Constructal design of passive micromixers with multiple obstacles via computational fluid dynamics

Passive micromixers have applications mainly in the chemical, pharmaceutical, and materials industries. Two or more fluids mix while flowing through microchannels in these devices. Due to the small dimensions and low flow rates, the flow is essentially laminar, and mixing takes place mainly by mass diffusion. One way to increase the mixing rate in micromixers is the addition of obstacles that increase the advective effects. This work aimed to introduce high-performance designs of passive micromixers with multiple obstacles. These designs were obtained by combining the Constructal Design method with the Response Surface Optimization method and Computational Fluid Dynamics (CFD). The micromixers were Y-shaped tubes with grooves and circular obstacles in cells that repeated along the device. From the first design inspired by a high-performance design from the literature, the evolutionary design of the system was achieved by increasing the number of obstacles and finding the best configuration for each evolution level (number of obstacles per cell, from three to seven). The effects on mixing percentage, pressure difference, and mixing energy cost (MEC) of obstacles’ vertical and horizontal distances were investigated with CFD simulations. Increasing the number of obstacles made it possible to increase the mixture percentage of the micromixer. At the same time, the total pressure drop rises faster than the mixing percentage. However, analyzing the pressure locally, it was shown that the lower the number of obstacles, the greater the local pressure drop, which could cause flow obstructions. The vertical distance of the obstacles had a more significant impact on the mixing than their horizontal distance. Both vertical and horizontal distances had a substantial effect on the pressure drop. As the number of obstacles increased, the effect of the horizontal distance became weaker as its variation was limited. The three-obstacles design presented a MEC equal to 2.47 and a mixing percentage equal to 67.12% mixing index. The latest design evolution (i.e., seven obstacles) achieved the best mixing percentage, 70.30%, with MEC equal to 2.97. By modifying the degrees of freedom, it was possible to understand and propose a path to the evolutionary design of the system to increase its performance while still using simple designs.