Laboratoire de Chimie et Physique Quantiques - UMR5626

The ’Modelling, Aggregates, Dynamics’ (MAD) team, headed by Aude SIMON since 2020, is composed of three associate professors and four CNRS researchers.

The MAD team develops modelling and numerical simulation activities to study complex atomic and molecular aggregates ranging in size from a few atoms to several thousands of atoms. Its field of research is therefore at the interface between physics and chemistry and involves close interaction with several experimental teams. A significant part of the work carried out by the MAD team concerns the development and implementation of state-of-the art theoretical methods to explore complex potential energy surfaces in order to determine a panel of properties:

 structural (isomer differentiation, structural transition)

 thermodynamic

 infrared spectra

 reactivity

 aggregation and fragmentation

 excited electronic states and electronic spectroscopy

 real-time dynamics of the system after energy deposition (by photon or collision) in the ground state or in excited states

 long-time behavior (evaporation, multifragmentation, etc.)

Left to right, second row : Xavier Gadéa, Mathias Rapacioli, Sophie Hoyau, Fernand Spiegelman ; first row : Camille Alauzet, Quentin Desdion, Fabienne Bessac, Aude Simon, Maia Courtiel, Héloïse Leboucher, Maysa Yusef Buey.

The approaches used by the MAD team are mainly based on molecular dynamics (MD) simulations including biased MD and path-integral MD where the electronic structure is explicitly described by density functional theory (DFT) and DFT-based tight binding (DFTB) approaches. A recent research area of the team is the description of aggregates using supervised machine learning approaches, where interaction potentials are obtained by training neural networks or Deep Neural Network Potentials (DNNP). This is at the core of the recently obtained team ANR project DIAPASONS (2025-2028). The team is currently involved in several collaborative ANR projects, including GROWNANO (2022-2025) and BIRD (2024-2028) among the most recent. It is also heavily involved in the CNRS’s GDR EMIE 3533.