Solution mining

Solution mining entails the injection of brine solutions into underground potash-bearing or other salt seams. The solution dissolves soluble potash-bearing minerals from the seam, and the pregnant potash-bearing solution is then recovered to the surface for processing. Solution mining techniques, focused on caverns in halite, are discussed in detail in Chapter 13, Warren (2016). Solution mining can substitute for conventional shaft mining in some potash deposits at depths of more than 1,100m, which is the current limit for conventional potash mining. It is not an all-encompassing mining alternative to be used whenever potash zones are too deep or too variable for conventional mining methods. Currently, there are six active and a number of planned solution mining operations, focused specifically on potash recovery (Warren, 2016). Thickness, mineralogy, and structure/ore-continuity, as well as other technical considerations of the mining operation, must be evaluated to determine the suitability of solution mining.

Belle Plaine-Mosaic_Sask9-cropped%20copy

Intrepid potash solution mines potash-rich brine from deformed salt beds in the Paradox Formation at Moab, Utah and then uses solar concentrators to achieve a product appropriate for processing. The Mosaic operation near Moose Jaw, Saskatoon, recovers potash brine by solution mining a sylvinite bed (Belle Plaine Member) 9 to 15 metres thick at a depth of 1,650 m using a patented solution and recovery method. To capture the ore, hot undersaturated brine feed is pumped into the dissolution cavity. Concentrated potash brine is then returned to the surface processing plant, which uses a sequential evaporation-crystallisation process to recover potash and recycle hot NaCl-rich brine to the subsurface for another leach cycle. In the dry steam-driven evaporation stage in the processing plant, moisture is removed, and the recovered brine is brought to saturation point. But halite and sylvite have different temperature responses. Halite is crystallised first by evaporating the recovered brine, and then potash is recovered by cooling the remaining hot, saturated solution.

Today, most single well solution mining operations utilise a modified annular injection method, known as variable point injection, where both the tubing string and the casing string are positioned near the lowermost level of salt to be dissolved. Undercuts of more than 100 m diameter are achieved using a gaseous or liquid roof pad. In captive wells, brine production rates of more than 13 litres/sec have been obtained once the blanket is withdrawn. As solution proceeds both the tubing string and the casing strings are raised clear of the accumulating insolubles (snubbed). This type of well is also capable of achieving planned-cavity shapes suitable for waste storage.

An interesting test of the efficiency of both single-well and two-connected-wells solution mining systems targeting a thin bed of sodium borate was described by Taylor (1970). An 8.8-metre thick layer of borate ore, which was 107 metres below the land surface with a 72% ore grade, was solution-mined in a series of tests. At the completion of the tests the cavities were physically mined (entered) and inspected. No roof padding was employed during the tests. Hot water at 102° to 110°C was injected as a feed into the 21°C formation and a 15 litre/min flow produced an “almost saturated” borax solution with an exit temperature around 77°C. As the test progressed, feed temperature was steadily decreased and the allowable flow rate of the saturated solution steadily increased. On later cavern entry it was found that broad morning glory holes had formed and large amounts of insolubles had accumulated on the cavern floor. Clay and shales from the roof had fallen as intraclasts with a “sand and gravel” structure. Cavity bottoms were dominated by this material, rather than by fine bottom slimes usually postulated to dominate cavity bottoms. Some slabs of borate remained encased in the pile of insolubles, where they remained buried as unleached blocks. 

Classic "joined-well" solution mining technique

Expert design of solution wells is for the most part dependent on geological understanding of the salt to be exploited. Required knowledge includes: dip and thickness variations of the target bed; mineralogical composition, homogeneity, insoluble levels and dissolution behaviour of the target salt bed and adjacent salt layers; distribution and mineralogy of intrasalt beds; structural integrity of confining beds. Caverns engineered in salt domes tend to be more reliable and predictable structures than caverns engineered in bedded salts and are more suited to product storage especially if the product is stored under pressure. Compared to dome salt, bedded salt in the vicinity of a solution well tends to be thinner and bound above and below by more permeable formations. Bedded salt is also more likely to enclose layered intrasalt beds with varying levels of solubility and fracture intensity (anhydrite, shale and dolomite intrabeds) and so tends to supply significantly higher quantities of impurities to the cavity floor. The most stable cavern shape for purpose-designed storage resembles a giant carrot or cucumber embedded deep in a mass of salt. This ideal shape is next to impossible in bedded salt, and to continue the vegetable garden analogy, jack-o-lantern pumpkins are a more likely outcome.

