Low uncertainty Fe isotope ratio measurements in sea-water samples by ICP-MS at medium mass resolution

Abstract

Single detector double focusing magnetic sector ICP-MS with high mass resolution capabilities (HR-ICP-SMS) is one of the most employed techniques for Fe isotope ratio measurements. Its application to sea-water samples is particularly challenging, mostly due to a combination of two problems, the ultra-low natural Fe contents and the presence of a large amount of salts. Matrix separation by means of Mg(OH)2 co-precipitation is efficient to only ∼96 to 98.5%, and the remaining inorganic salts fraction is a source of measurement problems.This paper describes the method we developed to overcome these difficulties, and to achieve 2–3% (k = 2) expanded uncertainty (U) on direct absolute n(56Fe)/n(57Fe) isotope ratio measurements by HR-ICP-SMS at medium mass resolution for ∼300–450 ng Fe kg−1 samples containing ∼0.1% residual salinity (i.e., following matrix separation on sea-water samples containing originally ∼30 ng Fe kg−1). There was a difference in results of up to 4% depending on whether mass discrimination effects were corrected for using matched conditions for both matrix composition and Fe concentration of the isotopic reference material. Furthermore, there were severe instabilities during the measurement scans in the form of fast high intensity signals, which were related to residual salts present in the samples. We observed that this noise could bias the end result by up to nearly 2%. Finally, the correction of our data also included the routine previously developed in our group to overcome the constraints posed by the instrument software, specifically for isotope ratio measurements at medium mass resolution. Overall, the stability of the repeatability over 150 min improved by a factor of ∼1.5, the reproducibility of the ratios improved by more than a factor of 2 and the average ratio changed by ∼3% (and by up to 11% in the worst case). Application of these developments to the determination of the iron content in sea-water by isotope dilution mass spectrometry led to a non-negligible ∼3% change of the final result and to a decrease of the U statement on this value by a factor of 2. The method was further validated using the CASS-2 certified reference sea-water material.