Effect of shock strength on dust entrainment behind a moving shock wave

Abstract

Secondary dust explosion is a serious industrial issue because it occurs under conditions corresponding to an increased quantity and concentration of dispersed, combustible dust when compared with the primary explosion. The problems of lifting and dispersion of a dust layer behind a propagating shock wave must therefore be understood to ensure safety regarding secondary dust explosion hazards. Using a new shock-tube facility for studying shock propagation over dust layers, limestone dust was subjected to Mach numbers ranging from 1.10 to 1.60. A shadowgraph technique was applied by using a high-speed camera (15,000 fps) for visualization of the dust-layer height change behind the moving shock wave. Also, the effect of dust-layer thickness on the entrainment process was observed by performing tests with two different layer depths, namely 3.2- and 12.7-mm thicknesses. New correlations were developed between the shock strength and the dust entrainment height as a function of time for each layer depth. In general, the results herein are in agreement with trends found in previous work, where there is a linear relationship between dust growth rate and shock Mach number at early times after shock passage. Also, new data were collected for image analyses over longer periods, where the longer observation time and higher camera framing rates led to the discovery of trends not previously observed by earlier studies, namely a clear transition time between the early, linear growth regime and a much-slower average growth regime. This second regime is however accompanied by surface instabilities that can lead to a much larger variation in the edge of the dust layer than seen in the early growth regime. In addition, for the linear growth regime, there was no significant difference in the dust-layer height growth between the two layer thicknesses; however, the larger thickness led to higher growth rates and much larger surface instabilities at later times.