A Geochemical model for iron oxidation, mine water acidity and CO2 degassing

The Problem

Venturi system used to treat AMD outflow

In 2006, Pennsylvania Department of Environmental Protection (PADEP) and the Federal Office of Surface Mining (OSM) developed a strategy to treat ~600 gpm of iron-laden mine drainage discharging from a recently closed underground coal mine. Prior to bankruptcy, the mine operator used sodium hydroxide to treat to NPDES effluent standards by raising the pH of the net-alkaline mine drainage to remove dissolved ferrous iron as Fe(OH)₂. This treatment scheme resulted in chemical and desludging costs of upwards of $15,000/month. PADEP and OSM’s treatment strategy focused in installing a venturi system to treat AMD outflow driven by artesian pressure from the steeply-dipping underground coal mine. The primary goals were to eliminate chemical usage by using the venturi to accelerate iron oxidation and to reduce desludging costs by precipitating the iron as the denser Fe(OH)₃ than the voluminous Fe(OH)₂.

Installation of the venturi system yielded significant cost savings, but it proved difficult to evaluate the relative effectiveness of the venturi system and associated settling pond. Primarily, OSM was interested in the relative contributions of heterogeneous iron oxidation, homogeneous oxidation and CO₂ exsolution to decreases in dissolved iron contents and increase fluid pH.

The Solution

Decrease in dissolved iron due to heterogeneous/homogeneous oxidation and CO2 dissolution

OSM asked RockWare to build a kinetic reaction path model to evaluate the relative effectiveness of the venturi system and settling pond in mediating dissolved iron contents and acidity. The model was built using The Geochemist's Workbench software, and incorporated the relative kinetic effects of heterogeneous and homogeneous oxidation, as well as CO₂ gas exsolution. Rate laws and constants were taken from Dempsey et al. (2001).ⁱ Temperature dependence was also incorporated. The final model presented to OSM provided a realistic representation of processes active at the site and helped drive further treatment planning. The model has also been successfully modified for use with a number of other related passive treatment systems.