Water/Rock Experiments to 300ºC and Comparisons to Chemical Interactions in Active Geothermal Systems

Abstract

Basaltic rocks have been reacted with synthetic groundwater in closed-system, Dickson-type rocking autoclaves for periods of up to 5,000 hours. The experiments were carried out isothermally at 100°, 200°, and 300° C, 300 bars, and an initial water/rock mass ratio of 10. These conditions were intended to simulate those present in active geothermal systems. Fluid compositions changed rapidly during the early stages of the experiments. In the long-term, however, most species approached steady-­state values. At that stage, temperature appears to be the most important factor controlling fluid composition; the effect of rock type and starting fluid composition were far less significant. Comparison of long-term, stabilized fluid compositions with those produced in other dilute-water/whole-rock experiments (basaltic and non-basaltic) shows that experiments conducted at the same temperature attain similar overall fluid chemistries. Such behavior may reflect an approach to equilibrium of these fluids with thermochemically-similar alteration mineral assemblages. Stable fluid chemistries produced in these and other dilute­-water/whole-rock experiments (80 °-300 ° C) were then compared with reservoir data from several geothermal fields. The comparisons show that experiments of this type can reproduce many properties of geothermal reservoir fluids. Like geothermal fluids, compositional parameters (pH, cation/proton ratios, cation/cation ratios, and neutral species concentrations) for species in experimentally derived fluids are temperature dependent but are relatively independent of rock, water, and water/rock mass ratio. At lower temperatures (less than or equal to 250° C), many of these experimental parameter-temperature trends agree with geothermal trends. However, at higher temperatures ( greater than or equal to 250° C), the geothermal fluids have calculated high-temperature pH values 1 to 2 pH units lower than do the experimentally-derived fluids. This produces a consistent offset between values of experimental and natural cation/proton activity ratios. Possible causes for the offsets include: pervasive metastable mineral formation in experiments; and absence, in experiments, of the equivalent of a magmatic or metamorphic gas input. Results from modified water/rock experiments indicate that, while metastability may be a partial cause for the offsets, geothermal fluid parameters can be duplicated by addition of CO2. Furthermore, maintenance of high /CO2, low pH conditions required a CO2 flux. Magmatic gas flux is not always considered in chemical models of geothermal systems and its requirement in the experiments may indicate that E CO2 should be taken as an independent variable. Finally, comparable offsets could also be caused by the addition of exotic co2 or other acid gases during upflow or sampling of wet steam discharges. Field influences such as these would give rise to lower than actual reservoir pH's. Thus the offsets could indicate that some geothermal fluid pH's are more basic than commonly calculated and closer to those attained in water/rock experiments. Water/rock experiments are capable of simulating many of the chemical features of geothermal reservoir fluids but interpretation of experimental data is not straightforward. The closed-system and relatively short-term nature of the experiments must be considered when making comparisons with natural phenomena. Of equal importance is a clear understanding of the natural phenomena being modeled.