Acoustic coagulation of aerosols

Abstract

Flue gas from combustion or syngas from gasification can contain fine particles, so called aerosols. These can be potentially harmful for downstream equipment or if emitted to the atmosphere, be inhaled by humans and cause respiratory problems. Standard practice to mitigate this is to use a cyclone which has a low filtration efficiency for sub-micron sized particles. In the 1920s it has been accidentally discovered that acoustic waves can coagulate aerosols to form bigger sized agglomerates. It has been suggested to use acoustic waves as a pre-treatment step before using a cyclone to increase the filtration efficiency without inducing a pressure or temperature drop on the flow. A study was done on the possible mechanisms for this coagulation and these are used in a population model based on Smoluchowski’s equation. The first mechanism is the orthokinetic particle interaction, which describes a relative velocity of two particles with different entrainment in relation to the velocity field of the acoustic wave. The second mechanism is the acoustic wake effect, an effect in which the distance between particles decreases due to the mutual slipstreaming in an oscillating flow. To verify if the correct physical relations are identified a lab scale setup was built. An of the shelf piezo-electric transducer intended for under water cleaning was taken apart and put into a FEM model. This model was validated with the use of various measurement methods and concluded was that the FEM model is a good tool to predict the mode shapes and Eigen frequencies, but unable to predict the magnitude of the vibration. The model was extended with the design of an ultrasonic horn and radiation plate. With these build, sound pressure levels of over 150dB at 41kHz were achieved which are a significant improvement of the transducer without the horn and plate. Three different experiment series were performed with aerosols of different composition, concentration and particle size. In the first series an aerosol was created by atomizing paraffin oil, which created a visible fog. The treatment of this with an acoustic field resulted in an increase of particle deposition in the coagulation chamber. In the second experiment series, the aerosol was formed by atomizing a brine solution to form NaCl crystals. The particles coagulated and a shift in the particle size distribution was measured in accordance with the model. The overall coagulation rate is still an uncertainty because of the inability to properly the measure the intensity of the acoustic field inside the chamber. A final experiment was done with brine atomization, but with helium as a carrier gas which has a lower density than air. As a result the flow rate of the atomizer has to be increased which resulted in a decrease of residence time. Coagulation was measured, but not entirely according to the model predictions. Concluded was that the model which is based on the orthokinetic particle interaction and the acoustic wake effect is able to predict the coagulation, but in some circumstances it appears that an effect is missing from the model.