Theoretical studies on aerosol agglomeration processes: Dissertation

In this thesis, theoretical modeling of certain aerosol systems has been presented. At first, the aerosol general dynamic equation is introduced, along with a discretization routine for its numerical solution. Of the various possible phenomena affecting aerosol behaviour, this work is mostly focused on aerosol agglomeration. The fundamentals of aerosol agglomeration theory are thus briefly reviewed. The two practical applications of agglomeration studied in this thesis are flue gas cleaning using an electrical agglomerator and nanomaterial synthesis with a free jet reactor. In an electrical agglomerator the aerosol particles are charged and brought into an alternating electric field. The aim is to remove submicron particles from flue gases by collisions with larger particles before conventional gas cleaning devices that have a clear penetration window in the problematic 0.1 - 1 mm size range. A mathematical model was constructed to find out the effects of the different system parameters on the agglomerator s performance. A crucial part of this task was finding out the collision efficiencies of particles of varying size and charge. The original idea was to use unipolar charging of the particles, and a laboratory scale apparatus was constructed for this purpose. Both theory and experiments clearly show that significant removal of submicron particles can not be achieved by such an arrangement. The theoretical analysis further shows that if the submicron particles and the large collector particles were charged with opposite polarity, significant removal of the submicron particles could be obtained. The second application of agglomeration considered in this thesis is predicting/controlling nanoparticle size in the gas-to-particle aerosol route to material synthesis. In a typical material reactor, a precursor vapor reacts to form molecules of the desired material. In a cooling environment, a particulate phase forms, the dynamics of which are determined by the rates of collisions and coalescence. In the thesis, it is first theoretically demonstrated how the onset of dendrite formation and primary particle size can be -predicted by studying the characteristic time scales of collision and coalescence. Then it is shown how the linear rate law for coalescence can be approximately applied to agglomerate structures by dividing the agglo-merates into sections. The developed models are then applied to a free jet material reactor. From the comparisons between theory and experiment it is obvious that such a model is able to capture the effects of the system parameters (temperature, velocity, volume loading of material and location of collection) on the primary particle size of the produced material.