Particle size-dependent viscosity behavior of a suspension using image processing

Abstract

In this work, interparticle distances for different powder particle sizes were determined using an image-processing method. Considering the theoretical model only available under several assumptions, this method enables calculation of the interparticle distance regardless of those assumptions. The method is based on the intensity difference between a powder and binder system, and enables the distinction of powder particles from the binder system. A mixed suspension composed of 17-4PH stainless steel powder and a wax-based polymeric binder system was polished with acetone to observe the powder distribution. The boundaries and centroids for the powder particles in the binder system detected by the image-processing method were employed to calculate the mean interparticle distance of a randomly selected particle. To obtain a representative value of interparticle distance for a given feedstock, the Monte Carlo method was used. Although the interparticle variation converged to 0 with repetitive iteration, outlier points satisfied the limitations to determine the iteration number. To overcome this problem induced by the outlier points, the summation of five points of interparticle variation as the limitation was considered. As the iteration number was based on the defined limitation, the interparticle distances for suspensions of the different-size-distribution powders were naturally obtained. With these obtained interparticle distances, the relationship between viscosity and powder particle size was analyzed. The powder characteristics in the viscosity model were analyzed in terms of the interparticle distance. The developed image-processing method to quantify the interparticle distances for powder particle sizes can enhance understanding of physical phenomena in various industrial applications, because the interparticle distance enables quantitative analysis of rheological characteristics such as viscosity or powder agglomeration. (C) 2018 Elsevier B.V. All rights reserved.