Abstract
The overriding aim of this project is to gain an understanding of the flowability of powders. Industrial interest for this study was provided by Centura Foods, U.K. who supplied three samples for study. These powder samples are starch, maltodextrin and hydrolysed vegetable protein (HVP). The first half of this project addresses the influence of adsorbed moisture on the flowability of these powders with strong emphasis on the value of cohesion. There is evidence in the literature suggesting that the prevailing relative humidity has an influence on powder cohesion. The general trend indicates an increase in cohesion with an increase in relative humidity. Some attempts have been made in the literature to relate this humidity dependence to powder properties notably the propensity for water sorption. Material characterisation studies revealed the morphological and surface characteristics of the samples through Scanning Electron Microscopy (SEM). It revealed the angular and amorphous appearances of the maltodextrin and HVP particles respectively. The starch particles appear to have distinct groups of larger and smaller particles. The larger particles are ellipsoidal and the smaller particles are near perfectly spherical. Water sorption studies using a gravimetric analyzer showed that the starch and maltodextrin powders had on average 19 and 25 % mass increase respectively when subjected to a relative humidity change from 0 % to 90 %. However, the HVP powders had the largest percentage mass uptake of over 100 %. This was supported by in-situ humidity microscopy results that showed significant swelling with dissolution when exposed to saturated air. Comparatively but to a lesser degree, similar behaviour was observed for maltodextrin powders. Starch particles also exhibited swelling under saturation conditions. Powder flowability studies were carried out using a Jenike-type shear cell. Commissioning of the shear cell was conducted using limestone powders. Measurements were carried out at dry, ambient and saturated conditions. This was facilitated by a newly designed in-situ humidity control system. Values of cohesion were obtained by constructing a family of yield loci for each material. HVP powders showed a significant increase in cohesion with humidity. Maltodextrin and starch powders showed no significant change in cohesion with humidity. However, both powders were more difficult to stir with changes to the relative humidity thereby highlighting that the shear cell in this study is not sensitive enough to detect the changes in fiowability. The phenomenon of stick-slip failure was observed for starch at all humidities, and for maltodextrin but only at 0 %RH. Published information implies that stick-slip behaviour is linked to the rate of stored elastic energy and frictional dissipation. A study of stick-slip observed with starch powder was conducted under four levels of consolidation over four different velocities. A mathematical model has been applied which describes the various stages of stick-slip and consequently identifying the parameters that influence stick-slip. A comparison of the model with experimental values highlighted the effectiveness of the model. Results from this study identified the levels of consolidation and drive velocities as the governing parameters influencing stick-slip, with the level of consolidation argued to have a greater impact. It was anticipated that the study would identify conditions to eliminate stick-slip. Although this was not achieved, the study however presents O, a measureable which represents the extent of stick-slip. On this basis, stick-slip is argued to be minimized under a combination of high consolidation loads and velocities. The study has identified possible areas for future work. Firstly, some irregular trends were observed when comparing some results under the two intermediate consolidation levels (4 kN/m2 and 12.5 kN/m2). It is possible to further investigate stick-slip failure with smaller increments to the level of consolidation. It is possible to add to the work done in this study by investigating the effects of higher consolidation and velocities and testing on other materials. Finally, a parallel study repeating the work done in this study (and incorporating the proposed ideas for future work) using a different type of shear tester remains a good candidate for research.