Scholarlink

<p>A global potential energy surface (PES) for the H+HCN reaction has been constructed using the fundamental invariants neural networks (FI-NN) fitting method, based on 38662 <i>ab initio</i> energy points calculated at the UCCSD(T)-F12a/AVTZ theoretical level. The PES not only covers the well-studied abstract channel, but also accurately describes the stationary points and reaction pathways of the exchange channel, achieving an overall root-mean-square error (RMSE) of 2.85 meV. Total integral cross sections of both channels were computed, and the branching ratios of HCN and HNC product isomers were also analyzed, with HNC further distinguished by its formation through three different pathways.</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

A Global Fundamental Invariants Neural Networks Potential Energy Surface and Dynamics Study of the H+HCN Reaction

  • 期刊论文

  • 发表日期：

    2025年9月13日

查看全文
  • Qun Chen,
  • Shu Liu,
  • Bina Fu,
  • Donghui Zhang

摘要

A global potential energy surface (PES) for the H+HCN reaction has been constructed using the fundamental invariants neural networks (FI-NN) fitting method, based on 38662 ab initio energy points calculated at the UCCSD(T)-F12a/AVTZ theoretical level. The PES not only covers the well-studied abstract channel, but also accurately describes the stationary points and reaction pathways of the exchange channel, achieving an overall root-mean-square error (RMSE) of 2.85 meV. Total integral cross sections of both channels were computed, and the branching ratios of HCN and HNC product isomers were also analyzed, with HNC further distinguished by its formation through three different pathways.