Safe long term storage sealing of CO₂ in hydrate

Injection of carbon dioxide for safe long terms storage in solid hydrate form is a concept which have been investigated in different universities and laboratories worldwide. University of Bergen has a leading role in developing this concept up to industrial level and was conducting the experimental and theoretical research that led to a pilot test in Alaska in 2012.

The goal of this project is to close some theoretical gaps that prohibit reliable design of specific storage projects. The overall goal is to develop a new hydrate reservoir simulator which take into account that hydrates in porous media never can reach thermodynamic equilibrium.

In order to achieve these goals 3 PhD projects have been establish. The first of these projects is the development of the reservoir simulator by itself. Incorporation of non-equilibrium into a reactive transport simulator (RetrasoCodeBright) which enables the incorporation of different hydrate phase transitions as pseudo reactions with specific kinetic rates. A second project is devoted to fundamental kinetic modelling using phase field theory. This theory is free energy minimization under constraints of heat and mass transport. To accomplish this a common reference level of thermodynamic properties, including adsorbed phase on mineral surfaces, needs to be developed. A third PhD project is dedicated to this aspect.

Unlike conventional oil and gas reservoir simulator s there are not any real hydrate production cases to use for comparing model results. The only test done with carbon dioxide injection for combines storage of carbon dioxide and release of in situ natural gas was limited to few weeks and a production technology (“huff and puff”) which is hardly the preferred technology. The demands for high level of rigor in the modelling tools is therefore an absolute must. In that sense the RCB Hydrate simulator is totally unique worldwide.

During 2015/2016 the project has made substantial progress in the understanding of the impact of mineral/water interactions. Water adsorbing on Calcite versus adsorbing oc Kaolinite (part of clay) appear to have very similar characteristics in the first adsorption layer of water but there is a substantial difference in the structure between the first and second peak , which is significantly broader for Kaolinite. For Kaolinite this implies that methane and/or carbon dioxide can be trapped in between these water layers and benefit restructuring over to hydrate. Preliminary simulations of larger systems indicate that bubbles of methane can be trapped close to the water structure. These are ongoing studies and we expect to publish results early autumn.

On the nano to pore scale modelling a totally new programming of the code has been initiated, along with some changes. The use of spectral solver combined with multi-GPU (graphical processing unit) parallel architecture makes it possible to study very large model systems over long time scales. Preliminary results confirm earlier findings from small systems but the efficient architecture makes it possible to include difficult samplings like for instance dynamic of surface effects and a more detailed analysis of various competing phase transitions. Publications are under writing.

On the RCB development we have completed the important phase transition for the CH4/CO₂ exchange and we are now including also additions of nitrogen. Several publications are submitted and in preparation.