Thermochemical Energy Storage Based on Salt Hydrates: A Comprehensive Review

Abstract

Thermal energy storage technologies are essential for balancing energy demand and supply. There are three main types: sensible heat, latent heat, and thermochemical energy storage. Among them, thermochemical energy storage offers the highest energy density (1–3 GJ/m³) and long-term storage capability. Salt hydrates have attracted attention as energy storage materials due to their low cost, wide availability, and operating temperatures being well-suited for residential and low-temperature applications. This review focuses on the use of salt hydrates in sorption-based thermochemical energy storage systems. It summarizes the current state of knowledge, including screening studies of various salt hydrates, their thermodynamic and operational limitations, advantages, and performance in composite materials. This review also covers recent projects and common reactor designs used in TCES applications. Based on the literature analysis, the most promising salt hydrates for sorption-based TCES systems include SrCl₂, SrBr₂, K₂CO₃, MgSO₄, MgCl₂, and CaCl₂. Despite the high theoretical energy density of many salt hydrates, future work should focus on experimental studies in large-scale reactor systems to better evaluate the practical discharge behavior of the energy storage system beyond theoretical hydration enthalpies or small-scale thermal analyses.

1. Introduction

The world is currently facing the challenge of growing energy demand, driven by population growth, urbanization, and technological development. Energy demand growth in the European Union (EU) has been increasing since 1994 and achieved 37,771 PJ in 2022 [1] (5744 MWh/capita [2]). Consequently, primary energy production in the EU in 2022 accounted for 23,566 PJ, marking a 5.9% decrease compared to the previous year, primarily attributable to a decline in energy production based on natural gas [1]. It is noteworthy that in the EU, there has been a significant decline (by approximately 57%) in energy production derived from solid fossil fuels since the 1990s. Concurrently, energy derived from renewable sources and biofuels has increased substantially (by 251%, reaching 10,453 PJ in 2022) [1].

The generation of energy from fossil fuels is linked to greenhouse gas emissions (GHGs) and resource depletion, and, in general, is associated with environmental pollution [3]. Europe is responsible for approximately 10.5% of global CO₂ emissions, due to the consumption of fossil fuels. In 2022, the total CO₂ emissions in the EU were approximately 3593 Mt, with 33% of these attributable to the generation of electricity and heat [2]. However, it is encouraging that, due to the increase in the share of renewable energy sources (24.5% in 2023 [4]), CO₂ emissions (from fuel combustion) decreased by 21% between 2000 and 2022. Moreover, the recently amended Renewable Energy Directive has augmented the binding target from 32% to a minimum 42.5% share of renewables in EU energy consumption, with the objective of attaining 45% [4]. According to the European Environmental Agency, achieving the target will necessitate a substantial Compound Annual Growth Rate (CAGR) of 8% in the EU’s share of renewables by 2030, more than double the rate observed over the past decade [4].

Despite the considerable growth in the use of renewable energy sources (RESs), these technologies face significant limitations due to their cyclical nature and limited operational flexibility. Wind energy (17.6% share in 2023) and solar energy (9.1% share in 2023), as two of the most prevalent forms of RES [4], demonstrate considerable variability in production, contingent on weather conditions. This variability complicates the balancing of electricity supply and demand. The inability to make precise forecasts and the irregularity of generation necessitate the implementation of capacity reserves and flexible grid management systems [5]. Moreover, the limited capacity to adjust generation to dynamically changing demand increases the risk of destabilizing the power grid, particularly in scenarios with a high share of RESs. Consequently, the development of efficient energy storage systems that can compensate for the variability of RES production and enhance the reliability of energy supply has become a critical component of the energy transition.

Energy storage technologies play a vital role in current energy systems, enabling functions such as load balancing, peak shaving, and the integration of renewable energy sources into the grid. Various storage methods have been explored, including mechanical (e.g., pumped hydro storage, flywheels, compressed air energy storage) [6], electrochemical (e.g., batteries and fuel cells) [7], and thermal energy storage [8]. Among these, thermal energy storage (TES) presents a promising solution, particularly for applications involving heat management and renewable energy utilization.

Thermal Energy Storage Technologies

TES is characterized by its ability to store thermal energy over a wide range of temperatures, from sub-zero temperatures to over 600 °C (high-temperature TES technologies) [9]. Typically, in TES systems, heat is transferred from one medium (called a heat transfer fluid, HTF) to another, which is an energy storage material. The process of raising the temperature of the energy storage material is called charging. During discharge, the thermal energy is transferred to the HTF and then used for a specific purpose. Preferably, concentrated solar energy or another form of renewable energy is used for the charging.

There are three main forms of heat storage: sensible heat, latent heat (phase transition energy), or chemical energy through thermochemical reactions (Figure 1). Sensible heat is the most elementary method of heat storage, which involves the storage of thermal energy by heating or cooling a liquid or solid storage medium (without changing phase). The capacity of sensible heat storage is directly proportional to the temperature change of the material and is related to its specific heat capacity (as shown by Equation (1)).

$$\mathrm{Q}=\mathrm{m}{\int }_{\mathrm{T}1}^{\mathrm{T}2}{\mathrm{C}}_{\mathrm{p}}\left(\mathrm{T}\right)\mathrm{d}\mathrm{T}$$

(1)

Q—stored energy (J), m—mass of the material (kg), C_p(T)—specific heat capacity as a function of temperature, (J/kg·°C), T₁, T₂—initial and final temperature of the material, respectively (°C).

Sensible heat storage (SHS) materials should have a high specific heat capacity and a high density, resulting in a high volumetric energy density (kJ/m³). This allows for smaller and simpler construction of storage tanks. Typical materials include water, natural and synthetic oils, sand, bricks, and concrete [10,11]. Their properties are shown in

Table 1. Examples of materials used in thermal energy storage systems.

Material	Heat Capacity, J/kg·°C	Average Density, kg/m3
Sensible heat storage (SHS) [8,12,14]
Alumina (Al2O3)	1180	2240
Aluminum	896	2707
Brick	840	1600
Cast iron	600	7800
Cast steel	600	7800
Copper	385	8954
Engine oil	1880	888
Glass	837	2710
Granite	892	2750
Graphite	1424	2250
Limestone	741	2500
Reinforced concrete	1500	2500
Rock	879–960	2560
Sand	830	1602
Silicone oil	2100	900
Sunflower oil	2969	753
Therminol 55	2400	-
Water	4190	1000
Latent heat storage (LHS) [15,16,17,18,19,20,21,22,23,24]
	Melting temperature, °C	Latent heat, kJ/kg
Organic compounds
Tetradecane	5.9	258
Hexadecane	18	211–236
Parraffin Rubitherm RT60	58–60	214
Parafin	64	210
Lauric acid	41–44	176–211
Myristic acid	49–54	175–204
Stearic acid	65–71	155–210
Palmitic acid	54–64	188–201
Paraffin RT110 (Rubitherm)	112	285
Erythritol	118–120	340
Salicylic acid	159	199
Urea	134	250
D-mannitol	167	326
Salt hydrates
CaCl2·6H2O	29	191
Na2SO4·10H2O	32	254
Na2S2O3·5H2O	48	209
Sodium acetate trihydrate	58	266
80% Mg(NO3)2·6H2O, 20% MgCl2·6H2O	60	150
Mg(NO3)2·6H2O	89	167–175
AlK(SO4)2·12H2O	91	269
MgCl2·6H2O	117	169
Salt mixtures and eutectics
Hitec XL (48 CN, 45% PN, 7% SN)	120	55
30% LN,60% PN, 10% CN	132	-
67% PN, 33% LN	133	170
Hitec (7% SN, 53% PN, 30% NaNO2)	142	80
55.4% LN, 4.5% SN, 40.1% KCl	160	266
57%LN, 43%SN	193	248
Solar Salt (60% SN/40% PN)	220–223	115
PN/KCl	320	74
56% KCl, 44% LiCl	348	170
60% MgCl2, 20.4% KCl, 19.6% NaCl	380	400
LiNaK Carbonate: (32.1% LC, 33.4% SC, 34.5% PC)	400	276
63% LiF, 37% LiCl	485	403
Metals and their alloys
Pb	328	23
48%Mg, 52%Zn	340	180
59% Al, 33% Mg, 6% Zn	443	310
66% Al, 34% Mg	450	310
20% Al, 80% Si	585	460
Al	660	397
49% Zn, 45% Cu, 6%Mg	703	176
Cu	1083	193

LC—lithium carbonate, LN—lithium nitrate, PN—potassium nitrate, CN—calcium nitrate, SN—sodium nitrate, SC—sodium carbonate, PC—potassium carbonate.

. Water is the most commonly used (up to 100 °C) as sensible heat storage material due to its low cost, availability, non-toxicity, and high specific heat capacity (C_p = 4184 J/kg·°C⁻¹ at 20 °C). Alternative materials such as oils, molten salts, and metal alloys are used primarily in higher temperature systems [12]. Sensible heat storage represents the most advanced technology, despite its limited storage capacity [13]. However, a large number of low-cost energy storage materials are available (Table 1).

In contrast to sensible energy storage, latent heat storage (LHS) systems utilize phase transition energy. During the charging process, the energy storage material undergoes a phase transition, typically from a solid to a liquid state. Subsequently, during the discharging, the material solidifies and releases energy called latent heat (or, more specifically, heat of fusion, when the process is melting/freezing). This energy can be used for specific applications [25,26,27].

Phase transitions can be classified as solid–liquid, liquid–gas, or solid–gas. However, phase change materials (PCMs) with solid–liquid phase transitions are the most widely used due to their advantageous properties, which include a wide operating temperature range suitable for a variety of applications, high storage capacity, and limited volume change during phase transition [25,28]. Furthermore, PCMs have a high energy density and the ability to maintain a constant temperature during phase transitions, making them attractive for temperature control and energy efficiency applications. The maximum storage capacity of latent heat storage can be described as follows:

$$\mathrm{Q}={\int }_{{\mathrm{T}}_1}^{{\mathrm{T}}_{\mathrm{m}}}\mathrm{m}{\mathrm{C}}_{\mathrm{p}1}(\mathrm{T})\mathrm{d}\mathrm{T}+\mathrm{m}{\mathrm{H}}_{\mathrm{f}}+{\int }_{{\mathrm{T}}_{\mathrm{m}}}^{{\mathrm{T}}_2}\mathrm{m}{\mathrm{C}}_{\mathrm{p}2}\left(\mathrm{T}\right)\mathrm{d}\mathrm{T}$$

(2)

Q—stored energy (J), m—mass of PCM (kg), C_p1, C_p2—specific heat capacity at temperature range T₁ to T_m and T_m to T₂, respectively (J/kg·°C), T₁—initial temperature (K), H_f—heat of fusion (J/kg), T_m—melting temperature (°C), T₂—final temperature of the PCM (°C); where T₂ > T₁ and T_m > T₁.

When the temperature of a material rises but does not undergo a phase transition (H_f = 0), the amount of energy stored corresponds to sensible heat.

Due to the wide range of melting temperatures exhibited by PCMs, their applications are diverse. The classification of PCMs based on their melting temperatures (T_m) can be categorized into three distinct groups: low temperature (T_m < 100 °C), medium temperature (T_m = 100–300 °C), and high temperature PCMs (T_m > 300 °C). Organic substances and their mixtures, including higher n-alkanes (e.g., n-dodecane, tetradecane, docosane), paraffin waxes, fatty acids (e.g., stearic, lauric and palmitic acid), sugar alcohols (e.g., erythritol, mannitol, xylitol), esters, and hydrates of inorganic salts (e.g., magnesium nitrate hexahydrate, sodium acetate trihydrate, calcium chloride dihydrate), are primarily utilized as low melting PCMs [17,29,30]. More information on the hydrated salts is given in Section 2.

In light of the selection of optimal parameters, a single substance is rarely used, but mixtures of them are. The eutectic mixtures in question (i.e., mixtures with a lower melting point than the individual components) consist of suitable salts that provide the desired melting point and the high heat of fusion, as well as good thermal conductivity. PCMs with melting points above 120 °C are mainly based on mixtures of inorganic salts. Mixtures of nitrates, nitrites, hydroxides, and halides of Li, Na, K, Mn, Mg, and Zn are commonly used for medium temperature PCMs [31]. PCMs with melting points above 400 °C are mainly mixtures of lithium and beryllium halides and carbonates (e.g., NaCl, KCl, LiCl, Li₂CO₃, K₂CO₃, MgCl₂) [16,29]. For materials with melting points exceeding 500 °C, the utilization of metal alloys, such as Zn-Mg, Zn-Cu-Mg, Mg-Cu, and Cu-Si-Mg, becomes prevalent [32,33]. Common inorganic salt mixtures include, but are not limited to, Solar Salt, Hitec^®, HitecXL^®, LiNaK Carbonate, or Sandia Mix. Their compositions and basic properties are shown in Table 1.

PCMs have the capacity to store and release substantial amounts of latent heat during phase transition, thereby enabling compact and efficient thermal energy storage. Due to their high energy density, a lower amount of material is required compared to sensible heat storage systems, resulting in space and weight savings. Moreover, in contrast to sensible heat storage materials, PCMs absorb and release heat at nearly constant temperatures, ensuring stable thermal management. Despite these advantages, PCMs also have some challenges.

The disadvantages of PCMs include the high cost of the components used, their toxicity, and hygroscopicity. High toxicity is a feature of, in particular, fluoride, barium, or manganese salts. High hygroscopicity (or even deliquescence) is mainly associated with lithium, magnesium, or calcium salts. Additionally, PCMs exhibit a low thermal conductivity (0.2–0.7 W/mK), which hinders heat transfer between the HTF and the energy storage [34]. Furthermore, some compounds, such as sugar alcohols or inorganic salt hydrates, exhibit phenomena of supercooling, incongruent melting, and phase separation [35,36]. Supercooling is defined as the process of lowering the temperature of a product below its usual freezing point without causing a phase change. In addition, some of the water in hydrates can be permanently lost once a certain temperature is exceeded, resulting in a change in the characteristics of the PCM. To address these limitations, various design modifications, such as different constructions of heat exchangers, circular fins, embedded high-conductivity structures [37,38,39,40,41,42,43], various additives [44,45,46,47,48], and PCM composites [49,50], have been used to increase the thermal conductivity of PCMs and to reduce supercooling and phase separation [51,52].

The literature on PCMs is abundant; for example, the Science Direct database for the search term “phase change materials” retrieved more than 255,000 research papers published in 2024. A similar search that included the term “inorganic salts” yielded more than 21,000 research papers published in 2024. A comprehensive review of the available PCM mixtures, their physicochemical properties, and the challenges associated with their use in energy storage applications can be found in a significant number of publications [14,16,17,18,22,53,54,55,56,57,58,59,60]. Many of the articles also consider the methodology for selecting appropriate PCMs depending on the application, for example, in buildings and heat domestic applications [61,62,63], for delaying ice/frost formation [64], in general for thermal energy storage [21,65,66,67], and the selection method based on the Rényi entropy [68]. Nevertheless, the most important parameters considered in the selection of an appropriate PCM are melting point, latent heat, specific heat, thermal conductivity, thermal stability, and toxicity. In general, the phase change temperature of the PCM must correspond to the operating temperature of a particular application.

As previously mentioned, in addition to energy storage in the form of sensible and latent heat (using the PCMs), the third option is thermochemical energy storage (TCES). TCES systems are based on reversible chemical reactions to store and release thermal energy, thereby providing high energy density and the potential for long-term energy storage (Figure 2). Generally, energy is stored during an endothermic chemical reaction and released during an exothermic reverse reaction. A comparison of TCES with sensible and latent heat storage reveals significant advantages and considerable interest in the former. TCES has been demonstrated to exhibit an energy storage density that is approximately from 8 to 10 times higher than that of SHS and approximately 2 times higher than that of LHS [69,70,71,72]. The stability of TCES is ensured by storing energy in the form of chemical bonds, which prevents degradation from chemical reactions. These characteristics position TCES as a key area of research and practical application in the field of long-term heat storage [69].

Reversible thermochemical reactions relevant to long-term energy storage can be categorized into four distinct classifications based on the state of the materials involved in these reactions. These categories are gas–gas reversible reactions, liquid–gas reversible reactions, liquid–liquid reversible reactions, and solid–gas reversible reactions [74]. However, solid–gas reversible reactions have attracted significant interest due to their wide range of transition temperatures and the ability of the reactants to separate independently into their respective components. For example, hydrates of inorganic salts have gained more attention, especially in domestic heating and building applications [63,75,76].

In addition to TCESs based on salt hydrates, other systems use high-temperature reversible solid–gas reactions to efficiently store and release heat [77]. Metal hydrides, such as MgH₂ or LaNi₅, offer exceptionally high energy densities through reversible absorption and desorption of hydrogen, making them promising for both thermal and chemical energy storage. Carbonate-based systems, including CaCO₃, SrCO₃, and MgCO₃, rely on the endothermic decomposition of carbonates into metal oxides (CaO, SrO, and MgO, respectively) and CO₂ at elevated temperatures, with recarbonation enabling heat recovery [78,79]. Metal hydroxides, such as Mg(OH)₂ and Ca(OH)₂, undergo endothermic dehydration at moderate to high temperatures, forming metal oxides (MgO and CaO) and releasing water vapor. This water vapor can later recombine with the oxide, regenerating the hydroxide and releasing heat. Metal peroxides and oxides, such as Na₂O₂, BaO₂, and Co₃O₄, store energy through reversible oxygen release and uptake [80]. These materials are particularly well-suited for high-temperature industrial applications, benefiting from abundant raw materials and long-term stability. However, significant challenges must be addressed to enhance the practical implementation of these systems. These challenges include high reaction temperatures, slow kinetics, particle sintering, and material degradation [72].

Salt hydrates offer several advantages over the previously mentioned TCES systems. Their reaction with water, a non-toxic and non-explosive substance that is widely available, ensures safer and more practical operation. Conversely, metal hydrides require hydrogen, a flammable gas that must be stored under high pressure. Furthermore, salt hydrate reactions typically occur at moderate temperatures, making them compatible with low- and medium-grade heat sources such as solar thermal energy or industrial waste heat. In contrast, the decomposition of carbonates, hydroxides, and peroxides requires significantly higher temperatures (350–1100 °C), which can lead to greater energy losses [80]. Moreover, salt hydrates also allow for efficient energy storage in compact systems, as their hydration-dehydration cycles can be performed at atmospheric pressure with relatively simple infrastructure.

This article focuses on thermochemical energy storage based on reversible hydration and dehydration reactions using inorganic salt hydrates. While numerous reviews have addressed thermal energy storage [8,28,72,74,79,80], many of them tend to cover all three main categories, i.e., SHS, LHS, and TCES. In addition, many of the review papers focus primarily on PCMs [25,30,54,60]. A bibliometric analysis made by Calderón et al. reveals that thermochemical energy storage has the lowest number of publications compared to other thermal energy storage systems [81].

A significant number of review studies that have been published address screening studies, with a primary focus on the identification and listing of potential salt hydrates or other sorption-based TCES materials [82,83,84,85,86,87,88]. Some of these also address materials such as salt hydrates, porous sorbents, and their composites, for example, in the works of Zbair and Bennici [89] Yang et al. [90]. However, most of these reviews concentrate on specific aspects, such as material properties or thermodynamic modeling, and often lack a broader systems-level perspective.

In contrast, the present study provides a comprehensive and up-to-date synthesis of the TCES field, with a continued focus on salt hydrates, while broadening the scope to include reactor designs, practical implementation aspects, and analysis of completed and ongoing demonstration projects. Moreover, we tried to present a holistic approach to TCES materials, including their properties, applied and investigated reactors, discussion of completed and ongoing projects, as well as future recommendations. This integrative and application-oriented approach distinguishes our work from previous reviews and makes it a valuable reference for both academic researchers and professionals involved in the design and deployment of TCES systems.

This paper is organized into six sections. Section 2 presents information on salt hydrates used as PCMs and describes the sorption and hydration/dehydration processes of salt hydrates. Then, their advantages and disadvantages are discussed. Section 3 focuses exclusively on salt hydrates. In addition to summarizing recent findings, a comprehensive synthesis of screening studies has been provided, bringing together the scattered literature into a coherent and accessible resource. Section 4 discusses composites and other enhancement strategies. In this section, the key results from the recent studies of composites (i.e., salts in porous matrices) are summarized in tabular form. Moreover, the scope is expanded beyond material selection by incorporating practical considerations such as reactor designs and completed TCES-related demonstration projects (Section 5). Future recommendations are presented in Section 6, and conclusions in Section 7.

2. An Overview of TCES

Since salt hydrates are used in both TCES and PCM systems, an understanding of the differences between these applications is essential. Section 2.1 examines the hydration/dehydration processes and related limitations of hydrated salts when used as PCMs, while Section 2.2 describes sorption-based TCES, highlighting the key differences between these energy storage systems.

2.1. Salt Hydrates as PCMs

Salt hydrates (or hydrated salt), such as CA·nH₂O, are defined as ionic compounds consisting of a cation (C) and an anion (A), with the additional water molecules chemically bound within the salt crystal lattice. This water is called water of crystallization or water of hydration. Hydrated salt contains water of crystallization (e.g., magnesium sulfate heptahydrate, MgSO₄·7H₂O), while anhydrous salt does not (MgSO₄). The formation of hydrates is contingent upon the ionic structure of the salt, with a limited number exhibiting thermodynamic stability. For instance, the stable forms of calcium chloride comprise monohydrate (CaCl₂·H₂O), dihydrate (CaCl₂·2H₂O), tetrahydrate (CaCl₂·4H₂O), and hexahydrate (CaCl₂·6H₂O) [91].

Salt hydrates (CA∙nH₂O) are attractive candidates for TCES applications due to their high heat storage density, low cost, safety, and environmental acceptability [63]. It should be noted at the outset that the observed phase transition of salt hydrates is not a physical phenomenon observed during the typical melting of anhydrous substances. The phase change of hydrates is a chemical process. Namely, as the temperature of the salt hydrate increases, there is a release of water of crystallization. At this point, the resulting liquid phase (and sometimes even a liquid–solid mixture) is a mixture of partially or completely dissolved salt in the released water [18]. In other words, the hydrated salt chemically decomposes, producing water and the lower hydrate CA·mH₂O (n > m), as shown in reaction 3. In the case of salt hydrates used as LHS, the process is controlled so that complete dehydration (reaction 4) does not usually occur. The amount of water of crystallization that is removed depends on the given substance and on the temperature. The process of dehydration is endothermic, while hydration is exothermic.

$$\mathrm{C}\mathrm{A}·\mathrm{n}{\mathrm{H}}_2\mathrm{O}+\mathrm{H}\mathrm{e}\mathrm{a}\mathrm{t}\to \mathrm{C}\mathrm{A}·\mathrm{m}{\mathrm{H}}_2\mathrm{O}+\left(\mathrm{n}-\mathrm{m}\right){\mathrm{H}}_2\mathrm{O} (\mathrm{dehydration} \mathrm{to} \mathrm{lower} \mathrm{hydrate})$$

(3)

$$\mathrm{C}\mathrm{A}·\mathrm{m}{\mathrm{H}}_2\mathrm{O}+\mathrm{H}\mathrm{e}\mathrm{a}\mathrm{t}\to \mathrm{C}\mathrm{A}+\mathrm{m}{\mathrm{H}}_2\mathrm{O} (\mathrm{further} \mathrm{dehydration} \mathrm{to} \mathrm{anhydrous} \mathrm{salt})$$

(4)

where CA is salt; n and m are the numbers of moles of water of crystallization (H₂O). Interestingly, the number of moles of water of crystallization does not have to be a total number.

