Reactivity of neodymium carriers in deep sea sediments: Implications for boundary exchange and paleoceanography

The dissolved neodymium (Nd) isotopic distribution in the deep oceans is determined by continental weathering inputs, water mass advection, and boundary exchange between particulate and dissolved fractions. Reconstructions of past Nd isotopic variability may therefore provide evidence on temporal changes in continental weathering inputs and/or ocean circulation patterns over a range of timescales. However, such an approach is limited by uncertainty in the mechanisms and importance of the boundary exchange process, and the challenge in reliably recovering past seawater Nd isotopic composition (epsilon(Nd)) from deep sea sediments. This study addresses these questions by investigating the processes involved in particulate-solution interactions and their impact on Nd isotopes. A better understanding of boundary exchange also has wider implications for the oceanic cycling and budgets of other particle-reactive elements. Sequential acid-reductive leaching experiments at pH similar to 2-5 on deep sea sediments from the western Indian Ocean enable us to investigate natural boundary exchange processes over a timescale appropriate to laboratory experiments. We provide evidence that both the dissolution of solid phases and exchange processes influence the epsilon(Nd) of leachates, which suggests that both processes may contribute to boundary exchange. We use major element and rare earth element (REE) data to investigate the pools of Nd that are accessed and demonstrate that sediment leachate epsilon(Nd) values cannot always be explained by admixture between an authigenic component and the bulk detrital component. For example, in core WIND 24B, acid-reductive leaching generates epsilon(Nd) values between -11 and -6 as a function of solution/solid ratios and leaching times, whereas the authigenic components have epsilon(Nd) approximate to -11 and the bulk detrital component has epsilon(Nd) approximate to -15. We infer that leaching in the Mascarene Basin accesses authigenic components and a minor radiogenic volcanic component that is more reactive than Madagascan-derived clays. The preferential mobilisation of such a minor component demonstrates that the Nd released by boundary exchange could often have a significantly different epsilon(Nd) composition than the bulk detrital sediment. These experiments further demonstrate certain limitations on the use of acid-reductive leaching to extract the epsilon(Nd) composition of the authigenic fraction of bulk deep sea sediments. For example, the detrital component may contain a reactive fraction which is also acid-extractible, while the incongruent nature of this dissolution suggests that it is often inappropriate to use the bulk detrital sediment elemental chemistry and/or epsilon(Nd) composition when assessing possible detrital contamination of leachates. Based on the highly systematic controls observed, and evidence from REE patterns on the phases extracted, we suggest two approaches that lead to the most reliable extraction of the authigenic epsilon(Nd) component and good agreement with foraminiferal-based approaches; either (i) leaching of sediments without a prior decarbonation step, or (ii) the use of short leaching times and low solution/solid ratios throughout. (C) 2013 Elsevier Ltd. All rights reserved.