Experiments were performed under oxic and anoxic conditions to constrain the influence of oxygen on Se reduction. Different concentrations of organic carbon sources and selenate were examined since they are key limiting factors for microbial Se reduction. Temporally resolved data on selenate and selenite concentrations at the reactor outflow and spatially resolved data on solid phase Se species within the aggregates was collected. We show that the coupling between physical and biogeochemical factors in our systems leads to increasing concentrations of reduced selenium in the solid phase towards the core of aggregates under a diverse set of conditions and that the investigated experimental factors have a predictable yet aggregate dynamics specific impact on reduction rates. Insights gained here have the potential to improve our capacity to predict Se transport and attenuation in structured soils, thereby facilitating the management of selenium contaminated soils.Effluent samples and solid phase extraction products from flow-through experiments were analyzed for total Se and selenite concentrations using a Perkin Elmer 5300 DV inductively coupled plasma-optical emission spectrometer . Total Se was determined using the ICP-OES operating in regular mode and selenite concentrations were measured via ICP-OES coupled to a hydride generation set-up based on the phase separator set-up described by Bosnak and Davidowski and with flow rates as those used by Brooks in manifold 2 .
A sample injection time of 120 seconds prior to optical emission reading and a rinse time of at least 30 seconds with an additional 100 seconds following samples exceeding 150 ppb were used,procona flower transport containers due to a significant memory effect. Elemental and speciation standards were prepared from certified reference stock solutions in a matrix matching the composition of measured samples . Selenate concentrations were computed as the difference between total selenium and selenite concentrations. All selenite concentrations measured in effluent samples were corrected from trace concentrations in the input solutions . Selenite export rates for each reactor were then computed by multiplying effluent concentrations at quasi-steady-state by the flow-rate and normalizing by aggregate dry mass. These quasi-steady-state concentrations were obtained by averaging across a timespan of at least 64 h, or eight consecutive 8 h samples.We used selenite export rates as a proxy for bulk selenium reduction. While as discussed in section 3.1 total reduction rates are expected to be somewhat higher then selenite export rates due to solid phase deposition inside the aggregate , the export rates do provide a hard lower bound for reduction rates. Selenite export rates were lower under oxic than under anoxic conditions, indicating lower selenium reduction rates in the presence of oxygen. Lower selenium reduction rates reflect the ability of E. cloacae to reduce selenate in the presence of oxygen, but at lower rates than in anaerobic conditions . However, since there is no evidence that T. selenatis can carry out selenate reduction in the presence of oxygen, the observed selenite export for the oxic experiments is an indication of the existence of anoxic micro-zones within the aggregate.
Anoxic zones may arise if the consumption of oxygen by T. selenatis via oxic respiration exceeds diffusive supply of oxygen into the aggregate from the surrounding solution. These zones enable selenium reduction within the aggregate, similar to what has long been known in the context of denitrification within natural soil aggregates . Reactive transport modelling conducted for analogous systems investigating arsenic desorption demonstrates that anoxic zones arise from aerobic respiration in artificial aggregates of this size and composition. The lower solid phase selenate concentrations observed under anoxic conditions as compared to oxic conditions indicate that transport of selenate from the surrounding solution into the aggregate is slower than consumption within the aggregate via reduction. This is further evidence that soil structure and diffusion-dominated transport within aggregates can impact selenium reduction. Aggregates may thus promote selenium reduction and retention in a soil by creating anoxic environments conducive to the process and drawing mobile, bio-available selenium from the surrounding soil solution.In this work we have shown that aggregate scale heterogeneity arising from microbial selenium reduction can be observed within simple artificial systems. Spatial heterogeneity comes into play in creating anoxic microzones that enable selenium reduction under conditions that are otherwise unfavorable and spatial heterogeneity can be observed in the increasing concentrations of the reduced product towards the core of aggregates. Aggregates may thus improve selenium retention in a soil by creating anoxic environments conducive to the reduction process and by accumulating reduced selenium at their cores. Promoting soil aggregation on seleniferous agricultural soils, through the addition of organic matter and reduction of tillage, may be an effective management practice to reduce the impacts of selenium contaminated drainage water on downstream ecosystems. This result may also be useful in the development of treatment systems designed to remove selenium from waste water, since they suggest that the usage of a dual-porosity flow-through system may be fruitful approach. The application of aggregate-reactor systems to microbial selenium reduction was thus successful in shedding light on the dynamics that control reduction in structured soils and led to the discovery of a gradient in reduced selenium concentrations at the aggregate scale that is likely to also occur in aggregates under field conditions. We have also shown that while reduction rates at the aggregate scale depend on the same controlling factors that have been found to be important in bulk , a simple bulk model misrepresents aggregate scale dynamics of selenium reduction under oxic conditions.
