.png){kind=icon}
返回
Distortion/interaction analysis via machine learning
.png){kind=icon}
OA
AI
DOI：10.1039/d4dd00224e.png)
.png){kind=icon}
原文求助.png){kind=icon}
分享.png){kind=icon}
收藏
摘要
En 中文
Machine learning (ML) models have provided a highly efficient pathway to quantum mechanical accurate reaction barrier predictions. Previous approaches have, however, stopped at prediction of these barriers instead of developing predictive capabilities in reactivity analysis tasks such as distortion/interaction-activation strain analysis. Such methods can provide insight into reactivity trends and ultimately guide rational reaction design. In this work we present the novel application of ML to the rapid and accurate prediction of distortion and interaction DFT energies across four datasets (three existing and one new dataset). We also show how our models can accurately predict on unseen, high impact literature examples where DFT-level distortion/interaction analysis has previously been used to explain reactivity trends for cycloadditions. This work thus provides support for ML to be utilised further in reactivity analysis across different reaction classes at a fraction of the cost of traditional methods such as DFT.
Keyword:
DIELS-ALDER REACTIONS
REACTION BARRIERS
CYCLOADDITIONS
CYCLOALKENONES
SELECTIVITIES
REACTIVITIES
CHEMISTRY
ORIGINS
期刊
IF:
5.6
论文数:
997
被引数:
1.7K
机构
引用论文
Self-assembly of five new organic–inorganic hybrids based on two new flexible tricationic templates基于两个新的柔性三阳离子模板的五个新的有机-无机杂化物的自组装
PREAI
分享收藏
Distortion-accelerated cycloadditions and strain-release-promoted cycloreversions in the organocatalytic carbonyl-olefin metathesis.png)
CHEMICAL SCIENCE
CHEMICAL SCIENCEIF7.4
PREAI
分享收藏
查看更多内容