PREDICTING ALTERATION OF WELL BORE CEMENT USING NUMERICAL MODELS

The integrity of well-bore cements after reaction with CO₂ is an area of concern in CO2 sequestration. Portland cement will react readily with injected CO2. Predicting these reactions and the resulting changes to well bore integrity is an important part of evaluating the overall fate and transport of CO₂ in a storage reservoir. Numerical models are a powerful tool in the prediction of reaction and transport in porous media and will play a large role in developing confidence in CO₂ storage and sequestration. To use reactive transport models to predict interactions between cement and CO₂ rich fluids it is necessary to first compare results from numerical simulations to field and experimental data to gain confidence in these models.

In a study by Wigand et al. (Chem. Geol., 2009) class G Portland well cement was reacted with brine and supercritical CO₂ at reservoir pressure (pore pressure 19.9 MPa) and temperature (54^oC). Chemical alteration of the cement was observed ~ 5 mm into the cement at after 99 days of exposure to supercritical CO₂. This experiment was simulated using the reactive transport code PFLOTRAN. Complete dissolution of the amorphous Ca-Si-H phase in the cement is predicted along with precipitation of calcite (+/- aragonite depending on relative reaction rates) and amorphous Si. These mineralogical changes are consistent with those observed in the laboratory experiment. A decrease in porosity associated with cement alteration is also accurately predicted with the numerical model. The use of numerical reactive transport models to predict cement alteration is promising given the success of the simulations presented here and those reported in Carey et al. (Int. J. Greenhouse Gas Control, 2008).