Scholarlink

<p>Tandem solar cells enhance the absorption of photons by utilizing materials that have different band gaps, allowing them to exceed the efficiency limits set by standard single-junction cells. In these configurations, the upper cell captures high-energy photons, whereas lower-energy photons are transmitted to the lower cell for additional conversion. This research presents a computational study of a two-terminal tandem solar cell that merges a lead-free CsSnI<sub>3</sub> top cell with a silicon bottom cell. The examination was carried out using SCAPS-1D simulations, which were improved with machine learning techniques to fine-tune device parameters like layer thickness and doping levels to achieve current matching and minimize losses from recombination. The CsSnI<sub>3</sub> absorber was meticulously adjusted to improve charge carrier behavior and optimize optical absorption, while the silicon sub-cell was modeled based on the filtered light spectrum from the top cell. The optimized tandem configuration reached a PCE of 29.68%, featuring a Voc of 1.675&#xa0;V, a Jsc of 20. 47&#xa0;mA/cm<sup>2</sup>, and an FF of 86.53%. These findings suggest the potential of machine learning-enhanced design in the development of high-efficiency, lead-free perovskite/silicon tandem solar cells. This methodology not only achieves impressive efficiency but also promotes sustainable and cost-effective photovoltaic technologies for future solar energy advancements.</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

Smart Solar Breakthrough: Machine Learning Powers Next-Gen CsSnI3/Si Tandem Cells

  • 期刊论文

  • 发表日期：

    2026年2月18日

查看全文
  • Mohamed Ait Oufakir,
  • Younes Chrafih,
  • Oussama AitMellal,
  • Loubna Boughlima,
  • Said Khaireddine

摘要

Tandem solar cells enhance the absorption of photons by utilizing materials that have different band gaps, allowing them to exceed the efficiency limits set by standard single-junction cells. In these configurations, the upper cell captures high-energy photons, whereas lower-energy photons are transmitted to the lower cell for additional conversion. This research presents a computational study of a two-terminal tandem solar cell that merges a lead-free CsSnI3 top cell with a silicon bottom cell. The examination was carried out using SCAPS-1D simulations, which were improved with machine learning techniques to fine-tune device parameters like layer thickness and doping levels to achieve current matching and minimize losses from recombination. The CsSnI3 absorber was meticulously adjusted to improve charge carrier behavior and optimize optical absorption, while the silicon sub-cell was modeled based on the filtered light spectrum from the top cell. The optimized tandem configuration reached a PCE of 29.68%, featuring a Voc of 1.675 V, a Jsc of 20. 47 mA/cm2, and an FF of 86.53%. These findings suggest the potential of machine learning-enhanced design in the development of high-efficiency, lead-free perovskite/silicon tandem solar cells. This methodology not only achieves impressive efficiency but also promotes sustainable and cost-effective photovoltaic technologies for future solar energy advancements.