It can be seen in the literature that when salt hydrate is used as a typical PCM, the heat release during solidification is often referred to as latent heat (heat of fusion). This simplification probably results from the fact that the observed phenomena during salt dehydration are visually similar to melting [60]. However, the phase transition of salt hydrate (in the LHS system) involves partial dehydration and dissolution, not a clean phase transition. Therefore, energy is partly stored in breaking weak hydrate bonds (enthalpy of hydration reverse reaction 1 or 2 [72]) and partly in dissolving processes (enthalpy of salt dissolution). As discussed below, the “melting” behavior of salt hydrates is more complicated.

Three distinct types of phase transition have been identified during the salt dehydration: congruent, semi-congruent, and incongruent [92,93]. Congruent “melting” occurs when the salt completely dissolves in the separated water, leaving only a liquid phase. Semi-congruent “melting” occurs when a salt hydrate with a lower water content is formed (reaction 3), and there is a separation into a liquid and a solid phase. Incongruent “melting” occurs when the salt does not completely dissolve in the hydrate water at the melting point. Separation into liquid and solid phases also occurs [92].

Incongruent and semi-congruent melting can be related to the phenomenon that, during solidification, the salt that has settled to the bottom (due to its greater density) does not have sufficient contact with the released water of crystallization and does not return to its original state (i.e., initial degree of hydration). This causes a reduction in the amount of heat released and a decrease in the efficiency of the energy storage as the number of charge/discharge cycles increases [91].

In order to prevent incongruent melting, the following potential methods are employed. A thickening agent such as carboxymethyl cellulose (CMC), xanthan gum, or polyvinyl alcohol (PVA) can be added to keep the anhydrous solid component suspended and prevent it from settling [93,94,95,96]. In order to ensure adequate rehydration and complete dissolution of the salt, it may be necessary to introduce additional water [97]. Moreover, the material can be stirred, vibrated, or rotated, and the storage material should be encapsulated so as to prevent the loss of water of hydration [91,98].

Another negative aspect often encountered when using hydrate salts as PCM is the phenomenon of supercooling [36]. It is a phenomenon that occurs when there are insufficient nucleation sites for crystallization (discharging process). Normally, when a liquid reaches its melting point, it begins to crystallize; however, in supercooled conditions, this process is hindered, allowing the liquid to cool below its melting point without solidifying. Consequently, the release of stored heat occurs at a temperature different from what one might expect, which can lead to inefficiencies in thermal management systems. This discrepancy in temperature can create challenges, especially in applications where precise temperature control is critical. To mitigate the risk of supercooling, one effective strategy is to introduce small amounts of a nucleating agent. By providing additional nucleation sites, these agents facilitate the crystallization process, thereby promoting more consistent and predictable heat release at the intended melting point [36,98]. The choice of nucleating agents should be close to the lattice parameter of the target materials. For example, sodium tetraborate (Borax) is generally considered the optimal nucleating agent for sodium sulfate decahydrate, capable of reducing supercooling levels from 15 °C to approximately 2–3 °C [93].

2.2. Sorption-Based TCES System

2.2.1. The Differences Between LHS and TCES Systems

It is important to note the difference between the use of salt hydrate as an LHS medium and its use in TCES. Specifically, when salt hydrate is used as a conventional PCM, the salt is heated or cooled within a closed system during the charging and discharging processes. The water of hydration is not removed from the system. Under the elevated temperature, the given salt hydrate releases only a part of the water of crystallization (weakly bound to the molecule), which leads to only partial dehydration of the salt. The water is released in the liquid state, and, therefore, the dissolution of the salt occurs with the formation of a saturated solution. For example, Na₂SO₄·10H₂O first releases water of crystallization at 32 °C.

In contrast to PCM, in a sorption-based TCES, water in the form of vapor is removed from the salt during the charging process to form an anhydrous or less hydrated salt. This is a solid–gas reaction. In this system, salt is kept separately from the water vapor. Importantly, such a salt, isolated from moisture, can potentially act as an energy store for an indefinite period of time, which is a feature that distinguishes it from other TES systems (Figure 3). During the discharging, water vapor must be added to the system, resulting in the hydration of the salt and the release of thermal energy (enthalpy of hydration). In other words, in the sorption-based TCES systems, the charging process (i.e., the storage of heat) occurs during the desorption of water vapor. Conversely, the discharging process (i.e., the release of heat) occurs during the sorption of water vapor. From a chemical point of view, the released energy is the enthalpy of the hydration reaction (more details are described below). In addition, since the water reacts in the gaseous state, the enthalpy of condensation or evaporation must be taken into account. This means that more energy can be stored than when salt hydrate is used as LHS. The differences between LHS and TCES systems are shown in

Table 2. The differences between using salt hydrate in LHS and sorption-based TCES systems [99].

Feature	Salt Hydrate in LHS System (as PCM)	Salt Hydrate in TCES System
Storage mechanism	Enthalpy of partially dehydration and enthalpy of dissolution and mixing	Enthalpy of dehydration (bond breaking and recombination)
Energy release	During solidification	During hydration (water vapor sorption)
Energy density	Generally lower than TCES	Generally higher, even 45% higher than the corresponding LHS [99]
Heat loss	Prone to self-discharge via heat dissipation	Negligible heat loss during storage

.

2.2.2. Hydration and Dehydration

Hydration temperature refers to the minimal temperature at which the sorption reaction takes place under specific vapor pressure and system conditions. The hydration temperature is a theoretical upper limit for the discharge temperature of the TCES system at a given hydrate and specific water vapor pressure. The dehydration temperature, on the other hand, is the minimum temperature that must be applied (at a given water vapor pressure) to remove the hydration water from the salt, i.e., to charge the system. The relationship to water vapor pressure and temperature is discussed below.

Both the hydration temperature and the dehydration temperature are affected by the properties of the material, making them particularly important for practical applications since the hydration temperature limits the output temperature of the system (also referred to as the discharge temperature). Therefore, two points are essential for the discharge of the TCES system: the appropriate water vapor pressure and a generally low temperature (depending on the specific salt). Conversely, charging requires a higher temperature than discharging, and the process can occur in the absence of vapor pressure [100].

In the TCES systems, gas (water vapor)–solid reactions play a pivotal role in the hydration and dehydration processes. Therefore, the sorption process needs to be discussed.

Sorption processes can be divided into absorption and adsorption. Adsorption is the process by which molecules (adsorbate) attach to the surface of a solid material (adsorbent) through physical or chemical interactions. In TCES, adsorption enables heat storage and release by reversibly binding water vapor (or other gases) to porous materials such as zeolites, silica gel, or MOFs (metal–organic frameworks) [71,101]. This process can occur via physisorption, involving weak van der Waals forces, or chemisorption, where stronger chemical bonds are formed, affecting energy density and operating conditions. Physical adsorption allows rapid and reversible cycles but has a lower energy capacity. In contrast, chemical adsorption (chemisorption) forms strong bonds that allow higher energy storage, but often require higher temperatures during the charging phase [8,101].

In adsorption systems, the energy storage density depends on the vaporization enthalpy of the adsorbate, the strength of the binding forces between the adsorbate and the adsorbent, and the temperature difference between the charging and discharging phases [8].

Absorption is defined as the process by which a substance (absorbate) is taken up by the bulk of another material, known as the absorbent. The energy storage density of an absorption system is determined by the molar mass of the absorbate, the affinity and physical state of the absorbent, and the properties of the absorbate.

The maximum amount of energy stored in thermochemical systems Q_TCES (using salt hydrates) can be described as follows:

$${\mathrm{Q}}_{\mathrm{T}\mathrm{C}\mathrm{E}\mathrm{S}}={\mathrm{n}}_{\mathrm{R}}{\mathsf{\Delta }\mathrm{H}}_{\mathrm{R}}$$

(5)

Here, n_R is the number of moles of the reactant involved in the reaction (e.g., moles of water or salt) and H_R is the enthalpy of the relevant reaction, expressed in kJ/mol.

However, it is important to note that the maximum energy from the enthalpy of hydration is not fully realized due to several factors, including incomplete hydration or dehydration, limited heat and mass transfer, and side reactions. A more thorough discussion of these issues can be found in Section 2.3 (advantages and limitations).

The gas–solid sorption reaction based on the hydration of salt can be described as follows:

$$\mathrm{C}\mathrm{A}\left(\mathrm{s}\mathrm{o}\mathrm{l}\mathrm{i}\mathrm{d}\right)+\mathsf{\nu }{\mathrm{H}}_2\mathrm{O}\left(\mathrm{g}\mathrm{a}\mathrm{s}\right)\leftrightarrow \mathrm{C}\mathrm{A}·\mathsf{\nu }{\mathrm{H}}_2\mathrm{O}\left(\mathrm{s}\mathrm{o}\mathrm{l}\mathrm{i}\mathrm{d}\right)+\mathsf{\Delta }{\mathrm{H}}_{\mathrm{R}}^0$$

(6)

Here, ν is mole of the water; $\mathsf{\Delta }{\mathrm{H}}_{\mathrm{R}}^0$ is standard enthalpy of the reaction, J/mol. CA, in addition to anhydrous salt, can also mean lower hydrate.

Gibbs energy change is 0 and can be described as follows:

$$∆{\mathrm{G}}_{\mathrm{R}}=∆{\mathrm{G}}_{\mathrm{R}}^0+\mathsf{\nu }\mathrm{R}{\mathrm{T}}_{\mathrm{e}}\mathrm{l}\mathrm{n}{\mathrm{K}}_{\mathrm{p}}=0$$

(7)

A monovariant heterogeneous equilibrium system, under the assumption of ideal behavior of water vapor, the equilibrium constant K_p can be described as follows [102]:

$${\mathrm{K}}_{\mathrm{p}}=\left({\displaystyle \frac{{\mathrm{p}}_{\mathrm{e}\mathrm{q}}}{{\mathrm{p}}^0}}\right)$$

(8)

where: p_eq is equilibrium pressure of the gas phase (e.g., water vapor), Pa; p⁰ is standard (reference) pressure, Pa. R is a gas constant (8.314 J/mol·K⁻¹); T_e—temperature in equilibrium, K; ΔS⁰—standard entropy, J/mol·K⁻¹.

Considering that the Gibbs energy change of the reaction is expressed by the following equation:

$$∆{\mathrm{G}}_{\mathrm{R}}^0=∆{\mathrm{H}}_{\mathrm{R}}^0-\mathrm{T}∆{\mathrm{S}}_{\mathrm{R}}^0$$

(9)

The relationship between temperature and vapor pressure can be described by the Clausius-Clapeyron Equation (10) [103]. It can be obtained by combining Equations (7)–(9).

$$\mathrm{l}\mathrm{n}\left({\displaystyle \frac{{\mathrm{p}}_{\mathrm{e}\mathrm{q}}}{{\mathrm{p}}^0}}\right)=-{\displaystyle \frac{\mathsf{\Delta }{\mathrm{H}}_{\mathrm{R}}^0}{\mathsf{\nu }\mathrm{R}{\mathrm{T}}_{\mathrm{e}}}}+{\displaystyle \frac{\mathsf{\Delta }{\mathrm{S}}_{\mathrm{R}}^0}{\mathsf{\nu }\mathrm{R}}}$$

(10)

A saturated solution of an inorganic salt will exhibit an equilibrium vapor pressure in the aqueous phase that is contingent on the surrounding temperature and, to a minor degree, the surrounding pressure. The equilibrium water vapor pressure at varying temperatures (or vice versa) can be calculated using the standard enthalpy and entropy of the hydration reaction.

Equation (10) describes at what temperature a given salt hydrate will adsorb or desorb water for a given water vapor pressure. There is an equilibrium temperature above which salt dehydrates and below which it absorbs water for a given equilibrium water vapor pressure.

Thus, it seems evident that the process of water sorption by salt is influenced by temperature and water vapor pressure. At a given temperature, a salt will absorb moisture from the air if the relative humidity (RH) is greater than the deliquescence relative humidity (DRH). Deliquescence is the process by which a substance absorbs water from the air and eventually becomes an aqueous solution. DRH is defined as the RH at which deliquescence begins to occur.

In cases where the RH is between the equilibrium humidity for the hydration–dehydration process and the DRH, the salt absorbs water vapor, thereby becoming more hydrated. Conversely, when the RH exceeds the DRH, the salt undergoes dissolution into the absorbed water until reaching equilibrium [104].

The DRH can be described as follows [100]:

$$\mathrm{D}\mathrm{R}\mathrm{H}={\displaystyle \frac{\mathrm{p}{\mathrm{H}}_2{\mathrm{O}}_{\mathrm{s}\mathrm{a}\mathrm{l}\mathrm{t}}}{\mathrm{p}{\mathrm{H}}_2{\mathrm{O}}^0}}$$

(11)

pH₂O_salt is the vapor pressure of saturated aqueous solution of the given salt, Pa; pH₂O⁰ is vapor pressure of the pure water, Pa.

The TCES systems should work under DRH to avoid salt dissolution. More information on this issue can be found in the following Section 2.3.

2.3. Advantages of Sorption-Based TCES Systems

Sorption-based TCES salt hydrates present numerous advantages that make them an attractive option for energy storage solutions. They have a high energy density, meaning they can store a significant amount of energy in a relatively small volume, which is crucial for applications where space is limited. For example, calcium hydroxide, Ca(OH)₂, has a maximum theoretical energy density of 3.11 MJ/m³ [105], and MgSO₄·7H₂O has a theoretical storage density of 2.8 GJ/m³ [106]. Generally, TCES materials have an average volumetric energy density of 1.8 GJ/m³, whereas PCMs typically reach about 0.36 GJ/m³, and sensible heat materials about 0.18 GJ/m³ [107]. To illustrate, popular lithium-ion batteries have a capacity of from approximately 1 to 2.5 GJ/m³ (200–300 Wh/kg) [108].

Since that energy is stored in chemical bonds, thermal losses over time are not a concern, enabling long-term storage without degradation. Sensible and latent heat storage systems rely solely on the temperature changes of the storage medium. As a result, they are susceptible to self-discharge due to heat loss.

The applications for TCES materials are diverse, ranging from the capture and use of industrial waste heat—improving overall energy efficiency—to the provision of seasonal heat storage for buildings [75,101,109], helping to balance energy consumption throughout the year. Furthermore, the utilized salt hydrates are typically cost-effective, rendering them economically viable for extensive application. The prices in relation to the energy density of the most common salt hydrates are discussed in Section 3.2.

2.4. Barriers and Limitations of Sorption-Based TCES Systems

Despite their promising advantages, several challenges must be addressed to optimize the performance of sorption-based TCES systems (Figure 4). The kinetics of the hydration reaction can be slow, leading to inefficiencies in energy retrieval [106,110]. The slow kinetics of salt hydration and dehydration in TCES can be caused by several factors related to mass transfer limitations. A major limitation is diffusion resistance, where water vapor must penetrate the material to reach reaction sites [111]. It can be hindered by particle agglomeration, compacted structures, or poor porosity.

In addition, cyclability, defined as the ability of a material to undergo repeated hydration and dehydration cycles without significant loss of performance, is a critical parameter in evaluating the suitability of salt hydrates for TCES. Over multiple cycles, some materials may experience degradation due to factors such as incomplete reversibility and structural changes (discussed below) or loss of water absorption capacity. These changes can result in reduced energy density, slower reaction kinetics, and material instability. The extent of degradation depends on several factors, including temperature, relative humidity, heating and cooling rates, and the presence of additives or supporting matrices (discussed in more detail in Section 4.1).

Another major concern in sorption-based thermochemical systems is low thermal conductivity [36]. Pure salts exhibit poor heat transfer characteristics, resulting in uneven temperature distribution and localized areas of over- or under-heating, which ultimately affect the reaction rate. In addition, particle sintering or structural degradation can reduce the accessible reactive surface area, further limiting water uptake and heat release. Moreover, material degradation can occur due to chemical reactions (e.g., hydrolysis, decomposition at high temperature during dehydration), posing a risk to the long-term reliability of TCES systems.

Deliquescence is also a challenge for sorption-based TCES systems, especially those using highly hygroscopic salt hydrates [112,113]. Deliquescence occurs when a salt absorbs water vapor beyond its crystalline hydration capacity (also known as overhydration) and eventually dissolves to form a solution, which can significantly affect system efficiency and lifetime.

Once the salt is dissolved, it may not be able to recrystallize back to its original form after dehydration, resulting in material degradation, reduced cycle stability, and reduced energy storage capability. In addition, the formation of a liquid phase can lead to clogging or agglomeration within the storage medium, impeding mass and heat transfer. It also increases the risk of corrosion, as the resulting salt solution can aggressively corrode system components, especially metals, requiring the use of special protective coatings or corrosion-resistant materials [114]. In addition, phase separation can occur, causing uneven redistribution of salt and water, resulting in localized performance degradation and ineffective energy storage. To overcome these challenges, operating conditions such as humidity and temperature must be controlled. The salt should operate in the TCES system below its DRH point, i.e., the moisture content of the gas during the discharge process should be lower than the DRH [107]. An alternative approach to mitigate this problem is the use of porous matrices, which is discussed in detail in Section 4.1.

Furthermore, a hydration/dehydration reaction may occur in different steps, resulting in the formation of an intermediate hydrate. As a result, the material does not fully return to its original state, negatively affecting the efficiency of the reaction. In addition, in some cases, the dehydration reaction may not be fully reversible, affecting the cyclability of the system.

In the context of cyclic stability, caking and agglomeration have been identified as prevalent degradation issues that can compromise performance over repeated cycles [87,115]. Agglomeration, defined as the sticking together of particles into larger clusters, is typically caused by moisture exposure, partial hydration, or cycling-induced changes. These changes can reduce surface area and hinder vapor diffusion, which is essential for high performance. Caking is a more severe form of agglomeration, where particles bond into a hard, compact mass through localized deliquescence and recrystallization, leading to loss of material permeability. The occurrence of these phenomena negatively affects heat and mass transfer, reducing the efficiency and stability of the storage system.

Loss of structural integrity can also occur if the salt hydrate melts prior to dehydration, which is a highly undesirable phenomenon. After melting, the material may not revert into the same hydrate form, making the reaction irreversible or inconsistent in subsequent cycles. In a solid state, dehydration occurs through the controlled release of water molecules bound within the crystal lattice. If the material melts first, water may be released in a less predictable manner, disrupting the desired thermochemical reaction profile. Melting can cause phase segregation or changes in particle morphology (as mentioned above, caking or agglomeration), all of which reduce the long-term cyclability and mechanical strength of the material. Eventually, the melting may cause the salt to leak from the energy storage reactor or porous matrix, if used (see Section 4.1 for more details).

Another challenge to maintaining stable system operation is the changing volume (density) of the granules. During the hydration and dehydration cycles, the structural integrity of the reactive grains can undergo significant changes. Dehydration results in an increase in grain density, causing the grains to shrink and, in some cases, collapse or fragment into smaller particles, a process known as pulverization. This changes the permeability and packing of the material bed within the reactor, which directly affects mass and heat transfer performance. Reduced permeability can hinder the flow of air or steam, while uneven packing can lead to temperature gradients and reduced overall efficiency. In addition, repeated swelling during hydration and shrinkage during dehydration can create mechanical stresses within the heat exchanger structure, potentially compromising its mechanical stability and long-term durability.

Considering the above-mentioned issues, it is clear that the cyclic stability of salt hydrates is a critical factor affecting their practical application and long-term performance. Due to the many reversible hydration and dehydration reactions, structural changes (both mechanical and chemical) are observed that can reduce performance over cycles.

Optimizing particle size and modifying material structures through the use of the porous matrix or compacting compositions with binders can help improve reaction kinetics, reduce swelling and volume changes, enabling the production of cyclically stable particles with consistent discharge performance [116]. Impregnating salt hydrates into porous polymer matrices can improve stability by preventing agglomeration and mechanical degradation. These enhancement methods are described in Section 4.

3. Salts Used in Water Vapor Sorption-Based TCES

At the outset, it is worth mentioning that, in addition to salt hydrates, several physical sorption materials are employed in thermochemical energy storage systems, including silica gel, zeolites, and metal–organic frameworks (MOFs) [114,117]. These porous materials are capable of storing heat through the reversible adsorption of water vapor onto their internal surfaces. While these materials generally offer lower energy densities compared to salt hydrates, they compensate for this with faster sorption kinetics, excellent cycling stability, and high resistance to deliquescence. Zeolites and MOFs, in particular, allow for the adjustment of sorption properties due to their well-defined pore structures and chemical versatility, making them attractive for applications requiring rapid and repeated thermal cycling.

Silica gel and zeolites are porous materials commonly used for water vapor adsorption and are also widely applied as desiccants due to their strong moisture uptake capabilities. Silica gel offers fast kinetics and good cycling stability, but they have low energy density and thermal conductivity. Zeolites, with their crystalline structure, provide higher thermal stability and stronger water adsorption, making them suitable for higher-temperature TCES applications, though they may exhibit slower desorption rates. These materials have high regeneration temperatures of 150–300 °C, which makes them unsuitable for solar energy storage. Information on these materials applicable to TCES systems can be found in other publications [88,114,118,119].

The following sections describe the desired properties of the sorption-based TCES materials and collect literature data on the materials used in the sorption systems. The review of materials focuses on salt hydrates because of their widespread use and advantages (discussed previously in Section 2.3).

3.1. Selection of TCES Material

First of all, an ideal salt hydrate-based TCES material should have high energy density to maximize storage capacity while minimizing material volume and maintaining long-term, reliable operation. No less importantly, it must have a well-defined reversible hydration reaction with an appropriate equilibrium temperature that matches the specific applications. The reversibility of the hydration reaction is key to the cyclic operation of the energy storage system [87]. Lack of reaction reversibility or unacceptable energy storage density precludes the use of a given salt as an effective thermochemical storage material.

Stability over multiple cycles is critical to prevent material degradation and irreversible side reactions. Fast hydration and dehydration kinetics improve system responsiveness and reduce charge and discharge times. The material should also have good heat and mass transport properties to minimize diffusion limitations. Resistance to side reactions, such as hydrolysis or decomposition, is essential to maintain consistent performance over multiple cycles. Generally, good thermal conductivity enhances heat transfer and allows adequate heat exchange between the water vapor and the salt so that the appropriate temperature for the heat transfer medium can be achieved.

It is also important that deliquescence does not occur during the hydration process. As mentioned above, deliquescence can lead to irreversible material degradation, loss of structural integrity, and impaired reversibility of the hydration–dehydration cycle. Therefore, operating conditions must remain below the DRH of the salt to maintain its solid-state reactivity. The DRH of a salt determines the humidity range in which it can safely work; salts with low DRH are suitable for dry environments, while those with higher DRH require more humid conditions, which directly affects their suitability for specific TCES applications.

In addition, it is highly desirable that the “melting” point of the salt be significantly higher than its dehydration temperature. If dehydration occurs near or above the “melting” point, partial melting may occur, resulting in sintering, agglomeration, or irreversible morphological changes that reduce the material’s reusability and heat storage efficiency. Maintaining a sufficient temperature differential between dehydration and “melting” ensures thermal stability, structural integrity, and consistent energy performance over multiple cycles.

