Learning general pair interactions between self-propelled particles

Area of Science:

• Active matter physics
• Soft condensed matter
• Statistical mechanics

Background:

• Synthetic active matter systems, including active colloids, exhibit complex interactions (hydrodynamic, chemical, electrostatic) that are difficult to model from first principles.
• Understanding these interactions is crucial for predicting and controlling the emergent behavior of these systems.

Purpose of the Study:

• To develop a method for learning general pair interactions, including transverse forces and torques, between self-propelled Janus particles from experimental trajectory data.
• To analyze the nature of these interactions and their influence on system dynamics.

Main Methods:

• Utilized stochastic force inference to learn inter-particle interactions from experimental trajectories of Janus particles.
• Employed numerical simulations with learned interactions to reproduce experimental observables and test system behavior at different densities.

Main Results:

• Inferred interactions show a predominantly repulsive and isotropic radial component.
• A complex angular dependence in the angular interaction was identified as the key factor controlling system behavior.
• Transverse interactions were found to be negligible.
• Symmetry analysis indicated that the interactions have a hydrodynamic component, ruling out electrostatics as the sole origin.

Conclusions:

• Stochastic force inference is a powerful tool for uncovering complex inter-particle interactions in active matter.
• The learned interactions accurately predict system behavior and can be extrapolated to different conditions.
• The findings highlight the significant role of anisotropic angular interactions and hydrodynamic forces in driving the collective dynamics of Janus particle systems.