Ab initio Simulation Explains the Enhancement of Catalytic Oxygen

14. The scarcity of non-renewable fossil fuels, along with the increasing demand for energy has. 15 attracted significant attention to the development...
5 downloads 16 Views 4MB Size
Subscriber access provided by READING UNIV

Article

Ab initio Simulation Explains the Enhancement of Catalytic Oxygen Evolution on CaMnO 3

Tian Qiu, Bingtian Tu, Diomedes Saldana-Greco, and Andrew M. Rappe ACS Catal., Just Accepted Manuscript • DOI: 10.1021/acscatal.7b03987 • Publication Date (Web): 31 Jan 2018 Downloaded from http://pubs.acs.org on January 31, 2018

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 Catalysis 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 19 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 Catalysis

Ab initio Simulation Explains the Enhancement of Catalytic Oxygen Evolution on CaMnO3 Tian Qiu,† Bingtian Tu,‡ Diomedes Saldana-Greco,† and Andrew M. Rappe∗,† 1

†Department of Chemistry, University of Pennsylvania, Philadelphia, PA 19104-6323, USA ‡State Key Lab of Advanced Technology for Materials Synthesis and Processing, Wuhan University of Technology, Wuhan, China E-mail: [email protected]

Abstract

2

3

Experimental results have shown promising catalytic activity for the oxygen evo-

4

lution reaction (OER) on the perovskite–type material CaMnO3 . Through density

5

functional theory investigation, we study the OER mechanism on CaMnO3 , based on a

6

thermodynamic stability approach. Our results reveal that the formation of Mn vacan-

7

cies caused by the solubility of Mn enhances lattice oxygen activity, which then reduces

8

the energy of the adsorbate *OOH, and therefore loosens the lower overpotential limit

9

predicted from the “scaling relationship”. This effect suggests that by doping mangan-

10

ite oxides with soluble elements, we could enhance their OER catalytic activities due

11

to the high surface lattice oxygen activity induced by surface metal ion vacancies.

12

Keywords: electrocatalysis, oxygen evolution, calcium manganite, density functional theory,

13

aqueous surface phase diagram

1

ACS Paragon Plus Environment

ACS Catalysis 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

14

Introduction

15

The scarcity of non-renewable fossil fuels, along with the increasing demand for energy has

16

attracted significant attention to the development of sources of alternative and renewable en-

17

ergy. One of the most promising alternatives is the field of electrochemical energy conversion,

18

such as fuel cells, metal-air batteries, and electrolysis. It is clear that reducing the energy

19

loss of the oxygen evolution reaction (OER) and the oxygen reduction reaction (ORR) is

20

vital in enhancing overall electrochemical energy conversion efficiency. For the electrolysis

21

of water, IrO2 and RuO2 have been reported as promising electrocatalysts due to their high

22

activity for OER. 1–3 However, since these materials are rare and expensive, current research

23

is focused on searching for inexpensive alternative materials with comparable or even higher

24

catalytic activity. Manganese compounds have shown significant advantages in this compe-

25

tition. Their high abundance and relatively low price ensure their availability for large-scale

26

energy conversion. Studies of the Mn4 O4 -cubane show its crucial role in photosynthesis, 4

27

and recent work shows that nanostructured α-Mn2 O3 is an excellent bi-functional catalyst

28

for OER and ORR. 5 Perovskites with Mn+3/+4 as B–site cations also exhibit impressive

29

catalysis activity. Du et al. 6 and Kim et al. 7 have found that the required OER potential on

30

CaMnO3−x decreases from URHE ≈ 1.6 V to URHE ≈ 1.5 V when x increases from 0 to 0.25,

31

where x stands for vacancy cocentration, RHE stands for “reversible hydrogen electrod”,

32

and URHE stands for electrode potential reference to RHE Noting that the required OER

33

potential on IrO2 is also 1.5 V, it suggests that manganites can provide the highest level of

34

OER catalysis activity. However, to our knowledge, there is no systematic theoretical work

35

investigating the OER process on CaMnO3 . In order to make further development designing

36

new materials, a thorough understanding of the catalytic mechanism is required.

