The Rheology of Viscoelastic Fluids for Oil Recovery: A Study of Micellar/Polymer Flow Over a Wide Range of Shear Rates in Porous Media

The rheological behavior of selected polymer solutions used for mobility control in waterflooding, micellar fluids used in low-tension chemical flooding, and micellar/polymer mixtures has been examined. Poly-mers used were Kelco's Xanflood (a xanthan gum) and Tiorro's Watercut 110 and 160. The micellar solutions contained Witco's TRS 10-80 or TRS 10-410 (both petroleum sulfonates), isobutyl alcohol, sodium chloride, and one of several hydrocarbons (iso-octane, n-decane, or Nectone 37 oil). A variety of porous media were utilized, ranging from packs made with glass beads to sandpacks and consolidated porous media (Berea rock) . Use was made of the modified Blake-Kozeny model of the porous medium to correlate the steady flow viscometric behavior of the fluids with that of the apparent viscosity as the fluid is displaced through a porous medium. Measurements were made to extend the shear rate range of investigation to values beyond those used in earlier experiments. Static measurements were made of phase volumes and viscosities of various mixtures of surfactant, water, alcohol, sodium chloride, polymer, and oil to obtain a more complete characteriza-tion of the phase behavior and viscosities of the fluids used in the flow experiments. Three different rheological measurements were made for each fluid. A couette viscometer was used for steady state measurements at low shear rates, a capillary viscometer measured the steady flow viscosity at medium to high rates of shear, and the-apparent viscosity was calculated from pressure drop-flow rate data as the fluid was flowing through the porous medium. For the micellar solutions studied, use of the modified Blake-Kozeny relation proved to be adequate to characterize their rheological behavior in porous media. Micellar fluids containing Witco's TRS 10-410 or TRS 10-80 petroleumsulfonates,isobutyl alcohol, and brine were mixed with various hydrocarbons. The apparent viscosity for these fluids as they flowed in glass bead packs was correlated with steady viscosity and good agreement between the values was achieved. A single experiment was done with a micellar fluid in Berea rock and a good correlation between apparent viscosity and the steady flow values was also realized, despite the existence of a residual water phase. The maximum viscosity at low shear rates and the onset of pseudoplastic behavior for these fluids was a strong function of oil type, oil concentration, and salinity. No shear degradation was observed for any of these fluids up to several thousand -1 sec The apparent viscosities of two Xanflood polymer solutions in Berea rock did not show very good agreement with the measured steady flow viscosity values. The apparent viscosity values were lower than the measured bulk viscosities over the middle shear rate range. At very high shear rates, the apparent viscosity of the fluids became slightly greater than the viscosity as measured in the capillary viscometer. A dispersion experiment done for these same Xanflood polymer solutions in Berea rock has indicated some polymer transport problems characterized by a polymer breakthrough lag behind that of tritiated water. v An experiment to characterize the rheological properties of a micellar/polymer mixture in a glass bead pack showed that the modified Blake-Kozeny relation was adequate under the conditions used. A polymer cross-linking experiment utilizing the cationic Watercut 160 and the anionic Watercut 110 was conducted with the cation Al3+. Alternating slugs of the anionic Watercut 110 and the aluminum ion solution resulted in no significant changes in the resistance factor in a high permeability sandpack. Alternating slugs of the cationic Watercut 160 and the anionic Watercut 110 produced a slight increase in resistance factor; however, further injections of Al3+ and Watercut 110 resulted in no change in resistance factor.