Scholarlink

<p>The present work comprehensively investigates the design, optimization, and predictive modeling of a high-performance, lead-free RbGeI<sub>3</sub>-based perovskite solar cell employing WS₂ as the ETL and CuI as the HTL. Numerical simulations performed using SCAPS-1D were validated against theoretical efficiency limits and electrostatic consistency checks, confirming the model’s physical reliability. The device achieved an optimized <i>η</i> of 24.15%, with <i>V</i><sub>oc</sub> = 1.1184 V, <i>J</i><sub>sc</sub> = 25.999 mA·cm<sup>−2</sup>, and FF = 83.09%, demonstrating excellent charge extraction and minimal recombination losses. Systematic parametric analyses revealed the critical influence of absorber doping, transport layer defect density, resistive losses (<i>R</i><sub>s</sub>/<i>R</i><sub>sh</sub>), absorber thickness (t-Abs), defect density (Nt), temperature, and solar irradiance on overall performance. Optimal operation was achieved for Nt = 10<sup>14</sup> cm⁻<sup>3</sup>, where light absorption and carrier transport are well balanced. Furthermore, machine learning (ML) algorithms, including XGBoost, random forest, and gradient boosting, were employed to predict photovoltaic outputs with near-perfect accuracy (<i>R</i><sup>2</sup> ≈ 1.0). The XGBoost model successfully identified absorber defect density, series resistance, and illumination intensity as the most dominant performance-determining features. The results demonstrate that the synergistic combination of WS<sub>2</sub>/CuI transport layers and ML-guided optimization establishes a promising framework for stable, efficient, and eco-friendly RbGeI<sub>3</sub>-based PSCs, paving the way for next-generation lead-free photovoltaic technologies.</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

Integrated simulation and machine learning framework for high-performance lead-free RbGeI3 perovskite solar cells with WS2/CuI transport layers

  • 期刊论文

  • 发表日期：

    2026年1月5日

查看全文
  • Umar Farooq Ali,
  • Qasim Ali,
  • Asif Ali,
  • Hussain Noor,
  • Usama Sohail,
  • Sabah Haider

摘要

The present work comprehensively investigates the design, optimization, and predictive modeling of a high-performance, lead-free RbGeI3-based perovskite solar cell employing WS₂ as the ETL and CuI as the HTL. Numerical simulations performed using SCAPS-1D were validated against theoretical efficiency limits and electrostatic consistency checks, confirming the model’s physical reliability. The device achieved an optimized η of 24.15%, with Voc = 1.1184 V, Jsc = 25.999 mA·cm−2, and FF = 83.09%, demonstrating excellent charge extraction and minimal recombination losses. Systematic parametric analyses revealed the critical influence of absorber doping, transport layer defect density, resistive losses (Rs/Rsh), absorber thickness (t-Abs), defect density (Nt), temperature, and solar irradiance on overall performance. Optimal operation was achieved for Nt = 1014 cm⁻3, where light absorption and carrier transport are well balanced. Furthermore, machine learning (ML) algorithms, including XGBoost, random forest, and gradient boosting, were employed to predict photovoltaic outputs with near-perfect accuracy (R2 ≈ 1.0). The XGBoost model successfully identified absorber defect density, series resistance, and illumination intensity as the most dominant performance-determining features. The results demonstrate that the synergistic combination of WS2/CuI transport layers and ML-guided optimization establishes a promising framework for stable, efficient, and eco-friendly RbGeI3-based PSCs, paving the way for next-generation lead-free photovoltaic technologies.