Device Characterization and Degradation

To properly assess the degradation of the perovskite solar cells, each device was identically prepared, stored in a nitrogen-filled glovebox, and transferred to a sealed environmental box without exposure to ambient air. A constant flow of either nitrogen, nitrogen and oxygen (80:20, v/v), or humidified nitrogen (30% RH) was connected to an environmental box. The oxygen percentage was controlled by adjusting the relative flow rate of O₂ to N₂ and monitored by a zirconia sensor (Cambridge Sensotet, Rapidox 2100) continuously before being connected to the environmental box. All of the devices were operated under simulated AM 1.5 sunlight at 100 mW cm^–2 irradiance (Abet Sun 3000 Class AAA solar simulator) for 10 h (in open-circuit condition) and then “rested” in the dark for 2 h before another 2 h measurement period, bringing the total experiment time to 14 h. This rest period was performed in response to Nie et al., who demonstrated that short rest periods could “heal” degraded devices.¹⁴ The J–V measurements were performed with a Keithley 2450 source measure unit. The cells were scanned from forward bias to short circuit and back at a rate of 0.5 V s^–1 after being held under illumination at 1.2 V for 2 s. The light intensity was calibrated with a Si reference cell (NIST traceable, VLSI) and corrected by measuring the spectral mismatch between the solar spectrum, the spectral response of the perovskite solar cell, and the reference cell. The mismatch factor was calculated to be approximately 11%.