37

Previous work has shown that there are at least two types of OER mechanisms, 8–10

38

namely the adsorbate–evolution mechanism (AEM), and the Mars–van Krevelen type mech-

39

anism (MKM). In a common AEM, oxygen is evolved at one active site on the surface

40

through the intermediates of *OH, *O, and *OOH, and then is released as O2 . In MKM, 2

ACS Paragon Plus Environment

Page 2 of 19

Page 3 of 19 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 Catalysis

41

the lattice oxygen can directly participate in the formation of O2 , and in some cases two

42

cooperative sites could be involved in the formation of a single O2 molecule. Since surfaces of

43

perovskite-type materials may undergo reconstructions in aqueous environment, 11 multiple

44

local geometries at the active site could be expected on different materials or under differ-

45

ent conditions. Therefore, exploring the OER mechanism on reconstructed perovskite-type

46

materials may disclose new features that provide novel opportunities for enhanced catalysis.

47

The required electrode potential for OER is the minimal voltage under which all four

48

steps of intermediate evolution are spontaneous. Knowing that the Gibbs free energy change

49

of the total oxygen evolution reaction 2H2 O → 4H+ + 4e− + O2 is independent of the

50

reaction path, the required electrode potential reaches its minimum (URHE = 1.23 V) if

51

four intermediates are evenly spaced on the free energy scale, i.e. all four single-electron

52

transfer reactions have the same reaction free energy. Any deviation from the evenly-spaced

53

case causes at least one single-electron transfer reaction having higher reaction free energy

54

than the average, and therefore requires extra electrode potential referring as “overpotential”

55

(η) to be spontaneous 9,12,13 . η is directly related to energy losses during the electrochemical

56

reaction. Recent theoretical studies of AEM have stated that the adsorption energy difference

57

between *OOH and *OH is approximately 3.2 eV regardless of metal ion identity at the

58

active site. 9 A direct conclusion from this “scaling relationship” is that there would be a

59

lower limit for η: η > (3.2/2 − 1.23) = 0.37 (eV). However, since surfaces of perovskite-type

60

materials can undergo reconstructions, it is possible that the local chemical environment at

61

the active site on the surface influences *OH and *OOH in a different way such that this

62

scaling relationship may not apply.

63

Since the thermodynamically stable (oxy)hydroxides are the most relevant structures in

64

OER 10,14 , we follow the thermodynamic–stability based approach to study the OER mech-

65

anism on the perovskite-type material CaMnO3 . The adsorbates could still be identified as

66

*OH, *O, and *OOH on different surface sites to some extent, but the high activity of the

67

lattice oxygens makes them able to host hydrogen atoms as well as interact with protons

3

ACS Paragon Plus Environment

ACS Catalysis 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 19

68

from the adsorbates. We notice that the proton from *OOH has a strong interaction with

69

the lattice oxygen, and even transfers to the lattice oxygen site, while the proton from *OH

70

is less active and could not be further stabilized by transferring to the lattice oxygen site.

71

Since this process lowers the energy of *OOH, and consequently reduces the adsorption en-

72

ergy difference between *OOH and *OH, it may guide the design of catalysts with lower

73

overpotential for OER.

74

Methods and Results

75

Prior to the investigation of the OER mechanism, the bulk stability region of CaMnO3 in

76

aqueous environment must be evaluated in order to estimate the concentrations of aqueous

77

ions involved in further simulations. The scheme for computing the bulk stability region

78

is well established, 11 based on finding the pH-U region such that the following reaction is

79

spontaneous:

X A

+ − nA Hx AOz− y −→ CaMnO3 + nH2 O H2 O + nH+ H + ne e

(1)

80

Where the sum accounts for all elements “A” in the bulk, nA , nH2 O , nH+ , and ne stands

81

for number of A element, H2 O, H+ , and e− involved in the reaction, respectively. Hx AOz− y

82

is the most thermodynamically stable aqueous form of “A” under given pH and U. Its exact

83

form is determined by selecting the species with the lowest formation energy, according to

84

the following equation:

+ − A + nH2 O H2 O −→ Hx AOz− y + nH + H + ne e

(2)

85

The result is usually plotted as a Pourbaix diagram 11,15 (Figure 1). Noticing that under

86

experimental condition, URHE = 1.6 eV and pH = 13, 7 the concentration of aqueous species

87

is approximately 10−6 M. This value is used in the rest of the calculations.

