A Review of Evaporite Beds Potential for Storage Caverns: Uncovering New Opportunities

Abstract

Salt caverns serve as underground storage for crude oil, natural gas, compressed air, carbon dioxide, and hydrogen. Key stages of cavern development for storage purposes include design, construction, storage, and abandonment. The design phase addresses optimal cavern shape, size, pillar dimensions, number of caverns, the impact of interbeds, and cyclic loading while considering the creep behavior of salt and the mechanical behavior of surrounding layers. During this phase, geological factors such as depth, thickness, and the quality of salt are considered. For construction, two main methods—direct leaching and reverse leaching—are chosen based on design specifications. The storage stage includes the injection and withdrawal of gases in a cyclic manner with specific injection rates and pressures. After 30 to 50 years, the caverns are plugged and abandoned. The geological limitation of salt domes makes it essential to look for more bedded evaporites. This study provides a comprehensive review of bedded evaporites, including their origin and depositional environment. The stability of caverns in all these stages heavily relies on geomechanical analysis. Factors affecting the geomechanics of bedded salts such as mineralogy, physical properties, and mechanical properties are reviewed. A list of bedded evaporites in the U.S. and Canada, including their depth, thickness, and existing caverns, is provided. Additionally, this study discusses the main geomechanical considerations influencing design, solution mining, cyclic loading, and abandonment of caverns in bedded salt caverns.

1. Introduction

Salt caverns are known to be suitable for subsurface gas storage purposes due to their essential characteristics such as impermeability, a lower cushion gas requirement compared to other storage media, and high injectivity and deliverability rates [1]. They have long been used for the storage of crude oil [2,3], natural gas [4,5], and compressed air energy (CAES) [6,7]. Additionally, salt caverns serve as disposal sites for nuclear waste [8,9] and industrial waste [10,11]. With the shift towards cleaner energy, salt caverns are also being studied and considered for carbon dioxide (CO₂) storage [12,13,14] and hydrogen (H₂) storage [15,16,17].

Regardless of the storage application, the integrity and stability of salt caverns heavily depend on a comprehensive understanding of geomechanics, which is essential throughout the design, construction, storage, and abandonment phases [18,19,20,21,22]. While several review papers briefly mention and describe the geomechanics of salt cavern storage [23,24,25,26,27,28,29,30,31,32,33,34,35,36], others examine the topic in greater detail.

Table 1. Review papers considering the geomechanics of salt cavern storage.

Author(s) and Year	Title	Focus of Paper
Tackie-Otoo and Haq (2024) [37]	A comprehensive review on geo-storage of H2 in salt caverns: Prospect and research advances	The potential of hydrogen storage in salt caverns; the integrity of caverns with respect to geochemical reactions, microbial activities, and geomechanical considerations; risks associated with salt cavern hydrogen storage; experimental works for underground hydrogen storage in salt caverns from the geomechanics perspective; and cavern design.
Tarifard et al. (2024) [38]	Review of the creep constitutive models for rocks and the application of creep analysis in geomechanics	Creep behavior of rocks, including salt, sandstone, shale, and soft rocks; classification of creep models, including empirical, component, and mechanism-based models; comparison of the models; and application of creep analysis in geomechanics.
Ramesh Kumar et al. (2023) [39]	Comprehensive review of geomechanics of underground hydrogen storage in depleted reservoirs and salt caverns	Salt cavern construction, usage, and deformation mechanisms; potential challenges of hydrogen storage in salt caverns; salt constitutive and numerical models; and risks associated with hydrogen cyclic loading/unloading, fault reactivation, rock property alteration, and well and borehole integrity.
Vandeginste et al. (2023) [40]	Mineralogy, microstructures, and geomechanics of rock salt for underground gas storage	Mineralogy, geochemistry, and microstructure characterization of salt rocks, including a wide range of rock compositions, presence of impurities, and different structures (e.g., domal and bedded salt), and their effect on the geomechanical properties of salt rocks at the macroscale; physical and geomechanical properties of rock salt (e.g., halite, anhydrite, and gypsum minerals and rocks); and geomechanical experiments such as uniaxial compression, triaxial compression, and creep tests.
Minougou et al. (2023) [41]	Underground hydrogen storage in caverns: Challenges of impure salt structures	Hydrogen storage in salt caverns with the presence of impurities; impact of impurities on shape and hydrogen leakage; salt creep and damage models; existing impurities and mechanical stability; and influence of cyclic loading and fatigue on the thermo-mechanical behavior of rocks.
Cyran (2021) [42]	The influence of impurities and fabrics on mechanical properties of rock salt for underground storage in salt caverns—A review	The effect of existing impurities in rock salt, either in the form of interlayers, laminae, or aggregates, on its mechanical properties; short-term mechanical properties; and long-term mechanical properties in three main stages of transient, steady-state, and tertiary creep.
Małachowska et al. (2022) [43]	Hydrogen storage in geological formations—The potential of salt caverns	Underground storage experience in salt caverns; the advantages of salt caverns for hydrogen storage such as suitable physiochemical and geomechanical properties; and salt characteristics for stable cavern development with the objective of hydrogen storage.
Zhang et al. (2021) [44]	Creep properties and constitutive model of salt rock	Rock salt creep properties from the perspectives of macrocreep properties and microscopic creep deformation mechanisms; and rock salt constitutive models, including the empirical model, the component combination model, the fractional derivatives creep constitutive model, the nonlinear creep constitutive model, and the creep constitutive model considering the self-healing ability of damaged rock salt.
Cerfontaine and Collin (2018) [45]	Cyclic and fatigue behaviour of rock materials: review, interpretation and research perspectives	Cyclic loading and behavior of rocks under cyclic loading and fatigue tested under different loading conditions and various available experiments in addition to a review of typical results such as stress–strain curves, deformation evolution, dilatancy, and crack development and measurement.

provides a summary of review papers discussing the geomechanics of salt cavern storage and the associated parameters [37,38,39,40,41,42,43,44,45].

As shown in Table 1, the studies reviewed here primarily focus on salt caverns and related geomechanical issues. However, most of these studies either broadly define salt formations without differentiating the geomechanical challenges of bedded salt caverns or address bedded salts only briefly. Specifically, the salt formations in these studies are mostly halite [46], with little attention given to various bedded salt deposits, including beds with potash and trona. It is worth noting that while salt domes are recommended for storage purposes (e.g., for hydrogen storage in [39]), the limited presence of this salt structure in certain geographical regions [31] highlights the need for further investigation into bedded evaporites. This along with the suitability of salt caverns for underground gas storage, underscores the importance of smaller caverns in bedded salts that are available in different regions for local use, such as those in Teesside, UK [16]. It is also important to note that the areal extent of evaporite beds can play an important role in constructing multiple smaller caverns, such as those in trona beds in Wyoming [47]. The current status of research shows a lack of thorough investigation of bedded evaporites, including variations in their properties and associated geomechanical characteristics. Therefore, this review provides a broader perspective on bedded evaporites and geomechanical studies related to salt caverns across various evaporite formations.

