Swarm behavior of self-propelled rods and swimming flagella

Yang, Y.; Marceau, V.; Gompper, G.

doi:10.1103/PhysRevE.82.031904

Journal Article

PreJuSER-11236

Swarm behavior of self-propelled rods and swimming flagella

Yang, Y.FZJ* ; Marceau, V.FZJ* ; Gompper, G.FZJ*

2010
APS College Park, Md.

Physical review / E 82(3), 031904 (2010) [10.1103/PhysRevE.82.031904]

This record in other databases:

Please use a persistent id in citations: http://hdl.handle.net/2128/9311 doi:10.1103/PhysRevE.82.031904

Abstract: Systems of self-propelled particles are known for their tendency to aggregate and to display swarm behavior. We investigate two model systems: self-propelled rods interacting via volume exclusion and sinusoidally beating flagella embedded in a fluid with hydrodynamic interactions. In the flagella system, beating frequencies are Gaussian distributed with a nonzero average. These systems are studied by Brownian-dynamics simulations and by mesoscale hydrodynamics simulations, respectively. The clustering behavior is analyzed as the particle density and the environmental or internal noise are varied. By distinguishing three types of cluster-size probability density functions, we obtain a phase diagram of different swarm behaviors. The properties of clusters such as their configuration, lifetime, and average size are analyzed. We find that the swarm behavior of the two systems, characterized by several effective power laws, is very similar. However, a more careful analysis reveals several differences. Clusters of self-propelled rods form due to partially blocked forward motion and are therefore typically wedge shaped. At higher rod density and low noise, a giant mobile cluster appears, in which most rods are mostly oriented toward the center. In contrast, flagella become hydrodynamically synchronized and attract each other; their clusters are therefore more elongated. Furthermore, the lifetime of flagella clusters decays more quickly with cluster size than of rod clusters.

Keyword(s): J

Classification:

Note: We thank Jens Elgeti and Roland Winkler for stimulating discussions. Y.Y. acknowledges support by the International Helmholtz Research School on Biophysics and Soft Matter (IHRS BioSoft). V. M. is grateful to the RISE program of the DAAD (Germany) and to NSERC (Canada) for financial support. This work was supported in part by the VW foundation through the program "Computational Soft Matter and Biophysics."

Contributing Institute(s):

Research Program(s):