4.2.1. Potential Scale-Forming Minerals
A major objective of this study was to identify potential scale-forming minerals within the surface infrastructure of the proposed DeepStor project at the KIT campus. Scaling poses a risk to many geothermal plants, as it reduces the efficiency of heat exchangers by blocking passages and reducing heat transfer between fluids. Additionally, in extreme cases, scaling can become so pervasive that it completely blocks the movement of fluid to, from, or along the surface, at which point the system must be shut down for cleaning. This cleaning process is time-consuming, and taking a plant offline should be avoided in order to preserve the productivity of the system. Scaling risks are well known from geothermal energy production sites, but less so from HT-ATES reservoirs. The high temperature of the injection fluid at the DeepStor site is above almost all other HT-ATES systems’ temperatures [
7] and therefore needs to be investigated due to the retrograde solubility of calcite.
We identified calcite, anhydrite, antigorite, and amorphous silica as potential scale-forming minerals. Calcite and anhydrite are expected to form during fluid heating in the summer thermal recharge season. Amorphous silica may precipitate during the winter heating season, i.e., when the thermal fluid cools. Antigorite’s precipitation kinetics are believed to be too slow to occur from a fluid. No effort was made to differentiate dolomite from calcite precipitation due to the low rate of precipitation of dolomite from natural systems (22). Anhydrite only occurs as a scaling phase during the period from 0 to 6 months at the hot well, after which no more anhydrite precipitates at the surface. The fluid is supersaturated with respect to amorphous at the cold well throughout this period, with ~30 mg/kg required to be precipitated to bring the fluid back to equilibrium. If the fluid is able to equilibrate before reinjection, all of these minerals will be present in the surface infrastructure and will need to be removed.
4.2.2. Scaling Risk Mitigations
Anhydrite is a well-known geochemical risk in geothermal systems [
27,
28]. It poses a unique challenge, as it has retrograde solubility with respect to temperature, becoming less soluble with increasing temperature. This leads to precipitation during interaction within high-temperature portions of geothermal processes [
28,
29]. Anhydrite scaling also poses a problem in that acid treatments are often not effective at removing scale [
28]. Instead, alkaline solutions and chelating agents have been used which effectively remove scale from pipes at geothermal conditions [
29]. These chemical methods may be used in lieu of mechanical removal of scale.
Calcite is also a well-known scaling risk associated with geothermal systems [
28,
30,
31,
32]. Calcite may be precipitated through decompression, reaction with pipe walls, boiling, and degassing of CO
2 [
32]. In high-temperature geothermal systems, boiling is often a major concern [
32,
33]. Similar to anhydrite, calcite may precipitate in the DeepStor system during the summer thermal recharge season, due to its retrograde solubility. Several methods currently exist to deal with the scaling issues that calcite presents. One of these methods is controlling wellbore pressures with a deep pump. This method reduces calcite scaling and allows for control over the depth of scale deposition [
33]. Additionally, chemical techniques have been developed which include the use of phosphate molybdenum inhibitors [
32]. These treatments carry environmental concerns. Inhibitors based on polycarboxylates, however, have been shown to be effective at reducing calcite scaling risks under geothermal conditions [
32,
33]. Chemical inhibitors have been shown to be cost-effective because they mitigate the need for mechanical cleaning of the wellbore. Thus, they are a popular method of controlling calcite scaling [
30,
33].
No effort was made to differentiate dolomite from calcite precipitation due to the low rate of precipitation of dolomite from natural systems [
22]. Calcite precipitation at the surface is variable during the summer thermal recharge, ranging from ~1.4 to nearly 100 mg/kg. Anhydrite only occurs as a scaling phase during the period from 0 to 12 months at the hot well, after which no more anhydrite precipitates at the surface.
Amorphous silica is present as a scaling phase in this system, as it is in many geothermal systems. Silica species can be present in geothermal systems as amorphous silica, chalcedony, opal-A, and opal-CT [
28,
34,
35]. Several chemical mitigation techniques have been developed for silica scales, many of which involve the modification of specific properties of the brine such as pH, or inhibition of colloid formation through rapid brine cooling [
35,
36,
37]. There are also chemicals which inhibit the growth of silica colloids by adhering to the surface of a particle, thereby slowing further growth [
34]. The dominant method of preventing silica scaling thus far has been to drop silica out of the solution at specified points within the system, e.g., designed settling ponds that often utilize seed crystals to accelerate the precipitation process [
35,
36,
37].