4

ACS Paragon Plus Environment

Page 5 of 19 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 Catalysis

ACS Paragon Plus Environment

ACS Catalysis 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 19

102

reference surface (Rref ). Under this definition, the surface with the most negative relative

103

formation energy is the most stable one under given pH and electrode potential U . A Hess’s

104

law related approach proposed by Rong et al. is selected to calculate ∆G, 11 as shown below.

X A

Rref −→

X

nA A + R j

∆G1

nA A + nH2 O H2 O −→

X

+ − nA Hx AOz− y + nH+ H + ne e

∆G2

A

A

∆G = ∆G1 + ∆G2

105

∆G1 can be calculated from density functional theory (DFT), and ∆G2 can be evaluated

106

from experimental thermodynamic data of free energy per A at standard state relative to

107

standard hydrogen electrode (SHE), ∆G◦A,SHE

∆G1 =

∆Hj +

X

nA ∆HA − ∆Href

X

nA ∆EA − ∆Eref

A

≈ ∆G2 =

∆Ej +

A

X A

X A

=

!

−T

−T

∆Sj +

X A

X

nA ∆SA − ∆Sref

!

nA ∆SA

A

nA (µ◦Hx AOz− − µ◦A ) + nH+ µ◦H+ + ne µ◦e − nH2 O µ◦H2 O + y

(4)

kT ln aHx AOz− − 2.3nH+ kT pH + ne eUSHE y

X A

!

+ ∆G◦Hx AOz− y ,SHE

X A

kT ln aHx AOz− − 2.3nH+ kT pH + ne eUSHE y

108

where the entropy differences between surfaces are neglected. 16 Next, to study the hydration

109

and (de)hydrogenation processes on surfaces, the relative thermodynamic stability under

110

˜ k of the following given pH and U can be established through the reaction free energy ∆G

111

reaction for each hydration and (de)hydrogenation state k

6

ACS Paragon Plus Environment

Page 7 of 19 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 Catalysis

˜ ref + nH2 O H2 O −→ R ˜ k + nH+ (H+ + e− ) R

˜k ∆G

(5)

112

where the most stable surface predicted from the metal-ion exchange step can be chosen as

113

˜ ref . R

114

The number of electrons (or protons) that are released in this reaction can be treated as a

115

˜ k on an oxidization ladder. Knowing that it is always possible to write label nk of the state R

116

−e −e ˜ 0 −−−− ˜ 1 −−−− OER as a sequence of four single-electron transfer reactions: R −−−→ R −−−→ + +

117

˜ 2 −−−−−−−→ R ˜ 3 −−−−−−−→ R ˜ 0 + O2 , in which two H2 O molecules (or OH− ) are included R + +



(+H2 O)−H

−e−

(+H2 O)−H



(+H2 O)−H

−e−

(+H2 O)−H

118

˜ k is increased by one successively from R ˜ 0 to R ˜ 3, at certain steps, the label nk of each state R

119

due to the removal of an electron at each step. With this relationship, the OER path can

120

be naturally determined instead of being arbitrarily selected. To determine the OER path,

121

notice that all four reactions must be made spontaneous by applying the required electrode

122

˜ 0 > ∆G ˜ 1 > ∆G ˜ 2 > ∆G ˜ 3 > ∆G ˜ 0 − 4e(U − φ◦ potential, i.e. ∆G O2 /H2 O ) under pH and U ,

123

˜ k with the lowest ∆G ˜ k under pH and U it can be immediately concluded that the state R

124

˜ 3 in the OER cycle. Once R ˜ 3 is determined, the labels nk of the other three takes place of R

125

intermediates are automatically determined, and therefore the identity of each of them can

126

be determined by selecting the most thermodynamically stable surface with the correct label

127

nk for this OER step.

