The Spin-MInt Algorithm: an Accurate and Symplectic Propagator for the Spin-Mapping Representation of Nonadiabatic Dynamics

Lauren E. Cook, James R. Rampton, Timothy J. H. Hele · 2025

Read The Spin-MInt Algorithm: an Accurate and Symplectic Propagator for the Spin-Mapping Representation of Nonadiabatic Dynamics by Lauren E. Cook, James R. Rampton, Timothy J. H. Hele (2025) 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-2512-01579.txt.

Introduction

Lauren Cook, James Rampton, and Timothy Hele present a propagator for spin-mapping dynamics. Mapping methods turn quantum dynamics into classical paths. Spin-mapping is one such method for nonadiabatic acts. Good propagators must be both accurate and symplectic.

Their Spin-MInt algorithm meets both needs. It builds on the Momentum Integral idea, the sole known symplectic choice for a kin method. Tests check its accuracy on standard dynamics cases. Sound propagators let chemists simulate light-driven reactions where electrons and nuclei move together. Symplectic form guards long runs, since energy drift in lesser propagators can fake reaction rates over the long times chemistry needs.

Catalog record

CREATORLauren E. Cook, James R. Rampton, Timothy J. H. Hele
YEAR2025
COLLECTIONPAPERS
FORMATTXT
FILE SIZE77 KB
VERSION1.0
LICENSEhttp://creativecommons.org/licenses/by/4.0/
RIGHTSCC BY
SOURCEarXiv

Source: https://arxiv.org/abs/2512-01579

Browse the full library · About the Tower of Babel

Full text

The opening of the work, from the library's hosted text:

