Recognizing heterogeneous distribution of platinum group elements (PGE) in geological materials by means of the Re-Os isotope system

The identification of uncertainties caused by sample inhomogeneity, as distinct from those caused by sample preparation and measurement, is a challenging task. Use of chemometric methods to separate and estimate these contributions to the combined standard uncertainty of a measurement (uc) of an analytical result requires complex experiments. The difficulty of platinum group element measurement makes this task even more complex. But unless it can be demonstrated that sample inhomogeneity is the major contributor to the high variability of an analytical result one should be careful not to mistakenly attribute this to a nugget effect. In this contribution we are able to demonstrate in two special cases that irreproducible results (up to 90% RSD) for analysis of Os and Re in the pg g(-1) to ng g(-1) range are truly caused by a nugget effect and not by inadequacies of the analytical method.