of Rh and In, further effort to enhance their atomic
31 effectiveness in catalyst fabrication including core-shell design,
32 mixed ion structure, and atom implantation is needed.
75 [15]
76
77 [16]
78
M. S. Frei, C. Mondelli, R. García-Muelas, K. S. Kley, B. Puértolas,
N. López, O. V. Safonova, J. A. Stewart, D. Curulla Ferré, J. Pérez-
Ramírez, Nat. Commun. 2019, 10, 3377.
79
80 [17]
81 [18]
82
B. Ouyang, W. Tan, B. Liu, Catal. Commun. 2017, 95, 36–39.
F. Studt, I. Sharafutdinov, F. Abild-Pedersen, C. F. Elkjær, J. S.
Hummelshøj, S. Dahl, I. Chorkendorff, J. K. Nørskov, Nat. Chem.
2014, 6, 320–324.
33 Acknowledgements
83
84 [19]
85
E. M. Fiordaliso, I. Sharafutdinov, H. W. P. Carvalho, J. D.
Grunwaldt, T. W. Hansen, I. Chorkendorff, J. B. Wagner, C. D.
Damsgaard, ACS Catal. 2015, 5, 5827–5836.
34 Financial support of this work from the EPSRC research council
35 of UK is acknowledged. The authors wish to thank beamline B07
36 of Diamond Light Source, UK for accessing its AP-XPS
37 facilities. M.M.-J.L. is grateful to the Swire Scholarship of Oxford
38 for her DPhil study and H.Z. is indebted to China Scholarship
39 Committee to fund her sabbatical study at the University of Oxford.
40 The authors acknowledge Dr. A. Robertson for the HR-TEM
41 images and Mr. Ping-Luen Ho for performing nonlinear filtering to
42 improve the quality of the TEM images.
86
87 [20]
88
O. Martin, A. J. Martín, C. Mondelli, S. Mitchell, T. F. Segawa, R.
Hauert, C. Drouilly, D. Curulla-Ferré, J. Pérez-Ramírez, Angew.
Chemie Int. Ed. 2016, 55, 6261–6265.
89
90 [21]
91
J. Wang, G. Li, Z. Li, C. Tang, Z. Feng, H. An, H. Liu, T. Liu, C. Li,
Sci. Adv. 2017, 3, e1701290.
92 [22]
93
M. M.-J. Li, C. Chen, T. Ayvali, H. Suo, J. Zheng, I. Teixeira, L. Ye,
H. Zou, D. O’Hare, S. C. E. Tsang, ACS Catal. 2018, 8, 4390–4401.
Y. Yin, B. Hu, X. Li, X. Zhou, X. Hong, G. Liu, Appl. Catal. B
Environ. 2018, 234, 143–152.
94 [23]
95
This article is protected by copyright. All rights reserved.

10.1002/anie.202000841
Angewandte Chemie International Edition
RESEARCH ARTICLE
1 [24]
N. Rui, Z. Wang, K. Sun, J. Ye, Q. Ge, C. Liu, Appl. Catal. B,
Environ. 2017, 218, 488–497.
2
3 [25]
J. L. Snider, V. Streibel, M. A. Hubert, T. S. Choksi, E. Valle, D. C.
Upham, J. Schumann, M. S. Duyar, A. Gallo, F. Abild-Pedersen, et
al., ACS Catal. 2019, 9, 3399–3412.
4
5
6 [26]
J. Graciani, K. Mudiyanselage, F. Xu, A. E. Baber, J. Evans, S. D.
Senanayake, D. J. Stacchiola, P. Liu, J. Hrbek, J. Fernández Sanz,
et al., Science 2014, 345, 546–50.
7
8
9 [27]
L. Lin, S. Yao, Z. Liu, F. Zhang, N. Li, D. Vovchok, A. Martínez-
Arias, R. Castañeda, J. Lin, S. D. Senanayake, et al., J. Phys.
Chem. C 2018, 122, 12934–12943.
10
11
12 [28]
13
S. Yang, S. H. Pang, T. P. Sulmonetti, W. Su, J. Lee, B. Hwang, C.
W. Jones, ACS Catal. 2018, 8, 12056–12066.
P. Anres, P. Fossati, J. P. Bros, J. Alloys Compd. 1998, 266, 241–
246.
14 [29]
15
16 [30]
17 [31]
18 [32]
19
J. VIS, J. Catal. 1985, 95, 333–345.
G. M. Nuñez, A. J. Rouco, J. Catal. 1988, 111, 41–49.
Y. Okamoto, N. Ishida, T. Imanaka, S. Teranishi, J. Catal. 1979, 58,
82–94.
20 [33]
21
Z. M. Detweiler, S. M. Wulfsberg, M. G. Frith, A. B. Bocarsly, S. L
Bernasek, Surf. Sci. 2016, 648, 188–195.
22 [34]
23
S. J. Kang, J. Baik, H.-J. Shin, J. Inorg. Organomet. Polym. Mater.
2013, 23, 1519–1523.
24 [35]
25
N. Iwasa, T. Mayanagi, N. Ogawa, K. Sakata, N. Takezawa, Catal.
Letters 1998, 54, 119–123.
26 [36]
27
J. A. Rodriguez, J. Evans, L. Feria, A. B. Vidal, P. Liu, K. Nakamura,
F. Illas, J. Catal. 2013, 307, 162–169.
28 [37]
29
J. A. Herron, J. Scaranto, P. Ferrin, S. Li, M. Mavrikakis, ACS Catal.
2014, 4, 4434–4445.
30 [38]
31
C. Liu, T. R. Cundari, A. K. Wilson, J. Phys. Chem. C 2012, 116,
5681–5688.
32 [39]
33
L. Dietz, S. Piccinin, M. Maestri, J. Phys. Chem. C 2015, 119, 4959–
4966.
34 [40]
35
G. C. Chinchen, P. J. Denny, J. R. Jennings, M. S. Spencer, K. C
Waugh, 1988, Appl. Catal. 36, 1–65.
36 [41]
37
Y. F. Zhao, Y. Yang, C. A. Mims, C. H. F. Peden, J. Li, D. Mei,
2011, J. Catal. 281, 199–211.
38 [42]
S. Kattel, P. J. Ramírez, J. G. Chen, J. A. Rodriguez, P. Liu,
Science 2017, 355, 1296–1299.
39
40
This article is protected by copyright. All rights reserved.

10.1002/anie.202000841
Angewandte Chemie International Edition
RESEARCH ARTICLE
Entry for the Table of Contents
RESEARCH ARTICLE
Molly Meng-Jung Li†, Hanbo Zou†,
Jianwei Zheng†, Tai-Sing Wu, Ting-Shan
Chan, Yun-Liang Soo, Xin-Ping Wu,
Xue-Qing Gong, Tianyi Chen, Kanak
Roy, Georg Held, Shik Chi Edman
Tsang*
Page No. – Page No.
Methanol synthesis at a wide range of
H2/CO2 ratios over Rh-In bimetallic
catalyst
Bimetallic Rh-In catalyst offers selective sites for capturing hydrogen and CO2 to approach methanol formation under H2-deficient
feedstock compositions with high methanol yield but minimise reverse water-gas shift reaction. Using this catalyst, a convenient
methanol synthesis based on renewable biomass-derived feedstocks with lower energy costs can be established.
This article is protected by copyright. All rights reserved.