Quantifying silica reactivity in subsurface environments: Controls of reaction affinity and solute matrix. 1998 annual progress report

'The authors goal is to develop a quantitative and mechanistic understanding of amorphous silica, SiO2 (am), dissolution kinetics in aqueous solutions. A knowledge of fundamental controls on the reactivity of simple SiO2 bonded phases is the compositional baseline for understanding highly complex silica phases. In the Earth, silicate minerals comprise >70% of the crust and dominate virtually every subsurface system. More importantly for the objectives of this EMSP project, silicate minerals and materials are significant because compositionally complex silicate glasses will become the front line of defense in containing radioactive wastes in the nation''s long term and interim storage strategies (Dove and Icenhower, 1997). To date, the behavior of SiO2 (am) is largely inferred from studies of the better known crystalline polymorphs (e.g. a-quartz). In the first step towards constructing a general model for amorphous silica reactivity in the complex fluid compositions of natural waters, the authors are determining the dissolution behavior as a function of temperature, solution pH and NaCl concentration. With these data they are determining relationships between SiO2 glass structure and dissolution rates in aqueous solutions, as described below. This report outlines the first year''s progress and the resulting publications to date. In this experimental investigation, the dissolution kinetics of SiO2 (am) (fused and flame pyrolysis silica) were measured in solutions over the pH range of 4 to 10 containing 0.0 (deionized water, DIW) to 0.15 M NaCl at 40 to 275 C. Dissolution rates were determined in low temperature (40 to 80 C) and hydrothermal (120 to 275 C) reactor systems, using flow-through reactors that are broadly similar in design. Rate data collected from these two reactor designs are consistent with each other and yield the first comprehensive model of amorphous silica reactivity in deionized water and electrolyte solutions (Icenhower and Dove, 1998). Measurements of rates show important similarities and differences between the corrosion behavior of SiO2 (am) and a-quartz. They find that the experimental energy of activation, E a,xp , for the dissolution of SiO2 (am) is 75 261 5 kJ mol -1 in DIW. The introduction of up to 0.05 M NaCl yields similar E a,xp values of 80 261 5 kJ mol -1 . These values are 30510 kJ mol -1 higher than previous estimates of E a,xp for SiO2 (am) but are consistent with reported values of E a,xp for a-quartz. Dissolution rates measured at 200 C in DIW show that SiO2 (am) dissolves 3053 to 30X faster than a-quartz. A possible explanation for this difference is that SiO 2 (am) has a fraction of Si-O-Si bonds (angles up to 180260) that have a greater ionic character, and are therefore more reactive than a-quartz constituents (mean angle of 152260 ) (Icenhower and Dove, in prep.). Measurements of SiO2 (am) dissolution rates versus NaCl concentrations at 200 C show that sodium enhances rates by a factor of 30510 to 30X compared to rates measured in DIW, which are less than rates for a-quartz under identical experimental conditions. In addition, they find that the dissolution rates of the two forms of SiO2 (am) (fused and flame pyrolysis silica) are similar within the experimental error of the early experiments. Results of this study suggest that the role of physical (structural) properties (e.g., Si-O-Si bonds) in governing reactivities of crystalline versus amorphous SiO2 polymorphs is significant.'