Comparison of analytical methods: ICP-QMS, ICP-SFMS and FF-ET-AAS for the determination of V, Mn, Ni, Cu, As, Sr, Mo, Cd and Pb in ground natural waters

Three analytical methods, namely, inductively coupled plasma sector field mass spectrometry (ICP-SFMS); inductively coupled plasma quadrupole mass spectrometry (ICP-QMS) and filter-furnace electrothermal atomic-absorption spectroscopy (FF-ET-AAS) for the determination of V, Mn, Ni, Cu, As, Sr, Mo, Cd and Pb in ground natural water samples were compared and evaluated for their capacity to provide reliable and precise results. Two certified reference materials (SLEW-3 Estuarine Water; SLRS-4 River Water) were analysed to prove that accurate results could be obtained by using all the listed methods with properly optimised parameters. The limit of detection (LOD) for V, Mn, Ni, Cu, As, Sr, Mo, Cd and Pb provided by the ICP-MS methods ranged from 0.001 to 0.05 µg L-1. Such LOD proved sufficient for the reliable determination of the listed elements in ground natural waters. However, the LOD of the FF-ET-AAS was approximately two orders of magnitude higher than that of ICP-MS, which made it impossible to quantify V, Mn, Ni, Mo and Pb. The effects of the usage of the collision cell mode in ICP-QMS and of the desolvation system Apex for ICP-SFMS to eliminate oxide ions levels were investigated. For all three analytical methods, the influence of the matrix effect on the results of the determination of the investigated elements using matrix model solution, external calibration and standard addition methods was evaluated. A comparison using a paired Student's t-test between the results obtained by both ICP-MS methods for V, Mn, Ni, Cu, As, Sr, Mo, Cd and Pb concentrations in ground natural waters showed that there was no significant difference on a 95% confidence level. The precision of the results for ICP-SFMS, ICP-QMS and FF-ET-AAS varied between ~0.5 and 11; 2.5 and 12.5; 3 and 13.5%, respectively. Moreover, ICP-SFMS equipped with the desolvation system APEX proved a better choice for As, Cu and Mn analysis due to its better LOD (0.008, 0.03 and 0.02 µg L-1, respectively) and precision (Sr <= 5.0; 7.5; 9.0%, respectively) compared to ICP-QMS and FF-ET-AAS.