Hydrous ferric oxide precipitation in the presence of nonmetabolizing bacteria: Constraints on the mechanism of a biotic effect

We have used room temperature and cryogenic Fe-57 Mossbauer spectroscopy, powder X-ray diffraction (pXRD), mineral magnetometry, and transmission electron microscopy (TEM), to study the synthetic precipitation of hydrous ferric oxides (HFOs) prepared either in the absence (abiotic, a-HFO) or presence (biotic. b-HFO) of nonmetabolizing bacterial cells (Bacillus subtilis or Bacillus licheniformis, similar to10(8) cells/mL) and under otherwise identical chemical conditions, starting from Fe(II) (10(-2), 10(-3), or 10(-4) mol/L) under open oxic conditions and at different pH (6-9). We have also performed the first Mossbauer spectroscopy measurements of bacterial cell wall (Bacillus subtilis) surface complexed Fe, where Fe(III) (10(-3.5)-10(-4.5) mol/L) was added to a fixed concentration of cells (similar to10(8) cells/mL) under open oxic condition; and at various pH (2.5-4.3). We find that non-metabolic bacterial cell wall surface complexation of Fe is not passive in that it affects Fe speciation in at least two ways: (1) it can reduce Fe(III) to sorbed-Fe2+ by a proposed steric and charge transfer effect and (2) it stabilizes Fe(II) as sorbed-Fe2+ against ambient oxidation. The cell wall sorption of Fe occurs in a manner that is not compatible with incorporation into the HFO structure (different coordination environment and stabilization of the ferrous state) and the cell wall-sorbed Fe is not chemically bonded to the HFO particle when they coexist (the sorbed Fe is not magnetically polarized by the HFO particle in its magnetically ordered state). This invalidates the concept that sorption is the first step in a heterogeneous nucleation of HFO onto bacterial cell walls. Both the a-HFOs and the b-HFOs are predominantly varieties of ferrihydrite (Fh), often containing admixtures of nanophase lepidocrocite (nLp). vel they show significant abiotic/biotic differences: Biotic Fh has less intraparticle (including surface region) atomic order (Mossbauer quadrupole splitting), smaller primary particle size (magnetometry blocking temperature), weaker Fe to particle bond strength (Mossbauer center shift), and no six-line Fh (6L-Fh) admixture (pXRD, magnetometry). Contrary to current belief, we find that 6L-Fh appears to be precipitated directly, under a-HFO conditions, from either Fe(II) or Fe(III), and depending on Fe concentration and pH, whereas the presence of bacteria disables all such 6L-Fh precipitation and produces two-line Fh (2L-Fh)-like biotic coprecipitates. Given the nature of the differences between a-HFO and b-HFO and their synthesis condition dependences, several biotic precipitation mechanisms (template effect, near-cell environment effect, catalyzed nucleation and/or growth effect, and substrate-based coprecipitation) are ruled out. The prevailing present view of a template or heterogeneous nucleation barrier reduction effect, in particular. is shown not to be the cause of the large observed biotic effects on the resulting HFOs. The only proposed mechanism (relevant to Fh) that is consistent with all our observations is coprecipitation with and possible surface poisoning by ancillary bacteriagenic compounds. That bacterial cell wall functional groups are redox active and the characteristics of biotic (i.e.. natural) HFOs compared to those of abiotic (i.e., synthetic) HFOs have several possible biogeochemical implications regarding Fe cycling, in (he photic zones of water columns in particular. Copyright (C) 2005 Elsevier Ltd.