Meng Zhou


Hi, I am Meng ZHOU, I joined Pr. Philippe Coussot’s research group in Nov. 2015 as a first year PhD student. After the engineering courses at ESPCI ParisTech, I developed my interest in research field of materials and now my work focus on the moisture transport in wood. Wood is a natural porous medium whose microscopic structure is quite different from common construction materials. My co-supervisors are Sabine Care, Denis Courtier-Murias, Paméla Faure, and Stéphane Rodts.

Besides the research, I am passionate about traveling, photographs and also cuisine!


Water transfers in wood

The phenomena of water transfers in wood have been studied at least during one century. However, due to its complexity and the lack of appropriate methods, the mechanisms of water transfers is still under the debate. Wood is indeed a porous medium, made of dead cell with cavity (considered as pores) inside. It is already known that, water could exist in several states inside the wood and thus interacts with this special medium in different ways:

  • Free water, which situate in the cavity of cells (called lumen) and is thought to circulate freely between pores or channels under capillary forces;
  • Bound water, means the water molecules absorbed inside the cell-walls by the formation of hydrogen bond with the wood polymers;
  • Water vapor in pores air.

What’s more, the adsorption of bound water inside cell-walls is the cause of wood deformation (swelling or shrinkage, according to different situation). This means that the structure of wood evaluates within the time.

To better understand this phenomenon we applied the method of MRI, which is a powerful tool capable to identify water molecules in different states (typically free water and bound water) and even to localize them in the space. The 2D images of a wood sample during imbibition (see fig.1) show us that water penetrate in the wood in a heterogeneous way, which could be explained by the distribution of tube sizes in the transversal section. However, 2D images only show us the presence of free water, thus conduct us to further investigate with 1D profiles (see fig.2). For the reason that water molecules in different environment give different MRI signals, by careful data treatment we are finally capable to visualize the penetration of bound water and free water separately during imbibition tests.

Fig.1 : 2D MRI images of the water distribution in wood sample at different times during imbibition: 15 min, 12 h, 24 h and 60h. A view of the sample cross-section is inserted in Figure 2c, showing the correspondence between the main water paths and the sample structure.


Fig.2 : .1D Distribution of absorbed free water (continuous lines) and bound water (short dash) per unit section area (in layers of 1.25 mm thickness) along the sample axis (parallel to imbibition direction) at different times: from left to right, 3 h after first contact with water, then every 6 h. The resulted deformations are also shown here in order to compare with bound water distribution. We could see that, in one hand, free water doesn’t penetrate as fast as expected under the action of capillary pressure; in the other hand, bound water diffuse in the cell-walls under osmotic pressure and advance faster in height than free water, leaving a saturation zone behind its advancing front. Accordance could also be observed between the form of deformation profile and that of bound water.

Marie Goavec


Hello, I’m Marie. I began my Ph.D. in October 2015 to study the drying mechanisms of bitumen emulsions used to make pavements. These emulsified road materials have a substantial environmental potential but their long term mechanical properties are not easy to predict, due to the presence of water. Understanding the impact of water and its movements on the structure of the material, using MRI technology, would be a big step forward.

My work is financed by IFSTTAR, and is co-supervised by Vincent Gaudefroy, Pamela Faure, Stephane Rodts and Emmanuel Keita.

Outside of the lab, I love to dance, sail, and travel around the world!


MRI of bitumen emulsion drying in porous medium

The preparation of cold mix asphalts, composed of aggregates and bitumen emulsion, does not require any heating, unlike hot mixtures. Accordingly, they represent a substantial environmental potential. The presence of water however delays the strengthening of the material, which progressively acquires its mechanical properties as the water evaporates. Our objective is to better understand how the mix dries and how this affects the structure and properties of the material.

We followed the mass of a bitumen emulsion and sand mix exposed to a dry air flux along its free surface. The evolution of the saturation deduced from these measurements shows that the drying of a bitumen emulsion in a porous medium is much slower than that of pure water. The drying rate decreases from the very beginning, a result that strongly differs from drying of a pure liquid in a porous medium, but is similar to what has been observed for the drying of colloidal particle suspensions.

Figure 1 Drying curves for porous medium initially saturated with water and bitumen emulsion

Magnetic resonance imaging (MRI) enabled us to measure the spatial distribution of the two emulsion components in the porous sample during drying. Indeed, due to their very different relaxation times water and bitumen are imaged separately. Recording the signal in thin cross-sectional layers located at a given position we obtain the water and bitumen distributions, or profiles, along the sample axis. We observed that the water distribution does not evolve homogeneously: far from the free surface, the pores remain saturated while an apparent dry front progresses from the free surface. Furthermore, the drying rate measured is much lower than if water diffusion was simply diffusing over the increasing length of the dry front. Although transport and accumulation of bitumen around the free surface of the sample could induce a decrease of the drying rate, it is not the case here: we observed that the bitumen distribution remains homogeneous throughout drying.

