Defect Engineering in MoSe2 for the Hydrogen Evolution Reaction

Nov 20, 2017 - The green and red triangles denote that the introduction of vacancies and holes into the MoSe2 basal plane leads to Se and Mo defect at...
0 downloads 7 Views 2MB Size
Subscriber access provided by READING UNIV

Article

Defect Engineering in MoSe2 for Hydrogen Evolution Reaction: From Point Defects to Edges Haibo Shu, Dong Zhou, Feng Li, Dan Cao, and Xiaoshuang Chen ACS Appl. Mater. Interfaces, Just Accepted Manuscript • DOI: 10.1021/acsami.7b12478 • Publication Date (Web): 20 Nov 2017 Downloaded from http://pubs.acs.org on November 21, 2017

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a free service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are accessible to all readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

ACS Applied Materials & Interfaces is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 22

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Applied Materials & Interfaces

Defect Engineering in MoSe2 for Hydrogen Evolution Reaction: From Point Defects to Edges Haibo Shu,†, ‡,* Dong Zhou,† Feng Li,† Dan Cao,§ and Xiaoshuang Chen ‡ †

College of Optical and Electronic Technology and §College of Science, China Jiliang University, 310018 Hangzhou, China ‡ National Laboratory for Infrared Physics, Shanghai Institute of Technical Physics, Chinese Academy of Science, 200083 Shanghai, China ∗ Corresponding author. Haibo Shu, Phone: +86-0571-86875622, E-mail: [email protected] ABSTRACT: Superior catalytic activity and high chemical stability of inexpensive electrocatalysts for the hydrogen evolution reaction (HER) are crucial to the large-scale production of hydrogen from water. The nonprecious two-dimensional (2D) MoSe2 materials are emerged as a potential candidate, and the improvement of their catalytic activity depends on the optimization of active reaction sites at both the edges and the basal plane. Herein, the structural stability, electrocatalytic activity and HER mechanisms on a series of MoSe2 catalytic structures including of point defects, holes, and edges have been explored by using first-principles calculations. Our calculated results demonstrate that thermodynamically stable defects (e.g., VSe, VSe2, SeMo, and VMo3Se2) and edges (e.g., Mo-R and Se-R) in MoSe2 are very similar to the case of MoS2 but their HER activity is higher than that of the corresponding structures in MoS2, which is in good agreement with experimental observations. Furthermore, a Fermi-abundance model is proposed to explain the fundamental correlation between the HER activity of various MoSe2 catalysts and their intrinsic electronic structures, and this model is also applicable for assessing the HER activity of other types of catalysts, such as MoS2 and Pt. Moreover, two different HER mechanisms have been revealed in the MoSe2 catalytic structures: the Volmer-Tafel mechanism is preferred for the VSe and VSe2 structures, while the Volmer-Heyrovsky mechanism is more favorable for other MoSe2 catalytic structures. The present work suggests that MoSe2 with appropriate defects and edges is able to compete against the Pt-based catalysts and also opens a route to design highly active electrocatalyts for the HER. KEYWORDS: transition-metal dichalcogenides, defect, edge, hydrogen evolution reaction, densityfunctional theory.

1 Environment ACS Paragon Plus

ACS Applied Materials & Interfaces

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 2 of 22

1. INTRODUCTION Electrochemical water splitting is a sustainable method to generate molecular hydrogen (H2),1-3 which has been regarded as one of the most promising clean energy solutions. Employing efficient catalysts for driving hydrogen evolution reaction (HER) is critical for the H2 production because they can accelerate the conversion from proton-electron pairs to H2 at low overpotentials (η).4 Among a variety of available catalysts, Platinum (Pt)-based catalysts exhibit outstanding electrocatalytic HER performance with high exchange current density and small Tafel slope.5 However, high cost and scarcity of Pt-based catalysts limit their widespread use.6,7 To gain a sustainable H2 production, the development of cost-effective and earth-abundant catalysts with high electrocatalytic activity and stability is still desirable. Recently, twodimensional (2D) layered transition metal dichalcogenides (TMDs), such as MoS2, MoSe2 and VS2,8-12 have emerged as an alternative to Pt-based catalysts. However, the HER active sites of TMD materials have been identified from the exposed metallic edges and their basal planes are catalytically inert,13,14 which limits the applications of TMD materials as the HER electrocatalysts. To improve the HER activity of TMDs-based electrocatalysts, extensive effort has been devoted to exploiting efficient methods to reach this goal. One feasible strategy strives to increase exposing edge sites through synthesizing nanostructured TMDs, such as nanoparticles,15 nanowires,16 nanoflakes,17,18 and nanofilms19. Another strategy was proposed to enhance the catalytic activity of TMD basal planes through phase structural engineering and defect engineering. Most of TMD materials with 2H phase are semiconductors, thus the transition from the semiconducting 2H to the metallic 1T phase contributes to the improvement of intrinsic HER activity on the basal plane.20,21 However, the unstable nature of 1Tphase TMD materials restricts the use of phase engineering method.22 Therefore, extensive studies focus on the defect engineering to promote the HER activity of basal planes.23-26 For instance, the inert basal planes of MoS2 and MoSe2 could be activated by introducing vacancies, grain boundaries, and holes. Recently, strained S-vacancies in the basal plane of 2H-phase MoS2 were found to have higher HER activity than the MoS2 edge sites.27 The improvement of HER activity in the MoS2 basal plane was ascribed to the produced gap states around the Fermi level which allows H to bind at exposed Mo atoms. Despite these preliminary achievements, our knowledge on the HER mechanisms and pathways of nanostructured and defective TMD materials is still limited, which becomes a main bottleneck for the further improvement of HER performance. As a representative TMD material, the HER performance of MoSe2 have been widely explored,28-30 but there is no consensus on the activity and HER mechanism of MoSe2 electrocatalysts. For instance, is the HER activity of MoSe2 catalysts better than MoS2? What is the best active site of MoSe2 for the HER, defects or edges? How is the HER mechanisms at the defect and edge sites of MoSe2? To answer these questions, it is necessary to understand the HER mechanisms on different active sites of MoSe2 catalysts and recognize the origin of catalytic activity. On the other 2 Environment ACS Paragon Plus

Page 3 of 22

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Applied Materials & Interfaces

hand, trial-and-error approaches are still used to data for the development of highly active catalysts, but they are still time-consuming and ineffective. Generally, monitoring the reaction intermediates formed on the catalytic surfaces is an effective approach for understanding the pathway and mechanism of HER catalysis, but it is difficult to detect in situ these adsorbed intermediates in experiment due to their extremely short lifetimes. For the rational design of TMD-based catalysts, it is critical to develop a fundamental guiding principle or descriptor for assessing their HER activities, which will accelerate the search for highly active electrocatalysts for the HER. In this work, we perform systematic density-functional theory (DFT) calculations to investigate the HER activities and mechanisms on a series of MoSe2 catalytic structures from point defects to edges. The activity trend of the MoSe2 catalytic structures is established by the correlation with their hydrogen adsorption free energies. Based on the electronic structures of catalysts, the origin of HER activity is explained by a Fermi-abundance model that is verified as a good activity descriptor for TMDs-based catalysts. Furthermore, the HER mechanisms and reaction dynamics on the MoSe2 defect and edge sites have been revealed by using ab initio molecular dynamics (AIMD) simulations. The present study paves a way for the design of efficient TMDs-based catalysts in electrochemical water-splitting applications.

