Partenaires

Logo CNRS Logo UPMC


 
 
 

Rechercher

Sur ce site

Sur le Web du CNRS

 
 
 



Accueil > Productions scientifiques > Séminaires à PHENIX > 2021 > Seminar : Adrien Izzet (New York University) - 11/06/2021 at 2pm (!! unusual time !!)

Seminar : Adrien Izzet (New York University) - 11/06/2021 at 2pm (!! unusual time !!)

par Pierre Illien - 11 mai

Seminar : Adrien Izzet (New York University)

Dynamics and contact interactions of emulsion droplets

Friday 11 June 2021 at 2pm (online)

Abstract

The mechanical response and the dynamics of heterogeneous media, such as sand or emulsions, depending on the nature of the physical interactions between their elementary bricks.

While dense granular media exhibit solid over-damped contacts dominated by friction, emulsions droplets are characterized by liquid-liquid interfaces, which properties can be tuned by chemistry. We designed an experimental system that consists of a bio-inspired emulsion in which the droplets are covered with cadherins to mimic cell-cell adhesion. We quantify the binding energy and provide evidence of cadherin crystallization in adhesion patches. We show that cadherin-induced cell-cell adhesion is a mechano-sensitive process, where confining pressure activates binding.

Fig. 1 : a) Live, cadherin-expressing cells form a confluent packing through cadherins (Priya et al. 2017). b) Biomimetic emulsion droplets functionalized with fluorescent E-cadherin extracellular domain forms a confluent tissue-like structure. Scale bar : 20μm.

In a dilute limit, emulsions droplets can be made to self-propel via a Marangoni flow at their interface. We show that this propulsion mechanism can be tuned to independently control the swimming speed and the persistence time of these particles. The behaviour of small droplets ( 30µm diameter) can be described as Active Brownian Particles (ABP). Yet, this model does not apply to larger droplets because their trajectories exhibit self-avoidance with their own trail. In that case, we introduce a non-Markovian theoretical framework to numerically investigate the memory effects on the mean square displacement (MSD). We use the experimental MSD to provide a measure of the noise in the system and fit both the Active Brownian Particle and non-Markovian model. We use Machine Learning techniques to classify simulated and experimental trajectories and determine to which extend droplet trajectories can be considered as ABP or as non-Markovian random walkers. Our findings show the limit of the ABP model to characterize the behaviour of these active particles and beyond which the non-Markovian model is required. The fitting of the theory to the experimental data provides insights into how the physical chemistry of the system modifies the different contributions in the model, namely the negative chemotaxis causing the memory effects and the spatio-temporal noise.

Fig. 2 : negative chemotaxis observed in a droplet meeting a previous one’s trail. a) experiment (colours represent time, from orange to blue : 2sec.), b) simulation of the oil concentration field (represented in colours).