Numerical fitting of transient decays in the high defect density limit

Abstract

A numerical fitting method based on the deep level transient spectroscopy (DLTS) technique is presented. This method deals with a situation where the standard rate window DLTS is no longer sufficient, i.e., the assumption that the defect density NT is much less than the donor doping density ND is no longer valid. Digitized capacitance transients are numerically fit to extract the electron emission rate, defect density, and energy level. The defect center under study is EL2 in n-type liquid-encapsulated Czochralski gallium arsenide. The fitting method gives an EL2 thermal activation energy of 0.76 eV, different from the 0.82 eV obtained by standard DLTS, which only examines the maximum emission conditions. The advantages, as well as the limitations, of this fitting method are discussed.