2. COMPUTATIONAL DETAILS Three groups of MoSe2 models are built to investigate the electrocatalytic activity and HER mechanisms, including of point defects, holes and edges. Prior to the construction of various models, a single-layer MoSe2 primitive cell is relaxed to determine the optimal lattice parameters. The calculated in-plane lattice constants is a=b= 3.323 Å with the angle of 120o between the two in-plane vectors, which is in good agreement with reported theoretical and experimental values.30,31 Based on the optimized unit cell, the point-defect and hole models are created by introducing the intrinsic point defects or holes into a 7×7 unit cell. As shown in Figure 1, we consider five potential point-defect models (see Figures 1a-e), including the Mo vacancy (VMo), monoselenium vacancy (VSe), diselenium vacancy (VSe2), Mo antisite (MoSe), and Se antisite (SeMo), and three hole models (see Figures 1f-h), including VMoSe3, VMo3Se2, and VMoSe6, respectively. The edge models are constructed on the basis of zigzag MoSe2 nanoribbons with different configurations (see Figures 1i-n), including the perfect Mo and Se edges (Mo-P and Se-P), the (2×1) reconstructed Mo and Se edges (Mo-R and Se-R), the Mo edge with the Se termination (Mo-Se), and the Se edge with the Mo termination (Se-Mo), respectively. To evaluate the stability of various defective structures, their formation energies have been calculated by the following equation, E f = ED − ET + ∆ni ∑ µi i

(1)

where ED and ET are total energies of monolayer MoSe2 with and without intrinsic defects, respectively. µi is the chemical potential of atomic species i (i = Mo and Se), and ∆ni is the difference of the number 3 Environment ACS Paragon Plus

ACS Applied Materials & Interfaces

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 4 of 22

of Mo and Se atoms between perfect and defective structures. To maintain thermodynamic equilibrium (i.e., µMoSe2 = µMo +2µSe), the allowable range of µSe is µSe(bulk) –∆Hf/2 < µSe < µSe(bulk), where the upper (lower) limit corresponds to Se-rich (Mo-rich) condition and ∆Hf is the heat of formation. Here ∆Hf is defined as ∆Hf = EMoSe2 – EMo – 2ESe, where EMoSe2, EMo, and ESe are the energies of MoSe2 monolayer, Mo and Se atoms in bulk, respectively. The computed ∆Hf is 2.06 eV and agrees well with the previous report.32 The stability of MoSe2 edges is evaluated by the comparison of edge formation energies, the corresponding calculated details have been demonstrated in our previous studies.33

Figure 1. Optimized atomic structures of point defects and holes on MoSe2 basal plane and zigzag MoSe2 edges. (a)-(e) point defect configurations, (f)-(h) hole configurations, and (i)-(n) zigzag edge configurations. Blue and yellow balls represent Mo and Se atoms, respectively. The green and red triangles denote that the introduction of vacancies and holes into the MoSe2 basal plane leads to Se and Mo defect atoms, respectively.

To evaluate the electrocatalytic HER activity, the relative hydrogen adsorption free energies (∆GH) on various MoSe2 catalytic structures have been calculated in the acid solution using the standard hydrogen electrode (CHE) model as follows,34 ∆GH = ∆EH + ∆EZPE – T∆S

(2)

where ∆EH is the adsorption energy for adding one H atom onto MoSe2 catalysts and is defined as ∆EH = EMoSe2+nH – EMoSe2+(n-1)H – 1/2H2. Here the H adsorption configurations are created by placing H atoms at the defect sites of a 7×7 supercell or the edge sites of 4×1 nanoribbon models, and the most stable 4 Environment ACS Paragon Plus

Page 5 of 22

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Applied Materials & Interfaces

configurations (see Figure S1 of Supporting Information (SI)) are obtained by the energy comparison. ∆EZPE and ∆S are zero-point energies (ZPE) and entropy differences between the adsorbed system and gas phase, respectively and T is temperature which refers to the room temperature (T = 298.15 K). EZPE is defined as EZPE = 1/ 2∑ hωi , where ω is the vibration frequency and N is the number of H i= N

independent freedom degree in adsorbed system or H2 gas molecule. The ∆S term is defined as ∆S = 1/2SH2 – SH*, where SH* is the vibration entropy of adsorbed H and defined as S H * = k B ∑ ln(1 − e − hωi / kBT ) , i=N

where kB and T are Boltzmann constant and temperature respectively,

35,36

and SH2 is the entropy of H2

gas molecule obtained from the standard thermodynamic database.37 The detailed data about various energies and entropies have been listed in Tables S1-S3. All DFT calculations were performed using the projector augmented wave (PAW) method38 as implemented in the Vienna ab initio Simulation Package (VASP)39,40. The electronic exchangecorrelation energy was treated by generalized-gradient approximation (GGA) of Perdew-BurkeErnzerhof (PBE).41 A kinetic cutoff energy of 400 eV was used for the plane-wave expansion set. The kpoint sampling in the Brillouin zone was implemented by the Monkhorst-Pack scheme with the grids of 16×16×1, 4×4×1 and 1×6×1 for MoSe2 primitive cell, 7×7 supercells, and nanoribbons, respectively. For the geometry optimization, the convergence criteria of energy and forces acting on each atom were 10-3 eV and 10-2 eV/Å, respectively. Simulated scan tunneling microscopy (STM) images were obtained using the Tersoff-Hamann approximation based on calculated charge densities, and the STM tip height was set to ~1.8 Å in the simulation. The energy barriers of transition states for the understanding of HER mechanisms were determined using climbing image nudged elastic band (CLNEB) calculations.42 In order to reveal the HER dynamics on defect and edge sites of MoSe2, ab initio molecular dynamics simulations have been carried out. In the calculations, the MoSe2/electrolyte interfaces were modeled by putting one layer of water film on defective MoSe2 basal planes or MoSe2 nanoribbons, and the PBE functional with van der Waals (vdWs) correction (DFT-D2)43 was used for the H-bonding interactions between water molecules. The AIMD simulations were performed in the canonical (NVT) using 6000 time steps with a 1.0 fs time step at 300 K.

3. RESULTS AND DISCUSSION 3.1 Geometries and Stability. The geometries and stability of MoSe2 defect and edge structures are firstly investigated with the purpose of elucidating the most thermodynamically favorable structures in three groups of MoSe2 catalysts. The formation energies (Ef) of monolayer MoSe2 with various point defects and holes as a function of Se chemical potential have been shown in Figure 2a. Among all pointdefect structures, the Se vacancy (VSe) presents the lowest formation energy in the allowed Se chemical

5 Environment ACS Paragon Plus

ACS Applied Materials & Interfaces

potential range, suggesting the high stability in thermodynamics. The result is in good agreement with extensive experimental observations24,32 where the Se vacancies are dominant point defects in fabricated MoSe2 samples. In addition to VSe, VSe2 and SeMo are the other two relatively stable defects. Especially for VSe2, its formation energy is lower than twice as much as that of VSe. In other words, a VSe2 defect is more stable than that of two isolated VSe defects, thus it should be produced in the synthesized MoSe2 samples. For other point defects, such as VMo and MoSe, they are thermodynamically unfavorable due to the large formation energies. The holes in MoSe2 basal plane is a type of defective structures located between the vacancies and edges. Among considered hole structures, VMo3Se2 is the most stable one, in particular for the Se-rich condition. Compared to the point defects such as VSe, VMo3Se2 shows relatively larger formation energies due to the increase of defect atoms with the dangling bonds.

