Insights From the Perovskite Database Project. What we Can Learn From Data Collected From 45000 Perovskite Solar Cells Reported in the Literature

The recent development of halide perovskite solar cells has been impressive, both in terms of increased performance and in the number of published papers. The sheer number of papers does, however, make it increasingly difficult to get a good overview of the field. It is also difficult to truly take advantage of all device data generated when it is inconsistently formatted and scattered over thousands of papers. We therefore initiated the Perovskite Database Project, which is an open-source communal bottom-up effort to gather all generated perovskite device data in one place, provide consistent formatting, and make it adhere to the FAIR data principles. So far, we have collected device data for around 45000 photovoltaic devices with up to 100 parameters per device. In this talk I will introduce the Perovskite Database Project and give examples of insights gained from analyzing the present dataset. Specific examples will include a quantitative discussion about a discrepancy between short circuit currents extracted from JV and EQE measurements, and a reflection about herd behavior in experimental design. I will also discuss what we can expect from the future in terms of follow up studies, challenges, opportunities, and how we can continue to make as good use as possible of our collective research efforts.