Iron (hydr)oxides not only serve as potent sorbents and repositories for nutrients and contaminants but also provide a terminal electron acceptor for microbial respiration. The microbial reduction of Fe (hydr)oxides and the subsequent secondary solid-phase transformations will, therefore, have a profound influence on the biogeochemical cycling of Fe as well as associated metals. Here we elucidate the pathways and mechanisms of secondary mineralization during dissimilatory iron reduction by a common iron-reducing bacterium, Shewanella putrefaciens (strain CN32), of 2-line ferrihydrite under advective flow conditions. Secondary mineralization of ferrihydrite occurs via a coupled, biotic-abiotic pathway primarily resulting in the production of magnetite and goethite with minor amounts of green rust. Operating mineralization pathways are driven by competing abiotic reactions of bacterially generated ferrous iron with the ferrihydrite surface. Subsequent to the initial sorption of ferrous iron on ferrihydrite, goethite (via dissolution/reprecipitation) and/or magnetite (via solid-state conversion) precipitation ensues resulting in the spatial coupling of both goethite and magnetite with the ferrihydrite surface. The distribution of goethite and magnetite within the column is dictated, in large part, by flow-induced ferrous Fe profiles. While goethite precipitation occurs over a large Fe(II) concentration range, magnetite accumulation is only observed at concentrations exceeding 0.3 mmol/L (equivalent to 0.5 mmol Fe[II]/g ferrihydrite) following 16 d of reaction. Consequently, transportregulated ferrous Fe profiles result in a progression of magnetite levels downgradient within the column. Declining microbial reduction over time results in lower Fe(II) concentrations and a subsequent shift in magnetite precipitation mechanisms from nucleation to crystal growth. While the initial precipitation rate of goethite exceeds that of magnetite, continued growth is inhibited by magnetite formation, potentially a result of lower Fe(III) activity. Conversely, the presence of lower initial Fe(II) concentrations followed by higher concentrations promotes goethite accumulation and inhibits magnetite precipitation even when Fe(II) concentrations later increase, thus revealing the importance of both the rate of Fe(II) generation and flow-induced Fe(II) profiles. As such, the operating secondary mineralization pathways following reductive dissolution of ferrihydrite at a given pH are governed principally by flow-regulated Fe(II) concentration, which drives mineral precipitation kinetics and selection of competing mineral pathways.

Owing to its high surface area and intrinsic reactivity, ferrihydrite serves as a dominant sink for numerous metals and nutrients in surface environments and is a potentially important terminal electron acceptor for microbial respiration. Introduction of Fe (II), by reductive dissolution of Fe(III) minerals, for example, converts ferrihydrite to Fe phases varying in their retention and reducing capacity. While Fe(II) concentration is the master variable dictating secondary mineralization pathways of ferrihydrite, here we reveal thatthe kinetics of conversion and ultimate mineral assemblage are a function of competing mineralization pathways influenced by pH and stabilizing ligands. Reaction of Fe(II) with ferrihydrite results in the precipitation of goethite, lepidocrocite, and magnetite. The three phases vary in their precipitation extent, rate, and residence time, all of which are primarily a function of Fe(II) concentration and ligand type (Cl, SO4, CO3). While lepidocrocite and goethite precipitate over a large Fe(II) concentration range, magnetite accumulation is only observed at surface loadings greater than 1.0 mmol Fe(II)/g ferrihydrite (in the absence of bicarbonate). Precipitation of magnetite induces the dissolution of lepidocrocite (presence of Cl) or goethite (presence of SO4), allowing for Fe(III)-dependent crystal growth. The rate of magnetite precipitation is a function of the relative proportions of goethite to lepidocrocite; the lower solubility of the former Fe (hydr)oxide slows magnetite precipitation. A one unit pH deviation from 7, however, either impedes (pH 6) or enhances (pH 8) magnetite precipitation. In the absence of magnetite nucleation, lepidocrocite and goethite continue to precipitate at the expense of ferrihydrite with near complete conversion within hours, the relative proportions of the two hydroxides dependent upon the ligand present. Goethite also continues to precipitate at the expense of lepidocrocite in the absence of chloride. In fact, the rate and extent of both goethite and magnetite precipitation are influenced by conditions conducive to the production and stability of lepidocrocite. Thus, predicting the secondary mineralization of ferrihydrite, a process having sweeping influences on contaminant/nutrient dynamics, will need to take into consideration kinetic restraints and transient precursor phases (e.g., lepidocrocite) that influence ensuing reaction pathways.

Tens of millions of people in south and southeast Asia routinely consume ground water that has unsafe arsenic levels. Arsenic is naturally derived from eroded Himalayan sediments, and is believed to enter solution following reductive release from solid phases under anaerobic conditions. However, the processes governing aqueous concentrations and locations of arsenic release to pore water remain unresolved, limiting our ability to predict arsenic concentrations spatially (between wells) and temporally (future concentrations) and to assess the impact of human activities on the arsenic problem. This uncertainty is partly attributed to a poor understanding of groundwater flow paths altered by extensive irrigation pumping in the Ganges-Brahmaputra delta, where most research has focused. Here, using hydrologic and (bio)geochemical measurements, we show that on the minimally disturbed Mekong delta of Cambodia, arsenic is released from near-surface, river-derived sediments and transported, on a centennial timescale, through the underlying aquifer back to the river. Owing to similarities in geologic deposition, aquifer source rock and regional hydrologic gradients, our results represent a model for understanding pre-disturbance conditions for other major deltas in Asia. Furthermore, the observation of strong hydrologic influence on arsenic behaviour indicates that release and transport of arsenic are sensitive to continuing and impending anthropogenic disturbances. In particular, groundwater pumping for irrigation, changes in agricultural practices, sediment excavation, levee construction and upstream dam installations will alter the hydraulic regime and/or arsenic source material and, by extension, influence groundwater arsenic concentrations and the future of this health problem.

No abstract