A coupled-trajectory approach for decoherence, frustrated hops and internal consistency in surface hopping
Lea M. Ibele, Eduarda Sangiogo Gil, Peter Schürger, Federica Agostini · 2024
Read A coupled-trajectory approach for decoherence, frustrated hops and internal consistency in surface hopping by Lea M. Ibele, Eduarda Sangiogo Gil, Peter Schürger, Federica Agostini (2024) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2412-04958.txt.
Introduction
Lea Ibele and co-authors present a chemistry paper from 2024 on simulating excited molecules. When light excites a molecule, its atoms move while electrons hop between energy surfaces. A common simulation lets each atom path run on its own and hop when rules say so. That scheme suffers from fading coherence, blocked hops, and clashes between its parts.
The team's scheme couples the paths so they move as one swarm. The swarm as a whole keeps its energy fixed, and paths share energy when a hop occurs. That single change tackles all three flaws at once. Tests on two light-driven molecules gave sound results for electron and vibration traits over time. The authors base the scheme on an exact split of the full quantum problem.
Catalog record
| CREATOR | Lea M. Ibele, Eduarda Sangiogo Gil, Peter Schürger, Federica Agostini |
|---|---|
| YEAR | 2024 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 28 KB |
| VERSION | 1.0 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2412-04958
Browse the full library · About the Tower of Babel
Full text
The opening of the work, from the library's hosted text:
A coupled-trajectory approach for decoherence, frustrated hops and internal consistency in surface hopping Lea M. Ibele, Eduarda Sangiogo Gil, Peter Schürger, Federica Agostini Abstract We address the issues of decoherence, frustrated hops and internal consistency in surface hopping. We demonstrate that moving away from an independent-trajectory picture is the strategy which allows us to propose a robust surface hopping scheme overcoming all these issues at once. Based on the exact factorization and on the idea of coupled trajectories, we consider the swarm of trajectories, that mimic the nuclear dynamics in nonadiabatic processes, as a unique entity. In this way, imposing energy conservation of the swarm and allowing the trajectories to share energy when hops occur clearly indicates the route towards a new surface hopping scheme. Encouraging results are reported, in terms of electronic and vibrational time-dependent properties on the photodynamics of fulvene, modeled with a full-dimensional linear vibronic coupling Hamiltonian. Nonadiabatic molecular dynamics is an irreplaceable method in the toolkit of physical chemists to model time-resolved photochemistry, the study of the coupled electron-nuclear evolution of molecules upon light excitation González and Lindh 2021 . Among the many theories available nowadays, surface hopping, since its first appearance in the literature in 1971 Tully and Preston 1971 , has grown to become the method of choice to simulate the nonadiabatic dynamics of molecules and to accompany experimental observations. Tully 1990 ; Ibele et al. 2020 ; Pathak et al. 2020 ; Squibb et al. 2018 ; Avagliano et al. 2022 ; Wang et al. 2016 ; Akimov 2022 ; Tully 2012 ; Qiu et al. 2023 Despite alternative theories being continuously developed, surface hopping remains very popular thanks to the unrivaled accuracy-cost compromise, to its relative implementation simplicity and especially to the fact that it conveys a very intuitive picture of a nonadiabatic process. Mai et al. 2018 ; Barbatti 2011a ; Agostini and Curchod 2019 ; Crespo-Otero and Barbatti 2018 ; Persico et al. 2024 ; Barbatti 2011b In surface hopping, the nuclear wavepacket is represented in terms of a swarm of independent trajectories that classically evolve on the support of electronic potential energy surfaces obtained as the solution of the time-independent Schrödinger equation at a given nuclear configuration. The key nonadiabatic effects, beyond the Born-Oppenheimer approximation, i.e., transitions between electronic states, are mimicked by stochastic hops of the trajectories from one potential energy surface to another usually occurring in localized regions of configuration space. Each trajectory of the swarm is coupled to the quantum-mechanical evolution of the electronic time-dependent state – even though not all surface hopping implementations require this. Suchan et al. 2020 ; Pieroni and Agostini 2021 Surface hopping has its weaknesses, which have been reported extensively in the literature: it suffers from overcoherence, due to the disconnect