Band gap and band alignment prediction of nitride-based semiconductors using machine learning

Abstract

Nitride has been drawing much attention due to its wide range of applications in optoelectronics and remains plenty of room for materials design and discovery. Here, a large set of nitrides have been designed, with their band gap and alignment being studied by first-principles calculations combined with machine learning. Band gap and band offset against wurtzite GaN accurately calculated by the combination of screened hybrid functional of HSE and DFT-PBE were used to train and test machine learning models. After comparison among different techniques of machine learning, when elemental properties are taken as features, support vector regression (SVR) with radial kernel performs best for predicting both band gap and band offset with prediction root mean square error (RMSE) of 0.298 eV and 0.183 eV, respectively. The former is within HSE calculation uncertainty and the latter is small enough to provide reliable predictions. Additionally, 2 when band gap calculated by DFT-PBE was added into the feature space, band gap prediction RMSE decreases to 0.099 eV. Through a feature engineering algorithm, elemental feature space based band gap prediction RMSE further drops by around 0.005 eV and the relative importance of elemental properties for band gap prediction was revealed. Finally, band gap and band offset of all designed nitrides were predicted and two trends were noticed that as the number of cation types increases, band gap tends to narrow down while band offset tends to go up. The predicted results will be a useful guidance for precise investigation on nitride engineering.

An innovative technique for electronic transport model of group-III nitrides

“…To better validate the accuracy of proposed empirical pseudopotential method inbuilt in FullBand simulator 15 , used for the full band structure simulation in current manuscript, the authors have compared and contrasted the energy values at gamma valley with previously reported experimental data taken from references 5 , 7 , 23 – 27 . Band gaps for AlN, GaN, InN, and Al 0.2 Ga 0.8 N, In 0.2 Al 0.8 N and In 0.2 Ga 0.8 N alloys are also compared with the results obtained by various density functional theories including LDA 5 , 7 , 28 , LDA-1/2 5 , PBE 7 , 12 , 28 and HSE 7 , 12 , 28 methods. All the comparison data are tabulated in Table 3 .…”

“…LDA local density approximation, LDA-1/2 approximately includes the self-energy of excitations in semiconductors, PBE Perdew–Burke–Ernzerhof (PBE) exchange energy theory, HSE Heyd–Scuseria–Ernzerhof exchange–correlation functional uses an error function screened Coulomb potential to calculate the exchange portion of the energy. *calculated from modified Vegard’s law 11 , 12 where A and B represent band gap values for binary nitride alloys and b is bowing parameter 12 .…”

“…*calculated from modified Vegard’s law 11 , 12 where A and B represent band gap values for binary nitride alloys and b is bowing parameter 12 .…”

“…Ensemble Monte Carlo (EMC) technique is the best suited numerical approach for accurate solution of the Boltzmann equation (BTE) under nonlinear response conditions 12 – 14 . The aim of the present paper is to demonstrate a simple and inexpensive theoretical approach for prediction of the low and high field mobility model for binary and ternary wurtzite nitride alloys for entire concentration range.…”

“…The carrier transport properties including multiple appropriate scattering mechanisms 12 – 16 associated with nitride alloys have been simulated by the solution of BTE through EMC method (using TNL’s ElecMob simulator) 12 – 17 . ElecMob simulator is capable to simulate carriers transport on full energy band including random scattering events due to impurities, lattice vibrations, etc.…”