Mechanism of olivine and glass alteration under experimental H2O-CO2 based supercritical gas: Application to modern and ancient Venus
Résumé
Extreme conditions encountered in some geological contexts (deep serpentinization,interaction of Venus atmosphere with its basaltic surface, volcanic degassing) activate mechanismsand rates of silicate alteration that are poorly understood. In the present study, we investigate themechanisms of mineral reactions in a natural geological system at high temperature, under conditionswhere the low solvation of cations by fluids likely promotes surface reactions such as surface diffusionand/or local recrystallization. We focus on vitreous glasses and olivine, reputed to be the mostalterable phases in volcanic rocks, by reacting samples for one week in a Ni-based alloy experimentalvessel. For the framework of our experimental study, we chose to apply the deep atmosphereconditions on Venus: 470°C and 90 bar of reconstituted Venus-like gas. We also tested the effect ofwater (Early Venus or wet volcanic degassing) by adding water vapor at up to 320 bar total pressure.The mineral reactions affecting the samples were identified by a set of spectroscopic surface analysesof the altered samples: Scanning Electron Microscopy, Energy Dispersive X-ray Spectroscopy, X-RayDiffraction in grazing incidence mode, X-ray Photo electron Spectroscopy and Raman spectroscopy.Samples of obsidian and tholeiitic glasses are found to be sensitive to a threshold waterpressure, depending on glass composition, below which the reaction is limited to some elementalmobility in the glass (alkali enrichment, calcium loss) leading to a possibly more stable surface layerof tens to hundreds of microns. Above this threshold water pressure (ca. 50 bar H2O for the obsidianbut >250 bar H2O for the tholeiitic glass), water promotes the depolymerization of the glass and thecrystallization of stable minerals. This crystalline rim is less protective that the chemically modifiedlayer.Olivine samples react differently depending on whether the olivine is isolated or included in abasaltic rock. In the latter case only, iron coatings are formed, which are identified as hematite,suggesting that this phase is not fed by olivine itself but rather by surface diffusion from neighboringFe-rich phases. This supports the conclusions from experimental studies and orbital observations onthe short-term visibility of unaltered olivine in Venus lava flows: such a coating is enhanced when Febearingminerals are in the proximity of olivine. Under high water vapor pressure, Fe-bearing talc (andnot serpentine) forms by a likely topotactic reaction that also incorporates silica from the gas. This talclayer may form a protective layer, implying that serpentinization of ultramafic rocks at hightemperature may not be as prevalent as one might think in a gas-dominated system like the EarlyVenus surface.