Effect of Electrical Bias in the Dark on the Flexible Perovskite Modules

In this study, the electrical bias stability of the perovskite flexible module was conducted in the dark state. In general, the solar cell is maintained in the MPP state, but when the module is disconnected during illumination, a voltage close to Voc is applied. Electrical bias causes module degradation by ion migration, charge carrier accumulation, and trap generation, so it is necessary to study the electrical bias stability when the perovskite module is operating and disconnected. The perovskite module was fabricated on a PEN/ITO flexible substrate. The structure of this module is a standard n-i-p structure and uses SnO2 as the ETL material and Spiro-OMeTAD as the HTL material. Au was deposited as the top electrode and all patterns were scribed using a laser and connected monolithically. The cell width is about 0.5cm wide with 5*5cm2. Module A was biased with voltage at the maximum power point(MPP) under 1sun and Vmpp was applied to module B in the dark.<br/>The initial efficiency of module A was 12.82% (Voc: 9.89V, Jsc 2.04mA/cm2, FF: 63.58%) Maximum Power Point (MPP) is tracked in ambient air. The temperature was maintained at approximately 33°C and the humidity was kept below 40%. Modules encapsulated with a diffusion barrier were placed under 1 sun. After 3h MPPT, the efficiency decreased to about 75% of the initial efficiency. The biggest cause of the decrease in efficiency is the decrease in Jsc, which is most likely caused by the decomposition of the perovskite layer, which is a light absorption layer, at a specific location. When the perovskite layer decomposes, Jsc appears to decrease because light absorption and carrier generation decrease. The distribution of elements in the module along the depth direction was measured through SIMS analysis. Measurements were performed on fresh, 25-hour working modules. The intensity of Au near the pattern was higher than that of the neat area. This shows that Au diffused from P2 and the top electrode. In addition, the I2 peak broadened as a result of perovskite decomposition. We obtained electro-luminescence (EL) images before and after MPPT and the intensity of EL decreased around the P2 pattern. Non-radiative recombination sites can be distinguished from EL images in which the perovskite layer disintegrates after ion migration has occurred. Series resistance increased significantly due to the increase of non-conductive materials such as PbI2, which is a by-product of perovskite layer decomposition. Also, the shunt resistance decreased slightly, showing a similar trend to LIT (Lock-in Thermography) results. A thermal imaging camera attached to the LIT system detects Joule heating in the shunt. The number and temperature of hot spots increased.<br/>In the case of module B, the initial efficiency was 14.11% (Voc: 10.43V, Jsc: 2.24mA/cm2, FF: 60.44%). Vmp was applied in the dark state. The temperature was maintained at 24.9°C and the humidity was kept below 50%. Modules are encapsulated in the same way as module A. After 16 hours of application, the efficiency decreased to 11.73(Voc: 9.97V, Jsc 2.17mA/cm2, FF: 54.11%). The biggest change is the reduction of Voc and FF. When a voltage is applied in the absence of light, ions are expected to move along the direction of the electric field, creating non-radiative recombination centers. The result is a decrease in Voc due to recombination. However, ions migrate in the opposite direction when the electric field disappears. When the recovery characteristics were measured again 3 days after the end of the voltage application of the module, the efficiency was 15.23% (Voc: 10.72V, Jsc: 2.23mA/cm2, FF: 63.84%), similar to before voltage application. In the perovskite module, electric bias causes ion migration, creating non-radiative recombination centers, resulting in Voc losses, and light causes decomposition, resulting in Jsc losses.