Identification of a matrix effect in the MC-ICP-MS due to sample purification using ion exchange resin: An isotopic case study of molybdenum

Abstract

Multiple-collector inductively coupled plasma mass spectrometry (MC-ICP-MS) has become the preferred method for precise and accurate measurements of the relative abundances of many radiogenic and stable isotopes in natural materials. Isotopic analyses by MC-ICP-MS require a correction for instrumental mass-dependent isotopic fractionation (“instrumental mass bias”). Two techniques have been used to correct for instrumental mass bias in the MC-ICP-MS: (1) standard-sample bracketing (SSB) or (2) double spiking. SSB is often cited as the preferred method, but it is more susceptible to matrix effects. Here we demonstrate that a matrix effect in the MC-ICP-MS may arise indirectly from the chemical separation and purification of molybdenum using anion exchange resin. The results of our experiments show that a Mo standard passed through a column of anion exchange resin or a Mo standard added to a Mo-free solution that had been collected from anion exchange resin appears to be isotopically lighter than expected from direct analysis of the same standard. Using amounts of Mo similar to what might be expected from most natural samples (∼3 μg per column cut), these offsets span a significant fraction (∼10–60%) of the total known range of mass-dependent Mo isotopic variation in nature. This “column matrix effect” appears to be caused by organic material stripped from the resin. All of our attempts to eliminate or control this column matrix effect have failed, making it difficult (if not impossible) to obtain accurate measurements of mass-dependent Mo isotopic variations in natural materials using the anion exchange resin procedure described in this study and SSB techniques to correct for instrumental mass bias in the MC-ICP-MS. It is currently unknown if this type of column matrix effect will affect measurements of other stable or radiogenic isotopes by MC-ICP-MS when SSB is used to correct for instrumental mass bias.