Laboratory and Numerical Study of Scaling Parameters Used in Modelling of CO2 Storage in Rocks

Abstract

Storage of CO2 in deep saline formations is currently the most promising option for mitigating the impact of climatic changes. The main concern related to CO2 storage in geological formation is safety. It is necessary that injected CO2 is under control at all times and its behaviour is predictable. Therefore, it is important to understand flow processes and distribution of forces acting underground on CO2 during and after injection. It is also crucial to know what happens with CO2 after tens, hundreds and even thousands of years after it has been injected. The only way to predict movement of CO2 in such a long time span is by modelling processes that take place underground. This work presents findings of both experimental and numerical modelling of flow processes taking place during and after CO2 injection.

Special experiments were designed in order to demonstrate the influence of gravitational, viscous, and capillary forces on the flow of CO2. In laboratory experiments fluid representing CO2 was injected into a 2D porous medium saturated with fluid representing brine. Two sets of fluids characterized by different interfacial tension (IFT) were tested. Results of the experiments demonstrate that at increasing injection rate viscous forces become stronger. This leads to a higher total displacement of brine. Such performance facilitates dissolution and residual trapping of CO2 and is desired at the field scale. However, such conditions can lead to pressure increase in the near-well zone due to injection and this can possibly damage the geological formation by fracturing, which in turn, can compromise safety of the storage site.

At low injection rates and high permeability, gravity effects increase their influence. This is demonstrated by lower volumes of the in-situ fluid displacement what is not favourable during CO2 storage. Therefore, reservoirs giving low influence of gravity forces are more suitable for CO2 storage.

Although the high-IFT fluid system had an IFT corresponding to the value of CO2-brine systems at possible reservoir conditions the flow in laboratory model was dominated by capillary forces. This kind of behaviour is less likely to be observed at the field scale. However, observations at the low-IFT fluid system resemble better the field scale flow behaviour.

A scaling analysis of the experiments and reservoir cases was performed based on dimensionless numbers. It showed that the experimental capillary number and viscous-to-gravity ratio at high-IFT and low injection rate agree reasonably well with calculations for some of the sedimentary basins and storage sites. However, low-IFT experiments scale far from the field cases when the range of flow velocities is assumed to be the same both in field cases and in experiments. Range of velocities observed in experiments is expected to occur in the reservoir far from the injection point, where the gravity forces dominate. The scaling analysis showed importance of various parameters in the process of site characterisation for CO2 storage. Representation of the reservoir conditions by means of the dimensionless analysis provides possibility of comparing various storage sites and predicting the flow regimes that may occur when CO2 is injected.

The laboratory experiments were modelled using numerical reservoir simulation software. In case of high-IFT system the flow was dominated by channelling. These features were caused by strong capillary effects and were challenging to model. This problem was solved by modifying properties of the simulation grid and results of sensitivities are presented. Simulations of low-IFT displacements showed accurate reflection of the experiments.

Sensitivities on influence of capillary pressure were also performed. In the laboratory experiments capillary pressure was negligibly small but in the field-scale modelling on generic models results proved to be very susceptible to this parameter. It strongly influences migration speed and thickness of the CO2 front.

Further analysis on full field case showed that low permeability of the storage site will have negative impact on the storage capacity and well injectivity. Due to low permeability the injected CO2 will not be able to reach all parts of the reservoir, hence not all available pore space will be utilized. Additionally, limited injectivity will result in lower injection rates, as increase of pressure in the near-well zone has to be avoided, or introduction of additional injection wells will be necessary to compensate for lower injection rates.