Organic-inorganic perovskite lead halide solar cells (PSCs) have received increasing interest in the last decade. PSCs already achieve record efficiencies above 25%. However, long-term stability is still a problem, especially when transitioning from cells to modules.

One of the described long-term stability problems at the module level is Potential-induced degradation (PID), caused by the voltage build-up by modules connected in series.

PID is extensively investigated within crystalline silicon PV, while very little is known for perovskites. Carolus et al. reported nearly complete efficiency losses within 18 hours of a PID stress test, indicating their high susceptibility to PID[1]. In order to tackle PID-related stability issues and make commercialization a reality, it is crucial to retrieve insights into the physics of the PID mechanism.

This study investigates and compares the PID mechanism within p-i-n CsFAPbIBr perovskite modules with either a copper (Cu) or an indium tin oxide (ITO) rear contact. Several soda-lime glass-glass configured 5.5 x 5.5 cm² mini-modules were PID stressed from the p-side at 40⁰C and 1000 V for 192 hours using the foil method. Intermediate current-voltage measurements (IV) and electroluminescence (EL) images were taken to investigate the PID progress.

The ITO contacted perovskite modules illustrate an incubation period of about 100 hours. A slight drop in short-circuit current (I_sc) and a modest increase in series resistance can be observed within this incubation period.

Subsequently, a significant drop in I_sc and increase in series resistance are noticeable. Evidently, after 192 hours, an additional decrease in shunt resistance is noticeable, resulting in a total relative loss of efficiency of almost 80%. The EL images illustrate no significant differences in the incubation period, although the formation of inhomogeneities can be observed further in the degradation process[2].

The Cu-contacted perovskite modules are significantly more prone to PID than their ITO counterpart. Similarly, the drop in I_sc and increase in series resistance can be observed; however, no incubation period is present.

Analogous to crystalline silicon, it is hypothesized that positively charged sodium ions migrate out of the cover glass towards the PV cell. Hence, the ions can alter the conductivity of the contacts or migrate deeper into the stack and form inhomogeneities in the perovskite material[3]. However, additional PID stress tests are ongoing to validate the hypothesized findings. Furthermore, microstructural analysis is necessary to designate this degradation mechanism's root cause and retrieve more insights into its physical behavior and kinetics.