Organic photovoltaics (OPVs) have emerged as an attractive option for solar energy conversion. Their potential for low cost, fast manufacturing, and varied applications ranging from small flexible devices to industrial-scale power supply have put them at the forefront of research efforts. Many strategies are being developed to increase the efficiency of these cells, including the design and synthesis of low bandgap donor polymers and non-fullerene acceptors. However, research efforts must also be directed towards the optimization of these new materials in real devices. Variables like film thickness, donor to acceptor ratio, total concentration, solvent, and solvent additive all play crucial and convoluted roles in the efficiency of an OPV . In this study we applied a powerful combined design of experiments and machine learning approach to optimize the bulk heterojunction of a high efficiency PM6:BTP-4Cl composition. This approach elevates the optimization process from simple one-variable-at-a-time screening to a more rational, and predictive, analysis of the interdependence of the most important variables in OPV performance. The aspect of OPV manufacturing we focused upon was the solvent annealing treatment, as it is determinant with respect to the morphology and electronic properties of polymeric thin films, in particular the annealing of the bulk heterojunction promotes phase segregation and domain changes which enhance the properties of the the device.

Pushing the boundaries of OPV optimization: a Design of experiments and machine learning treatment of the high-efficiency PM6:BTP-4Cl system

ACS Spring Meeting 2021

The official archive and repository for all ACS Meetings & Expositions abstracts and select posters or presentations.