ANALYTICAL ERRORS IN THE DETERMINATION OF HIGH-FIELD STRENGTH ELEMENTS AND THEIR IMPLICATIONS IN TECTONIC INTERPRETATION STUDIES

The quality of analytical data obtained from cost-effective commercial multi-element "packages" is examined with respect to its use in tectonic interpretation studies. Good accuracy and precision in the determination of high field strength elements, Zr, Y, Ti, Nb, Hf, Ta and the rare-earth elements (REE) is essential in extracting the correct and maximum information for discrimination diagrams. Misinterpretation can result from the use of noisy and biased analytical data; this is shown by the poor quality of results for Y and Nb where the concentrations measured are close to the method's detection limits. A "round-robin" study to determine Zr, Nb, Hf and Ta demonstrates the significant variability to be expected amongst laboratories, even for elements with relatively high abundance levels, such as Zr. The round-robin results are compared with a sensitive analytical method based upon inductively coupled plasma-mass spectrometry (ICP-MS) and LiBO2 fusion. The inefficiency of the common mixed-acid attack (HF-HClO4-HNO3-HCl) to dissolve the heavy REE in some samples is proven by comparison with results employing a fusion decomposition and with instrumental neutron activation analysis. Such a partial attack prior to analysis can lead to an erroneous chondrite-normalised pattern, incorrectly suggesting light-REE enrichment and masking positive or negative Eu anomalies.