This review paper examines the geomechanics of bedded salt caverns. First, a review of the origin of evaporites, their depositional environment, and mineralogy is provided. The main soluble minerals in those evaporite beds, including halite, potash, and trona, and their impurities are described in this section. After that, a list of bedded evaporite deposits, including their depth, thickness, present caverns, and their usage in the U.S. and Canada, is presented. It is important to distinguish evaporite beds with dominant minerals of halite, potash, trona, and the present impurities because these minerals impose differences in the properties of the beds [40,42]. Thus, the physical properties of each evaporite bed, including density, porosity, and permeability, are reviewed. Then, the mechanical properties of evaporite beds as crucial inputs for geomechanical studies are reviewed. Physical and mechanical properties are both affected by mineralogy and depositional environment and in turn influence salt cavern geomechanics. The key geomechanical issues in bedded caverns are then addressed. From a cavern life-cycle point of view, the geomechanical issues are divided into cavern design, solution mining, cyclic loading, and ultimately, cavern abandonment. The design of caverns in bedded salts is highly affected by the depth, thickness, areal extent, and presence of impurities, which in turn affects the shape, size, number of caverns, and pillar size. In terms of solution mining, the two main issues encountered are wellbore integrity (mainly casing) and cavern stability (formations present including salt and interlayers) during solution mining based on different cavern designs such as single and double well caverns. Once the cavern is leached, gases are injected and withdrawn in a cyclic mode to overcome seasonal shortages and demands. In this case, during cyclic loading, pressure and temperature are two main parameters changing and impacting the geomechanical factors such as creep, which in turn affects cavern shrinkage and size reduction. This reveals the importance of monitoring, which is essential during the entire life of salt caverns from construction to abandonment. Referring to abandonment as the action after gases are no longer stored in the cavern, the method of cavern filling, the type of infilled fluid, and the plugging method can lead to geomechanical issues.

2. Evaporites and Bedded Evaporite Deposits

Broadly defined, evaporite deposits refer to chemical sedimentary rocks formed from the precipitation of solid mineral crystals in a concentrated brine solution [48]. The formation of these deposits can result from processes such as freeze-drying or evaporation [49], primarily from the evaporation of seawater [50]. Figure 1 presents a simple schematic of the two main components involved in forming evaporites: solar heating and surface or near-surface brine concentration [51]. Evaporite deposits form when the rate of evaporation exceeds the rate of precipitation [52].

This section reviews various theories explaining the origin of evaporites, different depositional environments, and evaporite minerals. In addition, bedded evaporites in the U.S. and Canada are described. This section also includes a list of the number of existing caverns in the U.S. and Canada, including their locations and usage. The depth and thickness of caverns in bedded salts are also provided, which will be useful for future design and development of caverns in similar formations.

2.1. Origin of Evaporites

Several theories have been developed to explain the origin and formation of evaporites from various perspectives. The first theory, known as the Bar Theory [53], was created to describe the formation of thick saline deposits [54]. Proposed initially by Ochsenius (1888), based on ideas from Bischof (1864), Bar Theory suggests that evaporites are deposited through the precipitation of dissolved salts in a restricted arm of seawater, isolated by a barrier that limits inflow [55]. Bar Theory outlines the deposition process in five stages (Figure 2) [56] as follows:

• Isolation of an arm of the sea by a barrier with periodic inflow of fresh seawater;
• Increase in mineral concentration within the arm;
• The continued flow of seawater into the arm further raises mineral concentration and brine density;
• Deposition of various evaporites leads to a reduction in basin depth;
• Separation of the basin from the sea due to geological factors, like uplift.

While the Bar Theory explains the marine origin of evaporites, other theories have emerged to account for non-marine evaporite origins. For example, the Desert-Basin Theory, proposed by Walter (1894), addresses the absence of fossils in saline deposits. Grabau (1913) used the Desert-Basin Theory to explain the origin of the Silurian Salina Formation in the U.S. by connecting evaporite deposits to desert redbeds and suggesting that connate water in pre-existing sediments provides a sufficient salt source for evaporites [54].

Branson (1915) introduced the Ring Theory (Modified Bar Theory), proposing that different evaporites are deposited in separate basins. This theory explains the occurrence of monomineralic evaporites by hypothesizing separate basins for deposition: the outer basin is where gypsum is deposited, while the inner basin is reserved for halite [53,54,56]. Figure 3 illustrates the concept of evaporite deposition according to the Ring Theory [56].

Another modified Bar Theory, known as the Reflux Theory, was introduced by King (1947) to explain the formation of thick, extensive gypsum deposits. This theory posits that when seawater enters an evaporation basin through a channel in the barrier, two layers of brine with different densities develop within the basin. The denser brine remains at the bottom, and a process called backflow, or reflux, allows this denser brine to maintain a constant density and composition, facilitating the deposition of thick gypsum layers without halite formation [53,54,57]. Scruton (1953) further expanded on the Reflux Theory by suggesting a lateral salinity gradient in basins where evaporites precipitate (Figure 4) [53,54,55].

Further theories have been developed to explain both deep- and shallow-water evaporite depositions. The Mediterranean evaporite serves as an example used in [54] to illustrate three distinct theories: the “Deep-water Deep-basin Model” by Schmalz (1969), the “Shallow-water Shallow-basin Model” by Ogniben (1957), and the “Desiccating Deep-basin Model” by Hsu (1972). Figure 5 provides a schematic representation of these three models of evaporite deposition [58].

The Fractional Sedimentation Theory, proposed by Jeremic (1965), Braitsch (1971), and Yumuang (1983), describes the cyclic nature of evaporite deposition. This theory attributes cyclic evaporite deposition of substantial thickness to marine transgressions and regressions (Figure 6), resulting in various sequences of evaporite layers due to the precipitation of brines with differing mineral concentrations [56].

The theories reviewed in this section address diverse aspects of evaporite formation. While some theories focus on a general mineralogical description of evaporites (e.g., Bar Theory), others provide detailed models by differentiating mineral types during the formation of evaporite deposits (e.g., Reflux Theory). Factors such as the depth, thickness, and mineral composition of evaporite beds can be better understood through these theories. The depositional environment of evaporite deposits is another crucial factor that influences their properties and is discussed in the next section.

2.2. Depositional Environment of Evaporites

The depositional environment refers to the conditions in which sedimentary rocks form. The depositional environment influences characteristics such as rock lithology, mineral composition, texture, and sedimentary structures [59]. Evaporite deposits are generally classified as marine, non-marine, or a combination of both, depending on the source of their primary brine [60]. According to Ochsenius’s Bar Theory, the depositional environment for thick evaporites is a lagoon separated from the sea by a barrier. This setting is typically warm, arid, and limited in freshwater influx [54]. Ancient marine evaporites, which are the dominant type, are further divided into two settings: platform evaporites (with beds less than 50 m thick) and basinwide evaporites (with units greater than 50–100 m thick) [61]. Walter’s Desert-Basin Theory describes non-marine environments for thick evaporite deposits [54], with such deposits often forming in saline lakes, such as perennial lakes, ephemeral lakes, playa lakes, playas, salinas, and other similar settings [62].

Other researchers have proposed different classifications for evaporite depositional environments. For instance, three main types of evaporite basins—lagoon basin, salina basin, and saline lake—are identified in [63]. Figure 7 illustrates these three types of evaporite basins.

A classification that applies to both marine and non-marine environments was proposed in [64]. The three depositional environments identified are marginal (a mix of shallow subaqueous and subaerial), shallow subaqueous, and deep subaqueous. Figure 8 displays and describes these environments.

