•  Thesis on HAL.
•  Code developed for the project (mainly C/C++ [simulation] and  Python [analysis]).

•  Code developed for the project (mainly C/C++ [simulation] and  Python [analysis]).
•  Wiki of the project with details on our models and methods, theoretical developments, various plots and movies.

We have worked on a numerical model system of polydisperse active Brownian disks with purely repulsive interparticle harmonic potential, inspired by Fily et al. [Soft Matter, 2014, 10, 2132-2140]. Despite its simplicity, this model displays interesting collective behaviours: motility-induced phase separation (MIPS) [Cates and Tailleur, Annu. Rev. Condens. Matter Phys., 2015, 6 (10), 219-244] and long-ranged displacement correlations, even on time scales greater than the rotational diffusion time. We showed that the transition from the fluid to the phase separated regime, which characterises MIPS, is accompanied by increased dynamical heterogeneity. Moreover, we provided evidence of an \(r^{-2}\) decay of shear strain correlations on time scales short compared to the rotational diffusion time, both in the dense region of our phase separated system at high activity and in the truly glassy phase [Keta and Rottler, EPL, 2019, 125, 58004]. We attributed the origin of the \(r^{-2}\) decay to elastic interactions that lead to similar behaviour in passive glass formers.

• We have come to think that Pearson correlation coefficients — which are simple measures of the linear relationaship between two variables — may not be fitted to characterise the correlation between displacements in 2D when these are not colinear [Lange and Grubmüller, Proteins, 2006, 62, 1053–1061]. Finding other measures of the displacement correlation which would address this concern is an interesting challenge.
• Since we have shown that elastic interactions took place in our model on short enough time scales, an other interesting challenge would be to characterise the nature of and correlations between plastic events in active matter systems as it has been done for supercooled liquids [Chattoraj and Lemaître, Phys. Rev. Lett., 2013, 111, 066001, Puosi et al., Phys. Rev. E, 2014, 89, 042302].

•  Code developed for the project (mainly  Python).
•  Wiki of the project with details on our model and methods as well as various plots.

We adapted models and methods which were initially used to study shear jamming of frictionless soft 2D disks in order to consider 3D ellipsoids, and in particular spheroids — ellispoids of revolution.

• It has been shown that the nematic order parameter \(S_2\) is a non-monotonic function of the packing fraction in packings of non-spherical particles undergoing simple shear (see Fig. 1 in [Marschall et al., Phys. Rev. Lett., 122 (2019) 188002]) which maximum is independent of the shear rate. Qualitatively, the existence of a maximum \(S_2\) seems to be a consequence of the competition between elastic collisions — which makes the rotation velocity of particles a nonuniform function of the angle — and excluded volume effects — particles alignment is not compatible with dense packings. It would be interesting to develop a mean-field theory which would quantitatively predict this non-monotonic behaviour of the nematic order parameter.

•  Code developed for the project (mainly C).
•  Wiki of the project with a summary of some recent findings on shear jamming and details on our model and methods.

We aimed at using data obtained from drop impacts on single micrometric defects to predict the phenomenology of impacts on more complex networks.