www.acsprf.org

Over geologic time, major events such as the opening of ocean basins, or large changes in sea level, have periodically led to relatively rapid shifts in the oxygenation of marine bottom waters. Widespread anoxia in the deep ocean is of great interest, as it can lead to mass extinctions and create an ideal environment for the accumulation of organic-rich sediments that may eventually become the source of massive oil-shale deposits. Marine geochemists have long searched for chemical markers that leave a clear and permanent signal of these episodes in the sedimentary record and can therefore be utilized as a so-called ‘paleo-redox’ proxy. The distinct redox states that certain metals possess under oxidizing and reducing conditions are differentially sorbed on settling particles, which might cause their concentrations in the underlying sediment to rise and fall with the oxygenation of deep water at the time of deposition. A particularly promising example is cerium (Ce), the only member of the YREE (yttrium and rare earth element) series that can be oxidized to Ce(IV) at the normal temperatures and pressures of seawater. Since the higher charge and smaller ionic radius of Ce(IV) make it more particle-reactive than its strictly trivalent REE neighbors, it is more strongly sorbed on common components of marine particles like Fe oxides, Mn oxides, and organic matter. The argument then typically goes as follows: in the presence of dissolved oxygen, cerium is oxidized to Ce(IV) and more strongly sorbed on settling particles than the other YREEs; a positive Ce anomaly will thus be recorded in the underlying sediments. In anoxic water, cerium is present as Ce(III) and sorbs as strongly as the other YREEs; no Ce anomaly will develop. Hence, positive Ce anomalies in the sedimentary record are thought to indicate periods of oxygenated bottom water and the absence of a Ce anomaly periods of bottom water anoxia. Whereas many geochemists have forged ahead with interpretations of Ce anomaly records in various sedimentary settings, others have pointed out that such reasoning is subject to a number of caveats, particularly that the sedimentary Ce anomaly record, once formed, must not subsequently be disarranged by Ce mobility during early diagenesis. In this project we endeavor to demonstrate that interpretations of the sedimentary record are further complicated by a more fundamental problem, namely that the Ce anomaly of settling particles does not exclusively depend on the oxygenation of ambient seawater.

Graduate student Kathleen Marshall, who will defend her thesis in Nov 2011, has performed a comprehensive investigation of YREE sorption on hydrous Fe oxides (HFO) and Mn oxides (HMO) in 0.5 M NaCl as a function of pH. Both experiments were conducted under anaerobic conditions (N₂ atmosphere). The HFO study has been published in Marine Chemistry and a manuscript about the HMO study is being prepared for Chemical Geology. Sorption on HFO in the pH range 4-8 was modeled with a non-electrostatic surface complexation model (SCM) that accounts for (de)protonation of the amphoteric hydroxyl functional groups, plus interactions of free YREE cations and YREE-chloride complexes with the neutral and negatively charged sites. The latter also interact with YREE-hydroxide complexes at elevated pH. Each YREE species forms surface complexes with either one (monodentate) or two (bidentate) hydroxyl groups, releasing 1-3 protons per sorbed cation. Ce sorbs as strongly as the trivalent YREEs, yielding no evidence of Ce oxidation. Unlike HFO, which spontaneously precipitates from Fe(III) at pH > 3 within the experimental solution, HMO particles had to be prepared externally by oxidation of Mn(II) with Mn(VII) (as permanganate) in alkaline solution. The reaction could produce Mn(III), yet based on iodometric titrations the resulting solids are nearly 100% Mn(IV)O₂. X-ray powder diffraction analysis moreover shows them to be amorphous with no recognizable crystalline phases. Application of the SCM that was developed for HFO indicates that YREE sorption on HMO, generally stronger than on HFO, does not involve interactions that release 3 protons although this may reflect greater analytical uncertainty at high pH, where the YREEs are almost completely sorbed. Prominent Ce oxidation, catalyzed by the HMO surface, causes Ce to sorb about a hundred times more strongly than the trivalent YREEs.

Given that Fe and Mn oxides coexist in most marine sediments, Kathleen has also studied YREE sorption on Fe/Mn oxide composites, synthesized by oxidation of mixed Fe(II)/Mn(II) solutions with alkaline permanganate. This work is scheduled to be submitted to Geochimica et Cosmochimica Acta. Composites containing 10%, 50%, and 75% HFO were prepared and used in separate sorption experiments in 0.5 M NaCl over a similar pH range. Their Fe and Mn contents were verified by ICP-MS, while X-ray powder diffraction analysis again revealed no crystalline phases. The sorption experiments indicate that distribution coefficient patterns for the trivalent YREEs are well represented by linear combinations of the end member SCMs. The same is not true for their absolute concentrations because the composites are stronger sorbents than either of the pure phases individually, probably due to a larger specific surface area or higher density of sites. On the other hand, Ce sorption cannot be explained with an interpolated model. The SCM for pure HMO does not correctly predict Ce oxidation, as this is not a reversible equilibrium process, nor is the Ce distribution coefficient for Fe/Mn oxide composites simply a linear combination of the measured end member values. It was found instead that Ce distribution coefficients for the composite containing 25% HMO are about 2 log units higher than values for pure HFO, but less than 0.5 log units lower than those for pure HMO. Just a minor fraction of Mn oxides in marine sediments can ostensibly give rise to a substantial positive Ce anomaly under anaerobic conditions, which consequently is not a reliable proxy for the redox state of the overlying bottom waters. Bearing this in mind, future interpretations of sedimentary Ce anomaly records should be made with appropriate caution.