According to [65], evaporites can form in various settings such as sabkhas, barred basins, and shallow epeiric shelves. This source also links evaporites to black shale and limestone deposits, suggesting deposition in euxinic (low-oxygen) conditions. Most North American evaporites are classified as intra-basin deposits (formed in the center of basins), while others are basin-margin deposits. Intra-basin evaporites are further classified based on sill types as tectonically silled or reef-enclosed, whereas basin-margin evaporites are divided into back-reef and topographically silled basin-margin deposits [66]. A separate study in [67] suggests that modern marine evaporites are deposited in supratidal and coastal salina margins of marine basins, reflecting the hot, arid conditions conducive to their formation. This study also examines the impact of meteoric water on the mineralogy and sediment characteristics of evaporite sequences. Other studies have investigated how microbial activities influence evaporite sediment structures [68,69,70]. Although direct precipitation is identified as the main factor driving evaporite accumulation, the depositional environment is complex, with additional factors such as sediment transport, currents, biological activity, groundwater gradients, and underground movements influencing deposition [71].

Several studies focus on evaporite depositional environments in the U.S. and Canada [72,73,74,75,76,77,78,79]. For instance, bedded halite in the Permian San Andres Formation was deposited in perennial brine-pool conditions, according to [73]. Additionally, ref. [74] defined two cycles in the McNutt Potash Zone of the Permian Salado evaporites: one cycle in a shallow, marginal marine basin and another in a shallow saline lake. Another study on Permian Basin salt deposition proposes a coastal sabkha salt pan environment [72]. Research on Michigan basin salt indicates a deep-water depositional setting, whereas salt in the Appalachian basin is linked to shallow, turbulent water environments [75]. For the Williston Basin, three depositional environments—marine open-circulation, euxinic, and evaporitic—have been identified [77]. The Elk Point Basin in Canada is described by a deep-basin model [80]. Furthermore, the cyclic deposition of potash is detailed in [78], and trona deposition in large lakes is explained in [79].

Various methodologies have been applied to understand evaporite depositional environments. These methods include environmental marker analysis [81], mathematical modeling [82], sedimentological analysis [83], laboratory experiments [84], geochemical field studies [85], and seismic and well data analysis [86,87,88,89,90,91].

The depositional environment—including factors like water chemistry, evaporation rate, and other conditions—determines the mineral composition of evaporites. The sequence of mineral precipitation, along with the depositional setting of these minerals, is discussed in the next section.

2.3. Evaporite Minerals

In the previous section, the marine and non-marine sources of brine and depositional environments of evaporites were explained. The types of evaporite minerals that form depend on the brine chemistry in these environments. For marine evaporites, the primary minerals are halite and/or anhydrite, though other minerals like evaporitic carbonates and potash salts are also present. Non-marine brines, on the other hand, produce three main types of evaporites: soda ash, salt-cake, and borate deposits [60]. According to [92], non-marine evaporite basins contain five major water types from which various saline minerals are produced: (1) Ca–Mg–Na–(K)–Cl, (2) Na–(Ca)–SO₄–Cl, (3) Mg–Na–(Ca)–SO₄–Cl, (4) Na–CO₃–Cl, and (5) Na–CO₃–SO₄–Cl waters. Some minerals, such as halite and gypsum, are found in both marine and non-marine environments [62,93], as shown in

Table 2. Major saline minerals (marine, non-marine, or both settings) [94].

Mineral	Composition	Marine	Non-Marine
Anhydrite	CaSO4	X	X
Aphthitalite (glaserite)	K2SO4·(Na, K)SO4		X
Antarcticite	CaCl2·6H2O		X
Aragonite	CaCO3	X	X
Bischofite	MgCl2·6H2O	X	X
Bloedite (astrakanite)	Na2SO4·Mg SO4·4H2O	X	X
Burkeite	Na2CO3·2Na2SO4		X
Calcite	CaCO3	X	X
Carnallite	MgCl2·KCl·6H2O	X	X
Dolomite	CaCO3·MgCO3	X	X
Epsomite	MgSO4·7H2O	X	X
Gaylussite	CaCO3·Na2CO3·5H2O	X	X
Glauberite	CaSO4·Na2SO4	X	X
Gypsum	CaSO4·2H2O	X	X
Halite	NaCl	X	X
Hanksite	9Na2SO4·2Na2CO3·KCl		X
Hexahydrite	MgSO4·6H2O	X	X
Kainite	MgSO4·KCl·11/4H2O	X	X
Kieserite	MgSO4·H2O	X	X
Leonhardtite	MgSO4·4H2O	X
Leonite	MgSO4·K2SO4·4H2O	X	X
Mirabilite	Na2SO4·10H2O	X	X
Nahcolite	NaHCO3		X
Natron	Na2CO3·10H2O		X
Pentahydrite	MgSO4·5H2O	X
Pirssonite	CaCO3·Na2CO3·2H2O		X
Polyhalite	2CaSO4·MgSO4·K2SO4·2H2O	X	X
Shortite	2CaCO3·Na2CO3		?
Sylvite	KCl	X	X
Tachyhydrite	CaCl2·2MgCl2·12H2O		X
Thenardite	Na2SO4		X
Thermonatrite	Na2CO3·H2O		X
Trona	NaHCO3·Na2CO3·2H2O		X

, which lists major saline minerals and their occurrence in different settings [94].

Regardless of their marine or non-marine origin, evaporite minerals can be categorized into three main groups, as identified in [48]: carbonate minerals (e.g., calcite and dolomite), sulfate minerals (e.g., anhydrite and gypsum), and chloride minerals (e.g., halite and sylvite). A list of the most common evaporite minerals can be found in [95].

Research has also indicated that evaporite minerals precipitate in a specific sequence [94,96,97]. Carbonates are the first to precipitate when brine reaches twice the concentration of seawater, at a density of approximately 1.10 g/cm³. As brine concentration increases to five times that of seawater, with a density of around 1.13 g/cm³, gypsum begins to precipitate. Halite, which precipitates at the highest density of approximately 1.22 g/cm³, forms when the brine concentration reaches 10 to 12 times that of seawater [97].

To understand the mineralogy of local evaporites, it is necessary to investigate specific evaporite deposits in various locations. Examples of such research include studies on the Saline Valley in California [85], the Upper Miocene Solfifera Series in Sicily, Italy [71], the Permian Salado Formation in Texas [74], the evaporites of the Abqaiq Field in eastern Saudi Arabia [91], and evaporite mineral formation in lakes of Western Canada [98]. Various methods are used to identify evaporite minerals and their precipitation sequences, such as X-ray powder diffraction (XRD) analysis [99], scanning electron microscopy (SEM) [100], and remote sensing [101].

According to [102], halite is the dominant mineral in rock salts, though other saline minerals like gypsum, anhydrite, carnallite, kieserite, and polyhalite may also be present. Salt beds can be associated with shale and sandstone. In potash beds, as defined in [103], the dominant potassium-bearing minerals are sylvite and carnallite, with polyhalite also reported in some deposits. In trona beds, trona is the most common mineral, with additional sodium carbonate minerals like nahcolite and wegscheiderite found in some deposits. Some trona beds also contain a mix of halite and trona [104].