between the dynamics of the classical nuclei and of the quantum electrons Granucci and Persico 2007 ; Shenvi et al. 2011 ; Subotnik and Shenvi 2011 ; Granucci et al. 2010 ; Subotnik et al. 2016a ; Jaeger et al. 2012 ; Persico et al. 2024 ; Martens 2019 ; Zhu et al. 2004 ; Shu and Truhlar 2023 ; when hops occur, energy conservation for the hopping trajectory is enforced, which causes frustrated hops if a trajectory hopping to a state of higher potential energy does not have enough kinetic energy to compensate for the increase in potential Gómez et al. 2024 ; velocity rescaling related to the energy conservation after the hops is applied quite arbitrarily and strongly influence the dynamics; Plasser et al. 2019 ; Vindel-Zandbergen et al. 2021 ; Ibele and Curchod 2020 ; Toldo et al. 2024 ; Dupuy et al. 2024 using a swarm of independent trajectories to mimic the evolution of a quantum wavepacket misses non-local quantum-mechanical effects Subotnik and Shenvi 2011 ; Martens 2016 ; Pieroni and Agostini 2021 ; Pieroni et al. 2024 ; Subotnik et al. 2016b . These weaknesses are consequences of the fact that surface hopping was not deduced originally from first principles, even though derivations Amati et al. 2023 ; Martens 2016 ; Subotnik et al. 2013 and qualitative justifications Abedi et al. 2013 ; Agostini et al. 2015 were proposed later on. In this work, we address the question: What is missing in surface hopping that would cure all its weaknesses at once? We demonstrate that the idea of coupled trajectories emanating from the exact factorization Abedi et al. 2010 ; Ibele et al. 2024 ; Min et al. 2015 ; Min et al. 2017 is capable to indicate the route to follow. Specifically, this route implies that the “surface hopping” idea is used as a tool to simplify conceptually and numerically the evolutions equations derived from the exact factorization, rather than as the working framework. In this way, we do not need to cure the weaknesses of surface hopping, since the starting point is a formulation which is naturally free from these weaknesses. Therefore, we propose a theoretical framework that fully exploits the coupling of the trajectories and that is combined with the algorithmic efficiency of surface hopping, bypassing introducing the issues of decoherence, frustrated hops and internal consistency. Note that a coupled-trajectory-based surface hopping scheme has been developed previously Pieroni and Agostini 2021 ; Pieroni et al. 2024 , but, as we will show below, it inherits some of the issues of the original surface hopping since the idea of coupled trajectories was not fully “embraced” by all aspects of the algorithm. Instead, our new proposition is robust, stable and accurate, and, thus, we believe that our strategy is an important step forward in the development of novel ideas for nonadiabatic molecular dynamics schemes. The idea of coupled trajectories emerges naturally in the framework of the exact factorization (EF). Abedi et al. 2010 ; Ibele et al. 2024 When the time-dependent molecular wavefunction is factored as the product of a marginal nuclear amplitude, or wavefunction, and a conditional electronic amplitude, the molecular time-dependent Schrödinger equation (TDSE) can be decomposed as an effective nuclear TDSE and an electronic evolution equation. The nuclear dynamics is generated by time-dependent vector and scalar potentials, reminiscent of classical electromagnetic potentials, representing the effect of the electrons. Formally, an elegant and rigorous procedure can be introduced to derived classical-like Hamilton’s equations from a TDSE, which is the procedure followed in context of EF to deduce classical-like nuclear trajectories. Agostini et al. 2018 In turn, the nuclei affect the electronic dynamics entering the evolution equation with their momentum/velocity field and with their spatial distribution. As simple as this, the coupled-trajectory idea emerges in the exact factorization: while the momentum field can be interpreted classically, as it is a local property that can be encoded in a trajectory, the very nature of the spatial nuclear distribution is non local. Therefore, to determine the quantum momentum, as we dubbed the term depending on the nuclear distribution, the trajectories become coupled since only the whole ensemble yields the necessary non-local information. Extensive analyses of the time-dependent potentials of the EF have justified additional approximations, namely (i) the use of purely adiabatic forces to drive the nuclear dynamics, since the time-dependent scalar potential reproduces piecewise the adiabatic shapes Abedi et al. 2013 , and (ii)