Reliability of photoconductance decay measurements of samples with significant carrier lifetime inhomogeneity

Carrier lifetime measurements are performed using the photoconductance decay (PCD) technique on samples exhibiting varying degrees of inhomogeneity in carrier lifetime. The PCD measurement is regarded as the prevailing method for evaluating the electrical performance of silicon wafers or surface passivation layer structures for photovoltaic applications by reporting the effective carrier lifetime versus the injection level τ(∆n). However, we found that in case of substantial carrier lifetime inhomogeneity within the detected area, the reliability of the results is highly questionable.We used a simple model to compute the expected shape of the recorded τ(∆n) curves. While in the case of a uniform carrier lifetime distribution the results are independent of the measuring parameters, they are strongly influenced if measuring samples with inhomogeneous carrier lifetime. It was observed that changing the excitation light power or the duration of the light pulse resulted in a substantial and consistent shift in τ(∆n) curves. Furthermore, the τ(∆n) results of the decay measurement exhibited a significant deviation from the areal average of steady-state carrier lifetime depending on the degree of carrier lifetime inhomogeneity.This phenomenon was investigated experimentally as well, using a laser-based PCD setup on various sample types. These samples featured substantial areal carrier lifetime fluctuations, either in bulk lifetime or surface passivation quality. The experimental results obtained confirmed the expectations derived from the modeling.Recording the PCD curves with multiple laser excitation parameters and using the steady state carrier lifetime results, more precise and reliable carrier lifetime results can be obtained from inhomogeneous samples.