Solubility and stability of scorodite, FeAsOn'2HrO: Reply

The discussions by Nordstrom and Parks (1987) and Robins (1987) enumerate some of the pitfalls (real and prospective) that might be encountered in experiments to measure the solubility of the mineral scorodite. In this reply, we present our opinions of the relative dangers of these various pitfalls and the values of the avoidance strategies proposed by Nordstrom and Parks (1987). In addition, we present an explanation for the failure offerric hydroxide to nucleate in our experiments, and after a detailed analysis, we conclude that the solubility product of scorodite determined by Dove and Rimstidt (1985) and Robins (1987) is the best value currently available and is sufficiently accurate for most geochemical applicatrons. Nordstrom and Parks begin with a list of criteria that they believe are required for a "successful" solubility measurement. We shall comment on these in the order in which they were presented. 1. Although demonstrating the reversibility of a reaction is desirable in some cases, it is often not practical and certainly not required in order to obtain reasonable and useful thermodynamic values. For example, no measurements of amorphous silica solubility have been reversed, yet the published values are widely accepted and utilized. In addition, almost none of the reported quartz solubility values come from reversed reactions; in fact, the data from the only comprehensive attempt to measure quartz solubilities by reversing the reaction (Crerar and Anderson, l97l) were discounted in favor of unreversed measurements in the reviews of quartz solubility by Fournier (1979) and Fournier and Potter (1982). 2. We determined the stoichiometry of the reaction that we studied by measuring the pH along with the Fe3* and the arsenate concentration in the solution. The equal molal concentrations of Fe and arsenate in solution demonstrate that the scorodite sample dissolved congruently, and the lack of a shift to a lower pH shows that there was no significant ferric hydroxide precipitation (this point will be discussed in detail in a later section). The technique ofvarying the concentration of one or more aqueous species described by Nordstrom and Parks is commonly used to determine the stoichiometry of aqueous complexes (see for example, Barnes, 1981). However, if the stability and stoichiometry of the aqueous species are known, as in this case, this approach is redundant.