Image-based mesoscale modeling of concrete submitted to high temperature

General information

Project: ANR MULTI-FIRE

Host laboratory: Institut de Mécanique et d’Ingénierie de Bordeaux (I2M) – porous media team.

Type of contract: 3 years PhD, full-time position. Gross salary ≈ 2100 €/month.

Starting date: as soon as possible.

Topic Description

This PhD position is part of the research project ANR MULTI-FIRE (Jan. 2024 - Dec. 2027), collaborative project between the laboratory 3SR of Grenoble (UMR 5521) and the I2M of Bordeaux (I2M, UMR 5295).

Cementitious materials and concrete in particular are the most widely used building materials and their global daily consumption is second only to water. Although a broad body of research has extensively investigated mechanical properties and durability, the predisposition to explosive spalling, when subjected to high temperatures, is often ignored or underestimated. The term spalling refers to the detachment (violent or not) of layers of the surface of concrete exposed to high temperatures; this process reduces the working cross section of the structural element and can expose the reinforcements, leading to structural failure. Spalling is a complex phenomenon that is currently beyond the reach of the typical structural design. The interaction between the thermal load, the evolving moisture content and the evolving permeability of the medium is poorly understood at the fundamental level and their influence on the structural response on the lacks a coherent formalisation into practical criteria. This project aims to further our understanding of the fundamental mechanisms driving spalling at different scales (from molecular to material scale) through a synergic combination of advanced experimental investigations and state-of-the-art numerical modelling.

We believe that standard procedures based on point-wise measurements or mass-loss experiments have intrinsic limitations and that only with a strong interconnection of numeric and innovative experiments, efforts will succeed in understanding the spalling phenomenon. 

Objective of the PhD thesis and Role of the PhD candidate

In this general context, the objective of the thesis addressed in this offer is to develop a digital physical twin of concrete at the material-macroscopic scale accounting explicitly for coarse aggregates. The overall aim is to have a comprehensive mathematical model which will be firstly calibrated with existing imaging data of X-ray/neutron experiments (see video 1) and then used to predict the behavior of a new set of experiments performed during the project (with a different geometry of the specimen and different heating conditions). This methodology will allow us to cross-validate the mathematical model since the parameters identified in the first experimental test will be used to predict the behavior of other experiments, thus testing the data extrapolation potential of our mechanistic approach.

As customary in the thermo-hygro-chemo-mechanical (THCM) modeling approach of I2M and 3SR teams concrete is modelled as a multiphase porous material consisting of a solid phase, s, a liquid phase, l, and the gaseous phase, g, binary mixture of water vapour and dry air. This formulation will be enhanced accounting for the mesoscale structure of concrete thanks to a methodological pipeline which will enable 3D meshing including sub-domains from imaging data. The current bottleneck of this approach is in the structured nature of the mesh which is materially supported and determined by the geometrical arrangement of inclusions. This leads to an excessive geometry-driven mesh refinement which has an important computational cost. A mapping mesh method (MMM) constitutes a promising tool to solve this issue. MMM is essentially based on the use of two meshes. A refined mesh, obtained from imaging and called here material mesh, is adopted to label by superposition the elements of a regular (and less refined) mesh, i.e., the computational mesh. Three distinct domains, each with specific material properties, will be considered: the cement matrix zone (mortar area), the inclusion zone and the interfacial zone. The labeling of the computational mesh will indicate the material properties (according to the three areas). The shape function of the FE computational mesh will be exploited to smooth variation of properties when shifting from a zone to another.(Master) in Mechanics, Physics, Applied Mathematics or a related discipline.

Scientific university degree

(Master) in Civil engineering, Mechanics, Physics, Applied Mathematics or a related discipline.

Required skills of the candidate

The doctoral candidate will be immersed in an inherently multidisciplinary project. The researched student should be curious, open-minded, and capable of original initiatives. He/she should possess a strong background in solid mechanics and physics of heat and mass transport in porous material and a predisposition for the development and implementation (programming) of mathematical models on open-source finite element platforms such as Cast3M and Fenics

Mise à jour le 30/11/2023

Informations

Please send CV, cover letter and letter(s) of recommendation with the reference “Image-based mesoscale modeling of concrete submitted to high temperature” to:

Contact

Giuseppe SCIUME
I2M


Contacter par courriel

Didier LASSEUX
I2M

Contacter par courriel