pH as a control on interactions of methanogens and iron reducers

Abstract

A growing body of evidence demonstrates that methanogenesis and Fe(III) reduction can occur simultaneously. However, environmental controls on interactions between each are poorly understood. In this study we considered pH as a control on interactions between Fe(III) reduction and methanogenesis in anoxic sediment bioreactors. The reactors consisted of 100mL of synthetic aqueous media, and 1 g of marsh sediment amended with goethite (1mmol). One set of reactors received acidic media (pH 6), and the other alkaline media (pH 7.5). Each set received media containing acetate (0.25 mM) to serve as an electron donor. Control reactors, deficient in acetate, were also included. We maintained a fluid residence time of 35 days by sampling and feeding the reactors every seven days. For pH 6.0 and pH 7.5 reactors, the measured pH of effluent samples averaged 6.33 and 7.37, respectively. The extent of Fe(III) reduction and methanogenesis varied considerably between each set of reactors. More Fe(III) was reduced in the pH 6 reactors (646.39 μmoles on avg.) than the pH 7.5 reactors (31.32 μmoles on avg.). Conversely, more methane formed in pH 7.5 reactors (127.5 μmoles on avg.) than the pH 6 reactors (78.9 μmoles on avg.). Alkalinity concentrations during the middle and end of the experiment averaged 9.6 meq/L and 5.2 meq/L in pH 6 and pH7.5 reactors, respectively Although much less Fe(III) reduction occurred in pH 7.5 reactors, the relative abundance of Fe(III) reducers in them decreased little from levels observed in the pH 6 reactors. Sequences classified within Geobacter, a genus of bacteria known primarily as dissimilatory metal reducers, accounted for 22% and 13.45% of the sequences in the pH 6 and pH 7.5 reactors and only 0.8% of the sequences in the marsh sediment inoculum. In contrast, sequences classified within orders of methanogens were low in abundance, making up only 0.47% and 1.04% of the sequences in the pH 6 and pH 7.5 reactors, respectively. Mass balance calculations demonstrate that the amount of electron donor consumed by each group varied considerably between the sets of reactors. Expressed as a quantity of acetate, the reactions consumed about 160μM of electron donor each in pH 6 reactors. In contrast, methanogenesis consumed over 30 times more electron donor than Fe(III) reduction in the pH 7.5 reactors. Thus, the results of our experiment indicate that the decrease in electron donor consumption by Fe(III) reduction at basic pH was nearly matched by the increase in electron donor consumption by methanogens. Results of geochemical modeling calculations indicate that more energy was available for Fe(III) reduction in the pH 6.0 reactors than the pH 7.5 reactors, matching variation in Fe(III) reduction rates, and that the density of sorbed ferrous iron was higher in pH 6 reactors than pH 7.5 reactors. Thus, the calculation results are consistent with bioenergetics, but not variation in ferrous iron sorption, as a potential mechanism driving variation in the balance between each reaction with pH.