# Séminaire

##
Lundi 11 Décembre 2023 à **11h00**.

## Collective effects far from equilibrium: from active flows to minimal models

**Alexis PONCET**

( Laboratoire de Physique de l ENS de Lyon)

**Salle des séminaires LIPPMANN**

Invité(e) par

**Nicolas Bain**

présentera en 1 heure :

''

The theoretical description of the structure and dynamics of matter far from thermal equilibrium is at the heart of modern statistical physics. In this seminar, I will describe two complementary projects in this direction. The first one starts from the experimental observation of active flows in a hydraulic network and aims at building a theoretical description of them in terms of spin-ice physics. The second project goes in the reverse direction: starting from minimal models in 1D, how far can one go analytically in their characterization to hope to describe realistic systems in terms of few parameters?

Biological and synthetic active fluids defy the usual linear laws of confined hydraulics: active flows in pipes are indeed bistable. Nevertheless, the characterization of their emergent patterns in channel networks lacks both quantitative experiments and theory. In the first part of my talk, I will show that the experimental flows of active colloids in trivalent networks realize dynamical spin ices. The resulting streamline patterns split into two distinct classes of self-similar loops that I characterized both numerically using a double spin model, and theoretically with a series of mappings on loop O(n) models [**1**].

In the second part of my talk I will focus on paradigmatic models of out-of-equilibrium statistical mechanics: single-file systems (1D systems of particles that cannot pass each other) and in particular their versions on lattice. Starting from a simple example where two particles are driven apart from one another and may "unbind", I will introduce a framework for the description of density fields. This framework can be extended to all orders of displacement-density correlations, providing a complete characterization of the one-tag process in single-file lattices and paving the way for a description of such systems in and out of equilibrium in terms of simple transport coefficients [**2**].

[**1**] Jorge, Chardac, Poncet & Bartolo (2023), to appear in Nature Physics, arXiv:2305.06078

[**2**] Grabsch, Poncet, Rizkallah, Illien & Bénichou (2022), Science Advances 8 (12)

''