Abstract
Proton activity is the master variable in many biogeochemical reactions. To control pH, laboratory studies involving redox-sensitive minerals like manganese (Mn) oxides frequently use organic buffers (typically Good's buffers); however, two Good's buffers, HEPES and MES, have been shown to reduce Mn(IV) to Mn(III). Because Mn(III) strongly controls mineral reactivity, avoiding experimental artefacts that increase Mn(III) content is critical to avoid confounding results. Here, we quantified the extent of Mn reduction upon reaction between Mn oxides and several Good's buffers (MES, pK(a) = 6.10; PIPES, pK(a) = 6.76; MOPS, pK(a) = 7.28; HEPES, pK(a) = 7.48) and TRIS (pK(a) = 8.1) buffer. For δ-MnO(2), Mn reduction was rapid, with up to 35% solid-phase Mn(III) generated within 1 h of reaction with Good's buffers; aqueous Mn was minimal in all Good's buffers experiments except those where pH was one unit below the buffer pK(a) and the reaction proceeded for 24 h. Additionally, the extent of Mn reduction after 24 h increased in the order MES < MOPS < PIPES < HEPES << TRIS. Of the variables tested, the initial Mn(II,III) content had the greatest effect on susceptibility to reduction, such that Mn reduction scaled inversely with the initial average oxidation number (AMON) of the oxide. For biogenic Mn oxides, which consist of a mixture of Mn oxides, bacterial cells and extracelluar polymeric substances, the extent of Mn reduction was lower than predicted from experiments using abiotic analogs and may result from biotic re-oxidation of reduced Mn or a difference in the reducibility of abiotic versus biogenic oxides. The results from this study show that organic buffers, including morpholinic and piperazinic Good's buffers and TRIS, should be avoided for pH control in Mn oxide systems due to their ability to transfer electrons to Mn, which modifies the composition and reactivity of these redox-active minerals.