Uploaded December 2025 | Updated September 2026, 2 weeks ago
NQCDynamics.jl: Nonadiabatic Quantum Classical Dynamics in Julia by Henry Snowden, Alexander Spears
PreTalx: pretalx.com/juliacon-2025/talk/BY3B8P
Accurate and efficient methods to simulate nonadiabatic and quantum nuclear effects in high-dimensional and dissipative systems are crucial for the prediction of chemical dynamics in the condensed phase. Non-adiabatic effects arise from the transfer of energy between electrons and vibrational motion and are an essential process in the study of photocatalysis, inelastic scattering and chemical reactions. A full quantum dynamical treatment based on first principles is unfeasible for anything but the simplest model systems. To facilitate effective development, code sharing, and uptake of newly developed approximate and efficient mixed quantum-classical dynamics methods, it is important that software implementations can be easily accessed and built upon.
Using the Julia programming language, we have developed the NQCDynamics.jl package and Non-adiabatic Quantum-Classical Dynamics (NQCD) code suite, which provides a framework for established and emerging methods for performing semiclassical and mixed quantum–classical dynamics in the condensed phase (J. Gardner et al., J. Chem. Phys. 2022). At its core, NQCDynamics.jl offers efficient implementations of numerous dynamics methods and offers the community an open space for proof-of-principle implementations of newly published methods. The code wraps around the powerful DifferentialEquations.jl framework, therefore providing users with fine-grained control over simulation accuracy, stability, and performance.
As part of the NQCD suite, NQCModels.jl offers common model Hamiltonians and integrates with the Atomic Simulation Environment (ASE) for performing ab initio electronic structure calculations. The suite also features interfaces to cutting-edge machine learning potentials such as MACE, SchNet, and other neural networks, enabling rapid on-the-fly interatomic potential electronic structure property predictions (W. G. Stark et al. J. Phys. Chem. C 2023). Together, these components create a flexible, high-performance ecosystem for studying complex surface chemistry dynamics with both accuracy and efficiency, which we will showcase as part of this talk (G. Meng et al., Phys. Rev. Lett. 2024, J. Gardner et al., J. Chem. Phys. 2023).
NQCDynamics.jl: Nonadiabatic Quantum Classical Dynamics in Julia by Henry Snowden, Alexander Spears
PreTalx: pretalx.com/juliacon-2025/talk/BY3B8P
Accurate and efficient methods to simulate nonadiabatic and quantum nuclear effects in high-dimensional and dissipative systems are crucial for the prediction of chemical dynamics in the condensed phase. Non-adiabatic effects arise from the transfer of energy between electrons and vibrational motion and are an essential process in the study of photocatalysis, inelastic scattering and chemical reactions. A full quantum dynamical treatment based on first principles is unfeasible for anything but the simplest model systems. To facilitate effective development, code sharing, and uptake of newly developed approximate and efficient mixed quantum-classical dynamics methods, it is important that software implementations can be easily accessed and built upon.
Using the Julia programming language, we have developed the NQCDynamics.jl package and Non-adiabatic Quantum-Classical Dynamics (NQCD) code suite, which provides a framework for established and emerging methods for performing semiclassical and mixed quantum–classical dynamics in the condensed phase (J. Gardner et al., J. Chem. Phys. 2022). At its core, NQCDynamics.jl offers efficient implementations of numerous dynamics methods and offers the community an open space for proof-of-principle implementations of newly published methods. The code wraps around the powerful DifferentialEquations.jl framework, therefore providing users with fine-grained control over simulation accuracy, stability, and performance.
As part of the NQCD suite, NQCModels.jl offers common model Hamiltonians and integrates with the Atomic Simulation Environment (ASE) for performing ab initio electronic structure calculations. The suite also features interfaces to cutting-edge machine learning potentials such as MACE, SchNet, and other neural networks, enabling rapid on-the-fly interatomic potential electronic structure property predictions (W. G. Stark et al. J. Phys. Chem. C 2023). Together, these components create a flexible, high-performance ecosystem for studying complex surface chemistry dynamics with both accuracy and efficiency, which we will showcase as part of this talk (G. Meng et al., Phys. Rev. Lett. 2024, J. Gardner et al., J. Chem. Phys. 2023).
