PDF A ﬁrst-principles study on the hydrogenation of acetone on H ... - NJU

(1)

A first-principles study on the hydrogenation of acetone on H x MoO 3 surface

Qiyun Pan

^a

, Liang Huang

^b,^*

, Zhong Li

^a

, Juanjuan Han

^a

, Nian Zhao

^a

, Yunlong Xie

^a

, Xiang Li

^a

, Meifeng Liu

^a

, Xiuzhang Wang

^a

,

Jun-Ming Liu

^a,c

aInstitute for Advanced Materials, Hubei Key Laboratory of Pollutant Analysis&Reuse Technology, Hubei Normal University, Huangshi 435002, China

bThe State Key Laboratory of Refractories and Metallurgy, College of Materials and Metallurgy, Wuhan University of Science and Technology, Wuhan 430081, China

cNational Laboratory of Solid State Microstructures, Nanjing University, Nanjing 210093, China

a r t i c l e i n f o

Article history:

Received 19 December 2018 Received in revised form 4 February 2019

Accepted 6 February 2019 Available online 7 March 2019

Keywords:

First-principles study Acetone hydrogenation HxMoO3surface H content

a b s t r a c t

Hydrogenation of acetone on the (010) surface of hydrogen molybdenum bronzes was investigated by density functional theory (DFT) calculations with periodic slab models. The formation of H-bond between the carbonyl oxygen of acetone and the terminal OH group of the surface leads to a stable adsorption of acetone. The effect of hydrogen concentration in the bronzes on the hydrogenation of acetone was systematically investigated, indicating the hydrogenation reaction is a one-step concerted and exothermic process regardless of the hydrogen contents in the bronzes surface. The 8H surface with increased H-content shows a significantly exothermic reaction process and exhibits the smallest kinetic barrier compared with 4H or 6H surfaces. Additionally, the selectivity for hydrogenation acetone could increase owing the absence of CeC bond activation. The findings in this study can help with designing of high-efficient and low-cost metal oxide catalysts for hydrogenation of unsaturated substances.

Introduction

Hydrogenation of unsaturated substances is a widely prac- ticed process in catalytic reforming of petroleum feedstocks and numerous other chemical production processes[1e7]. As known, acetone is one of the typical unsaturated organic compounds. With the development of current industrial technology, the industrial yield of acetone has been over- saturated, while the usage of acetone as solvent is

decreasing in recent years, so as the other fields. In order to extend the usage of acetone, substantial experimental and theoretical studies have been devoted onto acetone hydrogenation in the recent years[8e14]. Among the hydrogenated products, isopropanol as the highly selective product of acetone is a critical solvent in industry for fine chemical synthesis, and enlarging the scale of isopropanol production by various approaches has drawn a lot of attention. In addi- tion, isopropanol is in high demand for direct isopropanol fuel cells for hydrogen storage[15e18], and relevant technologies

*Corresponding author.

E-mail address:[email protected](L. Huang).

Available online atwww.sciencedirect.com

ScienceDirect

journal hom epa ge: www.elsev ier.com/locate/he

https://doi.org/10.1016/j.ijhydene.2019.02.032

(2)

efficiency in acetone hydrogenation, they are more expensive and easy to be poisoned. Moreover, these solid heterogeneous catalysts work un-effectively, due to complicated experimental procedure at high operating temperature.

