Nanoscale dielectrophoresis for life science and memory applications: Theory and modelling

Dielectrophoresis (DEP) refers to the force that is exerted on a dielectric (= polarizable) object that is placed in a non-uniform electric field, typically in a liquid. This is a powerful technique that allows the selective manipulation, capture or sorting of objects based on their dielectric properties. Applications of this technique cover a very wide range. In life sciences, DEP has been successfully used for cell analysis and separation since the 1960’s [1]. More recently, biomanufacturing of therapeutic molecules, vaccines, food products, etc., has turned the focus towards much smaller bioparticles such as proteins, vesicles and viruses. In the seemingly unrelated field of computer memories, a novel colloidal memory concept is based on the selective capture and detection of two types of designer nanoparticles with different dielectric properties [2]. In all these cases the objects of interest are only a few nm to a few 10 nm in size.

Current understanding of DEP is based on theories developed for large objects, and is unable to correctly describe the DEP behavior of these nanometer-scale entities [3]. The Clausius-Mossotti (CM) factor of traditional Maxwell-Wagner models is based on homogeneous particles, placed into a homogeneous liquid, and with an infinitesimally thin interface between both. It predicts a DEP force that scales with the volume of the particle, such that Brownian motion is expected to dominate for particles smaller than a few 10’s of nm in size. However, successful experimental manipulation of proteins as small as 3 to 4 nm shows that this is not the case.

Recent insight points to two major effects: (1) the contribution of the electric double layer (EDL) surrounding a (homogeneously) charged particle becomes important when the decreasing particle size becomes comparable to the characteristic size of the double layer (the Debye length). (2) The presence of charged residues on proteins results in a permanent dipole moment, as well as in a modified dielectric behavior of the water molecules in the hydration shell, resulting in a modified induced moment. These two effects are currently studied separately, using different approaches. Molecular dynamics modeling is used for the local influence of the charged residues and the permanent dipole. The Maxwell stress tensor can be used for the longer-range hydrodynamic phenomena in the double layer, as well as for interactions with other particles or nearby walls.

The goal of this PhD is to develop multiscale modeling aimed at an improved understanding of the interplay between these various phenomena. The ultimate aim is to benchmark the modeling results to experimental data obtained by colleagues in the group, e.g. specific proteins, vesicles or nanoparticles, in specific device geometries.

[1] Pohl, “Separation of Living and Dead Cells by Dielectrophoresis,” Science 152, 647–49 (1966); https://doi.org/10.1126/science.152.3722.647.b.

[2] Rosmeulen, “Liquid Memory and the Future of Data Storage,” 2022 IEEE International Memory Workshop (IMW); https://doi.org/10.1109/IMW52921.2022.9779295.

[3] Pethig, “Protein Dielectrophoresis: A Tale of Two Clausius-Mossottis – Or Something Else?” Micromachines 13, 261 (2022); https://doi.org/10.3390/mi13020261.