Thu, 16 September, 2021
Understanding thermodynamics for geological hot brines is a challenge, not so much because of thermodynamics itself (already a well understood topic from a phenomenological point of view) but is due to interaction with the physical chemistry and mineralogy of the reservoirs under quite extreme conditions. Deep boreholes with aggressive liquids and salt steam up to almost red-hot temperatures, attacks sensors, pipes, heat exchangers and other technical installations thereby making direct observations almost impossible. Additionally, electrically charged units (cations and anions) in the fluids form so-called electrolytes lack good engineering model descriptions creating further problems. These models are necessary for us to be able to calculate - in a predictable way - how industrial harvesting of geothermal energy from said liquids can be carried out in a techno-economically safe and environmentally friendly operation.
The thermodynamic modelling of these liquids in the GEOPRO project is divided into two tasks: One task is to get better data for medium temperature brines up to 250 °-300° C (CNRS, France), another task is to close the gap to super-hot steam with dissolved salts (NTNU, Norway). Superhot here means temperatures up to 450°-500° C which is sufficiently high that the steam does not condense on its way up through the borehole. The steam is dry, so to speak, but at the same time much more energy-rich than hot water (on a mass basis). In addition to the challenges that can be attributed to our limited physical understanding of hot brines and salt steam, there is another challenge, and it is how we put the model to work inside the liquid flow simulator that becomes indispensable for dealing with issues such as: flow protection and plant operation, material fatigue and technical safety, and the return on financial investment.
The scientific marriage between thermodynamics and fluid mechanics is not particularly happy, and there are many obstacles to overcome in coding a simulator that is robust enough to allow routine calculations for such demanding tasks as the GEOPRO project requires, especially when the physical models that are needed to be able to describe the fluids are not yet known. In order to do the calculations we have agreed to switch from a pressure explicit to a density explicit model formulation (using a Helmholtz energy formalism). It is assumed that this is necessary to handle near-critical reservoir conditions, but at the same time the new approach also requires additional variable transformations which in turn require thermodynamic partial derivatives of several kinds. The hand coding of all the required model expressions (for several trial models) is a time-consuming and error-prone task that we want to overcome with the help of a code exporting tool developed at NTNU in a previous project, and which has been shown to deliver robust and semantically correct code.
Article courtesy: NTNU