4.2.3. Formation Damage
The potential for formation damage in this system is caused by mineral alteration in the reservoir. Anhydrite, daphnite, dolomite, minnesotaite, muscovite, quartz, saponite-Ca, saponite-Na, and siderite are all alteration products found in the reservoir during the simulation. Of the minerals forming in the reservoir, anhydrite, daphnite, saponite-Ca, saponite-Na, and siderite are not part of the original reservoir mineralogy. Their presence is the result of non-isothermal fluid circulation through the reservoir. Dolomite, minnesotaite, muscovite, and quartz occur naturally in the reservoir. Their concentrations also vary throughout these simulations. These minerals appear to have local phenomena affecting their precipitation and dissolution, leading to both elevated and depleted levels of each mineral at differing distances from wellbores in the model. Some minerals have a localized depletion point. Others have spikes of formation at discrete locations. The wellbores of this model act as local centers of mineral alteration.
Anhydrite forms only during the first 6 months of the simulation and is removed from the reservoir completely by 18 months. The initial formation of anhydrite is due to its retrograde solubility and the heating of the reservoir during the first injection periods. Its subsequent disappearance is due to its dissolution during a thermal extraction phase (cooling) and subsequent precipitation as mineral scale. The formation of anhydrite as mineral scale at the surface presumably removes enough sulfate from the circulating fluid to prevent additional anhydrite formation in the reservoir for the duration of the simulation.
All other alteration product minerals are present during every step of the simulation. Albite, annite, and calcite occur naturally in the reservoir, but their extent and location of their presence vary as the simulation progresses. This model shows albite and annite altering in the reservoir to clay minerals, namely, minnesotaite, saponite, daphnite, and muscovite. This alteration of albite and annite is detailed in the reactions found in
Table 6 below. The products of these reaction products may serve as reactants in subsequent reactions. It is unlikely, however, that minerals are completely dissolved and precipitated. Rather, these reactions occur largely due to cation leaching and exchange reactions. Such reactions are also seen in the cation exchange between calcite and dolomite, which, combined with iron leached from annite, subsequently leads to the formation of siderite. Additionally, the porosity and permeability both increase at the hot well in each time step, likely due to the lower levels of alteration and removal of components from the surrounding medium at that location.
Albite is altered to clay minerals during this simulation through reactions 1, 2, 3, and 4. Reactions 3 and 4 produce excess SiO2, resulting in potential silica deposition around the hot and cold wells whilst altering to daphnite and muscovite at the hot well. This is the only reactions shown above that produces SiO2; therefore, we conclude that albite alteration is the source of the additional silica in the reservoir around the wellbores. This excess silica can also be consumed in reaction 2 to produce saponite. The alteration extent from the wellbores is the same as the alteration extent of daphnite, minnesotaite, muscovite, and saponite; therefore, albite is altered to all four of these minerals in these simulations.
Reactions involving annite are defined in this system by reactions 5, 6, 7, and 8 above. Annite may alter to the same minerals in this system as albite. This is evidenced by the spatial extent of alteration being the same for these four clay minerals and albite. Additionally, annite alteration is also evidenced by the increased annite concentration beyond its depleted zone around the wellbores. This annite increase corresponds to decreases in minnesotaite and muscovite at the same locations. Thus, annite is altered to clay minerals proximal to the wellbore, whilst clay minerals alter to annite directly adjacent to this zone. Partial dissolution of albite and annite provides the components that produce the clay minerals seen in these assemblages.
Calcite and dolomite display a leeching and exchange of magnesium, as well as the addition of bicarbonate to the solution. The presence of iron released by annite weathering, accompanied by the bicarbonate released by calcite and dolomite, leads to the presence of siderite. Siderite occurs around the cold well, reaching a peak after the first 12 months and subsequently reducing. Dolomite follows an inverse trend to siderite, increasing after each injection cycle at the area closest to each wellbore, while calcite is removed from these areas. Distally from either wellbore, calcite is the dominant carbonate phase. From these observations, we conclude that calcite is altered to dolomite at the wellbore, and dolomite is altered to calcite in areas further from the wells. No other carbonate minerals exist to produce the necessary ions for alteration mineral formation; thus, calcite is the sole mineral reactant that produces dolomite and siderite, as shown in reactions 9 and 10 above.