Metal-halide perovskite (MHP) semiconductors are highly relevant candidates for the fabrication of next generation solar cells thanks to their very good opto-electronic properties as well as ease of processability. That said, considerable challenges are still ahead for MHP solar cells: Indeed, under continuous external stressors, such as heat, moisture, bias and light, MHP materials suffer from various (irr)reversible chemical reactions [1, 2, 8]. These phenomena impede devices from attaining the technological readiness levels required for deployment. Therefore, it is crucial to understand the initial stages of perovskite material evolution under external stressors in order to mitigate long term degradation of devices.

As part of long-term degradation of solar cell devices, bulk absorber stability under continuous illumination is needed in order to conserve appropriate device function. Stable bulk properties would also avoid second hand reactions at interfaces with available decomposition products of the bulk [3].

Striking evidence for absorber instability under continuous irradiation has been previously described, where steady state photoluminescence (PL) either increases or decreases, depending on conditions of continuous illumination [4,5,6] or quasi-continuous illumination [7]. Additionally, these reports also highlight the crucial role of the sample environment on PL evolution under continuous illumination [8]. For all these works, the common and favored explanation for PL evolution under continuous illumination is the introduction and/or annihilation of new trapping sites which respectively causes a drop or increase in PL intensity. Though correlations to kinetic models or DFT calculations have been shown previously [7], experimental evidence for the effect of newly introduced trapping sites on the dominating recombination mechanisms at play is lacking. This is where time resolved photoluminescence (trPL) is a very useful tool: it can be used to identify what type of recombination mechanism is dominating from carrier-density dependent lifetimes, using so called differential lifetime plots [9].

In this presentation, we will focus on methylammonium-lead-iodide (MAPI) thin films and the (ir)reversible introduction of traps triggered by quasi-continuous illumination. For this, we develop a method that uses long pulses of light in combination with trPL counting schemes, calling it long pulsed trPL (LP-trPL). From the method, we observe the inclusion of long lived and non-deep trapping sites due to continuous illumination. The data also suggests a highly asymmetric mechanism of trap formation, where trap annihilation is much slower than observed formation. We conclude that previously described mechanisms of iodine outgassing [7] is compatible with the observed shallow nature of traps introduced as well as the asymmetric process of formation/annihilation.