Suppr超能文献

线性响应密度累积理论用于激发电子态。

Linear-Response Density Cumulant Theory for Excited Electronic States.

机构信息

Department of Chemistry and Biochemistry , The Ohio State University , Columbus , Ohio 43210 , United States.

出版信息

J Chem Theory Comput. 2018 Aug 14;14(8):4097-4108. doi: 10.1021/acs.jctc.8b00326. Epub 2018 Jul 5.

Abstract

We present a linear-response formulation of density cumulant theory (DCT) that provides a balanced and accurate description of many electronic states simultaneously. In the original DCT formulation, only information about a single electronic state (usually, the ground state) is obtained. We discuss the derivation of linear-response DCT, present its implementation for the ODC-12 method (LR-ODC-12), and benchmark its performance for excitation energies in small molecules (N, CO, HCN, HNC, CH, and HCO), as well as challenging excited states in ethylene, butadiene, and hexatriene. For small molecules, LR-ODC-12 shows smaller mean absolute errors in excitation energies than equation-of-motion coupled cluster theory with single and double excitations (EOM-CCSD), relative to the reference data from EOM-CCSDT. In a study of butadiene and hexatriene, LR-ODC-12 correctly describes the relative energies of the singly excited 1B and the doubly excited 2A states, in excellent agreement with highly accurate semistochastic heat-bath configuration interaction results, while EOM-CCSD overestimates the energy of the 2A state by almost 1 eV. Our results demonstrate that linear-response DCT is a promising theoretical approach for excited states of molecules.

摘要

我们提出了密度累积量理论(DCT)的线性响应公式,该公式可以同时对许多电子态进行平衡和准确的描述。在原始的 DCT 公式中,仅获得有关单个电子态(通常是基态)的信息。我们讨论了线性响应 DCT 的推导,展示了其在 ODC-12 方法(LR-ODC-12)中的实现,并针对小分子(N、CO、HCN、HNC、CH 和 HCO)中的激发能以及乙烯、丁二烯和己三烯中的挑战性激发态对其性能进行了基准测试。对于小分子,LR-ODC-12 显示出比单重激发和双重激发的运动方程耦合簇理论(EOM-CCSD)更小的激发能平均绝对误差,相对于 EOM-CCSDT 的参考数据。在对丁二烯和己三烯的研究中,LR-ODC-12 正确描述了单重激发 1B 和双重激发 2A 态的相对能量,与高度准确的半随机热浴组态相互作用结果非常吻合,而 EOM-CCSD 高估了 2A 态的能量近 1 eV。我们的结果表明,线性响应 DCT 是一种有前途的分子激发态理论方法。

文献检索

告别复杂PubMed语法,用中文像聊天一样搜索,搜遍4000万医学文献。AI智能推荐,让科研检索更轻松。

立即免费搜索

文件翻译

保留排版,准确专业,支持PDF/Word/PPT等文件格式,支持 12+语言互译。

免费翻译文档

深度研究

AI帮你快速写综述,25分钟生成高质量综述,智能提取关键信息,辅助科研写作。

立即免费体验