Alternatively, a class of hydrogen bronzes materials in the form of HxTMOy, where TM denotes transition metal, have been attracting attention in recent years. They can be pre- pared from transition metal oxides such as MoO3, WO3, and V₂O₅, via hydrogen spillover mechanisms. Hydrogen spillover phenomenon in MoO3and WO3was extensively studied since the 1970s[32]. In fact, earlier investigations indicated that the formation of hydrogen bronzes was favorable and H atoms have high mobility in these transition metal oxides[33e37]

with low energy barrier pathways. The low barriers for proton diffusion on MoO3(010) surface allow highly mobile pro- tons at near ambient temperature[38]. On the other hand, a series of previous works did demonstrate the highly selective and effective catalytic kinetics of hydrogen bronzes materials in the hydrogenation ofp-conjugated organic molecules with carbon-carbon double bond[39e42]. Recent theoretical calculations also predicted that molybdenum oxide hydrogen bronzes such as HxMoO3were highly effective to the hydrogenation of ethylene [42]. Compared with transition metal catalysts, several reactions catalyzed by hydrogen bronzes generally occur at lower temperature without unfavored by- products[43,44]. All these earlier investigations suggest the great potentials of hydrogen bronzes as promising hydrogenation catalysts of unsaturated substances. We believe, therefore that hydrogen molybdenum bronzes may be one kind of effective catalysts with promising applications in acetone hydrogenation owing to the ability of hydrogen molybdenum bronzes to reversibly capture and release hydrogen[45].

Nevertheless, so far, the hydrogenation of acetone by hydrogen bronzes remains less touched and there is almost no study on microscopic mechanism for hydrogenation of C]

O onto CeOH with these hydrogen bronzes. Therefore, the study of hydrogenation of C]O using acetone as probe molecule on the bronzes surface is one of the most important topics. Given the issues discussed above, it is of essential in- terest to address possible kinetic landscape of the acetone hydrogenation process on the hydrogen bronzes surface, using the first-principles calculations based on the density- functional theory (DFT). As reported, the hydrogen concentrations in the lattice influence the physicochemical properties of MoO3[45]and the reaction activity[39,40]. Therefore,

layers, as done in our recent work [42]and shown inFig. 1 for a schematic illustration. The calculated adsorption energies of acetone on the (22) MoO3(010) slab, (33) MoO3

(010) slab and two-layer (22) MoO₃(010) slab were4.9,3.5 and 4.5 kcal/mol, respectively. Since MoO3 is principally layer-packed along the [010] orientation by means of weak van der Waals interactions, the effect of further increasing the atom layers is virtually negligible. Meanwhile, high coverage is more conducive to the adsorption of reactant. So we chose a supercell containing one (22) MoO3(010) slab, which is large enough to accommodate surface reactive species. Between adjacent slabs, a vacuum space of 15A was inserted to elim- inate the artificial inter-cell interactions.

All the calculations were performed on the commercially available code VASP (Vienna ab-initio simulation package) in the framework of spin-polarized generalized gradient approximation (GGA) [46,47]. The exchange-correlation potential was described by the PBE functional[48], and the cut- off energy of solving Kohh-Sham equation was chosen as 400 eV. The Monkhorst-Pack grid (331) was applied to sample the Brillouin zone during the geometry relaxation, while the Methfessel-Paxton technique was adopted to include the electron smearing effect in order to minimize the errors in Hellmann-Feynman forces and facilitate self- consistent field convergence[49]. The geometry was considered as an equilibrium state when the energy difference in two subsequent optimization steps was smaller than 10³eV and the force was converged to be less than 0.03 eV/A. The structures of transition states were obtained by the climbing image nudged elastic band (CI-NEB) method[50,51], where six images between reactant and product were inserted to ach- ieve a smooth energy curve with the force tolerance of 0.05 eV/

A. The minimum energy path calculations were performed to explore the detailed catalytic process of acetone hydrogenation. This computational protocol has been proved to be well- accurate in producing structural and energetic data for MoO3

and HxMoO3systems[42].

