High Throughput Screening of Highly Efficient Non-fullerene Acceptor based Organic Solar Cells Assisted by a Multi-Dataset Scientific Robot
Enrique Pascual-San-José a, Xabier Rodríguez-Martínez a, Martin Heeney b, Roger Guimerà-Manrique c, Mariano Campoy-Quiles a, Fei Zhuping b
a Institut de Ciència de Materials de Barcelona (ICMAB-CSIC, Campus UAB, Bellaterra, Spain
b Department of Chemistry and Centre for Plastic Electronics, Imperial College London, South Kensington Campus, London, United Kingdom
c Departament d’Enginyeria Química, Universitat Rovira i Virgili, Tarragona, Spain
Proceedings of International Conference on Hybrid and Organic Photovoltaics (HOPV19)
Roma, Italy, 2019 May 12th - 15th
Organizers: Prashant Kamat, Filippo De Angelis and Aldo Di Carlo
Oral, Enrique Pascual-San-José, presentation 183
Publication date: 11th February 2019

Organic photovoltaics (OPVs) have experienced an impressive performance enhancement over the past few years, recently going beyond the 15% milestone [1]. Current record devices rely on the combination of low-bandgap polymers blended with small acceptor molecules (also called non-fullerene acceptors, NFA). The vast majority of studies in the OPV field follows a traditional Edisonian optimization approach for the optimization of cells based on these systems. This, however, entails manufacturing tens to hundreds of devices in order to optimize the solar cell parameters such as: donor:acceptor ratio, thickness of the active layer, solvent system and thermal annealing. Thus, requiring a large amount of semiconductor materials, time and human resources.

We present a novel gradient methodology [2] based on controlled variation of the parameters of interest allowing to save both resources and research time. This methodology consists in three basic steps. First, donor (low bandgap polymer): acceptor (NFA) composition gradient is produced and measured using RamBIC. This tool extracts the composition and thickness from Raman hyperspectral images, and simultaneously and co-locally determines the photocurrent using the same laser to produce LBIC (laser-beam induced current) images. These images contain typically 10.000 data points. The comparison between images identifies the optimum values for thickness and composition. Then, the RamBIC predictions are double checked (and extended to the rest of the cell parameters) by producing doctor-bladed OPV devices with optimum thickness and composition combinations. Finally, the large dataset are used to train a multi-dataset scientific robot that uses artificial intelligence to try to identify an empirical equation that models all RamBIC data (Figure 1) for different OPV system.

The authors would like to acknowledge financial support from the Ministerio de Economía y Competitividad of Spain through the “Severo Ochoa” Programme for Centres of Excellence in R&D (SEV‐2015‐0496) and project MAT2015‐70850‐P; and the European Research Council (ERC) under grant agreement No. 648901.

© Fundació Scito
We use our own and third party cookies for analysing and measuring usage of our website to improve our services. If you continue browsing, we consider accepting its use. You can check our Cookies Policy in which you will also find how to configure your web browser for the use of cookies. More info