Further improvements in the performance of perovskite solar cells requires a deep understanding of their device physics and the different power loss mechanisms. In the literature, defect densities at the perovskite/transport layer interfaces, ion-mediated non-radiative recombination and imperfect charge extraction have been cited as the main sources of these power losses. However, the magnitudes of the parameters of these loss mechanisms, such as defect densities, ion densities, capture coefficients and time constants/lifetimes are not well known. This is due to the difficulty of discriminating between multiple mechanisms that respond simultaneously in a typical characterization measurement, in addition to the lack of optoelectronic models describing these loss mechanisms clearly. In this talk, I will provide improved data analysis methods for typical optoelectronic methods (both in the time and frequency domain) used to characterize perovskite solar cells, to accurately discriminate between the loss mechanisms and determine their parameters.

In the case of capacitance measurements to determine defect/doping/ion densities, I will show that the capacitance response of the perovskite layer is hidden by the response of the electrodes and the resistive transport layers. This effect dominates the response in several capacitance measurements reported in literature, leading to the calculated defect densities and related parameters largely being artefacts of measurement. Analytical resolution limits are derived for these techniques to distinguish a real defect response from a measurement artefact. Finally, I will show experimental measurements on single crystal devices that overcome these limits and allow identifying the actual defect density in the device.

I will furthermore develop an optoelectronic model that explicitly accounts for the extraction of charge carriers through the resistive transport layers, and apply it to the analysis of small perturbation methods in the time and frequency domain. This model predicts the existence of a time constant for charge carrier extraction, in additional to bulk and electrode-mediated recombination time constants. By developing modified transfer functions in the frequency domain, the charge extraction and recombination time constants can be accurately extracted using these methods. I will also develop a figure of merit that determines the efficiency of charge extraction in the perovskite solar cell. This figure of merit depends on the time constants of charge extraction and recombination, and can be calculated at each bias point of the current-voltage curve. Figure of merit values between 0.7–0.95 at or close to the 1 sun open-circuit voltage are obtained experimentally, indicating a significant electric field exists in the transport layers in these conditions.