Evaporites contain both soluble and insoluble minerals. Soluble minerals in evaporites include halite, potash, and trona [56]. Halite (NaCl) and sylvite (KCl), the most common potash mineral, are simple chloride salts, while trona (NaHCO₃·Na₂CO₃·2H₂O) is a carbonate double salt [63]. However, evaporites also contain insoluble minerals, such as gypsum and anhydrite [48].

This study focuses on evaporite beds containing halite, potash, and trona. These beds are significant for two main reasons: their economic importance due to the industrial use of these minerals and the potential to use caverns left by solution mining for storage purposes [105].

2.4. Bedded Evaporite Deposits and Caverns in the U.S. and Canada

Salt deposits used for constructing salt caverns can be either salt domes or salt beds. Initially, all salt deposits are formed as salt beds; over time, salt domes develop when salt flows through overburden layers [106]. Figure 9 shows caverns developed in both salt domes and bedded salt. As seen in the figure, solution mining methods, cavern shape, and the presence of insoluble interlayers are the primary differences between caverns in bedded salt and those in salt domes.

Salt domes are geographically limited in distribution. The locations of salt domes and salt beds in the U.S. are shown in Figure 10. With the increasing importance of underground energy storage, investigating and developing caverns in bedded salts is crucial.

2.4.1. Bedded Evaporites

Significant evaporite deposits in the U.S. include the Gulf Coast Basin, Paradox Basin, Permian Basin, Supai Basin, and Williston Basin, as described in [102]. In Canada, the major evaporite beds are found in the Elk Point Basin [109].

Table 3. Evaporite deposits and their locations in the U.S. and Canada [110].

Country	Salt Deposits	Basin	Location
U.S.	Major salt deposits	Gulf Coast Basin	Alabama, Arkansas, Louisiana, Mississippi, Texas
		Paradox Basin	Colorado, Utah
		Permian Basin	Colorado, Kansas, Oklahoma, West Texas–Eastern Mexico
		Salina Basin	Maryland, Michigan, New York, Ohio, Pennsylvania, West Virginia
		Supai Basin	Arizona, New Mexico
		Williston Basin	Montana, North Dakota, South Dakota, Wyoming
	Other salt resources	-	Eastern United States: Alabama, Florida, Virginia
			Western United States: Arizona, California, Colorado, Green River Basin in Wyoming, Hawaii, Idaho–Wyoming, Lusk Embayment, Nevada, New Mexico, Oregon, Texas, Utah, Washington
Canada	Major salt deposits	Elk Point Basin	Alberta, Manitoba, and Saskatchewan
	Other salt resources	-	Alberta, British Columbia, Labrador, Manitoba, New Brunswick, Newfoundland, Northwest Territories, Nova Scotia, Ontario, Prince Edward Island, Quebec, Saskatchewan, Yukon

lists the salt deposit basins and their locations in the U.S. and Canada according to [110].

A summary of bedded salts in the U.S. and Canada, including basin, deposit, depth, and thickness, is provided in

Table 4. Bedded salts in the U.S. and Canada [111].

Country	Basin Name	Deposit Name	Depth	Thickness
U.S.	Michigan Basin [112,113]	Salina Salt [113]	130–>2000 m	100–>500 m
		Detroit River Salt [114]	>500–>1250 m	5–150 m
	Appalachian Basin	Salina Salt [115]	>50–>3100 m	3–>150 m
	Paradox Basin	Paradox Salt [116]	>140–>2100 m	6–240 m or thicker
	Williston Basin [117,118]	Salt of Madison Group	>800–>3000 m	A few meters to >110 m
		Salt of the Opeche Formation	>1600–>2300 m	Max. 50 m
		Pine Salt	1200–2400 m	Max. 100 m
		Dunham Salt	1400–2300 m	Max. 30–40 m
	Denver Basin [119]	Permian Salt [120]	>800–2500 m	Max. >150 m
	Permian Basin [121]	Hutchinson Salt Member [122]	>40–>1000 m	Max. >190 m
	Permian Basin–Anadarko Basin	Lower Clear Fork Salt [72]	>200–>1700 m	2–8 m, max. 160 m
		Upper Clear Fork Salt	>120–>1500 m	2–8 m, max. 190 m
	Permian Basin–Palo Duro Basin	Salt of the San Andres Formation [123]	>90–>1100 m	150–200 m, max. >500 m
	Permian Basin–Delaware Basin	Salt of the Castile Formation [124]	max. 270 m	500–1100 m
	Midland, Delaware, and Palo Duro Basins	Salt of the Salado Formation [125]	<50–>790 m	30–300 m, max. 500 m
	Holbrook Basin	Salt of the Supai Formation [126]	>140–>800 m	2–9 m or max. >180 m
	Luke Basin	NA	300–>2000 m	>1000 m
	Red Lake Basin	NA	450–>2000 m	>1000 m
	Virgin Valley Basin	NA	Near or at surf., max. >1000 m	Max. >300–400 m
	Sevier Valley Basin	NA	Normally >1800 m–3600 m	20–>600 m
	Eagle Valley Basin	NA	>450–1600 m	10–60 m
	Piceance Basin	NA	>550 m	60 m
	Green River Basin	NA	200–750 m	<1 m
	Great Basin and Range	NA	Near or at surf., max. >2500 m	2–9 m
	Appalachian Basin	Salt of Saltville Area	Near to the surface, max. >1000 m	Total 240 m
	South Florida Basin	NA	3300–>3600 m	3–10 m (total)
	Verde Valley Basin	NA	NA	NA
	Safford Basin	NA	>250 m	>200 m max.
	Picacho Basin	NA	450 m	>250 m max.
	Higley Basin	NA	700 m	NA
	Detrital Valley Basin	NA	125 m	210 m max.
	Date Creek Basin	NA	NA	NA
	Northern Gulf Coast (Louann Salt Basin)	Louann Salt	Near to the surface, max. >18,000 m	450–>1500 m
	Gulf Coast Basin	NA	Near to the surface, max. >18,000 m	450–>1500 m
U.S., Canada	Western Canada Basin–Williston Basin	Prairie Salt [118]	>200–>3800 m	20–>300 m
Canada	Western Canada Basin–Williston Basin [127]	Lower Lotsberg Salt	1050–>2100 m	>60 m
		Upper Lotsberg Salt	750–>2100 m	28–>150 m
		Hubbard Salt	NA	Max. 18.9 m
		Cold Lake Salt	550–>2400 m	Max. >80 m
	Maritimes Basin	NA	>100 m max. >6000 m (diapirs)	>500 m or thicker in diapirs
	Mackenzie Basin	Saline River and Mount Cap Salt	Near to the surface and >2000 m	100 s of meters to >1000 m
	Sverdrup Basin	NA	Near to or at the surface, >4000 m	10 s of meters to >1000 m

[111]. Additional information on salt deposits in other regions of the world is also available in [111].

Numerous researchers have described the geology, depth, thickness, and distribution of evaporite beds in the U.S. and Canada. Salt deposits in the U.S. are detailed in [115], while the main salt deposits in Canada and their characteristics are discussed in [128]. Potash deposits in the Elk Point Basin (U.S. and Canada) are covered in [129]. The largest trona deposits in the U.S. are reported in [130], with additional details on the geology and mineralogy of these beds provided in [104].

