Formation of granule nuclei under static and dynamic powder bed conditions

Andrew C. S Lee, Purdue University

Abstract

This thesis reports results from a combined experimental/computational investigation into initial falling drop-powder bed interactions and subsequent formation of a liquid-powder agglomerate called a nucleus. Formation of a nucleus is an important process that occurs commonly in wet granulation. Experimental results start with the influence of impact conditions (reported in terms of impact Weber, We, and Reynolds, Re, numbers) on nucleus formation for the case of a single drop striking a static glass bead bed. Results from high speed images showed that the nucleation rate is not influenced by liquid physical properties (e.g. density, surface tension, and viscosity) for drops that spread significantly (62≤ We≤233). However, results also showed that nucleus size is determined by how much liquid penetrates into the bed during drop spreading, so liquid physical properties do play an important role in that aspect. A corresponding numerical model, derived form first principles, predicts nucleus size to within 1.5% using only liquid physical and powder bed properties, plus the experimentally measured drop spreading behavior. In subsequent experimental work single drops fell onto the surface of a flowing heap and the effects of powder bed motion on nucleation were investigated. For 62≤We≤233, initial drop spreading and nucleation appeared almost identical for the dynamic and static bed cases regardless of the speed of bed surface. The numerical model developed for the static bed was applied to the dynamic bed results; model accuracy was noticeably poorer. The disagreement was partly attributed to the possibility of the nucleus retraction process being impeded by the continuous powder bed motion under dynamic bed conditions, which could make a nucleus appear larger when viewed from the top.

Degree

Ph.D.

Advisors

Sojka, Purdue University.

Subject Area

Chemical engineering|Mechanical engineering

COinS