A Defect-Informed Phase-Field Model for Gate Oxide Breakdown in SiC MOS Structures

View at publisher
Authors: Jinglin Li, Aditya Shekhar, Yaqian Zhang, Willem Dirk van Driel, G Q Zhang
Cited by: 1

Abstract

Silicon carbide (SiC) MOS structures are promising for high voltage and high temperature power electronics, but gate oxide breakdown remains a major reliability concern because of its stochastic nature and the limited accessibility of the underlying failure pathway. In this work, a defect-informed phase- field framework is developed to investigate oxide breakdown in SiC MOS structures. The model is initialized using a depth- resolved pre-existing defect profile extracted from capacitance and conductance frequency response, thereby linking electrical characterization with physics-based breakdown simulation. The results show that breakdown starts from localized nucleation at defect-rich weak points near the SiO2/SiC interface, followed by directional growth, interaction, and coalescence of degraded regions into a through-thickness conductive path. Combined analysis of the phase-field variable and electric field distribution shows that local electric field redistribution plays a key role in guiding conductive path propagation and determining the final breakdown trajectory. Post-breakdown focused ion beam scanning electron microscopy (FIB-SEM) reveals a tortuous and spatially confined channel morphology, consistent with the defect-assisted and field-driven breakdown picture predicted by the model. The proposed framework provides a useful tool for analyzing breakdown trajectories and gate oxide reliability in SiC power devices.

Fields of Study & Subjects

Silicon Carbide Semiconductor Technologies Semiconductor materials and devices Advancements in Semiconductor Devices and Circuit Design

Publication details

Journal / source
2026 27th International Conference on Thermal, Mechanical and Multi-Physics Simulation and Experiments in Microelectronics and Microsystems (EuroSimE)
Publisher
IEEE
Pages
1–7
Publication date
19 April 2026