Uranium isotope fractionation in Saanich Inlet: A modern analog study of a paleoredox tracer

Uranium isotope fractionation was studied in the sediment and water of Saanich Inlet, a seasonally anoxic fjord on the east coast of Vancouver Island. The concentration of dissolved U is the same above and below the redoxcline at ∼120 m depth, with an average δ238U value of −0.45 ± 0.06‰ (2 s.e.), which is indistinguishable from seawater values reported in other studies. This finding is consistent with water renewal times for the inlet that are roughly seasonal in frequency, thus hiding the impact of the U losses to the anoxic sediment of the inlet. Complete digests of the anoxic sediment yielded higher δ238U values than seawater (−0.21 ± 0.11‰ (2 s.e.)). By contrast, complete digests of the sediment trap samples yielded lower δ238U values than seawater (−0.52 ± 0.10‰ (2 s.e.)), and marine plankton from the inlet yielded the lowest δ238U value of −1.24 ± 0.17‰ (2σ). Cross-plots of δ238U and δ234U vs. Th/U reveal strong correlations indicative of two-component mixing when the data from traps and anoxic inlet sediment are combined. One mixing end-member is fractionated uranium of seawater origin, with a δ238U value of 0.17 ± 0.14‰ (2σ). The other is detrital uranium with a δ238U value of −0.83 ± 0.12‰ (2σ). The detrital end-member is lower than the range of δ238U values reported in the literature for granitoid igneous rocks (−0.44‰ to –0.17‰, 2σ) (Telus et al., 2012), suggesting that continental weathering fractionates uranium isotopes, with preferential release of 238U. Development and application of U isotopes as a paleoredox proxy has its basis in the nuclear volume fractionation. The data from Saanich Inlet meets this expectation, with a positive fractionation factor (Δaq(VI)sed(IV)) of 0.62 ± 0.17‰ (2σ) calculated as the difference in δ238U between authigenic U in anoxic sediments and sediment traps (+0.17‰) and U dissolved in seawater (−0.45‰). However, it is widely believed that U(VI) reduction in the marine environment occurs on the surfaces of particles and that the negative isotope effect associated with U(VI) sorption to plankton (Δseawaterplankton=-0.79±0.17‰ (2σ) opens up the possibility that the particulate pool of seawater derived U(VI) is fractionated from the dissolved pool. Accordingly, the reduction of U(VI) in the marine environment might involve two steps: (1) U(VI) sorption to particles with a negative fractionation, and (2) reduction to U(IV) on particles with a positive fractionation. Environmental factors that are not yet well understood may influence the relative reaction rates for the two steps, thus affecting the magnitude and sign of the overall fractionation. This has the potential to explain conflicting results reported for U(VI) reduction experiments in the literature (laboratory and field), where both positive and negative fractionations have been observed, and in some instances, no fractionation at all. Variation in the fractionation factor has the potential to complicate application of U isotopes as a paleoredox proxy, but this study and a study of the Black Sea (Romaniello, 2012) give the same results (within their respective uncertainties) suggesting that 0.62 ± 0.17‰ (2σ) is a robust assessment of the apparent U isotope fractionation factor associated with reductive deposition of seawater U(VI) in both anoxic and euxinic marine sediments of the present day.