I. Introduction
Transient opto-electronic measurements are an important tool for the characterization of photovoltaic absorber materials and completed solar cells. Because of the high radiative efficiency of metal halide perovskites (MHP), these methods are especially convenient and popular for these materials. Transient photoluminescence (TRPL) has been successfully applied to perovskite absorbers to disentangle the different recombination mechanisms like Auger recombination, radiative recombination, and Shockley-Read-Hall recombination [1], [2]. The interpretation of TRPL measurements for the combination of absorber films with contact layers or for completed solar cells is complicated by the fact that charge carriers accumulate at the interfaces and/or are extracted to those additional layers [3], [4], as illustrated in Fig. 1. Especially for completed cells the charging of the contact layers is an important, generic effect that is closely related to the overall functionality of the solar cell, namely the generation of the photovoltage. In addition, these contact effects are also present in the results from transient photovoltage (TRPV), impedance spectroscopy (IS), and intensity modulated photovoltage (IMPV) [5].
Charge carrier kinetics following a pulsed light excitation of a bare perovskite absorber film (upper row) and of a completed perovskite solar cell (lower row). For the absorber film, the recombination kinetics typically is described in dependence of only a single variable, namely the split of the quasi-Fermi levels. Different recombination mechanisms, like radiative recombination or Shockley-Read-Hall (SRH) recombination are dominant at different carrier concentrations. In a completed solar cell, charge carriers can be also extracted to the contact layers where they charge the electrodes and build up an external voltage . As soon as the voltage exceeds these carriers are reinjected into the absorber where they subsequently recombine. This kinetics requires a description by a model that accounts for and as two independent variables.