Scaled.-Model Tests of In-Situ Combustion in Massive Unconsolidated Sands

Abstract

Criteria were developed for operating scaled models to evaluate the effect of process and reservoir variables on in-situ combustion. These scaling criteria were tested in 2 cylindrical models, one having a volume 325 times greater than the other. The results demonstrate the validity of the scaled-model approach, especially for reservoirs in which heat generation is controlled by mass transfer of oxygen. Process and reservoir variables were studied in 7-spot pattern models initially saturated with connate water and 2,100-cp oil. At practical injection rates, recoveries from 60-70% pore volume were achieved at reasonable air-oil ratios from a relatively thick, shale- free sand. Either reduced sand thickness or interbedded shale in the models increased requirements for injected air or oxygen to get equivalent recoveries. To test the usefulness of redistributing heat in the reservoir, water and oxygen were injected. At a water-to-oxygen mass ratio of about 5, oil recovery was increased and equivalent air- oil ratios were sharply reduced.