CO₂-Brine-Rock interactions in reservoir chalk rock - a coupled experimental and numerical approach for obtaining hydrochemical results at reservoir conditions

Sampling and analysis of water chemistry in reservoir condition experiments with high CO₂ pressures is challenging since CO₂ can easily degas from the samples and cause changes in the chemical composition before chemical analyses are performed. In the literature, water samples during dynamic experiments for chemical analyses are often obtained behind a back pressure (BP) regulator located at the downstream end of the flooding specimen. As a consequence, chemical samples are obtained at a lower pressure (often atmospheric pressure) via the BP regulator. Another approach is to sample directly from the device where reservoir conditions are applied (e.g. core holder or batch) and subsequently carry out chemical analyses or preserve samples “as fast as possible” after sampling. In both approaches, there is essentially no control on the CO₂ pressure and therefore CO2 degassing during the sampling procedure may cause both precipitation of carbonate and an increase in pH. As a result, the measured chemical composition may differ substantially from the true chemical composition at reservoir conditions. 
We present a novel methodology that, on the basis of obtaining a consistent set of hydrochemical data at one set of well-known physical conditions (CO₂ pressure and temperature), enables the estimation of true reservoir conditions by the application of numerical calculations with the hydrogeochemical code PHREEQC. The technique is demonstrated for an experimental study where reservoir chalk was flooded with either CO₂ saturated water or slugs of CO2 and seawater in a WAG (water alternating gas) scheme at reservoir conditions. The methodology is validated by various independent check calculations such as comparison of forward modeling results with measured results and carbonate mass balance calculations. 
At present, the methodology is considered applicable for flooding of chalk material where the dominant chemical reactions taking place are dissolution/precipitation of calcite and gas/water equilibrium with a CO₂ gas phase. In the case of a water-alternating-with-gas (WAG) flood, the numerical calculations may be refined by application of a true 2-phase numerical code such as TOUGHREACT in a 2D or 3D model.