Thèse Modélisation Numérique et Physique des Mélanges Grains-Fibres. H/F

Doctorat.Gouv.Fr

Établissement : Université Grenoble Alpes École doctorale : MSTII - Mathématiques, Sciences et technologies de l'information, Informatique Laboratoire de recherche : Laboratoire Jean Kuntzmann Direction de la thèse : Vincent ACARY ORCID 0000000297824051 Début de la thèse : 2026-10-01 Date limite de candidature : 2026-06-09T23:59:59 Les matériaux granulaires interagissent avec des structures élancées déformables dans de nombreux contextes industriels et naturels (végétation stabilisant les dunes, racines, filets de protection), mais les mécanismes physiques régissant ces interactions restent mal compris. Bien que l'efficacité des fibres pour renforcer les sols granulaires soit empiriquement prouvée (protection des sols, amélioration des matériaux de construction), leur influence sur les comportements dynamique et quasi-statique des milieux granulaires en présence de fibres reste peu étudiée, notamment en régime d'écoulement.
Ce projet de thèse vise à comprendre et modéliser les écoulements de mélanges grains-fibres, en identifiant le rôle de paramètres microscopiques clés (concentration en fibres, rapport d'aspect et rigidité en flexion des fibres) sur le comportement macroscopique. Il se focalisera sur les développements théoriques (modélisation mathématique) et numériques (DEM, MPM), tout en conservant une interaction forte avec les expériences en laboratoire menées au FAST dans le cadre d'une collaboration.

Le projet s'articulera autour de trois axes :

- Étudier le couplage microscopique entre l'élasticité des fibres et le contact avec les grains via des simulations DEM d'écoulements granulaires dans des forêts de fibres flexibles.
- Développer et implémenter un modèle mésoscopique biphasique (couplage MPM-structure), où la phase granulaire est décrite par une loi rhéologique continue et les fibres sont simulées individuellement.
- Étendre les modèles continus existants pour les milieux granulaires secs aux mélanges grains-fibres, en proposant une loi rhéologique homogénéisée. Interactions between granular materials and slender deformable structures occur in a wide range of contexts,
from industrial to biological or environmental settings. On the latter case, examples
notably include sand dunes stabilised by vegetation (see Figure 1), plant roots anchoring in sandy
soil, and protective nets installed on mountain slopes to reduce avalanche risks.

The presence of flexible structures significantly restricts the mobility of granular materials. For this
reason, a widespread strategy for improving the mechanical strength of granular media is to add a
small amount of flexible fibres to the grains (Maher & Gray, 1990). This strategy has proven effective
in stabilising soils and sand dunes (Sharma et al., 2016; Shukla, 2017), protecting coastlines and
mountain slopes from erosion (Imran et al., 2022), and enhancing the strength of construction
materials such as adobe (Yetgin et al., 2008) and concrete (Anas et al., 2022).

While the behavior of dry granular materials has been extensively studied over the past decades-
with a well-established physical framework to describe it (see Andreotti et al. (2013) and Forterre &
Pouliquen (2008) for reviews)-the effect of fibres within a granular medium remains poorly
characterised. Investigations of fibre-reinforced granular materials have primarily examined how
fibres influence the static yield strength of the material, largely driven by applications to soil
reinforcement (Zafar et al., 2023). These studies, rooted in soil mechanics, have focused on specific
systems-often involving highly polydisperse particles and fibres with poorly controlled properties,
sometimes complicated by the presence of capillary bridges. As a result, these analyses does not
enable a clear identification of the role of key parameters-such as fibre concentration, aspect ratio
or flexural stiffness-particularly in flow regimes involving significant interaction between fibre
deformation and fibre-grain contacts at the microscopic scale. A unified analytical framework is therefore still
needed to establish the constitutive laws governing the macroscopic behaviour of
grain-fibre mixtures.

The ambition of this PhD thesis is to address these challenges by identifying and quantifying the
influence of key parameters, with the ultimate goal of developing a macroscopic mechanical model
for grain-fibre mixtures.

This project will combine mathematical and numerical modelling, experiments and physical
analyses to study the complex interactions between elasticity and friction at the scale of fibres and
grains, and to assess their influence on the global mechanical properties of the material.
Characterising the microscopic mechanisms that determine the macroscopic properties of the grain-
fibre mixture will enable the development of a homogenised continuous model capable of describing
yield thresholds and internal dissipation through constitutive laws. The implementation of such a
model will provide an effective, robust and validated tool for predicting the behaviour of grain-fibre
mixtures in real-world scenarios while accounting for terrain-specific details at a reasonable cost. In
the long term, this could guide coastal stabilisation campaigns against erosion, as well as efforts to
combat desertification.