Modelling CaCO3 Scale in CO2 Water Alternating Gas CO2-WAG Processes

摘要

CO2 Water Alternating Gas (CO2-WAG) is one of the main Enhanced Oil Recovery (EOR) techniques which is currently being implemented in the emerging pre-salt projects offshore Brazil. CO2-WAG consists of the alternated injection of CO2 and water in the reservoir for tertiary oil recovery. However, this process may lead to the enhanced deposition of CaCO3 in production wells. This may occur since, in a CO2-WAG scheme, CO2 dissolves in the water slug causing a decrease in the pH. At lower pH levels, carbonate rock is dissolved causing an increase in the carbon and calcium content in the water slug. As fluids are later produced, the operating pressure is reduced and dissolved CO2 is evolved from solution, causing an increase in the pH. At these less acidic pH levels, CaCO3 may become oversaturated and precipitate. In order to address each process involved in CaCO3 formation, an integrated modelling approach between aqueous scale prediction modelling, Vapour-Liquid Equilibria (VLE) modelling and reservoir modelling is proposed. In particular, acid equilibria and precipitation reactions coupled with the Pitzer equations are used to build the scale prediction model. The facility to calculate CaCO3 co-precipitation with other minerals, such as BaSO4, FeCO3, FeS, etc., is also introduced. Five different equations of state (namely SRK, PR, PRSV, PT and VPT) are used in VLE calculations to model the solubility of CO2 (or a mixture of CO2, H2S and CH4) in brine. To model the reactive flows in the reservoir, the advection-diffusion equation is coupled with the scale precipitation equations. The reactive transport model addresses carbonate rock-brine interactions and fluid flow through a porous medium. Once integrated in one single model, these equations have been shown to address all steps in CaCO3 formation relevant for CO2-WAG, i.e., the geochemical processes taking place in the reservoir and at the production wells, thus capturing the full dimension of the scaling problem. In addition, the proposed model has been successfully validated with experimental data on the injection of seawater adjusted to various pH levels into a CaCO3 packed column. The measured effluent pH levels and calcium concentrations were used to validate the model.