Scholarlink

<p>Perovskite solar cells have emerged as leading photovoltaic technologies, owing to their exceptional efficiencies. However, challenges such as stability and scalable manufacturing hinder commercialization. Precursor additive engineering addresses these issues, yet the explosive growth of literature makes it difficult to efficiently utilize domain knowledge. Here, we introduce Perovskite-R1, a specialized large language model with advanced reasoning capabilities tailored for the discovery and design of perovskite precursor additives. By systematically mining and curating 1,232 high-quality scientific publications and integrating a comprehensive library of 33,269 candidate materials, we constructed a domain-specific instruction-tuning dataset. Experimental validation of several model-proposed strategies confirms their effectiveness in improving material stability and performance. Our work demonstrates the potential of domain-adapted large language models in accelerating materials discovery and provides an integrated workflow for intelligent, data-driven advancements in perovskite photovoltaic research.</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

Perovskite-R1: a domain-specialized large language model for intelligent discovery of precursor additives and experimental design

  • 期刊论文

  • 出版模式：

    开放获取

  • 发表日期：

    2026年2月10日

查看全文
  • Xin-De Wang,
  • Zhi-Rui Chen,
  • Peng-Jie Guo,
  • Ze-Feng Gao,
  • Cheng Mu,
  • Zhong-Yi Lu

摘要

Perovskite solar cells have emerged as leading photovoltaic technologies, owing to their exceptional efficiencies. However, challenges such as stability and scalable manufacturing hinder commercialization. Precursor additive engineering addresses these issues, yet the explosive growth of literature makes it difficult to efficiently utilize domain knowledge. Here, we introduce Perovskite-R1, a specialized large language model with advanced reasoning capabilities tailored for the discovery and design of perovskite precursor additives. By systematically mining and curating 1,232 high-quality scientific publications and integrating a comprehensive library of 33,269 candidate materials, we constructed a domain-specific instruction-tuning dataset. Experimental validation of several model-proposed strategies confirms their effectiveness in improving material stability and performance. Our work demonstrates the potential of domain-adapted large language models in accelerating materials discovery and provides an integrated workflow for intelligent, data-driven advancements in perovskite photovoltaic research.