T36 Basic and Applied Aspects of Clays and Clay Minerals in Continental Settings (Posters)

Sunday, 9 October 2011
9:00 AM - 6:00 PM
Hall C

A COMPARISON BETWEEN POPULAR GEOCHEMICAL THERMODYNAMIC DATA SETS AND NEAR-EQUILIBRIUM EXPERIMENTAL DATA

Paper number: T36-8
Program: Technical Sessions
Time: 9:00 AM – 6:00 PM

Authors

Abstract

The conversion of alkali feldspars to lower free energy clay mineral phases plays a significant role in the mineralogy and hydrogeology of an array of natural and engineered systems, ranging from granite weathering to CO2 sequestration in feldspathic sandstones. Discussions in the geochemical literature from the 1970s to present have focused on discrepancies in the aluminum mineral and aqueous species data provided by Helgeson and others in 1978, which was eventually included in SUPCRT92. Geochemists are increasingly turning to other sources, particularly Holland and Powell’s 1998 dataset (HP98) for aluminum mineral data, Rimstidt’s 1997 (R97) low T quartz solubility measurements for SiO2(aq) data, and Tagirov and Schott’s 2001 (TS01) dataset for Al speciation data. Here, we use results from carefully performed mineral solubility experiments involving a range of Al-bearing chemical systems and temperatures to assess the accuracy of predictions created using these thermodynamic datasets.

When evaluated against 80˚C, pH 8.8, near-equilibrium albite solution compositions from Burch et al. (1993), the HP98 data with and without TS01 and R97, the SUPCRT92 data with and without TS01 and R97, and the PHREEQC data sets produce saturation indices (SIs) of 1.5 – 2.1, 1.0 – 1.5, and -0.07 – -0.40, respectively. When compared with 150˚C, pH 9, P=40 bars, near-equilibrium K-feldspar solution compositions (approached from under-saturation) from Gautier et al. (1994), the three datasets for maximum microcline (HP98 does not include K-Feldspar data) produce SIs of -0.37 – -0.02 and -0.88 – -0.52, and N/A (150˚C is out of the range of PHREEQC), respectively. When compared with 80˚C, pH 3, equilibrium kaolinite solution compositions (approached from under- and super-saturation) from Nagy et al. (1991), the three datasets produce SIs of 3.1 – 4.2, 1.2 – 2.5, and 0.3 – 0.5, respectively. Overall, HP98 produces SIs around 0.5 to 2 SI units higher than SUPCRT92 for all cases, which has implications for both equilibrium- and kinetics-based predictions. Based on the results of these comparisons, the conversion to HP98 mineral data is unwarranted and altering datasets to reflect TS01 and R97 data produces the most consistently accurate results.