Spin–Spin Relaxation Time Investigation of Oil/Brine/Sand Systems. Kinetics, Effects of Salinity, and Implications for Wettability and Bitumen Recovery

Abstract

Improving the efficiency of secondary and tertiary oil recovery is one of the major challenges of the crude oil industry. Arguably the most significant advance made in recent years has been the realization of the importance of water chemistry during oil recovery. This has led to the concept of low-salinity waterflooding for conventional oil reservoirs, for which reducing the total salinity of the injection water has been found to improve oil recovery rates. Although the precise mechanisms responsible for the improvements are not completely understood, it is acknowledged that specific interactions in the crude oil/brine/rock (COBR) systems will modify wettability and interfacial energy. The present paper extends investigations of COBR interactions to higher viscosity oils and considers implications for heavy oil recovery. We have used NMR relaxation time measurements to study interactions in oil/brine/sand (OBS) systems containing either natural bitumen or a polybutene hydrocarbon (Glissopal). Exposing the oil-coated sands to water or aqueous group 1 and group 2 metal chloride solutions enabled the T2 relaxation time spectrum to be determined as a function of time. The observed decreases in the geometric mean T2 values with time obey first-order kinetics; for bitumen-coated sands, the rate constants are consistent with diffusion of water through the bitumen. Opto-digital microscopy verified the formation of ∼1–3 μm diameter water droplets in the initially dry bitumen coating, suggesting that water nucleation and growth also occur. This was not observed for the Glissopal-coated sand samples. No evidence was found for displacement of either viscous oil from the sand grains, although optical microscopy did reveal rearrangement of the bitumen coating, which possibly exposes fresh sand surfaces to the aqueous phase. This behavior is consistent with the finding that the original T2 parameters determined for fresh sand are not fully restored simply by contacting the bitumen-sand surface to water or aqueous salt solutions under the ambient experimental conditions. Glissopal-coated sands exhibited smaller time-dependent T2 changes compared with bitumen-coated sands. While not displacing viscous oils from the sand surface under the experimental conditions used, it is conjectured that water ingress into the surface oil layer could weaken oil/sand interactions which by analogy with recent studies on conventional oil recovery could provide an additional heavy oil recovery mechanism under more dynamic or higher temperature conditions.