(a) 15

(b) e6 M oS

V

M o3

6

Se M

o

Mo Se

Se2

VMo

- Mo

0.9

0.6

Se-P

Se-R

V Se2

3

Mo edges Se edges

Se

VMoSe3

9

1.5

1.2

Ef (eV/Å)

V

12

Ef (eV)

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 6 of 22

0.3

Mo-P

Mo

Mo-R

V Se 0

-Se

0.0 -0.9 Mo-rich

-0.6

-0.3

∆µSe (eV)

0.0 Se-rich

-0.9 Mo-rich

-0.6

-0.3

∆µSe (eV)

0.0 Se-rich

Figure 2. (a) Formation energies (in eV) of MoSe2 basal plane with various point defects and holes as a function of Se chemical potential difference (∆µSe = µSe -µSe(bulk)). (b) Edge formation energies (in eV/Å) of various zigzag edges as a function of Se chemical potential difference ∆µSe. The blue and red lines denote the results of Mo and Se edges, respectively.

Among various potential MoSe2 edges, the most commonly encountered type is the zigzag edges which are easily formed during the growth. Therefore, we only focus on the stability of zigzag edges. Owing to the lack of inversion symmetry, there are two types of zigzag edges: the Mo edges and Se edges. Figure 2b shows formation energies of various zigzag MoSe2 edges as a function of Se chemical potential. For the Mo edges, the (2×1) reconstructed Mo edge (Mo-R) is the most stable under the Morich condition and the Mo-Se structure is energetically favorable under the Se-rich condition. For the Se edges, the (2×1) reconstructed Se edge (Se-R) is still the most stable edge configurations in the allowed range of chemical potential. Such a result is very similar to the case of MoS2 in which the reconstructed zigzag edges also exhibit high stability.44 6 Environment ACS Paragon Plus

Page 7 of 22

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Applied Materials & Interfaces

In order to further understand the structural characteristic of MoSe2 defect and edge structures, the STM images of thermodynamically favorable defect and edge configurations, including of point defects (VSe, VSe2 and SeMo), hole (VMo3Se2), and edges (Mo-R, Mo-Se and Se-R) have been computed, as shown in Figure 3. The VSe image indicates a triangular light spot around the S vacancy due to the electron localization of unsaturated Mo atoms (see Figure 3a). The triangular spot in the STM image of VSe2 is brighter because of the increase of dangling-bond electrons (Figure 3b), which contributes to the stronger binding of H atoms at the defect sites. The formation of SeMo leads to three bright points around the defect in the STM image (Figure 3c), implying that the nearest-neighboring Se atoms have higher activity to trap H atoms than the defect site during the HER. The STM image of VMo3Se2 indicates highly bright points around the hole (Figure 3d), suggesting that the hole edges are active for the HER. For three stable zigzag edges, the STM images of Mo-R and Se-R display bright spots at their edges, but the image of Mo-Se at its edge is dark and the highly bright spots appear at the second outermost edge. The result means that the Mo-R and Se-R edges should be active for capturing H atoms but the active sites of Mo-Se may be located at the second outermost edge.

Figure 3. Simulated STM images of defective MoSe2 basal planes with the configuration of (a) VSe, (b) VSe2, (c) SeMo, and (d) VMo3Se2, and MoSe2 edges with the configuration of (e) Mo-R, (f) Mo-Se and (g) Se-R, respectively. The simulated STM images are obtained at the voltages of -1.0 V using the constant-height mode.

3.2 Electrocatalytic Activity and Origin. Now we focus on the HER activity of various point defects, holes, and edges on monolayer MoSe2. The Gibbs free energy change for hydrogen adsorption (∆GH) is a good parameter for evaluating the HER activity. Smaller ∆GH will induce strong binding of adsorbed H, while larger ∆GH will make the protons bonded difficultly to the catalytic surface, both leading to the slow HER kinetics. Hence, an optimal electrocatalyst should facilitate HER with the value of ∆GH at the vicinity of zero (∆GH ~ 0). The calculated ∆GH of MoSe2 basal planes with different point defects and 7 Environment ACS Paragon Plus

ACS Applied Materials & Interfaces

holes as well as various MoSe2 edges is presented in Figure 4a. Compared to the perfect MoSe2 basal plane (∆GH = 1.78 eV), the formation of defects and edges leads to the significant reduction of ∆GH values, indicating the enhanced HER activity. Among various defect structures including point defects and holes, calculated ∆GH values of VMoSe3, VMoSe6, and SeMo are very negative, suggesting that the interaction of the catalysts with H is too strong to release H2 from the catalytic sites. In contrast, the ∆GH values of other defect structures (-0.09 eV~0.08 eV), especially for the VSe and VSe2 structures, close to zero, indicating the high HER activity. For the MoSe2 edges, both the reconstructed Mo and Se edges (i.e., Mo-R and Se-R) present the smaller ∆GH values (~0.04 eV) relative to other edge configurations, which should be responsible for the high catalytic activity of MoSe2 nanostructures reported in experiment.28,29,45 Compared to the widely studied MoS2 catalysts, the formation of similar vacancy defects and edges in MoSe2 exhibits lower ∆GH values (see Table S4). The result suggests the superior HER activity of MoSe2 catalysts relative to MoS2 catalysts, which is supported by recent experimental reports.29,46 Some theoretical investigations on the HER activity of MoSe2 and MoS2 edges have been also reported,30,47 but their calculated models were generally based on the directly cleaved edges without the edge reconstruction.

(b) VMo VSe2

-0.5 -1.0

Defects Edges

-1.5

Mo-Se

Se-Mo

10

VMoSe6

10

VSe VSe2

-0.2

-27

Defects Edges Pt(111)

-32

VMoSe3

0.0

-22

10

-1.0

(d)

0.6

∆GH (eV)

(c)

VMoSe6

i0 (A/site)

0.0

VSe2 VSe Mo-R VMo3Se2 VMo Se-R SeMo Se-P Mo-P

-17

10

VMo3Se2 Mo-P Mo-R Mo-Se Se-P Se-R Se-Mo

VSe Mo Se SeMo

∆GH (eV)

(a) 0.5

∆GH (eV)

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 8 of 22

0.4

-0.5

0.0 0.5 ∆GH (eV)

1.0

VMo3Se2 Mo-R Se-R

0.2

-0.4 0.0 -0.6 0

5 10 15 20 Defect concentration (%)

0

20

40

60

80

100

Coverage (%)

Figure 4 (a) The calculated ∆GH values of various point defects, holes, and edges on monolayer MoSe2. (b) Volcano plot between the exchange current density i0 and ∆GH. The star, triangular, and circle symbols represent the data of Pt(111), MoSe2 defective and edge structures, respectively. (c) The ∆GH values of VSe and VSe2 as a function of defect concentration. (d) The ∆GH values of VMo3Se2, Mo-R, and Se-R as a function of H coverage.

