Numerical analysis of scaling formation in geothermal systems: application in bubble columns

Scaling summarizes precipitation of solids on the surface of pipes, heat exchangers, and other equipment in various industrial processes, including geothermal systems. Scalings in carbonate systems, which make up one of the most important geothermal reservoirs, are caused by a disruption of the lime-carbonic acid-equilibrium. Increasing temperatures (e.g. at the motor of a pump) lead to oversaturation. Decreasing pressure itself also leads to oversaturation but also to the formation of gas bubbles and stripping of CO2. The latter causes a shift to higher pH-values and massive oversaturation and affects most facilities in the North-Alpine Foreland Basin. The omnipresent scalings reduce the efficiency and may cause significant downtimes. It is therefore important to predict the amount of scalings along the geothermal cycle.

While current hydrogeochemical models are capable to predict the risk and position of scalings, they are falling short with regard to the temporal development. They generally over-predict the amount of scalings which indicates that limiting processes, e.g. diffusion limited crystal growth, partial volume effects, and local equilibria have to be considered. This requires a combination of a fluid dynamics model and a hydrogeochemical model. Since the geothermal fluids are quite heterogeneous in their hydrochemical composition (from fresh water to brine) and both, equilibrium constants and kinetic rate constants, depend on the hydrochemical composition, coupling to an established hydrogeochemical model is favored to shorthand implementations of individual reactions.

Benchmark data for such improved models can be obtained from bubble columns. Here, a gas is injected at the bottom of a usually transparent pipe filled with a fluid of known chemical composition and the effects of stripping (or gas augmentation) can be monitored with high temporal and spatial resolution. The fluid flow is accessible through tracking of particles and gas bubbles. Bubble columns are also used in different industrial processes, providing additional applications for the developed model.

In this contribution we show the coupling of OpenFOAM, a very versatile computational fluid dynamics model, with PhreeqC, a widely used hydrogeochemical model, to simulate the effects of stripping on the formation of carbonate precipitates on the pipe walls and in dispersion. The model is compared to experimental data and a hybrid hydrogeochemical model which used an effective mass transfer rate for the gas-water exchange reaction as fitting parameter to cover the rate limiting processes.