Figure 2  Water saturation profiles measured every 12h for a porous medium initially saturated with bitumen emulsion with a thickness of 19mm (increasing time from right to left)

These surprising results have led us to consider the drying of bitumen emulsion on its own. MRI measurements during the drying along the free surface of the emulsion show that near the free surface a compacted front of bitumen droplets forms, before progressing in the sample. A significant decrease in the drying rate accompanies this phenomenon, which seems to indicate that it is the bitumen emulsion in itself, which controls the drying rate in the porous medium.

Xiao Zhang


I joined professor Philippe Coussot as a PhD student in October 2015. My research topic is wall-slip and elongational flow of yield stress fluids. During most of my time in lab I make new born materials (yield stress fluids), make them sit happily on rheometers, sometimes under microscope or MRI, and diagnose them. Through my examinations some theories (e.g., wall-slip) become no longer valid, some others (e.g., 3D flow of complex fluids) turn into open questions. It may seem depressing at first glance but that’s why science is amazing: it’s true whether or not we believe in it!



Wall-slip and elongational flow of yield stress fluids: from rheological measurements to applications on the free surface flow.

Yield stress fluids are encountered in a wide range of applications: cements, mortars, foams, muds, mayonnaise, etc. On smooth surfaces, the materials slip. When wall-slip occurs, the bulk material can move as a rigid block for a very small stress, thus overturn the standard continuum mechanics description assuming adherence. This can be used to facilitate the transport of products such as in food digestion, cosmetic sensory perception, coal water slurry in pipes, fresh concrete pumping over long-distance, removal of food debris, and microbial films.

The wall-slip has been studied in rheometrical conditions for which the sample is confined between two solid surfaces. We carried out experiments aimed at observing the wall-slip of yield stress fluid heaps on an inclined smooth surface. The results support the wall-slip process described in rheometry.

The experimental validation of jamming for yield stress fluids essentially went through simple shear experiments. However, in real flow conditions such as extrusion, blade-coating, squeezing, extension, etc., the flow is more complex as it involves some elongational components. Another project during my PhD is to try to understand how the yielding properties under such conditions are related to the yield stress observed in simple shear, or more generally to the material structure.

Photo Oumar Fadoul

Oumar-Abdoulaye FADOUL

Hi, I’m OUMAR and I’m a second-year PhD student. I joined Professor Philippe Coussot as a PhD student in October 2015. I work on the instability of flow of yield stress fluid in porous medium, and more particularly on the Saffman-Taylor instability which appears when a viscous fluid is pushed by a less viscous fluid in a porous medium. This problem is related to industrial applications such as the last phase of enhanced oil recovery assisted during which one injects into the rock, around the well, different types of non-Newtonian fluids. I am trying to understand the physical origin of the appearance of fingering when air or a liquid pushed a yield stress fluid.


My experiment consists to inject air through a hole in the middle of two parallel plates between which a yield stress fluid layer (emulsion) has been set up. I am looking at the development of the interface in time as the injected air volume growth. While advancing the interface exhibits different shapes depending on the velocity and gap between plates: stable (circular) or unstable (with either fingers of tips).
In a first stage I also studied this instability during the separation of two plates with a layer of paste between them. It appears that the flow is unstable in any case with rough surfaces but can be stable under some conditions when the plates are perfectly smooth. However the empirical criterion of instability still does not correspond to the existing theoretical one.

Photo Gaétan GERBER

Gaétan Gerber

Hi everyone, I’m Gaétan. After I graduated from ESPCI Paris (, I chose in 2015 to broaden my Material Sciences skills with a second Master degree in Innovation and Entrepreneurship, coordinated by Ecole Polytechnique and U.C. Berkeley.

In October 2016, I joined Navier Laboratory and P. Coussot to focus on the transport, clogging and removal of diverse species (particles, pollutants) in model porous media. Techniques to fulfill this goal include Magnetic Resonance Imaging (under the supervision of S. Rodts), microfluidics, microtomography and diverse microscopy approaches.

This PhD project is co-financed and shared between ENPC ( and the Experimental Soft Condensed Matter Group at Harvard University (D.A. Weitz –, which gives me the chance to cultivate my taste for travels and discoveries.

PHOTO Thibault Lerouge

Thibault LEROUGE

Hi ! After graduated from Chimie ParisTech, I started in October 2016 my Ph.D. about multifunctional bio-based porous materials for sustainable construction (funded by LabEx MMCD). This project intends to understand the impact of a double porosity on some transfer and mechanical properties of porous media.

Polymers materials with biporous network are designed as a reference system for multi-scale building materials, and elaborated with Daniel Grande, Co-director of Complex Polymer Systems team, (ICMPE Thiais – CNRS), whilst the physical properties research is supervised by Olivier Pitois (Navier) and Ph. Coussot in Laboratoire Navier. Those materials aims to become bio-based with various surface specificities.

Besides my work, I am a passionate cinephile who tries to travel worldwide.



Hi everyone, I’m Nidal and I’m a first year Ph.D. student financed by The French Alternative Energies and Atomic Energy Commission (CEA de Marcoule). I’m specialized in Hydraulics and Fluid Mechanics.

