WEB Experimental and numerical investigation of the fine particle fractionating using a cross-flow module with superimposed electric field
The increasing significance of fine particles dispersed in a liquid phase with highly specific physical properties is leading to the development or advancement of classification processes. These suspensions are often intermediates for the production of coatings, structured reinforced plastic or for printed electronics. While production processes like precipitation or crystallization often generate particles with broad particle size distributions, a downstream classification process is needed. In known classification processes such as sedimentation, centrifugation or the usage of hydrocyclone, the particles are classified according to a certain grain or equivalent size. However, in many applications in which finely dispersed multicomponent mixtures of particles with sizes < 10 μm with different properties are to be separated in industrially relevant quantities, the usage of a single separating feature is often no longer sufficient. In this work, a novel cross-sectional superficial electric field filtration technique is being developed, which is a promising method for the highly specific separation of micro- and sub-micron suspensions. The method not only allows fractionation with regard to the particle size, but also the particle shape, the chemical composition and the physical properties of the individual particles. In this study, the test plant is developed and constructed. Based on CFD simulations, the flow channels and the size of the filter area were designed. The hydrodynamic forces on the particles in the laminar boundary layer are investigated experimentally. The possibility for the use the cross-flow as an effective classification process was proofed. A successful fractionation of particles < 5 µm was achieved. By varying the process parameters, the limits of the presented technique are detected. The influence of the electric forces on the particle motion is investigated with CFD methods.