Additionally, salt hydrate-based TCES material should be non-toxic, relatively inexpensive, and widely available to ensure feasibility for large-scale applications. The economic viability of large-scale implementation is strongly influenced by the cost of materials, with abundant and inexpensive salts, such as magnesium sulfate (MgSO₄) or calcium chloride (CaCl₂), being more attractive than rare or expensive alternatives. Corrosion resistance is another key factor, as previously mentioned, some salt hydrates (Na₂S, CaCl₂) can be corrosive, especially when deliquescent.

The key parameters that must be considered during the selection of TCES material are summarized in

Table 3. Key parameter used for selection of salt hydrate used in TCES [82,100,114,120].

Parameter	Description and Example Criteria
Energy density	Amount of energy stored per unit mass or volume of the salt hydrate. Theoretical energy density is determined by the reaction enthalpy, while practical energy density is influenced by factors such incomplete reactions and system design (open or closed). >500 kWh/m3 (1.8 GJ/m3) for practical deployment. Other filters: >1 GJ/m3 or >2 GJ/m3.
Operating temperature (Hydration and dehydration temperature)	The temperature at which energy can be stored or released (dehydration and hydration temperature). It influences the application of TCES system. Dehydration temperature < 100 °C for solar application and <140 °C for waste heat. Temperature of melting should be higher than dehydration. Temperature lift during hydration should be as high as possible.
Chemical and thermal properties	The material should not decompose (which can happen at higher temperatures, especially during dehydration) or undergo a hydrolysis reaction. RH of air at operating condition should be lower than DRH. Thermal stable to 150 °C. DRH > 60% (at 25 °C). No decomposition reaction.
Cycle stability	The ability of the salt hydrate to maintain its performance over repeated charging and discharging cycles without degradation. It determines the lifetime of the material and costs of stored energy. Stability: >10 stable cycles: hydration/dehydration.
Cost and availability	The cost of the salt hydrate material and the overall system cost are critical for economic viability. Good mechanical strength. Price: <3.5 USD/kg or <1–2 €/MJ.
Safety and environmental considerations	Materials should non-flammable and non-explosive (strong oxidizer are not preferred). Possibly the lowest corrosivity and non-toxicity for safe operation. It is important that no toxic gases (e.g., HCl, H2S) are released during dehydration. Safety: salt can be applied in open and closed system.

DRH—deliquescence relative humidity, RH—relative humidity.

. An optimal balance of these properties can ensure reliable, efficient, and economically viable thermal energy storage.

Different salt hydrates exhibit unique hydration and dehydration behaviors, making them suitable for various temperature ranges and applications. The following, Section 3.2, provides an overview of the most commonly used salts as TCES materials.

3.2. Salt Hydrates Used in Sorption-Based TCES Systems

Many salt hydrates have been studied for TCES applications. The most common materials include strontium chloride (SrCl₂), strontium bromide (SrBr₂), magnesium sulfate (MgSO₄), sodium phosphate (Na₃PO₄), magnesium chloride (MgCl₂), potassium carbonate (K₂CO₃), sodium sulfite (Na₂S), calcium chloride (CaCl₂), lithium chloride (LiCl), and lithium bromide (LiBr) [63,82,87]. There are many scientific papers that have reviewed and screened a large number of salt hydrates. These are discussed below.

Ruby-Jean Clark et al. [82] conducted an experimental screening of salt hydrates for TCES for building heating applications. Selection criteria included energy density > 500 kWh/m³, maximum dehydration temperature of 100 °C, and material cost of USD < 3.5/kg. Additionally, the material should not be toxic, flammable, explosive, corrosive, or melt during dehydration. The salts selected were MgSO₄, SrBr₂, SrCl₂, Na₃PO₄, and MgCl₂. However, SrBr₂ and SrCl₂ have shown the most promise.

Richter et al. carried out an impressive screening of 308 different inorganic salts [84]. The target application was the heat conversion and reintegration of process waste heat up to 300 °C. Salt selection was based on the theoretical and experimental analysis, including reversibility of dehydration reactions (in N₂ atmosphere), hydration tests (at 96 kPa of H₂O), and cycling stability (10 cycles of hydration/dehydration). The first steps of the screening involved a theoretical analysis and thermodynamic calculations. In general, materials that are carcinogenic, deliquescent, or hazardous were rejected. Details of the screening criteria are given in

Table 4. The most prominent papers related to the selection of salt hydrates and their major results.

Selection Criteria	Selected Salts	Application of TCES	Reference
E > 500 kWh/m3, maximum TDH of 100 °C, material cost of <3.5 USD/kg.	MgSO4, SrBr2, SrCl2, Na3PO4, and MgCl2.	Building heating	Ruby-Jean Clark et al. [82]
1 selection: E > 2 GJ/m3, TH > 65 °C at 12 mbar of H2O, TDH < 100 °C at 20 mbar of H2O. Tm > TDH 2 selection: E > 2 GJ/m3, TH > 50 °C at 12 mbar of H2O, TDH < 120 °C at 20 mbar of H2O. Tm > TDH	262 salts considered and 563 reactions. 25 salts preselected. The most promising: Na2S, LiCl, EuCl3 and GdCl3. The best salt is K2CO3.	TCM reactor in the built environment. Energy stored 10 GJ, deliver hot tap water at 65 °C and charged in summer using the solar panels.	Donkers et al. [120]
Toxic, flammable, explosive materials rejected (80 salts). Discharging temperature 60 °C. E > 480 kWh/m3, Upper limit for the charging of 105 °C.	(45 materials preselected and then evaluated by TGA) The most promising: SrBr2, MgSO4, LaCl3	store 80 kW h of heat generated by micro-CHP for household applications, in a storage unit of 1 m	N’tskoupoe et al. [83]
Rejected materials: harmless, deliquescent, having a side reaction or undergo decomposition, having a Teq > 150 °C at 96 kPa of H2O. Experiment of reversibility: TH > 150 °C at 96 kPa of H2O. TDH (at 5 kPa of H2O) < TH (at 96 kPa of H2O). Stability over 10 cycles	From 308 salts, 32 selected to experimental analysis. Only SrBr2 meets all criteria.	Reintegration of process waste heat up to 300 °C.	Richter et al. [84]
E ≥ 1.3 GJ/m3 TDH ≤ 120 °C at 12 mbar of H2O 10 ≥ cycles of hydration/dehydration	Selection from 24 double-salts. (NH4)2Zn(SO4)2·6H2O meets all criteria.	TCES	Kooijman et al. [121]
According to Strunz and Nickel mineral classification. Reversibility of charging and discharging up to 140 °C and 250 °C	Selection of minerals as potential TCES hydrates. 29 mineral specimens selected and experimentally evaluated	TCES	Afflerbach et al. [85]
Toxicity, energy demands, reaction temperature, stability, reaction hysteresis and reversibility. Three case scenarios T > 30 °C and T > 55 °C discharging temperature, and charring temperature: T < 160 °C.	454 salts hydrates and 1073 reactions evaluated. Eight salts and nine reactions (K2CO3 0–1.5, LiCl 0–1, NaI 0–2, NaCH3COO 0–3, (NH4)2Zn(SO4)2 0–6, SrBr2 1–6, CaC2O4 0–1, SrCl2 0–1 and 0–2) met all of the criteria. Including salt stabilization: additional four salts: CaBr2, CaCl2, LiBr, LiCl and ZnBr2.	Domestic heating TCES. Using heat sources available in the built environment.	Mazur et al. [63]
Three different case scenario and six filters including 12 parameters. Among others: E from >1 to >2 GJ/m3, Price: from EUR <0.5 to EUR <2/MJ. Chemical stability from 30 to 150 °C, hydration temperature < 30 °C, dehydration < 150 °C	10 salts experimentally validated: CaCl2, Ca(NO3)2, CaSO4, CuSO4, MgCl2, Mg(NO3)2, MgSO4, SrBr2, Zn(NO3)2, K2CO3. The most preferred: MgSO4 Candidates for closed systems: CaSO4 and CuSO4	TCES	Palacios et al. [100]
Among others, based on E (gravimetric and volumetric). Temperature hysteresis could not be larger than 50 °C. Metal salts more expensive than Li rejected. Unstable compounds eliminated (with distances > 50 meV/atom)	265 hydration reactions were screened. Novel salt: CrF3·9H2O 17 promising reactions. Novel compounds as TCES materials: AlF3·9H2O, CrF3·9H2O, FeBr2·4H2O, Na2Cu(OH)4, NaOH·7H2O, and ZnCl2·3H2O	TCES Three temperature operating ranges: <100 °C, 100–300 °C, >300 °C	Kiyabu et al. [122]

E—energy density, T_H—hydration temperature, T_DH—dehydration temperature, T_eq—equilibrium temperature of hydration/dehydration, T_m—melting temperature.

. Only SrBr₂ met all the criteria. The volumetric storage density for an open SrBr₂/H₂O system was 140 kWh/m³ (i.e., 0.504 GJ/m³), assuming the formation of a monohydrate and a bulk density of 1.9 g/cm³.

N’Tsoukpoe et al. [83] performed an extensive, multi-step screening of 125 salt hydrates. They evaluated the analyzed materials in terms of storage density, dehydration temperature, toxicity, and price, among other aspects. The authors performed theoretical calculations and thermogravimetric analysis (TGA), and hydration/dehydration tests of 45 preselected salts. The target application was a storage system that would meet both heating and domestic hot water needs, with a minimum outlet (discharging) temperature of 60 °C. Finally, SrBr₂, MgSO_4, and LaCl₃ (lanthanum chloride) were selected as the most promising TCES materials.

Donkers et al. reviewed salt hydrates for seasonal heat storage in domestic applications [120]. Remarkably, they collected nearly 600 thermodynamic data points of hydration reactions. They used two filters for the evaluation. Filter 1 included three criteria, i.e., an ideal hydration reaction with a capacity greater than 2 GJ/m³, a hydration temperature of 65 °C (suitable for domestic hot water) or higher, and a dehydration temperature below 100 °C. In this way, four salts were selected: Na₂S, LiCl, EuCl_3, and GdCl₃. However, due to their corrosiveness, toxicity, and the high price of rare metals (Eu, Gd), they were not considered viable for large-scale use.

The second selection [120] included hydration reactions with a capacity greater than 1.3 GJ/m³, a hydration temperature of 50 °C or higher, and a dehydration temperature below 120 °C. This resulted in the list of 25 TCM [120]. Some that seem interesting (not very rare or highly toxic compounds) are iron (II) chloride FeCl₂, CaCl₂, magnesium nitrate Mg(NO₃)₂, potassium aluminum sulfate KAl(SO₄)₂, K₂CO₃, and MgCl₂. Finally, K₂CO₃ was selected as the most promising candidate, despite having the lowest energy density among the salts analyzed. However, considering the dehydration temperature, melting point, deliquescence, and price, K₂CO₃ performs sufficiently well.

Van Essen et al. [123] conducted an analysis of four salts, MgSO₄, Al₂(SO₄)₃, CaCl₂, and MgCl₂, employing thermogravimetric and differential scanning calorimetry (TG-DSC) and a small-scale fixed bed reactor. Among the examined salts, MgCl₂ was identified as the most promising due to its theoretical energy density of 2.8 GJ/m³. However, further water absorption was hindered by the hygroscopic nature of MgCl₂ and CaCl₂, as evidenced by their tendency to form a gel-like substance during hydration experiments. Moreover, the results showed that only a small temperature increase (4 °C) was observed during the hydration of MgSO₄ at low pressure. At atmospheric pressure, MgSO₄ was unable to absorb water (and release heat) under practical conditions.

Pandey et al. performed a screening of salt hydrates and cellulose nanocrystal composites using life cycle assessment (LCA) [86]. They found that MgSO₄, zinc sulfate ZnSO₄, and CaCl₂ were the most preferred because of their low environmental impact, followed by magnesium and strontium chloride and strontium bromide. MgSO₄ was the favored composite material, followed by ZnSO₄, Na₂S, and SrCl₂.

Kooijman et al. [121] carried out experimental research to screen double salts, a relatively new class of TCES compounds. In their work, 24 different double salt sulfate hydrates were synthesized and screened for the desirable properties of the TCES system. The major criteria were as follows: energy density ≥ 1.3 GJ/m³, dehydration temperature < 120 °C, 10 cyclic operations at water vapor pressure of <12 mbar. Only (NH₄)₂Zn(SO₄)₂·6H₂O met all the set criteria. As stated by the authors, the selected salt has an energy density of 1.78 GJ/m³, a dehydration temperature of 84 °C (after 1 cycle), and is fully cycleable for at least 10 cycles at both 14 and 12 mbar without loss of performance.

Kiyabu et al. [122] performed a computational screening of hydration reactions based mainly on gravimetric and volumetric energy densities. Three operating temperature ranges were considered (see Table 4). 265 hydration reactions were screened, and 17 promising reactions were identified (as shown in Figure 5). Six compounds were considered novel in the context of use as TCES materials (presented in Table 4). In addition, two property-performance databases were created, one for hydrates and one for hydroxides.

Mazur et al. evaluated 454 salt hydrates and 1073 hydration reactions in their comprehensive work [63]. The hydrates were screened based on their availability, stability, energy density (>1 GJ/m³), and the use of three case scenarios. These scenarios included (1) application of TCES for space heating, T = 30 °C, p_H2O of 12 mbar; (2) for hot water production, T > 55 °C, p_H2O of 12 mbar; (3) using the heat sources for charging the energy storage system with a temperature of T < 160 °C while hydration T > 55 °C and p_H2O of 31 mbar (graphically shown in Figure 6). Finally, eight salts and nine reactions were found to meet all the criteria. In addition, the use of the salt stabilization method allowed an additional 4 salts and reactions to be distinguished (as shown in Table 4).

Palacios et al. [124] made an extensive literature review of the 22 most relevant salt hydrates, such as bromides, sulfates, carbonates, chlorides, hydroxides, and sulfides. A very detailed description of the most studied hydrate salts in TCES systems can be found in their work.

An impressive search algorithm for the selection of potential thermochemical systems (not only salt hydrates) was created by Markus Deutch et al. [125]. Reaction with different gases, such as water vapor, CO₂, NH₃, O_2, or SO₂, can be considered. Reactions used in the algorithm were selected from the HSC chemistry database.

In another work, by Palacios et al. [100], the authors performed a theoretical screening (first step) and an extensive practical screening (second step). In addition, different case studies were applied. In the second step, hydration and dehydration tests were performed at different heating rates (1, 2.5, and 5 °C/min). Other thermal parameters, such as specific heat, hydration kinetics, conversion rate, and chemical stability, were also studied to thoroughly characterize potential TCES candidates. The salts studied included CaCl₂, Ca(NO₃)₂, CaSO₄, CuSO₄, MgCl₂, Mg(NO₃)₂, MgSO₄, SrBr₂, Zn(NO₃)₂, and K₂CO₃. Finally, MgSO₄ was selected as the most promising material, featuring high energy density above 2 MJ/m³ and low cost (EUR 1/MJ). CaSO₄ and CuSO₄ were revealed as promising candidates.

A comprehensive review of the literature shows that a wide range of salt hydrates have been investigated for their potential use in TCES systems. Many of the data, including those presented in

Table 5. Data on the most popular salt hydrates used in TCES systems.

Salt Hydrate	nH2O	TDH, °C	TH, °C	Tm, °C	dHS, kg/m3	E, GJ/m3	ΔHR, kJ/molH2O	Remarks	Ref.
Al2(SO4)3·18H2O	8	100		88	1690	1.41	55.4	Very low temperature rise during hydration. Not fully dehydrated even at 300 °C.	[83]
Al2(SO4)3·6H2O
	0	306 *	293			3.19			[120]
K2CO3·1.5H2O	0	93			2043	1.3	61–66	Quite low energy. Good reversibility and excellent stability. Slow hydration rates. Promising material.	[126]
				>80	2150	1.24			[83]
		65	59	>150	2180	1.3 O/0.96 C			[120]
		120 (2% RH at 25 °C)	25 (38% RH)						[127]
KAl(SO4)2·3H2O	0	63 *	57	92	2200	1.39 O/1.01 C	55		[120]
CaCl2·6H2O	1	45		29	1710	2.16	55	Very low Tm and < TDH. Gel formation during hydration. Deliquescence. Often used in composites (see Section 4.1).	[83]
		71 *	32		1840	2.58 O			[120]
CaCl2·2H2O	0					1.44			[128]
CaBr2·6H2O	4	63		38	2295	2.64	62	Tm < TDH	[83]
	0	81	74			2.67			[120]
Ca(NO3)2·4H2O	2	44 *	38		1820	0.86	52	Strong oxidizer	[120]
	0	50	43			1.71
CaSO4·0.5H2O	0	120			2320	0.51	61–66	Poor reversibility in practical applications.	[84]
		102 *	94			0.60			[120]
MgCl2·6H2O	4	<100, 57			1569	1.1	72	HCl can be produced during dehydration > 115°C. Deliquescence. Low cost. Promising material when using in composites (see Section 4.1)	[82]
	2	96				1.97			[129]
	2	104 *	61	117	1560	1.93 O/1.24 C			[83]
						2.48			[120]
	1	117							[128]
MgSO4·7H2O	1	85		49	1670	2.27	56	Good hydration only under high RH. Stable under cycling. High energy density. Needs a lot of hydration time. Slow kinetic of hydration and low temperature lift.	[83]
			24			2.27 O			[120]
MgSO4·6H2O	1		22			2.08 O	55
MgSO4·6H2O	2	130 (5% RH)		88–93		2.0			[83,110]
	1.3
MgBr2·6H2O	4	<100		152–165	2000	0.99	72		[83]
MgBr2·H2O	0	116 *	109		1569	1.19	69		[120]
Mg(NO3)2·6H2O	2	68 *	61	-	1670	1.53 O/1.04 C	58	Decomposition can occur and release of N2	[120]
LaCl3·7H2O	1	86–105		40	2230	2.13	59	High cost.	[83]
	3	73 *	66			1.48 O/1.03			[120]
	0	158 *	48			2.41
La(NO3)3·6H2O	1.5	<100		40	2347	0.98	58	Oxidizer, high cost.	[83]
						1.41
LiCl·H2O	0			99	1700	1.75		Quite expensive and high deliquescence. Corrosive.	[83]
		72 *	66		1760	2.08 O/1.36 C	71.3		[120]
		87					62		[82]
					1780
LiNO3·3H2O	0	<100		30	1550	1.24	55	Oxidant, expensive	[83]
		34	28			2.13 O	55.6		[120]
LiBr·H2O	0	110 *	103	-	2670	2.01 O/1.37 C	69.5	Deliquescence	[120]
		90			1570		70		[82]
SrCl2·6H2O	2	42–50			1930	2.02		Stability under cycling, fully reversible under 65% RH, 22 °C. Very promising.	[82]
	0				1960	2.51	57		[83]
	2	33 *	27			1.58	23.7		[120]
	0	202 *	28	61		2.99
	0					2.4 ± 0.1			[130]
SrBr2·6H2O	1	90						Stability under 10 cycles. Reversible and reasonable energy density. Total dehydration from 80 °C. Very promising. Expensive. Irritant.	[82]
	1				2386	2.26	67.4		[83]
	1				1900	0.5			[84]
	2	33	28			1.61	54.6		[120]
	0	122	48	88		2.49
CuSO4·5H2O	1	40	30			2.05		Maximum temperature of hydration is 60 °C. Harmful.	[120]
		65–94			2286	2.06	57		[84]
Na3PO4·12H2O	8/7/6/4/0.5	37/46/52/78/84			1620	3.6	53	Not reversible. Uptake of 6 mol H2O at RH 60–65% and 20–22 °C.	[82]
	0	33	27				55		[120]
Na2SO4·7H2O	0	81.8		23	-	2.21	53.7	Low melting point of hydrates (about 30 °C). Require high RH to fully hydrate	[82,131]
Na2SO4·10H2O	0			32		2.48	50.5		[120,132]
Na2S2O3·5H2O	4/3/1	44.1/54.6/71.6					56	Formation of layer which inhibit the water release	[83]
	0	32	26	48	1690	1.91	54		[120]
Na2S·9H2O	5/2/0.5	60/92/100			1430	2.66	62	Toxic, corrosive, H2S can be released. Partial melting.	[83]
Na2S·5H2O	0.5	82 *	66			2.79 O			[120]
Zn(NO3)2·6H2O	4	39	34	36		0.86	62.1		[120,132]
ZnSO4·6H2O	1	27 *	21		1970	2.0	52.1		[120]

O/C—open and closed system, respectively, calculated by Donkers et al. [120]; DRH—deliquescence relative humidity; d_HS—density of hydrated salt; E—volumetric energy density (of pure salt); n_H2O—moles of water in dehydrated salt; T_DH—dehydration temperature; T_H–hydration temperature at 12 mbar H₂O or *—at 20 mbar of H₂O; T_m—melting temperature; Ref.—reference.

, are derived from calculations based on the standard enthalpy and entropy of dehydration reactions and using the Clausius–Clapeyron relationship (to calculate equilibrium temperature or water vapor pressure) [63,120]. However, experimental studies often reveal significant discrepancies between theoretical and actual energy densities, underscoring the importance of empirical validation.

Leslie Glasser’s extensive thermodynamic databases have been instrumental in identifying trends among common inorganic salts [103]. Glasser’s data, as well as those in Table 5, indicate that the dehydration enthalpy for most salts is in the range of 50–70 kJ/mol H₂O (g). Accordingly, the volumetric energy density for the most promising systems ranges from about 1 GJ/m³ to 2.7 GJ/m³.

Despite the large number of theoretically viable hydration reactions (numbering in the hundreds), only a limited number of salt hydrates meet the practical criteria necessary for real-world application. This is due to several factors, which are discussed below.

While many salts have suitable hydration enthalpies, their reactions may lack reversibility (e.g., phosphates), preventing their use. The energy that can be derived from salt hydration is less than theoretical, as shown in

Table 6. Energy density and reaction temperature in TCES based on hydration/dehydration for selected inorganic salts. Based on [88].

Reaction (Solid ⇄ Solid + Gas)	Theoretical Energy Density (GJ/m3)	Experimental Energy Density (GJ/m3)	Reaction Temperature (Charging/Discharging) (°C)	Ref.
MgCl2·6H2O ⇄ MgCl2·H2O + 5H2O	2.5	0.71	150/30–50 *	[123,133]
MgCl2·4H2O ⇄ MgCl2·2H2O + 2H2O	1.27	1.10	118/NA *	[134]
CaCl2·2H2O ⇄ CaCl2 + 2H2O	1.1	NA	95	[123]
CaCl2·6H2O ⇄ CaCl2 + 6H2O	2.8	1.47	NA/200	[135]
Al2(SO4)3·6H2O ⇄ Al2(SO4)3 + 6H2O	1.9	NA	150	[123]
MgSO4·6H2O ⇄ MgSO4·H2O + 5H2O	2.37	1.83; 0.38	72/NA *; NA/150 °C *	[133,136]
Na2S2·5H2O ⇄ Na2S2·1/2H2O + 9/2H2O	2.7	NA	80/65 *	[137]
SrBr2·6H2O ⇄ SrBr2·H2O + 5H2O	2.3	2.08	NA/23.5 **	[138]
Li2SO4·H2O ⇄ Li2SO4 + H2O	0.92	0.80	103/NA	[134]
CuSO4·5H2O ⇄ CuSO4·H2O + 4H2O	2.07	1.85	92/NA	[134]

NA—non available; *—water vapor pressure of 13 mbar; **—water vapor pressure of 20 mbar.