2.4.2. Caverns in the Bedded Evaporites

According to [131], there are 286 cavern fields in bedded salt deposits globally. The characteristics of these deposits, such as depth, thickness, and mineralogy, play a vital role in the suitability of cavern development for storage purposes. Salt caverns are artificial cavities created through a process called solution mining, which involves injecting water into the salt deposit to dissolve it, producing brine, and leaving a cavern suitable for storage. The location and storage uses of existing caverns in the U.S. and Canada are listed in

Table 5. Existing caverns in bedded salts (U.S. and Canada) [111].

Country	Location	Existing Caverns	Usage
U.S.	Arizona	11	LPG storage
		>4	LPG storage
		>4	Brine production
	Utah	4	LPG storage
	West Virginia	10 s	Brine production
	Virginia	NA	Brine production
		4	Gas storage
	Kansas	21	LPG storage
		14	LPG storage
		85	LPG storage
		43	LPG storage
		100	Natural gas storage
		71	LPG storage
		72	LPG storage
		53	LPG storage
		24	LPG storage
		NA	Brine production
		NA	Brine production
	Ohio	1	LPG storage
	Michigan	NA	Brine production
		8	LPG storage
		8	LPG storage
	New York	2	LPG storage
		15	LPG and natural gas storage
	West Texas	16	LPG storage
		18	LPG storage
		8	LPG storage
		10	Natural gas storage
		3	Natural gas storage
		6	Natural gas storage
Canada	Alberta	6	Natural gas storage
		3	Liquids storage
		4	Liquids storage
		2	Liquids storage
		2	Liquids storage
		4	Liquids storage
		11	Liquids storage
		17	Liquids storage
		13	Liquids storage
	Ontario	5	Liquids storage
		8	Liquids Storage
		6	Liquids storage
		4	Liquids storage
		9	Liquids storage
		10	Liquids storage
		19	Liquids storage
		9	Liquids storage
	Saskatchewan	4	Liquids Storage
		6	Liquids storage
		3	Liquids storage
		21	Natural gas storage
	Nova Scotia	3	Natural gas storage

.

Salt caverns for storage purposes are primarily found in halite-rich bedded salt deposits. The reported caverns in bedded rock salts (Table 5) have been mainly used for natural gas storage and liquid storage. However, no caverns have been reported for storage in potash- and trona-rich beds.

3. Properties and Geomechanics of Bedded Evaporites

This section describes the physical and mechanical properties of salt, potash, and trona beds, which make them suitable for gas storage caverns. These properties are influenced by the mineral composition of the rock, which is in turn determined by depositional environments and post-sedimentary processes [42].

3.1. Physical Properties

Density, porosity, and permeability are three key physical properties that impact the geomechanical behavior of salt caverns. Two definitions are provided in [132] for rock density: the density of rock constituents, which includes minerals and pore fluids, and the bulk density of rocks, which depends on the volume fraction and densities of the constituent minerals. The densities of halite, sylvite, and trona are 2.163, 1.987, and 2.170 (10³ kg/m³), respectively. Additional mineral densities are available in [133]. If rock pores are fluid-filled, the density of the pore fluid can vary depending on the type of fluid. In terms of bulk rock density, the mineral volume fraction and density are used to calculate an estimate. Variations in density can occur within the same crystal or among different rock crystals, as noted in [134]. It is also important to consider fluid inclusion, which refers to small amounts of fluid trapped within crystals during rock deposition or diagenesis, as this can affect overall rock density [135]. In terms of the application of density in geomechanics, overburden stress (${\sigma }_v$) can be calculated based on average density (${\rho }_a$), the acceleration due to gravity ($g$), and depth ($z$), based on Equation (1) [136]. Several researchers have used this method to calculate overburden stress, which is also known as vertical stress [21,46,137].

$${\sigma }_v={\rho }_{a}gz$$

(1)

Another important physical property that influences geomechanical characteristics is porosity. From a petroleum reservoir engineering perspective, porosity is defined as the pore volume within the rock, which indicates its storage capacity. In other words, porosity is the ratio of pore volume to the rock’s bulk volume [138]. Rock salts are typically characterized by low porosity, making them suitable for cavern storage purposes [139,140,141]. A porosity range of 0.26–3.00% has been reported for rock salt [142]. Research on the porosity of potash and trona beds is limited. According to [143,144], sylvite exhibits no porosity, whereas carnallite can reach porosity levels of up to 65%. Trona beds are reported to have a porosity range of 0.47–2.35% [145]. Although these three evaporite deposits generally have low porosity, the porosity of insoluble interlayers, such as gypsum, mudstone, and anhydrite, is also critical to be considered to ensure cavern tightness [139,140,146,147,148].

An inverse relationship between porosity and density was determined in [132]. Regarding rock mechanics, the influence of porosity on rock mechanical properties can be determined through rock models [149]. For example, as porosity increases in rock salts, rock strength decreases, as demonstrated in [150]. Relationships between porosity and permeability are also discussed in [151].

The low permeability of rock salts (10⁻¹⁷–10⁻²¹ m²) plays a crucial role in maintaining the integrity and tightness of salt caverns throughout their operational lifespan [139]. Generally, permeability is defined as a property of rock that indicates the ease with which fluids can flow through it [152]. Despite the inherently low permeability of undisturbed rock salts, changes in stress, pressure, and temperature during cavern construction, storage, and post-abandonment necessitate an investigation into how these parameters affect permeability. Studies focusing on the permeability of rock salts under such conditions are presented in [12,151,153,154,155]. Research has also considered the thermal, hydraulic, mechanical, and chemical coupling processes to understand rock salt permeability under these combined factors [148,156,157]. No reliable sources were found detailing permeability values for potash and trona beds.

While seismic data, well logs, and experimental data are commonly used to determine the physical properties of various rocks, studies on evaporites are limited. According to [105], this is because oil companies are generally less interested in collecting data on evaporites, which primarily function as caprocks in petroleum systems. Despite this data shortage, ref. [86] utilized seismic and well data to interpret evaporite minerals such as halite, anhydrite, and carnallite. In terms of well logs, ref. [158] included various log responses for halite and two major potash minerals: carnallite and sylvite. For trona beds, both [159] and [160] discussed the use of well logs—including gamma-ray, neutron, sonic, and density logs—to determine the physical properties of trona beds. Log responses for trona beds are detailed in [161]. Based on these log responses, it is possible to estimate the physical properties of halite, potash, and trona beds. Most core analyses have been conducted to examine rock salt’s mechanical properties, which will be discussed in the following section. It is also important to note that other physical factors such as the solubility of evaporites and interlayers affect geomechanical properties. For this purpose, the effect of insolubles is further described in later sections.

3.2. Mechanical Properties

The mechanical properties of rocks play a pivotal role in geomechanical applications, including wellbore stability, cavern tightness, and subsidence. The primary mechanical properties of rocks include elastic modulus (Young’s modulus), Poisson’s ratio, and rock strength [162]. Two other mechanical properties, shear modulus and bulk modulus, are derived from Young’s modulus and Poisson’s ratio. The stability and safety of cavern storage are significantly impacted by the creep behavior of salt [44].