8 Environment ACS Paragon Plus

Page 9 of 22

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Applied Materials & Interfaces

To more directly compare the catalytic ability of HER for different defect and edge configurations, the exchange current densities (i0) are calculated based on the results of ∆GH. The exchange current density reflects the intrinsic rate of proton transfer from the solvent to the catalytic surface. If the proton transfer is exothermic (∆GH < 0), i0 can be expressed as48 i0 = −eko

1 1 + exp(−∆GH / kT )

(3)

where k0 is the rate constant (k0 = 200 s-1site-1) includes all effect relating to the reorganization of the solvent during the proton transfer to the catalytic surface, and k is Boltzmann constant. The hydrogen evolution was evaluated as standard conditions corresponding to pH = 0 with T = 300 K in the present study. If the proton transfer is endothermic (∆GH > 0), i0 will be expressed as

i0 = −eko

1 exp( −∆GH / kT ) 1 + exp(−∆GH / kT )

(4)

Based on eq. (3) and (4), the exchange current densities of various catalytic structures as a function of ∆GH are plotted in Figure 4b. The calculated i0 indicates a volcano curve with the change of ∆GH. We find that some defect and edge structures, such as VSe, VSe2, Mo-R, and Se-R, are located at the top of the volcano curve and exhibit even higher exchange current densities (~10-17 A/site) than the Pt catalyst due to their relatively lower ∆GH values relative to the Pt(111) surface (∆GH = -0.09 eV). Such a result confirms that the stable defects and edges in MoSe2 exhibit extremely high HER activity. Nevertheless, it should be emphasized that most of surface sites on Pt-based catalysts could be efficient catalytic sites, while the active sites of MoSe2 catalysts are mainly confined at the defect and edge sites. Owing to the coverage dependence of ∆GH, the ∆GH values of MoSe2 catalysts with different H coverage have been also investigated. For the point-defect structures (e.g., VSe and VSe2), the H coverage on these defect structures depends on the defect density. We find that increasing defect concentration can lead to the reduction of ∆GH (see Figure 4c) which implies the enhanced H adsorption on the defect sites of VSe and VSe2. The result originates from that the increase of defect concentration causes the large structural distortion around defect sites (or highly strained defect sites), which is responsible for the enhanced H adsorption strength. For the H adsorption on hole and edge structures (e.g., VMo3Se2, Mo-R and Se-R), ∆GH increases with the increase of H coverage (see Figure 4d), and the HER is easily driven by the H adsorption with 10%~20% coverage. In order to understand the origin of HER activity on various MoSe2 catalytic structures, we attempt to establish a relationship between the electrocatalytic activity and the intrinsic electronic structure due to the fact that the HER activity of a catalyst is determined fundamentally by its electronic properties. The d-band center model (εd) is still regarded as an efficient activity descriptor for the transition-metalbased catalysts.49 However, our result indicates that the calculated ∆GH does not exhibit a regular trend 9 Environment ACS Paragon Plus

ACS Applied Materials & Interfaces

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 10 of 22

with εd (see S4 of SI for the details). Therefore, the d-band center model is not suitable to describe the HER activity of MoSe2-based catalysts, similar result has also been found in other TMD-based HER catalysts.11,26 Here the failure of d-band center model originates from that this model refers to the collective contribution of d-electron states in the whole energy range to correlate with the reactivity. However, not every electronic state contributes equally to the surface bonding and reaction. Generally, the closer electronic states are to the Fermi level (EF), the greater their contribution to the bonding interaction.50,51 The reactivity of a catalyst is thus determined by both its density of states (D(E)) and a weight factor (w(E)) that quantifies the contribution of every electronic state to the surface reaction (see Figure 5a). For example, VSe displays higher distribution of density of states (DOS) around the Fermi level than the perfect basal plane (see Figure 5b), which is responsible for the enhanced surface activity of MoSe2 basal plane by the introduction of Se vacancy. Hence, the HER activity of a catalyst actually depends on the abundance of occupied states near the Fermi level (FA) that is obtained by the sum of the weight contribution of occupied states as follows, FA = ∫

EF

−∞

D( E )w( E )dE

(5)

The previous studies have validated that the derivative of the Fermi-Dirac function ( − fT′ ( E − EF ) ) is a -1 good weight function,51 and the Fermi-Dirac function is defined as fT ( E − EF ) = {exp[( E − EF ) / kT ] + 1} .

Using the − fT′ ( E − EF ) as the weight function ensures that the weight value reaches a maximum at EF and reduces to zero with the increase of |E-EF|, as shown in Figure 5b. The spreading of the w(E) is sensitive to the magnitude of kT that is identified reaching an optimal value at 0.4 eV. Based on above analysis, a Fermi-abundance (DF) model is proposed to describe the HER activity of MoSe2 catalysts as follows, EF

DF

∫ = ∫

ED( E ) fT′ ( E − EF )dE

−∞ EF −∞

(6)

D( E ) fT′ ( E − EF )dE

Compared to the d-band center model, Fermi-abundance model can quantify the main contribution of electronic states at the vicinity of Fermi level to the surface reactivity. Owing to the fact that density of states of MoSe2 catalysts near the Fermi level are mainly arising from Mo atoms, here D(E) is thus referred to the projected DOS of Mo atoms in these structures (see Figures S3 and S4). It needs to be mentioned that the DFT-GGA method can underestimate the band gap, so the Heyd-Scuseria-Ernzerhof screened hybrid functional (HSE06)52 and GGA+U method have been used to provide a comparison. However, we find that the band gap of MoSe2 monolayer (1.43 eV) calculated by the PBE functional more approaches the experimental value (1.58 eV)53 as compared to the band gap obtained by the HSE06 and GGA+U methods (see S6 of SI for the details). Thus, the PBE functional can provide a reasonable prediction for the electronic structures of MoSe2 defect and edge structures. 10 Environment ACS Paragon Plus

Page 11 of 22

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Applied Materials & Interfaces

Figure 5c presents DF values of various MoSe2 defect and edge structures as a function of ∆GH. The DF values of various MoSe2 structures indicate an approximately linear trend in relation to their ∆GH values, thus DF can be used as a good activity descriptor for the HER. A larger DF value means the high distribution of occupied states at the vicinity of the Fermi level, resulting in the catalyst having a strong surface activity to trap H atoms (e.g., VMoSe3 and VMoSe6). In contrast, a catalyst with a smaller DF value will have a relatively weak surface activity to capture H atoms (e.g., Mo-Se). However, a catalyst with too strong or too weak surface activity cannot lead to high HER activity based on the Sabatier principle. Therefore, the MoSe2 catalysts with the medium DF values in the range from -0.64 to -0.54 eV, such as VSe, VSe2, Mo-R, and Se-R, have higher HER activity (i.e., ∆GH ≈ 0). The result originates from that these catalysts can balance the H adsorption and the H2 desorption, facilitating the HER. To examine whether the Fermi-abundance (DF) model can be used to describe the HER activity of other catalysts, we plot DF values of various MoSe2 and MoS2 catalytic structures and Pt(111) surface as a function of ∆GH (see Figure S5). It can be found that there is a good linear relation between the DF and ∆GH values. Therefore, the Fermi-abundance (DF) model is of generality for describing the HER activity of catalysts.

