Single and Multi-Objective Optimization of a Three-Dimensional Unbalanced Split-and-Recombine Micromixer

Area of Science:

• Fluid Dynamics
• Microfluidics
• Computational Engineering

Background:

• Micromixers are crucial for efficient fluid manipulation in microfluidic devices.
• Enhancing mixing in microchannels at low Reynolds numbers remains a challenge.
• Asymmetrical split-and-recombine (SPAR) mechanisms offer potential for improved mixing.

Purpose of the Study:

• To optimize the 3D geometry of an asymmetrical SPAR micromixer for enhanced fluid mixing.
• To investigate the trade-offs between mixing performance and pressure drop.
• To achieve superior mixing efficiency at a Reynolds number of 20.

Main Methods:

• Utilized particle swarm optimization and a genetic algorithm for single and multi-objective optimization.
• Employed surrogate modeling with Kriging metamodels based on computational fluid dynamics (CFD) simulations.
• Solved convection-diffusion and Navier-Stokes equations for mixing and flow analysis.
• Used Latin hypercube sampling for efficient design space exploration.

Main Results:

• Single-objective optimization achieved a 58.9% increase in mixing effectiveness compared to the baseline design.
• Multi-objective optimization yielded Pareto-optimal solutions with up to 48.5% higher mixing index.
• Multi-objective optimization also demonstrated a significant reduction in pressure drop by up to 55.0%.

Conclusions:

• The optimized asymmetrical SPAR micromixer geometry substantially improves fluid mixing efficiency.
• The study successfully balanced mixing enhancement with pressure drop reduction through multi-objective optimization.
• Computational optimization techniques are effective for designing high-performance microfluidic devices.