Young’s modulus is defined as stress over strain [163]. Two main types of Young’s modulus measurement, including static and dynamic measurement, from both experiments and well logs are described in [164]. Since static Young’s modulus is essential for geomechanical calculations, empirical relations converting dynamic Young’s modulus to static Young’s modulus are also provided. An example of Young’s modulus calculation based on the uniaxial compression test is described in [165]. The experimental data are used to illustrate the stress–strain curve under a uniaxial test and show three different methods for calculating Young’s modulus, including tangent, average, and secant Young’s modulus.

Poisson’s ratio is the ratio of lateral to axial strain resulting from uniform axial stress [166]. The uniaxial compression test was used in [167] to estimate Poisson’s ratio for various rocks. Similar to Young’s modulus determination, three methods for calculating Poisson’s ratio, tangent, average, and secant Poisson’s ratio in intact rocks are suggested in [168].

Rock strength reflects a rock’s ability to withstand stress without fracturing or reaching the yield (permanent deformation) point. Rock strength can also be determined in the laboratory via uniaxial or triaxial compressive strength tests. If only an estimate of compressive strength is required, other tests, such as point-load and Brazilian tests, may be used [169]. Figure 11 illustrates stress–strain curves of different rocks under the triaxial compression test, including (1) an incomplete post-peak curve, (2) a complete post-peak curve, (3) strain softening, and (4) a plastic yield curve [170].

Salt caverns are constructed in bedded salts deep underground, making it essential to understand the failure mechanisms of all overlying layers. Cyclic loading during gas storage in caverns may lead to failure in the surrounding rock. Various failure criteria have been developed to predict rock failure under different loading conditions, including but not limited to the Mohr–Coulomb, Hoek–Brown, Modified Lade, Modified Wiebols and Cook, Mogi, and Drucker–Prager criteria, which are compared in [171]. Additionally, ref. [172] discusses the von Mises, Griffith, and Mogi–Coulomb failure criteria to describe rock failure.

A unique characteristic of rock salt is its self-healing property, controlled by its creep behavior. Creep behavior, defined as time-dependent deformation under constant stress, is crucial for the long-term stability of salt caverns. A review of creep constitutive models for various rocks is presented in [38]. Empirical models, which are described as models established based on creep test results and examples of utilized fitting functions such as the Power function, Norton Power Law, Bailey–Norton Law, etc., are reviewed. In addition, a review of component-combined creep models (e.g., Maxwell model, Kelvin model, and Burger model), as well as the common elements used in developing those models (e.g., Hooke’s body, Newton’s dashpots, and St. Venant elements), is provided. A thorough review of mechanism-based creep constitutive models aiming to describe microscopic damages because of mechanical reactions is also given. More specifically, ref. [44] reviewed the creep properties and a constitutive model of salt rock in detail.

Several researchers have investigated the mechanical properties of bedded rock salts. For experimental studies, refs. [173,174,175,176,177,178,179,180,181] reported findings on rock salt mechanical properties, including interlayers. In addition, ref. [182] compared the Mohr–Coulomb (MC), Hoek–Brown (HB), and Drucker–Prager (DP) criteria using experimental data. A review of the impact of impurities on rock salt mechanical properties is provided in [42]. Few reports are available on the creep behavior of potash and trona; however, the mechanical behaviors of potash and trona are described in [183,184]. A comparison of the creep behavior of salt, potash, and trona in [185] highlights differences in trona’s creep behavior, which is further elaborated in [145].

3.3. Main Geomechanical Considerations in Bedded Evaporite Caverns

The four main phases of a cavern’s lifespan are design, solution mining, storage of gases, and abandonment. In addition to mineralogy and mechanical properties described in previous sections, other factors affecting geomechanics during each phase must be considered. This section reviews these considerations.

3.3.1. Cavern Design

In thin-bedded salt caverns, three main issues must be addressed during the design phase: the potential fracturing of heterogenous materials, slippage along bedding planes due to differential deformation in various layers, and propagation of damage leading to rock failure around the cavern [186]. A well-thought-out design can mitigate issues such as roof instability, fractures in surrounding formations leading to leakages, and subsidence. A detailed review of cavern behavior, design methods, and stability criteria is provided in [187].

Key parameters such as size, shape, number, distance between caverns (pillar size), and roof thickness are critical for both storage capacity and stability. The storage capacity of a cavern depends on its shape, diameter, and height as shown in a study in [188]. This study used these calculations for selecting hydrogen storage sites in salt caverns in China. With respect to stability, factors like cavern height, diameter, height-to-diameter ratio, and roof thickness were evaluated in the U.S. thin-bedded salt caverns in [186]. The geometry of bedded salts and the impurities in bedded salts of the Midland basin are described in [189]. A methodology for selecting underground salt cavern gas storage sites based on these parameters was developed in [190].

Impurities in salt beds, which can include non-soluble minerals or interlayers, present unique challenges during cavern design. These impurities occupy a large volume of the cavern, reducing storage capacity [191]. These interlayers are also the cause of irregularly shaped caverns [41,192]. A concept proposed in [22] suggests using pore spaces within these impurities for storage. The study in [193] highlights the stabilizing effect of interlayers on caverns. Various factors affecting the irregular shape of caverns, such as impurities, blanket position and duration, and tubing failure are explored in [194].

Three cavern shapes, including cylindrical, enlarged top, and enlarged bottom, were considered in [195] to investigate their stability. For bedded salts, ref. [107] noted that constructing vertical caverns is challenging due to significant impurities, making horizontal caverns more suitable. U-shaped caverns were proposed in [11] for the disposal of drilling waste in salt beds with high insoluble content. The study in [196] investigated the stability of U-shaped horizontal salt caverns for underground natural gas storage. In addition, the long-term stability of horizontal elliptical cylindrical caverns was analyzed in [197], while ref. [198] developed a design separating caverns into upper and lower sections, with gas pressure dictating cavern shape and dimensions. Based on the comparison performed in [192], ellipsoid-shaped caverns are the most stable, while cylinder-shaped and cuboid-shaped caverns are the least stable. The effects of geological conditions such as depth, thickness, and mineralogy on the shape of caverns are discussed in [131].

The space between caverns, known as pillars, is another crucial factor. Pillar dimensions and cavern stability were studied in [199,200], while ref. [201] investigated how pillar width and the number of caverns impact stability. Based on this study, a pillar width of 2–3 times the cavern diameter leads to instability in the center of the pillar, while thinner pillars cause instability at the edges. Variations in rock salt properties, as noted in [202], are the main reason for the lack of systematic pillar design. An experimental work in [203] optimized the width between new and old caverns, suggesting a distance of at least twice the maximum diameter of the new cavern.

The internal pressure of the cavern is another main parameter to be determined. Studies have shown how internal pressure affects displacements around caverns [19,20,204]. Internal pressure is influenced by cavern height as mentioned in [205], while ref. [206] demonstrated its impact on cavern stability. The minimum allowable internal pressure of hypothetical caverns in bedded salts in the U.S. was studied in [207]. The simulation utilized in this study considered the effect of various parameters on minimum allowable pressure, such as cavern roof span, depth, roof salt thickness, shale thickness, and shale stiffness. The internal pressure under high-frequency injection and production cycles of hydrogen was determined in [208]. A numerical simulation in [209] revealed a direct relation between the roof span and internal pressure and an inverse relation between the roof span and the depth of the salt cavern. Finally, ref. [210] used a geomechanical model to calculate the maximum allowable pressure for two caverns in China.

