Mixing by chaotic advection in a magneto-hydrodynamic driven flow

Abstract

A new device containing three circular electrodes and where very small quantities of a weakly electrically conductive liquid are propelled and mixed by chaotic advection is designed and constructed. The liquid, a copper sulfate solution, is propelled by the Lorentz body force, i.e., a magnetic field perpendicular to an electrical current. When the potentials of the electrodes are constant and the Lorentz force is small enough so that at the free surface the vertical velocity is practically zero, the flow field exhibits there a saddle point when the three circular electrodes are not in a concentric position. By modulating the electrical potential between the electrodes, the position of the saddle point changes. This slowly varying system is far from integrable and exhibits large-scale chaos, the non-integrability is due to the slow continuous modulation of the position of the saddle stagnation point and the two streamlines stagnating on it. Dye advection experiments are compared successfully to a numerical solution of the 3D equations of motion under these assumptions. We have also defined a potential mixing zone to predict the location of the chaotic region and calculated Poincaré sections. These two tools give results which are in excellent agreement, they are used, with others, to adjust the mixing protocol parameters and the geometry in order to improve mixing.

A study of MHD-based chaotic advection to enhance mixing in microfluidics using transient three dimensional CFD simulations

“…Their magnetic flux density is 0.05T whereas ours is 0.36T. There is also another major difference between the works of Dufour et al [24] and ours in the electric potential boundary conditions. They used an electrode potential of 0.4 V in their simulations, whereas we used 0.04V after accounting for the steep double layer drop.…”

“…Our results show that efficient mixing is possible in microscale devices with time periods of potential switching and elapsed time, a surrogate for a lab-on-a-chip response time, an order of magnitude less than reported by Dufour et al [24]. The main reason for the shortening of the mixing time is the choice of the number of disk electrodes, their placement and the switching schemes.…”

“…It'd be helpful to compare our results to Dufour et al [24] and highlight similarities and differences. They also used a shallow cylindrical cavity, but of a larger radius (17.5 mm), and the electrodes were non-intrusively placed on the cavity floor.…”

“…By doing so, three typical flow sequences were created, and lamination, stretching and mixing performance were investigated. Dufour et al [24] performed experiments in a shallow cavity and compared their results to those using linearized equations under the Stokes flow assumption. Gopalakrishnan and Thess [25] studied glass melt homogenization by stirring and mixing of flow in a pipe mixer subjected to electromagnetic forces by using computational methods.…”

Electromagnetohydrodynamic (EMHD) flow between two transversely wavy microparallel plates

“…Recently, analytical [5][6][7][8][9][10][11][12][13], numerical [14][15][16][17], and experimental [4,[18][19][20][21][22][23][24][25][26] studies on EMHD flow in microchannels were widely investigated. EMHD can be used not only for the purpose of propelling fluids but also for generating secondary complex flow that may be beneficial for mixing and stirring [27][28][29][30][31].…”