Verification and comparison of two thermodynamic databases through conversion to PHREEQC and multicomponent geothermometrical calculations

The databases SOLTHERM and Yucca Mountain Project (YMP) were converted from GeoT/TOUGHREACT to PHREEQC format, to not only allow a higher accessibility of freely available data for thermodynamic modelling in the temperature range of 0−350 °C at the pressure of water vapour saturation, but also show possible causes of differences between simulations using two programs. Converted databases were verified by comparing the results obtained with the same simulations performed with PHREEQC and GeoT. Four natural geothermal solutions from previously published works were used in the simulations, covering a wide range of temperatures, salinities, pH, and equilibrium minerals. Consecutively, a comparison was performed through geothermometrical calculations for the test solutions, by using the converted databases with PHREEQC and applying the same methods as in the original publications. Minimal differences were found between simulations using the original and converted databases, especially when solutions with low ionic strengths were involved. Differences found using solutions with ionic strengths above 1.0 mol/kg are likely caused by differences in the calculation methods for activity coefficients of dissolved species between converted databases in PHREEQC and original databases in GeoT. The geothermometrical calculations showed a good agreement between published results, using different databases and programs, and results obtained using the transformed databases in PHREEQC, especially considering calculated saturation indices of rock-forming minerals such as quartz, feldspars, calcite and dolomite, and muscovite. A higher deviation was found for clay minerals, as the thermodynamic data for this type of minerals are more variable and differ per type (and version) of database.