Influence of the Bias Voltage on Effective Electron Velocity in AlGaN/GaN High Electron Mobility Transistors

The small-signal S parameters of the fabricated double-finger gate AlGaN/GaN high electron mobility transistors (HEMTs) were measured at various direct current quiescent operating points (DCQOPs). Under active bias conditions, small-signal equivalent circuit (SSEC) parameters such as Rs and Rd, and intrinsic parameters were extracted. Utilizing fT and the SSEC parameters, the effective electron velocity (nu e-eff) and intrinsic electron velocity (nu e-int) corresponding to each gate bias (VGS) were obtained. Under active bias conditions, the influence mechanism of VGS on nu e-eff was systematically studied, and an expression was established that correlates nu e-eff, nu e-int, and bias-dependent parasitic resistances. Through the analysis of the main scattering mechanisms in AlGaN/GaN HEMTs, it has been discovered that the impact of VGS on nu e-eff should be comprehensively analyzed from the aspects of nu e-int and parasitic resistances. On the one hand, changes in VGS influence the intensity of polar optical phonon (POP) scattering and polarization Coulomb field (PCF) scattering, which lead to changes in nu e-int dependent on VGS. The trend of nu e-int with changes in VGS plays a dominant role in determining the trend of nu e-eff with changes in VGS. On the other hand, both POP scattering and PCF scattering affect nu e-eff through their impact on parasitic resistance. Since there is a difference in the additional scattering potential corresponding to the additional polarization charges (APC) between the gate-source/drain regions and the region under the gate, the mutual effects of PCF scattering on the under-gate electron system and the gate-source/drain electron system should be considered when adjusting the PCF scattering intensity through device structure optimization to improve linearity. This study contributes to a new understanding of the electron transport mechanisms in AlGaN/GaN HEMTs and provides a novel theoretical basis for improving device performance.