Challenges
Datasets
Workspaces
Papers
Discussions
Leaderboard
Log inSign up
Challenges
Datasets
Workspaces
Papers
Discussions
Leaderboard
Blog
Job Board
Q3AS

© 2026 Aqora Quantum S.A.S.

TermsPrivacyLegal Notice

arXiv:2609.40135

Smite: A quasiclassical trajectory (QCT) program for bimolecular collisions and unimolecular dynamics on ab-initio, and machine-learned potential energy surfaces

Péter Szabó, Jenne van Veerdeghem, Jérôme Loreau, Jean-François Müller, Jeremy N. Harvey

physics.chem-phquant-phSubmitted Sep 30, 2026• Announced Oct 1, 2026
arXivHTMLPDF

Abstract

We present Smite, a Python toolkit for quasiclassical trajectory (QCT) simulations of reactive and inelastic bimolecular collisions, unimolecular dynamics, and molecular collisions with finite surface models. A central feature is the explicit control of reactant preparation: harmonic normal modes admit fixed quantum numbers, prescribed energies, thermal quantum populations, or ground-state Wigner sampling; rotation is prepared at fixed angular-momentum magnitude or from thermal distributions appropriate to diatomic and polyatomic rotors. Rotating-Morse sampling provides a coupled anharmonic treatment of diatomic vibration and rotation, while saved molecular-dynamics phase points provide an alternative source of initial conditions. The same nuclear propagators operate with on-the-fly electronic-structure calculations or user-supplied analytical and machine-learned potential energy surfaces. Additional modules provide constrained rigid-fragment dynamics, thermostat-based preparation, photoionization initial conditions with energy and recoil constraints, and fewest-switches surface hopping (FSSH) on multiple PESs with approximate curvature-derived couplings. Geometry optimization, transition-state searches, crossing-point optimization, intrinsic-reaction-coordinate following, and thermochemistry use the same energy and derivative interfaces. Analysis tools connect trajectories to product energy partitioning, stereodynamical correlations, time-resolved vibrational signatures, and independent-atom x-ray scattering.

Related papers

Simulating Chemical Reaction Dynamics Using Sample-Based Quantum Diagonalization: A Quantum-Centric Supercomputing Approach

Sumit Kumar, Kalpak Ghosh, Deepak Kumar +1 more

Quantum computers offer the potential to simulate chemical systems beyond the capabilities of classical computers. However, current quantum computing approaches to ab initio molecular dynamics (AIMD) and vibrational analysis largely rely on the Variational Quantum Eigensolver (VQE), whose circuit depth, iterative optimization, and measurement overhead hinder the practical simulation of molecular systems. Here, we extend Sample-based Quantum Diagonalization (SQD), a quantum sampling-based electronic-structure method, to AIMD and vibrational analysis on noisy quantum hardware. The resulting framework enables nuclear forces and Hessians to be derived from quantum-computed electronic energies, facilitating molecular dynamics and characterization of potential energy surfaces. We demonstrate the approach using diverse molecular processes, including H2 bond vibration, the proton-exchange reaction (H+ + H2), the bimolecular nucleophilic substitution (SN2) reaction (Cl- + CH3Cl), and proton-transfer dynamics between two water molecules involving transient formation of the Zundel cation (H2O···H+···OH2). Across these systems, SQD-AIMD trajectories and vibrational frequencies obtained with noisy quantum hardware/noise-models show excellent agreement with classical reference calculations and substantially improved accuracy relative to previously reported VQE-based simulations. Importantly, the SQD-based approach eliminates costly variational optimization and significantly reduces measurement requirements relative to VQE, enabling AIMD simulations spanning approximately 500 timesteps (∼200 fs), substantially longer than previously reported AIMD simulations on quantum hardware. These results establish SQD-AIMD and SQD-vibrational analysis as practical and scalable frameworks for simulating chemical reaction dynamics and elucidating reaction mechanisms on near-term quantum hardware, extending quantum computing applications beyond static electronic structure calculations.