Results and discussions

The structures and formation thermodynamics of HxMoO3

with various hydrogen contents can be found in details in our previous publication [34]. In general, there are three types of O atoms on the MoO3(010) surface, denoted as Ot (terminal site), Oa (asymmetric site) and Os (symmetric

(3)

site) (Fig. 1.). Upon the hydrogenation of MoO3, the Ot sites are most preferentially occupied by H atoms, followed by the occupations at the Oa and Os sites. Since H atoms are highly mobile in MoO3 lattice, we can rationally assume that the Ot sites are always occupied by H atoms during the hydrogenation process. As shown in hydrogenation of

ethylene, the increased hydrogen concentration could lead to an enhanced kinetics [42]. A series of HxMoO3systems with varying H contents (labeled as 4H, 6H and 8H for illustration (Figs. 2e4), were considered in our calculations to evaluate the reactivity of HxMoO3with various hydrogen concentrations.

Fig. 1e(a) Side view of MoO3(010) surface, (b) top view of MoO3(010) surface.

Fig. 2e(a) Calculated energy diagram of the hydrogenation of acetone on the 4H molybdenum bronzes surface. (b) The top view of the acetone on the 4H molybdenum bronzes surface. Optimized configurations of acetone absorption geometries on the 4H surface (c) initial state (AB), (d) transition state (TS) and (e) final state (CD). (AþB presents the state of acetone and HxMoO3before adsorption, CþD is the state of isopropanol and HxMoO3after hydrogenation).

(4)

nation profile is shown inFig. 2, and several key structural parameters of the reactant, transition state, and product on the 4H surface are displayed inTable 1. Initially, the acetone molecule is attracted by theeOH groups on the surface with a moderate adsorption strength of 4.9 kcal/mol, which is much larger than the adsorption energy of ethylene (2.1 kcal/

mol)[42]. The enhanced anchoring strength of acetone should originate from the interaction between the electron enriched O atom of C]O and the protonic H ofeOH groups on molybdenum bronzes surface[42]. This allows a relatively strong H- bonding given the calculated O3eH2 distance of 1.756A. On the other hand, a relatively long distance between H1 and C1 is observed due to the steric repulsion between the methyl group of acetone and theeOH on the molybdenum bronzes surface.

Comparing with the gas phase acetone, the CeC bonds of acetone are essentially unchanged upon the adsorption, sug- gesting the fact that CeC bonds of acetone are not activated via the H-bonding interaction. As a result, only the C]O bond is accessible for hydrogenation, implying the superior selectivity of HxMoO3for the hydrogenation.

Subsequently, the isopropanol (hydrogenation product) is structurally relaxed on the HxMoO3surface to model the reaction product (CD), as shown in Fig. 2(a). The optimized adsorption structure of isopropanol on the surface is virtually identical to its gas-phase configuration. The formation of isopropanol is almost thermally neutral with a negligible

double-bonding. In the meantime, the distance of both C1eH1 and O3eH2 bonds is reduced to 2.074A and 1.036A, respectively, implying the formation of hydrogenation product. The short distance of O3eH2 (1.036A) in the transition state is originated from the strong H-bonding interaction, which fa- cilitates the H migration process. As a consequence, a moderate activation barrier of 17.8 kcal/mol is required to accomplish the hydrogenation reaction. Since the isopropanol is still adsorbed on the surface through a relatively week H- bonding interaction between theeOH of isopropanol and the terminal O atom on the surface with a distance of 2.080A, a small desorption energy of 2.1 kcal/mol is necessary to release the product into gas phase. Our results suggest that HxMoO3

with the 4H surface is capable of hydrogenating acetone molecules with superior activity. It is notable the desorption of isopropanol is facile, reducing the likelihood of subsequent reactions, which is helpful to improve the selectivity.

Hydrogenation of acetone on6H and 8H molybdenum bronzes surface

Similar calculations were performed on the 6H and 8H surfaces to address the effect of H content on the hydrogenation reaction. For the 6H surface, two possible starting points were considered in our calculations due to the asymmetric location of the H atom in the second layer, as shown inFig. 3(b and c).

