| Prof. | Bertrand DUPLANTIER | Reviewer | Prof. | Joseph KLAFTER | Reviewer | Prof. | Yves POMEAU | Reviewer | Prof. | Olivier BENICHOU | Examiner | Prof. | Jean-Philippe BOUCHAUD | Examiner | Prof. | Antoine GEORGES | President of the jury | Prof. | Pierre LEVITZ | Invited member | Prof. | Satya MAJUMDAR | Invited member | Prof. | Michel ZINSMEISTER | Invited member |
The surrounding world surprises us by the beauty and variety of complex shapes that emerge from nanometric to macroscopic scales. Natural or manufactured materials (sandstones, sedimentary rocks and cement), colloidal solutions (proteins and DNA), biological cells, tissues and organs (lungs, kidneys and placenta), they all present irregularly shaped "scenes" for a fundamental transport "performance", that is, diffusion. Here, the geometrical complexity, entangled with the stochastic character of diffusive motion, results in numerous fascinating and sometimes unexpected effects like diffusion screening or localization. These effects control many diffusion-mediated processes that play an important role in heterogeneous catalysis (reaction rate and overall production), biochemical mechanisms (protein search and migration), electrochemistry (electrode-electrolyte impedance), growth phenomena (solidification, viscous fingering), oil recovery (structure of sedimentary rocks), or building industry (cement hardening). In spite of a long and rich history of academic and industrial research in this field, it is striking to see how little we know about diffusion in complex geometries, especially the one which occurs in three dimensions.
In this work, we present our recent results on restricted diffusion, and describe further research development through ongoing and future projects. We look into the role of geometrical complexity at different levels, from boundary microroughness to hierarchical structure and connectivity of the whole diffusion-confining domain. We develop a new approach which consists in combining fast random walk algorithms with spectral tools. Up to now, our research was mainly focused on studying diffusion in model complex geometries (von Koch boundaries, Kitaoka acinus, etc.), as well as on developing and testing spectral methods. From now on, we aim at extending this knowledge and at applying the accomplished arsenal of theoretical and numerical tools to structures found in nature and industry (X-ray microtomography 3D scans of a cement paste, 2D slices of human skin and the placenta, etc.).