Polymer flooding is one of the most mature enhanced oil recovery (EOR) methods with many field implementations including those in China, Germany, Oman, and USA. The primary role of polymer is increasing the injected water viscosity, hence reducing the displacing front mobility and thereby improving the macroscopic sweep efficiency. Polyacrylamide, the main polymer used in EOR applications, achieve this increase in viscosity due to the large molecular size of its chains as well as the ionic repulsion between the chains. Those same properties result in significant interactions between the transported polymer molecules and the porous medium, including adsorption, mechanical entrapment, and hydrodynamic retention. Those phenomena, in turn, can lead to polymer losses, injectivity reductions and inaccessible pore volumes. Despite the maturity of polymer flooding, few implementations and research studies have targeted carbonates. Thus, a clear understanding of the magnitude and significance of those interactions and effects for carbonates is lacking. Those phenomena are critical for both numerical predictions and actual performances of polymer flood.
Therefore, in this work we investigate thoroughly polymer losses, injectivity reductions and inaccessible pore volumes for a slightly viscous Arabian carbonate reservoir that exhibits high salinity and high temperature conditions. For this purpose, we perform single phase displacement experiments at reservoir conditions. Representative materials were used including simulated brines reflecting connate and injection brine salinities, dead crude oil, and aged reservoir plugs. Core plugs with a wide permeability range from 45.2 md to 12836 md were used for the tests. A pre-screened polyacrylamide was used at an injection concentration of 5,500 ppm. A 2,000 ppm tracer was added into the polymer solution to assess polymer interactions. The effluent polymer concentrations were determined by total organic carbon (TOC) method, and tracer concentrations were analyzed by gas chromatography (GC).
Results showed that resistance factor (RF) tended to be higher for tighter samples. RF increased with increasing injection rate for lower permeability samples and decreased with increasing injection rate for higher permeability samples. Residual resistance factor (RRF) slightly decreased with increasing injection rate. RRF correlated well with pore size, with larger pore size corresponding to lower RRF. The effective in-situ viscosity of the polymer was constant at lower injection rates. However, at higher rates, the effective in-situ viscosity increased with injection, exhibiting a shear thickening behavior. Moreover, the polymer exhibited dynamic retention ranging from 0.155 to 0.530 mg/g-rock, and showed a decreasing trend for more permeable core sample. Finally, the studied carbonate constituted of 15.2% to 20.9% pore-volume that was inaccessible to the polymer. Those results besides being essential for numerical-based upscaling of polymer flooding, shed light on some of the similarities and differences between sandstones and carbonates when it comes to chemical EOR application.