Comparing with the 4H surface case, the internal H-bonding interaction between the H atom in the second layer and the O atom in the third layer leads to a downward movement of the second layer Oa atoms, resulting in two reaction sites. Upon adsorption of acetone on the surface, the acetone is somewhat locked by the foureOH groups on the surface. The reaction space at site A is more open (:OeMoeO¼94.1), while the adsorption region at site B is somewhat embedded with a reduced:OeMoeO¼85.4. Therefore, the site A should be more accessible for acetone molecule during the hydrogenation. Indeed, the optimized adsorption structures indicate that both the O3eH2 and C1eH1 distances at site A are much shorter than the value at site B (Table 1). As a consequence, the acetone adsorption at site A is energetically more favorable than that at site B by 1.9 kcal/mol in the energy difference.

As described in the 4H case, the acetone hydrogenation on the 6H surface also follows the one step reaction mechanism.

Table 1 summarizes the structural parameters of the optimized geometries along the reaction. At the transition state of site A, a small surface relaxation is observed with negligible Table 1eOptimized Structural Parameters of reactant

complex(R), transition state(TS), and product complex(P) on the 4H, 6H, and 8H molybdenum bronzes surface^a.

bond length (A)

C1eO3 C1eH1 O3eH2 O1eH1 O2eH2

4H R 1.234 2.964 1.756 0.981 1.003

TS 1.309 2.074 1.036 1.002 1.555

P 1.442 1.106 0.979 2.334 2.080

6H site A R 1.234 2.787 1.745 0.977 0.999

TS 1.322 2.219 1.013 0.997 1.696

P 1.443 1.108 0.976 2.579 2.307

6H site B R 1.233 2.821 1.966 0.983 0.990

TS 1.352 1.906 0.988 1.017 2.028

P 1.444 1.104 0.976 2.226 2.307

8H R 1.235 2.815 1.765 0.978 1.001

TS 1.337 2.286 0.992 0.991 1.845

P 1.446 1.103 0.976 2.330 2.352

a C1, C2, H1, H2, O1, and O2 are denoted inFig. 2(b).

(5)

increase of angle :OeMoeO. Here, the O3eH2 distance is slightly shorter than the value obtained on the 4H surface, in line with the facile accessibility of site A. The calculated activation energy of 16.8 kcal/mol (Fig. 3(a)) is also lower than the value on the 4H surface. Thermodynamically, the hydrogenation of acetone at site A become more exothermic than the 4H case with the calculated reaction energy of7.3 kcal/

mol (Fig. 3(a)). Alternatively, the hydrogenation reaction could also proceed at site B, where the foureOH groups are pushed

slightly toward each other. In principle, the acetone molecule should be a little bit far from the surface at the transition state due to the poor accessibility of site B. However, the optimized structure at the transition state shows that the acetone is very close to the surface and both the O3eH2 and C1eH1 distances are even shorter than the values at site A. The unexpected approaching of acetone is attributed to the considerable surface relaxation during the hydrogenation, as reflected by the significantly increased angle :OeMoeO from 85.4 of the Fig. 3e(a) Calculated energy diagram of the hydrogenation of acetone on the 6H molybdenum bronzes surface. The two possible reaction pathways are denoted as black for site A and red for site B. Optimized configurations of possible acetone absorption geometries on the 6H molybdenum bronzes surface with (b) 6H site A (1) initial state(AB), (2) transition state(TS) and (3) finial state(CD). (c) 6H site B. (4) initial state(AB), (5) transition state(TS) and (6) finial state(CD). (For interpretation of the references to colour in this figure legend, the reader is referred to the Web version of this article.)

