Statistical physics deals with systems with large number of degrees of freedom, such as disordered systems, Coulombic systems, and fluid phase transitions. Many domains are extending, in particular the granular media and the complex networks (e.g., the propagation of epidemics).

The range of actitivies for the group has been extended to new applications of statistical physics over the last few years, i.e. applications to problems from other disciplines.

The employed methods are of a large diversity : sooner or later, each technique of statistical physics can be used to solve a problem for which it was not designed initially.

### Complex networks

A. Barrat and A. Vespignani are interested in the dynamical phenomena occuring on networks, as well as their links with their large-scale structure and their topology.
These works exploit some methods and models typical of statistical phsycis and have an interdisciplinary range, from the structure of the Internet and airline networks to the propagation of epidemics.

### Car traffic

Packs are queues of vehicles separated by a very short amount of time - some time less than the average reaction time of the drivers. This potentially dangerous situation creates compression waves. C. Appert-Rolland has proposed a model to study collective effects in these structures and to identify the most dangerous configurations. She has also developed models of cellular automatons to link the individual behaviour of the rivers and some macroscopic phenomena, with a specific interest in the influence of the reaction time of the drivers.

### Granluar media

A. Barrat studies the granular media, which constitute
a good example of intrisically out-of-equilibrium systems : in order to kepp them in a dynamical state, one has to inject energy endlessly. For diluted granular media, he has abalysed the behaviour of granular gases in various contexts with the help of numerical simulations and analytical tools. His most recent researches are

(i) the mathematical properties of the Boltzmann equation, which describes the evolution of granular gases under some simplifying assumptions, and which provides the speed distribution,

(ii) the fluctuations of global observables such as the total energy of the system or the power that has to be injected to keep a stationnary situation.

### Long range interactions, theory and simulation of liquids

J.-M. Caillol develops methods in statistical physics, statistical field theory and theory of liquids. J.-J. Weis and D. Levesque are interested in ferrofluids and in the
the adsorption in porous disordered solids. These three scientists are specialists in molecular dynamics and in Monte Carlo simulations.

B. Jancovici works on systems of charged particles, in particular their properties in bidimensional models that can be solved exactly.

In order to interpret experiments on liquids at negative pressure such as water and helium, C. Appert-Rolland studies the focalisation of a spherical wave in the case of equations of states with different degrees of non-linearity, leading to the creation of shock waves.

M. Mazars studies the behaviour of almost bidimensional systems (surface properties of solids, Langmuir monolayers, Langmuir-Blodgett films, smectic films, cell membranes...) through numerical simulations. He studies also the model of the free joined string, which provides an elementary model for linear macromolecules, and for which
he has obtained an excellent agreement with simulations of molecular dynamics.

### Out-of-equilibrium systems

F. Cornu studies the Lorentz model with overlap in presence of a uniform magnetic field, which describes the non-colliding movement of a particle in a bidimensional medium consisting of immobile hard disks scattered randomly.

To study the chaotic properties of out-of-equilibrium systems, C. Appert-Rolland has extended an approach based on the thermodynamical formalism to stochastic systems with a continous time, in order to apply it to more realistic systems.

H. Hilhorst has a long-standing interest for phase transitions and for very diverse applications of random processes. Over the last few years, he has studied problems of reaction-diffusion, and more recently he has made progress in the random geometry of bidimensional cellular systems, which are a model for a large range of natural systems.