.

Sometimes, if complete dehydration is not achieved, the energy density may even drop by half during the subsequent hydration step. In addition, materials that are toxic (e.g., salts of Mn, Co, Cd, Ba, Cr), flammable or oxidizing (perchlorates, periodates), and may pose safety hazards such as fire or explosion risks are generally excluded from the outset.

The hydration temperature is, after the energy density, one of the key parameters that determines the real use of a given salt. It will determine if a salt is suitable only for domestic heating or also for hot tap water [120]. In some cases, the temperature rise during hydration (i.e., discharging the energy storage unit) may not be sufficient to provide useful heat, even if the reaction is technically feasible. According to N’Tsoukpoe [83], the most promising salts suitable for domestic application are SrBr₂ (from hexa- to monohydrate) and LaCl₃ (from heptahydrate to anhydrous salt).

According to Donkers [120], some of the reactions with the highest hydration temperatures are Mg(NO₃)₂ (from 0 to 2 moles of water), LiBr (from 0 to 1 mole of H₂O), and LiI (from 1 to 3 moles of H₂O), as shown in Figure 7.

The dehydration temperature should be as low as possible to facilitate charging of the TCES system. The best salt hydrates require dehydration temperatures below 100 °C. This allows them to be used with solar thermal collectors and other low-temperature heat sources. Among the 25 salts selected by Donkers et al. that have the lowest dehydration temperature are salts of FeCl₂, CuCl₂, KAl(SO₄)₂, and MnI₂ [120].

The exclusion of a particular salt can also be determined by the undesirable phenomenon that the salt begins to melt (i.e., release water and dissolve in it) before it is dehydrated (i.e., T_m < T_DH). The negative consequences of this situation are described in more detail in Section 2.4. For example, in the graph (Figure 7), it can be seen that the melting temperatures (marked as x) of the selected salts are higher than the dehydration temperature.

Another factor that greatly affects the energy efficiency of the storage and may even make the use of a particular salt unreasonable under certain conditions is deliquescence. This phenomenon has already been described in Section 2.2 and Section 2.4. Here, DRH for some salts is presented in Figure 8. The higher the DRH, the better. When the storage is discharged, air with an RH higher than the DRH will eventually lead to the formation of a solution of the salt in question, i.e., deliquescence.

Regarding the deliquescence, it is worth mentioning the data presented by Clark et al. [82] and shown here in

Table 7. Temperature and relative humidity (RH) required for application under different conditions for salt hydrate formation. Adopted from [82].

Salt	Low Temperature Application (0 °C)	Indoor Temperature Application (20 °C)
MgSO4·7H2O	35–95% RH	43–92% RH
SrCl2·6H2O	25–78% RH	33–72% RH
Na3PO4·12H2O	–	>28% RH
SrBr2·6H2O	5–65% RH	9–46% RH
MgCl2·6H2O	–	4–33% RH

. These data determine the minimum humidity (as RH) required to obtain the maximum degree of hydration. This, in turn, could lead to the maximum heat energy that can be obtained from a given salt. Clark et al. considered the use of seasonal storage, so they used the temperature in both winter and summer. Very importantly, if the RH of the air exceeds the level given in the table, then overhydration (i.e., deliquescence) can occur. As can be seen from Table 7, MgSO₄ can be used seasonally even with relatively dry air (RH of about 40%). There is also no concern about its dissolution. On the other hand, in the case of MgCl₂ or SrBr₂, it should be noted that overhydration occurs in air with RH above 33% for MgCl₂ and with RH above 46% for SrBr₂ (considering 20 °C). Therefore, MgCl₂ would form an aqueous solution under ambient conditions, rendering it unsuitable for the intended application.

Economic factors also play a critical role (please see

Table 8. Estimated costs of some salt hydrates proposed in TCES materials.

Salt	Moles of Water (Hydration–Dehydration)	Density of Hydrate, kg/m3	Theoretical Energy Density, GJ/m3	Estimated Price *, €/kg	Price of Stored Energy, €/GJ
CaCl2	6–1	1710	2.16	0.16	142
	6–2		1.85		166
	6–0		2.82		109
CaBr2	6–0.3	2295	2.64	3.5	3047
LaCl3	7–1	2223	2.13	2.2	2297
La(NO3)3	6–1.5	2347	1.41	3.5	5821
LiCl	1–0	1700	1.75	10	9713
MgBr2	6–4	2000	0.99	3.5	7045
MgCl2	6–2	1569	1.97	0.15	120
	6–1		3.14		75
MgSO4	7–1	1690	1.90	0.2	177
	7–0	1690	2.81		120
	6–0.1	1810	2.20		165
	6–1	1810	1.69		214
SrBr2	6–1	2386	2.26	7	7393
SrCl2	6–0	1960	2.51	2	1561
Na2S	9–0	1430	3.17	0.45	203
	5–0	1580	2.93		243

* Prices based on chemicals in industrial quantities from Alibaba.com and Indiamarkt.

). Although certain rare earth salts, such as those of Gd, Eu, or La, have good thermochemical properties, their extremely high cost (e.g., about EUR 5000 /kg of GdCl₃·6H₂O and EUR 1400/kg of LaCl₃·7H₂O [140]) and limited availability make them impractical for real-world applications. Lithium salts or SrBr₂ are also expensive, and their main demand is in the electric cell industry. Moreover, LiCl has a very low DRH point (11%). In contrast, magnesium, potassium, sodium, and calcium salts are relatively cheap.

The aforementioned issues demonstrate the difficulties associated with the transition of numerous salt hydrate systems from laboratory-scale conceptual frameworks to viable energy storage technologies.

Comparing the properties of different salt hydrates is clear, as they have a variety of properties. Nevertheless, despite many screening tests made by researchers, few salt hydrates have proven to be particularly effective in TCES systems due to their favorable thermodynamic properties and practical application. In particular, strontium chloride hexahydrate (SrCl₂·6H₂O) and strontium bromide hexahydrate (SrBr₂·6H₂O) have shown great promise. Experimental research shows that these salts have a high energy storage capacity(SrCl₂ of 2.4 GJ/m³) and retain their structural integrity after numerous hydration-dehydration cycles, making them ideal for building heating applications [120,130].

SrCl₂ stands out as the most effective salt over a wide range of dehydration temperatures [120,141]. At lower dehydration temperatures, the energy density is somewhat reduced. SrCl₂ is particularly advantageous in a range of moderate to high humidity conditions. The storage density for the hydration of the anhydrous SrCl₂ to the dehydrate is 1.7 GJ/m³ [141]. Hydrating to the monohydrate, temperatures of about 100 °C can be reached even at vapor pressures of only 1.2–2 kPa [141]. Measured hydration enthalpy is in the range of 50–63 kJ/mol_H2O [130]. Compared to sulfate salts, strontium chloride has the advantage of requiring lower vapor pressures to obtain the complete hydrate (SrCl₂·6H₂O).

SrBr₂ has a high energy density and stability. In an open system, dehydration can occur within a temperature range of 60–80 °C, while hydration takes place at temperatures between 20 and 35 °C [142]. These temperature ranges make SrBr₂·6H₂O an attractive candidate for use with solar heat and low-grade waste heat, which are two important energy sources in renewable-based systems. However, due to the tendency of SrBr₂ to overhydrate, low humidity conditions must be maintained [82]. Moreover, it is also quite expensive.

The use of SrBr₂ for low-temperature energy storage and building applications was reviewed by Fopah-Lele et al. [143]. Fopah-Lele et al., in other studies [144], also performed the dehydration of SrBr₂ at 58 °C and hydration at 45 °C. Mauran et al. [145] conducted the dehydration process at 80 °C. TCES based on SrBr₂/SrBr₂·H₂O has been experimentally validated on a 1 kW lab-scale storage unit [146]. SrBr₂ and SrCl₂ have also been tested in a variety of composite systems and reactors, as presented in the next sections (Section 4 and Section 5).

Magnesium sulfate (MgSO₄·7H₂O) has also been the focus of extensive research [70,111,133,147,148,149]. It offers a high theoretical energy storage capacity of 2.27 MJ/m³ (from heptahydrate to monohydrate), but faces problems related to sluggish reaction kinetics and partial irreversibility under real-world conditions, which could limit its efficiency in multiple cycles. Very slow kinetics of hydration of MgSO₄ can lead to negligible temperature lift of used air as a heat transfer fluid. It has been found that the formation of the amorphous phase during dehydration of MgSO₄·6H₂O is associated with a slow and inhomogeneous reorganization of the crystal structure and is responsible for this issue, as shown in Figure 9 [133]. Moreover, MgSO₄·7H₂O cannot be produced during rehydration when water vapor pressure is 13 mbar (as set in seasonal heat storage systems) [133].

Research has shown that the reaction from hexahydrate to monohydrate is the most promising [70,149]. In research conducted by Donkers et al. [120], MgSO₄ was rejected due to the low hydration temperature (discharging temperature) that does not fit the demands for domestic heating and hot tap water. In other experimental studies, conducted by Ferchaud [133], the measured enthalpy of dehydration of MgSO₄·6H₂O to MgSO₄·H₂O was 243.1 kJ/mol, while, during hydration, only 110 kJ/mol of energy was released. Moreover, during cycling tests at 13 mbar, 0.76 GJ/m³ of energy density was obtained, which is quite low considering the maximum theoretical enthalpy.

Salama’s experimental research using MgSO₄ composite material has shown that the minimum required dehydration temperature of monohydrate is 85 °C. The experimental energy density was in the range of 0.336–0.91 GJ/m^3, and the discharging temperature was 15 °C.

Calcium chloride (CaCl₂·6H₂O) seems to be another potential candidate, with a significant enthalpy of reaction (from 1.03 to 3.1 GJ/m³, depending on the hydration level). However, due to its low melting points of 30 °C (hexahydrate) and 45 °C (tetrahydrate), and high deliquescence, its practical application is limited. Moreover, its hydration temperature is also low. Nevertheless, CaCl₂ is utilized in composite sorbents for thermochemical and adsorption applications due to its high water uptake capacity, low cost, and good availability. When incorporated into porous matrices (e.g., silica gel, as discussed in more detail in Section 4.1), CaCl₂ benefits from improved stability, reduced deliquescence, and enhanced mass transfer. These composites have found application in low-temperature energy storage and for building power to heat and cooling applications utilizing sorption reactors [150].

Although potassium carbonate (K₂CO₃) has a relatively low energy density (1.3 GJ/m³) compared to other candidates, it stands out as a highly practical and reliable material for thermochemical energy storage. The enthalpy of the reaction, based on experimental hydration/dehydration results, is in the range of 60–64 kJ/mol_H2O. The estimated energy density for the bulk material (including porosity) is about 0.75 GJ/m³ [151]. Moreover, the discharging temperature is suitable for domestic applications.

The main advantage of K₂CO₃ is its excellent chemical stability, quite high DRH, and its non-toxicity, which allows it to be used safely in open systems without any health or environmental risks [151]. Unlike many other salts, K₂CO₃ does not release harmful gases during hydration or dehydration, nor does it present significant corrosion or degradation problems over multiple cycles. These favorable characteristics make it an attractive option for long-term applications where safety, system design simplicity, and material durability are more important than maximizing energy density. Research carried out by Sögütoglu et al. confirmed that K₂CO₃ is recommended for both open and closed systems [152]. The energy density was 1.28 GJ/m³ and 0.95 GJ/m³ for open and closed systems, respectively. Nevertheless, the discharge temperature of 33–45 °C is too low for the production of hot tap water, but sufficient for domestic space heating purposes [152].

Among the various salt hydrates studied for TCES, Na₂S and MgCl₂ have attracted interest due to their relatively high energy densities (about 2.79 GJ/m³ and 1.93 GJ/m^3, respectively). Uptake of H₂O by MgCl₂·2 H₂O to form MgCl₂·6H₂O releases an average of 58.3–62.8 kJ mol_H2O. However, the practical application of Na₂S and MgCl₂ is limited by significant safety and corrosion concerns, resulting from their chemical and thermal instability, respectively. Sodium sulfide releases hydrogen sulfide (H₂S) in the presence of moisture and CO₂ from the air [153]. H₂S poses serious health and safety risks. In addition, Na₂S is known to be highly corrosive, complicating reactor design and material compatibility. Similarly, MgCl₂, while considered less hazardous, can release highly corrosive hydrogen chloride (HCl) during dehydration at elevated temperatures (above 140 °C) [154,155]. Complete dehydration of MgCl₂ is impossible without decomposition of the salt. These drawbacks require careful material selection, reactor sealing, and gas handling strategies if such salts are to be considered for TCES applications.

The extensive study of the thermochemical performance of K₂CO₃, MgCl_2, and Na₂S was performed by Sögütoglu et al. [152]. Due to chemical instability, MgCl₂ and Na₂S were rejected, and the most promising material was K₂CO₃. However, the authors state that Na₂S can be stable in closed systems. In an open system, it degraded fully after four cycles. Moreover, experimental research [152] has shown that water uptake by MgCl₂ decreased rapidly over cycles. On the other hand, MgCl₂·6H₂O can be considered promising and recommended for seasonal heat storage, as shown in other research by Ferrchard [133]. The minimum hydration temperature should be 40 °C to avoid overhydration and liquefaction of the salt. As in the case of MgSO₄, the authors found that the energy recovered during the hydration process is less than the energy required for dehydration, 0.71 and 0.84 GJ/m³, respectively.

MgCl₂·2H₂O was also considered a promising TCES material in research conducted by Zondag et al. [129]. Using hydration from dehydrate to hexahydrate to dihydrate, the temperature rise in the reactor was from 25 °C to 87 °C. A similar temperature rise was observed using CaCl₂.

To address the limitations associated with the use of pure salt hydrates, such as agglomeration, deliquescence, and insufficient porosity, various material modifications have been developed. These strategies aim to improve the structural stability, sorption performance, and overall reliability of the storage media. The following, Section 4, discusses these approaches in more detail.

4. Materials Enhancement Strategies

As described in the previous sections, numerous challenges related to material properties hinder the widespread use of salt hydrate in TCES systems. The use of pure salt is not very practical as it has low porosity and therefore poor heat and mass transfer properties. For the effective work of the energy storage based on gas–solid reactions, a sufficiently large reaction surface between the water vapor and the salt is required. Other material challenges that restrict practical applications of pure salts include agglomeration and caking, deliquescence, and low thermal conductivity (as discussed in more detail in Section 2.4).

In order to address these challenges, various material enhancement strategies have been developed, including the integration of a porous matrix, the use of additives, and the use of mixed salts. These systems are most commonly referred to as composites and will be discussed in this section.

4.1. Salt in Porous Matrix Composite

A salt in a porous matrix composite (SPMC) is defined as a hybrid material formed by incorporating a salt hydrate into the structure of a porous support material, such as silica gel, activated alumina, or expanded graphite [156,157]. Interestingly, the origins of this type of composite extend to the 1990s, when Levitskij and co-authors conducted research on a composite based on CaCl₂ and silica gel [156].

The properties of the composites fall between those of inorganic salts and porous support [157]. Importantly, SPMCs can use both the adsorption enthalpy of the adsorbent (porous matrix) and the absorption enthalpy of salt hydrate [89]. Physical adsorption occurs when H₂O is adsorbed near the sorbent surface (e.g., zeolite) and internal pores, driven by van der Waals forces. Chemical sorption (chemisorption), in turn, occurs in salt hydration reactions, where molecular bonds are formed.

The use of SPMCs aims to combine the high energy storage capacity of the salt with the advantageous structural and transport properties of the matrix. The porous nature of the host material provides a substantial surface area, thereby enhancing vapor diffusion and reducing agglomeration, phase separation, swelling, and salt leakage during hydration and dehydration cycles [157]. As a result, the composite material enhances the thermal and mechanical stability of the TCES material, improving cyclic stability and energy performance. Moreover, salt-impregnated adsorbents have much higher water adsorption than adsorbents alone [158,159].

A variety of porous matrices have been investigated to date. These include, among others, zeolites, both synthetic and natural [160,161,162], silica gels [163,164,165,166], activated carbons [167,168], metal-organic frameworks (MOFs) [169,170], aluminophosphates (AIPOs) and silico-aluminophosphates (SAPOs) [171], carbon nanotubes [172], expanded graphite (EG) [173,174], vermiculite [175,176,177], and other natural clays [113,178,179].

The characteristics of many porous matrices were thoroughly described by Yang et al. [90]. Experimental results of some sorption materials are presented in

Table 9. Characteristics of sorption materials (used as porous matrix).

Material	Desorption Temperature, °C	Adsorbed Water, kg/kg	Maximum Temperature in the Storage, °C	Energy Density, Wh/kg	V Macro, cm3/g	V Meso, cm3/g	Pore Size	λ (W/mK)
LiLSX [171]	200	0.24	107	225	-	-	-	-
NaLSX [171]	200	0.23	97	185	-	-	-	-
SAPO [171]	100, 150	0.23	72	154	-	-	-	-
Silica gel [171]	120	0.18	123	62	-	-	-	-
Silica gel [180,181]	-	0.35	-	240	1.310	0.381	2.45 nm	0.0569
Vermiculite [180,181]	-	<0.1	-	-	4.110	0.017	3.68 µm	0.0527
Zeolite 13X [180,181]	>180	0.30	-	310–680	1.527	0.292	1.21 nm	0.0737
Zeolite 13X [159]	-	0.17–0.22 50–80% RH	-	-	-	-	3.36 nm	-
Activated carbon [180,181]	-	0.35	-	-	2.365	0.395	1.98 nm	0.0790

λ—thermal conductivity.

[171,180,181]. As can be seen from this table, water adsorption is in the range of 0.20–0.35 g/g, and the temperature required for regeneration is higher than for pure hydrate salts. In addition, depending on the supplier of a particular sorbent and the test method, data for the same sorbent may vary.

An ideal support matrix for salt hydrate composites should possess several essential properties to ensure effective operation and long-term stability. A high degree of porosity and a significant specific surface area are necessary to support adequate salt loading and to enhance vapor transport during hydration and dehydration processes. The matrix must also have good thermal conductivity to facilitate effective heat transfer within the composite. Mechanical integrity is required to withstand repeated cycling without fragmentation or structural failure. Chemical compatibility with the salt is essential to prevent unwanted reactions, deterioration, or degradation of performance over time. In addition, the matrix should be sufficiently hydrophilic to promote effective interaction with the water vapor, while remaining sufficiently inert to prevent deliquescence or premature hydration at ambient conditions. Ideally, it should be cost-effective, readily available, and environmentally friendly to allow for widespread use [158,171].

Silica gel is a porous matrix that has gained widespread popularity due to its large surface area, chemical stability, corrosion resistance, cost-effectiveness, safety, and good compatibility with various salt hydrates. Moreover, silica gel requires a low charging temperature, i.e., below 120 °C [171]. However, the material’s low thermal conductivity and varying conditions can affect its properties; for example, different salt hydrates exhibit different dispersion phases on the surface of silica gel [182]. The main drawbacks of silica gel are limited temperature lift during the discharge phase and its relatively low energy density in practical applications (about 0.2 GJ/m³ kg) [171]. The reason for this is that water adsorption is low under the typical discharge conditions of the TCES system. Mesoporous silica materials are more promising sorbents [183].

Zeolites (microporous crystalline aluminum silicates) with high water sorption capacity and homogeneous microporous structure, such as 4A and 13X, provide high water affinity and good thermal stability. However, at high salt content, their small pore diameters (4–10 Å) and low thermal conductivity (0.15 W/mK) can hinder salt mobility and cause pore blocking effects. Furthermore, when using zeolites, the potential ion exchange process must be carefully considered. For example, exposure of zeolite 13X to CaCl₂ solutions can result in the replacement of Na⁺ ions in the zeolite framework with Ca²⁺ ions from the solution [89]. This ion exchange can alter the structural and sorptive properties of the zeolite, thereby affecting the overall performance and stability of the composite. Therefore, the chemical interactions between the salt and the zeolite support play a critical role in the design and optimization of these systems.

Zeolites can be used in cycles with high load temperatures above 150 °C and have a high energy density (0.2–0.6 GJ/m³) [117,161,184]. The high regeneration temperature can be a major disadvantage if a lower temperature source is available (e.g., heat source from solar panels) [171]. Moreover, weak mechanical strength can result in performance degradation and sorbent losses. Among the many zeolites investigated, 13X appears the most promising.

Higher energy density, up to 0.86 GJ/m³, and corresponding temperature lift are demonstrated by AlPOs and SAPOs at moderate charging temperatures of 75–140 °C [185]. However, the major drawback is the high price of these materials.

Vermiculite, a naturally occurring group of hydrated lamellar minerals (aluminum-iron-magnesium silicates), is valued for its low cost, ability to intercalate salt, large pore volume, and relatively good water retention. However, the material’s relatively low surface area and tendency to undergo structural changes due to swelling/shrinkage during cycling can pose significant challenges [158]. Thermal techniques can also be used to expand the vermiculite, which results in a porous structure with slit-shaped pores that are a few microns in size. Expanded vermiculite (EV) can form numerous layers with abundant porosity that facilitate water vapor transport [175]. Depending on the salt hydrate used, the high porosity of vermiculite allows high salt loading and thus high energy density (0.3–0.9 GJ/m³ of bed) [186].

Expanded graphite and other carbon materials (such as carbon nanotubes) offer superior thermal conductivity and a layered configuration that accommodates significant salt loading, effectively improving heat transfer within the composite. In addition, there is no phase separation or salt leakage. They also exhibit good cycling stability, even with the use of highly deliquescent salts and elevated water vapor content [187]. However, the high cost compared to mineral-based supports can pose challenges.

Metal-organic frameworks (MOFs) have emerged as promising porous supports in TCES systems due to their high surface areas, tunable pore structures, and strong hydrophilicity. MOFs such as MIL-101 (Cr), aluminum fumarate, and CPO-27 (Ni) have demonstrated significant water uptake capacities and structural stability, making them suitable as salt hydrate hosts [188,189]. For example, MIL-101 (Cr) has been used to support salts such as CaCl₂ and Na₂S₂O₃, resulting in composites with high energy storage densities and excellent cycling stability [190]. The incorporation of salts into MOFs can increase the water vapor adsorption capacity, although excessive salt loading can clog the pores and reduce performance. Therefore, optimizing the salt content within the MOF structure is critical to balance adsorption capacity and structural integrity. In addition, MOFs offer the advantage of low desorption temperatures, which is beneficial for low-grade heat utilization. However, challenges such as hydrothermal stability and high cost remain, and further research is needed to address these issues for large-scale TCES applications.

An interesting candidate for thermal energy storage applications is MXene, an advanced two-dimensional (2D) material characterized by a layered interconnected architecture [191]. MXenes exhibit remarkable properties such as high thermal conductivity, ultralow density, large specific surface area, abundant surface active sites, and excellent sorption performance. This unique class of 2D materials consists of transition metal carbides or nitrides, generally described by the formula M_n+1X_nT_x, where M represents a transition metal, X represents carbon and/or nitrogen, T_x represents surface functional groups, and n can be 1, 2, or 3 [192].

