Open AccessDetermining the degree of powder homogeneity using PC-based program
Author(s) -
M Olivera Djuragic,
Radmilo Čolović,
D Jovanka Levic,
A Slavica Sredanovic,
Aleksandar Selakov
Publication year - 2010
Publication title -
hemijska industrija
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.147
H-Index - 19
eISSN - 2217-7426
pISSN - 0367-598X
DOI - 10.2298/hemind100606044d
Subject(s) - homogeneity (statistics) , materials science , tracer , filter paper , particle size , analytical chemistry (journal) , process engineering , chromatography , computer science , chemistry , physics , nuclear physics , machine learning , engineering
The mixing of powders and the quality control of the obtained mixtures are critical operations involved in the processing of granular materials in chemical, metallurgical, food and pharmaceutical industries. Studies on mixing efficiency and the time needed for achieving homogeneity in the powder mashes production have significant importance. Depending on the characteristic of the materials, a number of methods have been used for the homogeneity tests. Very often, the degree of mixing has been determined by analyzing images of particle arrays in the sample using microscopy, photography and/or video tools. In this paper, a new PC-based method for determining the number of particles in the powder homogeneity tests has been developed. Microtracers®, red iron particles, were used as external tracer added before mixing. Iron particles in the samples of the mixtures were separated by rotary magnet and spread onto a filter paper. The filter paper was sprayed with 50% solution of ethanol for color development and the particles counted where the number of spots presented the concentration of added tracer. The number of spots was counted manually, as well as by the developed PC program. The program which analyzes scanned filter papers with spots is based on digital image analyses, where red spots were converted through few filters into a black and white, and counted. Results obtained by manual and PC counting were compared. A high correlation was established between the two counting methods