Iron Cycling in Marine Sediments – New Insights from Isotope Analysis on Sequentially Extracted Fe Fractions

Abstract:

Reactive Fe (oxyhydr)oxides preferentially undergo early diagenetic cycling and may cause a diffusive flux of dissolved Fe²⁺ from sediments towards the sediment-water interface. The partitioning of Fe in sediments has traditionally been studied by applying sequential extractions based on reductive dissolution of Fe minerals. We complemented the sequential leaching method by Poulton and Canfield [1] in order to be able to gain δ⁵⁶Fe data for specific Fe fractions, as such data are potentially useful to study Fe cycling in marine environments. The specific mineral fractions are Fe-carbonates, ferrihydrite + lepidocrocite, goethite + hematite, and magnetite. Leaching was performed with acetic acid, hydroxylamine-HCl, Na-dithionite and oxalic acid. The processing of leachates for δ⁵⁶Fe analysis involved boiling the samples in HCl/HNO₃/H₂O₂, Fe precipitation and anion exchange column chromatography.

The new method was applied to short sediment cores from the North Sea and a bay of King George Island (South Shetland Islands, Antarctica). Downcore mineral-specific variations in δ⁵⁶Fe revealed differing contributions of Fe (oxyhydr)oxides to redox cycling. A slight decrease in easily reducible Fe oxides correlating with a slight increase in δ⁵⁶Fe for this fraction with depth, which is in line with progessive dissimilatory iron reduction [2,3], is visible in the top 10 cm of the North Sea core, but not in the antarctic sediments. Less reactive (dithionite and oxalate leachable) fractions did not reveal isotopic trends. The acetic acid-soluble fraction displayed pronounced δ⁵⁶Fe trends at both sites that cannot be explained by acid volatile sulfides that are also extracted by acetic acid [1]. We suggest that low δ⁵⁶Fe values in this fraction relative to the pool of easily reducible Fe oxides result from adsorbed Fe(II) that was open to isotopic exchange with oxide surfaces, affirming the experimental results of Crosby el al. [2]. Hence, δ⁵⁶Fe analyses on marine sediments appear to be particularly useful when targeting specific Fe pools, rather than the total highly reactive Fe fraction, since isotopic trends may be masked within the latter pool.

[1] Poulton and Canfield (2005), Chemical Geology 214, 209-221.

[2] Crosby et al., Geobiology 5 (2007), 169-189.

[3] Staubwasser et al., Geology 34 (2006), 629-632.