Efficient SnS Solar Cells via Plasmonic Light Trapping and Alternative Buffer Layers: A Combined Machine Learning and FDTD Analysis

AbstractIn this work, we propose a novel design framework based on combined finite-difference time-domain (FDTD) simulations and machine learning (ML) analysis, aiming to improve the light trapping and carrier transport management in the SnS-based thin-film solar cells (TFSCs). The present study aims to identify the key governing solar cell parameters throughout the structure designing processes in order to apply them to develop efficient eco-friendly SnS TFSCs. In this regard, a new SnS solar cell structure based on alternative electron transport layers (ETLs) combined with plasmonic-based light management approach is proposed. Extensive ML-FDTD analysis is performed to assess the influence of the gold nanoparticles (Au-NPs) distribution and carrier transport layers on the TFSC performance. It is revealed that the proposed design framework can predict the best radius and spatial distribution of Au-NPs and the appropriate ETL. This is attributed to the use of ML approach, allowing the selection of the Au-NPs configuration for enhanced light trapping and carrier transport management. Moreover, the numerical simulated current and voltage values show a consistency with the ML-FDTD predictions. Through ML-FDTD analysis, SnS2 material is found to be the best ETL, while radius and spatial distribution of Au-NPs are other key governing parameters to attain over 12% efficiency. The optimized device shows enhanced open circuit voltage of 0.63 V and improved short circuit current of 27...
Source: Plasmonics - Category: Biomedical Science Source Type: research