Our research is driven by our interest for the coupling between the small scale physical, biological and chemical mechanisms underlying the processes that control soil-like systems, characterised by confined fluids flow, transport & mixing. We use and develop laboratory experiments (mostly microfluidics and video-microscopy), numerical simulations and minimal-ingredient theoretical models. The comparison between micro-scale and field data represents a first step in discovering the laws linking the micro-scale processes to the larger scale behavior, which is not direct in general. The upscaling issue is a major component of our research.
Vortices in dead-end pores control anomalous dispersal
The role of microscopic structure and flow for the dispersal of particles and solutes in disordered systems characterized by dead-end pores has been only poorly understood, in part due to the stagnant nature of these microscopic regions in the global scale of the system. We study how particles are transported, retained and dispersed in such systems. We observe strong tailing of arrival time distributions at the outlet of the medium characterized by power-law decay with an exponent of 2/3. Our results demonstrate how microscopic flow structures can impact macroscopic particle transport. This work is under review.
Transport in heterogeneous unsaturated porous media
This research investigates colloid transport in unsaturated porous media, focusing on how air, water, and the solid matrix shape flow pathways and mobility. Breakthrough curves (BTCs) are measured across different water saturations to quantify how saturation conditions influence colloid arrival dynamics. Pore‑scale imaging is used to distinguish air, water, and solid grains, enabling the reconstruction of their spatial distribution and characteristic sizes. These structural observations are combined with velocity‑field simulations to identify links between phase configuration, flow organization, and colloid transport behavior. Automated time‑lapse microscopy supports long‑duration acquisitions across multiple regions and configurations, providing a dynamic view of colloid movement in partially saturated environments.
Mixing in confined flows
Mixing in heterogeneous confined porous media arises from the interaction between pore‑scale geometry and local flow structure. Continuous (Darcy‑scale) descriptions, which do not resolve the solid matrix, cannot capture the no‑slip and no‑flux boundary conditions that govern mixing at the pore scale. As a result, they fail to represent key mechanisms controlling how solutes and microbial suspensions mix in confined environments. Although flows in porous media typically occur at low Reynolds numbers, mixing remains complex. Some behaviours resemble those found in turbulent or chaotic systems, yet important differences stem from the presence of solid boundaries, stagnant zones, and heterogeneous velocity distributions. These features shape mixing kinetics and spatial patterns, making pore‑scale resolution essential for understanding and modelling mixing dynamics in complex porous structures.
Microbial transport & chemotaxis
Natural soils are host to a high density and diversity of microorganisms, and even deep-earth porous rocks provide a habitat for active microbial communities. In these environments, microbial transport by disordered flows is relevant for a broad range of natural and engineered processes, from biochemical cycling to remineralization and bioremediation. Yet, how bacteria are transported and distributed in the sub-surface as a result of the disordered flow and the associated chemical gradients characteristic of porous media has remained poorly understood. Using a microfluidic model system that captures flow disorder and chemical gradients at the pore scale we quantify the transport and dispersion of the soil-dwelling bacterium Bacillus subtilis in porous media.
Bacterial Colonization Dynamics in Confined Vascular Networks
This project investigates how bacterial plant pathogens colonize confined vascular environments under flow. The plant xylem forms a connected network of vessels that transports water and nutrients, creating a dynamic physical landscape for microbial movement. To study these processes, a microfluidic leaf replica was designed to reproduce the geometry and flow conditions of the xylem. This platform enables direct visualization of pathogen migration against the flow and allows disentangling the biological behaviours and physical constraints that govern colonization within complex vascular networks.
Evaporation in heterogeneous porous media
Description coming soon.
Air bubble generator
This project focuses on the controlled generation of air bubbles in microfluidic porous systems to study how they move and are displaced by an imposed continuous flow. Bubble formation at the microscale follows mechanisms that differ from those observed at larger scales, and is strongly governed by the capillary number, which reflects the balance between viscous forces, fluid velocity, and interfacial tension. At low capillary numbers, surface tension dominates and bubbles form through a squeezing mechanism, where the expanding interface obstructs the continuous phase until a stable neck pinches off. Geometry, wettability of the channel walls, and the pressure dynamics at junctions play a central role in determining bubble size and detachment.
Microfluidics chip in PDMS & NOA81
Boreholes drilled into the ground have traditionally provided the main source of information about subsurface processes, due to the opaque nature of soil and rocks. To overcome these limitations, we build two‑dimensional micro‑model analogs of simplified confined media. These systems consist of two parallel solid layers—at least one transparent—separated by a thin gap filled with impermeable solid obstacles whose size and shape can be precisely designed using photolithography, from micrometric to millimetric scales. In addition to PDMS, we also fabricate microfluidic chips using NOA81, a UV‑curable polymer that is fully impermeable to air. This material property allows us to reproduce confined environments where air retention, air–water interfaces, and unsaturated flow conditions can be controlled with high fidelity.