A 2-D Visualisation Study of the Mechanisms behind Increased Recovery due to Polymer Injection in Sandstone
MetadataShow full item record
The potential of synthetic polymer injection, such as hydrolysed polyacrylamide (HPAM), in reservoirs has been a field of study since the 1960's . The main intension of polymer injection is to improve the problems with water injection due to high mobility ratio between oil and water and reservoir heterogeneity. Polymers are added to injection water in order to increase the viscosity and reduce the mobility ratio. Though polymers have been thought to only improve the volumetric sweep and not the microscopic sweep, recent research argue the opposite due to viscoelastic effects and high residual oil saturation after water injection. The application of water and polymer injections in viscous to heavy oils were also regarded as non beneficial, but new research proposes a favourable recovery of viscous oils. However, because the mechanisms behind polymer injection in viscous oil reservoirs are poorly understood in virtue of high fluid complexity and the non Newtonian behaviour, more research is required. In this thesis, a two-dimensional core scanner is used to perform a combination of in-situ high resolution X-ray imaging of fluid movements and gamma-ray counts of fluid saturation. Together with rheological effluent analysis, this is used in order to identify mechanisms behind increased recovery of viscous crude oil by the use of HPAM. Two identical Bentheimer sandstone rock samples (named WF+PF and PF) are used to observe the difference between a primary and secondary polymer injection. To investigate the effect polymer may have on the displacement of oil, an initial condition with dampened capillary pressure is preferred. Ageing the rock material with crude oil at elevated temperature is assumed to alter wettability and thereby dampen capillarity. Earlier experiments performed with the two- dimensional core scanner at the Centre for Integrated Petroleum Research (CIPR), have proved the scanner suitable for in-situ high resolution X-ray imaging in order to observe finger behaviour at both miscible and immiscible displacements [2, 3]. For this thesis, effluent dispersion tests revealed homogeneous samples, where flow was little affected by dead-end pore volume, but more affected by the configuration of the inlet and outlet. The ageing process at 70°C for six weeks, gave a moderate change in wettability from strongly water-wet. This was supported by the relative oil permeability for both samples and the observed fingering and capillary end-effects in the WF+PF sample. No distinct oil bank was observed in the secondary polymer injection in the WF+PF sample. However, the water-cut was kept constant, and it is impossible to exclude the effect of the polymer. The primary polymer injection in the PF sample showed a more piston-like displacement compared to the water injection in the WF+PF sample. The polymer front arrived 0.4 pore volumes earlier than the dispersion front in the WF+PF sample, something which indicates inaccessible pore volume and depleted layer effects. The final recovery in both samples after polymer injection was similar, but it was achieved in different time scales. In the PF sample it only took approximately one pore volume (PV) of polymer solution to obtain the same recovery as it took for the WF+PF sample, with approximately 1 PV of...