Initial results on the coupling of sedimentation field-flow fractionation (SdFFF) to inductively coupled plasma-tandem mass spectrometry (ICP-MS/MS) for the detection and characterization of TiO2 nanoparticles

Manufactured TiO2 nanoparticles (TiO2 NPs) are nowadays widely present in products accessible to the mass market like paints, cosmetics or sunscreens. Despite the great increase in the use of these nanoparticles, until now their potential effects on the environment and biological systems have not been sufficiently studied. Reliable analytical methods are therefore required for better characterization of these emerging materials. We suggest the hyphenation of inductively coupled plasma-mass spectrometry (ICP-MS) with a separation technique like sedimentation field-flow fractionation (SdFFF) to obtain information related to the size and state of agglomeration of the investigated nanoparticles. In this work, initial experiments related to the on-line coupling of SdFFF to inductively coupled plasma-tandem mass spectrometry (ICP-MS/MS) have been conducted allowing the detection and separation of TiO2 NPs. The application of the ICP-MS/MS technology using a NH3 mass shift mode allowed the removal of the different molecular and isobaric interferences that complicate the reliable detection and quantification of Ti. Under optimised conditions, the achievable instrumental detection limits in matrix solutions were below 10 ng L-1 of Ti. The separation parameters of the SdFFF system were optimized using two commercially available model TiO2 NPs (nominal sizes: 21 and 50 nm). These materials were characterised concerning hydrodynamic diameters and the state of aggregation also using a multiangle light scattering (MALS) analysis detector. A transmission electron microscopy (TEM) technique was also applied to confirm the sizes and the shapes of the TiO2 NPs as well as the presence of aggregates. To demonstrate the applicability of the combination of SdFFF and ICP-MS/MS for the detection of TiO2 NPs at trace levels as well as to gain information about their hydrodynamic diameter and agglomeration state, the optimized method has been applied to the analysis of real water samples from a local lake in Germany.