Journal of the American Chemical Society•Oct 12, 2026•0 citations

Algebraic Semi-Classical Model for Reaction Dynamics

Tim Wendler

We use an algebraic method to model the molecular collision dynamics of a collinear triatomic system. Beginning with a forced oscillator, we develop a mathematical framework upon which inelastic and reactive collisions are modeled. The model is considered algebraic because it takes advantage of the properties of a Lie algebra in the derivation of a time-evolution operator. The time-evolution operator is shown to generate both phase-space and quantum dynamics of a forced oscillator simultaneously. The model is considered semi-classical because only the molecule's internal degrees-of-freedom are quantized. The relative translation between the colliding atom and molecule in an exchange reaction (AB+C ->A+BC) contains no bound states and any possible tunneling is neglected so the relative translation is treated classically. The purpose of this dissertation is to develop a working model for the quantum dynamics of a collinear reactive collision. After a reliable model is developed we apply statistical mechanics principles by averaging collisions with molecules in a thermal bath. The initial Boltzmann distribution is of the oscillator energies. The relative velocities of the colliding particles is considered a thermal average. Results are shown of quantum transition probabilities around the transition state that are highly dynamic due to the coupling between the translational and transverse coordinate.

•Jan 1, 2014•0 citations

Quantum-centric machine learning for molecular dynamics

Yanxian Tao, Lingyun Wan, Xiongzhi Zeng +4 more

Accurate and efficient prediction of electronic wavefunctions is central to ab initio molecular dynamics (AIMD) and electronic structure theory. However, conventional ab initio methods require self-consistent optimization of electronic states at every nuclear configuration, leading to prohibitive computational costs, especially for large or strongly correlated systems. Here, we introduce a quantum-centric machine learning (QCML) model-a hybrid quantum-classical framework that integrates parameterized quantum circuits (PQCs) with Transformer-based machine learning to directly predict molecular wavefunctions and quantum observables. By pretraining the Transformer on a diverse dataset of molecules and ansatz types and subsequently fine-tuning it for specific systems, QCML learns transferable mappings between molecular descriptors and PQC parameters, eliminating the need for iterative variational optimization. The pretrained model achieves chemical accuracy in potential energy surfaces, atomic forces, and dipole moments across multiple molecules and ansatzes, and enables efficient AIMD simulations with infrared spectra prediction. This work establishes a scalable and transferable quantum-centric machine learning paradigm, bridging variational quantum algorithms and modern deep learning for next-generation molecular simulation and quantum chemistry applications.

physics.chem-phCross-listedarXiv:2511.07771•Nov 12, 2025•1 citation

Translational cooling versus vibrational quenching in ultracold OH−–Rb collisions: A quantum assessment

M. Tacconi, Francesco Antonio Gianturco

A new three-dimensional potential energy surface (PES), which describes the interaction between the OH(-)(X (1)Sigma(+)) anionic molecule and the Rb((2)S) atomic gas has been obtained by using a highly correlated ab initio model of the electronic structure and a large atomic basis set supplemented with a set of midbond Gaussian functions. This accurate PES is being employed to evaluate the OH(-)(X (1)Sigma(+)) vibrational quenching efficiency induced by collisions with Rb((2)S) at low and ultralow energies. Calculations employ the coupled-channel ansatz to describe the nuclear quantum dynamics and finally show a very marked dependence of the inelastic cross sections and rates, at vanishing collision energies, on the initial vibrational energy content of the partner molecules. The present findings allow us to make specific suggestions on the likely outcomes from energy interplay between the title species within a mixed MOT/Paul Trap environment.

The Journal of Chemical Physics•Sep 2, 2009•9 citations

Benchmarking various nonadiabatic semiclassical mapping dynamics methods with tensor-train thermo-field dynamics

Zengkui Liu, Ningyi Lyu, Zhubin Hu +3 more

Accurate quantum dynamics simulations of nonadiabatic processes are important for studies of electron transfer, energy transfer, and photochemical reactions in complex systems. In this comparative study, we benchmark various approximate nonadiabatic dynamics methods with mapping variables against numerically exact calculations based on the tensor-train (TT) representation of high-dimensional arrays, including TT-KSL for zero-temperature dynamics and TT-thermofield dynamics for finite-temperature dynamics. The approximate nonadiabatic dynamics methods investigated include mixed quantum-classical Ehrenfest mean-field and fewest-switches surface hopping, linearized semiclassical mapping dynamics, symmetrized quasiclassical dynamics, the spin-mapping method, and extended classical mapping models. Different model systems were evaluated, including the spin-boson model for nonadiabatic dynamics in the condensed phase, the linear vibronic coupling model for electronic transition through conical intersections, the photoisomerization model of retinal, and Tully's one-dimensional scattering models. Our calculations show that the optimal choice of approximate dynamical method is system-specific, and the accuracy is sensitively dependent on the zero-point-energy parameter and the initial sampling strategy for the mapping variables.

