Scholarlink

Our attempts to build somewhat realistic models of surface- and nano-catalysisNano-catalysis at finite temperature motivate this progress review. We are bringing in machine learningMachine Learning (ML) (ML), most notably active learningActive Learning (AL) (AL) to, ultimately, define the relevant reaction coordinates. We report progress made on a longer journey towards this goal, by using ALActive Learning (AL) for the global optimization of structures of nanocatalysts such as Ni-Ceria, including vacancies and interactions with water, on the one hand, and data-base searches for better materials for thermo-, electro- and photo-activated catalysis, on the other. Developed protocols and software, QMLMaterial, GAMaterial, MLChem4D, RLMaterial are featured. Initial forays into MLMachine Learning (ML) on quantum computers (QMLQuantum Machine Learning (QML) and QAL) are highlighted.

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

Towards ML- and QML-Accelerated Discovery of Catalytic Materials and Mechanisms—A Progress Review

  • 会议论文

  • 发表日期：

    2025年

查看全文
  • Dennis R. Salahub,
  • Lizandra Barrios Herrera,
  • Mosayeb Naseri,
  • Sankha Ghosh,
  • Hatef Shahmohamadi,
  • Christoph Simon,
  • Hadi Zadeh-Haghighi,
  • Daya Gour,
  • Colin Bellinger,
  • Sergey Gusarov,
  • Jiří Hostaš,
  • Alain Tchagang,
  • Patrizia Calaminici,
  • Andreas M. Köster,
  • Maicon P. Lourenço

摘要

Our attempts to build somewhat realistic models of surface- and nano-catalysisNano-catalysis at finite temperature motivate this progress review. We are bringing in machine learningMachine Learning (ML) (ML), most notably active learningActive Learning (AL) (AL) to, ultimately, define the relevant reaction coordinates. We report progress made on a longer journey towards this goal, by using ALActive Learning (AL) for the global optimization of structures of nanocatalysts such as Ni-Ceria, including vacancies and interactions with water, on the one hand, and data-base searches for better materials for thermo-, electro- and photo-activated catalysis, on the other. Developed protocols and software, QMLMaterial, GAMaterial, MLChem4D, RLMaterial are featured. Initial forays into MLMachine Learning (ML) on quantum computers (QMLQuantum Machine Learning (QML) and QAL) are highlighted.