(440f) DEM Simulations of “Dry Cohesion” Effects in Powder Compaction

Powder compaction is widely used in industries such as pharmaceutical, food, catalyst, ceramic and automotive to manufacture products with complex geometries and high strength. In past few decades, various numerical techniques to simulate the compaction process have proven to be useful for obtaining insights into the microstructures and the deformation behavior of the granular compact. However, almost no work has been reported on the effect of “dry cohesion” (which is much more commonly found in real pharmaceutical materials) on compaction or tableting. In this work, we focus on the effect of cohesion and tableting speed on compression of both monodisperse and polydisperse granular systems. A three dimensional discrete element model (DEM) which incorporates static and dynamic friction is used in this study to simulate die filling, compaction and decompaction of cohesive granular system as in a simulated tablet press (a.k.a. “pelletizer”). The simulation includes discharge of the particles through a slit similar to the discharge port of commercial feeder systems, flow into the die, and packing of particles under gravity. The magnitude of the cohesive force is represented in terms of a parameter K = Fcohes/mg, (K is usually known as the “bond number” and is the measure of cohesiveness). The value of K explored ranges from 0 (free flowing case) to 75 (cohesive system).

Before compaction, a pre-determined volume of granular material is filled into a die for both monodisperse and polydisperse cohesive systems. It is observed that the cohesive forces hinder the mobility of the particles while flowing into the die and the time required to fill the die is strongly dependent on the cohesion of the material. This is in good qualitative agreement with the common observation that cohesion increases the weight variability of pharmaceutical tablets, and that this variability increases with the compression speed. In this deposition step, a low density, highly porous microstructure is established within the die. Thereafter, the simulated compaction cycle mimics the procedure employed in industrial tableting process. The material is compressed by the downward movement of the upper punch at prescribed velocities of 0.05m/s, 0.2m/s and 0.6m/s, which capture the range normally used in commercial applications. This compression step is followed by an upward retreat of the upper punch (the decompaction process). In this first study, the lower punch is kept stationary during both the compaction and decompaction process. Force-displacement curves are used to characterize the compression and deformation properties of the materials and are obtained by measuring the force on the upper punch as a function of its vertical displacement. In a large number of experimental studies, force-displacement curves have been used to characterize the compression properties and to show correlation between net work and tablet strength. Our results show that a considerable more energy is needed to compress cohesive material as compared to free flowing material. The effect of the compaction speed is also quantified and it is observed that the energy for tableting process is directly proportional to the upper punch speed, irrespective of the degree of cohesion of the systems. Compression force required for monodisperse systems is larger than that for polydisperse systems. This effect is due to the “crystallization” (emergence of large-scale microstructural order) in monodisperse systems during the compaction process.