(419d) Mixing Enhancement in Microfluidic Devices Using Contactless Dielectrophoresis (cDEP)

Mixing in microfluidic devices has an important role in numerous biological and chemical applications. Rapid mixing is necessary in many lab-on-a-chip (LOC) and Micro Total Analysis Systems (µTAS) devices for chemical processes, including specific applications such as chemical reactions, and for biological processes, such as enzyme reactions, DNA or RNA analysis, and protein folding.

Despite the small length scale of microdevices, rapid, efficient mixing is difficult to achieve. Due to small length scale the Reynolds number, the ratio between inertia and viscous forces, is small. Thus mixing without any intentional stretching and folding of interfaces is dominated by molecular diffusion, which takes a long time relative to the typical operating timescale of these microdevices. To address this need, various techniques for enhancing mixing in microdevices have been proposed, including both active and passive techniques that are based on phenomena such as chaotic advection. There are many active mixing techniques, including those that use pressure, magnetohydrodynamic, acoustic, or thermal disturbances to enhance mixing. Electrokinetic mixing has proven to be an efficient method for actively mixing solutions or microparticles.  

DEP, the motion of a particle in a suspending medium due to the presence of a non-uniform electric field, has shown great potential for particle separation, manipulation, and identification of micro- and nano-scale particles. However, studies in the literature using DEP for mixing enhancement are fairly limited. 

In this study, we have investigated mixing in microdevices based on a new technique, termed contactless dielectrophoresis (cDEP). In cDEP, an electric field is created in a microchannel by electrodes that are inserted into two side channels filled with conductive solution.  These side channels are separated from the main channel by thin insulating barriers that exhibit a capacitive behavior.  An electric field is induced in the main channel by applying an AC field across the barriers. Not having direct contact between the sample and the metallic electrodes eliminates issues that plague conventional DEP such as bubble formation, electrode delamination, and sample contamination. Furthermore, the fabrication process is relatively simple because it is not necessary to pattern micro-electrodes in the main channels.  This method thus is well suited to traditional mass fabrication techniques such as hot embossing and injection molding.  

We consider four variations of a device that consists of two “mixing chambers” located on opposite sides of a 50 mm or 100 mm rectangular microchannel. The chambers are either rectangular or semi-circular, and they are either placed symmetrically on opposite sides of the channel or are staggered with an axial spacing of 300 mm.  In all cases, the depth of the chambers is 50 mm to match the channel depth.

In our cDEP device, there are electrode channels which are separated from the mixing chamber by 20 mm insulating barriers. The electrode channels are filled with a conductive solution, and electric field gradients are generated in the microchannel using electrodes that are inserted into two electrode channels. The capacitive nature of the barrier between the chambers and the electrode channels generates a nonuniform electric field when an AC signal is applied. In some locations, the electrode channels are far from the mixing chamber to prevent their electric effect on the mixing chamber, thereby creating an electric field similar to that given by three isolated electrodes.

Before applying the electric field, the two inlet streams flow side-by-side through the devices with negligible diffusive mixing. After applying the electric field, the beads suspended in the bottom fluid are manipulated by the induced DEP force.  The mixing chambers exhibit a resulting secondary flow. This secondary flow is observed to stretch and fold the fluid, which is an important characteristic of rapid mixing. The stretching and folding in each device occurs between the symmetrically placed mixing chambers. In devices with two separate mixing chambers, the first mixing chamber begins the stretching and folding, and then the second chamber amplifies this stretching and folding and enhances mixing.

We quantified the effectiveness of these devices by evaluating mixing index, which is equal to 1 minus the ratio of the standard deviation of the concentration after mixing to the standard deviation of the concentration before mixing. We ran mixing experiments for voltages between 0 and 300 V_rms (0, 50, 100, 150, 200, 250, 300 V_rms) at a constant frequency of 600 kHz and a flow rate of 5 µL/hr. We found that both devices with rectangular mixing chambers achieve a maximum mixing index of 80%.  However, the maximum mixing index obtained in circular mixing chamber was 57%. This occurs because the rectangular shape of mixing chamber decreases the velocity of particles inside the mixing chamber, which give more time to DEP to influence the particles and create mixing. However, circular mixing chambers cannot have the same effect on particle’s trajectories; thus, we do not observe efficient mixing.

We have also considered the dependence of mixing on the flow rate for a fixed voltage and frequency. The mixing efficiency decreases by increasing the flow rate, which is equivalent to decreasing the flow time scale. The results also showed that a higher mixing index can be obtained at higher voltages for a given flow rate. 

In addition to mixing beads with a fluid, these microdevices can also be used for mixing two fluids when one of those fluids contains beads. To prove this claim, we suspended beads in a water-based blue dye solution and pumped the sample into one of the microdevice inlets. Deionized water was pumped through the other inlet. After applying an AC signal of 300 V_rms and 600 kHz, the dye solution got mixed with the deionized water due to the secondary flow generated by the DEP force on the beads.