After I graduated from ENSEEIHT Toulouse, I joined in October 2016 Navier Laboratory and CEA MArcoule to focus on the development of an innovative process for the decontamination of porous materials deeply contaminated by radioelements. The chosen orientation consists in developing a process based on the use of poultices. I am co-supervised by Alban Gossard and Stéphane Rodts.

Nuclear Magnetic Resonance (NMR) appears to be an accurate, non-destructive and reliable method to determine transient moisture distribution during various transport processes. Therefore, NMR is used in this study to characterize the moisture transport between the poultice and the porous substrate during the different steps of the process: imbibition then drying.

Photo Elie N'GOUAMBA


Hello, I am Elie. After graduating from ENSEEIHT in Fluid mechanics, I joined Professor Philippe Coussot as a PhD student in November 2017.

My work is focused on the rheophysics of suspension of salts of vanadium. The general context which motivates this work is the improvement of vanadium flow redox battery which generally works with liquid-liquid electrolytes of vanadium. The idea here is to use liquid-solid (salts) vanadium electrolytes to increase the concentration of vanadium and thus the energy density, to bypass the limitation due to the low solubility of vanadium. However, the electrolyte is a complex fluid whose behavior depends greatly on the other components of the suspension such as carbon black particles added to increase the conductivity of the electrolyte. We aim at understand at best the link between microstructure and rheological behavior of these suspensions. After that I will attempt to predict the flow properties of our suspension in the configuration of our battery.

This work is funded by the French National Research Agency (ANR) and co-supervised by Dr. Julie Goyon (Navier Laboratory).

In my spare time I enjoy running, traveling to discover other cultures and read about theology.



After my PhD (2017) on thermal and hygroscopic transfers in bamboo board, carried out at LOCIE (Univ. de Savoie Mont Blanc), I am currently a postdoc working on the physics of water transfers in softwood under the supervision of Prof. Philippe Coussot and Dr. Sabine Care.

The physical phenomena of water transfers in porous structural materials have been studied long time ago. However, the mechanisms of water transfers (imbibition, drying, and filtration) are still not well understood, specially, with complex structural materials. The recently developed techniques allow to visualize the spacial distribution of absorbed water in the studied materials. They also help to deeper understand these physical phenomena of water transfers.

This project is focusing on the water transfer mechanisms in softwoods. Firstly, the structure of softwoods is assessed through different techniques (X-Ray microtomography, optical microscopy or Scanning Electronic Microscopy techniques).Then, the mechanisms of water transfers, particularly the bound and spatial distribution of free water absorbed into the softwoods, in different initial conditions are carried out by using MRI and NMR techniques. This work gives also a new insight the relationship between microstructure and moisture.

Rakshith SHETTY

Rakshith SHETTY

Hi, I am a third year PhD student at Dept. of Civil Engineering, Indian Institute of Technology Bombay (IIT-B), India, where I am being supervised by Prof. D. N. Singh.  Major theme of my PhD work deals with understanding the rheological behavior of soil and sediment slurries. Soil and sediment are often encountered in their slurry state when dealing with problems related to navigational dredging, mud-flow, debris-flow, offshore landslides and slope-stability to name a few.

At Laboratory Navier (from Sept. to Dec, 2017), I am being advised by Prof. Philippe Coussot in determining the rheological behavior of clay-water mixtures/pastes under shear and elongation flow. The objective is to identify a truly intrinsic property, and their evolution with water content (solid concentration), which could be representative of the mechanical behavior of clay-water pastes as they transcend their liquid limit value, from a consistency lower than that of liquid limit.

The research collaboration is a part of Geotechnical and geological Responses to climate change: Exchanging Approaches and Technologies on a world-wide scale (GREAT) project funded by European Commission.

Former students


Ilham Maimouni

Hello! I am Ilham Maimouni and I’m a second-year PhD student. I discovered my passion for research during my Civil Engineering school years where I realized how much an upstream research work on construction materials could improve the final structures performance!

Currently, I work for Schlumberger who funds my research work about the understanding of stable and unstable flows of yield-stress fluids, namely mud and cement, in an oil well. The motivation behind this comprehension is to ensure a good cementation of the well, which is at the same time a great economic and ecological issue in the oil and gas industry!
My co-supervisors are Julie Goyon (Navier Laboratory) and Etienne Lac (Schlumberger Riboud Product Center).

Besides my passion for research, I easily get fulfilled when I hear the sounds of a new land, taste the flavours of a different cuisine and learn about a different culture!


Rayleigh-Taylor instability for yield stress fluids

One of the problems encountered in the cementation of oil wells is the mixing of two fluids during the upward movement of drilling mud induced by cement. One of the causes of this mixing phenomenon is the Rayleigh-Taylor instability that occurs at the interface between two fluids- in our case yield stress fluids given the rheological behaviour of the cement and the mud- of different densities when the heavy fluid is above the lighter one.

