Excited state molecular dynamics simulations with pyUNIxMD

  • POSTED DATE : 2021-10-29
  • WRITER : 화학과
  • HIT : 149
  • DATE : 2021년 11월 4일(목) 오후 4시 30분
  • PLACE : Webex

세미나가 다음주 목요일(11월 4일) 오후 4시 30분에 개최됩니다.

많은 참여 부탁드립니다.


제  목 : Excited state molecular dynamics simulations with pyUNIxMD
연  사 : 민승규 교수(UNIST)
일  시 : 2021년 11월 4일(목) 오후 4시 30분


방번호: 170 974 2739



Excited state molecular dynamics simulations with pyUNIxMD


Seung Kyu Min

Department of Chemistry, Ulsan National Institute of Science and Technology (UNIST), S. Korea


The excited state molecular dynamics simulation is a theoretical tool to understand light-matter interactions in nature. The most exact way to describe excited state phenomena is to handle the entire system quantum mechanically, i.e. quantum dynamics. Even though there are several algorithms to perform quantum dynamics with realistic molecules such as multi-configurational time-dependent Hartree (MCTDH) and full multiple spawning (FMS) simulations, one of the most promising ways for excited state molecular dynamics simulations is the mixed quantum-classical (MQC) dynamics which perform quantum mechanical simulations for electrons coupled to classical nuclear dynamics approximately. Within the MQC dynamics, the quality of approximations for the electron-nuclear correlation becomes crucial to handle quantum phenomena such as nuclear wave packet branchings and quantum coherences.

Our group has recently developed the python-based program package, namely pyUNIxMD[1], which provides various MQC algorithms to perform excited state molecular dynamics simulations coupled to various quantum chemistry program packages. The pyUNIxMD provides decoherence corrected surface hopping dynamics based on the exact factorization (DISH-XF) which can treat the proper electron-nuclear correlation. [2] In addition, we compared various decoherence corrections with independent-trajectory approaches. [3] Furthermore, both coupled-trajectory and independent-trajectory approaches are possible with the pyUNIxMD program with potential energy surfaces constructed from machine-learning. [4] In this presentation, I will present a brief introduction of the pyUNIxMD program, the development of additional algorithms implemented in the pyUNIxMD, and the recent numerical applications with the pyUNIxMD. 



[1] I.S. Lee, J.-K. Ha, D. Han, T.I. Kim, S.W. Moon, S.K. Min, J. Comp. Chem., 2021, 42, 1755.

[2] J.-K. Ha, I.S. Lee, S.K. Min, J. Phys. Chem. Lett., 2018, 9, 1097.

[3] P. Vindel-Zandbergen, L.M. Ibele, J.-K. Ha, S.K Min, B.F.E. Curchod, N.T. Maitra, J. Chem. Theory Comput., 2021, 17, 3852.

[4] J.-K. Ha, K. Kim, S.K. Min, J. Chem. Theory Comput., 2021, 17, 694.