Figure 5 (a) The schematic abundance of occupied sates (FA) defined as the sum of weight contribution of occupied density of states, here the derivative of Fermi-Dirac function is used as the weight function. (b) The density of states of MoSe2 basal plane without and with a Se vacancy. (c) The Fermi-abundance (DF) values of various MoSe2 catalytic structures as a function of ∆GH, and the DF value of Pt(111) surface has been indicated here.

11 Environment ACS Paragon Plus

ACS Applied Materials & Interfaces

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 12 of 22

Based on above results, MoSe2 catalytic structures exhibit an interesting activity trend: most of thermodynamically stable defect and edge configurations, such as VSe, VSe2, VMo3Se2, Mo-R, and Se-R, possess high HER activity, while these energetically unfavorable structures, such as VMoSe3, VMoSe6, MoP, and Se-Mo, indicating weak HER activity. This originates from that the highly stable point defects, holes, and edges can moderately modify electronic structures of MoSe2, facilitating the HER process, while the catalytic sites of these unstable MoSe2 structures are too strong interaction with H atoms, leading to the weak HER activity. Such a result suggests the huge potential of MoSe2 defects and edges as the highly active HER electrocatalysts.

3.3 HER Mechanisms and Reaction Dynamics. The understanding of HER mechanisms and dynamic process is crucial for enhancing the HER performance of catalysts. The catalyst-supported HER is a multistep electrochemical process. As shown in Figure 6a, the first step is the hydrogen adsorption onto a MoSe2 catalyst (i.e., Volmer reaction), which is described by H+ + e- + *  H*, where the prefix of * denotes the adsorption site of the catalyst. The subsequent step is the release of H2 molecules from the catalyst following either the Heyrovsky (H* + H+ + e-  H2 + *) or Tafel (2H*  H2 + 2*) reaction. To determine the HER pathway, the relative adsorption free energies (∆G) between H2* and 2H* on various MoSe2 catalysts are compared. Figure 6b shows ∆G values of 2H* and H2* on the stable MoSe2 defect and edge structures. We find that the ∆G values of 2H* are correspondingly lower than that of H2* on VSe and VSe2 catalysts, implying that the HER on the two catalysts prefers to the Volmer-Tafel (V-T) mechanism. For other MoSe2 catalysts including hole and edge structures, calculated ∆G values indicate that the HER following the Volmer-Heyrovsky (V-H) mechanism is preferred (see Figure 6b). Overall, these highly active catalysts mentioned above, such as VSe, VSe2, Mo-R and Se-R, present smaller ∆G values (< 0.25 eV) in the whole HER.

Figure 6. (a) Schematic illustration for the HER following the Volmer-Heyrovsky pathway and the Volmer-Tafel pathway on MoSe2 catalytic surface. (b) The change of Gibbs free energies (∆G) of 2H* and H2* on various MoSe2 catalytic structures.

12 Environment ACS Paragon Plus

Page 13 of 22

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Applied Materials & Interfaces

Figure 7. The energy profiles for the Tafel reaction and Heyrovsky reaction on MoSe2 defects and edges. (a) The reactions involved in the recombination of two H atoms in a Se vacancy (VSe), (b) the reactions involved in the recombination of three H atoms in a Se vacancy, (c) the reactions occur at the Mo-R edge, and (d) the reactions occur at the Se-R edge. The numbers indicate the energy barriers (in eV) for the reactions.

In addition to the consideration of thermodynamics, the reaction kinetics is also important for the understanding of the HER pathway on MoSe2 catalysts. Figure 7 shows the energy profiles for the Tafel and Heyrovsky reactions on the stable MoSe2 defect (i.e., VSe) and edges (i.e., Mo-R, and Se-R). For the Heyrovsky reaction, the release of a H2 molecule involved in a proton reacted with an adsorbed H needs to overcome a very small energy barrier (0.04 eV) on VSe, while the recombination of 2H* on VSe in the Tafel reaction induces the energy barrier of 0.61 eV (see Figure 7a). It seems that the V-H mechanism is preferred on VSe. However, it can be found that the energy of 2H* is 0.4 eV lower than that of H2. Moreover, the transition from H2 to 2H* on a Se vacancy only needs to overcome an energy barrier of 0.21 eV. In other words, it is easily dissociated into 2H* even if H2 is formed in Se vacancies. When the third proton attacks the Se vacancy with 2H*, the energy barrier for producing a H2 molecule is very small in both the Heyrovsky (0.09 eV) and Tafel (0.02 eV) reactions (Figure 7b), respectively. However, it will form a 3H* metastable state along the Heyrovsky reaction pathway. The result means that the HER on VSe prefers to the V-T mechanism. On the Mo-R and Se-R structures, there is nearly no barrier for releasing a H2 molecule along the Heyrovsky reaction pathway (see Figure 7c and 7d). In contrast, 13 Environment ACS Paragon Plus

ACS Applied Materials & Interfaces

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 14 of 22

the energy barrier for the recombination of 2H* into H2 is 0.73 eV and 0.84 eV for the Mo-R and Se-R, respectively. Although the stability of H2 is higher than 2H* on the Se-R structure (see Figure 7d), the formation of H2 doesn’t induce the dissociation into 2H* due to the lack of space limit as appeared in the Se vacancy. Hence, the V-H mechanism is preferred at the Mo-R and Se-R edges. To gain more straightforward insight into the HER mechanisms on MoSe2 catalysts, the dynamic HER process on three stable catalytic structures (i.e., VSe, VMo3Se2, and a MoSe2 nanoribbon with Mo-R and Se-R edges) is also explored by AIMD simulations. The AIMD simulations were carried out at 300 K within 6 ps. The MoSe2-water interface (or the Helmholtz layer) is simulated by putting one water layer on MoSe2 basal planes or nanoribbons with about 3 Å interlayer spacing, and the protons (or free H atoms) with 25% coverage are dispersed into the water layer. Figure 8 shows a few snapshots of trajectories for the HER on the VSe structure. In order to increase the reaction probability, we introduce three Se vacancies into the MoSe2 basal plane and each Se vacancy includes an adsorbed H atom (i.e., Volmer reaction has finished). We find that free H atoms do not directly attack the adsorbed H (or H* atoms) to form H2 molecules but tend to firstly adsorbed at the defect sites in the initial 0.4 ps. When the reaction time is beyond 1 ps, we can find the release of H2 from the catalytic surface by the bonding of surface H* atoms. A real-time movie for the HER process on VSe is shown in Movie S1 of SI. The result suggests that the HER on VSe following the Volmer-Tafel mechanism is kinetically preferred.

Figure 8. Snapshots of trajectories for the HER on the VSe structure following 6 ps AIMD simulation at 300 K. The blue, orange, red, green, and white balls represent Mo, Se, O, H atoms in water, and reacted H atoms, respectively.

