Trace element characterization of USGS reference materials by HR-ICP-MS and Q-ICP-MS

Abstract A trace-element characterization of United States Geological Survey (USGS) reference materials has been carried out using three different inductively coupled plasma mass spectrometer instruments: Nu Instruments AttoM, Thermo Scientific Element 2, and Agilent 7700x. The six USGS reference materials (BHVO-2, BCR-2, AGV-2, STM-2, RGM-2, and G-3) include both extrusive and intrusive rocks with compositions ranging from mafic to felsic that encompass a wide range of trace element concentrations and mineral phases. The prevalence of these 2nd (and 3rd) generation reference materials among geochemical laboratories and the added care that was taken during sample preparation make this generation of reference materials more suitable for characterization and use by the geochemical community. The mafic-intermediate reference materials (BHVO-2, BCR-2, and AGV-2) had the most abundant published characterization prior to this study, which allowed for thorough and direct comparison with previous studies (and USGS certificates). Published data for these reference materials are relatively well constrained for most elements (i.e., not highly variable) and the results of this study generally correspond well with the literature. This can, at least partially, be attributed to the relatively simple matrices of these reference materials together with their overall smaller range of trace element concentrations. Unlike the mafic-intermediate reference materials, STM-2 lacked any published trace element concentrations prior to this study and data for RGM-2 and G-3 were relatively sparse. Data for STM-2 and RGM-2 from this study are compared to preliminary USGS certificates, while data for G-3 are compared to the G-2 certificate values; for comparison with published data, RGM-1 and G-2 are included. Overall, the published data for RGM and G are variable across the entire suite of trace elements and display poorer agreement with the data from this study. This can be explained by the fact that, in addition to the comparison between generations that may add heterogeneity, the felsic reference materials have relatively complex matrices and larger ranges in trace element concentration. Given the new data that this study provides, the felsic reference materials can now be used by the geochemical community to produce more accurate trace element concentration analyses of felsic materials. By using three instruments with different configurations, this study also serves as a comparison of the performance of each instrument in response to an array of geologic matrices. The data reveal that the tested instruments are all capable of producing accurate and precise trace element concentration data over a range of geologic matrices, provided that each instrument is carefully operated within its dynamic range. In many cases, this could be accomplished simply by ensuring proper dilution of samples; however, when matrices are more complex, the use of medium and high resolutions can be useful in reducing the amount of preparative chemistry that is required to manage highly variable trace element concentrations within a single sample as well as any potential isobaric interferences.