(342b) Extraction of Aggregation Kernel from DEM Simulations Using Sustained Contact: A New Kernel Extraction Methodology for Drum Granulators

Drum granulator is a popular granulation equipment whose modeling is usually conducted using the population balance model. The population balance model requires the knowledge of breakage, aggregation, growth and nucleation kernels. Such kernels are either developed through phenomenological models [1] or inverse problems [2]. Additionally, DEM simulations can be used to model these kernels [3]. While numerous studies exist on deriving aggregation kernels from DEM simulations [4, 5, 6], they predominantly rely on relative velocity, specifically, the collisional velocity between particle pairs, to obtain these kernels. However, these methodologies cannot capture the physics of sustained contacts and resulting liquid/solid bridge formation. Hence a reliable method for kernel extraction is needed for systems where aggregation happens due to sustained contacts. In this work, we demonstrate a new method for the extraction of aggregation kernels using information about sustained contacts.

In the proposed approach, aggregation kernels are obtained as a time snapshot by analyzing the particle locations over a certain number of time steps (nΔt) around that particular time. The location of the particles provides the duration and average overlap and a few additional information on the contact (such as the number of attachments/detachments). Subsequently, this information is processed through a series of physically grounded criteria to identify contacts likely to result in successful aggregation, thus determining the aggregation efficiency. Finally, the obtained information is translated into the aggregation kernel by using appropriate expressions. This simple contact information-based kernel extraction methodology successfully reproduces several established kernels. To illustrate, we apply this methodology to extract the Brownian kernel (see Figure 1). After validating our kernel extraction method for a few additional known kernels, we extract the aggregation kernel for the case of a rotating drum containing wet particles. We then compare it with the kernels used in literature for granulation in case of dense flows.

References :

[1] Ramkrishna, D. (2000), Population balances: Theory and applications to particulate systems in engineering, Elsevier.

[2] Ramkrishna, D. (2000), Inverse Problems in Population Balances. In Population balances: Theory and applications to particulate systems in engineering (pp. 221-273), Elsevier.

[3] Gantt, J. A., Cameron, I. T., Litster, J. D. & Gatzke, E. P. (2006), ‘Determination of coalescence kernels for high-shear granulation using dem simulations’, Powder Technology 170(2), 53–63.

[4] Gantt, J.A. and Gatzke E. P. (2006), ‘ A stochastic technique for multidimensional granulation modeling’, AIChE Journal, 52(9):3067 – 3077. doi: 10.1002/aic.10911.

[5] Reinhold, A. & Briesen, H. (2012), ‘Numerical behavior of a multiscale aggregation modelâA˘Tcoupling population balances and discrete element models’, ˇ Chemical engineering science 70, 165–17

[6] Barrasso, D. & Ramachandran, R. (2015), ‘Multi-scale modeling of granulation processes: bi-directional coupling of pbm with dem via collision frequencies’, Chemical Engineering Research and Design 93, 304–317