For the AIMD simulations on the VMo3Se2 structure and the MoSe2 nanoribbon, the similar models have been employed including the adsorbed H atoms located at the active sites (see Figure 9). In the HER process, most of H protons tend to be adsorbed at the active sites on both of two structures and there is no production of H2 molecules by the reaction of H* atoms within 6 ps. Instead, we find the formation of a H2 molecule by the Heyrovsky reaction (i.e., H* + H+ + e-  H2 + *) at the Se edge of 14 Environment ACS Paragon Plus

Page 15 of 22

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Applied Materials & Interfaces

MoSe2 nanoribbon at ~0.5 ps (see Figure 9b). Moreover, we observe the formation of hydronium ions (H3O+) on both of two catalytic structures. Owing to the H proton of H3O+ can be transferred in the water layer (see Movies S2 and S3), thus it facilitates the Heyrovsky reaction by H* + H3O+ + e-  H2 + H2O + *. The similar mechanism has been found in the HER process on 1T-MoS2 catalyst.54 Although the Heyrovsky reaction has not been directly observed on the VMo3Se2 structure (see Figure 9a), it may derive from the limited simulation time (6 ps).

Figure 9. Snapshots of trajectories for the HER on (a) VMo3Se2 and (b) MoSe2 nanoribbon with the Mo-R and Se-R edges following 6 ps AIMD simulation at 300 K. The blue, orange, red, green, and white balls represent Mo, Se, O, H atoms in water, and reacted H atoms, respectively.

The above HER mechanisms on MoSe2 catalytic structures can be understood as follows: On VSe and VSe2 surface, the protons are adsorbed at the defect sites in the initial HER process. Owing to the small space of Se vacancies, the distance of the neighboring adsorbed H atoms is very small (~ 1.7 Å) and thus they have the large chance to react into H2 molecules. Hence, the Volmer-Tafel mechanism is preferred for the HER on the VSe and VSe2 structures. On the hole and edge structures (e.g., VMo3Se2, MoR and Se-R edges), the distance of neighboring adsorbed H atoms is largely equal to the spacing of two edge sites (~ 3.3 Å). Thanks to the larger H-H distance, two neighboring H* atoms are difficult to react and bond into a H2 molecule in the HER. Therefore, the Volmer-Tafel mechanism is unfavorable on the VMo3Se2 structure and MoSe2 edges. In contrast, the generation of H2 on hole and edge structures can be 15 Environment ACS Paragon Plus

ACS Applied Materials & Interfaces

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 16 of 22

achieved by the adsorbed H atoms with attacked protons and H3O+ ions. Hence, the Volmer-Heyrovsky mechanism is preferred for the HER on hole and edge structures.

4. CONCLUSIONS In summary, we performed systematic DFT calculations to gain deep insight into the structural stability, electrocatalytic activity, and HER mechanisms on a series of MoSe2 catalytic structures including point defects, holes, and edges. Our results indicate that thermodynamically stable defect structures (i.e., VSe, VSe2, SeMo, and VMo3Se2) and edge configurations (i.e., Mo-R and Se-R) are very active for the HER as comparable to the Pt-based catalysts, which originates from that the formation of defects and edges moderately modified electronic structures of MoSe2 facilitating the H adsorption and the H2 desorption. The fundamental correlation between the HER activity of MoSe2 catalytic structures and their intrinsic electronic structures can be understood by a Fermi-abundance model, and this model is also suitable to describe the HER activity of other types of electrocatalysts, such as MoS2 and Pt. Furthermore, we find that there are two different HER mechanisms on the MoSe2 catalytic structures: the Volmer-Tafel mechanism is preferred for VSe and VSe2 catalysts, while the Volmer- Heyrovsky mechanism is more favorable for other types of MoSe2 catalysts. Our work suggests that monolayer MoSe2 with high density of intrinsic defects and edges can be potential candidate as the highly active HER catalyst and lays out an effective scheme for assessing the HER activity of similar classes of materials (e.g., other TMD materials).

 ASSOCIATED CONTENT  Supporting Information H adsorption configurations on MoSe2 defect and edge structures, DFT energies, entropies, and ∆GH on various MoSe2 defect and edge structures, the comparison of HER activity between MoSe2 and MoS2 catalysts, the d-band centers of MoSe2 defect and edge structures, projected DOS of Mo atoms and DF values of MoSe2 defect and edge structures, the bandgap comparison of monolayer MoSe2 calculated by using different methods, and the validation of the adaptability of Fermi-abundance model. This material is available free of charge via the internet at http://pubs.acs.org.

 AUTHOR INFORMATION Corresponding Authors *E-mail: [email protected] 16 Environment ACS Paragon Plus

Page 17 of 22

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Applied Materials & Interfaces

Notes The authors declare no competing financial interest.

 ACKNOWLEDGMENT This work was supported in part by the National Natural Science Foundation of China (Grant no.

61775201, 11404309, and 51402275) and the Fund of Shanghai Science and Technology Foundation (Grant no. 13JC1408800). Computational resources from the Shanghai Supercomputer Center are acknowledged.

REFERENCES AND NOTES (1) Turner, J. A. Sustainable Hydorgen Production. Science 2004, 305, 972-974. (2) Lu. Q.; Yu, Y.; Ma, Q.; Chen, B.; Zhang, H. 2D Transition-Metal-Dichalcogenide-Nanosheet-Based Composites for Photocatalytic and Electrocatalytic Hydrogen Evolution Reactions. Adv. Mater. 2016, 28, 1917-1933. (3) Greeley, J.; Jaramillo, T. F.; Bonde, J.; Chorkendorff, I.; Norskov, J. K. Computational High-Throughput Screening of Electrocatalytic Materials for Hyrogen Evolution. Nat. Mater. 2006, 5, 909-913. (4) Yan, J.; Yao, Z.; Mietek, J.; Qiao, S. Z. Design of Electrocatalysts for Oxygen- and Hydrogen-Involving Energy Conversion Reactions. Chem. Soc. Rev. 2015, 44, 2060-2086. (5) Yin, H.; Zhao, S.; Zhao, K.; Muqsit, A.; Tang, H.; Chang, L.; Zhao, H.; Gao, Y.; Tang, Z. Ultrathin Platinum Nanowires Grown on Single-Layered Nickel Hydroxide with High Hydrogen Evolution Activity. Nat. Comm. 2015, 6, 6430. (6) Zheng, Y. ; Jiao, Y. ; Jaroniec, M. ; Qiao, S. Z. Advancing the Electrochemistry of the Hydrogen Evolution Reaction through Combining Experiment and Theory. Angew. Chem. Int. Ed. 2015, 54, 52-65. (7) Song, B.; Wang, C.; Deng, M.; Gong, M.; Bai, Y.; Jiang, J.; Xiong, Y. Surface Polarization Matters: Enhancing the Hydrogen-Evolution Reaction by Shrinking Pt Shells in Pt-Pd-Graphene Stack Structure. Angew. Chem. Int. Ed. 2014, 53, 12120-12124. (8) Jarmillo, T. F.; Jørgensen, K. P.; Bonde, J.; Nielsen, J. H.; Horch, S.; Chorkendorff. I. Identification of Active Edge Sites for Electrochemical H2 Evolution from MoS2 Nanocatalysts. Science 2007, 317, 100-102. (9) Hinnemann, B.; Moses, P. G.; Bonde, J.; Jørgensen, K. P.; Nielsen, J. H.; Horch, S.; Chorkendorff. I.; Nørskov, J. K. Biomimetic Hydrogen Evolution: MoS2 Nanoparticles as Catalyst for Hydrogen Evolution. J. Am. Chem. Soc. 2005, 127, 5308-5309.

