|M.Sc Student||Harduf Yuval|
|Subject||Analysis of Stability Transitions in a Superparamagnetic|
|Department||Department of Mechanical Engineering||Supervisor||Professor Yizhar Or|
|Full Thesis text|
Robotic microswimmers are a source of growing interest in the fields of physics and biomedical robotics. The famous work of Dreyfus et al (2005) introduced a robotic microswimmer composed of a chain of superparamagnetic beads and actuated by a planar oscillating magnetic field. Further experiments and numerical simulations of the swimmer's model revealed that for large enough oscillation amplitude of the magnetic field's direction, the swimmer's mean orientation and net swimming direction both flip from the mean direction of the magnetic field to a direction perpendicular to it.
In the current work, this phenomenon is analyzed theoretically by studying two models of microswimmers: a multilink model with a spherical head attached to a chain of paramagnetic links, and the simplest possible microswimmer model: two slender rigid links connected by an elastic joint, while one link is superparamagnetic. The dynamic equations of motion of both models are formulated explicitly. The multilink model is studied numerically, and the two-link model is studied using numerical methods and various asymptotical analyses. It is found that the changes in the swimmer's swimming direction are associated with stability transitions between two solutions, which are induced by nonlinear parametric excitations. Furthermore, the analysis leads to analytical conditions for stability transitions and explicit expressions of the swimmer's mean speed, both confirmed via numerical analysis of the model. It is found that there exist intermediate parameter regions of dynamic bi-stability where the solutions of swimming about the aligned and perpendicular directions are both stable under different initial conditions.
Finally, preliminary experimental results of our research collaboration with Li Zhang from Chinese U. Hong Kong are presented, corroborating our theoretical predictions.