Sampling particulate matter for analysis - Controlling uncertainty and bias using the theory of sampling

The Grouping and Segregation Error in the Rice Experiment

The grouping and segregation error, one of the seven sampling errors defined by Pierre Gy, is related to the combined effects of gravity and characteristics of the target analyte such as particle size, density, shape, and moisture content of the particulate materials being sampled. Kinetic energy acting on particulate materials that are moved, flow, transported, or stockpiled causes the spatial distribution of fragments relative to one another to change. The grouping and segregation error is identified, quantified, and measured in relative sampling variance terms by comparing the sampling variability due to fractional shovelling (scooping) with that using a Jones riffle splitter. The relative sampling variance of low concentrations, approximately 0.01%, of steel balls, lead balls, and flakes of tungsten carbide in the host substrate indicates that, in this specific sampling space, the grouping and segregation error is primarily a function of particle density. Conclusions from the experiments are that components of the grouping and segregation error, namely the grouping factor and segregation factor, can be identified, measured, and mitigated. Whereas the grouping and segregation error has historically been considered to be less than the fundamental sampling error, these experiments suggest that it can be up to four times the fundamental sampling error depending on the density of the segregated materials.