The most common salts used in porous matrix include MgSO₄, CaCl₂, LiBr, SrCl₂, SrBr₂, and LiCl [143,166,175,176,193,194,195,196].

Table 10. Summarized research on composite TCES material using salt hydrates.

Salt and Content in the Composite	Porous Matrix	Method of Preparation	Major Results and Other Information	Reference (Main Author and Date)
CaCl2, 16–66%	Macroporous vermiculite (V), expanded perlite (EP), pumice, mesoporous silica gel (SG), microporous Zeolite 13X	Vacuum wet impregnation, centrifugal draining and drying at 150 °C for 24 h.	For V-66%CaCl2 and EP-69%CaCl2, E = 0.55–0.65 GJ/m3 but slow reaction kinetics. For SG-CaCl2: superior reaction rates, temperature lift of 16 °C at 50% RH. High pressure drop in case of SG-CaCl2. The most promising for larger scale systems is EP-CaCl2.	Chen, 2025 [201]
CaCl2	AC four types	Wet impregnation and vacuum freeze drying	E = 2039 kJ/kg for AC with 43.4% of CaCl2. Enthalpy of 42.18 kJ/molH2O. Slight decline in water absorption after 25 cycles. AC host matrices enhance low-temperature water desorption. The best water adsorption of 1.07 g/g.	Zhang, 2024 [167]
CaCl2 62–63%	EG (novel design of perforated EG block)	Molten salt impregnation for 72 h and dried at 150 °C	E in the range of 135–277 kWh/m3 (of material) or 96–197 kWh/m3 (of reactor). Conditions in the reactor: (0.01 kgH2O/kg air, 20 °C, gas flow rate of 100–400 L/min. Temperature lift up to 24 °C. Impressive 90 hydration/dehydration test and material remained stable. Test of block: hydration at 20 °C and 40% RH for 17 h, dehydration at 150 °C overnight.	Galazutdinova, 2024 [187]
CaCl2, 63–77%	Graphite and sodium alginate matrix	Wet impregnation and mould method.	Energy density in the range of 1052–1281 kJ/kg. Salt with higher salt content had reduced energy efficiency. Thermal conductivity about 0.33 W/mK.	Reynolds, 2024 [202]
CaCl2 46–50.4%	Mesoporous carbon CMK-3. AC and EG.	Wet impregnation and vacuum drying in oven 150 °C	Excellent stability after 25 cycles. Promising composite below 120 °C. E = 2037 kJ/kg.	Gao, 2023 [168]
CaCl2, 32% and 42%	Silica-PEG	Sol–gel method	E = 782 kJ/kg, what is 70% higher than zeolite. Water adsorption of 0.37 g/g at regeneration temperature of 130 °C. Absorption at 30 °C and 42% RH. Kinetics slower than zeolite 13X. Lab-scale open reactor. 4 cycles were performed effectively.	Berut, 2023 [203]
CaCl2, 23–58%	EG		Dehydration < 120 °C. E = 1638 J/g. Water adsorption of 0.79 g/g.	Gao 2022 [204]
CaCl2, 25–60%	Aluminium fumarate MOF	Synthesis of MOF	Maximum water sorption of 0.68 g/g and heat of sorption of 1840 J/g for 58% CaCl2/MOF.	Touloumet, 2021 [205]
CaCl2, 15%	Silica gel, alumina, bentonite	Incipient wetness impregnation. Dried at 110 °C and dehydrated at 260 °C.	The highest water adsorption of 0.27 g/g for silica gel-CaCl2. Average heat of dehydration E = 746 J/g (of sample). Alumina-CaCl2: water adsorption of 0.17 g/g and E = 576 J/g. Bentonite-CaCl2: water adsorption of 0.23 g/g and E = 719 J/g.	Jabbari-Hichri, 2017 [113]
CaCl2	AC Attapulgite (Palygorskite)	Wet impregnation using 30% CaCl2	Water adsorption of 0.223 g/g, E = 580 kJ/kg. Maximum temperature in the storage 77 °C. Water adsorption of 0.18 g/g, E = 637 kJ/kg. Maximum temperature in the storage 63 °C. Apparatus: thermochemical storage unit for 1 kg of material	Jänchen, 2005 [171]
MgSO4, 7–13.4%	Zeolites 3A, 5A, 13X	Wet impregnation in 10% and 20% MgSO4 for 48 h. Dried at 150 °C and 300 °C.	The best composite was MgSO4-Zeolite 13X. Water sorption of 0.21 g/g (24% higher than pure Zeolite 13X). E = 438.4 kJ/kg. Improvement of the mass transfer. Hydration at 60% RH at 25 °C.	Li, 2024 [206]
MgSO4, 1–7.6%	Bead activated carbon (BAC)	Wet impregnation. Dried at 150 °C for 12 h.	Hydration at 30 °C and 60% RH and dehydration at 150 °C. Heat of hydration 920 J/g (for 7.6% MgSO4-BAC). Water adsorption of 0.3 g/g	Nguyen, 2023 [71]
MgSO4, 4.4–19.5%	Biochar from corn cobs	Wet impregnation	The best water adsorption of 0.24 g/g (for 19.5% MgSO4). E = 635 kJ/kg (Hydration at 30 °C and 6%RH).	Nguyen, 2023 [207]
MgSO4	Hydroxyapatite (HAP)	HAP synthesis by co-precipitation. Wet impregnation	Maximum water sorption of 0.14 g/g and heat of hydration of 464 J/g for 20% MgSO4 composite. Good long-term cycling operation over 20 cycles; dehydration at 150 °C and hydration at 30 °C and 60% RH.	Nguyen, 2022 [208]
MgSO4, 4–43%	AC	Incipient wetness impregnation	Increasing hydration enthalpy with increasing content of MgSO4 up to 30%. Hydration at 30% and 60%RH, at temp. 30 °C for 20 h. E max of 1.32 kJ/g (of dry material) at 60% RH for 30% MgSO4/AC. Dehydration at 150 °C. Stable after 8 cycles. Water adsorption up to0.24 g/g. Thermal conductivity 0.43 W/mK for 30% MgSO4/AC.	Bennici, 2022 [209]
MgSO4, 30–50%, 15–30% for Zeolite 13X	Mesostructured cellular foam MCF and Zeolite 13X	MCF by Microemulation templating method. Composite by wet impregnation. Dried at 150 °C under vacuum for 4 h.	30 °C, water vapor pressure of 25 mbar, Dehydration tests at 150 °C (1 °C/min) and water vapor pressure of 25 mbar. Slightly lower dehydration temperature than pure salt. 50% adsorbed water war removed at 65 °C and 70% at 75 °C (90 °C is necessary in case of pure salt). 3D porous structure exhibits higher water adsorption capacity and faster adsorption rate than mesoporous silica with 2D porous structure.	Liu, 2022 [183]
MgSO4, 50–80%	EG	Wet impregnation for 24 h. Dried in oven at 150 °C. Cylindrical mold.	Low and medium TES storage, open systems. Hydration tests at 85% RH at 25 °C for 12 h. The most promising was 60% MgSO4-EG; its heat of reaction of 718.0 kJ/kg, thermal conductivity of 0.6696 W/mK, BET 15 cm2/g. BET for pure EG of 44 cm2/g. Improvement of hydration time (shortened to about ¼) and significant increase in thermal conductivity (by 84.8%).	Miao, 2021 [173]
MgSO4, 30% and 60%.	Diatomite	Dry impregnation. Dried at 250 °C for 12 h. Formed into tablets.	E = 773 kJ/kg over 80–150 °C, water uptake of 0.37 g/g (for 60% MgSO4/diatomite) Cycling stability tests: hydration at 25 °C and 85% RH for 10 h. Dehydration at 250 °C for 6 h.	Zhang, 2021 [178]
MgSO4, 3–11%	Zeolite 13X	Wet impregnation (in 5–20% MgSO4), Dried at 150 °C for 4 h and at 300 °C for 2 h.	Tests at RH in the range of 50–80% for 25 h at 25 °C. Water adsorption in the range of 0.2–0.32 g/g (at 50–80% RH). The best composite was 15% MgSO4-Zeolite 13X, E = 632 J/g at 80% RH, where for pure zeolite E = 551 J/g.	Weng, 2019 [159]
MgSO4, 40–70%	Silicone foam	Silicon foam synthesis from PMHS and PMDS and catalyst	Dehydration at water vapor pressure of 23.4 mbar and temperature up to 150 °C. Silicone foam does not hinder the water vapor diffusion. Progressive loss of salt with increasing compression cycles.	Calabrese, 2019 [210]
MgSO4, 2–14%	Zeolites 3A, 4A, 13X	Wet impregnation in 10–20% MgSO4 solution	Water adsorption of 0.25 g/g (13X Zeolite-MgSO4) Test platform with regulated RH and temperature. Temperature lift about 50 °C.	Xu, 2018 [211]
MgSO4, 10–25%	Zeolite 13X	Wet impregnation. Dried at 150 °C.	Water adsorption of 0.14 g/g at 30 °C and at 17–19 mbar of water vapor pressure. For optimum composite, i.e., 15% MgSO4-Zeolite, E = 166 kWh/m3 at 50% RH. Minimum 50% RH is required to obtain temperature lift 15 °C in the reactor (200 g of sample). Obtained energy corresponds to 45% of theoretical value.	Hongois, 2011 [161]
MgCl2, 86% and 97%	MXene (MX)	Synthesis of MXene. Wet impregnation in 10 and 15% MgCl2 with ultrasonic processing.	Hydration/dehydration enthalpy of 2227 J/g at 85% RH. Water absorption of 1.9 g/g and 2.2 g/g. E = 0.55–0.77 GJ/m3 at 85% RH for 86%Mg/MX and 97% Mg/MX 15%.	Rehman, 2023 [192]
MgCl2	NH2-MIL-88 (Fe) framework coupled with MXene (carbon material)	Synthesis of MOF. Wet impregnation. Dried at 150 °C.	Water adsorption of 2.1 g/g at 75% RH. E = 0.54 GJ/m3 at 75% RH and E = 0.33 GJ/m3 at 65% RH. 20 successive cycling tests. Lower dehydration temperature than pure salt and prevent HCl formation.	Rehman, 2025 [212]
MgCl2, 33–36%	Coral aggregate (CA)	Pretreatment of CA. Vacuum and atmospheric impregnation.	Vacuum impregnation led to higher salt loading. E = 74.2 kWh/m3. Compared to Zeolite 13X (the same conditions), E = 121 kWh/m3.	Wang, 2025 [213]
MgCl2, 50–90%	Graphene oxide aerogel (GO)	Preparation of GO. Hydrothermal and freeze drying method.	Temperature of dehydration was decreased even by 90 °C. E = 1598 J/g for 90% Mg-GO.	Zhou, 2019 [214]
CaCl2 42% MgCl2 33% LiBr 48% CaCl2 31%	Mesoporous silica gel and alumina	Wet impregnation and samples molded under pressure of 150–200 bar	The steep increase in composite thermal conductivity occurs at the same fraction of pore volume occupied by the salt solution, regardless of the amount of salt trapped.	Tanashev, 2013 [215]
SrCl2, 9.9–21.6%	Activated alumina, Al2O3	Vacuum impregnation in 10–30% SrCl2.	The best 20.95% SrCl2-Al2O3. E = 141 kWh/m3 (on the reactor level) and 150 kWh/m3 (on material level). Water uptake of 0.282 g/g at 20 °C and 70%RH. Lower activation energy and faster dehydration than pure salt. Stability over 10 cycles. Energy efficiency up to 93.8%. Bench-scale reactor. Temperature lift of 24 °C.	Yang, 2025 [216]
SrCl2, 35–60%	Cement	Mixing with Portland cement powder.	Dehydration of SrCl2-cement can occur below 90 °C. Average E = 0.49 GJ/m3 for 50% SrCl2. Maximum outlet temperature in the range of 32–37 °C. Lab scale open packed bed reactor. Comparison to Zeolite 13X.	Clark, 2021 [217]
SrCl2 40% and 14% (in pumice)	Expanded clay and pumice	Wet impregnation	E = 29 kWh/m3 (0.104 GJ/m3) for Clay-SrCl2 40%. E = 7.3 kWh/m3 (0.026 GJ/m3) for pumice-14% SrCl2. Impregnation did not affect salt dehydration. Decreasing power output during 4 run for clay composite. Better cycling stability in case of pumice. Maximum air lift temperature of 10 °C. Lab scale open packed bed reactor.	Mehrabadi, 2018 [194]
SrBr2, 15–45%	EV	Wet impregnation method. Dried at 120 °C for 12 h.	For 45% SrBr2-EV, Water adsorption of 0.6 g/g at 30 °C and 60% RH. Desorption enthalpy of 645 J/g.	Ding, 2021 [218]
SrBr2, 63%	MOF MIL-101 (Cr)	Wet impregnation in 20–30% SrBr2 solution. Dried at 100 °C for 4 h.	The best performance for 63% SrBr2. Water adsorption of 0.303 g/g (of dry sample) and E = 233 kWh/m3 at pH2O of 1.25 kPa and 30 °C. Desorption at 80 °C.	D’Ans, 2019 [194]
SrBr2 58%	Silica gel	Incipient wetness impregnation using 40% SrBr2, for 1 h and then dried at 200 °C	E = 203 kWh/m3 (0.731 GJ/m3) Adsorption at 30 °C and desorption at 80 °C, pH2O of 13.5 mbar. Water adsorption of 0.22 g/g. High cycling stability. Average pore diameter in composite of 21 nm.	Courbon, 2017 [164]
SrBr2,30–63%	EV	Wet impregnation method in 10–40% SrBr2. Dried at 120 °C	63% SrBr2-EV, E = 105 kWh/m3 Hydration at 30 °C and 60% RH. The best water uptake up to 0.5 g/g.	Zhang, 2016 [175]
SrBr2, 40% and 80%	Graphite Consolidated composite *	Incipient wetness impregnation. Drying at 200 °C. Form in tablets of 13 mm diameter and thickens 1–2 mm.	For 40% and 80% SrBr2, heat of reaction 417 kJ/kg and 798 kJ/kg, and thermal conductivity 2.30 and 1.30, respectively. Pure SrBr2 E = 417 kJ/kg and thermal conductivity of 0.38 W/mK. Water uptake from 0.2 to 0.35 g/g.	Cammarata, 2018 [174]
SrBr2	Expanded natural graphite (EG), 6–10% Consolidated composite *	Wet impregnation. Form in tablets (40 × 40 × 10 mm).	1 kWh lab scale system. Temperature of charging and discharging were 80 °C and 35 °C, respectively. E = 189 kWh/m3. Sorption heat storage is about 59% of theoretical value. Heat storage efficiency of 91%.	Zhao, 2016 [219]
LiCl, 42%	Highly porous silica, HPS (with large pore volume)	HPS synthesized. Vacuum impregnation method.	Water adsorption of 0.53 g/g. Energy storage of 1.4 kJ/g. Very stable operation during hydration/dehydration.	Cherpakova, 2025 [220]
LiCl, 28–35.4%	Mesoporous silica gel	Dry impregnation. Drying at 160 °C for 24 h	The maximum E = 1200 J/g (of dry sample) for 35% LiCl-composite (0.65 GJ/m3). Leakage can occur in case of 35% LiCl which limits practical applications.	Frazzica, 2020 [221]
LiCl	Expanded Vermiculite	Wet impregnation	Specified for seasonal heat storage. E = 224 kW h/m3 (0.81 GJ/m3) and 253 kW h/m3 (0.91 GJ/m3) for seasonal and daily seasonal heat storage cycles at the charging temperature in the range of 75–85 °C.	Grekova, 2017 [177]
LiCl, 93.6%	EG Consolidated composite *.	Wet impregnation	E = 65.3 kWh/m3 (0.235 GJ/m3) charging at 85 °C and discharging at 40 °C. Experimental closed system reactor 10 kWh. Heat storage efficiency approximately 94%. Thermal conductivity 2.25 W/mK, bulk density of 510 kg/m3.	Zhao, 2016 [222]
LiCl, 42–53%	Carbon Nano-tubes (CNTs)	Wet impregnation, dried at 160 °C for 12 h.	For 42% LiCl-CNT, E = 470 Wh/kg at 35 °C and water vapor pressure of 0.87 kPa. Desorption at 75 °C.	Grekova, 2016 [172]
LiCl, 5–20%	Expanded Vermiculite (EV)	Wet impregnation in 5–40% LiCl solution for 48 h. Dried at 120 °C.	Sorption at 30 °C and 60% RH, EV-LiCl20 selected as optimal composite with water adsorption of 1.41 g/g and E = 171.6 kWh/m3. Maximum threshold concentration in composite is 32.6% or leakage may occur.	Zhang, 2016 [223]
LiCl, 6.5–43.6%	Silica gel, 2–3 nm (type A) and 8–10 nm (type C)	Wet impregnation in 10–40% LiCl for 12 h at 25 °C. Vacuum filtration. Dried at 120 °C.	Compete water desorption at temperatures from 60 to 100 °C. The most appropriate composite was 30% LiCl and heat storage density = 163.6 kWh/m3 (of composite) and 89.3 kWh/m3 of the system. Adsorption at 30 °C and desorption at 80 °C.	Yu, 2014 [224]
K2CO3, 76–95%	EG and octylphenol plyoxyethylene (10) ether—OP-10. Consolidated composite *	Wet impregnation in 50% K2CO3. OP-10 used as surfactant. Dried at 150 °C for 12 h. Molded to tablets.	Thermal conductivities of the composites increased by 9–15 times in comparison with pure salt (0.09 W/mK vs. 2.36 W/mK). E = 138–216 kWh/m3. Water uptake up to 0.18 g/g.	Zhao, 2022 [225]
K2CO3, 69%	EV	Wet impregnation in 53% K2CO3 solution, then dried under vacuum. Dried at 160 °C.	E = 0.9 GJ/m3 with deliquescence. Hydration rate higher than pure salt. Very good cycling stability (stable for 47 hydration/dehydration tests) Hydration at 40 °C (for 9 h) and dehydration at 120 °C (for 3 h). Water adsorption of 0.4–1.5 g/g at 30–50 °C and pH2O of 22–117 mbar.	Shkatulov, 2020 [186]
Na2S2O3, 62% and 94%	MOF, MIL-101 (Cr)	Synthesis of MOF. Then, wet impregnation in 10% and 30% Na2S2O3.	E = 1099.5 J/g (dehydration) at 120 °C for 61.8% Na2S2O3. Water uptake: 0.04–0.05 g/g at 30 °C and pH2O of 1.27 kPa. Good cyclic stability (9 cycles).	Padamurthy, 2022 [190]
Na3PO4 30–80%	AC	Melt impregnation	E = 1793 J/cm3, stable after 20 cyclic tests.	Lin, 2024 [226]

E—energy storage density (kWh/m³ or GJ/m³) or enthalpy of hydration/dehydration reaction (given in J/g), EG—expanded graphite, EV—expanded vermiculite, AC—activated carbon, MOF—metal organic framework, p_H2O—water vapor pressure, RH—relative humidity. *—For more details about the consolidated composite, please see Section 4.3.

shows some of the most recent and some older studies on composites using salt hydrates. It is evident that, given the extensive number of publications on composites, it was impossible to include all the latest research. Other review papers on composites, their properties, and prices were summarized by Liu et al. [197], Hua et al. [198], Yang et al. [90], and Zbair and Bennici [89]. Interesting material optimization for salt hydrates was presented by Zhao et al. [199]. Selection and synthesis of porous matrices for open TCES systems were described in the paper by Casey et al. [180]. A summary of the composites used at the reactor level can be found in the paper by Lin et al. [200].

In SPMC, the salt hydrates are distributed throughout the many pores. In general, the more salt in the support, the better. A higher salt concentration can result in greater water absorption (and higher energy density), but it can also decrease mass transfer efficiency (pores are clogged) [227]. Increasing the salt concentration decreases the BET surface area and pore volume. There is a maximum allowable salt content to ensure that the resulting salt solution occupies less volume than the pore space, preventing the solution from leaking out of the pores [157,159]. Moreover, higher salt content can decrease the heat conductivity of the composite [228].

The distribution of salt within the matrix is influenced by various factors, including the salt’s tendency for adsorption on the porous surface, the salt’s concentration in the solution, the viscosity of the solution, and the conditions under which drying occurs [158]. Research conducted by Miao et al. [173] has shown that increasing MgSO₄ content in the EG porous matrix reduced the area of the sorbent. However, the heat of reaction increased with the salt concentration. The most optimized salt content in the composite was 60% [173]. In this case, the salt was uniformly dispersed in the matrix pores.

The common methods for preparation of SPMCs are generally based on impregnation techniques that ensure uniform distribution of the salt throughout the pore structure of the host. The sorption properties of SPMCs are affected by the conditions at the impregnation of the porous matrix. In a typical procedure, known as the wet impregnation method (Figure 10), the porous support is first dried (at a temperature range of 200–300 °C and for about 8–12 h) and then immersed in a saturated or concentrated solution of the desired salt [142,194]. This process ensures complete penetration of the salt solution into the matrix. Subsequent drying or thermal treatment removes the salt solution, leaving the salt dispersed within the pores. Sometimes agitation or vacuum is used during impregnation. Furthermore, some modification of the process involves vacuum freeze-drying of the filtered matrix (after wet impregnation). According to the authors [163], this method reduces salt agglomeration and prevents salt from reaching the surface of the material.

Alternatively, melt impregnation involves heating the salt above its melting point and then infusing it into the porous matrix, where it solidifies upon cooling [187]. These processes are optimized to maintain the integrity of the matrix while achieving high loading of the salt.

The dry impregnation method (Figure 11), also called incipient wetness impregnation, is based on the addition of a salt solution in an amount equal to the pore volume of the matrix (volume of the solution is just enough to fill the pore volume of the support without exceeding it, achieving a “wet paste” consistency). In this case, the solution is only inside the pores, and the filtration step is omitted [158,229].

A more advanced method, particularly for MOF-based composites, is the spray-drying method. This process allows the uniform embedding of salt hydrates in porous supports. In the spray-drying method, a slurry containing the salt and support material is atomized into fine droplets and rapidly dried in a hot air chamber to form spherical, dry composite particles. This technique ensures good salt dispersion, controlled particle size, and compatibility with large-scale production. It also improves water vapor accessibility and cycle stability of the composite. However, care must be taken to control the drying temperature to avoid salt degradation or premature phase transitions. Spray-dried composites, such as salt/silica materials, have shown promising performance in TCES applications. Moreover, spray-drying is a practical technique to produce encapsulated salts within porous supports for TCES applications [90,231]. When the saturated salt solution is sprayed into the porous framework, forming a core–shell composite where the salt is the core and the framework acts as the shell, this is an encapsulation process.

In addition, the sol–gel approach is employed to prepare SPMCs. In the sol–gel process, a solution is converted to a solid gel phase. Typically, metal alkoxides (such as tetraethyl orthosilicate, TEOS) are hydrolyzed and polymerized in the presence of a salt (such as CaCl₂), forming a silica network that entraps or hosts the salt in its pores. This process produces SPMCs with an aero-gel structure that have a remarkably high sorption capacity and pore volume. However, this method is not widely used because the required equipment and reagents are expensive [158,210].

