473464 Simulation of Near-Size-Independent iDEP Separation Using Multiple Electric Fields

Tuesday, November 15, 2016: 10:15 AM
Embarcadero (Parc 55 San Francisco)
Benjamin G. Hawkins and Ngoc Huynh, Biomedical, Chemical, and Materials Engineering, San Jose State University, San Jose, CA

Dielectrophoresis (DEP) is a non-linear electrokinetic force that depends on the magnitude of gradients in the squared electric field and differences between electrical properties of a particle and its surrounding media. For a particle immersed in a semi-infinite media with a non-uniform electric field that is well-approximated by a truncated multipole expansion (i.e., not too nonlinear), the DEP force is:


( 1 )

where is the particle radius, is the ÒrealÓ operator, and is the so-called complex Clausius-Mossotti factor which is dependent on the electrical properties of a particle and its surrounding fluid. Owing to this dependence, DEP-based techniques have shown great potential to isolate, separate, enrich, or otherwise manipulate biological particles (e.g., cells, vesicles, DNA). Sensitivity to particle size arises in two locations: the Clausius-Mossotti factor (for particles with complex internal structures) and in the preceding term. The cubic dependence on particle radius means that biological variability in size will likely dominate any subtler Ð through much more relevant and interesting Ð electrical differences indicative of a particular phenotype or composition of interest.

We have designed a new dielectrophoresis separation that can reduce this primary dependence on particle size. The technique combines multiple DEP fields at different frequencies with an electrically insulating constriction in channel depth. When an electric potential is applied across an insulating constriction, the electric field is concentrated, forming a local gradient in the electric field where the channel cross-sectional area changes. This gradient drives DEP forces and has been used to separate particles in a number of different systems and is typically referred to as ÒinsulatorÓ-based or ÒiDEPÓ. The device layout is shown in Figure 1A and composed of a wide straight microchannel with an angled constriction across its width. A pressure driven flow is used to carry particles through the device. Electrodes are embedded along the sidewalls on either side of the constriction, and are energized with different electric potentials at different frequencies. The two field frequencies are chosen based on analytical and empirical data: one where the DEP response is sensitive to changes in parameters of interest (e.g., increasing cytoplasmic permittivity) and one that is opposite and insensitive to changes in parameters of interest. The latter is typically low frequency such that depends on membrane conductivity and drives nDEP; the former is determined by multishell modeling or empirical data. The ÒinsensitiveÓ or nDEP field is applied uniformly along the sidewall and exerts a uniform nDEP force that exceeds the fluid drag force. The resulting particle motion is parallel to the constriction and at an angle to the direction of pressure driven flow. The ÒsensitiveÓ or pDEP field is applied to the other electrode and Ð critically Ð increases along electrode length. The separation depends on the sum of three forces: the uniform nDEP force, the increasing pDEP force, and the fluid drag force. As the pDEP force increases, the sum of the pDEP and nDEP forces decreases until the nDEP force no longer exceeds the drag force and the cell is transported with the fluid out of the device. By driving the magnitude and spatial variation in nDEP and pDEP forces, it is possible to reduce the separations dependence on the drag force relative to the DEP force. In preliminary work, we have shown that, using a multishell model, it is possible to use this technique to distinguish bacterial cells with membrane permittivities differing only by a factor of two (Figure 1B).    

Key benefits of this design are: continuous flow separation, variable sensitivity based on the chosen electric potentials, and reduced sensitivity to particle size (Figure 4). One feature of the proposed technique is its ability to tune separation sensitivity by changing the applied electric potentials, the fluid volumetric flow rate, and the frequency of applied signals. Compared to other systems, this device can operate on a wider range of cell samples and operates in continuous flow.

A. B.
Figure 1: (A) Schematic of applied electric fields and resulting F_DEP. Flow from top to bottom, electric fields are applied on the left (nDEP) and right (pDEP) electrodes (red). Electric field gradients are generated by a diagonal constriction in channel depth. nDEP field (orange) is constant and pDEP field (green) increases along channel; particles are deflected along constriction by nDEP until |F_pDEP |=|F_nDEP | and fluid drag dominates. (B. a, top) 2D simulations of particle trajectories using the electrode scheme in (A). Flow is from left to right and colortable corresponds to total electric field magnitude (black is highest over the constriction). Particle trajectory is modeled based on a double shell model for f___CM. __(r,membrane) is the relative permittivity of the membrane. A 2-fold difference is in __(r,membrane) between (B. a) and (B. b) yields a significant change in particle position at the outlet16.


Extended Abstract: File Not Uploaded