Fig. 1: Multiscale modeling framework for defect identification in semiconductors under irradiation.

a Calculation of defect parameters including formation energies, energy levels, carrier capture cross-sections and migration energies by the density functional theory. The orange and light blue regions represent the conduction and valence bands, respectively. b Irradiation and annealing simulations by combining our self-developed Monte Carlo (MC) and object kinetic MC methods to acquire annealed defect concentrations. The inset shows the complex dynamic processes of defects and carriers (electrons and holes) in semiconductors under irradiation, including the migration and drift of mobile defects and carriers, the generation of self-interstitials (Is) and vacancies (Vs), the I-V and electron-hole recombination, the formation and break-up of impurity-point-defect pairs, the trapping and detrapping of I/V clusters and their complex clusters with impurities, and the electron/hole capture and emission of charged defects. The unfilled, black filled and red filled circles are represented by vacancies, self-interstitials and impurities, respectively. c Deep-level transient spectroscopy (DLTS) simulations by our improved rate theory model based on the Shockey-Read-Hall theory. t₁ and t₂ are determined by the rate window of DLTS experiments. d Determination of crucial deep-level defects through matching simulated and experimental DLTS peaks using our self-developed peak-level matching algorithm (Experimental data is reproduced from ref. ³³). FWHM represents the Full Width at Half Maximum of a DLTS peak.