Leaning Into The Problem, An Experimental Study of Mixing Groundwater With Geothermal Water, Towards Full Integration of Reykjavík’s Two District Heating Systems

Mixing of heated groundwater with geothermal water has proven problematic due to the different chemical properties of the two types of water. At their geological origins, the two waters are in equilibria with the reservoir they are extracted from and specific to the problem of mixing, the geothermal water contains a high monomeric silica concentration and a relatively high pH (>9.0). In contrast, the groundwater has a low silica concentration and lower pH but contains a small amount of magnesium. When mixed, silica and magnesium are introduced at alkaline conditions and thus amorphous magnesium silicate precipitates. The district heating system in the Reykjavik capital area is supplied with two sources of water, heated groundwater from two co-generative geothermal power plants and low temperature geothermal water from geothermal fields in Reykjavik. Because of this precipitation and resulting scaling issues, the district heating system is operated as two parallel systems with two types of water that cannot mix. Our proposed solution to this problem is to intentionally induce precipitation of magnesium silicate in the heated groundwater at the power plants in a controlled process to remove magnesium. We do this by mixing the high temperature geothermal water (post steam separation) with the heated groundwater to induce precipitation and filter it out. In this process we produce “synthetic” low temperature geothermal water with a geochemical profile as close as possible to the water from the low temperature geothermal fields in Reykjavík. Experiments to develop and validate this solution have been performed on a laboratory scale and pilot scale tests are ongoing. Our goal was to study the precipitation reaction through which magnesium silicate is formed. Firstly, high temperature geothermal water and heated groundwater were mixed in varying ratios to research whether scaling happened at any ratio of mixing and the effect of the mixing ratio on the rate of the reaction. We found that in mixtures with >20% geothermal water the reaction proceeds in a matter of days through homogenous nucleation, but only after a lag time that extended to about a week. Secondly, we mixed the two types of water in the presence of magnesium silicate seed. With the addition of magnesium silicate seed the lag time is eliminated and the reaction proceeds through heterogenous nucleation in a few hours to provide a solution with a magnesium concentration below 0.5 mg/L. The main factors that affected the rate of the reaction were pH, silica concentration, magnesium concentration, and the type, size, and amount of seed material. Thirdly, we mixed the resulting synthetic low temperature geothermal water from the previous experiments with real low temperature geothermal water to confirm that no further precipitation occurs through mixing of the two types of water. This served to validate our proposed solution to being able to operate a single district heating system. No precipitation was found to happen, solidifying our belief that once a reliable process to perform this precipitation reaction at scale is developed, the two separated district heating systems in Reykjavík can be joined.