The Journal of Chemical Physics•Jul 9, 2024•20 citations

Nonadiabatic Molecular Dynamics with Fermionic Subspace-Expansion Algorithms on Quantum Computers

Anthony Gandon, Alberto Baiardi, Pauline J. Ollitrault +1 more

We introduce a novel computational framework for excited-state molecular quantum dynamics simulations driven by quantum-computing-based electronic-structure calculations. This framework leverages the fewest-switches surface-hopping method for simulating the nuclear dynamics and calculates the required excited-state transition properties with different flavors of the quantum subspace expansion and quantum equation-of-motion algorithms. We apply our method to simulate the collision reaction between a hydrogen atom and a hydrogen molecule. For this system, we critically compare the accuracy and efficiency of different quantum subspace expansion and equation-of-motion algorithms and show that only methods that can capture both weak and strong electron correlation effects can properly describe the nonadiabatic effects that tune the reactive event.

Journal of Chemical Theory and Computation•Jul 5, 2024•12 citations
We present Smite, a Python toolkit for quasiclassical trajectory (QCT) simulations of reactive and inelastic bimolecular collisions, unimolecular dynamics, and molecular collisions with finite surface models. A central feature is the explicit control of reactant preparation: harmonic normal modes admit fixed quantum numbers, prescribed energies, thermal quantum populations, or ground-state Wigner sampling; rotation is prepared at fixed angular-momentum magnitude or from thermal distributions appropriate to diatomic and polyatomic rotors. Rotating-Morse sampling provides a coupled anharmonic treatment of diatomic vibration and rotation, while saved molecular-dynamics phase points provide an alternative source of initial conditions. The same nuclear propagators operate with on-the-fly electronic-structure calculations or user-supplied analytical and machine-learned potential energy surfaces. Additional modules provide constrained rigid-fragment dynamics, thermostat-based preparation, photoionization initial conditions with energy and recoil constraints, and fewest-switches surface hopping (FSSH) on multiple PESs with approximate curvature-derived couplings. Geometry optimization, transition-state searches, crossing-point optimization, intrinsic-reaction-coordinate following, and thermochemistry use the same energy and derivative interfaces. Analysis tools connect trajectories to product energy partitioning, stereodynamical correlations, time-resolved vibrational signatures, and independent-atom x-ray scattering.
Quantum computers offer the potential to simulate chemical systems beyond the capabilities of classical computers. However, current quantum computing approaches to ab initio molecular dynamics (AIMD) and vibrational analysis largely rely on the Variational Quantum Eigensolver (VQE), whose circuit depth, iterative optimization, and measurement overhead hinder the practical simulation of molecular systems. Here, we extend Sample-based Quantum Diagonalization (SQD), a quantum sampling-based electronic-structure method, to AIMD and vibrational analysis on noisy quantum hardware. The resulting framework enables nuclear forces and Hessians to be derived from quantum-computed electronic energies, facilitating molecular dynamics and characterization of potential energy surfaces. We demonstrate the approach using diverse molecular processes, including H2 bond vibration, the proton-exchange reaction (H+ + H2), the bimolecular nucleophilic substitution (SN2) reaction (Cl- + CH3Cl), and proton-transfer dynamics between two water molecules involving transient formation of the Zundel cation (H2O···H+···OH2). Across these systems, SQD-AIMD trajectories and vibrational frequencies obtained with noisy quantum hardware/noise-models show excellent agreement with classical reference calculations and substantially improved accuracy relative to previously reported VQE-based simulations. Importantly, the SQD-based approach eliminates costly variational optimization and significantly reduces measurement requirements relative to VQE, enabling AIMD simulations spanning approximately 500 timesteps (∼200 fs), substantially longer than previously reported AIMD simulations on quantum hardware. These results establish SQD-AIMD and SQD-vibrational analysis as practical and scalable frameworks for simulating chemical reaction dynamics and elucidating reaction mechanisms on near-term quantum hardware, extending quantum computing applications beyond static electronic structure calculations.