128

The structure and energetics of CaMnO3 with different surface reconstructions are com-

129

puted with first-principles DFT using the Perdew-Burke-Ernzerhof functional revised for

130

solids (PBEsol) 17 as implemented in the QUANTUM ESPRESSO package. 18 The calcula-

131

tions account for spin-polarized electronic densities. All atoms are represented by norm-

132

conserving, optimized, 19 designed nonlocal 20 pseudopotentials generated with the OPIUM

133

package, 21 treating the 2s and 2p of O, 3s, 3p, 3d, and 4s of Ca, and 3s, 3p, 3d, 4s, and

134

4p of Mn as semi-core and valence states. In addition to the inclusion of semi-core states,

135

the Mn pseudopotential also includes nonlinear core-valence interaction via the partial core

136

correction scheme. 22–24 A rotationally invariant DFT+U+J scheme 25 is applied for Mn. Ueff 7

ACS Paragon Plus Environment

ACS Catalysis 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

137

= 1.6 V is chosen from Ref [ 26] and J0 = 1.08 V is calculated from the linear response

138

method. 25,27 All calculations are performed with a 70 Ry plane-wave energy cutoff. 26 The

139

140

Brillouin zone is sampled using 4 × 4 × 1 Monkhorst-Pack 28 k -point meshes for the slab √ √ calculations. The 2 × 2 R45◦ nine-layer supercell is chosen for all surface configurations.

141

Surface-adsorbates and bare surface structures are relaxed with 13 ˚ A vacuum and a dipole

142

correction 29 to cancel the artificial interaction between the system and its periodic images.

143

MnO2 -terminated (001) surfaces and CaO-terminated (001) surfaces with various O

144

adatoms, O vacancies, Mn vacancies and Ca vacancies are considered. A subset of the in-

145

volved surfaces is shown in Figure (2). The most stable surface reconstructions are reported

146

as a phase diagram in Figure (3 – a). Surfaces are named with the format “termination ±

147

#atom”, where +, - represent adatoms and vacancies, respectively, and # represents number

148

of atoms introduced / removed per supercell relative to a reference surface with fixed termi-

149

nation, which can be either Mn2 O4 or Ca2 O2 . Figure (3 – a) shows that there are two surface

150

reconstructions whose stability regions extend above the O2 /H2 O equilibrium line, namely

151

Mn2 O4 + 2.0O and Mn2 O4 - 1.0Mn - 1.0O. Hence the hydration and (de)hydrogenation

152

process are evaluated on both surfaces.

153

Applying the scheme mentioned in equation (5), results of the most stable hydration

154

and oxidation states are shown in Figure (3 – b) and Figure (3 – c). Conditions are named

155

with the format “#H2 O ± #H”, where # is the number of H2 O and H introduced to or

156

removed from the surface per supercell. For the case of Mn2 O4 - 1.0Mn - 1.0O, since 2H2 O

157

- 1H is in the stability region at or above the O2 /H2 O equilibrium line, we first assume

158

˜ 3 in the OER cycle on surface Mn2 O4 - 1.0Mn - 1.0O. this state to be the intermediate R

159

˜ 0, R ˜ 1 , and Since the position on the oxidization ladder is only determined by “± #H”, R

160

˜ 2 should take the form of #H2 O + 2H, #H2 O + 1H, and #H2 O + 0H, respectively, and R

161

˜ k and R ˜ ′k are built up from their exact identities are listed in Table (1). Intermediates R

162

different surface reconstructions listed in the first row. Once the OER cycle is determined,

163

the required electrode potential could be calculated accordingly. For this reaction cycle on

8

ACS Paragon Plus Environment

Page 8 of 19

Page 9 of 19 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 Catalysis

ACS Paragon Plus Environment

ACS Catalysis 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 Paragon Plus Environment

Page 10 of 19

Page 11 of 19 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 Catalysis

ACS Paragon Plus Environment

ACS Catalysis 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 19

Table 2: Reaction free energy for each step in OER Mn2 O4 - 1.0Mn - 1.0O Reaction Step ∆GRHE /eV ˜ ˜ (1) R0 → R1 1.62 ˜1 → R ˜2 (2) R 1.09 ˜2 → R ˜3 (3) R 0.82 ˜3 → R ˜ 0 + O2 (4) R 1.39

Mn2 O4 + 2.0O Reaction Step ∆GRHE /eV ′ ′ ˜ ˜ 1.90 (1) R0 → R1 ′ ′ ˜ ˜ →R 1.23 (2) R 2 1 ′ ′ ˜3 ˜2 → R 0.44 (3) R ′ ′ ˜ →R ˜ + O2 (4) R 1.35 3

0

203

1.90 eV + 1.23 eV = 3.13 eV (referenced to RHE) is close to the proposed value (3.2 eV) in

204

the “scaling relationship”. The OER process on Mn2 O4 - 1.0Mn - 1.0O surface is different

