The application of MRI to enhanced oil recovery studies

• Main Author: Williams, John Lewis Anthony
• Published: University of Surrey 1992
• Subjects: 530.41; Nuclear magnetic resonance
• Online Access: https://ethos.bl.uk/OrderDetails.do?uin=uk.bl.ethos.305009

At the start of this work in 1988, apart from bulk relaxation experiments four authors had reported discrimination of oil from water in core samples under static conditions by MRI, and only one, Baldwin (1986) had reported visualisation of flowing oil and water using doping agents Mn++. The aim of this research was to develop the application of MRI to petroleum engineering and in particular to develop Enhanced Oil Recovery (EOR) techniques. A synergistic approach was adopted involving collaboration with AGIP Oil, Italy and the Petroleum Engineering Department of Imperial College. A systematic study of properties of the fluids and their interaction with rock types was mounted, to investigate suitable contrast mechanisms. This was followed by static and dynamic imaging experiments. The chemical shift imaging technique has been used to measure oil and water saturations during flooding experiments with Portland Limestone. For the first time MRI has been used to generate relative permeability, and fractional flow curves for a core plug. The capillary pressure gradient was also determined from the data. The new method was compared to the results of traditional core analysis. Miscible displacements in Lochaline Sandstone were then studied using D2O, glycerol and polymer solutions. The spin echo imaging technique was used to determine the angle of the interface between the fluids. This angle is determined by the ratio of viscous to capillary forces. From this information the core permeability (liquid) and polymer viscosity were measured. These measurements compared well with conventional methods and the results break new ground for MRI. The polymer viscosity measurements are of particular interest since the fluid is non-Newtonian. In separate experiments tertiary chemicals were identified directly during displacement experiments by MRI for the first time. Finally a new model system was developed for image calibration.