The rapid acceleration of vehicle drivetrain electrification is expected to be a major driver for the widespread adoption of silicon carbide (SiC) power devices such as metal-oxide-semiconductor field-effect transistors (MOSFETs) and junction-barrier-Schottky (JBS) diodes (the latter needed as the freewheeling diode in parallel with the MOSFET switch), due to their ability to increase the efficiency and reduce the volume of the power electronic inverter that ties the vehicle’s battery to the electric motor(s). The move from traditional silicon-based power electronics to SiC-based power electronics is necessary to achieve the U.S. Department of Energy (DOE) power electronics density target of 100 kW/L. However, gallium nitride (GaN)-based inverters may offer additional benefits beyond what can be achieved with SiC. For example, genetic-algorithm-based system-level simulations of the Pareto front trade-off between power density and efficiency of the inverter indicate that GaN power devices offer solutions superior to those achievable with SiC. However, given that the drivetrain DC bus voltage in future electric vehicles is expected to be 800 V or higher, commercially-available GaN power switching high electron mobility transistors (HEMTs), which typically have voltage ratings of 650 V or less, are unsuitable for a two-level inverter design, which is preferred by automotive system designers due to its simplicity and associated reliability. As such, vertical GaN power devices with voltage ratings of 1.2 kV or greater are necessary. This talk will describe our team’s efforts to develop both vertical GaN MOSFETs and JBS diodes. Designs for both trench MOSFETs and double-well MOSFETs (DMOSFETs) will be discussed. Experimental work has focused on trench MOSFETs, chiefly due to the need for selective-area doping in the DMOSFET, which is generally challenging in GaN and is particularly so in this structure due to the double well. The trench MOSFET presents significant challenges as well, including the need for a high-quality gate dielectric on the etched sidewall of the trench as well as mitigation of high electric fields at the bottom of the trench. Vertical GaN trench MOSFETs utilizing drift layers grown by metal-organic chemical vapor deposition (MOCVD) on native GaN substrates have been fabricated. Devices with an atomic-layer-deposited SiO2 gate dielectric have demonstrated a positive threshold voltage of ~8 V, an on/off ratio of ~108, and a current density of ~400 mA/mm. Additionally, GaN JBS diodes are being developed concurrently, and in this case selective-area doping cannot be avoided as alternating regions of p- and n-doped material are inherent to the operation of the device. Our team has utilized an etch-and-regrowth approach to demonstrate vertical GaN JBS diodes with reverse hold-off voltage >1.5 kV and forward turn-on voltage <1 V, consistent with JBS operation, although the reverse leakage current is higher than is typically observed for a continuously-grown GaN PN junction. Prospects to further improve device performance will be discussed, including reducing the reverse leakage current though advances in the selective-area p-GaN regrowth process. This work was supported by the Electric Drivetrain Consortium managed by Susan Rogers of DOE’s Vehicle Technologies Office and by the ARPA-E PNDIODES program directed by Isik Kizilyalli. Sandia National Laboratories is a multi-mission laboratory managed and operated by National Technology & Engineering Solutions of Sandia, LLC, a wholly owned subsidiary of Honeywell International Inc., for the U.S. Department of Energy’s National Nuclear Security Administration under contract DE-NA0003525. This paper describes objective technical results and analysis. Any subjective views or opinions that might be expressed in the paper do not necessarily represent the views of the U.S. Department of Energy or the United States Government.
Read full abstract