(563b) Continuation of Experts In Discrete Element Modeling (EDEM) Validation Study for Packed Bed Structural Properties

Experts in Discrete Element Modeling (EDEM)¹ is a discrete element method (DEM) based modeling software program, capable of providing a wealth of information about granular solid flows occurring in many industrial applications. The DEM is a numerical approach used to compute the motion of large number of particles^2.A computational study involving the application of EDEM to a pebble bed nuclear reactor (PBR) application is being performed at Missouri S&T. Prior to full involvement with EDEM based numerical simulations, an effort has been made to validate the packing algorithm used in EDEM; particularly for cylindrical packed bed applications. This is essential, as the local flow and transport properties are closely coupled with the structural characteristics of a packed bed which in turn depends upon the packing of the particles inside the container. Also, the packing of particles inside the container is the first and main step in EDEM based analysis. Packing algorithms available with commercial codes such as EDEM are used as a ‘Black-Box’ and are without any detailed validation exercise. In most cases, average porosity results are compared to benchmark the numerical packing results with available experimental results, which is not sufficient. Detailed technical information about the packing algorithm used in EDEM software is not available in open literature and hence there is a need to perform validation studies of the packing algorithm used in EDEM. The packing algorithm used in EDEM demands accurate input of elasticity (material properties) and frictional (interaction properties) parameters which are not readily available in the literature for materials of interest. The radial porosity profile, along with average porosity values, will be used as a means for this validation study. This study also involves a parametric sensitivity study of material and interaction properties on structural properties of cylindrical packed beds. This sensitivity study will highlight important material and interaction properties which need to be provided as an input from a reliable simulation point of view.  Also, a DEM based simulation study will optimize the values of material and interaction properties by mimicking an experimentally obtained radial porosity variation profile to a larger extent. Hence, obtained results will provide a new and better understanding about the packing algorithm used in EDEM as well as recommendations to simulate realistic packed bed structures using EDEM. The cylindrical packed bed has been chosen as a model for this validation study. There is much benchmark data available for packed bed structural properties from previous experimental measurements³ and from high-fidelity numerical packing algorithms4 for a wide range of aspect ratios. Also, the slow and dense granular flow encountered in a PBR can be approximated by static packed beds⁵. The various structural properties of cylindrical packed beds which can be used for this validation study are mean porosity, the distribution of particle centers, and the axially averaged radial porosity profile which exhibits a typical shape of damped oscillation. The radial distribution of particle centers serves as an input to Computational Fluid Dynamics (CFD) analysis of packed beds in order to simulate them more accurately⁶. Radial and axial porosity profiles and associated solid phase distributions are required as an input to hydrodynamic and thermal models of packed beds⁷. It has already been demonstrated in our previous work that the friction between wall and particle plays an important role in the structural characterization of packed beds⁸ and neglecting this friction results in simulations of tightly packed bed which are unreal. 

DESCRIPTION OF THE ACTUAL WORK 

A series of EDEM simulations is being carried out to perform a parametric sensitivity study to check the effect of static friction between particles. The Hertz-Mindlin (with no-slip) contact model⁹ has been used to model particle-particle and container wall-particle contact forces. Higher values of static friction coefficient between particles damps out the porosity variation profile quickly while moving towards the center. Neglecting friction between particles under predicts porosity variation near the container wall, resulting in tightly packed beds which are evident from mean porosity values.  The effect of other material and interaction properties on the structural properties of packed beds  needs to be investigated further. Such study might help in the development of a mechanistic model for the radial porosity variation of packed beds which will require inputs of material properties besides the aspect ratio.

Acknowledgements

The authors acknowledge the financial support provided by Department of Energy (DOE) Nuclear Energy Research Initiative (NERI) project (NERI-08-043).

REFERENCES

[1] EDEM™ User Manual, DEM Solutions (USA) Inc., Edinburgh, Scotland (2010). 

[2] P.A. CUNDALL, O.D.L.STRACK, A discrete numerical model for granular assemblies. Geotechnique, 29:47–65, 1979.

[3] G. MUELLER, Radial void fraction distributions in randomly packed fixed beds of uniformly sized spheres in cylindrical containers, Powder technology,72,pp.269-275, 1992.

[4] G. MUELLER, Numerically packing spheres in cylinders, Powder technology, 159, pp.105-110, 2005.

[5] C. du TOIT, The Numerical Determination of the Variation on the Porosity of the Pebble-Bed Core,  Proceedings of the conference on High-Temperature Reactors, Petten, The Netherlands, European Nuclear Society, 2002.

[6] N.J.  MARIANI, W.I. SALVAT, A. CAMPESI, , G.F. BARRETO,  O.M. MARTINEZ, Evaluation of structural properties of cylindrical packed beds using numerical simulations and tomographic experiments, International Journal of Chemical Reactor Engineering 7, Art. No. A82.,2009.

[7]A.J. SEDERMAN, P.ALEXANDER and L.F. GLADDEN, Structure of packed beds probed by Magnetic Resonance Imaging, Powder Technology, 117, pp. 255-269, 2001.

[8] V. KHANE, G.E. MUELLER, and M. AL-DAHHAN, Experts in Discrete Element Modeling (EDEM) Validation for Packed Bed Structural Properties, AIChE-2010 Annual Meeting, Salt Lake City, Utah, November 7-12, 2010.

[9] Y. TSUJI, T. TANAKA and T. ISHIDA, Lagrangian numerical simulation of plug flow of cohesion- less particles in a horizontal pipe, Powder Technology, 71 239-250, 1992.

[#] A.G. POLSON, Distinct element method analyses of fuel spheres in the PBMR core using RFC3D, Proceedings of the 2004 International Congress on Advances in Nuclear Power Plants, ICAPP’04, June 13-17, 2004.