On the use of iron radio-isotopes to study iron speciation kinetics in seawater: A column separation and off-line counting approach

Abstract

The determination of unidirectional fluxes of Fe in the seawater Fe-pools requires the use of Fe (radio)isotopes, which in turn necessitates off-line iron counting. To implement new off-line counting approaches, methods by King et al. (1991) [King DW, Lin J, Kester DR. Spectrophotometric determination of iron(II) in seawater at nanomolar concentrations. Anal Chim Acta 1991;247:125–32] were adapted for use with radio-iron for Fe(II) and total Fe analysis. The approach consists of passing samples through preloaded SepPak® C 18 cartridges with either ferrozine (FZ) for Fe(II) measurement or 1-nitroso-2-naphthol (1N2N) for total iron analysis. Cartridges are afterwards eluted with methanol and the eluate subsequently counted for radio-iron. A series of experiments was carried out to investigate the new method's applicability. This included loading and extraction behaviour, selectivity for the targeted iron species, iron species recovery, memory effects and the influence of iron concentration. The FZ-preloaded cartridge showed 85% Fe(II) recovery in seawater by direct determination. Recoveries of about 60% for Fe(II) in seawater were found when including 10–15 min waiting time and sample loops, which was probably caused by the instability of Fe(II) due to oxidation. Using 1N2N cartridges for total dissolved iron measurement, recovery was obtained as nearly 100%, but with high variability which was probably caused by the interaction time of Fe with the 1N2N-preloaded cartridges. The cartridges showed only small memory effects. Remaining radio-iron could only be removed from the cartridges with 0.5 M HCl, which destroyed the functionality of the cartridge. This result suggests that the Fe fraction, which could not be eluted by methanol, was irreversibly bound to the cartridge. Desferrioxamine B was applied as a model ligand to test for interference from naturally occurring organic ligands; the results indicate that interferences may occur and should be considered very carefully. The described approaches are tested for seawater iron concentrations in a 1.4–2.3 nM Fe range; further testing is necessary for other Fe concentration ranges.