Numerical Modeling of Sequential Segmentation for Enhancement of Mixing Inside Microchannels

Abstract

Microfluidics appeared recently as a new tool that facilitates many applications including chemical synthesis, electronics cooling and biological assays. However, flow on the microscale is characterized by laminar flow which renders mixing a challenge and reduces the effectiveness of mixing in many applications. Therefore, many techniques have been reported to enhance mixing inside microchannels. This thesis demonstrates the different mixing techniques inside microchannels and investigates sequential segmentation as an active mixing technique. In sequential segmentation, the solute and the solvent are divided into segments in the axial direction. Due to the parabolic velocity profile exhibited in laminar flow, Taylor-Aris dispersion can improve mixing by several orders of magnitude as compared with pure molecular diffusion. The present work comprises a comprehensive study to optimize this technique. A numerical model was built using COMSOL Multiphysics to calculate the concentration distribution by diffusion and advection in the entire microchannel. The different investigated parameters that were optimized are frequency, flow velocity, duty cycle (DC), aspect ratio, cross-sectional shape, and effect of inlet configuration of inlet branches. It has been found that changing the channel aspect ratio has the most significant effect on mixing efficiency. The analysis of aspect ratio (H:W) showed a moderate increase in mixing index for aspect ratio of 4:1. However, an aspect ratio of 1:4 is extremely inefficient. The best mixing efficiency achieved is that of sequential segmentation coupled with hydrodynamic focusing which is 99.7%. On the contrary, changing the shape of inlet branches to T-shape or arrow-head shape decreased the mixing index. Increasing segmentation frequency was found to increase the mixing efficiency up to a frequency of 150 Hz beyond which mixing efficiency decreased due to non-complete segmentation of both streams at higher frequencies. The duty cycle effect is mainly important for specific microfluidic applications that require different mixing fractions of the solutions. It is shown that by increasing the duty cycle of the solvent, mixing slightly increases, whereas decreasing it leads to a reduction in mixing. The effect of different cross-sectional shapes also has a small effect on the mixing index.