Towards a New Theory for Improved Oil Recovery from Sandstone Reservoirs

Abstract As discoveries of large new oilfields become a thing of the past, the residual oil in old oilfields becomes an increasingly important oil resource. Significant research and development efforts on increasing recovery have however yielded low results. We believe that the main reason for this is that the current models for how residual oil is trapped and mobilized are wrong. Several field observations have shown that mobilized residual oil moves more rapidly through the reservoir than the injected water assumed to be pushing it. This is not possible within the framework of current reservoir simulation models. We here present a new theory for the behaviour of residual oil in sandstone reservoirs. Based on field cases and a laboratory study we argue that the residual oil is collected in continuous oil strands extending from injector to producer and blocked only by water in a pore close to the producer. The blocking water of that pore throat can be removed by applying a pressure pulse to the oil strand or by reducing the capillary force that makes the water adhere to the wall of the pore throat. In a flooding experiment where the flow rate of oil and water was followed with oil and water soluble tracers, the results showed that oil moved 1000 times faster than the injection water through the core. Field observations involved oil fields, where analysis of H2S concentration at the producers showed that H2S had been transported with the oil and moved faster than the water through the reservoir. In an MEOR pilot increased oil production ceased almost immediately after the microbial activity was stopped, showing a continuous oil phase from injector to producer even under residual oil concentration. Applying this new knowledge can considerably increase the recovery from most oil fields.