To experimentally study this instability, we superimpose two immiscible fluids of different densities, a yield stress fluid under a heavier Newtonian one, and we observe the evolution of the interface. For a given density difference, the instability occurs below a critical yield stress in the form of fingers or mushrooms of one fluid abruptly spreading through the other one (see Figure 1). Above this critical yield stress, the interface remains undeformed. This set of critical characteristics provides an empirical criterion for the instability.



Figure 1: Illustration of an unstable case: Once there is contact between the two supeposed fluids, fingers of the bottom fluid, here a white direct emulsion of a yield stress equal to 9Pa and a density of around 1 kg/l, spread into the heavier fluid above which is here an iodure sodium solution of a density of 1.8 kg/l.

500 Jules Thiery

Jules Thiery

Current research associate in the field of Microfluidics (i.e. the science and engineering of fluid flow at the microscale) at Harvard University in the experimental soft condensed matter group (Prof. D. A. Weitz Lab.)

My PhD work focused on studying fluid transfers in sub-micron porous media during drying and imbibition (i.e. mechanism of fluid sorption by a porous media) using MRI and Electron microscopy. I was supervised by P. Coussot, D.A Weitz, Stéphane Rodts.


Cracking regimes and confined flow in Nano-porous media during drying

Drying of gels often leads to undesirable and irreversible material alteration such as cracking; therefore understanding the phenomenon appears of high technological concern for industries manufacturing concretes, cosmetics, paints or coatings. Macroscopic observations coupled with airflow simulations on fractured media enabled to depict the kinetics of drying of a porous media that may fracture. High-resolution MRI profiling granting access to the distribution of solvent in the material during desiccation enabled to differentiate and describe (1) two distinct cracking regimes, (2) a new drying regime in nano-porous medium that precludes the appearance of a dry front, and appears to constitute a new method for the observation of liquid nano-films.

Thiery Research picture 1

Figure 1: Aspects of drying gel layers (initial solid fraction ) on an adhesive (upper row) or non-adhesive (lower row) substrate at different stages of the process: (a) and (a’) initial state; (b) and (b’) Regime A (shrinkage associated (b) or not (b’) with open fractures); (c) and (c’) transition between regimes A and B; (d) and (d’) same concentrations, Regime B (new fractures). The white numbers indicate the current time to total drying duration ratio.

Thiery Research picture 2

Figure 2: Distributions of water content in time inside a non-adhesive gel (=20%) during drying. The continuous curves from top to bottom correspond to successive times (every 30 min.) from the test beginning. The symbols are the corresponding positions of the sample free surface as measured (by NMR) from the position of the water pot lying on the sample. The thick line corresponds to the transition between the two regimes (see text).

Perspective: Drying of complex gels mixtures

The development of a new MRI profiling observation technique enables to image and quantify Microgel (PNIPAm) migration. After a productive collaboration with Dimitris Vlassopoulos (FORTH, Crete), I am now currently working on the complex kinematics of drying of depleted silica-PNIPAm gels. I am also planning to investigate Microgel (PNIPAm) migration during imbibition/injection of a polymer mixture in a porous media.

Photo Alizée Lehoux

Alizée Lehoux

I’m a postdoc working on contaminated sediments at the University of Uppsala, Sweden. As a geochemist and a traveler I want to protect our soils and waters from pollution. Did you know that it takes 10 years in average for 1 mm of soil to form? I finished my PhD in November 2016, during which I was supervised by Paméla Faure, Denis Courtier-Murias and Eric Michel. In order to understand the transport of particles such as pollutants in soils, I wa studying the mechanisms of transport and retention of colloids in porous media. My two magical tools were MRI technology and modelling.


Colloid transport in soils

The ability to predict transport and retention of colloidal particles is a major environmental concern as such particles can carry adsorbed pollutants towards the groundwater or be pollutants themselves. The traditional experimental approach of the fate of colloids in soils consists in performing column experiments and studying the breakthrough curves (concentration of particles coming out of the column as a function of volume eluted). We applied this method and combined it with MRI measurements of the column during the transport experiment. Indeed, MRI provides spatial distribution of colloidal particles and water content in time along the sample axis during transport experiment through a porous medium of up to 20 cm height. We injected several pulses of superparamagnetic nanoparticles in columns of porous media of increasing complexity: glass beads, sand, soil aggregates, and undisturbed soil.

In glass beads and sand we were able to observe the transport of negatively charged particles. We measured the dispersion coefficient on different sizes of porous media and various flow rates; we found values for this dispersion coefficient almost ten times lower than expected from literature data based on breakthrough curves analysis. This can be explained by entrance and exit effects, which can induce flow heterogeneities and simulate a larger dispersion at the output.

In another series of experiments we focused on the adsorption. We used positively charged nanoparticles that can get attached to the sand (negatively charged in experimental conditions) very quickly after entering in the column. We were able to follow suspended and adsorbed particle concentrations during the experiment and we found that particles rapidly explore the pores and adsorb as soon as they meet available sites on grains.

Experiments in soil aggregates showed a strong adsorption but also a constant release, which indicates that more complex mechanisms are occurring in soil aggregates, due to heterogeneities of surface charges and complex porosity. We are working on a model to explain this particular behavior.