Foam synthesis [210] has emerged as a promising technique for developing porous matrices suitable for thermal energy storage applications, including TCES. This approach involves the creation of highly porous, interconnected structures that allow for efficient vapor transport and uniform salt distribution. One reported method [210] uses a polymer with high water vapor permeability to form a silicone-based foam. Salts such as CaCl₂ or MgSO₄ are mixed with the siloxane matrix to form a homogeneous slurry, which is then subjected to a controlled foaming reaction in an oven to produce the final composite. These salt-impregnated foams benefit from high porosity, low density, and enhanced water vapor diffusivity, which contribute to improved hydration–dehydration kinetics [232].

It is important to note that, in the SPMCs, the different pore structures of the matrices result in quite different salt loading patterns, and, consequently, in large differences in salt loading capacity, water sorption, and heat storage.

Mesopores (2 nm < pore size < 50 nm) increase vapor diffusivity, allowing for faster hydration and dehydration reactions. This is essential for effective heat transfer in TCES systems [89,168]. Micropores (pore size < 2 nm) provide additional sorption sites, increasing the overall energy storage capacity of the material [89,168,169]. Larger pore volume (macropores > 50 nm) allows for improved dispersion of salt within the matrix and higher salt loading, reducing agglomeration and increasing the uniformity of hydration and dehydration processes. This uniformity plays a key role in stable cycling performance.

Research by Ponomarenko et al. and Glaznev et al. [233,234] has shown that the size of the pores in the support matrix also plays a critical role in determining the hydration and dehydration behavior of the embedded salts. Results have shown that, when salts such as CaCl₂ are confined in smaller mesopores (e.g., 8.1 nm vs. 11.8 nm in SBA-15), the formation of CaCl₂·2H₂O occurs at significantly lower water vapor pressures compared to pure salt [233,234]. This is due to the higher surface-to-volume ratio and capillary water condensation. Consequently, smaller pores facilitate hydration at lower humidity levels and allow operation at lower evaporator temperatures in adsorption cycles. However, smaller pores also increase the strength of the water bond, resulting in higher desorption temperatures. On the other hand, if the pores are too small (3–8 nm), the sorption capacity may decrease due to disruptions in the solvation shell of the salt ions caused by interactions with the pore walls. Conversely, in wider pores (8–15 nm), the behavior of the trapped salt more closely resembles that of the bulk material [233,234].

Posern et al. [104] have also investigated the influence of porous host materials (various porous glasses) using MgSO₄ as a salt hydrate. In VitraPOR P5 glass frits (median pore diameter of 1.7 µm), the total enthalpy of the hydration reaction, leading to MgSO₄·6H₂O, was higher than in the bulk sample of lower MgSO₄ hydrates. The use of a matrix with smaller pores, i.e., below 200 nm, eliminated the kinetic hindrance of MgSO₄ hydration. The authors pointed out that the total enthalpy of reaction increases with decreasing pore size and correlates with the vapor sorption capacity of the porous material. As the pore size decreases, the contribution of the hydration reaction to the total heat of reaction becomes less significant, while the contribution related to adsorption effects becomes more dominant. For example, when the material with the smallest pores was used, the enthalpy of adsorption and hydration accounted for 56% and 44%, respectively, of the total reaction enthalpy. Moreover, the results showed that the volumetric energy densities also increased with decreasing pore size. In these studies [104], the higher desorption temperature was also observed when smaller pore size materials were used. It is interesting to note that hydration is less sensitive to pore size as long as the pores are small enough to overcome the kinetic barrier. The level of pore filling appears to be a considerably more important factor that should be optimized to achieve high storage densities through hydration reactions.

In a study, Zhang et al. [235] developed and characterized two composite sorbents, LiCl/SG30 (silica gel matrix, mesoporous structure) and LiCl/EV45 (expanded vermiculite matrix, macroporous structure), for adsorption desalination applications, demonstrating distinct sorption behaviors compared to standard microporous Siogel (a type of silica gel). The sorbents were evaluated for their sorption performance and mass transfer characteristics, highlighting a trade-off between capacity and kinetics due to differences in pore structure. LiCl/EVM45 showed the highest sorption capacity (2.4 g/g), while the silica-based composite exhibited faster sorption kinetics. The studies have shown that the sorption/desorption process in LiCl/SG30 is the combination of physisorption, chemisorption, and solution absorption. LiCl/EVM45 primarily exhibits the sorption characteristics of the incorporated LiCl crystals, given that the macroporous vermiculite exhibits an insignificant capacity for physisorption. Both composites outperformed standard silica gel in predicted specific daily water production, with LiCl/SG30 reaching up to 60 m³/ton/day, indicating strong potential for adsorption desalination applications.

The novel study of Eberbach et al. examined the influence of the porous matrix’s pore size on the hydration/dehydration of CaCl₂ [236]. As a porous matrix, various clay materials (such as vermiculite, halloysite, and sepiolite) and silica gels with different pore sizes were used. The results showed that the formation of CaCl_2⋅1/3H₂O during hydration is hindered in smaller pores (less than 15 nm average pore diameter) compared to larger pores or bulk CaCl₂. This is attributed to its larger unit cell volume compared to monohydrate (CaCl₂⋅1H₂O), making it more difficult to form in the confined space of small pores. In silica gel composites (11 nm and 6 nm average pore diameters) and sepiolite (8 nm), direct hydration from anhydrate to dihydrate (CaCl₂⋅2H₂O) occurs (CaCl₂·1/3H₂O phase is not detected). However, although different hydration steps may occur in different pore sizes, the onset temperatures for the same hydration transitions in this study appear to be less affected by either the matrix or the pore size.

Considering the studies discussed, it is clear that optimization of pore size is essential to balance sorption efficiency, operating temperature range, and overall system performance in thermochemical energy storage applications.

The use of composites is also very common in the case of hygroscopic salts, such as CaCl₂ or MgCl₂, which have a tendency to deliquesce. Experimental studies by Gao et al. [168] have shown that the use of SPMCs allows greater moisture absorption than pure salts. Therefore, it is easier to reach a state of equilibrium, i.e., maximum hydration. The gross water uptake was 1.6–2.0 times higher than the pure chemisorption capacity when EG-CaCl₂ and AC-CaCl₂ composites were used [168]. Generally, hygroscopic salts embedded in the porous structure can be loaded more with water, allowing recovery of more hydration energy while not causing salt leakage [112].

Moreover, as demonstrated in the research conducted by Gao et al. [204], the water absorption of EG/CaCl₂ composites comprises chemically bound water (resulting from the hydration) and water absorbed by deliquescence of the salt. The same authors emphasize that the role of solution absorption is nearly as significant as that of chemical sorption, since it affects both the capacity for water adsorption and the ability to store thermal energy.

At levels above the DRH, the salt dissolves in the adsorbed water and is subsequently absorbed into the porous matrix. Although exceeding the DRH allows for greater water absorption, thereby increasing the amount of heat provided by the reaction, the formation of salt solution can degrade the porous matrix [209].

As shown in this section, a considerable number of studies have been conducted on salt-in-matrix composites for thermochemical energy storage (see Table 10). The most commonly used porous materials are silica gel, zeolite 13X, vermiculite, and carbon materials, such as activated carbon or expanded graphite. A substantial proportion of these studies were conducted at the laboratory scale, comprising hydration and dehydration tests performed separately in various types of apparatus. This configuration complicates the direct comparison of results, particularly given the variation in water vapor pressures and temperatures. Nevertheless, research often targets conditions relevant to seasonal storage, such as air temperatures and humidity levels typical of summer and winter (necessary for charging and discharging a seasonal TCES system). In most cases, dehydration is carried out under constant conditions. The research studies reviewed focus on key parameters such as water uptake, energy released during hydration as a function of vapor pressure and temperature, and changes in sorbent structure. Commonly used salts in these studies include MgSO₄, CaCl₂, MgCl₂, SrCl_2, and SrBr₂, selected primarily for their high hydration enthalpy, low dehydration temperature, and availability. A major research focus is to optimize the amount of salt embedded in the matrix; as mentioned above, the salt loading must be compatible with the porosity and structure of the support. Despite the reduction in surface area due to salt incorporation, the composites show significantly improved water sorption capacity and thus energy density. In addition, the use of carbon-based materials such as graphite, expanded graphite, or carbon nanotubes improves thermal conductivity and helps mitigate problems such as deliquescence or salt leakage.

In the next two Section 4.2 and Section 4.3, the focus shifts to other strategies aimed at further improving the performance of salt hydrate-based materials. These include the use of mixed salt systems, where two or more different salts are combined to tailor sorption and hydration properties, and the incorporation of additives to improve thermal conductivity, hydration/dehydration rates, and stability.

4.2. Mixed Salts Composites

Another promising strategy to improve the performance of TCES systems is the use of mixed salt systems, typically binary salt combinations. While single salt hydrates are extensively studied, their limitations—such as narrow operating temperature ranges, low energy density, or deliquescence—can often be overcome by mixing different salts. The resulting mixtures can exhibit improved hydration/dehydration behavior, modified thermodynamic properties, and enhanced cycling stability. This section discusses the rationale for using mixed salt systems, common combinations explored in the literature, and the effects of salt pairing on key TCES performance parameters.

The example of binary salt systems is the mixing of sulfate and chloride [89,237]. Due to the high deliquescence of chloride (e.g., CaCl₂ or MgCl₂), it is easily overhydrated to form a solution in the hydration reaction, which significantly reduces the stability of the hydration reaction. Sulfate, e.g., MgSO₄, on the other hand, undergoes an incomplete reaction due to poor reaction kinetics and insufficient water vapor transfer [104]. To produce a higher hydrate state, the dehydrated sulfate can be partially dissolved in a solution of the chloride-hydrated salt by mixing these two salts (sulfate and chloride). This improves the hydration kinetics of the sulfate salt and avoids the instability of excessive chloride hydration. It is necessary to combine a mixture with a high sulfate ratio and a low desorption temperature to prevent chloride corrosion of the reactor.

Posern and Kaps [237] investigated the salt content ratio in a composition consisting of MgSO₄, MgCl_2, and attapulgite as a porous matrix. In this binary salt system, MgSO₄ has a high 90% DRH, and MgCl₂ has only 33% DRH (at 30 °C). The results have shown that substitution of some MgSO₄ by MgCl₂ will result in higher heat of sorption due to better water absorption in the created MgCl₂ solution. The measured heat of sorption was 1590 kJ/kg (at 30 °C and 85% RH) for a mixture of salts of 20% MgSO₄ and 80% MgCl₂ embedded in attapulgite. Dehydration tests were performed at 130 °C.

Zbair and co-authors [238] also studied the binary system of MgSO₄/MgCl₂ embedded on porous carbon. It was shown that this binary system has faster hydration kinetics and higher water sorption compared to the mono-salt MgSO₄. In this case, MgCl₂ facilitates the diffusion of water vapor and modifies the hydration of MgSO₄. Binary composite (48% MgSO₄/12% MgCl₂) achieved water uptake up to 0.75 g/g. The heat release of this composite was 1840 J/g, which was 70% of the heat release of pure mixed salts.

Gordeeva and colleagues in 2009 described a new “tool”, embedding in the matrix two salts that affect each other [229]. In this regard, the authors studied the fact that a mixture of LiCl-LiBr in silica gel matrix forms a solid solution (these salts have good solubility in each other) and, thus, distortion of the crystal lattice can occur. The specific type of solid solution formed (homogeneous solid solution enriched with LiCl (SS_Cl) or with LiBr (SS_Br), or a mixture of both) depends on the molar ratio of LiCl to LiBr. Importantly, the formation of these solid solutions changes the temperature at which the salts absorb water vapor.

For LiBr-rich composites (e.g., LiCl + LiBr with molar ratios of 1:3 and 1:1), the hydration temperature of LiBr (to monohydrate) decreases by about 5–10 K compared to the single salt LiBr composite [229]. This is attributed to the formation of a solid solution of LiCl in LiBr (SS_Br), which is accompanied by a narrowing of the crystal lattice spacing parameter. For the most LiBr-rich composite (1:6 ratio), the water sorption behavior is very similar to the sum of the individual salts. For LiCl-rich composites (e.g., LiCl + LiBr with ratios of 6:1 and 3:1), the hydration temperature of LiCl increases by 10–20 K. This occurs even with a smaller amount of LiBr dissolved in the LiCl lattice and is thought to be related to a broadening of the crystal lattice spacing parameter, which may facilitate water vapor uptake. Therefore, the temperature at which water vapor is absorbed can be adjusted by changing the ratio of LiCl to LiBr in the composite [229].

Other studies by Gordeeva and co-authors [239] on binary salt systems (LiCl/LiBr, CaCl₂/CaBr₂, and BaCl₂/BaBr₂) incorporated into porous materials such as silica and vermiculite also reveal that the formation of solid salt solutions shifts the water sorption equilibrium. Bromides, which have a higher affinity for water than chlorides, raise the equilibrium temperature and lower the water vapor pressure required for sorption. In contrast, the addition of chlorides decreases the equilibrium temperature and increases the vapor pressure. This allows the hydration behavior to be tailored to better suit specific thermal storage applications.

This effect, described by Gordeeva et al. in connection with solid solutions, was also studied by Grekova et al. [240] using binary salt composites of CaCl₂/CaBr₂ and silica gel as a porous matrix. Grekova et al. described that solid solutions of CaCl₂ in CaBr₂ form complexes with water at a pressure that depends on the molar ratio of the salts used [240]. The authors argue that the dissolution of CaBr₂ in the crystalline lattice of CaCl₂ leads to a broadening of the spacing parameter, which probably favors the incorporation of water molecules in the lattice and the decrease of the equilibrium pressure of solvate formation from the salt, so that also in this case, depending on the ratio, the sorbent can be optimized to specific conditions (e.g., RH of air in hot climates).

Chen and co-authors investigated LiCl/LiBr and 3A zeolite composites [160]. Salt concentrations of 5–25% (Z5-Z25) for wet impregnation and different LiCl/LiBr mass ratios were used. In these studies, it was observed that higher salt loading than Z15 decreased the heat storage densities, probably due to pore blockage and reduced mass transfer. The Z15 sample (i.e., prepared from a 15% salt solution and LiCl/LiB ratio of 1:3) exhibited both a high water adsorption rate and capacity and was tested in a laboratory-scale reactor using 3 kg of composite. The heat storage density was 434 J/g at 70% RH. Unfortunately, studies on the influence of different salt ratios were not performed.

Another example is the blend of two chlorides, namely, the SrCl₂ has a low adsorption capacity, while the CaCl₂ has a high adsorption capacity but is prone to deliquescence and has a limited cycle operation. Hu et al. in their novel research [241] described the synergistic effect between SrCl₂ and CaCl₂ in expanded perlite as a porous support. Binary salt composites with a mass ratio of SrCl₂ to CaCl₂ greater than 1:1 showed improved overall performance, including better thermal storage and cycle stability, compared to single salt composites. Furthermore, the addition of CaCl₂ was shown to enhance the hydration of SrCl₂, although this effect can be positive or negative depending on the ratio. When the CaCl₂ content is low, the effect is positive and significantly promotes the adsorption capacity of the composite (due to the hygroscopic nature of CaCl₂). At higher CaCl₂ content, the effect is negative. The optimum thermal energy storage performance was achieved with a mass ratio of 2:1 (SrCl₂:CaCl₂) and a salt impregnation concentration of 30%. This composite had a maximum water adsorption capacity of 1.06 g/g at 20 °C and 80% RH. This represented an 82.76% increase in adsorption capacity over the single SrCl₂ salt adsorbent under the same conditions. The heat storage density was found to be 1273 kJ/kg. This was 47.17% higher than the single SrCl₂ salt adsorbent. Moreover, the optimal composite (2:1 ratio) also demonstrated superior cycling stability over 20 cycles, with minimal deliquescence, agglomeration, and structural damage compared to the single-salt composites.

Rammelberg et al. [242] investigated the binary system of CaCl₂/MgCl₂, showed excellent cycling performance under all conditions evaluated, and exhibited improved kinetic properties, even in the absence of a support material. It maintained nearly constant performance over 55 hydration/dehydration cycles and showed faster and greater water sorption than the pure salts. There is evidence to suggest the formation of tachydrite (CaMg₂Cl₆⋅12H₂O) during the cycling operation of the CaCl₂/MgCl₂ blend. Conversely, MgCl₂/MgSO₄ showed poor cycle stability with respect to initial water content, with a significant decrease in normalized water absorption and hydration enthalpy.

Interestingly, the results presented by Liu et al. [243] highlight the potential of using binary salt composites in powder form for both fluidized bed and thermal storage systems. The composite of MgCl₂/MgSO₄/CMS (commercial mesoporous silica) effectively combines the strengths of its individual components: it inherits the rapid reaction rates, high water sorption capacity, and significant temperature rise of the MgCl₂/CMS system, while also benefiting from the improved particle dispersion and minimized agglomeration typical of the MgSO₄/CMS composite. Experiments conducted at the reactor scale showed that the MgCl₂/MgSO₄/CMS composite, with a 1:1 salt ratio, could be fluidized at relatively low gas flow velocities of around 0.01 m/s. Under hydration conditions of 80% relative humidity and 30 °C, the material achieved a temperature increase of up to 24.7 °C and delivered an energy density of 1018 kJ/kg. In the same paper, the authors tested a binary mixture of MgBr₂/MgSO₄/CMS, containing 50% salts with a 1:1 ratio [243]. The results showed that the MgBr₂/MgSO₄/CMS composite had better hydration kinetics than the MgSO₄/CMS composite and achieved higher water uptake than the MgBr₂/CMS composite.

Wang et al. [244] studied the binary system of MgSO₄/LiCl/Zeolite 13X (porous matrix) for heat storage. In this configuration, LiCl is a highly hygroscopic salt that enhances water vapor absorption. The composite exhibited a significantly faster water sorption rate than the single zeolite 13X/MgSO₄ composite, i.e., 90% of the maximum water uptake (0.24 g/g) was achieved in about 15 min compared to over 30 min for the single salt composite. In addition, the water adsorption capacity was 37.6% higher than that of the single system (MgSO₄/Zeolite 13X), which also resulted in a 41.6% increase in heat storage density (maximum of 458.3 kJ/kg).

Li, Zeng, and Wang [245] studied blend of MgSO₄/SrCl₂ (without any porous matrix). In this research, the addition of chloride salt (SrCl₂) also improved adsorption kinetics compared to pure MgSO₄. The developed composite salt hydrate showed good cyclability, indicating its potential for stable performance over multiple charging and discharging cycles. The blend with 20%MgSO₄ showed superior cycling stability (20 tests), adsorption kinetics, and water uptake compared to the individual salts forming the mixture, reaching a peak water uptake of 0.487 g/g (of dry material). The energy storage density was 595 kWh/m³. However, the MgSO₄ content should remain below 30% to prevent deliquescence under less favorable conditions, such as 30 °C and 55% RH.

Ata Ur Rehmann and colleagues studied the sulfate binary system of MgSO₄/ZnSO₄ [246]. The hydration behavior of this composite was found to be 34% better than pure MgSO₄ and 48% better than pure ZnSO₄. The 90% MgSO4/10% ZnSO₄ composite had a lower dehydration energy (120 °C) and an improved average hydration enthalpy than the pure salts (1422 J/g vs. 882 J/g and 693 J/g for MgSO₄ and ZnSO₄, respectively).

In summary, as shown in

Table 11. Summary of selected tested mixed salts.

Mixed Salts System	Selected Results	Reference
MgSO4/MgCl2/attapulgite	Higher water sorption in the formed MgCl2 solution. Heat of water sorption of 1590 J/g	[237]
MgSO4/MgCl2/porous carbon	Faster hydration kinetics and water sorption than MgSO4. Heat of water sorption of 1840 J/g	[238]
LiCl/LiBr/silica gel	The temperature of water sorption can be adjusted by changing the salt ratio.	[229]
LiCl/LiBr and CaCl2/CaBr2, and BaCl2 and BaBr2	Hydration behavior can be adjusted for specific applications. Bromides increase the equilibrium sorption temperature and lower the water vapor pressure, unlike chlorides.	[239]
CaCl2/CaBr2/silica gel	Equilibrium water pressure can be adjusted (as mentioned above)	[240]
LiCl/LiBr/Zeolite 3A	Heat storage of 434 J/g	[160]
SrCl2/CaCl2/expanded perlite	In 1:1 ratio system showed improved overall efficiency. Higher heat storrage than single salts. In 2:1 ratio, superior cycling stability.	[241]
CaCl2/MgCl2	Very good cycling performance.	[242]
MgCl2/MgSO4/CMS	Better hydration kinetics than MgSO4/CMS and higher water sorption than MgBr2/CMS	[243]
MgSO4/LiCl/zeolite 13 X	Higher water sorption kinetic than single composite (MgSO4/zeolite)	[244]
MgSO4/SrCl2	Good cyclability and better water sorption and kinetics than single salts.	[245]
MgSO4/ZnSO4	Better hydration behavior than individual salts. Lower dehydration temperature.	[246]

, the use of binary salt systems, as explored in these papers with various salt combinations (usually chloride/sulfate systems, e.g., MgSO₄/SrCl₂, MgSO₄/MgCl₂, and MgSO₄/LiCl or halide systems such as LiBr/LiCl or CaCl₂/CaBr₂), consistently demonstrated improved performance in TES by enhancing reaction kinetics, increasing water uptake and heat storage capacity, and improving cycle stability compared to the use of single salts. These improvements are attributed to factors such as altered hydration pathways, improved water vapor uptake through the use of hygroscopic chloride, reduced energy barriers to hydration, and mitigation of problems such as agglomeration or overhydration.

4.3. Consolidated Composites

Consolidated composites represent another approach to TCES materials. By compacting powdered or granular salt hydrate mixtures into a denser form, several key performance indicators critical to practical applications of TCES systems, such as high volumetric energy density, thermal conductivity, and heat and mass transfer within the storage reactor, can be improved.

Compacted salt hydrate composites are formed by applying pressure to a mixture of a salt hydrate and other additives to form a consolidated structure, typically in the form of pellets or blocks. This process reduces the void space between the particles, resulting in a denser material with potentially improved thermo-physical properties. Importantly, compaction improves the mechanical integrity of the composite, reducing problems such as pulverization, attrition, and channeling that can occur with loose granular materials during repeated cycling. This improved stability contributes to better long-term performance and cyclability [247]. In particular, pure salts often lose their structural integrity after just a couple of hydration/dehydration cycles [248]. Finally, compacted materials in the form of pellets or blocks can simplify the design and operation of TCES reactors by allowing for more predictable heat transfer fluid flow patterns through the material bed.

The additives, such as carbon materials and various binders, play a critical role in not only improving the intrinsic properties of the composite but also in enabling and stabilizing the compacted structure [247]. The incorporation of additives aims to modify the physical and chemical properties of TCES materials. Carbon-based materials can significantly enhance thermal conductivity, improving heat transfer within the compacted bed [174,248]. Expanded graphite (EG) is particularly effective due to its high conductivity and ability to form a conductive network within the compact [249]. Porous carbon materials also aid in water vapor diffusion and can improve structural stability. Generally, nanomaterials (e.g., carbon nanotubes and nanofibers, EG, metal oxide nanopowders) are particularly effective due to their unique structural and functional properties and can improve many of the desired properties (thermal conductivity, heat storage density, cyclic stability, etc.) [250,251].

