“…4,5 The negative transconductance dispersion, commonly observed in GaAs metalsemiconductor field-effect transistors, was believed to be due to the surface states existing on the ungated regions between gate and source/drain electrodes. 1,6,7 The magnitude of the negative transconductance dispersion was dependent on the surface leakage current from the gate 1 and could be reduced after the surface passivation with Si 3 N 4 . 6 On the other hand, the positive transconductance dispersion was observed in HEMT with the source layer of Si-doped AlGaAs, which was thought to be due to bulk traps, such as DX center and point defects produced by the neutron irradiation.…”