(28d) "Perfect Particles" - 3D Printing of Tuneable Agglomerates for Validation of DEM Breakage Models

One of the long term barriers to advanced and accurate DEM modelling of agglomerate breakage is the lack of a suitable set of test particles that can be used to validate DEM models. Generally the approach has been to take a specific, simplified particle system, measure the relevant properties as accurately as possible, and compare and fit the model to experimental measurements. There main limitation of this is approach is that agglomerate breakage is a destructive experimental tests, and so the experiment represents the date from a single individual agglomerate in a single test condition. Ideally, what we want is the ability to create hundreds of identical agglomerates, with systematic strength and structure variations, and to test each set of agglomerates under controlled variations. This has historically not been possible via experimental agglomeration production techniques.

3D printing offers the ability to design and produce test agglomerates with controlled properties to replicate simple or complex agglomerate structures, and to use these agglomerates to validate DEM models of agglomerate breakage (Ge et al, Powder Technology 2017). In this work, agglomerates with well-defined structure (tetrahedral and random structures, with varied porosity) and tuneable bond strength properties were designed in CAD and EDEM. A Stratasys Object500 polyjet printer was used to 3D print multiple, identical copies of each agglomerate design, and quasi-static breakage tests were then conducted to vary the strain rate and examine the breakage behaviour. The macroscopic agglomerate breakage experiments were compared with DEM simulations, which used Timoshenko Beam Bond Model (TBBM) contact model recently implemented by Brown et al (Granular Matter, 2014).

Qualitative comparisons of the failure mode between experimental tests and DEM simulations (using TBBM bond model) were successful. The quantitative load-displacement results shown that DEM simulation was an excellent match for the cubic tetrahedral agglomerate structure. However, DEM underestimated compressive loads for spherical random agglomerates. However, the results show that the approach is a significant step forward in the ability to "calibrate" and "validate" DEM models of agglomerate breakage.