We use an algebraic method to model the molecular collision dynamics of a collinear triatomic system. Beginning with a forced oscillator, we develop a mathematical framework upon which inelastic and reactive collisions are modeled. The model is considered algebraic because it takes advantage of the properties of a Lie algebra in the derivation of a time-evolution operator. The time-evolution operator is shown to generate both phase-space and quantum dynamics of a forced oscillator simultaneously. The model is considered semi-classical because only the molecule's internal degrees-of-freedom are quantized. The relative translation between the colliding atom and molecule in an exchange reaction (AB+C ->A+BC) contains no bound states and any possible tunneling is neglected so the relative translation is treated classically. The purpose of this dissertation is to develop a working model for the quantum dynamics of a collinear reactive collision. After a reliable model is developed we apply statistical mechanics principles by averaging collisions with molecules in a thermal bath. The initial Boltzmann distribution is of the oscillator energies. The relative velocities of the colliding particles is considered a thermal average. Results are shown of quantum transition probabilities around the transition state that are highly dynamic due to the coupling between the translational and transverse coordinate.
Accurate and efficient prediction of electronic wavefunctions is central to ab initio molecular dynamics (AIMD) and electronic structure theory. However, conventional ab initio methods require self-consistent optimization of electronic states at every nuclear configuration, leading to prohibitive computational costs, especially for large or strongly correlated systems. Here, we introduce a quantum-centric machine learning (QCML) model-a hybrid quantum-classical framework that integrates parameterized quantum circuits (PQCs) with Transformer-based machine learning to directly predict molecular wavefunctions and quantum observables. By pretraining the Transformer on a diverse dataset of molecules and ansatz types and subsequently fine-tuning it for specific systems, QCML learns transferable mappings between molecular descriptors and PQC parameters, eliminating the need for iterative variational optimization. The pretrained model achieves chemical accuracy in potential energy surfaces, atomic forces, and dipole moments across multiple molecules and ansatzes, and enables efficient AIMD simulations with infrared spectra prediction. This work establishes a scalable and transferable quantum-centric machine learning paradigm, bridging variational quantum algorithms and modern deep learning for next-generation molecular simulation and quantum chemistry applications.
A new three-dimensional potential energy surface (PES), which describes the interaction between the OH(-)(X (1)Sigma(+)) anionic molecule and the Rb((2)S) atomic gas has been obtained by using a highly correlated ab initio model of the electronic structure and a large atomic basis set supplemented with a set of midbond Gaussian functions. This accurate PES is being employed to evaluate the OH(-)(X (1)Sigma(+)) vibrational quenching efficiency induced by collisions with Rb((2)S) at low and ultralow energies. Calculations employ the coupled-channel ansatz to describe the nuclear quantum dynamics and finally show a very marked dependence of the inelastic cross sections and rates, at vanishing collision energies, on the initial vibrational energy content of the partner molecules. The present findings allow us to make specific suggestions on the likely outcomes from energy interplay between the title species within a mixed MOT/Paul Trap environment.
Accurate quantum dynamics simulations of nonadiabatic processes are important for studies of electron transfer, energy transfer, and photochemical reactions in complex systems. In this comparative study, we benchmark various approximate nonadiabatic dynamics methods with mapping variables against numerically exact calculations based on the tensor-train (TT) representation of high-dimensional arrays, including TT-KSL for zero-temperature dynamics and TT-thermofield dynamics for finite-temperature dynamics. The approximate nonadiabatic dynamics methods investigated include mixed quantum-classical Ehrenfest mean-field and fewest-switches surface hopping, linearized semiclassical mapping dynamics, symmetrized quasiclassical dynamics, the spin-mapping method, and extended classical mapping models. Different model systems were evaluated, including the spin-boson model for nonadiabatic dynamics in the condensed phase, the linear vibronic coupling model for electronic transition through conical intersections, the photoisomerization model of retinal, and Tully's one-dimensional scattering models. Our calculations show that the optimal choice of approximate dynamical method is system-specific, and the accuracy is sensitively dependent on the zero-point-energy parameter and the initial sampling strategy for the mapping variables.
We introduce a novel computational framework for excited-state molecular quantum dynamics simulations driven by quantum-computing-based electronic-structure calculations. This framework leverages the fewest-switches surface-hopping method for simulating the nuclear dynamics and calculates the required excited-state transition properties with different flavors of the quantum subspace expansion and quantum equation-of-motion algorithms. We apply our method to simulate the collision reaction between a hydrogen atom and a hydrogen molecule. For this system, we critically compare the accuracy and efficiency of different quantum subspace expansion and equation-of-motion algorithms and show that only methods that can capture both weak and strong electron correlation effects can properly describe the nonadiabatic effects that tune the reactive event.