Blank overcorrection in ICP-MS – influence of the torch material used on the molar mass of highly enriched silicon and on the redefinition of the kilogram

Abstract

Molar mass (M) measurements of silicon highly enriched in ²⁸Si carried out in the framework of the upcoming redefinition of the kilogram and mole were performed both with a specially designed sapphire torch almost free of natural silicon and a conventional torch completely built from quartz using a high resolution multicollector-inductively coupled plasma mass spectrometer (MC-ICP-MS). In both cases, relative uncertainties (u_rel(M)) were in the 10⁻⁹ range. M measured using the quartz assembly was systematically shifted to lower values by −3 × 10⁻⁶%, which is thirteen times the combined standard uncertainty of M measured with the sapphire assembly. This effect can be explained by the silicon released from the injector tube surface by the blank solution (aqueous tetramethylammonium hydroxide, TMAH) used to correct for the load of natural silicon in the samples and blends (containing the sample and ³⁰Si spike) necessary to avoid the otherwise large bias of the isotopic pattern of the enriched silicon in the presence of the ubiquitous silicon with natural isotopic composition. The TMAH solution mobilizes the silicon from the injector tube in the torch most prominently. In the case of silicon containing sample solutions, these are saturated due to the extreme content of Si, 5000 μg g⁻¹ in the sample x and 3000 μg g⁻¹ in the blend bx. The additional Si load during the blank measurements induced an overcorrection of the isotope ratios in the case of the quartz assembly.