GaAs field-effect transistor properties, as influenced by the local concentrations of midgap native donors and dislocations

Abstract

Results are presented of electrical parameter mapping for arrays of field-effect transistors (FET’s) fabricated in semi-insulating GaAs wafers, compared with mapping data of the dislocation density and the neutral concentration of the main midgap donor defect (known as EL2) in the same wafers. The work seeks to clarify whether a ‘‘shift’’ of FET parameters such as threshold voltage Vth results directly from dislocation proximity, or whether such a shift results from FET sensitivity to the local EL2 concentration. For a wafer having quite different spatial distributions of EL2 and dislocations, the FET parameters were found to correlate strongly with the local neutral EL2 density, a high density promoting a larger channel current and a more negative Vth. Such results suggest that a reported sensitivity of Vth to dislocation proximity may arise from the role a dislocation can play as a source or sink for point defects.