In hyper-productive research communities in applied research areas such as halide perovskite photovoltaics, often incremental and sometimes serendipitous minor changes lead to continuous optimization and improvement of devices in multiple performance dimensions. Research progress has been primarily reflected in the increase in reported power conversion efficiency. The sheer amount of publications reporting progress in perovskite solar cells is becoming overwhelming and hence difficult to keep track of and to a large degree the data sets generated are by no means findable, accessible, interoperable, and reusable - FAIR.

For this reason, we started a database project in 2019 in an effort to collect data based on "key performance indicators" reported in the domain of perovskite solar cells primarily as a tool to keep track of progress made in the published peer-reviewed literature.[1] In the initial data collection campaign, we collected data from over 42,400 individual perovskite photovoltaic devices that can be viewed, filtered, analyzed, and downloaded dynamically from the database.[2] A year after the launch of the project, we find that despite providing the tools and means to also upload new data, this is not being made use of to a large extent.

In this presentation, I will advocate for the need of our research community to make all their research data - from failed attempts to make small area test devices to long-term performance studies of perovskite modules outdoors - available through central secondary dissemination platforms to accelerate the technological exploitation of perovskites for large-scale solar energy conversion. This will also require expanding the data ontology and data infrastructure further to capture all performance dimensions of PV materials and device exploration and optimization on different technology readiness levels. What will be increasingly valuable for the industrial exploitation of perovskite PV will be a more systematic collection of stability data based on unified testing protocols such as the ISOS protocols [3] and a focus on large area devices to identify current challenges in the scale-up of the device technology. Equally relevant is the dedicated collection of fundamental material properties that enables the prediction of the efficiency and stability of photovoltaic materials to enable the theory and AI-assisted identification of novel materials that could revolutionize the next generation photovoltaics.