Validation of a Discrete Element Code using Magnetic Resonance Measurements

  • Mr Christoph Muller, University of Cambridge, Department of Chemical Engineering, United Kingdom
  • Dr Dan Holland, University of Cambridge, Department of Chemical Engineering, United Kingdom
  • Dr Andy Sederman, University of Cambridge, Department of Chemical Engineering, United Kingdom
  • Dr John Dennis, University of Cambridge, Department of Chemical Engineering, United Kingdom
  • Dr Stuart Scott, University of Cambridge, Department of Engineering, United Kingdom
  • Prof Lynn Gladden, University of Cambridge, Department of Chemical Engineering, United Kingdom

    • In the Discrete Element approach to simulate two-phase granular systems each particle is tracked as a single entity. Collisions between particles are detected and the resulting forces are calculated using contact mechanics. The position and velocity of each particle is then calculated using Newton’s second law. In addition, the gas phase interacts with the particles via drag forces. Discrete Element Modelling (DEM) is now established as a sophisticated modelling technique for two-phase granular systems. However, it is intrinsically difficult to obtain experimental measurements of these systems due to their complexity, e.g. evolution over rapid time-scales, visually opaque, etc.

    Recently, Magnetic Resonance (MR) imaging has been developed for use in two-phase granular systems achieving measurements of high temporal and spatial resolution. MR measurements of the voidage have been used here to validate an in-house DEM code using a small 2D-fluidised bed with width, transverse thickness and height of 44×10×1000 mm. Particles of diameter 1.2 mm and density 1000 kg/m3, supported on a porous frit distributor were fluidised by air. To evaluate the DEM code several “input” parameters, i.e. the coefficient of friction, the thickness of the bed that was simulated, the coefficient of restitution and the applied drag laws have been varied. The MR measurements show fair agreement with the DEM simulations.