Pushing the boundaries of acoustic particle separation: achieving high-throughput, avoiding spillover effects, investigating the effects of the particle concentration, and measuring acoustic properties

Sammanfattning: Acoustophoresis, i.e., the contactless manipulation of small cells or particles via acoustic waves is a gentle and label-free method that can, amongst others, be used to isolate rare cells such as circulating tumor cells from blood. To isolate enough cells, a large amount of sample needs to be processed in a reasonable amount of time. In a clinical setting, usually around 10 mL of blood can be collected from a patient. Typically, a throughput higher than 100 µL/min is expected to process the whole sample under one hour. Currently, the sample throughput achievable by acoustophoresis is around 5 to 20 µL/min. The underlying work aims at resolving this exact challenge. Several different strategies exist to increase the throughput. First, increasing the total flow rate or the sample flow rate are obvious candidates. However, at high flow rates inertia effects impact the flow dynamics and the spillover effect occurs, which pushes all particles in the center outlet and thus, renders all separation impossible. Herein, we show that the spillover effect can be reduced by tuning the outlet splitting ratio. We obtain separation with an unprecedented sample throughput of 1200 µL/min. A second strategy consists of analyzing the set up numerically, to find the optimal setting parameters in silico. To this aim, we set up a finite element model and validated it with experimental results. To our surprise, the position of the streamlines in the prefocusing channel depends on the prefocusing voltage at high flow rates, which raises new questions on the behavior of the chip at these flow conditions. By considering this effect in the finite element model, simulations and experiments give similar positions of the particle streamline in the main channel. A third strategy to increase throughput is to run the separation at high sample concentrations. However, this entails that the particles are closer to each other, which can lead to hydrodynamic particle-particle interactions. We made a few observations at high particle concentrations. An increase in apparent acoustic energy density was observed. Furthermore, the separation curve, which is used as a tool to characterize particle separations, showed that increasing the concentration or the flow rates hinders particle separation. Here, another surprising effect occurred: at very high flow rates, and rather high inlet splitting ratios, the gap between the two particle streamlines is much larger than what we would expect from the theory. This effect could be further investigated to improve the separation performance. Finally, we introduce a new method to measure the mobility ratio of cells and particles based on particle separation. The method is validated by measurements previously obtained with particle tracking. The advantage of particle separation with respect to particle tracking is that many particles can be evaluated at once. Moreover, it does not require any knowledge on the flow rate, the acoustic energy density, the viscosity of the fluid, or the length of the channel.