Evidence that High Polymer Viscosity Accelerates and Increases Oil Response in Grimbeek Manantiales Behr

Increasing oil recovery because of induced flow from low permeability to high permeability driven by the ratio polymer viscosity/oil viscosity has been around for more than 30 years. This phenomenon occurs when there is a surface of contact and you inject polymer which creates a higher pressure drop across the polymer slug.

Does an increase in the ratio polymer viscosity/oil viscosity always increase recovery and accelerate oil response? How is this ratio affected by the combination of geological heterogeneity oil viscosity? How do multiple surfaces in fluvial systems enable the crossflow or create bypassed oil during water injection? Is there any universality that captures this phenomenon across the combination of geological heterogeneity and fluids viscosity?

Initial simulations indicated that incremental oil will start to ramp up after 6 months of stable polymer injection at the target viscosity. Two groups of polymer injection units started polymer injection between the August-2019 and September-2019. During that time, we faced problems in water supply, therefore we had to reduce the water injection rate from Qi=100 m3/d to Qj=70 m3/d. The target polymer concentration was Ci=2500 ppm which means a 0.5 polymer/oil viscosity ratio at reservoir conditions. Conceptual pore-scale simulations give the insight that increasing the polymer viscosity could make a higher pressure in the polymer zone that could induce additional crossflow. We tested this hypothesis in the simulator by compensating the rate reduction by increasing the polymer dosage. The simulations show that increasing concentration can compensate for the reduction in incremental oil production because of the injection rate reduction. Thus, we injected 70 m3/d @ 3775 ppm-polymer concentration. The increase in concentration raised also to 1.15 the polymer/oil viscosity ratio at reservoir conditions. Unpredictably, the actual oil rate was 20% higher and the response faster than simulations.

Multiscale crossflow is one of the leading recovery mechanisms in polymer injection (Sorbie, 2019). We risk saying that standard modelling lacks the level of heterogeneity needed to estimate better the remaining oil in the subsurface and we normally underpredict the polymer flooding potential especially in heavily water flooded reservoir -more than 30-year waterflooding.

Therefore, we have undertaken enormous efforts to construct very detailed models that aim to capture the many possible surfaces of contact between different facies.

The good recovery values, thus, may indicate the efficacy of the proposed higher concentration dosage mechanism for inducing additional crossflow.