K%26S%20Mine locations-bethune
K&S Bethune Mine, Saskatchewan
well%20header%20K%26%20S%20bethune mine-12
Brine well head, K&S Bethune Mine, Saskatchewan

Techniques to solution mine potash

In conventional solution mining of potash intervals, all of the halite in the dissolution cavern must be leached along with the potash to keep the face of the solution cavity open and dissolving rapidly. Otherwise, a buildup of salt and insolubles on the active dissolution face cause blinding. Bottom casings are raised (snubbed) periodically to keep above the accumulation of insolubles. In all solution brine operations, including those targeting potash intervals, flow rates are kept low enough to produce a transparent “almost saturated” brine. Because the targeted potash bed is typically thin with respect to the halite host, well capital and operating costs are comparatively high, but are reported to be still less than conventional potash mining.

When potash is recovered from diapiric intervals, the bed inclinations can be steep and thicknesses variable. Differential rates of dissolution can complicate the control of the solution cavern shape and ore bed targeting, and the susceptibility of bed flow back into the cavity is controlled by mineralogy. Relevant published potash solution mining articles and processing patents include; Bach et al. (1985), Day (1967), Dillard et al. (1975) and Schlitt (1982). See Warren 2016, Chapter 13 for more detail.

Texas Gulf solution mines potash-rich brine from deformed salt beds in the Paradox Formation at Moab, Utah and then uses solar concentrators to achieve a product appropriate for processing. The IMC Kalium operation near Moose Jaw, Saskatoon, recovers potash brine by solution mining a sylvinite bed (Belle Plain Member) 9 to 15 metres thick at a depth of 1,650 m using a patented solution and recovery method based on different temperature and solutity responses acress halite, sylvite and carnallite (Figure).

To capture the ore, hot undersaturated brine feed is pumped into the dissolution cavity. Concentrated potash brine is then returned to the surface processing plant, which uses a sequential evaporation-crystallization process to recover potash and recycle hot NaCl-rich brine to the subsurface for another leach cycle. In the dry steam driven evaporation stage in the processing plant, moisture is removed and the recovered brine is brought to saturation point. But halite and sylvite have different temperature responses (Figure). Halite is crystallized first by evaporating the recovered brine, then potash (a prograde salt) is recovered by cooling the remaining hot, saturated solution.

Carnallitite, like trona, shows incongruent solubility so that in beds composed of mixtures of carnallite, sylvite and halite, the sylvite tends to be left behind as a less soluble component that can slow the rate of dissolution (surface blinding; Figure). Furthermore, the solubility of halite, sylvite, and kieserite is very low in strong MgCl2 solutions and this limits exploitation in intervals where these minerals are co-associated. 

Phase relationship between the equilibrium constant (K) and temperature for the incongruent dissolution of carnallite to sylvite at 1 bar and 1 kbar pressures (after Koehler et al., 1990). The ion activity product (IAP) is related to the product of the concentration of Cl- and Mg++ in the solution and is equal to K along the boundary between carnallite and sylvite stability fields (see Warren Chapters 11 and 13)(
Solubilities in water of the three main potash minerals at 25°C.
Solution mining of brines from flooded mines with surface processing of brine via solar evaporotation at Cane Creek, Utah and cryogenesis at Belle Plaine (see Warren 2016, Chapter 11 for details).

Solution mining of magnesium from bischofite 

The NedMag solution mining facility near Veendam in the Nederlands produces magnesium from magnesium chloride brine by solution mining a Zechstein salt diapir, targeting beds that are a mixture of carnallite (KCl.MgCl2.6H2O), bischofite (MgCl2.6H2O) and halite (NaCl), with some sylvite (KCl) and kieserite (MgSO4.H2O). Target intervals at Veenham average 100 metres (combined) thickness at depths, dipping around 20°, at depths between 1,400 and 1,800 m (Figure A). About a 100 metres of halite lies above the target magnesium salts and more than 1,400 m occur below (Figure B Fokker, 1995).

Beds were accessed by a total of 12 solution wells drilled between 1972 and 1991. Nine of these wells were still operational in 2000 with average brine production per well around 25 m3/h. NedMag has recently excavated one of the deepest solution mining cavities in the world at 2,890 metres. Each year the NedMag plant produces in excess of 150,000 tonnes of high purity synthetic dead-burned magnesia and more than 70,000 tonnes of magnesium chloride in liquid or solid form. A similar solution-mined operation producing MgCl2 brine by targeting bischofite beds is planned for a large deposit near Volgograd in Russia.