Binders are essential for creating mechanically stable compacts. They hold the particles together, providing sufficient strength to withstand handling and the stresses associated with repeated hydration-dehydration cycles. Example binders include polyvinyl alcohol, carboxymethylcellulose, gelatin, and polyvinyl pyrrolidone [247].

Studies conducted by Akcaoglu et al. showed that the use of both AC and EG helped maintain the shape of the pellets for a longer period of time [248]. Pellets of MgSO₄/EG/AC (1:1) were considered the most promising composite and were able to withstand 10 cycles of hydration/dehydration (see Figure 12). The heat of fusion for this material was 1320 J/g (after 10 cycles). The use of AC or ENG alone was not sufficient to maintain a stable structure of the pellets after dehydration. These studies also showed that higher compression pressure led to a decrease in permeability and porosity but increased thermal conductivity and mechanical strength. Nevertheless, it is worth mentioning that salts such as SrBr₂ and MgSO₄ mixed with AC and EG exhibited better permeability than compacted pure salts. In addition, the addition of AC + EG provided protection against deliquescence.

The strategy of improving the durability of MgSO₄ by forming consolidated composites was also investigated by Lasek and colleagues [252]. The authors investigated various additives such as kaolin, silica gel, bentonite, alginates, magnesium chloride, and process wastes (fly and bottom ash and slag taken from a coal-fired boiler, and gypsum). The highest durability (up to 25 cycles) was observed for the mixture of 90%MgSO₄/5% bottom ash (fraction below 0.2 mm) and 5% bentonite.

A consolidated composite of LiCl and EG was tested in a 10 kWh reactor by Zhao et al. [222]. The results were presented in Table 10 during a discussion on salt in porous matrix composites (SPMCs). The presentation of these results in Section 4.2 is due to the fact that often consolidated hydrate composites are produced first by impregnating the matrix, and then these mixtures, after evaporation of the solvent, are formed into pellets. Therefore, it can be generally said that the consolidated composites are SPMCs, which have additionally been molded. Sometimes the term “consolidated host matrices” is used. For the sake of clarity, however, it is noted in Table 10 that these are consolidated composites.

For example, in the studies by Cammarata et al. [174], the authors describe, in the abstract, that the composites were made of SrBr₂·6H₂O and natural graphite, which act as a supporting matrix. Further analysis of the paper shows that this composite was produced in the form of consolidated tablets (as shown in Figure 13). The results have shown that the use of graphite greatly improves the thermal conductivity of the composite. In addition, the hydration/dehydration kinetics were improved. Further details of this research are presented in Table 10.

Yu and colleagues studied the consolidated composite based on LiCl and activated carbon (AC, coal-based) as a porous matrix [253]. Expanded natural graphite treated with sulfuric acid (ENG-TSA) was used as an additive to improve heat transfer. Silica solution (SS) with a concentration of about 30% was used as a binder. The appropriate mass ratio between ENG-TSA and silica solution was found to be 1:4 (1:1.2 in dried composite). AC was impregnated in 40% LiCl solution (ALi40). Different mass ratios of ALi40 and ENG-TSA were investigated. The sample with an ALi40:ENG-TSA:SS mass ratio of 6:1:1.2 and bulk density of 758 kg/m³ was a suitable composite for the LiCl·H₂O system, considering both volumetric storage density and kinetic performance. The thermal conductivity of the composites was in the range of 2.0–2.83 W/mK, depending on the bulk densities (462–820 kg/m³).

The interesting consolidated composite based on MgSO₄ and RTEG (a room temperature expanded graphite) as a porous additive was studied by Salama et al. [70]. RTEG was treated with sulfuric acid (H₂SO₄) and ammonium persulfate. The method of preparing this composite is shown in Figure 14. This study focuses on the overall performance of the composite system under different operating conditions (charging/discharging temperatures); the effect of the additive in a comparative study with pure MgSO₄ or different composite compositions was not studied.

Li et al. [254] investigated the effect of hygroscopic additives such as poly(sodium acrylate), 13X zeolite, and nano-Al₂O₃ on the thermal storage performance of MgSO₄. Through hydrothermal treatment, bulk MgSO₄⋅6H₂O was transformed into nanoparticles (200–500 nm) when composited with poly(sodium acrylate), 13X-zeolite, and nano-Al₂O₃. The nano-Al₂O₃ modified composite material exhibited an activation energy of 28.5 kJ/mol and an energy density of 1305 kJ kg. The poly(sodium acrylate)-modified composite material showed a good heat storage energy density (1100 kJ/kg) and the lowest activation energy (22.3 kJ/mol). The investigated reactor design using these composite materials could emit heat at about 50 °C and be charged in the temperature range of 100–200 °C.

Interestingly, a very stable composite consisting of two salts, Al₂(SO₄)₃·18H₂O and FeSO₄·7H₂O, and carboxymethylcellulose (CMC, binder) and carbon nanotubes (CNTs), was prepared and investigated by Zhou et al. [255]. CNTs strongly improved the structural stability (over 100 cycles) and thermal conductivity of the composite. Without the CNTs, the samples expanded strongly after dehydration, which was unfavorable. In addition, at 200 °C, CMC by melting can connect the interfaces between CNTs and the hydrate, resulting in increased thermal conductivity. The operating temperature was 118 °C, and the absorbed heat was 420.4 J/g.

In Reynolds et al. studies [256], CaCl₂-based composites formed into spherical beads of 6 mm diameter were prepared using sodium alginate and expanded graphite (EG). The CaCl₂ content was up to 84%, and the bulk densities ranged from 0.37 to 0.55 g/cm³. The theoretical storage density was in the range of 0.21–0.37 GJ/m³. Furthermore, the results showed that the prepared composite materials can achieve better packing density and comparable energy density compared to the conventional vermiculite/CaCl₂ composite.

The research of Salviati et al. [257] describes the preparation of a novel composite material based on SrBr₂, EG, and using nanocellulose or polydiallyldimethylammonium chloride as a binder. Although not a typical consolidated composite, it contains additives such as binders and the commonly used graphite to improve thermal conductivity. It was found that polymeric binders improved salt distribution over the graphite lamellae, thus providing mechanical stabilization of the prepared composite. In addition, the use of nanocellulose allowed the freeze-drying process to be used to produce the foam structure. The energy storage density was up to 764 kJ/kg, and the nanocellulose-based composite increased the hydration kinetics compared to pure salt.

Fayazmanesh and colleagues [258] report on a composite material made of graphite flakes (or copper powder), silica gel, binder, and CaCl₂. The binders used were polyvinylopyrrolidone with different molecular weights (10,000—PVP 10 and 40,000—PVP 40, respectively). Impressively, the water uptake of the silica gel/CaCl₂/PVP40 composite was stable over 300 pressure cycling cycles.

More information about the stability of salt hydrates used in TCES can be found in a paper by Aarts and co-authors [116,259]. A compilation of sorbent consolidated composites (mainly based on physical sorption) and binders was compiled by Rocky et al. [247].

In summary, consolidated composites are a key area of research in TCES systems, primarily aimed at improving material properties and system performance. Based on the reviewed papers, these composites are typically formed into dense structures such as pellets or blocks. The main motivations for consolidating salt hydrates include increasing the volumetric energy storage capacity and improving the structural stability to withstand repeated hydration/dehydration cycles, which can cause expansion and cracking in pure salt hydrates. Various salts, such as LiCl, MgSO₄, CaCl₂, and SrBr₂, are used to form these composites in combination with porous host matrices and additives. Commonly mentioned matrix materials and additives are AC, EG, CNTs, and silica gel. Binders such as silica solution or other polymers (PVP, alginate) are often incorporated to provide mechanical strength to the consolidated structure. The incorporation of thermally conductive additives such as graphite, EG, or CNTs has been shown to significantly improve the effective thermal conductivity of the composite blocks or pellets, allowing faster heat transfer during charging and discharging. While consolidation generally improves thermal conductivity and bulk density, it can sometimes negatively affect mass transfer (sorption kinetics) by reducing porosity and increasing diffusion resistance, highlighting a trade-off that needs to be optimized through material selection and preparation methods. Overall, consolidated composites represent a promising approach to developing durable and efficient TCES materials suitable for practical applications.

5. Overview of TCES Concepts

5.1. TCES System Configuration

TCES systems can be classified as either open or closed [75] (as shown in Figure 15). In an open system, both mass and energy exchange take place with the surroundings. The basic idea of the system is to use a blower to pass sufficiently humid air through a layer of sorption material during discharging. The heat of reaction is transferred to the flowing air. During charging, hot, dry air is passed through the hydrated salt, causing it to dehydrate. The primary drawbacks include the need for a blower to push the humid air through the reactor; the potential need for a humidifier to achieve the desired water vapor pressure, and the limitation of the temperature differential across the reactor imposed by the thermal mass of the airflow, requiring a heat recovery unit to produce temperatures suitable for space heating and domestic hot water production. In addition, the mass exchange with the environment eliminates the use of hazardous materials. The main advantage of open systems is their simplicity, lack of pressurization, effective, and controlled heat transfer, and fewer components compared to closed systems. Moreover, this system provides better heat and mass transfer since it usually works under 1 atm, and flowing air increases the heat transfer coefficient [75].

In contrast to open systems, in closed systems, the water vapor circulates in a closed loop and is never exchanged with the outside environment. In this system, when the reactor is charged, heat is transferred (via heat transfer fluid, HTF) to the salt hydrate, and the released water vapor is directed to the second tank (condenser/evaporator), where it is condensed. During the discharging process, heat is supplied to the evaporator, and the released water vapor is directed to the salt. The heat of reaction is transferred to the HTF. These systems usually work under low pressure.

This configuration enables precise regulation of parameters such as pressure and humidity. Moreover, harmful compounds as storage material can be used because they work in a closed loop. While closed systems are well-suited for compact, mobile, or high-precision applications, they necessitate additional components such as condensers, evaporators, and vacuum-tight enclosures, which can lead to increased cost and complexity. Moreover, the closed system requires periodic evacuation due to the accumulation of incondensable gases, which impedes the flow of sorbate to the condenser [114,260].

The selection of one over the other depends on the target application, environmental constraints, and desired control over operating conditions.

A comparative summary of the open and closed TCES sorption systems is presented in

Table 12. Comparative summary of the open and closed TCES sorption [261].

Criterion	Closed System	Open System
System Configuration	Requires sealed, pressure-tight components	Operates with humid air in open environment, non-pressure
Steam Source	Generated internally via evaporator during hydration reaction	Supplied by ambient moist air
Heat and Mass Transfer	Requires internal heat exchanger to manage the heat of reaction Heat exchange is the main limiting factor	No internal heat exchanger HTF and reactive gas in single flow Mass transfer is the limiting mechanism
Thermal Management	Complex due to need for evaporator and heat exchanger	Simpler thermal configuration; higher thermal power due to absence of evaporator and heat exchanger
Reactor design, manufacturing and cost	Demands precise engineering for thermal control Strong technological and manufacturing constraints, especially for reactor and evaporator/condenser Needs evaporator/condenser systems Higher production cost	Depends on material permeability after reaction for bed dimensioning Simplified manufacturing and system design Requires blower to circulate humid air Lower production cost
Applications	Well-suited for compact, mobile or controlled environments requiring high thermal precision applications	Suitable for long-term storage in domestic and domestic hot water production applications

.

5.2. Reactor Design in TCES System

Reactor design is a critical factor in the performance of salt hydrate TCES systems. Key challenges in the field of reactor design include the following [262]: Firstly, the occurrence of non-uniform reactions can be attributed to uneven temperature and water vapor distribution, which can result in incomplete reactions and diminished efficiency. Secondly, liquefaction, defined as the formation of liquid phases due to deliquescence or melting, can impede mass transfer and hinder system operation. Thirdly, limitations in heat and mass transfer can reduce the overall reaction rate and power output of the system. Fourthly, excessive pressure drops across the reactor can reduce efficiency and increase energy consumption. Nevertheless, improving heat and mass exchange appears to be essential to maximize the benefits of salt hydrates used as TCES materials.

Gas–solid reactors play a key role in thermochemical energy storage systems. Depending on the mode of contact between the gas phase and the solid material, three different reactor types are distinguished in the literature [88,263]:

• Fixed bed reactors (packed beds) involve a stationary arrangement of solid particles through which the reactive gas flows. These systems are relatively simple in design and operation, but suffer from limited heat and mass transfer, especially in larger-scale units. When efficient thermal exchange is critical, this limitation significantly constrains their applicability [264].
• Moving bed reactors allow for periodic or continuous removal of the solid phase. While the flow characteristics of the gas phase resemble those in fixed beds, the capacity to exchange or regenerate solids introduces operational flexibility. This is particularly beneficial in processes requiring cyclic material replacement or thermal regeneration.
• Fluidized bed reactors utilize fine solid particles suspended by an upward gas stream. When the gas velocity exceeds the minimum fluidization threshold but remains below the entrainment limit, the solid phase achieves a dynamic, fluid-like state. This configuration offers superior heat and mass transfer performance, uniform temperature distribution, and enhanced reaction kinetics, making it particularly advantageous for large-scale or highly exothermic/endothermic processes.

In addition to the standard configurations mentioned above, rotary reactors have been developed as an alternative approach for specific thermochemical processes [265]. These systems offer continuous operation and enhanced mixing capabilities, albeit at the cost of increased mechanical complexity.

From a design perspective, the selection of a suitable gas–solid reactor depends on several interrelated factors: the intrinsic reaction kinetics of the solid particles, the particle size distribution, and the fluid–solid hydrodynamics within the reactor. The physical properties of the reactants and products—such as density, porosity, reactivity, and thermal conductivity—also significantly influence reactor choice and performance [88].

The typical configurations for reactors in gas–solid sorption-based heat storage systems include packed beds; however, various reactor designs have been investigated to address critical issues such as increasing heat capacity, reducing heat losses, and improving heat and mass transfer. The advantages and disadvantages of typical reactors used in TCES are shown in

Table 13. Comparative assessment of reactor types used in TCES—based on [88] and own study.

Reactor Type	Advantages	Disadvantages
Fixed Bed/Packed Bed	Simplified design and operation. Facilitates mathematical modeling due to well-defined flow characteristics.	Limited heat and mass transfer efficiency, especially in large-scale applications. Significant pressure drop across the bed.
Moving Bed	Direct heat exchange between the gas and solid phases. Allows continuous or periodic solids flow for regeneration.	Complicated flow dynamics and reactor control. Scale-up and uniform solids distribution challenges.
Fluidized Bed	High heat and mass transfer rates. Minimized risk of hot spots and improved temperature uniformity.	Complex fluid dynamics and process modelling. Requires careful control of fluid velocity to avoid elutriation. Risk of erosion of internal components due to particle movement.

.

5.3. Research and Demonstration Projects

This section provides an overview of several research projects focused on thermochemical energy storage (TCES) systems. The review places particular emphasis on research-oriented initiatives, highlighting different modeling approaches, reactor design strategies, and their experimental and numerical validation. Consequently, the scope also includes projects based on the broader concept of thermochemical energy storage, going beyond gas–solid sorption systems to capture relevant advances in reactor architecture and performance assessment.

Between 2003 and 2007, within the framework of the International Energy Agency Solar Heating and Cooling Programme (IEA SHC), Task 32 was conducted under the title “Advanced Storage Concepts for Solar and Low Energy Buildings”. This task was subdivided into four subtasks, of which Subtask B was dedicated to chemical and sorption heat storage technologies. Six research projects were selected for implementation under this subtask, namely [266]:

• “Compact chemical seasonal storage of solar heat”—ECN and Eindhoven University of Technology, The Netherlands;
• “Evaluation of thermo-chemical accumulator (TCA)”—SERC, Dalarna University, Sweden;
• “Sorption storage”—SPF Institute for Solar Technology, Switzerland;
• “Modestore (Modular high energy density heat storage)”—AEE INTEC, Austria;
• “Monosorp”—ITW, University of Stuttgart, Germany;
• “Closed NaOH absorption storage”—EMPA, Switzerland.

The final report indicates that among the investigated systems, only the project titled “Evaluation of thermo-chemical accumulator (TCA)” reached commercialization within the timeframe of the IEA SHC initiative (Figure 16). The system was commercialized by the Swedish company ClimateWell, which has operated under the name SaltX Technology since 2016. Over 35 thermal storage units integrated with heat pumps were sold, primarily for solar heating and cooling systems in Spain. Although ClimateWell was not directly represented in the project, the company collaborated closely with SERC.

The “Modestore” project led to the development of a demonstration-scale installation; however, the materials (energy carriers) used during testing were deemed unsuitable for seasonal storage applications. Three of the remaining projects—“Compact chemical seasonal storage of solar heat”, “Monosorp”, and “Closed NaOH absorption storage”—progressed to the design and testing phases of laboratory-scale prototypes. The “Sorption storage” project concluded at the stage of material characterization [266,267].

Hauer [268] presented a thermochemical heat storage system developed between 1998 and 2001 within the EU-funded project “High Energy Density Sorption Heat Storage for Solar Space Heating” (“HYDES”). The aim of the project was to develop and demonstrate a high-energy-density storage system based on closed-cycle adsorption processes, suitable for long-term storage of low-temperature thermal energy. The system was tested for seasonal solar space heating under various climatic and operational conditions, utilizing a silica gel/water working pair (Figure 17). During the charging phase in summer, thermal energy from solar collectors is transferred to three separate adsorbent beds, inducing water desorption from the silica gel. In the discharging phase during winter, the low ambient temperatures promote water evaporation in the evaporator/condenser units. The resulting adsorption process releases heat, which is then directed to the building’s heating system.

The subsequent “MODESTORE” project represented a continuation of the “HYDES” concept, further developing the design of the thermochemical reactor. The system also operated in a closed-loop configuration using silica gel as the adsorbent material. The prototype reactor had a volume of approximately 350 L and contained about 200 kg of silica gel. Unlike the initial “HYDES” configuration, the “MODESTORE” design integrated the packed-bed reactor and the condenser/evaporator within a single housing, eliminating the need for a valve and enabling a more compact system layout, as shown in Figure 18. Despite these design improvements, experimental studies on the demonstration installation revealed insufficient storage performance, as the achieved energy density was lower than that of conventional hot water tanks [267].

The “MonoSorp” project was a research initiative started in 2003 at the Institute of Thermodynamics and Thermal Engineering (ITW) at the University of Stuttgart, Germany, under the leadership of Kerskes [267,269]. The project investigated an open-cycle seasonal adsorption storage system integrated directly into the mechanical ventilation system of a building (as shown in Figure 19). The core storage material consisted of highly loaded zeolite A4 formed into honeycomb structures by thermoplastic extrusion. This geometry was chosen for its favorable adsorption kinetics and low pressure drop, which is particularly advantageous in open-cycle processes that rely on fan-driven airflow. The storage system was designed to recover moisture from the indoor air during the heating season by using the heat of adsorption to preheat the air stream, with temperature gains ranging from 15 to 25 °C. This preheated and dehumidified air was then passed through the building’s heat exchanger to raise the supply air temperature to approximately 40 °C, contributing directly to space heating. Desorption of the zeolite took place during the summer using excess thermal energy from vacuum tube solar collectors, increasing seasonal solar use and efficiency. The system was designed for simplicity, requiring only a second heat exchanger and three additional valves. Numerical simulations were performed using PDEX (for sorption behavior) and TRNSYS (for system-level analysis) and validated against laboratory-scale tests using honeycomb modules. Simulation results for a reference single-family house in Germany showed that the “MonoSorp” system with a 7.6 m³ sorption tank and 20 m² CPC solar collector area achieved annual energy savings of up to 70%, comparable to conventional systems with larger heat accumulators and collector areas.

The “SolSpaces” project is a follow-up of the “Monosorp” project, focusing on a thermal storage system that operates in an open-loop configuration using a mechanical ventilation system with heat recovery. Unlike the previous concept, which used traditional collectors and an air-to-water heat exchanger, the system now uses direct air heating in a solar air collector. The system can operate in three modes: “heating with collector”, “heating with sorption store”, and “charging of sorption store”. The design of the heat storage reactor has been modified compared to the previous project. This includes changes in the material and its formation process. The reactor has a rectangular shape and is divided into four vertical segments, each of which contains horizontal segments. Each horizontal segment consists of two layers filled with 13XBF zeolite spheres (2 mm in diameter) and air gaps. Air is introduced at the top of the reactor, distributed through a central channel, and then flows through the two air gaps in the active segment. After passing through the layers of the bed, the air is combined and discharged through an outlet channel in the corner of the reactor. The air flow in each segment is controlled by dampers to ensure that only one segment is active at a time. A schematic representation of a vertical section through the storage reactor and its appearance during build-up are shown in Figure 20. The system, containing 4.3 m³ of zeolite 13XBF, provides a storage capacity of 700 kWh. Experiments demonstrated efficient and uniform charging and discharging, with thermal power ranging from 790 to 565 W. The system’s performance can be adjusted by modifying the air mass flow rate, with higher humidity increasing thermal power output. An automatic control system for operation has been implemented, with ongoing monitoring and future evaluation planned [270].

The “SolSpaces 2.0” project builds on the results of the original “SolSpaces” initiative. The aim of the project was to further optimize and simplify the solar heating system with a sorption-based thermal energy storage. Key objectives include reducing the desorption temperature from ~180 °C to ~130 °C by pre-drying the air, allowing the use of low-cost flat plate collectors. The project also explores combined thermal and electrical desorption using PV systems, integration of sorptive cooling to prevent building overheating, and testing of switchable glazing. In addition, the manufacturing process of the storage unit will be evaluated for cost reduction, with a focus on adapting the system to different building types and climate zones. The project was scheduled to be completed in September 2019. No further updates have been found, apart from information available on the website of the coordinating institution [271].

Between 2012 and 2016, the “More Effective use of Renewables Including compact seasonal Thermal energy Storage” project (acronym: “MERITS”) led to the development of the world’s first prototype installation for thermochemical heat storage powered by solar energy [272,273]. The system (shown in Figure 21) was constructed in early 2015 in Warsaw (Poland), inside a modified shipping container divided into three zones: control, process, and simulation. Four solar collectors with a total absorber area of 12.4 m² were installed on the roof. The storage material used in the system was sodium sulfide (Na₂S), which undergoes reversible hydration from the hemihydrate to the pentahydrate form, storing and releasing thermal energy in the process. The storage battery consisted of two vessels containing heat exchangers: the reactor (upper section) and the evaporator (lower section), separated by an internal valve that is opened during charging and discharging. The storage medium was arranged in lamellae on the fins of the reactor’s heat exchanger to enhance heat and mass transfer. The thermal capacity of the two sets of thermochemical batteries included in the system is 350 MJ.

