Multiscale DFT-SCAPS-1D and explainable machine learning framework for optimizing lead-free A 3 BiBr 3 perovskite-derived solar cells

, and FF of 85.06% and 86.15%. To accelerate device optimization, 4374 SCAPS-generated data points are used to train and evaluate multiple machine-learning and deep-learning models. The best-performing models achieve near-unity predictive accuracy, demonstrating their ability to capture the nonlinear relationships between device parameters and photovoltaic efficiency. SHAP and LIME analyses further reveal that temperature, resistive losses, shunt resistance, and carrier concentration are among the most influential factors governing device performance. These findings provide insights into the photovoltaic potential of Bi-based materials containing mineral-resource-associated elements. Overall, the proposed physics-informed and explainable data-driven framework provides a systematic strategy for identifying promising device configurations and accelerating the computational design of lead-free perovskite-derived solar cells.

Authors

Institutions

Publication Details

Journal
RSC Advances
Published
2026-11-16
DOI
https://doi.org/10.1039/d6ra07527d
Primary Topic
Machine Learning in Materials Science
Type
article
Field-Weighted Citation Impact
0.00

Funders

Controls
|||
ALL TIME
JAN
FEB
MAR
APR
MAY
JUN
JUL
AUG
SEP
OCT
NOV
article

Multiscale DFT-SCAPS-1D and explainable machine learning framework for optimizing lead-free A 3 BiBr 3 perovskite-derived solar cells

Jehan Y. Al‐Humaidi, Samar O. Aljazzar, Md. Hafijur Rahman, Salah Knani et al.
3 links
RSC Advances
Machine Learning in Materials Science
article

Multiscale DFT-SCAPS-1D and explainable machine learning framework for optimizing lead-free A 3 BiBr 3 perovskite-derived solar cells

Jehan Y. Al‐Humaidi, Samar O. Aljazzar, Md. Hafijur Rahman, Salah Knani, Md. Azizur Rahman, Noureddine Elboughdiri, Mohamed Benghanem, Jothi Ramalingam Rajabathar
article en

Abstract

, and FF of 85.06% and 86.15%. To accelerate device optimization, 4374 SCAPS-generated data points are used to train and evaluate multiple machine-learning and deep-learning models. The best-performing models achieve near-unity predictive accuracy, demonstrating their ability to capture the nonlinear relationships between device parameters and photovoltaic efficiency. SHAP and LIME analyses further reveal that temperature, resistive losses, shunt resistance, and carrier concentration are among the most influential factors governing device performance. These findings provide insights into the photovoltaic potential of Bi-based materials containing mineral-resource-associated elements. Overall, the proposed physics-informed and explainable data-driven framework provides a systematic strategy for identifying promising device configurations and accelerating the computational design of lead-free perovskite-derived solar cells.

RSC Advances
Princess Nourah bint Abdulrahman University (SA), Northern Border University (SA), Begum Rokeya University (BD), University of Ha'il (SA), Islamic University of Madinah (SA), Saveetha University (IN)
Princess Nourah Bint Abdulrahman University, Northern Borders University, Islamic University of Madinah
Affordable and clean energy
Machine Learning in Materials Science
AI Navigator

Ask Laika to Summarize, Analyze, and Connect papers live on the map.

Summarize Papers & Methodologies

Extract key findings, datasets, and comparative methods across publications.

Benchmark Rankings & Visual Analytics

Rank top research institutions, authors, funders, topics, and journals by Field-Weighted Citation Impact (FWCI) and paper volume with instant charts.

Connect Distant Disciplines

Bridge topological clusters on the map to find hidden collaborative intersections.

Laika access coming soon