Microfluidic flow splitter shape optimization for high throughput

Sammanfattning: Due to an increasing incidence rate, the need for new effective methods for diagnosing cancer are of high importance. Studies have shown that circulating tumor cells, CTCs, have great potential as a diagnostic biomarker. Acoustophoresis is a label free method, capable of separating CTCs from red blood cells in a blood sample. In theory, this would enable better CTC detection than the gold standard method of today, and in turn a better way of diagnosing cancer. However, due to inertial forces, the particles are pushed closer to the centerline of the channel in a phenomenon known as the spillover effect. This occurs without the influence of acoustophoresis, and minimizing the effect is a prerequisite for acoustophoretic separation to work at high throughput. The aim of this project was to verify that the spillover effect could be decreased by changing the shape of the fluid splitter, as has been suggested by previous studies. This was done using genetic algorithms—algorithms that utilize the concept of biological evolution in order to reach optimal solutions. The results of the project show that it is indeed possible to push the particles further away from the center of the channel, however at the expense of raising the shear stress in the system. The best achieved result was a 40 percent increase in the distance from the center line. Provided that a physically fabricated version of the optimized chip works as well as the simulations showed, this would make the acoustophoresis method more clinically viable.