On the use of Machine Learning for the smart selection of surface passivant in efficient perovskite solar cells
Giulia Grancini a, Matteo Degani a, Mattia Ragni a, fabiola faini a, Sam Teale c, Bin Chen d, Edward Sargent d, Ian Postuma b
a Department of Chemistry, University of Pavia, Italy
b Department of Physics, University of Pavia, Pavia, Italy
c Clarendon Laboratory, University of Oxford, Oxford OX1 3PU, Reino Unido, United Kingdom
d Department of Chemistry, Northwestern University, Evanston, USA, Sheridan Road, 2145, Evanston, United States
NIPHO
Proceedings of International Conference on Perovskite Thin Film Photovoltaics and Perovskite Photonics and Optoelectronics (NIPHO24)
Sardinia, Italy, 2024 June 17th - 18th
Organizers: Giulia Grancini, Francesca Brunetti and Maria Antonietta Loi
Oral, Matteo Degani, presentation 017
Publication date: 25th April 2024

Solution-based approach used to fabricate hybrid halide perovskites (HPs) induces polycrystallinity in the layer so that a massive amount of defects (10 16 cm ^-3 ) is generally formed, with a consequent detrimental effect on the open circuit voltage (Voc ) of the respective perovskite solar cells (PSCs). To face this issue, surface passivation of HPs has been improved and large organic salts are ranked among the best candidates to perform it [1]. Unfortunately, the exploration of a plethora of organic salts is driven by the trial-and-error approach, which is time and money-consuming, with limited insight into the organic salts’ features which most affect the effectiveness of the passivation. In this context machine learning (ML) methods may emerge as a valuable tool to guide experimental efforts thanks to a deeper comprehension of the passivation mechanism itself. In this study, we propose a ML approach with Shapley additive explanation to get the organic cation’s features governing the Voc optimization. We found that low halide fraction and hetero atom carbon ratio correlate with increased PSCs Voc. Through optical and morphological characterization of HPs, we concluded that the beneficial role of low halide fraction is dominated by the light Cl- , whose, with its strong binding capacity to positively charged defects in HPs, reduces non-radiative recombination, while the low hetero atom carbon ratio affects the increased flexibility of the molecule, resulting in better coverage of the surface. Finally, we assessed and confirmed the force of the ML algorithm, by using it directly to make predictions about the experimental Voc that would obtain a PSC in which the HP is passivated with a new cation [2].

[1] Sam Teale, Matteo Degani, Bin Chen, Edward Sargent, Giulia Grancini, Nature Energy, accepted
[2] Mattia Ragni, Fabiola Faini, Matteo Degani, Ian Postuma, Giulia Grancini, submitted

The authors acknowledge the “HY-NANO” project that received funding from the European Research Council (ERC) Starting Grant 2018 under the European Union’s Horizon 2020 research and innovation program (Grant Agreement No. 802862) and the “SPIKE” project that received funding from the European Research Council (ERC) Proof of Concept 2022 under the European Union’s Horizon 2020 research and innovation program (Grant Agreement No. 101068936) and the support from the Ministero dell’Università e della Ricerca (MUR) and the University of Pavia through the program “Dipartimenti di Eccellenza 2023-2027".

© FUNDACIO DE LA COMUNITAT VALENCIANA 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