We also followed water content and particle concentration during a rain simulation under undisturbed soil. Many difficulties arise from this unsaturated complex porous media, so we first worked on understanding water content evolution during rain. Then, we added the particles and we were able to follow their concentration in the entire sample during the experiment. We detected more than half of the particles at the output.



figure alizee

500 Marine Fourmentin

Marine Fourmentin

I am a research engineer in Saint-Gobain Recherche, transersal research center of the Saint-Gobain Group. I mainly work on the chemistry and physico-chemistry of cementitious materials (classical cements and green cements containing supplementary cementitious materials).


Water transfers in hemp concrete

Hemp concrete is a mix of hemp shives, a binder that mainly consists of cement and hydrated lime, and water. Hemp shives have been used as an aggregate for lightweight concretes for more 20 years now. They are considered as a by-product of hemp industry, as hemp is mainly cultivated for its fibrous part. As a vegetal material, its structure is very close to that of wood, but more porous. As a consequence, it is a very hydrophilic material, which is able to absorb up to 3 or 4 times its weight in water. This can be at the origin of competition for water absorption between hemp and cement, which needs water to hydrate. Thanks to Nuclear Magnetic Resonance (NMR), we are able to separate and quantify water impregnated in hemp and water in the binder as they exhibit very different relaxation times, separately as well as in the concrete (see Figure). We are thus able to follow with these measurements the transfers between these two components during setting and also to monitor the adsorption of water in the two phases when the concrete is placed in a humid atmosphere once it was set.

 Fourmentin Research picture 1

 Figure : Distribution of relaxation for the different mixtures.

Setting of lime-cement pastes

In order to study the transfers in hemp concrete, we first need to understand the setting of the lime-cement paste that constitutes the binder, as it is related to water consumption. Therefore, we used NMR, which allows to follow the porosity of the sample, and rheometrical measurements that allow monitoring the evolution of elastic modulus in the sample. The combination of these to techniques added to the classical isothermal calorimetry allows to clearly identify the successive steps of the setting and the contributes to a better understanding of the evolutions of porosity in the cement paste. Comparing these results to a lime-cement paste, we can see that hydrated lime acts as an accelerator on cement during the setting. In this case, the evolution of porous structure is modified, as hydrates precipitate earlier in the pores.

Fourmentin Research picture 2

Figure: Schematic representation of the different stages of structure evolution in time during setting of a cement paste (upper): (a) initial state just after material preparation, (b) colloidal aggregation, (c) hydrate formation around the points of contact, (d) hydrates progressively filling pores; and a lime-cement paste (lower): (a) initial state just after material preparation, (b) colloidal aggregation, (c) and (d) hydrates progressively filling pores. The anhydrous cement grains are represented by larger circles, lime by smaller circles and the hydrates by small crosses

Imbibition in woody materials

As wood, hemp consists roughly in a group of very elongated cells that are parallel to each other. Thanks to NMR, we are able to follow the kinetics of imbibition of hemp particles. We can show that hemp imbibition is very fast during the very first minutes, and then slows down but goes on for three days. This very slow and long imbibition cannot be explained by the classical models known for classical porous media. In order to understand the imbibition of water in hemp shives, we decided to study more widely the imbibition of several wood species, as wood is more studied than hemp. It seems in fact that in wood also there are two types of water penetration in the material. A first one allows water to penetrate very fast in woods using preferential channels (latewood), and even to store at the top of a wood sample. The second one allows water to slowly diffuse through cell walls in the transverse direction to fill other cells (see Figure).

Fourmentin Research picture 3

500 Mathilde Maillard

Mathilde Maillard

Ph.D student (2012-2015) on the coating of yield stress fluids. Currently working at the French nuclear safety authority, in charge of the radioactive waste management unit.


Viscoplastic dip-coating

The process of dip-coating is widely used in industry (civil engineering, food industry, cosmetcis,…). It consists in immersing then withrawing an object from a bath of fluid to coat it, for example to paint or to treat it. I studied the case of a thin plate coated with simple yield stress fluids, that is Carbopol gels or a mix of Carbopol and glycerol. The key question was to understand the formation of the layer that stays on the plate once it is out of the bath. I used various techniques such as macroscopic measurements (weighing, force recording, direct observations), microscopic studies (determination of the flow field with PIV) and numerical simulations based on second-order cone programming. It allows to estimate the impact of macroscopic and rheological parameters – plate velocity, material yield stress and consistency.

Maillard Research picture 1

Figure: Regimes of viscoplastic dip-coating, as deduced from experimental data, as a function of dimensionless numbers.

Viscoplastic blade-coating

The process of blade-coating is a controlled way to spread fluids on a substrate. I studied experimentally this process in 2D with Carbopol gels and a plate that can be inclined. The major question concerned the dynamic of the roll of fluid that forms in front of the plate, like when a mortar is spread with a trowel. I performed macroscopic measurements as well as some numerical studies to understand the involved flow, and the impact of the plate velocity and the fluid characteristics.

