Table 1 NNP and DFT predicted lattice constants (a and c in Å), elastic constants (cij in GPa), polycrystalline shear (μ) and bulk (κ) moduli (in GPa) of wurtzite ZnO, wurtzite GaN and cubic SrTiO3

From: Neural network potential for dislocation plasticity in ceramics

 

a

c

c11

c12

c13

c33

c44

c66

μ

κ

Note

wurtzite ZnO

3.287

5.305

201

120

101

213

38

41

43.42

129.17

This work, NNP

 

3.287

5.306

188

107

94

201

37

40

41.22

129.59

This work, GGA-PBE

 

3.278

5.291

        

GGA-PBE23

   

191.5

108.7

95

206.7

41.4

38

  

PBE110

   

221.1

105.9

87.4

234.3

57.6

53.4

  

PBE+U110

 

3.195

5.160

        

LDA111

 

3.148

5.074

        

LDA+U111

 

3.25

5.21

        

Exp.112

   

190

110

90

196

39

40

  

Exp.112

   

206

117

118

211

44.3

44.6

  

Exp.113

   

209.7

121.1

105.1

210.9

42.47

44.29

  

Exp.114

wurtzite GaN

3.224

5.290

335

143

123

392

79

92

93.38

204.25

This work, NNP

 

3.248

5.282

329

128

96

355

86

100

99.42

183.44

This work, GGA-PBE

   

356

146

115

382

91

105

 

205

LDA115

 

3.190

5.189

        

Exp.116

 

3.189

5.178

        

Exp.117

   

390

145

106

398

105

123

  

Exp.118

cubic SrTiO3

3.901

 

331

106

  

98

 

109.20

189.00

This work, NNP

 

3.895

 

351

103

  

114

 

118.03

186.04

This work, GGA-PBEsol

   

319.3

97.5

  

113

   

GGA-PBE119

   

290.7

93.9

  

110.9

   

BLYP119

 

3.895

 

352

105

  

116

   

GGA-PBEsol28

   

317.2

102.5

  

122.35

   

Exp.120

  1. The theoretical values and experimental data previously published in the literature are also shown for comparison.