Data-driven optimization and machine learning analysis of compatible molecules for halide perovskite material

doi:10.1038/s41524-024-01297-4

	Data-driven optimization and machine learning analysis of compatible molecules for halide perovskite material
	Wang, Shaojun1 ; Huang, Yiru 2; Hu, Wenguang 2; Zhang, Lei 2
	2024-05-29
发表期刊	NPJ COMPUTATIONAL MATERIALS
卷号	10 期号:1
摘要	Optoelectronic stability of halide perovskite material in hostile conditions such as water is rather limited, preventing them from further industrial deployment. Here, we optimize and perform machine learning analysis on CH3NH3PbI3 materials with additives, solvents and post-treatment molecules using combined experimental and data-driven methods. A champion system consisting of a compatible tertiary molecular combination 'calcein+PbBr2 + DMSO' active at diverse surfaces is identified, delivering a large aqueous photoelectrochemical (PEC) photocurrent of 10-5 A/cm2 and an improved aqueous stability of 92.5%. Subsequently, machine interpretation is provided to decouple the multi-molecule contributions with the assistance of genetic programming (GP) and extra-trees (ET) machine learning models, highlighting the intricate molecular features for the target outputs. The post-hoc density functional theory (DFT) calculation suggests the presence of multiple hydrogen bond and anionpi surface interactions to stabilize the interfacial structures. The present 'PEC + GP + ET + DFT' approach is suggested to be an effective approach to design and comprehensively evaluate molecule-modified materials.
关键词	Additives Density functional theory Design for testability Genetic algorithms Hydrogen bonds Lead compounds Machine learning Molecules Perovskite Photoelectrochemical cells Condition Data-driven methods Data-driven optimization Extra-trees Halide perovskites Industrial deployment Machine-learning Photoelectrochemicals Post treatment Solvent treatment
DOI	10.1038/s41524-024-01297-4
收录类别	SCIE ; EI
语种	英语
WOS研究方向	Chemistry ; Materials Science
WOS类目	Chemistry, Physical ; Materials Science, Multidisciplinary
WOS记录号	WOS:001234702700001
出版者	NATURE PORTFOLIO
EI入藏号	20242316212332
EI主题词	Genetic programming
EI分类号	482.2 Minerals ; 702.1 Electric Batteries ; 723.1 Computer Programming ; 723.4 Artificial Intelligence ; 801.4 Physical Chemistry ; 803 Chemical Agents and Basic Industrial Chemicals ; 922.1 Probability Theory ; 931.3 Atomic and Molecular Physics ; 931.4 Quantum Theory ; Quantum Mechanics
原始文献类型	Article
EISSN	2057-3960
引用统计	被引频次：3[WOS] [WOS记录] [WOS相关记录]
文献类型	期刊论文
条目标识符	http://ir.lzufe.edu.cn/handle/39EH0E1M/36242
专题	信息工程与人工智能学院
通讯作者	Zhang, Lei
作者单位	1.Lanzhou Univ Finance & Econ, Dept Elect Commerce, Lanzhou, Gansu, Peoples R China; 2.Nanjing Univ Informat Sci & Technol, Sch Chem & Mat Sci, Dept Mat Phys, Nanjing, Peoples R China
第一作者单位	兰州财经大学
推荐引用方式 GB/T 7714	Wang, Shaojun,Huang, Yiru,Hu, Wenguang,et al. Data-driven optimization and machine learning analysis of compatible molecules for halide perovskite material[J]. NPJ COMPUTATIONAL MATERIALS,2024,10(1).
APA	Wang, Shaojun,Huang, Yiru,Hu, Wenguang,&Zhang, Lei.(2024).Data-driven optimization and machine learning analysis of compatible molecules for halide perovskite material.NPJ COMPUTATIONAL MATERIALS,10(1).
MLA	Wang, Shaojun,et al."Data-driven optimization and machine learning analysis of compatible molecules for halide perovskite material".NPJ COMPUTATIONAL MATERIALS 10.1(2024).