Abstract:Direct atomistic simulation of nonadiabatic molecular dynamics is a challenging goal that allows important insights into fundamental physical phenomena. A variety of frameworks, ranging from fully quantum treatment of nuclei to semiclassical and mixed quantum–classical approaches, were developed. These algorithms are then coupled to specific electronic structure techniques. Such diversity and lack of standardized implementation make it difficult to compare the performance of different methodologies when treating realistic systems. Here, we compare three popular methods for large chromophores: Ehrenfest, surface hopping, and multiconfigurational Ehrenfest with <i>ab initio</i> multiple cloning (MCE-AIMC). These approaches are implemented in the NEXMD software, which features a common computational chemistry model. The resulting comparisons reveal the method performance for population relaxation and coherent vibronic dynamics. Finally, we study the numerical convergence of MCE-AIMC algorithms by considering the number of trajectories, cloning thresholds, and Gaussian wavepacket width. Our results provide helpful reference data for selecting an optimal methodology for simulating excited-state molecular dynamics.The Supporting Information is available free of charge at <a class="ext-link" href="/doi/10.1021/acs.jpclett.1c00266?goto=supporting-info">https://pubs.acs.org/doi/10.1021/acs.jpclett.1c00266</a>.Transition densities showing the number of excited states included in the nonadiabatic dynamics and the initial state for each system; AIMC simulation parameters used to generate data plotted in <a class="internalNav" href="#fig5">Figure </a><a class="internalNav" href="#fig5">5</a>; plots of relative differences between adiabatic state populations obtained from EHR and AIMC simulations using different parameters compared to our most accurate AIMC results (<a class="ext-link" href="/doi/suppl/10.1021/acs.jpclett.1c00266/suppl_file/jz1c00266_si_001.pdf">PDF</a>)This article has not yet been cited by other publications.

A Novel Energy Conversion Based Method for Velocity Correction in Molecular Dynamics Simulations

Learning the Quantum Centroid Force Correction in Molecular Systems: A Localized Approach

Two Improved Electronegativity Equalization Methods for Charge Distribution in Large Scale Non-Uniform System

Coupling Accelerated Molecular Dynamics Methods with Thermodynamic Integration Simulations

Quantum Fragment Based ab Initio Molecular Dynamics for Proteins.

Effective conservation of energy and momentum algorithm using switching potentials suitable for molecular dynamics simulation of thermodynamical systems

Accelerated Quantum Mechanics/Molecular Mechanics Simulations Via Neural Networks Incorporated with Mechanical Embedding Scheme.

Nonadiabatic Excited-State Molecular Dynamics Methodologies: Comparison and Convergence

AN ADAPTIVE COUPLING MOLECULAR DYNAMICS-SMOOTHED MOLECULAR DYNAMICS (MD-SMD) METHOD FOR NANOMECHANICS

Energy-conserving molecular dynamics is not energy conserving

Branching Corrected Mean Field Method for Nonadiabatic Dynamics

Accelerating ab initio QM/MM Molecular Dynamics Simulations with Multiple Time Step Integration and a Recalibrated Semi-empirical QM/MM Hamiltonian

A quantitative study on the approximation error and speed-up of the multi-scale MCMC (Monte Carlo Markov chain) method for molecular dynamics

Momentum removal to obtain the position‐dependent diffusion constant in constrained molecular dynamics simulation

Free Energy Calculations Using the Movable Type Method with Molecular Dynamics Driven Protein-Ligand Sampling

Accelerating Ab Initio Quantum Mechanical and Molecular Mechanical (QM/MM) Molecular Dynamics Simulations with Multiple Time Step Integration and a Recalibrated Semiempirical QM/MM Hamiltonian.

Incorporating Nuclear Quantum Effects in Molecular Dynamics with a Constrained Minimized Energy Surface

Accurate interatomic force field for molecular dynamics simulation by hybridizing classical and machine learning potentials

Simultaneous Optimization of Biomolecular Energy Functions on Features from Small Molecules and Macromolecules.

Accurate Free Energy Calculation via Multiscale Simulations Driven by Hybrid Machine Learning and Molecular Mechanics Potentials