Dissolution rates of phyllosilicates as a function of bacterial metabolic diversity

• Published in: Geochimica et cosmochimica acta 2010-10, Vol.74 (19), p.5478-5493
• Main Authors: Balland, C., Poszwa, A., Leyval, C., Mustin, C.
• Format: Article
• Language: English
• Subjects: Bacteria; Continental interfaces, environment; Earth Sciences; Geochemistry; Sciences of the Universe
• ISSN: 0016-7037; 1872-9533
• DOI: 10.1016/j.gca.2010.06.022

Weathering experiments using biotite and phlogopite in the presence of bacteria were conducted to better understand biotic dissolution kinetics and processes (proton- and ligand-promoted dissolution) under aerobic conditions. Miniature batch reactors (300 μl in microplate wells) were used at 24 °C for 3 days with and without bacterial strains. Abiotic experiments were performed with organic and nitric acids in order to calibrate the biotite–phlogopite chemical dissolution. An empirical model was used to fit the pH dependence for iron release rate (r Fe) considering the influence of both protons and ligands from acidic to neutral conditions (pH ranging from 3 to 7): r Fe = k H ( a H + ) m + k L ( a L ) 1 where k is the apparent rate constant, a H+ and a L are the activities of protons and ligands, and m and l are the reaction orders. For both minerals in most cases at a given pH, the iron release rates in the presence of bacteria were in good agreement with rates determined by the chemical model and could be explained by a combination of proton- and ligand-promoted processes. Bacteria affect mineral dissolution and iron release rates through the quantities and nature of the organic acids they produce. Three domains were differentiated and proposed as biochemical models of mica dissolution: (1) below pH 3, only proton-promoted dissolution occurred, (2) in weakly acidic solutions both ligand- and proton-promoted mechanisms were involved, and (3) iron immobilization occured, at pH values greater than 4 for biotite and greater than 5 for phlogopite. This model allows us to distinguish the “weathering pattern phenotypes” of strains. Bacteria that are isolated from horizons poor in carbon appear more efficient at weathering micas than bacterial strains isolated from environments rich in carbon. Moreover, our results suggest that the mineral could exert a control on the release of organic acids and the “weathering pattern phenotypes” of bacteria.