Caleb Boyd1 Rongrong Cheacharoen1 Wanliang Tan1 Michael McGehee1

1, Stanford University, Stanford, California, United States

Hybrid organic-inorganic lead halide perovskite solar cells have made rapid advancements in efficiency, approaching and overtaking those of other thin-film technologies (1). Their cheap and facile deposition, alongside their tunable bandgap, makes them an ideal candidate for tandem devices, with promise to boost the existing efficiency of single-crystal silicon cells at low additional cost (2). Before commercialization can be achieved, however, the stability of perovskite solar cells must be improved. While moisture exposure can be mitigated through careful encapsulation, the thermal stability of the cell, with respect to both intrinsic degradation of the absorber material and extrinsic reactions with other layers, is critical.
We evaluate thermal stability of semitransparent FA0.83Cs0.17Pb(I0.83Br0.17)3 perovskite solar cells at 85C in a nitrogen environment for up to 1000 hours, and show that the primary factor in cell degradation is reaction with a metal contact. Using depth profiling in x-ray photoelectron spectroscopy alongside SIMS techniques, we show that silver, copper, and gold not only create a driving force for iodine migration from the perovskite, but also surprisingly have the potential to diffuse through a sputtered tin-doped indium oxide (ITO) window layer, an atomic layer deposited (ALD) tin oxide layer, and an evaporated fullerene electron transport layer into the perovskite, harming the performance of the perovskite solar cell.
The poor barrier quality of the transparent conducting oxide (TCO) is due largely to diffusion channels in domain boundaries created by a proliferation of the existing rough perovskite morphology. We investigate several solutions, including spin-coating the fullerene layer, adding an ALD titania barrier layer, and using amorphous indium zinc oxide (IZO) as an alternative TCO. We discuss the performance and viability of each solution as well as implications for perovskite/silicon and perovskite/perovskite tandem solar cell design. Solving this problem is critically important because metal grid lines are commonly used to reduce series resistance in modules. Moreover our experiments reveal pinholes that could enable water to enter solar cells or volatile cations (e.g. methylammonium or formamidinium) to leave. Having an extremely impermeable contact should prevent multiple types of degradation.

1 .Best Research-Cell Efficiencies (NREL, 2017);
2. K. A. Bush, A. F. Palmstrom, et al., Nat. Energy 2, 17009 (2017).