Scholarlink

<p>Achieving high efficiency in CZTSSe solar cells requires precise control over both structural and compositional parameters. This work aims to develop a predictive, data-driven design strategy for high-performance and eco-friendly CZTSSe solar cells by optimizing plasmonic and compositional features using deep learning and numerical simulation techniques. We present a comprehensive numerical and deep learning–driven investigation of key design features, including absorber layer thickness, charge transport layer properties, front-contact configuration, embedded gold nanoparticles (Au-NPs), and a tuned <i>S/(S</i> + <i>Se)</i> ratio, which is directly linked to bandgap tuning and electron affinity shifts within the CZTSSe absorber. By combining finite-difference time-domain (FDTD) simulations, the SCAPS-1D tool, and deep learning techniques, we systematically evaluate the role of each parameter in influencing the photovoltaic figures of merit (FoMs). Furthermore, we employ deep learning–enhanced FDTD analysis to explore the synergistic effects of Au-NP size, key design parameters, and <i>S/(S</i> + <i>Se)</i> ratio values on optical absorption and charge extraction performances. The results demonstrate that the integration of Au nanoparticles with optimized bandgap tuning significantly enhances light absorption and charge transport, resulting in an overall power conversion efficiency (<i>PCE</i>) exceeding 23%. These findings validate the effectiveness of our integrated approach and provide a pathway toward the rational design of next-generation sustainable kesterite solar cells.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Deep Learning–Driven Design of Eco-friendly CZTSₓSe₁₋ₓ Solar Cells: Impact of Plasmonic Light Trapping and Bandgap Tuning

  • 期刊论文

  • 发表日期：

    2025年6月18日

查看全文
  • Abdelhak Maoucha,
  • Tarek Berghout,
  • Faycal Djeffal,
  • Hichem Ferhati

摘要

Achieving high efficiency in CZTSSe solar cells requires precise control over both structural and compositional parameters. This work aims to develop a predictive, data-driven design strategy for high-performance and eco-friendly CZTSSe solar cells by optimizing plasmonic and compositional features using deep learning and numerical simulation techniques. We present a comprehensive numerical and deep learning–driven investigation of key design features, including absorber layer thickness, charge transport layer properties, front-contact configuration, embedded gold nanoparticles (Au-NPs), and a tuned S/(S + Se) ratio, which is directly linked to bandgap tuning and electron affinity shifts within the CZTSSe absorber. By combining finite-difference time-domain (FDTD) simulations, the SCAPS-1D tool, and deep learning techniques, we systematically evaluate the role of each parameter in influencing the photovoltaic figures of merit (FoMs). Furthermore, we employ deep learning–enhanced FDTD analysis to explore the synergistic effects of Au-NP size, key design parameters, and S/(S + Se) ratio values on optical absorption and charge extraction performances. The results demonstrate that the integration of Au nanoparticles with optimized bandgap tuning significantly enhances light absorption and charge transport, resulting in an overall power conversion efficiency (PCE) exceeding 23%. These findings validate the effectiveness of our integrated approach and provide a pathway toward the rational design of next-generation sustainable kesterite solar cells.