205

from what is described in the AEM as mentioned above, but it is still possible to identify

206

˜ 0, R ˜ 1 , and R ˜ 3 , respectively, as they all occupy the same site. The the *OH, *O, and *O2 in R

207

˜ 2 is different from *OOH, but we can treat it as *OOH with adsorbate at the active site in R

208

the proton transferred to the lattice oxygen site, and investigate the influence of this proton

209

transfer phenomenon. Adding up the reaction free energy of step (1) and step (2) shows

210

˜ 0 and R ˜ 2 is 1.62 eV + 1.09 eV = 2.71 eV. This that the free energy difference between R

211

value is notably smaller than the expected free energy difference between *OOH and *OH

212

(3.2 eV), suggesting the violation of the “scaling relationship”. In order to confirm whether

213

the violation is caused by the proton transfer, we move the hydrogen back to the adsorbed

214

oxygen site (Figure 5–a) and let the structure relax (Figure 5–b). The energy is increased by

215

0.16 eV, but there is still strong interaction between the lattice oxygen and the proton. In

216

order to eliminate this interaction, we first let the lattice oxygen bond with another hydrogen

217

atom pointing away from *OOH, remove one hydrogen atom from one lattice oxygen, and

218

only allow the three atoms in *OOH to relax, shown in Figure (5–c) and Figure (5–d). Then,

219

we replace the *OOH in Figure (5–b) with the relaxed one in Figure (5–d) to achieve the

220

final geometry, see in Figure (5–e), and perform the scf calculation for this geometry. From

221

Figure (5–b) to Figure (5–e), the energy is further increased by 0.20 eV, meaning that the

222

lattice oxygen–hydrogen interaction reduces the energy by approximately 0.36 eV in total.

223

This suggests that the proton transfer process indeed makes a significant contribution to the

12

ACS Paragon Plus Environment

Page 13 of 19 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

224

ACS Catalysis

weakening of the “scaling relationship”.

225

The lattice–oxygen hydrogen–transfer process is driven by the strong activity of the

226

lattice oxygen, which itself could be attributed to the formation of Mn vacancies. Surface

227

atoms are less coordinated, compared with atoms in the bulk, and the dissolution of Mn

228

ions further reduces the coordination number of the neighboring lattice oxygen atoms. In

229

order to compensate for the missing lattice Mn, lattice oxygen interacts with the hydrogen

230

atoms, as can be seen by noting the contrast (Figure 4) that half of the surface lattice oxygen

231

˜ 0 , the ˜ ′0 , while all surface lattice oxygen atoms are hydrogenated in R are hydrogenated in R

232

surface with Mn vacancies. It is also worthwhile to consider whether hydrogen transfer occurs

233

between *OH and lattice oxygen or not. An examination of Figure (4–a) and Table 2 shows

234

that reaction (1) has 0.8 eV higher reaction free energy than reaction (3). This difference

235

˜ 2 since it would be even larger if the hydrogen were bonded to *OO to make it *OOH in R

236

˜ 2 less stable. Although the over 0.8 eV reaction free energy difference could would make R

237

not be fully attributed to differences in O–H bond strength between *OH and *OOH, it still

238

exhibits that the hydrogen from *OH is much less likely to transfer to the lattice oxygen,

239

compared with the hydrogen in *OOH.

240

The enhanced activity of lattice oxygen also induces the formation of the MnO2 triatomic

241

ring as the intermediate of O2 formation, rather than *OOH standing on the surface, which

242

could be another factor leading to reduced energy difference between *OH and *OOH. If we

243

trace back further for the reason of Mn vacancy formation, the high solubility of MnO− 4 ion

244

plays the role. Following this analysis, we suggest that doping the B site element with a more

245

soluble element could be a possible way of reducing the overpotential for the OER process

246

on perovskite–type materials. It is constructive to compare our conclusion with the work in

247

Ref [ 31], where A site cation deficiency in perovskite–type materal LaFeO3 enhances the

248

OER and ORR performance. They attribute the improvement to the formation of lattice

249

oxygen vacancies on the surface induced by A site vacancies. Our claim, however, is based

250

on the enhanced activity of lattice oxygen atoms surrounding the metal ion vacancies that

13

ACS Paragon Plus Environment

ACS Catalysis 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 Paragon Plus Environment

Page 14 of 19

Page 15 of 19 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 Catalysis

261

the OER overpotential and increase the OER catalytic activity. Therefore, we suggest that

262

doping perovskite–type materials with soluble metal elements to induce metal ion vacancies

263

on the surface in aqueous environment could enhance their catalytic activity for the OER

264

process.

