Dual- Vs. Single-Peak Transconductance Evolution in Schottky $p$-Gan Gate Hemts: Influence of Partially and Fully Depleted $p$-Gan Layer

Le résumé fourni par la source

In this work, four groups of wafers with distinct Mg activation levels were fabricated, consisting of three partially depleted$\boldsymbol{p}$-GaN (PDP-GaN) layers and one fully depleted$\boldsymbol{p}$-GaN (FDP-GaN) layer. Reduced Mg activation induces anomalous dual-peak transconductance ($G \mathrm{m}$) in PDP-GaN devices, a phenomenon rarely reported in Schottky-type$p$-GaN gate HEMTs. Reduced activated Mg concentration transforms the$G_{\mathrm{m}}$profile from dual-peaks in PDP-GaN layers to a single$G_{\mathrm{m}}$peak in FDP-GaN, consistent with the evolution toward an MIS-like gate stack. For the dual-peaks$G_{\mathrm{m}}$in three PDP-GaN wafers, the first$G_{\mathrm{m}}$peak positions remain nearly identical across all activated Mg concentrations, while their amplitudes decrease with reduced Mg activation. The second$G_{\mathrm{m}}$peak positions shift to higher gate voltages, while the amplitudes exhibit activated Mg concentration dependent attenuation. TCAD simulations attribute the observed dual-peak behavior to a voltage partitioning mechanism across the back-to-back junctions in the gate stack. At higher gate voltages, lower Mg activation concentrations resulted in increased Schottky junction voltages, thereby reducing the voltage drop across the channel and thus affecting both the accumulation rate and density of two-dimensional electron gas (2DEG).

Ce résumé expose les affirmations des auteurs. BNTIC ne l’interprète pas comme une validation indépendante des résultats.