The mining procedure at Veendam utilizes a backstepping solution cavity method (Figure A ; Fokker, 1995). Cavity extent is limited to ≈100 m diameter (perhaps now extended to 150 m), and the brine is maintained at the original downhole (lithostatic) pressure. Wells are designed to be plugged and abandoned at full pressure. During production the magnesium salt-rich layers are preferentially dissolved, leaving behind many halite balconies on the cavity walls. Eventually the balconies collapse. This creates considerable insoluble rockfall and mud deposition on the cavity floor, compared to conventional cavities that target homogenous halite. Dealing with this extra volume of collapse debris requires frequent raising of both the injection (to control the cavity diameter) and the withdrawal pipes (to prevent blockages; Figure).

Cavern pressures are kept relatively low in the producing caverns (average of 10 MPa or 100 bars) and well below the far field rock stresses. This facilitates salt creep into the cavern, but the lower viscosity of the magnesium salts compared to halite means they are preferentially squeezed like toothpaste into an active solution cavern at a rate far greater than the creep of the adjacent halite (Figureb). Hence, the bischofite beds supply far greater volumes to the dissolution cavern than the adjacent halite beds, with convergence rates contributing 30-40% of the produced MgCl-brine volume. This method has been called “squeeze mining” and the brine product from squeeze caverns can be very pure. For example, four wells producing brine from “squeeze caverns” at Veendam are almost saturated with respect to bischofite, rendering a high quality brine product with less than 1% by weight of non-magnesium chloride salts.Although the degree of creep is not well understood,it has been established that halite shows a much lower creep than carnallite, which in turn shows a much lower creep than bischofite (Figure B; Drijkoningen et al., 2012).

Although the strain rate depends on differential stress, the strain rates can be estimated very roughly, in order of magnitude, as 1:10:100, respectively. The bischofite will therefore creep towards the caverns, resulting in a gradual thinning of the salt layer. Bischofite will creep due to pressure difference between lithostatic pressure depending on the depth and the pressure applied in the cavern. In the development of cavities it is important to note that squeeze-mining has a different effect than conventional solution mining under lithostatic conditions, because part of the created brine is replaced by solid salt. This enables production of more salt from the same cavern (Drijkoningen et al., 2012).

A counter-effect is that squeeze mining leads to a more rapid surface subsidence than conventional lithostatic mining, in which there is no net volume change in the salt layers. The squeeze mining leads to extra deformation of the overburden, which in turn causes extra subsidence of the Earth’s surface. Dutch regulations now impose a maximum subsidence of the surface of 65 cm for this mine, limiting production in the future. By carefully maintaining the proper MgCl2 concentration in the Veendam cavern brines and monitoring solution flow rates using variably pressurised waters, NedMag has minimised blinding in bischofite beds (Steenge 1979). For successful magnesium chloride recovery, the solution rate, the dissolving surface and the produced brine at Veendam must be maintained at all times at an ‘almost saturated” condition.

The operations at Veendam are a worldclass example of targeted and monitored solution mining with a set of process technologies specifically designed for the brine product. Historically, high levels of carnallite in some potash beds in the Zechstein have made them a poor choice for a KCl targeted beds in both solution and conventional mining operations. But NedMag has recently begun to use a patented process to produce brines that are also saturated with respect to carnallite from carnallitic beds previously considered subeconomic

NedMag Plant
Squeeze mining of bischofite, Nederlands A) Schematic of deviated wells in magnesium salts at the NedMag plant in Veendam, Netherlands, showing solution mining method of upward retreating successive cuts with positions controlled by the blanket (after Currie and Walters, 1985). B) Squeeze mining method utilised at Veendam targeting bischofite layers (in part after Drijkoningen et al., 2012).
Effect of salt geology on cavern shape in diapiric Zechstein Salt. A) Vertical section through a cavity in Zechstein Salt in Germany showing preferential enlargement into potash-rich Kaliflöz Stassfurt horizon (shaded grey). B) Horizontal slice showing strongly asymmetrical shape formed as cavity develops preferentially into the potash facies of the Kaliflöz Stassfurt. C) Horizontal slice in a cavity in two halite lithologies where differences in crystal size and proportion of insolubles create a slight asymmetry that is maintained during growth (after Wilke et al., 2001).

© 2016-2024 Saltwork Consultants Pty Ltd