Machine Learning–Assisted Optimization of Self-Assembled Monolayer Processing for Efficient p–i–n Perovskite Solar Cells
Gulay Zeynep Gunel a, Sevdiye Basak Turgut a, Burak Kahraman a, Ceylan Zafer a
a Ege University Solar Energy Institute, Ege University Solar Energy Institute 35100 Bornova Izmir Turkey, Izmir, 35100, Turkey
Proceedings of International Conference on Hybrid and Organic Photovoltaics (HOPV26)
Uppsala, Sweden, 2026 May 18th - 20th
Organizers: Gerrit Boschloo, Ellen Moons, Feng Gao and Anders Hagfeldt
Poster, Gulay Zeynep Gunel, 207
Publication date: 11th March 2026

Self-assembled monolayers (SAMs) have emerged as a pivotal strategy for enhancing energy level alignment and mitigating interfacial defects in p-i-n perovskite solar cells (PSCs).1-2 However, identifying the optimal deposition conditions for novel SAM molecules often requires extensive and time-consuming experimental trials. This study presents a systematic, data-driven optimization of the deposition process for a triazatruxene-based SAM on transparent conductive substrates. To navigate the complex, multi-dimensional parameter space—including solution concentration, solvent systems, and thermal annealing—we employed Bayesian optimization to guide the experimental workflow. A surrogate model, iteratively refined through experimental feedback, effectively mapped the relationship between processing parameters and device performance. Initial optimization cycles led to a maximum power conversion efficiency (PCE) of 15.9%, demonstrating the efficiency of the model in identifying promising processing windows with minimal experimental iterations. The chemical anchoring of the SAM to the substrate was confirmed via X-ray photoelectron spectroscopy (XPS). Furthermore, steady-state and time-resolved photoluminescence (PL/TRPL) measurements revealed significantly improved charge extraction dynamics and reduced non-radiative recombination at the SAM/perovskite interface compared to non-optimized devices. These results highlight that Bayesian optimization offers a powerful and scalable methodology for accelerating the development of high-performance interfacial layers in next-generation photovoltaics.

The authors acknowledge financial support from the Strategy and Budget Department of the Presidency of the Republic of Türkiye (Grant No. 16DPT002), Ege University Scientific Research Projects (BAP, Project No: 32965), and TÜBİTAK 1002 – Rapid Support Program (Project No: 125Z201). Gülay Zeynep Günel also acknowledges support from the TÜBİTAK 2211-C National PhD Scholarship Program.

© 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