The generation of H2 during serpentinization of ultramafic oceanic crust along Earth's mid-ocean ridges provides a key energy source for the deep biosphere and has important implications for metal transport and mineral stability in the subseafloor hydrothermal system. Nonetheless, the driving forces and rates of the coupled processes of Fe partitioning and H2 generation during serpentinization are complicated by many factors, but especially, the activity of SiO2(aq) in the serpentinizing fluid. To examine the role of this key variable in H2 generation and Fe partitioning between magnetite and Fe-serpentine during serpentinization, San Carlos olivine was reacted with a NaCl-SiO2(aq)-bearing fluid at 300 °C and 500 bars for approximately 90 days using flexible gold cell hydrothermal equipment. Time-series changes in solution chemistry and dissolved H2 are coupled with magnetic susceptibility, Mössbauer spectroscopy, and Fe isotope measurements of reactant olivine and coexisting alteration products to provide a complete description of Fe mass transfer and oxidation as a function of reaction progress. Utilization of talc as reactant not only provided a source of SiO2(aq) for serpentinization, but effectively prevented brucite formation, in keeping with constraints imposed by many natural hydrothermal systems, which are inherently open systems. Moreover, this experimental strategy allowed Fe derived from olivine dissolution to transfer completely to serpentine and magnetite, facilitating reaction path modeling. Initial changes in solution chemistry indicate “buffering” of dissolved SiO2(aq), owing to the apparently balanced rates of olivine dissolution and serpentine precipitation. During this stage of reaction, a negligible amount of H2 is generated, suggesting that Fe2+-serpentine is stable and the dominant Fe-bearing product. With continued reaction, the rate of SiO2(aq) production lessens likely due to reduction in surface area of talc, such that the continued hydration and hydrolysis of olivine titrates the previously buffered activity of SiO2(aq), eventually approaching serpentine-brucite coexistence. During this secondary stage of reaction, the rate of H2 generation abruptly increases, which reflects the formation of magnetite and Fe3+-serpentine. Temporal changes in the isotope composition of dissolved Fe, throughout these stages, likely reflect different mineral alteration and formation processes, where initially, 1) preferential mass transfer of isotopically light Fe into solution occurred during olivine alteration, without considerable oxidation, as demonstrated by the low H2 production rate, and, 2) the preferential removal of isotopically light Fe from solution in connection with enhanced magnetite and H2 formation during the later stages of reaction. Post experimental analysis of the Fe isotopic composition of Fe-bearing minerals demonstrates statistically negligible enrichment of δ56Fe of magnetite relative to olivine and serpentine, consistent with observations from natural altered oceanic crust. Observations from this study demonstrate the utility of using coupled chemical, isotopic, and analytical approaches to ascertain time series changes in rates of mass transfer and mechanisms of oxidation of Fe during serpentinization of olivine. Ultimately, these insights provide important constraints for the chemical and Fe isotopic evolution of fluids and minerals in ultramafic hosted hydrothermal systems at mid-ocean ridges.
Bibliographical noteFunding Information:
The authors thank Rick Knurr for providing detailed chemical analyses of the solution samples presented in this study and Dr. Nicholas Pester for providing assistance with H 2 analysis by GC. The authors would also like to thank the Institute of Rock Magnetism, Peter Solheid, and Dr. Hongluo Zhong, and Professor Bruce Moskowitz for providing help with Mössbauer analysis and modeling. The authors would like to thank Gen Ito for meticulously grinding and cleaning the olivine and talc reactants during his tenure at Minnesota. We thank Dr. Nicholas Seaton for providing his expertise in SEM imaging. Finally, we thank the editor Dr. David H. Tilton and the reviewer, Dr. Wolfgang Bach, for their constructive comments and suggestions, which made this manuscript clearer. This study was funded by the NSF-OCE awards 1232587 (DMB) and 1232704 (WES).
- Fe isotopes
- Fe mass transfer and oxidation
- H generation
- Mid-ocean ridge hydrothermal systems