Accurate and Precise Silicon Isotope Analysis of Sulfur‐ and Iron‐Rich Samples by MC‐ICP‐MS

Abstract

Silicon isotope determination of sulfur‐rich samples by MC‐ICP‐MS can be challenging because cation‐exchange chromatography used for Si purification does not efficiently remove anionic sulfur species. Results for pure Si standard solutions with addition of sulfate showed shifts of up to +1.04 ± 0.10‰ (2s) in δ30Si. Doping of both standard solutions and samples with S to a fixed S/Si ratio can eliminate the relative change in instrumental mass fractionation due to variable S/Si in samples and also boosts the relative sensitivity of Si by up to 66%. Moreover, Fe hydroxide precipitation during sample processing adsorbs Si resulting in isotopic fractionations. Tests using Fe‐rich samples showed that this could be a major factor for observed shifts in δ30Si. Acidification of the sample and standard solutions to a pH < 1 aggressively dissolved any Fe hydroxide precipitates, even in relatively Fe‐rich samples such as chondrite meteorites. The pH values of the sample solutions were subsequently adjusted to a range of 2–3 by adding ultra‐pure NaOH solutions. The combination of sulfur doping and the pH adjustment protocol ensured a full recovery of Si and proved to be an efficient and reliable method for Si isotope determination of S‐ and Fe‐rich materials.

Assessment of matrix effects induced by chloride, sulfate and dissolved organic carbon during Si isotope measurements by MC-ICP-MS under wet plasma conditions

“…Although the statistical approach revealed a shi in d 30 Si of −0.06& produced by sulfate and a mixture of sulfate and chloride contamination, this shi is substantially smaller than those reported by previous authors with values of +1.4&, 8 +0.28& 10 and +1.04&. 9 Moreover, considering the analytical uncertainties associated with Si isotope measurements by MC-ICP-MS, the shi in d 30 Si of −0.06& for both groups (sulfate only and sulfate + chloride) fall within errors of the undoped NBS-28, whose d 30 Si = −0.01 ± 0.09& (2SD), as well as within the long-term reproducibility of ±0.11& (2SD) obtained for the BHVO-2 standard. Consequently, we concluded that despite observing a matrix effect (−0.06&) produced by sulfate, chloride, or a mix of both anions, this effect is not distinguishable given our current measurement uncertainties.…”

“…Our results clearly indicate that no signicant matrix effect is induced by the presence of chloride for [Cl − ]/[Si] ratio < 5.65. In contrast to the ndings of Van Den Boorn et al, 8 Chen et al 9 and Hughes et al, 10 ]/[Si] = 0.12 and 0.15 that show a slightly larger deviation of −0.12& and −0.11&, respectively. However, all those values remain within uncertainties of the undoped NBS-28.…”

“…Van Den Boorn et al 8 reported a d 30 Si offset of +1.4& when [S]/[Si] > 0.02 (wt/wt) in samples from sulfur-rich rock. Chen et al 9 tested the inuence of sulfur by doping a pure silicon solution and reported a similar d 30 Si shi of +1.04& when [S]/[Si] > 0.02. Hughes et al 10 observed a shi of +0.28& in river samples when [SO 4 Table 1 Operating parameters for this work and studies that have tested the influence of DOC and anions (Cl − , SO Conversely, Georg et al 5 tested the effect of sulfates and nitrates on the accuracy of Si isotope measurements, but no signicant isotopic effect was reported.…”

High precision analysis of stable potassium (K) isotopes by the collision cell MC-ICP-MS “Sapphire” and a correction method for concentration mismatch

“…The term “concentration mismatch” was used throughout this study to be consistent with previous studies, 70,72,107 but we emphasize that ion intensity mismatch is the fundamental source for biased isotope measurements. Although, typically, ion intensity is directly related to the analyte concentration, it can also be affected by other factors, such as the presence of certain matrix elements, 109,110 and instrument drift (see our results below), which could also result in biased measurements. In this contribution, we use “concentration mismatch” in a broader sense to include “ion intensity mismatch” caused by any reason.…”

Silicon and iron isotopes in components of enstatite chondrites: Implications for metal–silicate–sulfide fractionation in the solar nebula

“…One additional problem of Si isotope analyses in metallic fractions of EC is the high abundance of sulfides and phosphides associated with Si‐bearing kamacite (Weisberg & Kimura, 2012). The cation exchange ion chromatography procedure does not effectively remove S and P from Si, and therefore, an overabundance of these anions in the samples might have an influence on the measured Si isotope composition (Chen et al., 2017; van den Boorn et al., 2009). The matrix effect of anions on mass spectrometric analyses of Si was tested at GFZ by doping purified NBS‐28 solution with phosphate, sulfate, and nitrate anions to produce different [anion]/[Si] mass ratios (μg μg −1 ) from 0.01 to 6.4 (Oelze et al., 2016).…”