(64ac) Particle Size Reduction on the 20L Sphere, a Computational Approach | AIChE

(64ac) Particle Size Reduction on the 20L Sphere, a Computational Approach

Authors 

Pinilla, J. - Presenter, Universidad de los Andes
Rios, N., Universidad de los Andes
Amin, M., Universidad de los Andes
Dufaud, O., Université de Lorraine, Reaction and Chemical Engineering Laboratory (LRGP)
Monnier, N., Université de Lorraine, Reaction and Chemical Engineering Laboratory (LRGP)
Dust explosions might occur when a combustible dust is dispersed near an ignition source in a confined facility.  This means that this kind of event is a main concern for any industry that handle or store metallic or organic particle material as part of their process. Nowadays, the 20L sphere has become the standard device for characterizing parameters associated with flammable dust explosions like the maximum pressure (Pmax), the maximum rate of pressure rise (dP/dt max), the minimum explosive concentration (MEC), and the deflagration index (Kst). The 20L sphere contains an explosion chamber, a dispersion system, an ignition system pressure measurement and an automatic control system. For several years, there have been generated some questions about the conditions in which the standard test is performed in order evaluate some parameters. One of those questions is associated with a reduction of the particle size due to fragmentation phenomenon in the 20 L sphere. When this type of phenomenon occur, a mischaracterization of the flammable dust takes place, as the Pmax and  dP/dt max tends to radically increase with particle size reduction. Also, depending of the cohesive characteristics of each material, agglomeration phenomenon may also occur, increasing the particle size.  According to the above, this study aims to assess the agglomeration and fragmentation phenomenon during the 20L standard test. For this purpose a CFD simulation was developed, where the particle interaction was modeled using DEM, and cohesion models were implemented to represent agglomeration and fragmentation of particles. Four different time frames were defined for the study: an initial time (1 ms), high turbulence time (20 ms), the explosion time according to ASTM standard (60 ms) and a time of low turbulence (100 ms). Also, for each time five different areas were stablished to evaluate and quantify the fragmentation and agglomeration. Each area was defined according to the particles flow around the sphere, which was study in a previous work.  An analysis of the 20L sphere standard test was presented. The phenomenon of fragmentation and agglomeration in dust explosions have an important effect on the severity of the explosion because it affects particle size distribution. This implies that values such as the MEC and Kst, calculated during the test, are related to different conditions than the ones needed. The results allowed to determine the degree of agglomeration and fragmentation in the different areas. It was obtained that the areas with high turbulence presented a higher degree of fragmentation than the areas with low turbulence.