CO2-induced release of copper and zinc from model soil in water

The eventual leakage of CO2 from storage sites into shallow aquifers may degrade the quality of potable groundwater, due to acidification caused by the intrusion of gas CO2. Experiments in continuous-stirred tank reactors (CSTRs) seems to be a fast and cost-effective way to quantify the kinetics of heavy metal release from soil material to groundwater. The goal of the present work is to evaluate the capability of a model to provide reliable values for the kinetic parameters of metal release from CSTR tests with reference to results from soil column and batch tests. Well-sorted silica sand is used as model soil with its composition being modified with the pH-controlled precipitation of Cu or Zn. Gas CO2 is dissolved in distilled water, and the carbonated water is fed at constant flow rate in a CSTR containing soil grains suspended in water. The concentration of dissolved metals in effluents is measured with atomic absorption spectroscopy (AAS). A dynamic mathematical model of the operation of CSTRs is developed by regarding the metal release as a pH-controlled two-site adsorption/desorption process. A sensitivity analysis is done, and the kinetic parameters of the sorption model are estimated, separately for each metal, with inverse modeling of the transient response of metal concentration in CSTR. In addition, the model validity is assessed by comparing the transient response of pH and redox potential with numerical predictions. In general, a significant uncertainty is embedded in estimated parameters of the 2-site sorption model. Compared to the corresponding kinetic parameter values estimated from soil column and batch tests, any differences should be interpreted in the light of the different experimental conditions and assumptions.