26th Australian Institute of Physics Congress and the Australian and New Zealand Conference on Optics and Photonics

Computationally Tractable Prediction of Impact Ionization for Semiconductor Materials Discovery

Not scheduled

1h 30m

Belinda Hutchinson Building (The University of Sydney )

Contributed Oral COMMAD - Optoelectronic and Microelectronic Materials and Devices Parallel sessions

Description

Impact ionization plays a central role in avalanche multiplication, electrical breakdown and high-field carrier transport in semiconductor devices. However, accurate prediction of impact-ionization behaviour typically requires computationally intensive first-principles calculations involving wavefunction overlap integrals, dielectric screening and high-dimensional Brillouin-zone integrations. These requirements have largely limited the application of impact-ionization modelling to small numbers of materials and hindered its use in large-scale materials discovery.

We present a computationally tractable framework for predicting impact-ionization characteristics using only publicly available electronic structure data from the Materials Project database. The framework comprises a hierarchy of physics-based proxies, including kinematic threshold optimisation, phase-space models, phase-space evaluation using reconstructed Brillouin-zone manifolds, and physically motivated scattering proxies incorporating momentum-transfer-dependent matrix elements. Together, these approaches progressively recover key aspects of impact-ionization physics while avoiding explicit wavefunction calculations.

Benchmark calculations on representative semiconductor materials demonstrate how impact-ionization thresholds, accessible phase space and scattering descriptors evolve across the hierarchy of models. The results reveal a clear distinction between the minimum kinematically allowed impact-ionization threshold and the energy at which significant impact-ionization phase space becomes available, highlighting the important role of Brillouin-zone topology in avalanche processes.

By leveraging pre-computed electronic structure databases, the proposed framework enables scalable screening of semiconductor materials for avalanche photodiodes, radiation detectors and high-field electronic devices. The methodology provides a practical pathway for incorporating impact-ionization descriptors into data-driven semiconductor materials discovery workflows while substantially reducing the computational cost associated with conventional first-principles approaches.