Breakage of Agglomerates: Attrition, Abrasion and Compression

In many industries particulate solids are handled in different forms. When producing particles breakage is an important wanted, in the case of grinding, or unwanted phenomenon. Granules often consist of more than one component and multiple phases. This means that granules are strongly anisotropic and inhomogeneous which makes mechanical characterization difficult, so the strength is difficult to measure. The objective of this thesis is to develop and validate new particle breakage testers from an industrial point of view. The philosophy behind the breakage testers is to isolate the forces that cause damage to the particles. This requires repeated stress cycles to be transfered to the test particles. To validate the improved attrition tester with an industrial application, the damage per unit energy is compared to the damage caused during dilute phase pneumatic conveying. Wear observed during Pneumatic conveying correlates well with results from the improved attrition tester for particles having surface asperities. But only during the so called rounding off phase. The asperities act as a fracture buffer absorbing energy from the impact. First the attrition process removes asperities. Then fracture occurs. Besides the validation of the attrition tester a new breakage tester testing tangential forces is developed in this work. The abrasion tester, is based on a particle box following a planetary motion. This motion causes the particles to rotate continuously over the particle box wall. The particle wear during abrasion testing, expressed in mass loss over transferred kinetic energy is in the same order as during attrition testing. This is explained by the same efficiency at which the kinetic energy is converted into particle deformation. The kinetic energy is converted into newly created surface energy. And the efficiency at which this occurs is a particle property. The major difference between abrasion and attrition testing is the absence of the accelerated removal of surface asperities. A qualitative difference between abrasion and attrition testing is also presented. Abrasion is characterized by the gentle polishing of the particles. Whereas attrition testing results in the removal of the asperities and "craters" are visible of the impact sites of the particle with the wall. Often granules are coated with a water soluble polymeric coating in order to improve the mechanical strength. The question arises what polymeric properties are important to form a mechanically strong coating on a particle? The work of Rowe is cited about the adhesion of the coatings to the particles. Rowe found an inverse relation between coating thickness and film adhesion. Indicating that a thick coating does not necessarily leads to stronger granules. To study the influence of different coating polymers on the overall particle strength, particle formulations are made using fluidised bed coating. The formulated particles are then tested using the developed attrition, abrasion and compression tester. Three attrition sub regimes are identified, the peeling, erosion and layer fatigue modes. The conclusion is that for an elastic deforming brittle polymer, the layer fatigue attrition sub mechanism is expected. Applying a plastic deforming coating leads to a reservoir system where the fragments are held together by the plastic deforming coating. A plastic deforming coating on sodium benzoate shows the erosion mechanism during attrition testing. The coated granules are abrasion tested. A linear relation between remaining mass and transfered energy is observed. During abrasion testing it was not possible to remove all the coating completely. Apparently the rate of strain in the abrasion tester is not high enough for the coatings to be removed. Applying higher strain rates by increasing the centripetal force could overcome this. At this moment it is mechanically not possible. Also the coated particles are much stronger towards abrasion testing than to attrition testing. This is explained by the perceived thickness of the coating. The compression strength of coated granules is not easy to quantify since there is a large spread in the observed critical stress. This large spread is due to the heterogeneous nature of the granules. Weibull statistics are successfully applied to statistically describe the observed strength of the coated granules. Using Weibull statistics a ranking can be made between the coatings and the compression strength. A plastic deforming coating tends to hold the particle together as a reservoir upon compression. An elastic deforming coating has a higher initial strength without regions of plastic deformation. Individual failure of different layers is observed. This compares also to the layer fatigue attrition mechanism where initially no attrition is observed but after a while the granules start to fail