17 Environment ACS Paragon Plus

ACS Applied Materials & Interfaces

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

(10)

Page 18 of 22

Qu, B.; Yu, X.; Chen, Y.; Zhu, C.; Li, C.; Yin, Z.; Zhang X. Ultrathin MoSe2 Nanosheets Decorated on

Carbon Fiber Cloth as Binder-Free and High-Performance Electrocatalyst for Hydrogen Evolution. ACS Appl. Mater. Interfaces 2015, 7, 14170-14175. (11)

Zhang, Y.; Chen X.; Huang, Y.; Zhang, C.; Li, F.; Shu, H. The Role of Intrinsic Defects in Electrocatalytic

Activity of Monolayer VS2 Basal Planes for the Hydrogen Evolution Reaction. J. Phys. Chem. C 2017, 121, 15301536. (12)

Wang, F.; Shifa, T. A.; Zhan, X.; Huang, Y.; Liu, K.; Cheng, Z.; Jiang, C.; He, J. Recent Advances in

Transition-Metal Dichalcogenide Based Nanomaterials for Water Splitting. Nanoscale 2015, 7, 19764-19788. (13)

Kibsgaard, J.; Chen, Z.; Reinecke, B. N.; and Jaramillo, T. F. Engineering the Surface Structure of MoS2 to

Preferentially Expose Active Edge Sites for Electrocatalysis. Nat. Mater. 2012, 11, 963-969. (14)

Ye, G.; Gong, Y.; Lin, J.; Li, B.; He, Y.; Pantelides, S. T.; Zhou, W.; Vajtai, R.; Ajayan, P. M. Defects

Engineered Monolayer MoS2 for Improved Hydrogen Evolution Reaction. Nano Lett. 2016, 16, 1097-1103. (15)

Li, Y.; Wang, H.; Xie, L.; Liang, Y.; Hong, G.; Dai, H. MoS2 Nanoparticles Grown on Graphene: An

Advanced Catalyst for the Hydrogen Evolution Reaction. J. Am. Chem. Soc. 2011, 133, 7296-7299. (16)

Chen, Z.; Cummins, D.; Reinecke, B. N.; Clark, E.; Sunkara, M. K.; Jaramillo, T. F. Core-Shell MoO3-

MoS2 Nanowires for Hydrogen Evolution: A Functional Design for Electrocatalytic Materials. Nano Lett. 2011, 11, 4168-4175. (17)

Zhang, Y.; Gong, Q.; Li, L.; Yang, H.; Li, Y.; Wang, Q. MoSe2 Porous Microspheres Comprising

Monolayer Flakes with High Electrocatalytic Activity. Nano Res. 2015, 8, 1108-1115. (18)

Hu, J.; Huang, B.; Zhang, C.; Wang, Z.; An, Y.; Zhou, D.; Lin, H.; Leung, M. K. H.; Yang, S.

Engineering Stepped Edge Surface Structures of MoS2 Sheet Stacks to Accelerate the Hydrogen Evolution Reaction. Energy Environ. Sci. 2017, 10, 593-603. (19)

Wang, H.; Kong, D.; Johanes, P.; Cha, J. J.; Zheng, G.; Liu, N.; Cui, Y. MoSe2 and WSe2 Nanofilms

with Vertically Aligned Molecular Layers on Curved and Rough Surfaces. Nano Lett. 2013, 13, 3426-3433. (20)

Voiry, D.; Salehi, M.; Silva, R.; Fujita, T.; Chen, M.; Asefa, T.; Shenoy, V.; Eda, G.; Chhowalla, M.

Conducting MoS2 Nanosheets as Catalysts for Hyrogen Evolution Reaction. Nano Lett. 2013, 13, 6222-6227. (21)

Lukowski, M. A.; Daniel, A. S.; Meng, F.; Forticaux, A.; Li, L.; Jin, S. Enhanced Hydrogen Evolution

Catalysis from Chemically Exfoliated Metallic MoS2 Nanosheets. J. Am. Chem. Soc. 2013, 135, 1027410277.

18 Environment ACS Paragon Plus

Page 19 of 22

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Applied Materials & Interfaces

(22)

Calandra, M. Chemically Exfoliated Single-layer MoS2: Stability, Lattice Dynamics, and Catalytic

Adsorption from First Principles. Phys. Rev. B 2013, 88, 245428. (23)

Li, G.; Zhang, D.; Qiao, Q.; Yu, Y.; Peterson, D.; Zafar, A.; Kumar, R.; Curtarolo, S.; Hunte, F.;

Shannon, S.; Zhu, Y.; Yang, W.; Cao, L. All the Catalytic Active Sites of MoS2 for Hyrogen Evolution. J. Am. Chem. Soc. 2016, 138, 16632-16638. (24)

Zhou, X.; Jiang, J.; Ding, T.; Zhang, J.; Pan, B.; Zuo, J.; Yang, Q. Fast Colloidal Synthesis of Scalable

Mo-Rich Hierarchical Ultrathin MoSe2-x Nanosheets for High-Performance Hydrogen Evolution. Nanoscale 2014, 6, 11046-11051. (25)

Xie, J.; Zhang, H.; Li, S.; Wang, R.; Sun, X.; Zhou, M.; Zhou, J.; Lou, X. W.; Xie, Y. Defect-Rich MoS2

Ultrathin Nanosheets with Additional Active Edge Sites for Enhanced Electrocatalytic Hydrogen Evolution. Adv. Mater. 2013, 25, 5807-5813. (26)

Ouyang, Y.; Ling, C.; Chen, Q.; Wang, Z.; Shi, L.; Wang, J. Activating Inert Basal Planes of MoS2 for

Hydrogen Evolution Reaction Through the Fromation of Different Intrinsic Defects. Chem. Mater. 2016, 28, 4390-4396. (27)

Li, H.; Tsai, C.; Koh, A. L.; Cai, L.; Contryman, A. W.; Fragapane, A. H.; Zhao, J.; Han, H. S.;

Manoharan, H. C.; Abild-Pedersen, F.; Nørskov, J. K.; Zheng, X. Activating and Optimizing MoS2 basal Planes for Hydrogen Evolution Through the Fomation of Strained Sulphur Vancancies. Nat. Mater. 2016, 15, 48-53. (28)

Eftekhari, A. Molybdenum Diselenide (MoSe2) for Energy Storage, Catalysis, and Optoelectronics. Appl.

