Table 1 Properties of amorphous carbon with different hydrogen contents in density of 2.4 gcm−3

From: Transferable machine learning model for multi-target nanoscale simulations in hydrogen-carbon system from crystal to amorphous

Amorphous Carbon

Young’s modulus (GPa)

Bulk modulus (GPa)

Shear modulus (GPa)

H content (atomic%)

DFT

MLP (Small)

MLP (Large)

DFT

MLP (Small)

MLP (Large)

DFT

MLP (Small)

MLP (Large)

0

291.71 (7.04)

298.03 (15.97)

291.72 (7.04)

193.76 (5.81)

215.83 (22.57)

199.78 (11.98)

115.57 (6.46)

118.82 (3.37)

118.29 (3.41)

10

281.27 (3.23)

262.40 (10.46)

278.42 (13.22)

216.91 (4.15)

198.63 (14.21)

193.01 (6.72)

112.11 (4.74)

102.17 (6.78)

106.36 (9.23)

20

267.97 (3.05)

254.69 (15.74)

240.76 (10.15)

199.03 (4.02)

199.47 (8.48)

191.64 (1.53)

106.52 (3.12)

108.56 (2.85)

102.57 (10.70)

30

216.43 (4.76)

196.67 (23.19)

194.63 (13.48)

136.62 (5.15)

130.44 (7.24)

127.15 (9.57)

87.71 (1.85)

85.59 (3.50)

81.22 (6.76)

40

231.13 (3.59)

238.98 (14.69)

234.01 (12.78)

198.47 (4.76)

196.26 (0.84)

195.16 (3.87)

90.17 (3.83)

90.37 (5.60)

87.76 (7.38)

  1. The structures of the a-C:H with different hydrogen contents were constructed through melt quenching process (see “Methods”) in AIMD simulations and MLP-based MD simulations, which, respectively, were calculated with DFT methods and pre-trained MLP model for their modulus. The structures for DFT and MLP (Small) contained the order of magnitude 102 atoms and that for MLP (Large) contained the order of magnitude 104 atoms. The numbers in parentheses are the standard deviation of the corresponding values calculated in three independent MLP-based MD simulations and AIMD simulations.
Back to article page