Soltani, Amir (Beicip-Franlab) | Decroux, Benoit (Beicip-Franlab) | Negre, Andres (Beicip-Franlab) | Le Maux, Thierry (Beicip-Franlab) | Djarir, Maâmar (Sonatrach) | Selmi, Farouk (Sonatrach) | Lantoine, Martin (Beicip-Franlab)
Abstract EOR surfactants are usually formulated at the initial reservoir temperature. Is this a correct approach? Field data from three Single-Well Chemical Tracer pilots in North Africa are used to answer this question. The objectives are, first, to provide a realistic image of the temperature variations inside the water-flooded reservoir; second, to demonstrate the impact of such temperature variations on the surfactant performances; and last, to introduce a new methodology for estimating the target temperature window for surfactant formulations. During pre-SWCTT pilot tests, water injection, shut-in and back-production were performed. The bottom-hole temperature was monitored to evaluate the reservoir temperature changes (initially at 120°C) and to calibrate a thermal model. The thermal parameters were applied to the reservoir model to simulate 30 years of water injection (with its surface temperature varying between 20°C and 60°C) and to obtain a full picture of the temperature variations inside the reservoir. Multi-well surfactant injection was modelled assuming that the surfactant is only efficient within ±10°C around the design temperature. The impact of this assumption on the additional oil recovery was analyzed for several scenarios. The rock thermal transmissivity was found to be the key parameter for properly reproducing the observed data gathered in the North African pre-SWCTT tests. The measured temperature during the back-production phase demonstrated the accuracy of the thermal model parametrization. It proved that the heat exchange between the reservoir and the injected fluid is considerably less than what industry expects: the injected water temperature inside the reservoir remains far below the initial reservoir temperature even after 11 days of shut-in. When simulating various historical bottom-hole injection temperatures and pre-flush durations, the thermal model showed an average cooling radius of 275m, larger than the industry recommended well-spacing for the EOR 5-spot patterns. This was mainly due to the significant temperature difference between the historical injected water and the initial reservoir temperature. Several simulations were performed for 3 representative bottom-hole injection temperatures of 20°C, 40°C and 60°C, varying the surfactant design temperature range between the injection temperature and the initial reservoir temperature. The results showed that regardless of the injection temperature, the simulated additional oil recovery is highest when the design temperature range is close to the injection bottom-hole temperature. This is an important subject since in the EOR industry, the surfactants are usually formulated at the initial reservoir temperature and thus, the impact of the reservoir cooling on the surfactant efficiency is seldom considered. In a water flooded reservoir, the injected chemicals are unlikely to encounter the initial reservoir temperature. This results in a dramatic loss of surfactant performance especially when there is a considerable difference between the initial reservoir and the injected fluid temperatures.