Self-Heating and Reliability-Aware Intrinsic Safe Operating Area of Wide Bandgap Semiconductors-An Analytical Approach

The emergence of several technology options and the ever-broadening range of applications (e.g., automotive, smart grids, solar/wind farms) for power electronic devices suggest both a need and an opportunity to develop unifying principles to guide the development of wide bandgap (WBG) semiconductors. Unfortunately, power electronic devices are typically evaluated with a variety of elementary figure of merits (FOMs), which offer inconsistent/contradictory projections regarding the relative merits of emerging technologies. Indeed, one relies on the empirical (extrinsic) safe-operating area (SOA) of a packaged device to ultimately assess the performance potential of a technology option. Unfortunately, extrinsic SOA can only be calculated a posteriori, i.e., after precise measurement of the fabricated device parameters, making it suitable only for relatively mature technologies. Based on the insights of material-device-circuit-system performance analysis of a variety of idealized WBG power electronic devices (e.g., GaN HEMT, beta-Ga2O3 MOSFET), in this paper, we analytically derive a comprehensive, substrate-, selfheating-, and reliability-aware intrinsic/limiting safe operating area (SOA) that establishes a priori, i.e., before device fabrication, the optimum and self-consistent trade-off among breakdown voltage, power consumption, operating frequency, heat dissipation, and reliability. We establish the relevance of the intrinsic-SOA by comparing its prediction with a broad range of experimental data available in the literature. In between the traditional FOMs and extrinsic SOA, the intrinsic SOA allows fundamental/intuitive re-evaluation of intrinsic technology potential for power electronic devices and identifies specific performance bottlenecks and suggests strategies to circumvent them.

Automated summary

This paper discusses the importance of developing unified principles to guide the development of wide bandgap semiconductors, proposing a comprehensive intrinsic safe operating area that optimizes trade-offs in advance. By comparing its predictions with experimental data, it provides a new perspective for evaluating the performance of power electronic devices.