MONITORING SERPENTINIZATION REACTIONS IN OLIVINE IN SITU USING SYNTHETIC FLUID INCLUSIONS

Serpentinization is an important geological process that takes place where ultramafic rocks are exposed to fluid circulation in the oceanic and continental crusts. Despite much recent attention, some aspects of the serpentinization reactions such as the order, timing and direction of reactions are still a matter of considerable debate in the literature.

We used the synthetic fluid inclusion technique to trap fluids of known composition at known P-T conditions in olivine crystals to follow in-situ serpentinization reactions in a closed system. Pre-fractured olivine crystals were loaded into platinum capsules along with a H₂O-NaCl-MgCl₂ fluid of seawater concentration and welded shut (Bodnar & Sterner, 1984; 1987). The loaded capsules then were placed into high-pressure vessels, and P-T were increased to 5.6 kbar and 600 ºC for 21 days. The P-T conditions were selected to trap the fluid inclusions around the 0.83 g/cm³ isochore, in order to have an internal pressure of the fluid inclusions of ~500 bars at ~280 ºC to simulate the serpentinization conditions at slow/ultra slow spreading ridges. After trapping the fluid inclusions the samples were placed into a furnace at ~280 ºC and 1 atm.

Preliminary results show that serpentinization reactions start after a few days in some of the fluid inclusions. Mineralogy was monitored by Raman analysis, and we observed the formation of brucite and serpentine. After 28 days, some of the fluid inclusions consumed nearly all the fluid, leaving the cavity filled with brucite and serpentine. In at least one case, H₂ was detected in the fluid inclusion, showing that the reducing conditions inside the fluid inclusion are similar to what occurs in nature. However, no magnetite was observed in any of the fluid inclusions based on raman analysis, suggesting that the reaction 2(FeO)_rock + H₂O ⟶ (Fe₂O₃)_magnetite + H₂ is not occurring, but Fe³⁺ is maybe incorporated into serpentine or some other phase. This observation concurs with thermodynamic calculations and observations of natural samples in which magnetite is formed by a secondary reaction of brucite and serpentine with more fluid (Bach et al., 2006, Frost & Beard, 2007). More experiments and thermodynamic modelling are being conducted to better understand these reactions and the overall implications of the serpentinization processes.