Experimental design for TBT quantification by isotope dilution SPE–GC–ICP–MS under the European water framework directive

Abstract

In Europe the maximum allowable concentration for tributyltin (TBT) compounds in surface water has been regulated by the water framework directive (WFD) and daughter directive that impose a limit of 0.2ngL−1 in whole water (as tributyltin cation). Despite the large number of different methodologies for the quantification of organotin species developed in the last two decades, standardised analytical methods at required concentration level do not exist. TBT quantification at picogram level requires efficient and accurate sample preparation and preconcentration, and maximum care to avoid blank contamination. To meet the WFD requirement, a method for the quantification of TBT in mineral water at environmental quality standard (EQS) level, based on solid phase extraction (SPE), was developed and optimised. The quantification was done using species-specific isotope dilution (SSID) followed by gas chromatography (GC) coupled to inductively coupled plasma mass spectrometry (ICP–MS). The analytical process was optimised using a design of experiment (DOE) based on a factorial fractionary plan. The DOE allowed to evaluate 3 qualitative factors (type of stationary phase and eluent, phase mass and eluent volume, pH and analyte ethylation procedure) for a total of 13 levels studied, and a sample volume in the range of 250–1000mL. Four different models fitting the results were defined and evaluated with statistic tools: one of them was selected and optimised to find the best procedural conditions. C18 phase was found to be the best stationary phase for SPE experiments. The 4 solvents tested with C18, the pH and ethylation conditions, the mass of the phases, the volume of the eluents and the sample volume can all be optimal, but depending on their respective combination. For that reason, the equation of the model conceived in this work is a useful decisional tool for the planning of experiments, because it can be applied to predict the TBT mass fraction recovery when the experimental conditions are drawn. This work shows that SPE is a convenient technique for TBT pre-concentration at pico-trace levels and a robust approach: in fact (i) number of different experimental conditions led to satisfactory results and (ii) the participation of two institutes to the experimental work did not impact the developed model.