3.3.2. Solution Mining

According to the Solution Mining Research Institute (SMRI), solution mining involves injecting water into evaporite formations, through one or more wells, dissolving water-soluble minerals such as salt, potash, and trona, and producing the saturated brine via a pipe string [211]. Some researchers refer to this process as leaching [212,213,214,215]. Solution mining can be performed using a single well or multiple wells. Single-well solution mining methods can follow either direct or reverse methods (Figure 12). Multiple-well solution mining is conducted using two or more vertical wells (Figure 13) or a combination of one horizontal (or inclined) and one or more vertical wells (Figure 14). During solution mining, the underground in situ stresses are disturbed, causing induced stresses around the wellbore and cavern. These induced stresses, coupled with the mechanical properties of the bedded salts and surrounding rocks, may lead to geomechanical issues such as wellbore or cavern instability.

Well casing failure and cement integrity are the primary challenges in maintaining wellbore integrity. In the context of hydrogen storage, ref. [218] highlighted the corrosive effects of hydrogen on casing and cementing corrosion. Another problem described in [219] is flow-induced vibration within the leaching tubing, which leads to bending and damage of tubing. Additionally, salt deformation poses another risk, potentially causing casing collapse after drilling. Temperature differences between the top and bottom of the cavern create differential creep rates and stresses, which can compromise casing and cement integrity by cracking the cement and creating pathways for fluid leakage [220].

Several methods have been developed to assess the wellbore integrity in bedded salt caverns. Factors affecting wellbore leakage were investigated in [221] and tightness tests, such as the Nitrogen Leak Test and Fuel-Oil Leak Test, were conducted.

These tests showed that an effective well design can prevent leakage through the cemented well. Another study [222] categorized tightness tests into two groups: those assessing the integrity of the wellhead and open hole and those evaluating cavern integrity post-completion before gas injection. The studies in [223,224] also examined drilling challenges in salt formations and their impact on well casing failure. Additionally, an ultrasonic cased-hole imager was proposed in [225] to evaluate cement quality. An ultrasonic cased-hole imager was proposed to evaluate the cement quality in [225]. The study in [226] concluded the presence of sediments in the cavern and contact of caverns with external aquifers as significant contributors to leakages. The effect of borehole spacing in solution mining on cavern geometry was modeled in [227], showing that roof displacement is primarily influenced by changes in borehole spacing.

Once the well is drilled into the salt formation, the leaching process begins. According to [155], three main factors control leaching for producing large symmetric caverns: alternating injection and withdrawal, a high flow rate, and the use of an oil blanket. Both single-well [214,228] and two-well [206,217] leaching methods have been investigated. A comparison between these methods, conducted in [229], revealed that caverns using the single-well method are axisymmetric with circular cross-sections, while those produced using the two-well method are non-axisymmetric with ovular cross-sections. The experimental work in [230] showed that freshwater injection rates significantly affect salt dissolution rates and pipe settings within caverns. Furthermore, ref. [231] developed a technique to control fluid motion, brine distribution, wall dissolution rate, and pipe settings within caverns.

A multi-step horizontal leaching method was proposed in [215], while ref. [232] analyzed the influence of multi-stage leaching, including factors like water injection rate, tubing withdrawal distance, and existing air cushions, on the shape and ultimately the stability of caverns. In the study in [214], the authors outlined four stages of cavern leaching: sump leaching, leaching of the main chamber, leaching of the cavern dome, and neck leaching. The authors also proposed creating niches in well walls to expedite leaching and achieve desired cavern shapes. With respect to multi-step retreating (MSR), ref. [212] conducted experiments to determine fluid velocity distribution and cavern shapes in simulated horizontal caverns.

Several studies have focused on predicting cavern stability during leaching. For instance, ref. [233] applied catastrophe theory to understand interbed failure during solution mining. In another model based on the mechanical properties of rock salt, ref. [19] developed a simulator to model cavern construction and integrity. The authors of [215] developed a mathematical model for multi-step horizontal cavern construction, incorporating factors such as dissolution rate, brine concentration, and accumulation of impurities and their distribution in the cavern. The stability of caverns as a function of cavern height was analyzed in this study. Another mathematical model developed in [234] investigated leaching in salt formation with high amounts of impurities, coupling cavern shape, brine concentration, and brine velocity. Another mathematical model was proposed in [235] to simulate the repair of the irregular salt caverns using a re-leaching method. A multiphysics coupled model based on heat transfer during horizontal cavern leaching was developed in [236], while coupled thermo–hydro-mechanical–chemical processes were described for solution mining in [237]. In addition, the study in [238] investigated various factors affecting deformation in brine-filled salt caverns after solution mining, including salt creep, brine thermal expansion, brine penetration, salt dissolution, and cavern compressibility. It is crucial to monitor changes in pressure due to these factors to ensure cavern stability.

Although most of the studies on solution mining are considering salt beds, a few studies are available on solution mining in potash and trona beds. The authors of [239] provided an overview of the geology of solution mining of potash in Saskatchewan, Canada. A feasibility study for horizontal solution mining was performed in [240] for trona beds in Wyoming, United States.

3.3.3. Cyclic Loading

Cyclic loading, characterized by periodic injection and withdrawal of gas, imposes significant stress on salt formations, often resulting in cavern instability. The main forms of damage caused by cyclic loading are fatigue, creep deformation, crack formation, and wellbore instability.

Fatigue, defined as the deterioration of material strength due to repeated stress, occurs in two forms: cyclic fatigue from repeated loading cycles and static fatigue (or creep rupture) from prolonged stress [241]. Both fatigue modes can happen in salt cavern gas storage because the rock salt is subjected to both cyclic gas storage and an extended storage period.

Cyclic loading experiments conducted in [242] demonstrated a logarithmic relationship between fatigue life and gas pressure. Fatigue models in [243,244] identified two phases of damage: rapid initial accumulation followed by a stable phase. A microstructural study was performed in [245] to record the differences in microstructures during cyclic loading, using fatigue and NMR tests. The damage evolution of rock salt showed three stages of damage, including minor damage in the first two stages and rapid damage in the third stage when the rock salt is near failure. Fatigue tests were also conducted in [246] to determine the effect of high-stress intervals during cyclic loading on the rock around the salt cavern. A mixture of shear and tensile cracks was observed in the rock samples in addition to intra-grain and inter-grain cracks in salt crystals. The main factors affecting rock fatigue behavior are maximum applied stress, amplitude, loading frequency and waveform, and temperature, based on experimental works in [247,248].

The creep behavior of salt results in cavern volume shrinkage under constant gas pressures. A study in [249] investigated the cavern creep shrinkage under constant gas pressures for a sphere- or cylindrical-shaped cavern. The authors derived equations for small and large deformations of cavern shrinkage. The behavior of salt caverns under both cyclic loading and average cycle pressure was studied in [250]. The results indicated a faster volume loss under cyclic pressures than under average cycle pressure. The authors of [173] investigated the effect of cyclic loading on rock deformations in bedded salts. The results showed a brittle deformation of gypsum and an elastoplastic deformation of rock salt. The study in [251] investigated the deformations of mudstone interbeds in salt caverns, demonstrating significant differences in the creep behavior of salt and mudstone under highly effective stress conditions. Using a geomechanical model, ref. [252] studied the deformation around a cavern. Large displacements around cavern walls were simulated in [253] using FLAC3D software (Version 5.0), showing significant deformations after 20 years of cavern creep. Another numerical simulation study of horizontal salt caverns in two mining layers showed larger displacement and volume shrinkage compared to caverns in single mining layers [155].