Maillard Research picture 2

500 Mamadou Diaga Seck

Mamadou Diaga Seck

I am actually working for the ETEX group Concrete Tiles Competency Centre as Concrete Roofing and Process Engineering Specialist. My work consist of carrying out R&D works into optimizing our concrete formulations and into improving the tiles manufacturing process within our factories all over the world.


Mechanism of plaster drying

Most building materials, such as cement, plaster, concrete, etc, are made by inducing a chemical reaction between a powder and water. During this process, for workability purposes, a surplus of water is added which, at the end of the chemical reaction, must be evacuated by drying. We showed that the drying rate of plaster pastes is significantly lower than that expected for ideal porous systems. This slowing effect is enhanced by the air flow velocity and the initial solid/water ratio. Further investigations with the help of additional non-destructive techniques (MRI, ESEM, Microtomography), for measuring the drying rate and local characteristics (water content, porosity), prove that this effect is due to the crystallization of gypsum ions below the sample free surface, which leads to a dewetting of the upper layers, i.e. the development of a dry region thicker than in the absence of gypsum (Figure 1).

Seck Research picture 1

Figure 1: Total water saturation as a function of time during drying tests inside a bead packing (squares) and a plaster (w/p=0.8, cylinder B) (crosses), air flow velocity: 0.27 m/s., as deduced from NMR measurements of the water distribution along the sample axis. [Paper 141]

Crystallization in polygonal capillaries

The drying rate of a polygonal capillary tube saturated with an ionic solution is significantly different from that expected for one saturated with pure water. Using a microscopic visualization technics to follow evolution of corner films and ions crystallization, we show that this difference finds its origin in the crystallization patterns of salts, which may induce some dewetting of the solid walls, i.e. a displacement of the line of contact towards the interior of the channel (see Figure 2). Our understanding of the process makes it possible to establish a simple diffusion-based model predicting the drying rate in capillary tube, in excellent agreement with experimental observations.

Seck Research picture 2

Figure 2: View from upper of the air-liquid interface during drying of the gypsum solution in a small channel. One can see the dewetting effect along the wall between the channel entrance (on the left) and the first visible gypsum crystals.


Rafael Mendes

I am currently a 3rd year Ph.D. student working on the rheological behavior of waxy crude oils and the flow restart of pipelines with gelled oil. I am a mechanical engineer (UFSC – Brazil, 2004) and with a Master degree in thermal sciences (UFSC – Brazil, 2007). I have been working since 2004 as petroleum engineer at the Flow Assurance dept. of Petrobras research center.

Fields of research : Modeling and simulation of multiphase flow in petroleum production systems, flow assurance analysis and rheology of complex fluids.


500 Emmanuel Keita

Emmanuel Keita

My research focus on drying in porous media. I graduated from Ecole Polytechnique in 2011. Then I join Prof. Philippe Coussot Group at Université Paris-Est to work on drying of porous media. I investigate the mechanism of liquid transfer in model building materials as water evaporates. I have been visiting Prof. David Weitz Group at Harvard University for 1.5 years to synthesize particles and look at particles aggregates with confocal microscopy. I completed my Ph.D. in Fall 2014 and I am now researcher at IFSTTAR (Laboratory FM2D). My co-supervisor was Paméla Faure.


Particles transport in drying porous media

Due to the migration of elements they induce imbibition-drying cycles are known to play a major role in the colloid-facilitated transport in many industrials process, for instance for pollutants migration in soils or pores clogging in building materials. We study the drying of a colloidal suspension in a porous media. The critical physical phenomenon at work here is the displacement and redistribution of colloidal particles induced by evaporation of the liquid phase from the porous medium. Using a new MRI technique to measure at the same time the distributions of water and particles we observed particles rising towards the free surface, as water remains homogenously distributed. The particles aggregation area is very large compared to their volume fraction in the pore volume; we succeeded to model the particle migration and the drying kinetics of the system. [Paper 123]


Keita Research figure 1


Figure 1 – Drying through the upper free surface of a cylindrical (height = 4 cm) porous medium made of glass beads (62% volume fraction) initially filled with a colloidal suspension at 5% volume fraction in water. MRI measurements were carried out every 2 h over 4 days giving us access to the distribution of colloids in the liquid phase (top), and the distribution of water (bottom) as a function of sample height. The profiles displayed are separated by 10 h. Red thick lines represent the initial state. [Paper 123]

Drying in model porous media

A capillary tube of rectangular cross section maintains water layers in its 4 corners and reproduce well the drying regimes of a porous medium. Using a simple geometry, we observe the water distribution and measure the shape of the air/water interface with good resolution in imaging and in time. Here we show that a small variation in the shape of the cross section modifies drastically the invasion of air due to equilibrium of capillary forces. Moreover not only the corners but a large part of the cross section remain wet in particular at the entrance of the tube allowing a high drying rate. Interpreting only the water mass loss as a function of time may lead to wrong conclusions considering basics drying regimes.

Keita Research figure 2

Figure 2 – Glass capillary of slightly modified rectangular cross section initially filled with water; superposition of images after 1, 17, 34, 51 and 68 min of drying. Scale bar is 1 mm.

