Novel 3-D T-Shaped Passive Micromixer Design with Helicoidal Flows

Abstract

Laminar fluid flow and advection-dominant transport produce ineffective mixing conditions in micromixers. In these systems, a desirable fluid mixing over a short distance may be achieved using special geometries in which complex flow paths are generated. In this paper, a novel design, utilizing semi-circular ridges, is proposed to improve mixing in micro channels. Fluid flow and scalar transport are investigated employing Computational Fluid Dynamics (CFD) tool. Mixing dynamics are investigated in detail for alternative designs, injection, and diffusivity conditions. Results indicate that the convex alignment of semi-circular elements yields a specific, helicoidal-shaped fluid flow along the mixing channel which in turn enhances fluid mixing. In all cases examined, homogenous concentration distributions with mixing index values over 80% are obtained. When it is compared to the classical T-shaped micromixer, the novel design increases mixing index and mixing performance values by the factors of 8.7 and 3.3, respectively. It is also shown that different orientations of ridges adversely affect the mixing efficiency by disturbing the formation of helicoidal-shaped flow profile.

Toward the Next Generation of Passive Micromixers: A Novel 3-D Design Approach

“…Nonetheless, further increase of flowrate after Re = 1, which is defined as an inflection point in the study, increases the mixing performance by means of chaotic advection formed in the mixing channel. In our previous work [ 18 ], a similar inflection point was seen at Re = 5 after which the effective use of advection was accelerated in convex semi-circular ridge (CSCR) micromixer design. Although the CSCR micromixer geometry developed a complex flow profile even at very low flow conditions (e.g., Re ≤ 5), and hence enlarged the contact surface between fluids, short residence time of fluids reduced the diffusive activity across the interfacial area formed.…”

“…To ensure stability in steady-state simulations, under-relaxation technique [ 31 ] is used with a factor of 0.9. Steady-state solutions are accepted to be converged when residuals reach 1 × 10 −6 threshold as typically practiced in several numerical micromixer studies [ 18 , 32 , 33 ]. In time-dependent simulations, temporal terms are discretized using Crank-Nicolson scheme with a blending coefficient of 0.9.…”

“…It should be noted that computing a mixing efficiency only relying on the scalar concentration distribution on a certain cross-section may provide imprecise mixing outcomes in cases where the distribution of scalar concentration is non-uniform in a flow profile. More information on this point can be found in References [ 18 , 33 ]. where A is area, u is velocity, σ and σ 2 are variance and the maximum variance, respectively, c is concentration, and is the average concentration.…”

“…Thus, contact surface area between fluids is enhanced, which in turn expedites diffusive mixing. Typical design and fluid mixing examples may be seen in References [ 18 , 19 , 20 , 21 ]. At very low Re numbers (e.g., Re < 5–10); however, the effective use of advection is inhibited drastically due to unidirectional fluid flow developed in microchannels.…”

Numerical Investigation of Mixing by Induced Electrokinetic Flow in T-Micromixer with Conductive Curved Arc Plate

“…The results of the present work have a lower mixing time and mixing length compared to the results of other studies that used passive T-shape micromixers. Moreover, the geometry used in the present work is clearly simpler [35,36]. In this section, the computational outcomes, obtained for improving the performance of the micromixer, are presented.…”