(6)

asymmetric O site (Oa) in the second layer on the 8H surface move downwards to interact with the O atom in the third layer, leading to an open interaction space surrounded by the four terminal OH groups (Fig. 4(b)). Considering the higher H content, the 8H surface is expected to present favorable reaction kinetic properties. Kinetically, the 8H surface shows a lower activation barrier (7.9 kcal/mol) than those of the 4H and 6H (site A) surfaces. Thermodynamically, the hydrogenation process on the 8H surface is extremely exothermic with the calculated reaction energy of25.2 kcal/mol. Therefore, the highly protonic nature of the terminal H is a dominant factor

surface. The twoporbitals around7.5 and 2.5 eV are attributed to the C]O bonds of acetone, accompanied with the minor change of orbitals of oxygen (Fig. 5(2)). After the formation of isopropanol (Fig. 5(3)), the electronic structure of Mo around the Fermi level experiences significant changes as expected, where the Mo atoms linked to the terminal OeH group are oxidized, and the C]O orbitals of acetone disap- pears as one CeH bond and one OeH bond are formed.

The DOS of 6H and 8H were also calculated, as shown in Fig. 6. Generally, the conductor-like electronic structure of bronzes surface is preserved along the reaction pathway. The

Fig. 4e(a) Calculated energy diagram of the hydrogenation of acetone on the 8H molybdenum bronzes surface. The structure of acetone on the 8H surface (a) initial state(AB), (b) transition state(TS) and (c) finial state(CD).

(7)

adsorption of acetone would not result in any obvious changes in the electronic structures of bronze with the selected surfaces, reflecting their weak physical adsorption properties.

While the formation of isopropanol could lead to significant changes around the Fermi level, due to oxidization of the terminal OeH group.

Fig. 7shows the relationship between the calculated thermochemical energies and activation barriers of acetone hydrogenation on the HxMoO3surfaces with selected hydrogen contents. This phenomenon fits well with the BEP (Brønsted- Evans-Polanyi) Relations, which is able to predict energy barriers of the reaction under different chemical environment of the reactive center[52]. From the calculate results we can see

that, as the content of hydrogen in these compounds increases, the catalytic activity of HxMoO3for acetone hydrogenation increases significantly, which fits to our previous study for ethylene hydrogenation elucidated the potential positive effects of increasing H content on the bronzes surface to suppress the energy barrier for hydrogenation [34]. The calculated activation energies range from 17.8 to 7.9 kcal/mol, depending on the content of H in the hydrogen bronzes.

Overall, the reaction pathway to hydrogenate acetone on the 8H surface is the most feasible in kinetics, with a barrier of only 7.9 kcal/mol. It fits to our previous study for ethylene hydrogenation elucidated the potential positive effects of increasing H content on the bronze surface to suppress the Fig. 5eCalculated electronic DOS of (1) the 4H bronzes surface, (2) the 4H bronzes surface with acetone, (3) the 4H bronzes surface with isopropanol after reaction.

(8)

energy barrier for hydrogenation. It is worth mentioning that, compared to the reported activation energies of acetone hydrogenation on the Pt(111) surface (14.3 kcal/mol) and on the Ni tetramer cluster (17.4 kcal/mol), hydrogenation of acetone catalyzed by HxMoO₃appears to be kinetically more favorable.

Compared with ethylene hydrogenation on hydrogen bronzes HxMoO3surface, it can be observed that the acetone hydrogenation is quite different from the case of ethylene hydrogenation (Table 2). Different from ethylene that requires the C]C bond activation, the polar C]O bond in acetone would definitely facilitate the proton transfer from the terminal OH group to the carbonyl oxygenviaan H-bond attack, so that hydrogenation of acetone exhibits lower activation energy. It is noteworthy that the hydrogenation activity and activation energy between hydrogenation of acetone and ethylene on hydrogen bronzes HxMoO3surface are different, indicating HxMoO3may be a promising and effective catalyst for selective hydrogenation of unsaturated substances containing C]O and C]C bond simultaneously. It may be a meaningful research for HxMoO3as the hydrogenation catalyst for methyl vinyl ketone[53].

Fig. 6eCalculated electronic DOS. (1)the bronze surface of 6H, (2) the 6H bronze surface upon acetone adsorption, (3) the 6H bronze surface upon hydrogenation to form isopropanol; (4)the bronze surface of 8H, (5) the 8H bronze surface upon acetone adsorption, (6) the 8H bronze surface upon hydrogenation to form isopropanol.

