Sampling of Particulate Material

Abstract

Particulate materials are often produced in very large quantities. In contrast, usually small quantities are used for characterization or applications. There are two reasons that small samples often give poor representation of the average quality of the bulk. One reason is the discrete nature of particles. This necessitates a minimum amount of sample to reach a desired variance for a stated PSD parameter. The second reason is the fact that both process fluctuations and segregation during transport and storage may lead to large differences in quality at different locations in a product lot. This necessitates analysis of several individual or combined sample increments, taken at different times during production or at different locations in the lot. Thus, adequate care is required for sampling of bulk powders in order to obtain a known and acceptable maximum error between lot and sample. In practice, often three different stages of sampling can be discriminated: primary sampling of several sample increments from the bulk product, which are then blended to a composite primary sample, secondary sample splitting in the laboratory and tertiary sampling within the measurement instrument.