On the isotope composition of reactive iron in marine sediments: Redox shuttle versus early diagenesis

• Terrigenous particulate reactive Fe has a light δ56Fe relative to the bulk crust.
• Partial loss of reactive Fe to the water column shifts δ56Fe of the remaining reactive Fe to higher values.
• Isotope composition of pyrite depends on the extent of pyritization.
• δ56Fe of highly reactive Fe pool (incl. oxide, carbonate and sulfide) may provide information on water column Fe cycling.
• In reducing settings with high Fe but low sulfide concentrations, reactive Fe may form authigenic silicate minerals.

The isotope composition of reactive iron (Fe) in marine sediments and sedimentary rocks is a promising tool for identifying Fe sources and sinks across ocean basins. In addition to cross-basinal Fe redistribution, which can modify Fe isotope signatures, Fe minerals also undergo diagenetic redistribution during burial. The isotope fractionation associated with this redistribution does not affect the bulk isotope composition, but complicates the identification of mineral-specific isotope signatures. Here, we present new Fe isotope data for Peru margin sediments and revisit previously published data for sediments from the California margin to unravel the impact of early diagenesis on Fe isotope compositions of individual Fe pools.Sediments from oxic California margin sites are dominated by terrigenous Fe supply with Fe release from sediments having a negligible influence on the solid phase Fe isotope composition. The highly reactive Fe pool (sum of Fe bound to (oxyhydr)oxide, carbonate, monosulfide and pyrite) of these sediments has a light isotope composition relative to the bulk crust, which is consistent with earlier studies showing that continental weathering shifts the isotope composition of Fe (oxyhydr)oxides to lighter values. Ferruginous sediments within the Peruvian oxygen minimum zone are depleted in Fe relative to the lithogenic background, which we attribute to extensive Fe release to the water column. The remaining highly reactive Fe pool has a heavier isotope composition compared to California margin sediments. This observation is in agreement with the general notion of an isotopically light benthic Fe efflux. Most of the reactive Fe delivered and retained in the sediment is transferred into authigenic mineral phases within the topmost 10 to 20 cm of the sediments. We observe a first-order relationship between the extent of pyritization of Fe monosulfide and the isotope composition of authigenic pyrite. With increasing pyritization, the isotope composition of authigenic pyrite approaches the isotope composition of the highly reactive Fe pool. We argue that the isotope composition of authigenic pyrite or other Fe minerals that may undergo pyritization may only be used to trace water column sources or sinks if the extent of pyritization is separately evaluated and either close to 100% or 0%. Alternatively, one may calculate the isotope composition of the highly reactive Fe pool, thereby avoiding isotope effects due to internal diagenetic redistribution. In depositional settings with high Fe but low sulfide concentrations, source and sink signatures in the isotope composition of the highly reactive Fe pool may be compromised by sequestration of Fe within authigenic silicate minerals. Authigenic silicate minerals appear to be an important burial phase for reactive Fe below the Peruvian oxygen minimum zone.