No abstract

In the ocean, yttrium and the rare earth elements (YREEs) show nutrient-like vertical profiles. Since the YREEs have no manifest biological function, their removal from solution (scavenging) is probably caused by sorption on particles rather than active microbial uptake, yet the exact nature of these particles is uncertain. An existing theoretical model describes scavenging as an equilibrium between complexation with dissolved inorganic ligands and with functional groups on particle surfaces. This model was able to predict input-normalized (i.e., shale-normalized) YREE abundance patterns in seawater without requiring poorly known parameters like particle concentrations or the site densities of functional groups. Employing well-established stabilities of inorganic YREE complexes, while assuming that the sorbent particles are organic with functional groups represented by a mixture of simple monocarboxylic acids, it reproduced some key features of seawater YREE patterns, specifically the characteristic increase of shale-normalized abundance with atomic number and distinctive anomalies of certain trivalent REEs (La and Gd). The familiar negative Ce anomaly of seawater, however, is due to redox reactions that were not explicitly accounted for.In this study, we refined calculations of YREE solution speciation by adding complexation with desferrioxamine B to gauge the influence of strong organic ligands prevalent near the ocean surface. The scavenging model was then inverted by subtracting high-quality YREE abundance patterns, reported for the open ocean, from this solution speciation to yield an average pattern of relative YREE affinities for the sorbent particles. The resulting affinity pattern is compared with patterns of distribution coefficients, derived from laboratory experiments, for YREE sorption on relevant solid phases including hydrated Fe and Mn oxides (HFO/HMO), calcite, and the green macroalga Ulva lactuca as a substitute for marine organic matter. For deep seawater, the best agreement is observed with HMO, which may thus be the dominant carrier of YREEs to the sediment. Since Ce is catalytically oxidized on manganese oxide surfaces, this could have implications for the evolution of the Ce anomaly. Direct comparisons of distribution coefficient patterns with YREE analyses of suspended particles from the Atlantic Ocean also favor HMO, but this may be dictated in one case by the use of a selective leaching method. For shallow seawater, particularly in the Atlantic Ocean, the best agreement is observed with U. lactuca and calcite, although the latter is offset by the low YREE contents of this biogenic mineral. Matches with HFO are generally less convincing. The aforementioned mixture of simple monocarboxylic acids is not a good proxy for the sorptive properties of marine organic matter.

No abstract