Temperature-dependent lifetime spectroscopy (TDLS) in silicon

Carrier lifetime is very sensitive to electrically active defects. Apart from detecting the presence of recombination-active defects, lifetime measurements allow for a direct determination of defect parameters if the temperature or injection dependence of carrier lifetime is analyzed. The advantage of TDLS compared to other lifetime spectroscopic methods is that it allows a direct determination of the defect energy level (W from the temperature dependence of LLI-SRH-lifetime measured on one single sample. As the accuracy of the (W-determination strongly depends on the width of the temperature interval over which the characteristic Arrheniusincrease may be observed, different methods for an optimized data evaluation are proposed in the present work. Applying these new evaluation methods to TDLS-measurements taken on intentionally metal-contaminated silicon samples, excellent agreement between TDLS and DLTS is achieved. Thus, with these new data evaluation methods, TDLS allows for an accurate determination of the energy level (W in a wide range of defect energies and doping concentrations. The main disadvantage of TDLS is its limited capability to resolve multiple active defect levels. In contrast to the prevalent interpretation of TDLS-data in literature, it is in no case possible to determine more than one defect level from fits to different linear regions in the Arrhenius-plot.