Selective particle separation using Dynamic Acoustic Fields

Abstract

Separation of particles from suspensions has been an integral part of many industrial, chemical and biological processes. Traditional methods such as sieving, flocculation, centrifugation etc. have many demerits. This has lead to an interest in unconventional separation methods that use physical properties of the particles. Acoustophoresis is one such method that has gained popularity in recent times. Acoustophoretic separation is based on the density contrast between the particles and the medium, as well as the size of the particles. A particle in a medium in which an acoustic field has been applied, experiences an acoustic force. This acoustic force is greater in standing waves than traveling waves, which is why acoustophoretic devices generally use standing wave fields. Particles with positive contrast factor are pushed to the pressure nodes; once the particles reach the nodes, they are trapped. If the medium is in flow, the particles also experience drag force. The interplay between the drag and the acoustic force forms the basis for separation. Standing wave fields can further be distinguished as static fields or dynamic acoustic fields (DAF). In a static field, the nodes are stationary, while in DAF, there is movement of the nodes, caused by variation of source frequency. Here, the tendency of different-sized particles to follow the nodal movement gives rise to separation. DAF devices seem preferable when separation needs to be done on a larger scale.In this thesis, two novel DAF devices were studied; one device works on the principle of frequency sweeping, while the other works on frequency modulation. Both devices are 3D printed, and have two inlets and two exits, with separation occurring in an acoustic chamber where the DAF acts. Both devices are multi-nodal & milli-fluidic. The samples to be separated consist of polyethylene particles of a continuous size distribution between 32-106 µm, dispersed in water. The devices were tested at a total channel flow rate of 1000 mlh-1 (Reynolds number of 20). The aim was to understand how varying the flow and acoustic field parameters affect the separation/filtration performance of the devices. This was done by defining a “threshold size” or the cut-off size at which the entire particle distribution is divided into two separate distributions. Flow and acoustic field parameters were adjusted to see how the threshold size increases/decreases.The analysis for each device was done by a parametric study using Design of Experiments (DOE). First the available features (both design and operation) were studied and suitable parameters and their values were selected. The DOE was then designed, having 27 combinations of parameters known as ‘runs’. Each run of the DOE was first simulated on an available 2D COMSOL model, followed by experimental validation. The DOE was then analysed using both theoretical and actual values, to find out the parameters that had the maximum effect on the threshold size, and their respective trends.