Accurate band gap prediction based on an interpretable Δ-machine learning. (December 2022)

• Record Type: Journal Article
• Authors: Zhang, Lingyao; Su, Tianhao; Li, Musen; Jia, Fanhao; Hu, Shuobo; Zhang, Peihong; Ren, Wei
• Is Part Of: Materials today communications. Volume 33(2022)
• Publication Date: 2022-12
• Subjects: 2D materials -- Bandgap -- Machine learning -- DFT calculation -- Interpretable; Materials science -- Periodicals; 620.11
• DOI: 10.1016/j.mtcomm.2022.104630
• Languages: English
• ISSNs: 2352-4928

Abstract

Abstract: Most materials science datasets are not so large that the accuracy of machine learning (ML) models is relatively limited if only simple features are used. Here, we constructed an interpretable ∆-machine learning (∆-ML) model to connect the hybrid functional HSE bandgap ( E g HSE ) with the PBE functional bandgap ( E g PBE ). The former can reproduce the band gap comparable with experiments, but the computational cost is much more challenging. The training is based on our high-throughput calculations on a set of two-dimensional semiconductors. Four complex descriptors, all based on the E g PBE are constructed using the sure independence screening and sparsifying operator (SISSO) algorithm. Using these descriptors, the ∆-ML can accurately predict the E g HSE of test set with a determination coefficient ( R 2 ) of 0.96. The error satisfies a normal distribution with a mean of zero. We provide a direct functional relationship between input descriptors and target properties. We find that E g HSE and the 5/6th power of E g PBE show a significant linear correlation, which may guide rapid prediction of E g HSE from E g PBE for materials with a E g HSE greater than 0.22 eV. We also discussed the correlation between the atomic radius and the E g HSE . Our work will provide an effective and interpretable model to construct the optimal physical descriptors for ML prediction on bandgaps in screening massive new 2D materials research. Graphical Abstract: ga1 Highlights: Constructing an interpretable ∆-machine learning (∆-ML) model to connect the hybrid functional E g HSE with the E g PBE . SISSO descriptor D 3 = E g PBE 5 / 6 can predict the E g HSE of 2D-semiconductors using equation E g HSE = D3 × 1.55 + 0.22. SISSO descriptor D 1 shows the atomic volume negatively correlated to E g HSE .