265

Acknowledgement

266

T.Q. was supported by the Department of Energy Office of Basic Energy Sciences, un-

267

der grant number DE-FG02-07ER15920. B.T. was supported by National Natural Science

268

Foundation of China, under grant number 51502219. D.S.G. was supported by the National

269

Science Foundation, under grant number DMR-1719353. A.M.R. was supported by the Office

270

of Naval Research, under grant number N00014-17-1-2574.

271

The authors acknowledge computational support from the High-Performance Computing

272

Modernization Office of the U.S. Department of Defense, as well as the National Energy

273

Research Scientific Computing center.

274

References

275

276

(1) Rossmeisl, J.; Qu, Z. W.; Zhu, H.; Kroes, G. J.; Nørskov, J. K. J. Electroanal. Chem. 2007, 607, 83–89.

277

(2) Youngblood, W. J.; Lee, S.-H. A.; Kobayashi, Y.; Hernandez-Pagan, E. A.; Ho-

278

ertz, P. G.; Moore, T. A.; Moore, A. L.; Gust, D.; Mallouk, T. E. J. Am. Chem.

279

Soc. 2009, 131, 926–927, PMID: 19119815.

280

281

282

(3) Lee, Y.; Suntivich, J.; May, K. J.; Perry, E. E.; Shao-Horn, Y. J. Phys. Chem. Lett. 2012, 3, 399–404, PMID: 26285858. (4) Dismukes, G. C.; Brimblecombe, R.; Felton, G. A. N.; Pryadun, R. S.; Sheats, J. E.;

15

ACS Paragon Plus Environment

ACS Catalysis 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

283

Spiccia, L.; Swiegers, G. F. Accounts Chem. Res. 2009, 42, 1935–1943, PMID:

284

19908827.

285

286

287

288

289

290

(5) Su, H. Y.; Gorlin, Y.; Man, I. C.; Calle-Vallejo, F.; Nørskov, J. K.; Jaramillo, T. F.; Rossmeisl, J. Phys. Chem. Chem. Phys. 2012, 14, 14010–14022. (6) Du, J.; Zhang, T.; Cheng, F.; Chu, W.; Wu, Z.; Chen, J. Inorg. Chem. 2014, 53, 9106–9114, PMID: 25133744. (7) Kim, J.; Yin, X.; Tsao, K.-C.; Fang, S.; Yang, H. J. Am. Chem. Soc. 2014, 136, 14646–14649, PMID: 25295698.

291

(8) Rong, X.; Parolin, J.; Kolpak, A. M. ACS Catal. 2016, 6, 1153–1158.

292

(9) Man, I. C.; Su, H.-Y.; Calle-Vallejo, F.; Hansen, H. A.; Mart´ınez, J. I.; Inoglu, N. G.;

293

Kitchin, J.; Jaramillo, T. F.; Nørskov, J. K.; Rossmeisl, J. ChemCatChem 2011, 3,

294

1159–1165.

295

(10) Martirez, J. M. P.; Kim, S.; Morales, E. H.; Diroll, B. T.; Cargnello, M.; Gordon, T. R.;

296

Murray, C. B.; Bonnell, D. A.; Rappe, A. M. J. Am. Chem. Soc. 2015, 173, 2939–2947.

297

(11) Rong, X.; Kolpak, A. M. J. Phys. Chem. Lett. 2015, 6, 1785–1789.

298

(12) Friebel, D.; Louie, M. W.; Bajdich, M.; Sanwald, K. E.; Cai, Y.; Wise, A. M.;

299

Cheng, M.-J.; Sokaras, D.; Weng, T.-C.; Alonso-Mori, R.; Davis, R. C.; Bargar, J. R.;

300

Nørskov, J. K.; Nilsson, A.; Bell, A. T. J. Am. Chem. Soc. 2015, 137, 1305–1313,

301

PMID: 25562406.

