Results of CO2-flooding experiments Exp-1, Exp-2, Exp-3, and Exp-4. Greensand Project Phase 1 WP2

In Greensand Project Phase 1 WP2 four CO2-flooding experiments, Exp-1, Exp-2, Exp-3 and Exp-4 were conducted on core samples from the Nini-4 well at reservoir conditions (60 oC, 200 bara). The experiments consisted of a complex sequence of CO2 and formation brine injection operations aimed at evaluating the injectivity and investigating the possible reactions between supercritical CO2, formation brine, and the greensand reservoir rock. Each experiment was conducted on a composite core sample with length approximately 15 cm, consisting of three 1.5” standard core samples.

The brine permeability measurements agree with the gas permeability determinations of the core samples from the conventional core analysis programme of Robertson Research International Ltd. (RRI). Brine permeability measurements made after CO2-flooding operations yield the same permeability as before the CO2-flooding operations, indicating that no significant damage is caused by the CO2-flooding. This agrees with macroscopic visual investigations of the core samples after the experiments that do not reveal significant changes to the greensand rock.

A total of 14 CO2 injection operations with permeability logging were conducted in the 4 flooding experiments. 11 operations used a rate of 800 ml/h, and 3 operations used a rate of 1740 ml/h. The CO2 injection operations make a strong contrast with the brine injection operations by showing a complex permeability behaviour. The 10 injection operations that started with a fully brine-saturated state (Sw=100%) shows an initial steep rise in permeability to CO2 from a low permeability level, which is as expected from relative permeability conditions. The permeability relationships to CO2 deviates in two ways from the permeability relationships to brine. First, the permeability to CO2 in the four experiments never recovers to more than between 22% and 38% of the permeability to brine. Second, in five of the 14 CO2 injection operations a total of seven instances of abrupt, spontaneous drops in permeability were recorded, with the drops in permeability ranging from 7% to 75%. The drops in permeability were not correlated with any rig operations.

After Exp-4 the three core samples of this experiment were analysed for water saturation by Dean Stark extraction. From core holder inlet towards core holder outlet, the three samples showed water saturation results of 28%, 40%, and 65%. This shows the presence of a strong water saturation gradient within the core sample from inlet towards the outlet.

The results of the CO2 injection operations are interpreted as being caused by channelling of CO2 in a core sample with high water saturation. The viscosity ratio µCO2/µw = 0.1 caused the CO2 to concentrate in channels, possibly more pronounced towards the outlet end of the core sample. At irregular time intervals the CO2 flow within the core spontaneously changed to another position. The fluid density log shows production of water from the core sample through most of the CO2 injection periods. Probably the shift in channel position was caused by changes in water distribution within the sample. Six of the seven instances of spontaneous permeability drop occurred before injecting more than 6 PV’s of CO2. Therefore, the probability of spontaneous drop in permeability is clearly greatest at the start of an injection operation when the water saturation is high.