Challenges with scaling
(see also text on ‘solutions’)
General comments
When the concentrations of dissolved components in water exceed the solubility of a mineral, solids can form. In operations, such precipitated solids are called scalings and they represent common challenges during geothermal exploitation (Boch et al. 2017, Hartog, 2016; Schreiber et al., 2016).
The scaling products lead to blockage of devices, wells and aquifers close to filters. Given the long time span that subsurface solutions have reacted with the reservoir rock, they are often nearly saturated with respect to the minerals there. However, changes to the system can affect solution composition or mineral solubility. Thus, solutions can become supersaturated with respect to minerals because of mixing of waters, geochemical reactions (such as gas exsolution), and changes in temperature and pressure.
Scaling minerals typically stick to existing surfaces and affects both productivity and injectivity because of clogging of installations, well pipes, pumps, filters, perforations, and pores of the reservoir rock (Ungemach, 2003, Blöcher et al., 2016, Gallup, 2009, Heshauss et al. 2013). An example of scaling problems observed in the infrastructure facilities is illustrated in Figure 1.
The scaling material commonly consists of carbonates (e.g., calcite, aragonite), sulphates (e.g., barite), iron compounds (sulphides, oxides), and at high salinity, chlorides (e.g., laurionite, halite) (Regenspurg et al., 2015).
Further details on scaling issues, along with additional illustration material, can be found in Eichinger (2015).
Calcite scale
Calcite is a very commonly occurring mineral in nature and likely to exist in the reservoir. In XRD data for Dutch reservoirs, for example, the mineral occurs in practically all samples that were characterised with X-ray diffraction. Figure 2 schematically depicts the formation of CaCO3 solids on the inside of a production well resulting from CO2 degassing as pressure decreases.
In a closed system, calcite equilibrates with the solution via the reaction:
CaCO3 = Ca2+ + CO32-,
until the product of the Ca2+ and CO32- activity (the effective concentration) equals the solubility product (Ksp):
Ksp = (CO32-) (Ca2+),
In this equation, (CO32-) and (Ca2+) denote the activities of the two ions, which are related to their concentrations through their activity coefficients, gCO32- and gCa2+.
Assuming that the activity coefficients are alike for the reservoir waters, this would lead to a negative relationship between the concentration of Ca2+ and CO32- if plotted on a logarithmic scale, i.e.:
log Ksp = log(CO32-) + log(Ca2+).
This is observed for the data in our database (Figure 3), although the trend is shifted upwards compared to that expected for activity coefficients of 1. This shift reflects the lower activity coefficients expected for saline solutions and, possibly the supersaturation that is required for minerals to nucleate.
The trend indicates that the formation waters are broadly poised at calcite equilibrium and that solutions can easily become supersaturated if conditions are disturbed. The solubility of calcite increases with decreasing temperature, meaning that cooling the water would not increase the thermodynamic drive for precipitation. However, a substantial drop in operational pressure may lead to de-gassing of carbon dioxide (CO2(g)), and the de-gassing process increases the calcium carbonate saturation in the thermal water, driving the reaction listed below to the right:
Ca2+ + 2 HCO3– = CaCO3 + CO2(g) + H2O.
This means that formation of bubbles as pressure decreases can cause precipitation of carbonate scales when solutions rise in the production well (Figure 3A). In our database, chemical analysis of sampled solids regularly yields significant Ca concentration, and for several of the datasets, bubbling was reported during the dissolution of the material – as would be expected if carbonate minerals were present. Thus, calcium carbonate most likely forms in many plants, albeit the amount needs not be large.
Figure 3. A): The correlation between Ca2+ and calculated CO32- concentration for all datasets in the database with measured Ca, HCO3 and pH and B): between Ba2+ and SO42- concentration excluding low salinity datasets. Lines denote the solubility of calcite and barite at infinite dilution, where activity coefficients are 1 (i.e. equilibrium at unity activity). For the saline waters, the coefficients are lower, which would give rise to concentrations at equilibrium plotting above the line. For barite, solutions require substantial supersaturation before nucleation occurs.
Barite scale
Barite is another common scaling mineral. In our dataset, Ba2+ is generally negatively correlated with SO42-. This is most clear for sites producing deeper seated, saline waters (plot of the correlation in Figure 3B), although the trend is less obvious than for Ca2+ and CO32-. Barite solubility decreases with decreasing temperature. For this mineral, cooling of water at the surface therefore decreases solubility and increases the driving force for precipitation. For solids to form they must, however, first nucleate which requires a threshold value for saturation index to be overcome (e.g., Vekilov 2010). In our dataset, calculations indicate that waters from several plants are supersaturated with respect to barite without substantial scaling occurring. We tentatively propose that more than a ~10 fold supersaturation at 25 °C is required for nucleation, and that this is part of the reason that datapoints in the Ba2+ and SO42- plot are shifted more upwards from the equilibrium line at unity activity compared to the plot for Ca2+ and CO32-. However, the actual threshold value is poorly defined and its dependence on solution composition is unknown.
Other scaling minerals
Other scaling minerals commonly observed in pipes and heat exchangers of geothermal plants producing from saline reservoirs are Sr-sulphate (SrSO4 or celestine) and Pb-sulphide (PbS or galena). In addition, minerals like pyrite (FeS2), laurionite (PbOHCl) and magnetite (Fe3O4) may form scales.
References
Boch, R., Leis, A., Haslinger, E., Goldbrunner, J. E., Mittermayr, F., Fröschl, H., Dietzel, M. (2017): Scale fragment formation impairing geothermal energy production: interacting H2S corrosion and CaCO3 crystal growth. Geothermal Energy, 5(1). doi:10.1186/s40517-017-0062-3.
Blöcher, G., Reinsch, T., Henninges, J., Milsch, H., Regenspurg, S., Kummerow, J., Huenges, E. (2016):
Hydraulic history and current state of the deep geothermal reservoir Groß Schönebeck.
Geothermics, 63, 27-43. doi:10.1016/j.geothermics.2015.07.008
Eichinger, F. (2015): Geothermal ERA-NET: OPERA, Operational Issues of Geothermal Energy Installations in Europe. Country Overview, Germany. Reference is made to: http://www.geothermaleranet.is/media/publications/OpERA_AnnexII_reduced.pdf
Gallup, D. L. (2009): Production engineering in geothermal technology: A review.
Geothermics, 38(3), 326-334. doi:10.1016/j.geothermics.2009.03.001
Hartog, (2016): Carbonate scalings in deep geothermal systems. In: Operational issues in Geothermal Energy in Europe, Chapter: 2.2. Publisher: Coordination Office Geothermal ERA NET. Editors: Stephan Schreiber, Andrej Lapanje, Paul Ramsak, Gerdi Breemboek.
Hesshaus, A., Houben, G., & Kringel, R. (2013): Halite clogging in a deep geothermal well – Geochemical and
isotopic characterisation of salt origin. Physics and Chemistry of the Earth, Parts A/B/C, 64, 127-139.
doi:10.1016/j.pce.2013.06.002
Regenspurg, S. et al., (2015): Mineral precipitation during production of geothermal fluid from a Permian Rotliegend reservoir. Geothermics, 54, pp. 122–135.
Ungemach, P. (2003): Reinjection of cooled geothermal brines into sandstone reservoirs. Geothermics, 32(4-6),
743-761. doi:10.1016/s0375-6505(03)00074-9
Vekilov, P. G. (2010): Nucleation. Cryst. Growth Des. v. 10, pp. 5007–5019.