Interpretation of interlaboratory trials based on accuracy profiles.

Validation is a very active field in analytical chemistry as illustrated by the numerous publications addressing this topic. According to most definitions, method validation is required to confirm the fitness-for-purpose of a particular analytical method. When considering interlaboratory validation, organization and calculation of parameters that describe method performance follow well-recognized standards. But the strategy used to verify whether the method is fit for a particular purpose is still under discussion. This paper presents a method for assessing the fitness-for-purpose of analytical methods based on results produced during interlaboratory trials. It is based on the construction of accuracy profiles, which can be used as a graphical decision support tool, and demonstrates how accuracy profiles, initially developed for in-house validations, can be extended to interlaboratory studies. These interlaboratory accuracy profiles use measurements collected under reproducibility conditions to give an interval where an expected proportion of future measurements will be located. This interval can be compared to an acceptability interval defined by the end-user to simply decide whether a method is fit-for-purpose or not. Several examples of application illustrate how data can be interpreted to draw conclusions about produced accuracy profiles and fitness-for-purpose. Hence, the accuracy profile could be used as a harmonized procedure to assess the performance of analytical methods that undergo interlaboratory validation.

Direct solid analysis of powdered tungsten carbide hardmetal precursors by laser-induced argon spark ablation with inductively coupled plasma atomic emission spectrometry.

The potential of the laser-induced argon spark atomizer (LINA-Spark atomizer) coupled with ICP-AES as a convenient device for direct analysis of WC/Co powdered precursors of sintered hardmetals was studied. The samples were presented for the ablation as pressed pellets prepared by mixing with powdered silver binder containing GeO2 as internal standard. The pellets were ablated with the aid of a Q-switched Nd:YAG laser (1064 nm) focused 16 mm behind the target surface with a resulting estimated power density of 5 GW cm(-2). Laser ablation ICP-AES signals were studied as a function of ablation time, and the duration of time prior to measurement (pre-ablation time) which was necessary to obtain reliable results was about 40 s. Linear calibration plots were obtained up to 10% (m/m) Ti, 9% Ta and 3.5% Nb both without internal standardization and by using germanium as an added internal standard or tungsten as a contained internal standard. The relative uncertainty at the centroid of the calibration line was in the range from +/- 6% to +/- 11% for Nb, Ta and Ti both with and without internal standardisation by Ge. A higher spread of points about the regression was observed for cobalt for which the relative uncertainty at the centroid was in the range from +/- 9% to +/- 14%. Repeatability of results was improved by the use of both Ge and W internal standards. The lowest determinable quantities calculated for calibration plots were 0.060% Co, 0.010% Nb, 0.16% Ta and 0.030% Ti with internal standardization by Ge. The LA-ICP-AES analyses of real samples led to good agreement with the results obtained by solution-based ICP determination with a relative bias not exceeding 10%. The elimination of the dissolution procedure of powdered tungsten (Nb, Ta, Ti) carbide is the principal advantage of the developed LA-ICP-AES method.