Fig. 7eThe relationship between calculated

thermochemical energies (DEr) and the activation energies (Ea) of acetone hydrogenation on the HxMoO3with various hydrogen concentrations.

(9)

Conclusion

The surface catalytic activity of hydrogen bronzes MoO3with different hydrogen contents towards acetone hydrogenation was investigated by means of plane wave density functional theory. The adsorption of acetone on the HxMoO3surfaces is favorable due to the formation of H-bond between acetone and the surfaces. There is virtually no CeC bond activation during the adsorption of acetone, allowing for excellent selectivity for hydrogenation. Our calculations suggest that the hydrogenation of acetone is a one-step concerted and exothermic reaction. Increase of H concentration (i.e., 8H surface) expectedly decreases the kinetic barrier. The hydrogenation of acetone on 8H surface is extremely exothermic (25.2 kcal/mol) with the smallest activation barrier (7.9 kcal/

mol), even appearing to be kinetically more favorable than Pt (111) surface and Ni tetramer cluster. Additionally, the difference of hydrogenation activity of acetone and ethylene on HxMoO3 surface indicate the potential application of hydrogen bronzes as selective hydrogenation catalysts. Our results suggest HxMoO3 could be an effective catalyst for acetone hydrogenation and provide a novel insight for designing low-cost hydrogen bronzes MoO₃catalysts.

Acknowledgements

This work was financially supported by National Natural Sci- ence Foundation of China (Grant No. 51702241), Hubei Key Laboratory of Pollutant Analysis&Reuse Technology (Grant No. PA20170203), the Research Project of Hubei Provincial Department of Education (Grant No. B2018147).

Appendix A. Supplementary data

Supplementary data related to this article can be found at https://doi.org/10.1016/j.ijhydene.2019.02.032.

r e f e r e n c e s

[1] Esmaeili J, Rahimpour F. Regeneration of spent nickel catalyst from hydrogenation process of edible oils: heat treatment with hydrogen injection. Int J Hydrog Energy 2017;42:24197e204.

[2] Fei S, Han B, Li L, Mei P, Zhu T, Yang M, et al. A study on the catalytic hydrogenation of N-ethylcarbazole on the mesoporous Pd/MoO3catalyst. Int J Hydrog Energy 2017;42:25942e50.

[3] Huang H, Yuan P, Yu Y, Chung KH. Electrochemical hydrogenation of organic sulfides. Int J Hydrog Energy 2017;42:18203e8.

[4] Knowles WS. Asymmetric hydrogenations (nobel lecture 2001). Adv Synth Catal 2003;345:3e13.

[5] Bozkurt G,Ozer A, Yurtcan AB. Hydrogen generation from€ sodium borohydride with Ni and Co based catalysts supported on Co3O4. Int J Hydrog Energy 2018;43:22205e14.

[6] Wang Y, Yang X, Wang Y. Catalytic performance of mesoporous MgO supported Ni catalyst in steam reforming of model compounds of biomass fermentation for hydrogen production. Int J Hydrog Energy 2016;41:17846e57.

[7] Bourane A, Elanany M, Pham TV, Katikaneni SP. An overview of organic liquid phase hydrogen carriers. Int J Hydrog Energy 2016;41:23075e91.

[8] Rositani F, Galvagno S, Poltarzewski Z, Staiti P, Antonucci PL.

Kinetics of acetone hydrogenation over Pt/Al2O3catalysts. J Chem Technol Biotechnol 1985;35:234e40.

[9] Saito Y, Yamashita M, Ito E, Meng N. Hydrogen production from 2-propanol as a key reaction for a chemical heat pump with reaction couple of 2-propanol

dehydrogenation/acetone hydrogenation. Int J Hydrog Energy 1994;19:223e6.

