Publication date: 11th March 2026
The inverse design of tailored organic molecules for specific optoelectronic devices of high complexity holds an enormous potential but has not yet been realized. Current models rely on large data sets that generally do not exist for specialized research fields. We demonstrate a closed-loop workflow that combines high-throughput synthesis of organic semiconductors to create large data sets and Bayesian optimization to discover new hole-transporting materials with tailored properties for solar cell applications. The predictive models were based on molecular descriptors that allowed us to link the structure of these materials to their performance. A series of high-performance molecules were identified from minimal suggestions and achieved up to 26.2% (certified 25.9%) power conversion efficiency in perovskite solar cells. Recent works went beyond 27 % single junction perovskites. But, more important is the general applicability of the workflow, that already shows success in discovering radiation hard materials, photooxidative stable materials or transparent semiconductors.
That milestone underlines the feasibility of developing autonomous research strategies that discover materials tailored for specific applications with highly interconnected workflows including synthesis, purification, characterization and device optimization. However, before such workflows are capable of solving real world problems, optimization has to be thought in a much broader sense, going beyond single-objective optimization (such as efficiency) to multi-objective optimization (such as efficiency, lifetime, toxicity, …). Such workflows could specifically develop fully optimized solar cells, LEDs, photodetectors or X-Ray detectors that can be directly transferred from the lab to the fab. The outlook will summarize the advantages but also the limitations of data driven methods and will give first examples of such campaigns.
