Hydrogen generation and iron partitioning during experimental serpentinization of an olivine–pyroxene mixture. (1st August 2020)

• Record Type: Journal Article
• Authors: McCollom, Thomas M.; Klein, Frieder; Moskowitz, Bruce; Berquó, Thelma S.; Bach, Wolfgang; Templeton, Alexis S.
• Is Part Of: Geochimica et cosmochimica acta. Volume 282(2020)
• Page Start: 55
• Page End: 75
• Publication Date: 2020-08-01
• Subjects: Geochemistry -- Periodicals; Meteorites -- Periodicals; Géochimie -- Périodiques; Météorites -- Périodiques; Geochemie; Astrochemie; Electronic journals; 551.905
• DOI: 10.1016/j.gca.2020.05.016
• Languages: English
• ISSNs: 0016-7037

Abstract

Abstract: A series of laboratory experiments was conducted to investigate serpentinization of olivine–pyroxene mixtures at 230 °C, with the objective of evaluating the effect of mixed compositions on Fe partitioning among product minerals, H2 generation, and reaction rates. An initial experiment reacted a mixture of 86 wt.% olivine and 14 wt.% orthopyroxene (Opx) with the same initial grain size for 387 days. The experiment resulted in extensive reaction (∼53% conversion), and solids recovered at termination of the experiment were dominated by Fe-bearing chrysotile and relict olivine along with minor brucite and magnetite. Only limited amounts of H2 were generated during the first ∼100 days of the experiment, but the rate of H2 generation then increased sharply coincident with an increase in pH from mildly alkaline to strongly alkaline conditions. Two shorter term experiments with the same reactants (26 and 113 days) produced a mixture of lizardite and talc that formed a thin coating on relict olivine and Opx grains, with virtually no generation of H2 . Comparison of the results with reaction path models indicates that the Opx reacted about two times faster than olivine, which contrasts with some previous studies that suggested olivine should react more rapidly than Opx at the experimental conditions. The models also indicate that the long-term experiment transitioned from producing serpentine ± talc early in the early stages to precipitation of serpentine plus magnetite, with brucite beginning to precipitate only late in the experiment as Opx was depleted. The results indicate that overall reaction of olivine and Opx was initially relatively slow, but reaction rates accelerated substantially when the pH transitioned to strongly alkaline conditions. Serpentine and brucite precipitated from the olivine-Opx mixture had higher Fe contents than observed in olivine-only experiments at mildly alkaline pH, but had comparable Fe contents to reaction of olivine at strongly alkaline pH implying that higher pH may favor greater partitioning of Fe into serpentine and brucite and less into magnetite. Despite the presence of brucite, dissolved silica activities during the long-term olivine-Opx experiment maintained levels well above serpentine-brucite equilibrium. Instead, silica activities converged on levels close to metastable equilibrium between brucite and olivine. It is proposed that silica levels during the experiment may have been regulated by exchange of SiO2 between the fluid and a silica-depleted, brucite-like surface layer on dissolving olivine.