EXPERIMENTS USING PRE-FRACTURED QUARTZ TO TRAP VOLATILE PHASES: CAN WE BELIEVE THE DATA?

The relative ore-forming potential of magmatic vapor and brine remains elusive. PVTX phase relations in the NaCl-H₂O system, an analogue for natural ore forming fluids, indicate that in a vapor + brine system the concentration of any chloride-complexed metal should be higher in the brine phase. However, data from natural vapor + brine fluid inclusion assemblages indicate that the vapor can contain higher concentrations of some metals. Over the past two decades geologists have increasingly used experimental techniques that allow the trapping of coexisting vapor + brine as fluid inclusions in pre-fractured quartz to quantify the partitioning behavior of metals in melt-vapor-brine systems as a function of pressure, temperature, oxygen fugacity, inter alia. The solute load of such manufactured fluid inclusions can be quantified by analytical techniques such as in situ LA-ICP-MS, micro-Raman and SXRF. A major question surrounding the data from these studies is whether or not the volatile phases trapped as synthetic fluid inclusions in quartz had fully equilibrated with the charge at the experimental conditions. To answer this question, we performed experiments at 800°C, 100 and 110 MPa, in a Au-saturated melt-vapor-brine system and used in situ LA-ICP-MS and INAA to quantify the Au concentration of synthetic brine inclusions trapped in quartz and melt vesicles, respectively. It should be noted that the melt vesicles contain brine that was trapped as the melt was cooled through the glass transition state. The Au concentrations of the synthetic fluid inclusions in both quartz and melt are on the order of 30 µg/g. Considering the very different nature of the fluid inclusions, the fact that the Au concentrations overlap clearly suggests that quartz heals on a slow enough time scale to permit entrapment of equilibrated fluids. These data evince that synthetic fluid inclusions trapped in quartz provide an accurate estimate of the solute load of experimental volatile phases.