Search

The measurement uncertainty (MU) in environmental analyses is usually considered as a combination of the additive contributions from sampling, sample preparation and analytical measurement (usually instrumental). The target MU for such studies is usually relatively high due to the natural sample heterogeneity and/or the complicated pretreatment procedure required. In the analytical practice, the contribution of the sample preparation to MU is rarely evaluated. Thus, it remains a hidden part of the one related to the analysis. However, the knowledge for the contribution of sample preparation might provide important information and further possibility for optimization of the entire analytical procedure and reduction of the expanded MU of the analytical result. From statistical point of view, the separation of uncertainty contributions from the different steps is, generally, not a trivial task. Selection of the proper statistical approach depends on the data structure and quality, variables distribution, etc. In the present study, three different statistical methods for evaluation of the uncertainty contribution of the sample preparation were applied, compared, and discussed. The considered approaches are based on, both, classical and robust analysis of variances (ANOVA) applied to data from instrumental analysis of marine algae and bee honey samples, undergoing microwave digestion. Some general recommendations on the statistical approach selection are revealed based on real experimental data set. [ABSTRACT FROM AUTHOR]

A tungsten-containing chemical modifier was used in a studyon lead determination, with model interpretation of analyte loss kinetics as its focus. Thermal stabilization effect, analyte loss kinetics and the shape ofthe adsorbance signal were the factors considered in assessing the influence ofmodifier concentration. Acceptable agreement was found between the experimentalresults obtained and with the theory of regular solutions.