Flexible Interconnect Design Using a Mechanically-Focused, Multi-Objective Genetic Algorithm

This paper presents a multi-objective optimization process using a non-dominated sorting genetic algorithm-II that aims to maximize the vertical displacement of MEMS-type mechanically flexible interconnects (MFIs) while simultaneously preventing these interconnects from experiencing plastic deformation. Given an initial MFI geometric body, this process seeks to optimize the photomask interconnect design so that no modification is required for the existing fabrication process flow. After the design optimization, a reflowed polymer-dome microfabrication process is employed to fabricate the NiW MFIs to provide a more complete analysis of this paper. Simulation data shows that the optimized design can achieve up to 29.7% greater vertical displacement before experiencing plastic deformation relative to the initial design. Experimental data shows an average of 34.2% less vertical plastic deformation relative to the initial design when the MFIs are indented to a depth of $45~\mu \text{m}$ . In both simulation and experimental scenarios, the optimized MFIs demonstrate less stress/plastic deformation relative to the initial MFIs for the same given indented depth.