The Dielectrophoresis Network
at the University of Surrey
Dielectrophoresis has often been used as a separation technology - two different particle types with sufficiently different electrical properties can be separated if hte conditions allow one particle type to experience positive DEP, the other negative DEP. We have been working on this with our partners, principally DSTL Porton Down but also with Smiths Detection, to develop sample processing technologyies to assist in the detection of airborne pathogens such as bacteria and viruses, prior to detection using surface-based sensor technology. As an initial sift mechanism, DEP is very effective at removing particles such as dust, soot and clay from bacteria in sample scollected directly from the environment, and we have developed and evaluated such devices based on "conventional" DEP technology.
Following the initial purification of the sample, there is then a requirement to concentrate it and draw the analyte to a sensor surface for analysis. Whilst there are many examples of DEP particle concentration and sparation that have been used to act upon cell-sized particles, when nanoparitlces such as viruses and proteins are examined, limitations to DEP begin to emerge. The dielectrophoretic force scales with particle volume, so in order to compensate for this, much higher field gradients need to be used. These tent to act overshort distances, making the calpure volume small. This is a problem for amy applications, such as environmental monitoring; harmful agents can remain airborne for very long distances. There are difficulties associated with the direct detection of particles in air, and hence samples collected must be re-suspended in liquid prior to detection, and must then be detected at potentially low concentrations and in the initial presence of confounding particles such as diesel and pollen. Furthermore, the movement of such small particles is governed by Brownian motion, which can make it very difficult to detect them using surface-based interactions, simply because the particles rarely come into contact with the surface.
In conjunction with DSTL, we have developed a new method of harnessing the phenomenon of induced particle collection due to electrostatic effects in fluids, such as electro-osmotic flow (AC-EOF) and dielectrophoresis (DEP), to pre-concentrate virus sized particles (100 nm latex beads) in a continuously flowing system, or to pull particles down onto a seonsor surface. The effect is largely size-independent, trapping particles form 20nm-20um in diameter at about the same speed, and works with or without an externally-imposed flow.
The technique uses electro hydrodynamic forces (EHD) caused by the interaction between the induced double layer at relatively low frequencies (less than 1kHz) and the electric field through that double layer. The electric field passess throguh tangentially, producing forces that both move the charges towards the electrode and away from the elelctrode edge. With appropriate elelctrode design, this can produce vortices that trap all particles above the electrode and pull them to the surface, from as far as 500um away, depositing them in the middle of a sensor surface. We've successfully used this technique to enhance QCM biosensors, and it should worl equally well for other surface sensing methods. The electrodes resemble a locked zipper, hence its name - the Zipper electrode.
Where particle concentration is required from flow, we achieve it by introducing asymmetric vortices into a laminar flow and thereby push particles to one side of a flow cell. This is achieved by flowing the sample over electrode tracks that are angled against the laminar flow pattern inside a flow cell. An applied AC field causes vortices to form over the electrode tracks; since the vortices are angled against the flow, they form an asymmetric pattern that pushes particles along on the electrode track to one side of the flow cell. This results in a higher concentration of particles at one side of the flow cell. By splitting the flow at the end of the flow cell into a low and a high concentration stream, particles can be recovered from the flow cell. The system has so far achieved a 7-fold increase in particle concentration on a 1/15 flow split; this represents an actual, real-world concentration value in the outlet stream, rather than a local concentration that needs to be dilutes in order to be released from the device.
Relevant papers on the Journal papers page: 25, 27, 39, 41, 63, 64.