Clog-Free Trilobite Filtration: Tunable Flow Setup and Velocity Measurements

Abstract

The ability to separate and filter out microscopic objects lies at the core of many biomedical applications. However, a persistent problem is clogging, as biomaterials stick to the internal chip surface and limit device efficiency and liability. Here, we review an alternative technique that could solve these clogging issues. By leveraging tunable flow fields and particle inertia around special trilobite-shaped filtration units, we perform filtration of plastic beads by size and we demonstrate sorting of live cells. The separation and filtration are performed completely without signs of clogging. However, a clog-free operation relies on a controlled flow configuration to steer the particles and cells away from the filter structures. In this paper, we describe the tunable flow system for such an operation and we describe an optical setup enabling hydrodynamical interactions between particles and cells with the flow fields and direct interactions with the filter structures to be characterized. The optical setup is capable of measuring particle and flow velocities (by Particle Tracking Velocimetry (PTV), Micro Particle Image Velocimetry (μPIV), and streakline visualization) in meters per second necessary to avoid clogging. However, accurate measurements rely on strict calibration and validation procedures to be followed, and we devote a substantial portion of our paper to laying out such procedures. A comparison between μPIV data and a known flow profile is particularly valuable for assessing measurement accuracy, and this important validation has not been previously published by us. The detail level in our description of the flow configuration and optical system is sufficient to replicate the experiments. In the last part of the paper, we review an assessment of the device performance when handling rigid spheres and live cells. We deconvolute the influences of cell shape from effects of size and find that the shape has only a weak influence on device performance.