Search

The effectiveness of hot-melt coating depends on its uniformity and the extent to which the surface is completely covered. Compared with solvent-based coating, spreading is more limited in hot-melt coating; thus, the coating uniformity is more affected by the process parameters. This study presents a new method for identifying and quantifying factors influencing coating uniformity. The proposed method facilitates the determination of coating-thickness distribution and non-covered surface proportion based on micro-computed tomography measurements. The proposed method is based on particles that have undergone hot-melt coating in a fluidized bed, and it is compared to common methods for layer-thickness determination. The influencing factors are quantified in terms of the dependency of coating uniformity on the coating amount and material. Material properties have a significant impact because stearin and palm fat create different coating layers. The proposed method is confirmed to be well suited for analyzing coating qualities.

While particulate products are often characterized by their median diameter or the width of the particle size distribution, information is rarely given about the agglomeration degree of the product. To obtain this information, a tool combining image analysis and discriminant factorial analysis (DFA) was introduced in previous works. The accuracy of that method depended on the number of image descriptors selected, i.e. measurements describing each particle: few image descriptors resulted in rather poor classification while too many lead to an overfitting of the data. The aim of this study is twofold: First, we want to compare the classification accuracy of artificial neural networks (ANN) and DFA which, contrary to ANN, forms linear classifiers. Second, we want to provide an easy-to-implement procedure for generating particle classifiers. We used a qualitative measure called Proportional Similarity to test whether a subset selection of image descriptors was necessary to avoid an overfitting. The influence of the training set size was investigated as well as the transferability of the classifier on data obtained under different experimental conditions. The chemical systems used were l -alanine/water and adipic acid/water and the classes considered were single crystals, agglomerates, and gas bubbles. The results show that an ANN classifier provides higher accuracy and is more effective when only few image descriptors are available while DFA is simpler to create. Moreover, we show good transferability of classifiers trained on data of different experimental conditions. Based on our results, we provide guidelines for classification of particulate systems.

In order to fully characterize crystal aggregates, the orientation of primary particles has to be analyzed. A procedure for extracting this information from three-dimensional microcomputed tomography ( μ CT) images was recently published by our group. We here extend this method for asymmetrical crystals and apply it for studying the disorientation angle distribution of four potash alum crystal samples that were obtained under various experimental conditions. The results show that for all considered supersaturation profiles, primary particle pairs tend to have the same orientation significantly more often than in theoretical considerations, in which the orientations of primary particles are assumed to be distributed randomly.

The degree of agglomeration and the aggregate shape influence the quality of crystalline products and the ease of downstream processing. Studying the shape of primary particles in an aggregate can lead to a better understanding of the underlying aggregation mechanism. We present an automatic image processing procedure for identifying the shape, size, and position of each primary particle in microcomputed tomography (μCT) images of potash alum aggregates. Splitting an aggregate into primary particles is based on recombining watershed-transform regions, where concavity points are considered as indicators of correct segmentation. The shape identification algorithm uses the Hough transform to identify visible face normals and matches them to the set of face normals defined by a crystal model. In principle, the algorithm is applicable to other crystalline compounds provided that sufficient symmetry is present to determine the shape of a primary particle from its visible part.