A Diffusion Controlled Model for Prediction of Long-Term Cement Sheath Carbonation

Abstract

The recent increase in production from reservoirs containing in-situ CO 2 and H 2 S; and injection of produced CO 2 for geological carbon dioxide storage and enhanced oil recovery (EOR) has prompted strict risk assessment, well construction best practices, and long-term monitoring. Isolation materials used in these wells, particularly casing and cementing slurries, are to be resistant to CO 2 attack. A long-term experimental approach might be too slow to meet industry needs; therefore, model-based simulation is essential to guide a design that helps ensure zonal isolation for the life of the well.This work presents a diffusion-controlled model for simulating the conversion of cement hydration products to calcium carbonate and subsequently to water-soluble calcium bicarbonate, predicting the depth of penetration of the two reaction fronts. Temperature, total pressure, and partial pressure of CO 2 are considered, corresponding to downhole conditions. After calibration with short-term experiments, the model is able to extrapolate the degradation to long term.A simulation tool based on this model was developed and validated using available literature results and experiments performed with this purpose. Simulation results show that progress of the carbonation reaction front is faster than the subsequent conversion to calcium bicarbonate. The extent of degradation is quantified by depth of penetration of the two reactions. Strong dependency on temperature, pressure, and partial pressure of CO 2 was observed. The formation permeability also plays an important role, acting as an additional mass transfer resistance.The results demonstrate that the need for long-term experiments can be reduced or eliminated using this model-based simulation tool. The tool does not only evaluate and compare current slurry formulations, but also guides the design of new CO 2 resistant materials. The extent of cement degradation attributed to CO 2 injection can be predicted in downhole conditions, providing one of the most important parameters for assessing the risk of CO 2 leakage in the long term, for long-term safety of wells.