Canada Spends Logo

Explorez notre analyse du budget 2026

Grant: Discovery Grants Program - Individual

Université de Montréal

$48,000.00

Development and application of multiscale modelling approaches to complex materials

EF 2024-25

Résumé
The aging of batteries, the noise from two-level systems in LIGO mirrors, impurity diffusion in high-value steels and aluminum alloys, and the aging of the core in nuclear reactors are just a few examples of problems that are completely determined by the atomistic motion occurring over large spatial and long time scales. Fully characterizing these evolutions experimentally poses significant challenges. Although modeling of these processes has been taking place for decades, the precision has been limited by the lack of reliable force fields that are light enough computationally to be applied to thousands or millions of atoms, as well as the absence of appropriate multiscale approaches. The recent development of machine-learned potentials based on density functional theory (DFT) offers a new opportunity to describe these complex environments. This advancement has reignited efforts to rethink multiscale modeling algorithms. Leveraging the methods I have been developing over the years, which are integrated into the kinetic Activation-Relaxation Technique (k-ART) package-an off-lattice kinetic Monte Carlo algorithm with on-the-fly cataloguing-I propose a research program to create an open-source multiscale framework. The aim of this program is to integrate off-lattice modeling with lattice and object-oriented Kinetic Monte Carlo (KMC) techniques. Development will take place on multiple levels including: (i) the integration of machine-learned potentials with on-the-fly adjustments;  (ii) the development of a multiscale approach to perform this on-the-fly adjustment with DFT precision while simulating  systems many thousands to tens of thousands of atoms;  the development of a bi-directional generation of kinetic objects, i.e. complex structures able to diffuse in materials. In parallel with these methodological developments, the algorithms will be tested and applied to problems in condensed matter and materials science. Initial work will focus on characterization of two-level systems in glasses used, for example, in the LIGO mirrors. Using our on-the-fly ML potentials, we will also look at the response of glass coatings to external strains, a related problem but with different chemistry, as well as aging of batteries, focusing on dendritic growth at the interface. With the integration of on-the-fly generated objects, we will turn our attention to more multiscale problems in materials science looking, in particular, at irradiated materials, in response to issues raised by the nuclear industry as they develop and plan the next generation of nuclear reactors, including small modular reactors. At the end of this major research program, we will have a comprehensive and unique package capable of accurately simulating complex materials over time and length scales relevant to experiments opening the door to significant advances for these systems.
Programme
Discovery Grants Program - Individual
Emplacement
Quebec, Canada
Type de bénéficiaire
Academia
Type
Grant
Source
NSERC's Awards Data

Tel que publié

Application id
RGPIN-2024-04682
Name nom
Mousseau, Normand
Department département
Physique - Physique
Organization id
63
Institution établissement
Université de Montréal
Province en
Québec
Province fr
Québec
Country en
CANADA
Country fr
CANADA
Fiscal year exercice financier
2024
Competition year année de concours
2024
Award amount
48000
Program id
RGPIN
Program name en
Discovery Grants Program - Individual
Program name fr
Programme de subventions à la découverte - individuelles
Group en
DISCOVERY RESEARCH
Group fr
RECHERCHE AXÉE SUR LA DÉCOUVERTE
Committee code
1505
Committee name en
Physics
Committee name fr
Physique
Area of application code
1200
Area of application group en
Advancement of knowledge
Area of application group fr
Avancement des connaissances
Area of application en
Advancement of knowledge
Area of application fr
Avancement des connaissances
Research subject code
3300
Research subject group en
Condensed matterphysics
Research subject group fr
Physique de la matière condensée
Research subject en
Condensed matterphysics
Research subject fr
Physique de la matière condensée
Field of research list names en
Not available
Field of research list names fr
Non disponible
Socioeconomic objective list names en
Not available
Socioeconomic objective list names fr
Non disponible
Application title
Development and application of multiscale modelling approaches to complex materials
Keywords
kinetic monte carlo; activation-relaxation technique; complex materials; defect diffusion; glasses and amorphous materials; machine-learning potentials; long-time relaxation; multiscale simulations; surface growth; batteries
Application summary
The aging of batteries, the noise from two-level systems in LIGO mirrors, impurity diffusion in high-value steels and aluminum alloys, and the aging of the core in nuclear reactors are just a few examples of problems that are completely determined by the atomistic motion occurring over large spatial and long time scales. Fully characterizing these evolutions experimentally poses significant challenges. Although modeling of these processes has been taking place for decades, the precision has been limited by the lack of reliable force fields that are light enough computationally to be applied to thousands or millions of atoms, as well as the absence of appropriate multiscale approaches. The recent development of machine-learned potentials based on density functional theory (DFT) offers a new opportunity to describe these complex environments. This advancement has reignited efforts to rethink multiscale modeling algorithms. Leveraging the methods I have been developing over the years, which are integrated into the kinetic Activation-Relaxation Technique (k-ART) package-an off-lattice kinetic Monte Carlo algorithm with on-the-fly cataloguing-I propose a research program to create an open-source multiscale framework. The aim of this program is to integrate off-lattice modeling with lattice and object-oriented Kinetic Monte Carlo (KMC) techniques. Development will take place on multiple levels including: (i) the integration of machine-learned potentials with on-the-fly adjustments;  (ii) the development of a multiscale approach to perform this on-the-fly adjustment with DFT precision while simulating  systems many thousands to tens of thousands of atoms;  the development of a bi-directional generation of kinetic objects, i.e. complex structures able to diffuse in materials. In parallel with these methodological developments, the algorithms will be tested and applied to problems in condensed matter and materials science. Initial work will focus on characterization of two-level systems in glasses used, for example, in the LIGO mirrors. Using our on-the-fly ML potentials, we will also look at the response of glass coatings to external strains, a related problem but with different chemistry, as well as aging of batteries, focusing on dendritic growth at the interface. With the integration of on-the-fly generated objects, we will turn our attention to more multiscale problems in materials science looking, in particular, at irradiated materials, in response to issues raised by the nuclear industry as they develop and plan the next generation of nuclear reactors, including small modular reactors. At the end of this major research program, we will have a comprehensive and unique package capable of accurately simulating complex materials over time and length scales relevant to experiments opening the door to significant advances for these systems.