The “CREATE” project [275] aimed to develop and demonstrate a compact, lossless thermal battery based on thermochemical materials for integration into existing buildings. The system used potassium carbonate (K₂CO₃) as the active storage material, selected for its stability despite a lower theoretical energy density compared to sodium sulfide. The storage module featured a prismatic design with internal heat exchangers and an integrated evaporator/condenser unit optimized for volumetric efficiency and thermal control. Laboratory tests and a six-month field demonstration in a residential building in Warsaw (Poland) confirmed the system’s ability to store heat without losses between charge and discharge cycles. The energy densities achieved were 115 kWh/m³ at the module level and 347 kWh/m³ at the material level—63% of the original target. While the thermal performance has been validated, economic feasibility remains a challenge with a projected system cost of EUR 30,000 and a payback period of 20–30 years at production scale. Subsequent projects in the Netherlands and Austria will persist in the development process, using different materials for the purpose of commercialization in two distinct applications: seasonal storage in a single-family residence and short-term storage for power-to-heat in a hotel or restaurant. The “CREATE” system combines traditional and new components. Standard elements include a low-temperature heat source, a heat pump, solar collectors, circulation pumps, and a large water buffer tank. The project developed thermochemical storage modules and an evaporator/condenser unit integrated into the heat battery. A schematic visualization of the system is shown in Figure 22a, while the corresponding demonstration installation is shown in Figure 22b.

The “Chemical heat storage using reversible solid/gas-reactions” (“CWS”) project proposed a concept for integrating a thermochemical heat storage system with a conventional solar thermal combi-system comprising a buffer water tank, solar collector array, and auxiliary (backup) gas boiler. The system was designed to store solar heat either in the buffer tank or by charging the thermochemical reactor through the dehydration of the storage material. The storage system operated in an open-cycle configuration and featured a distinct layout in which the reactor was physically separated from the storage material reservoir. Key components included an external fixed-bed reactor, a material transport mechanism, and an air/air heat exchanger for energy recovery. The reactor itself was composed of two chambers: a reaction chamber for hydration and a regeneration chamber for dehydration, separated by a water–air heat exchanger. Air was supplied in a crossflow arrangement at 180 m³/h. The experimental setup included a 20 L laboratory prototype reactor utilizing zeolite 13X as the storage material. Granular material was gravity-fed from a top-mounted reservoir at a controlled rate (≤5 kg/h) using a rotary feeder. This design allowed for continuous material flow, distinguishing it from typical static-bed reactors. The concept of the system is shown in Figure 23a, while the schematic of the reactor design flow modes—at hydration (heat recovery) and dehydration (heat storage) phase is shown in Figure 23b. In the “CWS” project, the process was analyzed both numerically using COMSOL Multiphysics and experimentally validated, confirming its feasibility and energetic advantages over conventional hot water tanks. Despite increased system complexity, performance gains and the use of low-cost storage materials justify further development. Key challenges remain in reactor design and system integration, and follow-up demonstration activities are planned.

The “ESSI” project investigated an open TCES system designed for high-density, long-term (seasonal) heat storage [277]. The system uses a SrBr₂/H₂O reactive pair and operates with humid air, providing a simpler and more cost-effective alternative to traditional pure steam-based systems. A full-scale prototype has been developed and tested, containing 400 kg of hydrated salt and offering a storage capacity of 105 kWh, with a reactor energy density of 203 kWh/m³. Specific energy densities of up to 388 kWh/m³ were recorded during hydration, with power outputs ranging from 0.75 to 2 W/kg. The thermochemical reactor was designed as a modular system consisting of eight stacked rectangular units, each filled with a fixed bed of hydrated SrBr₂. The final design omitted a planned ninth module due to manufacturing constraints. Each module holds approximately 50 kg of reactive material, for a total storage capacity of 400 kg and 105 kWh. The reactive bed in each module was designed to be 7.5 cm thick to provide both a manageable pressure drop (60–560 Pa) and adequate thermal performance, resulting in an energy density of 388 kWh/m³. The reactor was sized for ease of handling and operated at a constant humid air flow rate of 300 m³/h. The reactor concept and prototype are shown in Figure 24. The performance was supported by a previously validated 1D sharp-front model and experimental permeability measurements for the reactive bed in both hydrated and dehydrated states. Experimental results over several charge/discharge cycles demonstrated the feasibility and thermal stability of the system, with reaction kinetics stabilizing after six cycles. The reactor delivered sufficient thermal power for typical residential space heating applications under French climatic conditions. In addition, the system proved to be tolerant to intermittent operating modes without significant loss of thermal performance. The project also identified two key control parameters that influence the thermal performance of the reactor: the equilibrium drop and the humid air mass flow rate. While the latter showed a direct linear correlation with thermal power output, the outlet air temperature was mainly dependent on the inlet air conditions. These results suggest that power output can be effectively controlled by adjusting airflow, while temperature control requires careful management of inlet air characteristics. Overall, the results confirm the technical viability of humid air-driven thermochemical storage for seasonal applications and provide a basis for simple and effective control strategies tailored to the heating needs of the user [277].

Farcot et al. [278] presented a full-scale prototype moving bed thermochemical reactor, which was developed to demonstrate the feasibility of continuous heat storage for building applications using hydrated salt and humid air. The system used a SrBr₂-6H₂O/SrBr₂ as the reactive material and was designed to provide up to 1 kW of heating power. The reactor bed was 0.95 m high and wide, with a thickness of 10 cm, optimized to provide low pressure drop and quasi-uniform solids flow. The reactor model and prototype are shown in Figure 25. Air inlet conditions (temperature: 0–100 °C, humidity: 0–90%, flow rate: 60–400 m³/h) were controlled by a dedicated air handling unit. During steady-state operation, the reactor achieved uniform temperature distribution throughout the bed, reaching temperatures of up to 41 °C and specific heating powers of 1.7–4.6 kW/m³. The authors highlighted the strong influence of inlet air humidity on reactor performance and confirmed the operational feasibility of using moving bed configurations for thermochemical heat storage. However, limitations were identified in the solid transport mechanism, which remains manual in this prototype. The authors emphasize that further research is needed to improve material flow, automate the system, and evaluate long-term performance. The use of composite materials may address flow issues, although this would likely reduce the overall energy density of the system.

In the “Redox Materials-based Structured Reactors/Heat Exchangers for Thermo-Chemical Heat Storage Systems in Concentrated Solar Power Plants” (acronym: “RESTRUCTURE”) project, a pilot-scale thermochemical energy storage system based on the Co₃O₄/CoO redox pair was developed and experimentally tested at the Solar Tower Jülich facility in Germany [279]. The system used Co₃O₄-coated cordierite honeycomb structures stacked in two identical reactor chambers (each 0.8 m × 0.6 m) with a total reactive core volume of 0.181 m³ and a thermal storage capacity of 74 kWh (see Figure 26). Air was used as both the reaction medium and the heat transfer fluid, allowing the reactive material to function simultaneously as a heat exchanger and a storage medium. A two-dimensional, axisymmetric numerical model was developed to simulate the heat and mass transfer and the reaction kinetics during the charge/discharge cycles. The model was validated with extensive experimental data, including temperature measurements from 29 thermocouples and total stored/released energy. Relative errors between simulations and experiments ranged from 0.1% to 9.5%, confirming high predictive accuracy. Discrepancies observed during discharge were attributed to slower oxidation kinetics at lower temperatures. The project demonstrated the technical feasibility of cobalt-based redox heat storage and provided detailed insights into the thermal behavior and process propagation within the reactor. Future work will focus on extending the numerical model to include cascaded porous structures for combined thermochemical and sensible heat storage within a single reactor unit [280]. A techno-economic assessment of the “RESTRUCTURE” system showed that the estimated levelized cost of energy (LCOE) was at least 20% below the cost target of the US DoE SunShot Initiative (<0.15 €/kWh_th and fell well within the LCOE range for combined cycle power plants reported by the IEA (~EUR 0.07–14.5/kWh_th). The main cost drivers were identified as the high-temperature fan and the cobalt oxide active material. While further cost optimization is needed, particularly through materials development and scaling of blower production, the technology was deemed technically feasible and conceptually straightforward for future high-temperature CSP applications.

Another example of a pilot-scale thermochemical energy storage was developed in the “SEASTOR” project. This project was implemented under the Operational Programme Smart Growth 2014–2020, supported by the European Regional Development Fund (ERDF). The beneficiaries of this project were the Institute of Energy and Fuel Processing Technology (formerly the Institute for Chemical Processing of Coal) and the Polish Energy Group (PGE Group). The aim of the project was to design and develop a pilot-scale reactor. It was assumed that the air is heated by the MgSO₄ bed, and the nominal air flow rate is 100 m_n³/h. The bed mass was estimated to be 150 kg, and the heat storage capacity was estimated to be 171 kWh/m³. The size of the reactor was estimated to be 1000 mm in diameter and 2000 mm in height. The construction was divided into several levels (shelves) to achieve better control of heat and mass in the bed (see Figure 27). Unfortunately, this project was closed in 2020 due to development and financial barriers. Namely, the challenges were the cost-effectiveness of implementation, low energy density, size of the system in terms of its application as a main heating source. The project partners agreed that further industrial R&D work would not lead to the expected results and that future implementation of the project results would be pointless due to a lack of cost-effectiveness. One of the results of this project was the patent “Variable capacity heat storage” (PL245575B1) [281]. This idea solves a significant limitation. Namely, in the single module reactor or single layer bed, the evaluation of the bed condition in its central zone is very difficult. In addition, the reaction zone is moved gradually, so some over-reacted zone is useless. If the reactor is divided into zones, as a multi-zone solution, the heat storage can be controlled more efficiently, and a certain part of the bed can be loaded separately, step by step.

Zhang et al. [282] developed a tubular modular thermochemical energy storage reactor with a fixed bed and evaluated it numerically (COMSOL) and experimentally. The system consisted of 64 tubular-type modular beds filled with a CaCl₂-impregnated vermiculite composite and was designed in a compact 0.4 m cubic enclosure. Air served as both a heat transfer medium and a reactant carrier. Humidified ambient air was directed through the bed, and outlet temperature profiles were used to confirm agreement between simulated and experimental results. The authors compared tubular-type and plate-type configurations under identical inlet conditions. Model validation was performed using a laboratory-scale test rig containing a metal mesh-packed bed filled with the same composite material (see Figure 28). During discharge, the reactor achieved a peak outlet temperature of 45.1 °C, a thermal power of 501 W, a pressure drop of 60 Pa, and a high thermal efficiency of 96.3%. Simulated and experimental results were in close agreement (RMSD: 4.2%). While both achieved similar temperature lifts, the tubular reactor showed significantly lower pressure drops and higher equivalent thermal efficiency, particularly at elevated humidity. Design parameters such as the number of modules and inner tube radius influenced system performance by reducing bed thickness, improving airflow, and increasing efficiency. However, trade-offs between stack density and thermal performance must be considered in practical design. An experimental scale reactor is planned for future testing under real-world conditions.

To illustrate the current state of technological advancement in TCES,

Table 14. Comparative overview of selected TCES projects.

Project (Duration)	Storage Type	Storage Material	Reactor Design	Energy Density/Power	Economic Analysis/Cost	Technological Barriers	Remarks	Ref.
HYDES (1998–2001)	Sorption (closed system)	Silica gel (GRACE 127B); spherical beads, diameter 2–3 mm; bulk density 790 kg/m3; approx. 2 × 800 kg of material.	Experimental setup: 2 cylindrical adsorbers + 1 condenser/evaporator; nominal volume: 1.25 m3; bed volume: 1.1 m3.	115 kWh/m3 (adsorber 1); 123 kWh/m3 (adsorber 2)	No data available.	Energy density only 1.8–2.2× higher than water tanks; insufficient material stability under operating conditions.	Continued under the MODESTORE project.	[268]
Evaluation of ThermoChemical Accumulator (TCA) (2003–2006)	Absorption (three-phase system)	LiCl/water with inhibitors; solid–liquid–vapor transitions during charge/discharge cycles.	Cylindrical unit (2000 mm height, 700 mm diameter) with internal heat exchanger, pumps, salt filter basket, and condenser/evaporator. Includes connection unit with eight control valves.	56 kWh (cooling), 76 kWh (heating); 10 kW heating, 18 kW cooling; 94% design efficiency	No data available.	Non-condensable gases, poor valve and crystal control, heat exchanger wetting—all led to reduced real-world performance.	Commercial development ongoing; units sold and integrated with solar/district heating systems.	[267]
MODESTORE (2003–2006)	Sorption (closed system)	Silica gel; preparation method not specified.	Pilot-scale cylindrical reactor (350 L) with approx. 200 kg of material; bottom: evaporator/condenser; center: 20 cm steam channel.	50 kWh/m3 (material)	Approx. 4300 EUR/m3 (material)	Lower storage capacity than water tanks	Development discontinued.	[267]
MonoSorp (2003–2006)	Sorption (open system)	Zeolite 4A; compressed into honeycomb-shaped blocks; 70 kg of material.	Experimental installation: prismatic reactor, 100 L.	160 kWh/m3 (material); 120 kWh/m3 (system)	Approx. 2500–3500 EUR/m3 (material)	High desorption temperature (180 °C); high material cost.	Continued under the SolSpaces project.	[267,269]
SolSpaces (2012–2016)	Sorption (open system)	Zeolite 13XBF; binder-free spherical particles (~2 mm); 4.3 m3; bulk density: 680 kg/m3.	Pilot reactor: rectangular, vertically divided into 4 sections and horizontally into several segments. Each segment has 2 bed layers; air flows in through a central duct and exits through corner ducts. One segment operates at a time. Total reactor volume with insulation: 8 m3.	163 kWh/m3 (material)	No data available.	High desorption temperature (180 °C).	Continued under the SolSpaces 2.0 project.	[270]
SolSpaces 2.0 (2016–2019)	Sorption (open system)	Same material and configuration as in SolSpaces.			Focus on reducing desorption temperature to ~130 °C; testing PV system integration.	-	-	[271]
MERITS (2012–2016)	Absorption (closed system)	Na2S; preparation method not specified.	Cylindrical reactor: two tanks with internal heat exchangers—top (reactor), bottom (evaporator/condenser); sodium sulfide hydrate layered on finned structures.	350 MJ	No data available.	-	Continued under the CREATE project.	[273,274]
CREATE (2015–2020)	Absorption (closed system)	K2CO3; two variants: (1) Triangular prisms: 96% K2CO3·1.5H2O, 3% graphite, 1% pyrogenic silica; bulk density 1000 kg/m3. (2) Granules: 97.5% K2CO3·1.5H2O, 2.5% graphite; grain size: 73% 4–5 mm, 27% 1.4–2 mm; bulk density 1200 kg/m3.	Lab: 10 cm cube with thin (0.25 mm) horizontal copper fins, spaced 10 mm apart. Moisture supplied via 4 structural pipes. Pilot: absorber exchanger volume ~250 L.	128–153 kWh/m3 (material)	4700–9400 EUR/module (7 m3)	-	Follow-up projects maybe will be realized in NL and AT target seasonal and short-term storage applications for commercialization.	[275]
CWS (2008–2012)	Sorption (open, moving bed)	Zeolite 4A; preparation method not specified.	Lab-scale cuboid (20 L), dimensions 0.5 × 0.5 × 0.08 m.	185 kWh/m3 (material)	No data available.	Proof-of-concept confirmed; further development needed for material, reactor design, and system integration.	No further development reported.	[276]
ESSI (2014)	Thermochemical (open system)	SrBr2·H2O/SrBr2·6H2O	Rectangular lab reactor (99.3 × 77.5 × 72) cm, eight trays (8 × 70 × 65) cm; 50 kg per tray; bed height: 7.5 cm.	388 kWh/m3 (material); 203 kWh/m3 (reactor)	No data available.	-	No further development reported.	[277]
L. Farcot et al. (2019)	Thermochemical (open, moving bed)	SrBr2·H2O/SrBr2·6H2O	Prototype reactor: (0.95 × 0.95 × 0.1) m	1.7–4.6 kW/m3 of reactor bed	No data available.	-	-	[278]
RESTRUCTURE (2011–2016)	Redox (Co3O4/CoO)	31.5% Co3O4, 68.5% cordierite; honeycomb ceramic monolith.	Pilot: 2 symmetrical insulated cuboid chambers (0.8 m height × 0.6 m width).	135–205 kWh/m3 (system)	Yes, Overall LCOE for CSP plant < 0.12 €/kWhth	-	No further development reported.	[279,280]
SEASTOR (2019–2021)	Thermochemical (hydration/dehydration)	MgSO4·1H2O/MgSO4·7H2O	Pilot-scale reactor, 100 mn3/h air, diameter of 1 m and height of 2 m	150 kg of bed mass, estimated capacity 171 kWh/m3	CAPEX over 80,000 PLN (per reactor), OPEX about 2200 PLN	Cost-effectiveness of implementation, low energy density, size of system considered as a main heating source	Project was closed in 2020	This paper
Zhang et al. (2024)	Thermochemical (open system)	Vermiculite impregnated with CaCl2/ CaCl2∙6H2O	Two 3D models: tubular-type and plate-type modular reactorl 0.4 m × 0.4 m × 0.33 m (length × width × height)	Materials storage capacity: 42.50 L; peak power: up to 501 W	No data available.	Trade-off between efficiency and packing density; material stacking complexity for thin beds; need for improved scale-up and integration	Validated by COMSOL model and lab-scale setup; tubular design outperforms plate in pressure drop and efficiency; experimental reactor planned for real-world testing	[282]

compiles a detailed comparison of several research and demonstration projects carried out over the past two decades. The table includes system classifications, material, reactor specifications, energy density (power), and key outcomes related to performance and scalability.

Chen et al. [160] investigated an adsorption-based thermochemical heat storage system using a LiCl/LiBr-zeolite 3A composite as the active material. A series of laboratory experiments (using 3 kg of the composite material) and TRNSYS simulations were conducted to evaluate the thermal performance of the system under varying humidity and air flow conditions. The optimum composite (15% salt concentration) achieved a heat storage density of up to 585.3 J/g and a discharge efficiency of 74.3%. A simulation model of a residential building equipped with this system showed a high coefficient of performance (COP) of 6.67, indicating strong potential for seasonal space heating applications.

Wang et al. [283] investigated the thermal performance of a fixed-bed thermochemical heat storage reactor using silica gel as the storage material. The authors conducted both experimental investigations and numerical modeling (ANSYS FLUENT) to analyze the impact of various loading parameters on the discharge behavior. The model was validated against experimental results and showed good agreement with a maximum Root Mean Square Percentage Error (RMSPE) of 10.08%. The results showed that under incomplete charging conditions, reversing the airflow direction during discharge significantly improves the thermal performance, increasing the maximum outlet temperature lift. In addition, reducing the charging airflow velocity at a fixed charging temperature resulted in higher discharge temperatures. The study also showed that increasing the charge air temperature and reducing the ambient humidity improved performance. These results elucidated the mechanisms by which charging conditions affect discharge efficiency and provided practical guidelines for adapting TCES system operation to different climatic conditions. The approach also addressed the challenge of low discharge temperatures under partial charging, providing valuable insights into operational control strategies for fixed-bed thermochemical storage systems.

6. Future Perspectives

Several limitations have been identified for TCES systems that need to be addressed. To realize the full potential of the TCES system, the future work should prioritize the following:

• Improving material stability and performance: further research is needed to address issues such as deliquescence, agglomeration, and caking in salt hydrates, possibly through the development of more robust composite structures with optimized porous matrices and binders that can better maintain structural integrity and prevent salt leakage during cycling. Long-term testing to assess material degradation and thermal stability, as well as operational efficiency under real-world conditions, should be further investigated.
• Enhancing heat and mass transfer: future efforts should focus on improving the thermal conductivity and slow kinetics of salt hydrates. This could include exploring novel matrix materials (such as advanced carbon-based structures for improved thermal conductivity), optimizing the design and properties of composite materials, and investigating methods to improve mass transfer within the storage system.
• Optimizing material combinations: for porous composites, the influence of matrix type and pore size on thermochemical behavior should be further studied. Continued research into mixed salt systems, particularly binary combinations, is crucial to discover and optimize synergistic effects that can lead to improved energy storage density, faster reaction kinetics, and enhanced long-term stability under a wider range of operating conditions. Identifying optimal salt ratios and compatible pairings will be key. Ultimately, no single material is universally optimal, and choices must be adjusted to meet specific system requirements.
• Advancing TCES to real-world integration: large-scale cyclic testing of promising composites is needed to identify potential operational challenges. Economic considerations must guide material selection; while porous carbons and MOFs show excellent performance, their high cost limits practical use. Instead, attention should shift to natural, modified materials such as vermiculite and biochar. Similarly, research should aim to improve low-cost, abundant salts such as MgCl₂, MgSO₄, K₂CO₃, and CaCl₂. High material costs remain a major obstacle to the large-scale deployment of TCES, so cost-effectiveness is essential. Binary salt systems are particularly promising because they allow the tailoring of properties using inexpensive components and porous supports. Moreover, comprehensive techno-economic analyses should be undertaken to assess cost-effectiveness and identify the main cost drivers.
• The reactor design: future research should focus on finding the optimal balance between bed thickness, packing density, and airflow resistance, particularly in fixed-bed reactors. To achieve this, future work should explore multi-layer reactor structures, the incorporation of fins, and the use of composite materials with improved thermal conductivity. For moving-bed reactors, ensuring good flowability and uniformity of storage materials is critical to avoid clogging and reduce maintenance costs. Moving bed reactors, due to their structural complexity and abrasion issues, require further research into durable materials and robust system designs.
• Improving the operation of the system: optimization should focus on adapting reactor configurations to operating conditions, improving mass transfer in open systems through modular bed design, and improving heat transfer in closed systems through advanced heat exchanger integration. Well-developed control methods for air temperature, humidity, and flow rate are crucial to ensure consistent system performance and efficiency, especially in open-cycle configurations. In addition, more emphasis should be placed on research into long-term operational performance, environmental robustness, and compatibility with larger energy systems (e.g., district heating, CSP).

7. Conclusions

Salt hydrates are promising candidates for thermochemical energy storage due to their high energy density, long-term storage capability, and reversibility. Compared to sensible and latent heat storage systems, they can store 5–10 times more energy with minimal losses. Despite dozens of potentiating hydrate salts, only a few of them, such as SrBr₂, SrCl₂, CaCl₂, LiCl, K₂CO₃, MgSO₄, and MgCl₂ have shown real potential. However, these materials still face challenges such as deliquescence, low thermal conductivity, poor hydration kinetics, and decreased stability.

To overcome these drawbacks, composites using porous matrices have been developed to improve thermal performance, prevent leakage, and enhance structural integrity. Mixed salt systems offer further improvements through synergistic effects on reaction kinetics and stability. Consolidated composites form help increase volumetric energy density and mechanical strength, although they can sometimes impede mass transfer.

Extensive experimental and modeling studies have been conducted to evaluate TCES systems, including laboratory and pilot-scale demonstrations. Various reactor designs, such as fixed, moving, and fluidized beds, have been explored to optimize heat and mass transfer and minimize losses. Open-cycle systems, in particular, require precise control of air temperature, humidity, and flow to achieve reliable performance.

While the fundamental understanding of salt hydrates and their composites has advanced significantly, further work is needed to scale up, standardize system designs, and achieve commercial readiness. TCES continues to offer a unique path toward dispatchable, compact, and seasonal heat storage capable of bridging the gap between intermittent renewable supply and stable demand. With focused innovation and coordinated research efforts, TCES systems have the potential to play a significant role in the energy transition.

The key takeaways are as follows:

• Salt hydrates offer high energy densities and long-term stability.
• Composite materials improve thermal and mechanical performance.
• Mixed salts exhibit synergistic effects, enhancing kinetics and water uptake.
• Consolidated forms improve structural integrity but require careful design to avoid mass transfer issues.
• Promising reactor designs and test rigs exist, but further optimization and standardization are needed for scale-up.