Public defence, Engineering Physics, MSc Juan Carlos Verano Espitia

Title of the thesis: Statistical physics of creep rupture of heterogeneous materials

Thesis defender: Juan Carlos Verano Espitia 
Opponent: Assistant Professor Péter Dusán Ispánovity, Eötvös University, Hungary 
Custos: Professor Mikko Alava, Aalto University School of Science

The main objective of this thesis was to understand how materials deform and damage when they are subjected to constant loads, i.e., creep. Understanding the physics underlying damage localization as failure approaches is relevant to engineering, materials science, or geophysics. In the classical mechanics and engineering literature, empirical expressions commonly describe creep rupture, and there have been significant advances in a theoretical understanding. Nevertheless, a complete physical theory is lacking, thereby preventing predictions from a more theoretical perspective. 

This project aimed to track progressive damage of a material under creep loading from (i) tensile creep experiments in a heterogeneous material, e.g., common office paper, using monitoring tools such as acoustic emission, and from (ii) numerical models, e.g., the fiber bundle model in one dimension or the progressive damage model in two dimensions, with the last one reproducing the mechanical behavior of geomaterials, e.g., rocks. Then, to apply statistical physics theory to interpret the damage evolution and subsequently predict failure. 

Prior studies have shown that the initial microstructure plays an important role in determining the mean and the variability of failure time. The proposed methodology enhances the precision and reliability of characterizing this initial microstructure. Crucially, this work pushes beyond the current state of the art, in which failure prediction is possible only at the latest stages of deformation in most cases. Therefore, a further outcome of this research is the prediction of creep lifetime via Bayesian inference from early stages. 

The results of this work will help bridge the gap between local mechanical response and early-stage creep rupture prediction. Finally, the methodology developed here may be extended to other heterogeneous materials, thereby expanding its applicability across a wide range of material classes with diverse mechanical properties, including brittle materials that exhibit damage-induced memory effects. Beyond laboratory creep, it should be relevant to geophysical rupture phenomena, where precursor event sequences provide a window into underlying instability and failure, such as landslides, slope creep, or rockfalls.

Keywords: statistical physics, heterogeneous material, thermal activation

Thesis available for public display 7 days prior to the defence at Aalto University's public display page. 

Contact Information:
https://research.aalto.fi/en/persons/juan-carlos-verano-espitia/