Tensile failure is the primary cause of fractures in bedded salt caverns. According to [254], tensile fractures can occur if cavern pressure exceeds the operating limit. Another study of rock salt with mudstone interbeds in [251] concluded that tensile failure in mudstone interlayers could compromise cavern integrity. According to the study in [255], the risk of tensile failure in bedded salt rock caverns increases with higher extraction rates and repeated hydrogen storage cycles.

Wellbore instability is another challenge associated with cyclic loading. Gas production rates play a crucial role in cavern and wellbore stability. The study in [256] proposed optimal gas production rate methods to avoid issues such as cavern volume shrinkage, tubing erosion and corrosion, hydrate and condensate formation at the wellhead, and stability concerns. The effect of cyclic internal pressure and cavern volume shrinkage on cement sheath integrity was also studied through experimental work and numerical simulations in [257]. The results revealed an increased plastic strain in the cement sheath under cyclic loading. Cavern volume shrinkage can lead to cement debonding and gas leakage. A coupled analysis of cement sheath fatigue damage and rock salt creep was suggested for bedded salt caverns in [258]. The results indicated that the creep rate and volume shrinkage in bedded salts were lower than in pure rock salts, suggesting better wellbore integrity in such formations.

It is important to note that pressure and temperature are key factors in three different damage types. The study in [259] found that creep rates increase significantly with rising temperatures. Higher internal pressure combined with shorter gas production times resulted in less cavern volume reduction in specific cases. The study in [260] investigated the influence of minimum, maximum, and average internal pressure on the stability of caverns during hydrogen storage via a numerical calculation model. An increase in average internal pressure caused an increase in tightness evaluation indexes such as leak range, hydrogen pore pressure distribution, and overall hydrogen leakage to the rocks in the vicinity of the cavern. A coupled thermo-mechanical model was proposed in [261] to investigate the effects of temperature and pressure changes during cyclic loading on cavern creep variations. The study in [262] coupled both pressure and temperature to understand their effect on tensile failure and ultimately cavern stability. Thermally induced tensile fractures due to high-frequency cycling were also seen in [263] through experiments and modeling. An integrated 1D and 3D geomechanical model was proposed in [264] to indicate the optimal internal pressure of caverns during hydrogen storage.

Monitoring plays a critical role in observing changes in salt caverns, as they are inaccessible directly. Two indirect monitoring methods were proposed in [265], utilizing cavern pressure changes and liquid flow rate in open hole well sections. Several monitoring techniques, including geophysical tools and experiments, were described in [266]. Sonar survey, well pressure and temperature logging, cavern bottom sounding, micro-seismic monitoring, and subsidence survey as various methods of cavern monitoring were discussed in [267]. During integrity tests, factors such as thermal disequilibrium, brine permeation through cavern walls, additional salt dissolution, and transient creep must be considered to avoid misinterpretation of test results [268].

3.3.4. Abandonment

Salt caverns typically have a lifespan of 30–50 years, as suggested in [269]. The abandonment process commonly involves filling the cavern with brine and sealing the well with cement [270]. Pressure build-up following abandonment may cause fractures around the cavern and brine migrations. Key factors affecting pressure build-up include cavern compressibility, creep, equilibrium state, and brine thermal expansion [271]. The study in [272] proposed a model to forecast the behavior of abandoned salt caverns considering short-, medium-, and long-term implications. Another study in [146] assessed the suitability of abandoned caverns in China for hydrocarbon energy storage, aiming to mitigate safety and environmental risks. The stability of abandoned horizontal caverns for gas storage was studied in [273]. An 18-month test on a deep brine-filled cavern in [270] found that equilibrium pressure was significantly lower than geostatic pressure, reducing the risk of formation fracturing.

Surface subsidence is another issue that may occur when a salt cavern is abandoned. According to [274], subsidence occurs when cavern pressure decreases below initial stress levels, causing volume shrinkage. Salt creep is a significant contributor to subsidence and displacement in this scenario. However, a study suggested that certain conditions, such as filling caverns with alkali waste, could minimize subsidence risks [269].

4. Conclusions and Future Research

Future energy needs, such as cleaner energy production, secure energy storage, CO₂ emission reduction, and managing demand fluctuations, require expanded options for energy storage. Historically, salt caverns in halite formations have been widely considered for this purpose due to advantages like effective gas containment, self-healing properties, and high deliverability/injectivity rates. Yet halite is not the only evaporite mineral suitable for cavern storage. Other evaporite minerals, such as potash and trona, exhibit comparable behavior. All three—halite, potash, and trona—can be extracted through solution mining, which leaves a cavern for storage purposes. From a geomechanical perspective, these minerals display creep behavior; although, this investigation revealed some differences in the creep behavior among the three. Each mineral is deposited in bedded evaporite form. Despite their similarities, potash and trona deposits have primarily been considered for mineral extraction. The findings of this research suggest an additional advantage for these bedded evaporites—beyond their commercial value—by highlighting their potential for cavern storage, aligning with the growing need for more storage capacity.

Thus, this review paper underscores the importance of bedded evaporites not previously considered for cavern construction and subsequent storage, specifically potash and trona. It provides a comprehensive overview of evaporites, including their origin, depositional environment, and constituent minerals—factors that influence the physical and mechanical properties of bedded evaporites. To illustrate the potential for cavern development and storage, we present a list of bedded salts, including the number of caverns and their usage in the U.S. and Canada. This paper also reviews key rock properties, particularly those affecting the geomechanical behavior of caverns. Finally, a detailed review of the geomechanics of bedded salts through various stages of design, solution mining, cyclic loading, and abandonment is presented. Additional research on the following topics could further enhance our understanding of evaporite formations and unveil new opportunities for underground gas storage:

• Salt caverns have been used for storage purposes due to their unique properties. However, exploring additional evaporite beds, such as potash and trona, could expand opportunities for energy storage.
• The mineralogy of salt, potash, and trona caverns plays an important role in their physical and mechanical behavior. While the mineralogy of these beds is well understood, more studies focusing on physical and mechanical properties are necessary for broader applications.
• It is important to understand how different physical and mechanical properties affect solution mining and the stability of caverns in different evaporite beds. Reliable data on density, porosity, and permeability in potash and trona beds are currently limited and should be prioritized.
• Developing innovative design approaches compatible with the nature of bedded salts will improve storage functionality and practicality.
• Geochemical aspects, alongside geomechanical factors, are critical for cavern stability and must be considered in the design, operation, and abandonment phases.
• Although this study focused on three evaporite deposit types, future research should address how the characteristics of stored materials influence evaporite properties.
• Investigating geochemical reactions and microbial effects on cavern integrity in different bedded evaporites is recommended, considering the specific mineralogy, type of stored gas, and in situ microbial communities.
• Coupling these geochemical insights with geomechanical consideration will offer a broader perspective on cavern integrity in various bedded evaporites, which is an essential gap to be addressed in future research.