Università degli Studi di Milano Simulates Dynamics With Wavepackets

Researchers at Università degli Studi di Milano have developed a new semiclassical method for simulating non-adiabatic molecular dynamics that differs from traditional single-surface approaches by employing two coupled Gaussian wavepackets, each evolving on a separate diabatic potential energy surface. This technique allows for a more nuanced modeling of complex molecular behavior, particularly in scenarios where the Born-Oppenheimer approximation breaks down. To validate their method, the team first performed a check to ensure both approximations reproduce Rabi oscillations, before applying it to two non-adiabatic potential energy scenarios. The first involves two coupled displaced harmonic oscillators, as in a typical electron transfer reaction, and the second comprises a Morse potential coupled to an upper dissociative state, modeling a photo-dissociation process. In both scenarios, the variational thawed Gaussian approach is quite accurate, while the standard thawed Gaussian one fails to fully capture the non-adiabatic effects. Ultimately, the researchers report reproducing non-adiabatic molecular dynamics by means of two classical trajectories without introducing any artificial jump or other ad-hoc non-classical effects.

This departure from traditional single-surface approaches, detailed with a license date of July 10, 2026, promises a more accurate depiction of non-adiabatic processes, those where the Born-Oppenheimer approximation breaks down. The team, based at Dipartimento di Chimica, Università degli Studi di Milano, began with a check using Rabi oscillations. After this check, the methods were applied to two non-adiabatic potential energy scenarios: two coupled displaced harmonic oscillators, as in a typical electron transfer reaction, and a Morse potential coupled to an upper dissociative state, modeling a photo-dissociation process. The variational thawed Gaussian approach is quite accurate, while the standard thawed Gaussian approach fails to fully capture the non-adiabatic effects. Non-adiabatic molecular dynamics is reproduced by means of two classical trajectories without introducing any artificial jump or other ad-hoc non-classical effects, offering a potentially powerful tool for understanding complex chemical processes.

Current approaches to modeling non-adiabatic molecular dynamics, processes where the Born-Oppenheimer approximation breaks down, often rely on computationally intensive methods like multi-configurational time-dependent Hartree or surface hopping techniques. Simplified mixed quantum-classical methods have also gained traction, yet accurately capturing nuclear quantum effects remains a significant challenge. This new technique differs from traditional single-surface approaches by simultaneously propagating wavepackets on both electronic states, allowing for population transfer induced by the coupling between surfaces without introducing any artificial jump or other ad-hoc non-classical effects. The classical equations of motion are derived in one case by enforcing the thawed Gaussian ansatz, and in the other by applying the time-dependent variational principles to the thawed Gaussian ansatz. “The idea is to reproduce quantum electronic transitions by allowing the wavepacket shape to change, without introducing any artificial hops or forcing classical trajectories to change potential energy surface,” the researchers explain.

The team’s innovation centers on driving these wavepackets using classical equations of motion, derived in one case by enforcing the thawed Gaussian ansatz, and in the other by the time-dependent variational principles. To rigorously test their method, the researchers first performed a check with Rabi oscillations, confirming its ability to reproduce a well-understood quantum phenomenon before applying these methods to two non-adiabatic potential energy scenarios. In both scenarios, the variational thawed Gaussian approach is quite accurate, while the standard thawed Gaussian one fails to fully capture the non-adiabatic effects.

The pursuit of accurate molecular simulations has led researchers to refine methods for modeling non-adiabatic processes, crucial for understanding everything from photosynthesis to atmospheric chemistry. This dual-wavepacket strategy aims to capture the subtle interplay between electronic states during chemical reactions, offering a more complete picture of molecular behavior. The team rigorously tested their new non-adiabatic thawed Gaussian wavepacket dynamics (NA-TGWD) and non-adiabatic variational thawed Gaussian wavepacket dynamics (NA-VTGWD) methods, performing a check with Rabi oscillations. This ensured the models could accurately reproduce a well-understood quantum phenomenon before applying them to two scenarios: coupled harmonic oscillators, as in a typical electron transfer reaction, and a Morse potential coupled to an upper dissociative state, modeling a photo-dissociation process.

Conventional validation of molecular dynamics simulations often relies on reproducing established experimental data, but the team behind this new semiclassical approach performed a check with Rabi oscillations, a well-understood quantum phenomenon, to ensure the underlying physics was accurately captured. This deliberate step, while seemingly basic, underscores a commitment to rigorous model verification before applying the methods to two non-adiabatic potential energy scenarios. This pairing allowed them to model the exchange of population between electronic states, mirroring the behavior observed in Rabi oscillations.

A novel approach to modeling molecular behavior utilizes paired wavepackets evolving on separate potential energy surfaces, diverging from traditional single-surface methods and offering a more detailed picture of complex chemical processes. Researchers at Dipartimento di Chimica, Università degli Studi di Milano have developed methods employing two coupled Gaussian wavepackets, each propagating on a distinct diabatic potential, to simulate non-adiabatic dynamics, situations where the Born-Oppenheimer approximation breaks down. This technique aims to accurately represent electronic transitions without introducing any artificial jump or other ad-hoc non-classical effects. The team performed a check with Rabi oscillations, a well-understood quantum phenomenon. The methods were applied to two non-adiabatic potential energy scenarios: two coupled displaced harmonic oscillators, as in a typical electron transfer reaction, and a Morse potential coupled to an upper dissociative state, modeling a photo-dissociation process. In both scenarios, the variational thawed Gaussian approach is quite accurate, while the standard thawed Gaussian one fails to fully capture the non-adiabatic effects. This approach offers a promising pathway for simulating molecular dynamics with enhanced accuracy and physical realism.

Researchers are increasingly focused on methods that move beyond single-surface descriptions, seeking to model the interplay between electronic and nuclear motion with greater fidelity. The team performed a check using Rabi oscillations, confirming its ability to reproduce a well-understood quantum phenomenon before applying these methods to two distinct systems: coupled harmonic oscillators, as in a typical electron transfer reaction, and a Morse potential modeling a photo-dissociation process. This rigorous validation process demonstrates a commitment to ensuring the reliability of the approach.

Researchers at Dipartimento di Chimica, Università degli Studi di Milano are refining methods for simulating molecular behavior with quantum accuracy, but even the most sophisticated approaches have limitations. After a check where both approximations reproduce Rabi oscillations, the underlying approximations inherent in single-trajectory methods present challenges in accurately capturing complex dynamics. The researchers found that the standard thawed Gaussian approach fails to fully capture the non-adiabatic effects. This highlights a crucial point: while computationally efficient, relying on a single trajectory can compromise the fidelity of the simulation, especially when electronic transitions are prominent. The core issue, as the team explains, lies in the approximation of quantum behavior with classical trajectories.