Mater. Today 2017, 8, 1-17. (29)

Tang, H.; Dou, K.; Kaun, C.-C.; Kuang, Q.; Yang, S. MoSe2 Nanosheets and Their Graphene Hybrids:

Synthesis, Characterization and Hydrogen Evolution Reaction Studies. J. Mater. Chem. A 2014, 2, 360-364. (30)

Tsai, C.; Chan, K.; Abild-Pedersen, F.; Nørskov, J. K. Active Edge Sites in MoSe2 and WSe2 Catalysts

for the Hydrogen Evolution Reaction: A Density Functional Study. Phys. Chem. Chem. Phys. 2015, 16, 13156-13164. (31)

Brixner, L. H. Preparation and Properties of the Single Crystalline AB2-Type Selenides and Tellurides of

Niobium, Tantalum, Molybdenum and Tungsten. J. Inorg. Nucl. Chem. 1962, 24, 257-263.

19 Environment ACS Paragon Plus

ACS Applied Materials & Interfaces

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

(32)

Page 20 of 22

Lehtinen, O.; Komsa, H.-P.; Pulkin, A.; Whitwick, M. B.; Chen, M.-W.; Lehnert, T.; Mohn, M. J.;

Yazyev, O. V.; Kis, A.; Kaiser, U.; Krasheninnikov, A. V. Atomic Scale Microstructure and Properties of SeDeficient Two-Dimensional MoSe2. ACS Nano 2015, 9, 3274-3283. (33)

Cao, D.; Shen, T.; Liang, P.; Chen, X.; Shu, H. Role of Chemical Potential in Flake Shape and Edge

Properties of Monolayer MoS2. J. Phys. Chem. C 2015, 119, 4294-4301. (34)

Greeley, J.; Jaramillo, T. F.; Bonde, J.; Chorkendorff, I.; Nørskov, J. K. Computational High-Throughput

Screening of Electrocatalytic Materials for Hydrogen Evolution. Nat. Mater. 2006, 5, 909-913. (35)

Reuter, K.; Scheffler, M. Composition, Structure, and Stability of RuO2(110) as a Function of Oxygen

Pressure. Phys. Rev. B 2001, 65, 035406. (36)

Shu, H. B.; Tao, X.-M.; Ding, F. What are the Active Carbon Species During Graphene Chemical Vapor

Deposition Growth? Nanoscale 2015, 7, 1627-1634. (37)

Cramer, C. J. Essentials of Computational Chemistry Theories and Models, 2nd ed.; John Wiley & Sons,

Ltd.: West Sussex, England, 2004. (38)

Kresse, G.; Joubert, D. From Ultrasoft Pseudopotentials to the Projector Augmented-Wave Method.

Phys. Rev. B 1999, 59, 1758. (39)

Blöchl, P. E. Projector Augmented-Wave Method. Phys. Rev. B 1994, 50, 17953-17979.

(40)

Kresse, G.; Furthmüller, J. Efficient Iterative Schemes for ab Initio Total-Energy Calculations Using a

Plane-Wave Basis Set. Phys. Rev. B 1996, 54, 11169-11185. (41)

Payne, M. C.; Teter, M. P.; Allan, D. C.; Arias, T. A.; Joannopoulos, J. D. Iterative Minimization

Technologies for Ab Initio Total-Energy Calcualtions: Molecular Dynamics and Conjugate Gradients. Rev. Mod. Phys. 1992, 64, 1045-1097. (42)

Henkelman, G.; Uberuaga, B. P.; Jónsson, H. A Climbing Image Nudged Elastic Band Method for

Finding Saddle Points and Minimum Energy Paths. J. Chem. Phys. 2000, 113, 9901-9904. (43)

Grimme, S. Semiempirical GGA-Type Density Functional Constructed with a Long-Range Dispersion

Correction. J. Comput. Chem. 2006, 27, 1787-1799. (44)

Cui, P.; Choi, J.-H.; Chen, W.; Zeng, J.; Shih, C.-K.; Li, Z.; Zhang, Z. Contrasting Structural

Reconstructions, Electronic Properties, and Magnetic Orderings along Different Edges of Zigzag Transition Metal Dichalcogendide Nanoribbons. Nano Lett. 2017, 17, 1097-1101.

20 Environment ACS Paragon Plus

Page 21 of 22

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ACS Applied Materials & Interfaces

(45)

Mao, S.; Wen, Z.; Ci, S.; Guo, X.; Ostrikov, K.; Chen, J. Perpendicularly Oriented MoSe2/Graphene

Nanosheets as Advanced Electrocatalysts for Hydrogen Evolution. Small 2015, 11, 414-419. (46)

Wiensch, J. D.; John, J.; Velazquez, J. M.; Torelli, D. A.; Pieterick, A. P.; McDowell, M. T.; Sun, K.;

Zhao, X.; Brunschwig, B. S.; Lewis, N. S. Comparative Study in Acidic and Alkaline Media of the Effects of pH and Crystallinity on the Hydrogen-Evolution Reaction on MoS2 and MoSe2. ACS Energy Lett. 2017, 2, 2234-2238. (47)

Tsai, C.; Chan, K.; Nørskov, J. K.; Abild-Pedersen, F. Theoretical Insights into the Hydrogen Evolution

Activity of Layered Transition Metal Dichalcogenides. Surf. Sci. 2015, 640, 133-140. (48)

Nørskov, J. K.; Bligaard, T.; Logadottir, A.; Kitchin, J. R.; Chen, J. G.; Pandelov, S.; Stimming, U.

Trends in the Exchange Current for Hydrogen Evolution. J. Electrochem. Soc. 2005, 152, J23-J26. (49)

Hammer, B.; Nørskov, J. K. Electronic Factors Determining the Reactivity of Metal Surfaces. Surf. Sci.

1995, 343, 211-220. (50)

Li, F.; Shu, H.; Liu, X.; Shi, Z.; Liang, P.; Chen, X. Electrocatalytic Activity and Design Principles of

Heteroatom-Doped Graphene Catalysts for Oxygen-Reduction Reaction. J. Phys. Chem. C 2017, 121, 1443414442. (51)

Huang, B.; Xiao, L.; Lu, J.; Zhuang, L. Spatially Resolved Quantification of the Surface Reactivity of

Solid Catalysts. Angew Chem. Int. Ed. 2016, 55, 6239-6243. (52)

Paier, J.; Marsman, M.; Hummer, K.; Kresse, G.; Gerber, I. C.; Ángyán, J. G. Screened Hybrid Density

Functinal Applied to Solids. J. Chem. Phys. 2006, 124, 154709. (53)

Zhang, Y.; Chang, T.-R.; Zhou, B.; Cui, Y.-T.; Yan, H.; Liu, Z.; Schmitt, F.; Lee, J.; Moore, R.; Chen, Y.;

Lin, H.; Jeng, H.-T.; Mo, S.-K.; Hussain, Z.; Bansil, A.; Shen, Z.-X. Direct Observation of the Transition from Indirect to Direct Bandgap in Atomically Thin Epitaxial MoSe2. Nat. Nanotechnol. 2014, 9, 111-115. (54)

Tang, Q.; Jiang, D. Mechanism of Hydrogen Evolution Reaction on 1T-MoS2 from First Principles. ACS

Catal. 2016, 6, 4953-4961.

21 Environment ACS Paragon Plus

ACS Applied Materials & Interfaces

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

TOC Image

22 Environment ACS Paragon Plus

Page 22 of 22