中文摘要

論文題目:具橫條結構T型管微混合器混合效率改良
研究生:周洧志
指導教授:潘大知

本研究以計算流體力學作為工具，對T型管微混合器進行數值模擬實驗，探討在T型管主流道內加入橫條結構對混合器混合效率之影響，並以田口方法設計橫條得結構參數，利用直交表搭配控制因子反應表找出有利於提升混合效率的參數趨勢，進而找出高混合效率之設計參數。

關鍵字:計算流體力學；微混合器；T型管；田口方法

Mixing Efficiency Improvement of T-channel Micromixers with Crossbar Structures

Author: wei- Chih Chou

Advisor: Dar-Tzi Pan

Department of Aeronautics and Astronautics, National Cheng Kung University

SUMMARY
Numerical simulations are performed to study the mixing efficiency of T-channel micromixers with crossbar structures immersed in the main channel . A commercial CFD software and Taguchi Method are used to set up numerical experiments to search for the crossbar geometric parameters corresponding to high mixing efficiency.

Keywords: CFD; Micromixer; T-channel

INTRODUCTION
There are two kinds of micromixers:
Active micromixers: External sources of energy of various kind are introduced and actively controlled to enhance mixing.
Passive micromixers: No movable or controllable component is implemented. Mixing can be enhanced only by inducing vortical flows using geometric variations of the flow passage.

This thesis works on the improvement of passive T-channel with crossbar structures placed in the main channel to enhance mixing. A commercial software developed by CFDRC is used as the tool to compute the mixing phenomena of two fluids in the main channel of micromixers. The orthogonal tables of Taguchi method are employed to find the geometric settings of the crossbar structures for high mixing efficiency. The geometric dimensions of the crossbars and the main channel are less than 1mm.

IMPORTANT PARAMETERS
Reynolds Number:
The Reynolds number is defined as Re=(ρV_ref L_ref)/μ, where ρ is the density of the fluid(kg/m^3), V_ref is the characteristic velocity of the fluid flow (m/s), L_ref is a characteristic linear dimension(m), μ is the dynamic viscosity of the fluid( kg / m • s). Reynolds number is generally used to identify whether the flow studied is
laminar or turbulent. In this thesis, the Reynolds number is so low that the flow is laminar.
Schmidt Number:
Schmidt number (Sc) is a dimensionless number defined as Sc=μ/ρD, where μ is the dynamic viscosity of the fluid( kg / m • s), ρ is the density of the fluid(kg/m^3), D is the mass diffusivity (m^2/ s). It is an indicator of the effectiveness of mass diffusion as compared with momentum diffusion.

BASIC ASSUMPTIONS
The two fluids undergoing mixing are the same fluid with different mole concentrations. During and after mixing the fluid properties remain unchanged, but concentration changes. The following conditions are assumed:
(1)The fluid is Newtonian fluid.
(2)The flow is incompressible and laminar.
(3)The fluid density , viscosity and diffusivity are constant.
(4)No chemical reaction occur during mixing.
(5)The body forces can be neglected.

TAGUCHI METHOD
The Taguchi method is a statistical method developed by Genichi Taguchi. It has been widely used in factories to improve the quality of manufactured goods. The orthogonal table of Taguchi method is usually applied to minimize the number of experiments required to collect adequate statistical information about the manufacturing process in order to ensure the quality control of the product. It is termed as L_a (b^c), where “a” is the number of experimental runs, “b” is the number of possible settings of control factors, and “c” is the number of control factors of the processes. In this thesis, Taguchi method is used to setup the numerical experiments for finding the crossbar parameters corresponding to high mixing efficiency.

BASIC SMOOTH-WALL T-CHANNEL MICROMIXER
The height H of the square sectional plane of the main channel is used as the reference length in this study. It is 1 mm. The mean velocity U_mean at the branch entrance of the mixer is used as the reference velocity. The Reynolds number studied here ranges from Re=0.1 to Re=10, which is well within the laminar flow range. For smooth-wall micromixer, the mixing occurring in the main channel relies mainly on the molecular diffusion in the direction perpendicular to the main stream.
A mixing index σ can be defined for the concentration distribution on the cross sectional plane of the main channel. While σ =0 indicates no mixing occurs on the plane, σ=100% means a complete mixing. Here we take σ=99% or 98% as an indication of well mixing. The length of channel required for the occurrence of σ=99% is termed as L_(99%). For smooth wall micromixers with water as the working fluid, L_(99%)=28.5mm. We shall introduce structural variations in the main channel to enhance mixing, or equivalently, to reduce L_(99%).

MIXING ENHANCEMENT BY CROSSBAR STRUCTURES
Two rows of crossbar are placed in the main channel to enhance mixing. The control factors adapted in this study are:
The gap distance G between the upper and the lower crossbars.
The separation S between two consecutive crossbar structures.
The horizontal offset of the crossbars.
The oblique angle T of the crossbar relative to the main stream direction.
The height of crossbar, H.
The upper row and the lower row of crossbars can be placed in parallel to each other or in staggered Ⅹ-formation.
The combination of control factors derived from this study using water as the working fluid at Re=0.1 reduces the channel length required for a 99% mixing from 28.5mm for smooth-wall T-channel micromixer to only 9.76mm.

CONCLUSION
Numerical results have shown that the staggered Ⅹ-formation is less effective to enhance mixing. Instead, the parallel crossbar formation with proper geometric parameters can effectively reduce L_(98%) with working fluid of water, alcohol and hydrogen gas at Reynolds number less than 10.

參考文獻

[1]Nguyen, N. and Wu, Z ., “Micromixers - a review,” Journal of Micromechanics and Microengineering,NO.15 ,pp1-16,2005

[2] Glasgow, I., and Aubry, N.,“Enhancement of microfluidic mixing using time pulsing,” Journal of Lab on a Chip,No.3,pp.114-120,2003

[3]Wang . Y ., Zhe. J.and Chung, B . T. F. “A rapid magnetic particle driven micromixer” Microfluid Nanofluid 4:375-389,2008.

[4] Mülle, S.D ., Mezi, I, Walther,J.H.,Koumoutsakos, p., “Transverse momentum micromixer optimization with evolution strategies”, Computer&Fluids,No.33,pp521-531,2004

[5]Schwesinger, N., Frank, T., and Wurmus, H., “A Modular micro-fluidic System with an Integrated Micromixer, ”Journal of Micromechanics and Microengineering,Vol.6,No. 1,pp.99-102,1996

[6]Miyake, R., Lammerink T.S.J., Elwenspoek M., and fluitman, J.H.J., “Micromixer with Fast Diffusion, ” An Investergation of IEEE , pp248-283,1993.

[7]邱柏誠，利用橫條結構改善T型管微混合器之混合效率，國立成功大學航空太空研究所論文，2015。

[8]Stroock,A.D.,Dertinger,S.K.,Whitesides,G.M.and Ajdari,A., “Patterning Flows Using Grooved Surface ”, Analytical Chemistry, Vol. 74, No.20,pp.5306-5312,2002.