Finally we showed that metabolic differences between different selenium reducing microbes are reflected at the aggregate scale. While these results highlight the need to consider aggregate scale processes when discussing selenium mobility in soils, it is not evident how they will translate into the elevated complexity of natural systems. This question needs to be attacked on three fronts: The complexity of model systems needs to be incrementally increased to approach that of the field. We used 3D aggregate systems rather than simple flow through columns or 2D systems and added a level of complexity by including a sorbent phase in select experiments. Future work however, could focus on variations in aggregate size and shape, mixed microbial communities and unsaturated flow.We derived a simple relationship from our results that predicts an increase of reduced selenium concentrations with aggregate size in structured soils undergoing selenium reduction. This represents a testable hypothesis that could guide future field investigations. Reactive transport modeling must be employed to help interpret experimental results and bridge the gap between lab and field. The persistence of the radial solid phase gradients in reduced selenium here observed across all conditions tested is striking. A reactive transport model could help clarify whether diffusive intra-aggregate transport of selenate, selenite and electron donors coupled with microbial reduction would generally be expected to produce the observed pattern in a spherical geometry. If so, the model could be modified to test a broader range of geometries, scales, and biogeochemical processes that are expected under field conditions.Selenium is a trace element of great ecological significance,procona valencia due to its dual role as both an essential nutrient and environmental contaminant. The intake range between deficiency and toxicity of selenium for most animals is exceedingly small . For example, humans need around 40 µg/day for optimal health; however, an intake of more than 400 µg/day can cause selenosis . Selenium is heterogeneously distributed across terrestrial landscapes, with seleniferous and selenium-deficient soils sometimes occurring as close as 20 km from one another . The irrigation of seleniferous soils can lead to the production of sub-surface drainage water containing high selenium concentrations that cause ecological damage in downstream aquatic ecosystems. This process was first recognized in the late 1980s after the occurrence of numerous deaths and embryonic deformities in waterfowl populations near the agricultural evaporation ponds of the Kesterson reservoir, a former wildlife refuge in California . In the Western United States alone, nearly 400,000 km2 may be at risk of irrigation-induced selenium contamination and similar problems have also been observed in Canada, Egypt, Israel, and Mexico . Enhanced understanding of selenium transport and retention in surface soils will improve the management of seleniferous soils to diminish the risk of contamination for downstream ecosystems. The transport and biogeochemical behavior of selenium depend greatly on its chemical speciation. Whereas selenium occurs naturally in four oxidation states: Se, Se, Se and Se, the primary oxidation states associated with irrigation-induced contamination, Se and Se, occur as bio-available oxyanions: selenate and selenite . Elemental selenium is solid and immobile, whereas Se occurs in soluble and bio-available organo-selenides or in gaseous methylated forms . Transformations between these various species are kinetically hindered and thus the speciation of selenium is poorly predicted by thermodynamic models .
Redox reactions are catalyzed primarily by microorganisms, the transformation of Se to Se and Se for example, is driven predominantly by microbial dissimilatory reduction . This reduction reaction yielding solid Se is an important attenuation pathway for selenium in surface environments . The known microorganisms capable of selenium reduction are scattered across a very diverse set of prokaryotic taxa . We chose Enterobacter cloacae SLD1a-1 as the benchmark organism for the reactive transport model presented in this paper. Since its original isolation from seleniferous agricultural drainage water, this γ-Proteobacterium has received a particularly large amount of research attention . E. cloacae is known to grow both aerobically and anaerobically using a variety of electron donors, including pyruvate . It can reduce selenate via selenite to solid elemental Se and its membrane-bound selenate reductase has been identified and purified . Furthermore, it has been used as a model organism in several studies investigating the molecular genetics of selenium reduction and has been considered as a candidate for bioremediation schemes . E. cloacae can carry out selenium reduction under oxic conditions, though at rates that are an order of magnitude lower than under anoxic conditions . This inhibition is in accordance with thermodynamic expectations, since at neutral pH reduction of selenate to selenite is not energetically favored above a pE of 7.5 placing it generally between nitrate and manganese oxide on the redox ladder . Unsurprisingly, oxygen inhibition of microbial selenium reduction has also been observed in environmental samples . Selenium reduction rates are thus sensitive to local redox potential. Redox conditions in soils are known to display a large amount of spatial heterogeneity, owing to the heterogeneous size and distribution of pore spaces in a soil’s physical structure . A common approach to model this spatial heterogeneity in mechanistic bio-geochemical models has been to represent the soil matrix as a series of spherical microporous structures or aggregates . Soil aggregates are mm- to cm-sized structural units of mineral particles bound by roots, fungal hyphae, and organic matter that naturally occur in most soils . Whereas a variety of different shapes exist in nature, granular aggregates are particularly common in surface soils . Aggregates represent well-defined natural systems in which to study the impact of mm- to cm-scale spatial heterogeneity on bio-geochemical reactions . Dissolved chemical concentrations within aggregates depend on the local coupling of reactions and transport. Whereas fast advective solute transport is prevalent in the inter-aggregate macropores, transport inside aggregates is mostly diffusive, due to the low permeability of the microporous structures. In conjunction with local microbial metabolic activity this often leads to the formation of steep chemical gradients within aggregates , as the consumption of a chemical species inside an aggregate outpaces the diffusive supply from the surrounding macropore . Full transitions from oxic to anoxic have been observed in aggregates as small as 4 mm in diameter . Tokunaga et al. demonstrated that such anoxic zones can influence the mm-scale distribution and mobility of selenium based on synthetic, flat aggregate no-flow systems . We previously developed a system of saturated-flow reactors containing single idealized soil aggregates, for the investigation of aggregate-scale gradients in the microbial reduction and secondary mineral formation of iron and a mathematical model describing the coupled transport and reactions of iron within the system . This unique experimental approach, emulating field-like transport conditions at the aggregate-scale while allowing for controlled experiments, was later applied to investigations of arsenic mobility and heterogeneity in microbial selenium reduction .