302

303

304

305

(13) Lee, J. G.; Hwang, J.; Hwang, H. J.; Jeon, O. S.; Jang, J.; Kwon, O.; Lee, Y.; Han, B.; Shul, Y.-G. J. Am Chem. Soc. 2016, 138, 3541–3547, PMID: 26910187. (14) Burke, M. S.; Enman, L. J.; Batchellor, A. S.; Zou, S.; Boettcher, S. W. Chem. Mater. 2015, 27, 7549–7558. 16

ACS Paragon Plus Environment

Page 16 of 19

Page 17 of 19 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

306

307

308

309

310

311

ACS Catalysis

(15) Gerken, J. B.; McAlpin, J. G.; Chen, J. Y. C.; Rigsby, M. L.; Casey, W. H.; Britt, R. D.; Stahl, S. S. J. Am. Chem. Soc. 2011, 133, 14431–14442, PMID: 21806043. (16) Kolpak, A. M.; Li, D.; Shao, R.; Rappe, A. M.; Bonnell, D. A. Phys. Rev. Lett. 2008, 101, 036102 (1–4). (17) Perdew, J.; Ruzsinszky, A.; G´abor, C.; Vydrov, O.; Scuseria, G.; Constantin, L.; Zhou, X.; Burke, K. Phys. Rev. Lett. 2008, 100, 136406–1 – 4.

312

(18) Giannozzi, P.; Baroni, S.; Bonini, N.; Calandra, M.; Car, R.; Cavazzoni, C.; Ceresoli, D.;

313

Chiarotti, G. L.; Cococcioni, M.; Dabo, I.; Corso, A. D.; de Gironcoli, S.; Fabris, S.;

314

Fratesi, G.; Gebauer, R.; Gerstmann, U.; Gougoussis, C.; Kokalj, A.; Lazzeri, M.;

315

Martin-Samos, L.; Marzari, N.; Mauri, F.; Mazzarello, R.; Paolini, S.; Pasquarello, A.;

316

Paulatto, L.; Sbraccia, C.; Scandolo, S.; Sclauzero, G.; Seitsonen, A. P.; Smogunov, A.;

317

Umari, P.; Wentzcovitch, R. M. J. Phys.: Condens. Matter 2009, 21, 395502 (1–19).

318

319

(19) Rappe, A. M.; Rabe, K. M.; Kaxiras, E.; Joannopoulos, J. D. Phys. Rev. B Rapid Comm. 1990, 41, 1227–1230.

320

(20) Ramer, N. J.; Rappe, A. M. Phys. Rev. B 1999, 59, 12471–12478.

321

(21) http://opium.sourceforge.net.

322

(22) Louie, S. G.; Froyen, S.; Cohen, M. L. Phys. Rev. B 1982, 26, 1738–1742.

323

(23) Fuchs, M.; Scheffler, M. Comput. Phys. Commun. 1999, 119, 67–98.

324

(24) Porezag, D.; Pederson, M. R.; Liu, A. Y. Phys. Rev. B 1999, 60, 14132–9.

325

(25) Himmetoglu, B.; Wentzcovitch, R. M.; Cococcioni, M. Phys. Rev. B 2011, 84, 115108–

326

1–8.

327

(26) Lim, J. S.; Saldana-Greco, D.; Rappe, A. M. Phys. Rev. B 2016, 94, 165151 (1–10).

328

(27) Cococcioni, M.; de Gironcoli, S. Phys. Rev. B 2005, 71, 035105. 17

ACS Paragon Plus Environment

ACS Catalysis 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

329

(28) Monkhorst, H. J.; Pack, J. D. Phys. Rev. B 1976, 13, 5188–5192.

330

(29) Bengtsson, L. Phys. Rev. B 1999, 59, 12301–4.

331

(30) Grimaud, A.; Diaz-Morales, O.; Han, B.; Hong, W. T.; Lee, Y.-L.; Giordano, L.; Sto-

332

333

334

erzinger, K. A.; Koper, M. T. M.; Shao-Horn, Y. Nat. Chem. 2017, 9, 457, Article. (31) Zhu, Y.; Zhou, W.; Yu, J.; Chen, Y.; Liu, M.; Shao, Z. Chem. Mater. 2016, 28, 1691– 1697.

18

ACS Paragon Plus Environment

Page 18 of 19

Page 19 of 19 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 Catalysis

ACS Paragon Plus Environment