(475e) Effects of Electrothermally-Induced Flow On Electrodeless (insulative) Dielectrophoresis Devices
AIChE Annual Meeting
2010 Annual Meeting
Engineering Sciences and Fundamentals
Microfluidics and Small Scale Flows II
Wednesday, November 10, 2010 - 1:30pm to 1:45pm
We conducted numerical simulations of particle transport in electrodeless (insulative) dielectrophoresis (iDEP) devices. These simulations include coupled thermal, electrical, and fluid mechanical phenomena, including electrothermally-induced flows that arise from non-uniform fluid conductivity and permittivity where the thermal P?clet number is of order 1 (i.e., PeT ~ 1-10). Previous work has investigated the effects of electrothermally induced flows near electrodes in electrode-based DEP devices where PeTÂ¨ 0. In this work, we model diffusional heat transfer through the device substrate into the surrounding air as well as heat diffusion and convection in the fluid.
Dielectrophoresis (DEP) --- the motion of a polarized particle in a non-uniform electric field --- is an attractive technique for researchers attempting to manipulate particles or cells based on their characteristic dielectric properties. The DEP force on a particle is dependent on particle and fluid permittivity and conductivity as well as the frequency of the applied electric field. For a homogeneous, isotropic sphere in a semi-infinite, homogeneous, isotropic domain, the time-averaged DEP force can be written
where asterisks denote complex quantities,
is the complex Clausius-Mossotti factor,
. The applied electric field in this work is a DC-offset, AC electric field. We characterize the ratio of AC-to-DC electric field magnitudes with the parameter, a. Electrode-based DEP devices induce electric field non-uniformities via photolithographically patterned electrodes embedded in the microfluidic channel. Electrodes in iDEP devices are placed in external reservoirs at the inlet and outlet and rely on constrictions in channel geometry to alter the electric current path and induce non-uniformities. The simulated channel geometry and numerical mesh are shown in Figure 1.
Figure 1: Channel geometry with constriction in channel depth. Mesh resolution is high near channel boundaries and in the constriction region. Mesh consists of ~100,000 elements.
Electrothermally induced flow is the result of Coulombic body forces exerted on a localized charge density. In an electrically neutral fluid, localized bound charge densities arise from gradients in fluid permittivity and conductivity owing to localized variations in temperature. In the case of iDEP devices, these localized variations in temperature arise due to locally high Joule heating near geometric constrictions in the microfluidic channel.
Particle transport in iDEP devices is the result of fluid and electrokinetic forces. In this work, we model particle motion owing to electrophoresis and dielectrophoresis and fluid motion owing to electroosmosis and electrothermal body forces.
Modeling of the thermal conditions of the device incorporates heat diffusion within the fluid channel, through the substrate material, and into the surrounding air. These conditions correspond with experimental conditions found in our work: a plastic device placed on an inverted microscope stage. We linearize heat transport from the fluid into the plastic substrate using a temperature-dependent heat transfer coefficient for the fluid/plastic interface. In addition to heat diffusion within the channel, we model heat convection via electroosmotic fluid flow. The temperature distribution is coupled to fluid and electrical systems via temperature dependent material properties: specific heat, thermal conductivity, viscosity, permittivity, and electrical conductivity.
Figure 2: Particle deflection as a function of constriction ratio and AC-to-DC ratio, a. Deflection owing to DEP with (solid) and without (dotted) electrothermally induced flow. Electrothermal flow has the greatest influence at high a and at intermediate constriction ratios. Solution conductivity is 0.01 S/m.
We quantify the effects of electrothermal flow on particle transport in iDEP devices by calculating the difference between the input and output positions of a particle entering the middle of the channel (y=50Âµm). The effects of electrothermal flow as a function of constriction ratio (bulk channel depth vs. constriction channel depth) and the AC-to-DC ratio, a, in are shown in Figure 2. At low constriction ratios localized Joule heating is low and temperature gradients are not significant. As the constriction ratio increases, localized heating induces significant variation in solution conductivity and permittivity, leading to electrothermally induced flows in the constriction region. As the constriction ratio increases further, fluid viscosity and shear within the constriction region prevents electrothermal body forces from altering the flow field.
We characterize the temperature distribution and particle behavior as a function of relevant experimental parameters and define design principles for researchers working with iDEP devices. Our results examine the effects of solution conductivity, particle electrophoretic mobility, channel electroosmotic mobility, channel geometry, and DC electric field magnitude. Our results demonstrate that electrothermal flow can either be avoided through judicious choice of experimental design parameters or harnessed to enhance particle deflection.