[10] Meng N, Shinoda S, Saito Y. Improvements on thermal efficiency of chemical heat pump involving the reaction couple of 2-propanol dehydrogenation and acetone hydrogenation. Int J Hydrog Energy 1997;22:361e7.

[11] Jeffery EL, Mann RK, Hutchings GJ, Taylor SH, Willock DJ. A density functional theory study of the adsorption of acetone to the (111) surface of Pt: implications for hydrogenation catalysis. Catal Today 2005;105:85e92.

[12] Kawahara R, Fujita K, Yamaguchi R. Cooperative catalysis by Iridium complexes with a bipyridonate ligand: versatile dehydrogenative oxidation of alcohols and reversible dehydrogenationehydrogenation between 2-propanol and acetone. Angew Chem Int Ed 2012;124:12962e6.

[13] Balouch A, Ali Umar A, Shah AA, Mat Salleh M, Oyama M.

Efficient heterogeneous catalytic hydrogenation of acetone to isopropanol on semihollow and porous palladium nanocatalyst. ACS Appl Mater Interfaces 2013;5:9843e9.

[14] Peng W, Xu M, Huai X, Liu Z. 3D CFD simulations of acetone hydrogenation in randomly packed beds for an

isopropanoleacetoneehydrogen chemical heat pump. Appl Therm Eng 2016;94:238e48.

[15] Cao D, Bergens SH. A direct 2-propanol polymer electrolyte fuel cell. J Power Sources 2003;124:12e7.

[16] Buonaiuto M, De Crisci AG, Jaramillo TF, Waymouth RM.

Electrooxidation of alcohols with electrode-supported transfer hydrogenation catalysts. ACS Catal

2015;5:7343e9.

[17] Waldie KM, Flajslik KR, McLoughlin E, Chidsey CED, Waymouth RM. Electrocatalytic alcohol oxidation with ruthenium transfer hydrogenation catalysts. J Am Chem Soc 2017;139:738e48.

[18] Solis-Tobı´as JE, Dı´az-Guillen JA, Melendez-Gonzalez PC, Sanchez-Padilla NM, Perez-Hernandez R, Alonso-Lemus IL, et al. Enhanced catalytic activity of supported

nanostructured Pd for the oxidation of organic molecules usingg-Fe2O3and Fe3O4as co-electrocatalysts. Int J Hydrog Energy 2017;42:30301e9.

[19] van Druten GMR, Ponec V. Hydrogenation of carbonylic compounds: Part I: competitive hydrogenation of propanal and acetone over noble metal catalysts. Appl Catal, A 2000;191:153e62.

Table 2eThe comparison about hydrogenation of acetone and ethylene on molybdenum bronzes surface.

Contrast Unsaturated bond Reaction mechanism Lowest activation energy(kcal/mol) H concentration effect

Acetone C¼O One step 7.9 Fits to BEP

Ethylene C¼C One step 9.3þ0.9 Fits to BEP

(10)

versatile oxidation and hydrogenation catalyst. Acc Chem Res 2014;47:750e60.

[25] Li Y, Huang X, Zhang Q, Chen L, Zhang X, Wang T, et al.

Hydrogenation and hydrodeoxygenation of difurfurylidene acetone to liquid alkanes over Raney Ni and the supported Pt catalysts. Appl Energy 2015;160:990e8.

[26] Ji Y, Wu Y, Zhao G, Wang D, Liu L, He W, et al. Porous bimetallic Pt-Fe nanocatalysts for highly efficient hydrogenation of acetone. Nano Res 2015;8:2706e13.

[27] Polo V, Schrock RR, Oro LA. A DFT study of the role of water in the rhodium-catalyzed hydrogenation of acetone. Chem Commun 2016;52:13881e4.

[28] Konda SSM, Caratzoulas S, Vlachos DG. Computational insights into the role of metal and acid sites in bifunctional metal/zeolite catalysts: a case study of acetone

hydrogenation to 2-propanol and subsequent dehydration to propene. ACS Catal 2016;6:123e33.

