Coupling of multi-physical processes for the simulation of gas wells

In September 2010 the German government announced a new energy concept. The goals comprise essential changes of the power supply until 2050. The large challenges of this huge project are represented by both a failure-resistant energy-supply encompassed by a large contribution of strongly fluctuating energy production drawing on wind and solar power plants, as well as a sustainable reduction of greenhouse gas. For this goal, underground storage units can be utilized in different ways: 

• Hydrogen storage might be used to store huge amounts of electrical energy in geological subsurfaces. In this technology, electrical energy is transformed in chemical energy and, accordingly, can be stored in comparably small space. First, the surplus electrical energy is utilized to separate water into hydrogen and oxygen by means of electrolysis. The hydrogen is condensed and pressed into a geological formation. During periods of less power production, but high demand, the hydrogen is extracted and transformed into electrical energy by generators or fuel cells.
• POWER-to-GAS: Hydrogen and biogas can be injected into the current natural gas grid. Studies have shown that concentrations in the range of single-digit percentage are conceivable. In this concept, the natural gas grid and existing underground gas storages are used as storage facilities.
• CO₂-storage: The greenhouse gas carbon dioxide, which frequently is a byproduct of energy plants on the basis of fossil fuel, but also in various industrial processes, can be separated and stored in geological formations.

For all of the applications mentioned afore, gas wells and geological storage formations are the system critical components. In order to assess the permanent technical integrity, coupled simulations of the compound system are required.

The storage of natural gas is performed for more than 100 years in Germany to balance the daily and seasonable fluctuations of the gas consumption. Thus, it is a well-established technology. The experience using other gases such as hydrogen, carbon dioxide or mixtures of gases is limited. In Great Britain and the US there are occasional cavern storages in which comparably small total volumes of pure hydrogen (>95%vol) is stored. It is reported that this works well. However, using former gas storages and aquifers, where the gas in not stored in a large cavity, difficulties might arise for carbon dioxide and hydrogen. There is some experience using town gas with less than 50%vol of hydrogen. 

Especially, the integrity of bore holes must be mentioned. In this case, the change of the temperature and stress state and corrosion processes can lead to leakages. Possible pathways of the leakage are the connection between the cap and reservoir rock and the cement as well as between cement and casing. 

Storage wells in Germany are mostly older than thirteen years and were only established for natural gas. The usage for other gases might lead to problems for different reasons: 

• Molecular hydrogen might penetrate into the metal pipes' lattice leading to hydrogen embrittlement.
• Hydrogen storage might lead to micro-biological activities. In this context produced hydrogen sulfide is dangerous, since it leads to corrosive solvents.
• Moreover, CO₂ and biogas might contain corrosive components.
• Hydrogen can disappear much faster through small cracks due to their high diffusivity and small molecule size.
• The injection of particular gas components (e.g. hydrogen and carbon dioxide), which have not been initially in the reservoir, might lead to a number of bio- and geochemical processes, which change the properties of the rock. If this is coupled with the stress and temperature changes, cracks in the cement or in the connections can be introduced. This is very dangerous in the capstone region, since a leakage of the gas deposit might emerge.

The main goal of the project is the development of a coupled numerical model, which is able to describe the hydrodynamical, bio- and geochemical, thermal and geomechanical processes close to the well. For this reason, both the well as well as the cement and the geological formation (storage and cap rock region) have to be incorporated into the numerical model. Using surface coupling this is coupled with a constitutive model of the casing. Afterwards, the entire coupled model is used for studying different issues such as the storage of hydrogen, carbon dioxide and mixture of gases. In a first step, the program DuMuX (Flemisch et al., 2011), which can describe the hydraulic and chemical behavior in the subsurface, is coupled with a model of thermomechanics implemented in the finite element program TASAFEM (Hartmann, 2003). In the third year of the project, a geomechanical constitutive model with pressure-dependent yield function will be implemented into the finite element solver, in order to perform fully coupled thermal, mechanical, hydraulic and chemical computations treating the process during injection and withdrawal. In this context, the coupling tool must contain new concepts of time-integration, acceleration techniques for the partitioned approach using non-linear Gauss-Seidel procedure, and interpolation techniques for data transfer. Within the time integration step-size controlled, diagonally-implicit Runge-Kutta methods should be extended to incorporate different time-scales.

In this project, both modern and efficient procedures are taken into consideration to develop a coupling tool for multi-physics. Furthermore, preliminary studies will be performed which enter the research activities of the drilling simulator in Celle. These investigations contribute to its major goals, i.e. the combination of experiments, theory, simulation and application.

• Prof. Dr.-Ing. Stefan Hartmann
• Prof. Dr. Leonhard Ganzer