Photo Claire

Claire Marliere

Research Engineer in petrophysics and chemical Enhanced Oil Recovery- study of complex fluid flow in porous media, from 2014 – Current research : Study of two-phases flows of complex fluids applied to enhanced oil recovery

Post doc with Total at the Laboratoire Colloïdes et Matériaux Divisés on encapsulation of lubricant additives using multiple emulsions, in 2013

PhD thesis with Saint-Gobain on Water transfers in porous medium in presence of water retaining polymers: application to mortar from 2010 to 2013,


  • Water retention standard tests, rheology, visualization of the phenomenon by MRI, cellulose ethers characterization (microscopy, DLS), study of imbibition, drainage, flow under pressure, filtration in porous media.
  • Demonstration of the presence of polydisperse cellulose ethers aggregates;
  • Observation of a jamming phenomenon of a cellulose ethers solution through a porous medium (cf. Figures 1 et 2)
  • Statistical modeling of the phenomenon (probability of a pore jamming);
  • Establishment of a more relevant alternative filtration test than the retention test;
  • Study of physico-chemistry criteria of a good water retaining polymer.

Figure 1

Figure 1: Scheme of the filtration experiment

Figure 2

Figure 2 : MRI profiles at different times showing the saturation of a column of beads as a function of the column height in the presence of a cellulose ether solution through a stack of glass beads of 45-90 microns

Thibaud Chevalier

  • Since September 2016 : Research Engineer in charge of the NMR lab at IFP Energies Nouvelles, Rueil Malmaison, France – Current research : Study of complex fluids and porous media characterisation applied to enhanced oil recovery and catalysis.
  • 2015 – 2016 : Post doc on foam imbibition in fractured media at Department of Applied Physics, Aalto University, Finland and partly at ENS Lyon for experiments.
  • 2013 – 2015 : Post doc on complex flows in disordered media at Laboratoire FAST, Orsay, France.
  • 2010 – 2013 : Ph.D. Student on the flow of yield stress fluids through porous medium at Laboratoire NAVIER, France.


My research focuses on complex fluid flows in disordered media. I first developed a methodology to follow slow flows in porous media under magnetic resonance imager (MRI). Then I realized the experimental study of macroscopic flow laws  in porous media in a two phase flow system (2) and also during the injection of viscoplastic fluids. In the latter case, I showed, using a numerical approach, that in a 2D disordered medium, the quadratic flowing regime can be explained by the statistical properties of the non-flowing zones. Modeling this regime allowed us to capture the dynamics of opening of flowing paths by analogy with an avalanche dynamic. I studied in more details these dynamics by following the propagation of a reactive front in an adverse flow in disordered media (3). Finally, I developed different local approaches on these systems such that the experimental and theoretical description of a boundary layer at the pore scale in the flow of viscoplastic fluids (1). During my last-post-doc, I focused on the foam invasion in model fracture both numerically and experimentally (4).

Figure recherches Thibaud Chevalier


Jalila Boujlel

  • Since October 2014: Research Engineer in Complex Fluid Flows at IFP Energies Nouvelles, Rueil Malmaison, France.
  • 2013-2014:  post doc on chaotic mixing of complex fluids at the Joint Unit CNRS/Saint-Gobain, Aubervilliers, France
  • 2008-2012: Ph.D. Student on the flow of complex fluids along solid surfaces at laboratoire Navier, Champs/Marne, France



Two phase flow in porous media for Enhanced Oil Recovery application

  • Study of the microscopic mechanisms of oil mobilization by a surfactant solution in 2D micromodel.

Figure 1 : Oil blobs trapped in 2D porous system: grains in white, pores in black and oil in orange

  • Experimental evaluation and development of Enhanced Oil Recovery Methodologies.

Chaotic mixing of complex fluids

Study of the efficiency of chaotic advection for the mixing of complex fluid by using a rod-stirring protocol with a rotating vessel:

  • Quantification of the chaotic mixing rate of yield stress fluids: the mixing rate is found to be proportional to the vicoplastic boundary layer generated around the rods (Fig.1.d), which depends on the rod radius and the Bingham number. A quantitative model was developed.
  • Determination of the effect of the rheological properties of the material, the mixing geometry and the wall speed on the mixing characteristics

Figure 2. Mixing of a transparent yield stress fluid with a blob of dye by using two pairs of cylindrical stirring rods, counter-rotating. (a, b, c) Pictures of mixing showing the dependence of the thickness of the dye filament formed behind the rod (red arrow), on the rod radius (R (a) < R (b) < R(c)). (d) Schematic of the morphology of the flow around a cylinder moving at constant speed through a yield stress fluid, highlighting the boundary layer, where the shear is very intense and localized near the cylinder.