[29] Tezel E, Figen HE, Baykara SZ. Hydrogen production by methane decomposition using bimetallic NieFe catalysts. Int J Hydrog Energy 2019. Doi: org/10.1016/j.ijhydene.2018.12.151, https://www.sciencedirect.com/science/article/pii/

S0360319918341533.

[30] Deng X, Yang T, Zhang Q, Chu Y, Luo J, Zhang L, et al. A monolith CuNiFe/g-Al2O3/Al catalyst for steam reforming of dimethyl ether and applied in a microreactor. Int J Hydrog Energy 2019;44:2417e25.

[31] Pastor-Perez L, Gu S, Sepu´lveda-Escribano A, Reina TR.

Bimetallic CueNi catalysts for the WGS reactione cooperative or uncooperative effect? Int J Hydrog Energy 2019;44:4011e9.

[32] Vannice MA, Boudart M, Fripiat JJ. Mobility of hydrogen in hydrogen tungsten bronze. J Catal 1970;17:359e65.

[33] Smith RL, Rohrer GS. An atomic force microscopy study of the morphological evolution of the MoO3(010) surface during reduction reactions. J Catal 1996;163:12e7.

[34] Smith RL, Rohrer GS. The protonation of MoO3during the partial oxidation of alcohols. J Catal 1998;173:219e28.

[35] Noh H, Wang D, Luo S, Flanagan TB, Balasubramaniam R, Sakamoto Y. Hydrogen bronze formation within Pd/MoO3

composites. J Phys Chem B 2004;108:310e9.

materials. Energy Environ Sci 2008;1:338e54.

[42] Yang M, Han B, Cheng H. First-principles study of

hydrogenation of ethylene on a HxMoO3(010) surface. J Phys Chem C 2012;116:24630e8.

[43] Matsuda T, Sakagami H, Takahashi N. H2-reduced Pt/MoO3 as a selective catalyst for heptane isomerization. Catal Today 2003;81:31e42.

[44] Sakagami H, Asano Y, Takahashi N, Matsuda T. H2 reduction of hydrogen molybdenum bronze to porous molybdenum oxide and its catalytic properties for the conversions of pentane and propan-2-ol. Appl Catal A 2005;284:123e30.

[45] Sha X, Chen L, Cooper AC, Pez GP, Cheng H. Hydrogen absorption and diffusion in bulka-MoO₃. J Phys Chem C 2009;113:11399e407.

[46] Kresse G, Hafner J. Ab initio molecular dynamics for open- shell transition metals. Phys Rev B 1993;48:13115e8.

[47] Kresse G, Furthmu¨ller J. Efficiency of ab-initio total energy calculations for metals and semiconductors using a plane- wave basis set. Comput Mater Sci 1996;6:15e50.

[48] Perdew JP, Burke K, Ernzerhof M. Generalized gradient approximation made simple. Phys Rev Lett 1996;77:3865e8.

[49] Methfessel M, Paxton AT. High-precision sampling for Brillouin-zone integration in metals. Phys Rev B 1989;40:3616e21.

[50] Henkelman G, Uberuaga BP, Jonsson H. A climbing image nudged elastic band method for finding saddle points and minimum energy paths. J Chem Phys 2000;113:9901e4.

[51] Henkelman G, Jonsson H. Improved tangent estimate in the nudged elastic band method for finding minimum energy paths and saddle points. J Chem Phys 2000;113:9978e85.

[52] Loffreda D, Delbecq F, Vigne F, Sautet P. Fast prediction of selectivity in heterogeneous catalysis from extended BrønstedeEvansePolanyi relations: a theoretical insight.

Angew Chem Int Ed 2009;48:8978e80.

[53] Smith GV, Deany JF. Metal-catalyzed intramolecular hydrogen shift during hydrogenation and exchange of methyl vinyl ketone. J Catal 1966;6:14e9.