Impact of Salt Diffusion on Low-Salinity Enhanced Oil Recovery

Abstract

Abstract Low-salinity water flooding is an emerging EOR technique that has yielded significant incremental oil in core floods and single- and inter-well trials. Because low-salinity water flooding involves reducing salinity to a low level, of the order of 1000 ppm, there is a risk that its efficiency could be reduced by mixing between injected low-salinity water and high-salinity in situ water held in shales. This risk is more significant when the in situ salinity is very high and the amount of water held in shale is large, as is the case in some reservoirs found in deep-water depositional environments. To investigate the effect of salt diffusion on low-salinity EOR, low- and normal-salinity (sea water) water floods were simulated in fine-scale, 2D models of deep-water reservoirs incorporating dispersive and diffusive mixing of injected and in situ water and the wettability change caused by low salinity. In the model, dispersion predominates in sandstone, whereas diffusion controls the transport of salt ions within shale. The dispersivity and diffusion coefficients are based on published data. The simulation results show that low-salinity water-flooding of thin (<2.4m thick) sand layers sandwiched between thick shales is vulnerable to salt diffusion at practical flood rates when the in situ salinity is high (e.g. ≥100,000 ppm). In such layers, diffusion can reduce low-salinity EOR by more than 90%. In the reservoir models we studied, diffusion can reduce the overall low-salinity EOR to 60 – 95% of the value calculated when diffusion is neglected. The adverse effect of diffusion increases as the average sand layer thickness decreases. In thick, channel sands, salt diffusion has only a small effect. Apart from reservoir description, the most important controls on the effect of salt diffusion are intra-reservoir shale diffusivity and in situ salinity. The simulation results were used to calibrate a simple model for rapidly screening stratified reservoirs. The screening model calculates a ‘diffusion recovery factor (RF) multiplier’ which can be used to discount conventional low-salinity EOR predictions that neglect salt diffusion. The screening model matches simulation data to within 14% RMSE.