Refining the carbonate-associated iodine redox proxy with leaching experiments

Abstract

Iodate ions can substitute for carbonate in crystal lattices, which is why carbonate-associated iodine (CAI) is increasingly being used to track redox-sensitive iodine species, and by extrapolation, the redox conditions of ancient oceans. Extracting iodine from carbonate rocks is challenging, however, because iodine is susceptible to volatilization in acidic solutions. A broadly similar leaching scheme with different leaching times has been applied to all kinds of ancient carbonate rocks. However, the effects of non‑carbonate phases and iodine stability during leaching remain under-constrained, which limits its use for paleoredox reconstruction. For this study, we assessed these issues by conducting a series of leaching experiments on carbonate standards and ancient carbonate rocks of varying purity. We first developed an alternative stabilizer for iodine, i.e., 3% EDTA (ethylenediaminetetraacetic acid) in 3% NH4OH (ammonium hydroxide). This easily accessible new stabilizer effectively prevents the precipitation of metal ions from the alkaline solution, thus easing the preparation process and reducing the iodine signal drift during the CAI analysis. We also propose two interlaboratory limestone standards for iodine analysis with long-term (one year) I/Ca values of 0.43 ± 0.03 μmol/mol for limestone CRM-393 and 0.99 ± 0.06 μmol/mol for argillaceous limestone SRM-1c. Here we demonstrate that excessive leaching times (>60 min) can lead to the loss of nearly half of the iodine released under typical leaching conditions. Bulk carbonate digestion is shown to potentially underestimate primary CAI concentrations due to the presence of diagenetic carbonates. Moreover, we identified phosphates and organic matter as potential contaminants, the inadvertent leaching of which could lead to overestimated concentrations. Although the potential significance of such contamination warrants further study, our results show that contamination, if present, can be minimised by oxidative cleaning with 12% NaClO followed by dissolution in 0.3 M acetic acid. The possibility that measured CAI concentrations can be both under- and over-estimated may lead to reassessment of oxygenation events identified from the iodine concentrations of impure carbonate rocks.