CFD-DEM study in the 20L explosion sphere

CFD-DEM study of the influence of the particle clump shape on the degree of agglomeration and de-agglomeration in the 20L explosion sphere

The 20L sphere has become the standard device for the characterization of explosion parameters associated with flammable dust. One concern about this device is related with the variation of particle size due to particles agglomeration and de-agglomeration during the dispersion of the dust, this phenomenon can lead to the underestimation of the severity of dust explosions as this tends to increase when the dust elementary particles size is reduced. As a consequence, this could induce a bad design of process equipment. In order to assess the standard test based on the degree of agglomeration and de-agglomeration for different particle clumps shapes, a set of CFD-DEM simulations of the starch dust-air dispersion within the 20 L sphere were developed in  STAR-CCM+®.

The air was defined using a eulerian approach. The dust particle was modeled using the Discrete Element Method. Furthermore, the DEM particles were define as particle clumps that can be elastically deformed and can be broken. The Bonded Particle model was used to calculate the forces between spheres. Additional models like Linear Cohesion Model and Simple Failure Model were used.

The results shows fast and simultaneous de-agglomeration and agglomeration process after dispersion despite the particle clumps shape used. However, the degree of de-agglomeration and agglomeration differ for each particle clumps shape. These results allow our team to adapt the standard ignition time according the dust characteristics in order to reduce the mischaracterization of the severity of dust explosions.

Presenter(s): 
Universidad de los Andes
Mariangel Amin
Universidad de los Andes
Andres Pinilla
Session Time Slot(s): 
Time: 
06/03/2017 - 4:40pm-06/03/2017 - 5:05pm
Room: 
Room I
Presentation Image: 
Session Track: 

Session Tracks (STAR Global Conference 2017)

Co-Author(s): 
First Name: 
Andres
Last Name: 
Pinilla
Company / Institution: 
Universidad de los Andes
First Name: 
Carlos
Last Name: 
Murillo
Company / Institution: 
Université de Lorraine
First Name: 
David
Last Name: 
Torrado
Company / Institution: 
Université de Lorraine
First Name: 
Nicolás
Last Name: 
Ratkovich
Company / Institution: 
Universidad de los Andes
First Name: 
Olivier
Last Name: 
Dufaud
Company / Institution: 
Université de Lorraine
First Name: 
Felipe
Last Name: 
Muñoz
Company / Institution: 
Universidad de los Andes