The Spin-MInt Algorithm: an Accurate and Symplectic Propagator for the Spin-Mapping Representation of Nonadiabatic Dynamics Lauren E. Cook, James R. Rampton, Timothy J. H. Hele Abstract Mapping methods, including the Meyer–Miller–Stock–Thoss (MMST) mapping and spin-mapping, are commonly utilised to simulate nonadiabatic dynamics by propagating classical mapping variable trajectories. Recent work confirmed the Momentum Integral (MInt) algorithm is the only known symplectic algorithm for the MMST Hamiltonian. To our knowledge, no symplectic algorithm has been published for the spin-mapping representation without converting to MMST variables and utilising the MInt algorithm. Here, we present the Spin-MInt algorithm which directly propagates the spin-mapping variables. First, we consider a two-level system which maps onto a spin-vector on a Bloch sphere and determine that the Spin-MInt is a symplectic, symmetrical, second-order, time-reversible, angle invariant and geometric structure preserving algorithm. Despite spin-variables resulting in a non-invertible structure matrix, we rigorously prove the Spin-MInt is symplectic using a canonical variable transformation. Computationally, we find that the Spin-MInt and MInt algorithms are symplectic, satisfy Liouville’s theorem, provide second-order energy conservation and are more accurate than a previously-published angle-based algorithm. The Spin-MInt is significantly faster than the MInt algorithm for two electronic states. Secondly, we extend this methodology to more than two electronic states and present accurate population results for a three-state morse potential. We believe this to be the first known symplectic algorithm for propagating the nonadiabatic spin-mapping Hamiltonian and one of the first rigorously symplectic algorithms in the case of non-trivial coupling between canonical and spin systems. These results should guide and improve future simulations. I Introduction Nonadiabatic dynamics governs many interesting phenomena involving light and energy transfer. Simulation is particularly challenging to implement due to the coupled nuclear and electronic degrees of freedom (DoF). Full quantum methods, such as Multi-Configurational Time-Dependent Hartree Fock (MCTDH) and wavepacket methods, 7 ; 88 ; 91 ; 76 ; 88 ; 60 are extremely accurate but limited to smaller systems by the large computational expense associated. For single surface (adiabatic) systems, many classical-scaling methods have been introduced including ring-polymer molecular dynamics (RPMD), the associated centroid molecular dynamics (CMD) and thermostatted (T)-RPMD, and the Linearised Semiclassical Initial-Value Representations (LSC-IVR). 34 ; 13 ; 15 ; 16 ; 14 ; 68 ; 78 ; 59 ; 36 ; 35 ; 38 ; 43 ; 44 ; 24 ; 17 ; 31 ; 81 ; 80 ; 75 ; 90 ; 89 Several extensions for nonadiabatic dynamics are available such as mean-field RPMD, nonadiabatic (N)-RPMD and mapping variable (MV)-RPMD for RPMD, 37 ; 67 ; 5 and Mixed Quantum Classical (MCQ)-IVR for IVR methods. 4 ; 19 ; 6 ; 49 ; 26 ; 82 Additionally, several mixed quantum-classical methods exist including Ehrenfest methods, surface hopping and mapping methods. 45 ; 86 ; 87 ; 73 ; 74 ; 92 ; 58 ; 78 ; 79 ; 71 ; 54 ; 2 ; 1 ; 62 Here, we focus on trajectory-based mapping methods which map the discrete quantum system onto continuous variables which can be propagated classically. One well-established method is the Meyer-Miller-Stock-Thoss (MMST) mapping which obtains electronic positions and momenta and evolves under the MMST Hamiltonian using Hamilton’s equations of motion (EOM). 58 ; 78 Alternatively, spin-mapping obtains a spin-vector on the surface of a Bloch sphere (in the two-level case) which follows Heisenberg’s EOM. 71 ; 72 ; 69 This is an advantageous representation as it removes two redundant DoF compared to the MMST representation. 71 ; 72 The two representations are related, where elements of the spin-vector can be calculated from MMST variables and the Hamiltonians are equivalent under the addition of a zero-point energy (ZPE) parameter to the MMST Hamiltonian. 71 ; 72 ; 12 A ZPE fitting parameter was first introduced by Stock and Müller to mitigate ZPE leakage and many possibilities were tested. 61 ; 77 However, spin-mapping derives this as a function of the number of states and leads to a parameter that is close to what was previously found to be optimal. 71 Historically, spin-mapping was limited to two electronic states. 71 ; 10 However, spin-coherent states and the SU(N) Lie group have been harnessed to generalise to more states and we use this approach to extend this work. 72 ; 12 ; 11 While many of the nonadiabatic ring-polymer methods utilise the MMST mapping, a spin-mapping (SM)-NRPMD has been proposed, 10 ; 11 which alongside the flurry of spin-mapping methods including a partially linearized approach, 55 ; 53 ; 56 an ellipsoidal method, 3 generalisation to multiple states, 12 ; 11 ; 72 and the Mapping Approach to Surface Hopping (MASH), highlights the interest in spin-mapping approaches. 54 ; 47 ; 30 ; 65 ; 70 ; 28 As far as we are aware, none of the mapping methods developed thus far reproduce Rabi oscillations, scale classically with system size, conserve the quantum Boltzmann distribution, have a clear derivation from exact quantum dynamics and produce accurate dynamics. 21 Recent work showed that such a method is unlikely to be obtained through use of MMST variables and the related electronic state descriptors but an alternative metric, such as a spin-mapping metric, may achieve this in future. 21 To investigate dynamical properties of nonadiabatic systems, mapping methods are often utilised to approximate correlation functions of the form, \displaystyle C_{AB}=\frac{1}{Z}\mathrm{Tr}[e^{-\beta\hat{H}}\hat{A}(0)\hat{B}(t)]\text{,} (1) where e^{-\beta\hat{H}} is the quantum Boltzmann distribution at inverse temperature, \beta=1/k_{B}T , \hat{A} and \hat{B} are operators and Z is the partition function given by, \displaystyle Z=\mathrm{Tr}[e^{-\beta\hat{H}}]\text{.} (2) To approximate these, many molecular dynamics trajectories are required to ensure that enough phase-space is sampled. Therefore, we need accurate propagation algorithms. 22 For Hamiltonian systems, algorithms can be symplectic which arises from exact Hamiltonian or sub-Hamiltonian integration and results in no/little energy drift over time when propagating the dynamics. 48 This is also advantageous for methods that have a phase-factor as a symplectic integrator aids convergence. 20 ; 9 ; 59 Although many algorithms have been developed for the MMST Hamiltonian, 20 ; 46 ; 66 ; 89 a recent study has shown that the MInt algorithm is the only rigorously symplectic integrator. 20 ; 22 As far as we are aware, no such algorithm has been proposed for the spin-mapping representation. 71 While an angle-based algorithm has been utilised for the spin-mapping Hamiltonian, it is known to have instabilities and it was subsequently suggested that to obtain symplectic propagation, one should transform into MMST variables and propagate with the MInt algorithm before transforming back. 12 ; 10 ; 72 Therefore, we believe it would be advantageous to be able to propagate the spin-variables directly for the spin-mapping Hamiltonian to reduce both the computational cost and the opportunity for errors to arise. Other algorithms have been suggested for general spin-systems, 84 ; 64 ; 42 ; 50 ; 57 ; 27 ; 40 but many assume a separable Hamiltonian, use a Split-Liouvillian approximation (which we later show results in a loss of symplecticity for the spin-mapping Hamiltonian) or are not applicable to a system with a non-trivial coupling between the nuclear and electronic DoF. Hence, we do not consider them further in this work. In this article, we present a spin-mapping approach to the MInt algorithm (the

Read the complete text