Behavior of yield stress fluids along solid surfaces  : wetting, adhesion and boundary layer

Study of the flow generated along a thin solid plate in vertical movement through a yield stress fluid. Emphasis was placed on the study of the following phenomena:

  • Boundary layer: when moving a thin solid plate through a yield stress fluid, a uniform thickness of liquid layer is developed along the plate, while the rest of the fluid remains in its solid state.
  • Deposit of yield stress fluid on solid surfaces : a solid plate pulled out of a yield stress fluid bath leads with it a millimetric layer of fluid. Its  thickness depends on the thickness set in motion in the bath.
  • Surface tension of yield stress fluid : measured using a method inspired by the Wilhelmy plate technique

Figure 3.  Apparent surface tension scaled by the surface tension of the interstitial liquid (water) as a function of the capillary number for yield stress fluid times a gravity factor, for different blade thicknesses and for different Carbopol gels. The continuous line corresponds to an empirical model. The inset shows the picture of the meniscus formed along a glass blade when it is put in contact with a surface of yield stress fluid initially flat (Soft Matter, 2013,9, 5898-5908).

Photo Jorge Avendaño

Jorge Avendano

Chemical engineer graduated at ULA, Mérida (Venezuela). Lecturer from 2003 at ULA and researcher in FIRP Laboratory with Prof. Jean Louis Salager from 2004. From 2008 to 2012 I did my Ph.D. Theses at IFP Energies Nouvelles under the direction of Prof. Coussot and Dr. Benjamin Herzhaft studying the effect of viscoelasticity of displacing fluids for enhanced oil recovery. Back in Venezuela, as Research Associate Professor from 2014 I continued to study physicochemical interactions of polymers in chemical and polymer flooding for EOR process. From 2015 I´ve been in a Post-Doc position at LMMP in PUC-Rio de Janeiro with Prof. Marcio Carvalho studying the influence of dispersions and polymers in residual oil saturation for EOR process using porous media microfluidic devices.




Hello, I am Imane GUETNI. I am a Civil engineer with a Master degree on materials science. I was supervised by Prof. Philippe COUSSOT during my master project on the blade coating using yield stress fluids.
Currently, I am a Ph.D candidate at IFP Energies nouvelles and I am studying the injectivity of complex fluids in low permeability porous media applied to enhanced oil recovery.

photo Trium


Hi, I am Léo, student at Ecole Nationale des Ponts et Chausées in the department of Mechanial Engineering and Material Science. I worked as a Research Intern, for two months (Apr. 2015-June 2016), at Laboratoire Navier, studying wood imbibition and deformation in Apr. 2015, under the supervision of Philippe Coussot. Formerly intern at Saint-Gobain and currently at Millidrop, I am also passoniated by robotics, 3D printing and biomechanics, and I am planning to pursue my studies and works in one of these fields.


Hafsa Rahoui

  • Christophe Ancey (Ph.D., 1997) Professor, Ecole Polytechnique Fédérale de Lausanne
  • Philippe Alexandre (Ph.D., 1997) Engineer, Switzerland
  • Jean-Christophe Baudez (Ph.D., 2001) Group Leader “Pollution Control and Impact Assessment at Luxembourg Institute of Science and Technology (LIST)
  • Philippe Poullain (Ph.D., 2004) Associate professor, Univ. Nantes
  • Sébastien Jarny (Ph.D., 2004) Associate professor, Univ. Poitiers
  • Hervé Tabuteau (Ph.D., 2005) CNRS researcher, Univ. Rennes
  • T.L.H Nguyen (Ph.D., 2007) Engineer
  • Alexandre Ragouilliaux (Ph.D., 2007) Senior Expert, AREVA
  • Pierre Rognon (Ph.D., 2007) Senior lecturer, Univ. Sydney, Australia
  • Fabien Mahaut (Ph.D., 2009) Post-doc, Laboratoire FAST, Univ. Orsay
  • Elsa Bourguignon (Ph.D., 2009) Conservation scientist at Laboratoire de Recherche des Monuments Historiques
  • Brooks D. Rabideau (Post-doc, 2007-2010) Assistant Professor, Univ. of South Alabama, USA
  • Xavier Clain (Ph.D., 2010) Post-doc, Univ. Paris-Est
  • Jorge Avendado (Ph.D., 2012) Post-doc, Pontifical Catholiv Univ. Rio, Brazil
  • Jalila Boujlel (Ph.D., 2012) Researcher, IFPEN, Rueil-Malmaison
  • Claire Marlière (Ph.D., 2013) Researcher, IFPEN, Rueil-Malmaison
  • Thibaud Chevalier (Ph.D., 2013) Researcher, IFPEN, Rueil-Malmaison
  • Emmanuel Keita (Ph.D., 2014) Researcher, IFSTTAR, Marne la Vallée
  • Rafael Mendes (Ph.D., 2015) Petroleum Engineer, Petrobras, Rio
  • Mathilde Maillard (Ph.D., 2015) Manager, Authority of Nuclear Safety, France
  • Marine Fourmentin (Ph.D., 2015) Research engineer, Sigma Béton, France
  • Mamadou Diaga Seck (Ph.D., 2015) Product Engineer, Morgan Advanced Materials, Corby (UK)
  • Jules Thiery (Ph.D., 2016) Research Associate, Harvard Univ.
  • Alizée Lehoux (Ph.D., 2016) Post-doc, Uppsala, Sweden