Carbonation of an active serpentinization system

Carbonation of serpentinite has been invoked to be a promising tool to mitigate large-scale CO2 emissions, however, monitoring the reaction progress during or after CO2-injection and the interpretation of rapidly evolving fluid-rock equilibria remains a critical but challenging task. We report on a hydrothermal experiment where CO2 was injected into an ongoing serpentinization system in order to assess the changes in fluid chemistry and mineralogy during carbonation. In a first step olivine (Fo90) was reacted with a fluid of seawater chlorinity at 300 °C and 350 bars and fluid-to-rock mass ratio of 2. Under these conditions serpentinization of olivine is very rapid and causes the formation of serpentine, brucite and minor amounts of magnetite. Several fluid samples were taken and immediately analyzed for aqueous silica (SiO2,aq), hydrogen (H2,aq) and pH to monitor the reaction progress. As soon as the serpentine-brucite equilibrium was reached we lowered the temperature to 230°C to facilitate the subsequent carbonation of serpentine, brucite and olivine. The lower temperature was used since carbonation reactions appear to be more rapid and equilibrium CO2 levels are lower, facilitating carbonation reactions. Next, we injected about 9 milimoles of CO2 into the flexible-cell hydrothermal apparatus resulting in a dissolved concentration of about 180 mM CO2,aq. The injection of CO2 caused a drastic change in fluid composition. Within six hours the pH decreased from 9 to 6, while the increased levels of SiO2,aq and CO2,aq indicate talc-magnesite saturation. Two days after the injection the concentrations of SiO2,aq and CO2,aq increased to quartz-magnesite saturation. Subsequently SiO2,aq and CO2,aq decreased to values close to the serpentine-talc-magnesite quasi-invariant point and remained virtually fixed until the experiment was opened after 91 days. The solid reaction products were analyzed using a field emission SEM equipped with an Oxford EDS system. In agreement with the fluid chemistry the secondary mineralogy consists of serpentine, talc, magnesite and traces of magnetite; brucite and quartz are absent. Although relict olivine is present at the end of the experiment the fluid chemistry rapidly responded to the dominating secondary mineralogy, as suggested by the lack of quartz. This experiment shows how the fluid chemistry can be used to remotely monitor changes in mineralogy during carbonation of ultramafic rocks, changes that may be difficult to monitor otherwise.