High-throughput electrokinetic nanoparticle manipulation: Dealing with sub-10 nm nanoparticle contamination in advanced CMOS processing

Particle contamination has been the dominant yield-limiting factor throughout the entire history of semiconductor manufacturing, with estimates attributing up to 75% of yield loss to contamination. As critical device dimensions continue to shrink, the tolerable size of particulate defects decreases proportionally. With critical dimensions in today’s leading-edge logic and memory nodes approaching or already moving below 10 nm, sub-10 nm particles are now becoming killer defects [1].

This represents a fundamentally new regime with many unknowns, where the dimensions of particles, particle precursors and molecules in liquids begin to overlap.  In this size range, “particles” consist almost entirely of surface and interface, resulting in properties that differ from the bulk material, modified or enhanced chemical reactivity, interaction forces that depend on the microscopic shape of the interface, etc.

At the same time, the metrology techniques available in the fab are limited in their detection capabilities. Detection of particles on a surface using light scattering is ineffective below 10 to 15 nm. Particle counters in liquids are only able to measure down to 30-40 nm, and cannot differentiate between foreign particles, micelles, or bubbles [1]. Also chemical analysis becomes increasingly difficult.

As a result, the focus shifts increasingly to the prevention of particles, and to understanding their nature and origin. Particle contamination originates from multiple sources, including airborne particles, airborne molecular contamination that may result in the nucleation and growth of particle-like deposits on the surface, as well as liquid-based sources. In this thesis you will focus on particles in liquids, and in particular the electrokinetic phenomena associated with them.

When an electric field (DC or AC) is applied to an aqueous liquid containing nanoparticles, a variety of phenomena takes place. For large particles, the bulk properties and classical Maxwell electrostatics are sufficient to describe the electrophoretic (EP) or dielectrophoretic (DEP) behavior. However, for small particles the problem is much more complex. First, hydrodynamic phenomena associated with the mobile ions in the electric double layer surrounding any charged surface [2] result in a variety of electrokinetic forces that can no longer be neglected. Electroosmotic flows (EOF) and induced charge electrophoresis (ICEP) may dominate over the pure DEP force [3]. At an even smaller length scale, molecular dynamics details at the interface become important [4].

Electrokinetic phenomena have the potential to contribute significantly to the control of sub-10 nm particle contamination. It has been shown experimentally DEP allows to remove nanoparticles from processing liquids in a size range where traditional filtration techniques are struggling [5], although currently at low throughput and with limited fundamental understanding. DEP may also be used to upconcentrate particles for downstream analysis, or to differentiate between particles and chemical precursors, or between particles and micelles. An AC field applied to an electrolyte with unequal ion mobilities will set up a long range DC field that can levitate particles away from the surface [6], etc.

However, fundamental understanding of the involved phenomena is still far from complete. An associated practical challenge is the low throughput. The very large electric field gradients needed to manipulate nm-sized particles using DEP or ICEP can only be generated close to nanoscale electrodes (electrode-based DEP, eDEP) or nanoscale constrictions (insulator-based DEP, iDEP) and do not extend very far into the liquid. Hence a second challenge is to develop structures and devices that can handle large volumes of liquid, e.g. by extending dielectric filtration concepts [7] to the nm regime and to highly conductive liquids. Here as well the electrokinetic phenomena at the solid-liquid interfaces within the device are expected to play a key role.

The goal of this thesis is to increase the fundamental insight in the properties and behavior of nanoparticles and explore how to use this to study, manipulate, or remove nanoparticles from liquids. You will do this using a combination of theory/modelling and experimental approaches. The existing Maxwell stress tensor approach [3] will be generalized and combined with finite element methods to handle complex device geometries. Dielectric spectroscopy (DS) will be used to characterize the response of dilute particle solutions and validate the modeling results [8]. Nanopore-based devices will allow the study and manipulation of individual nanoparticles [9].

[1]        IEEE International Roadmap for Devices and Systems, “Yield Enhancement,” 2023, Institute of Electrical and Electronics Engineers. doi: 10.60627/TE8P-7W91.

[2]        M. Z. Bazant, M. S. Kilic, B. D. Storey, and A. Ajdari, “Towards an understanding of induced-charge electrokinetics at large applied voltages in concentrated solutions,” Advances in Colloid and Interface Science, vol. 152, no. 1–2, pp. 48–88, Nov. 2009, doi: 10.1016/j.cis.2009.10.001.

[3]        B.-Y. Shih et al., “Not only dielectrophoresis: The crucial role of electrokinetic phenomena in a dielectric particle’s response to an oscillating electric field,” Journal of Electrostatics, vol. 133, p. 104009, Jan. 2025, doi: 10.1016/j.elstat.2024.104009.

[4]        M. Heyden and D. V. Matyushov, “Dielectrophoresis of Proteins in Solution,” J. Phys. Chem. B, vol. 124, no. 51, pp. 11634–11647, Dec. 2020, doi: 10.1021/acs.jpcb.0c09007.

[5]        D. Lee et al., “Insulator-Based Dielectrophoresis for Purifying Semiconductor Industry-Compatible Chemicals with Trace Nanoparticles,” JACS Au, vol. 5, no. 5, pp. 2342–2349, May 2025, doi: 10.1021/jacsau.5c00307.

[6]        S. M. H. Hashemi Amrei, S. C. Bukosky, S. P. Rader, W. D. Ristenpart, and G. H. Miller, “Oscillating Electric Fields in Liquids Create a Long-Range Steady Field,” Phys. Rev. Lett., vol. 121, no. 18, p. 185504, Nov. 2018, doi: 10.1103/PhysRevLett.121.185504.

[7]        M. Lorenz, D. Malangré, F. Du, M. Baune, J. Thöming, and G. R. Pesch, “High-throughput dielectrophoretic filtration of sub-micron and micro particles in macroscopic porous materials,” Anal Bioanal Chem, vol. 412, no. 16, pp. 3903–3914, Jun. 2020, doi: 10.1007/s00216-020-02557-0.

[8]        R. Pethig, “Protein Dielectrophoresis: A Tale of Two Clausius-Mossottis—Or Something Else?,” Micromachines, vol. 13, no. 2, p. 261, Feb. 2022, doi: 10.3390/mi13020261.

[9]        T. Colburn and D. V. Matyushov, “Trapping proteins on nanopores by dielectrophoresis,” Journal of Applied Physics, vol. 133, no. 16, p. 164701, Apr. 2023, doi: 10.1063/5.0144564.