TRANSIENT BEHAVIOUR OF AEROSOL CLOUDS IN A CONFINED VOLUME: AN EXPERIMENTAL AND COMPUTATIONAL STUDY

Combustible transient aerosol sprays may be formed when pipe or vessel failures on process plant lead to a leak of high pressure hydrocarbon fluids. The measurement of such transient aerosol clouds poses two diagnostic problems: the first is to measure very small droplets distributed over a large volume, the second is to dynamically measure the change in drop size and concentration over a short time period. In this study, transient homogeneous aerosols of water and kerosene were created inside a confined rig of 1.67m length, 1.5m width and lm height by various combinations of half-cone and full-cone Schlick spray nozzles. The aerosol decay was measured by the classical light scattering technique using the Malvern 2600 and Mastersizer-X particle sizers. A standard 300mm lens was used to measure the fine spray and a novel composite Fresnel lens was also employed to combat vignetting due to the large dimensions of the rig. Experimental results with water showed an initial D_3.2 of between 15 and 32µm. The kerosene tests showed an initial D_3.2 of about 5.6µm which fell very quickly to about 2.8µm, and the resulting aerosol took about 200 seconds longer to fully disperse than a water aerosol created under the same conditions. This difference in dropsize arose because kerosene and water have a surface tension of 0.026 Nm^-l and 0.072 Nm^-l respectively, hence kerosene is atomised much more easily than water. The results from the Mastersizer-X show the aerosol development and decay, and its new features allow many measurements to be made over a short time interval. The aerosol decay was also modelled using FLUENT to predict the effects of the purges on the droplet fall-out, and size distribution over time.