An Overview of the Molten Salt Nanofluids as Thermal Energy Storage Media

Abstract

The research in the field of the nanofluids has experienced noticeable advances since its discovery two decades ago. These thermal fluids having minimal quantities of nano-scaled solid particles in suspension have great potential for thermal management purposes because of their superior thermophysical properties. The conventional water-based nanofluids have been extensively investigated so far with emphasis in their improved thermal conductivity. A novel class of nanofluids based on inorganic salts has been developed in the last few years with the goal of storing and transferring thermal energy under high temperatures. These molten salt-based nanofluids can in general be recognized by an enhanced specific heat due to the inclusion of the nanoparticles. However, it should be emphasized that this does not always happen since this thermophysical property depends on so many factors, including the nature of the molten salts, different preparation methods, and formation of the compressed layer and secondary nanostructures, among others, which will be thoroughly discussed in this work. This peculiar performance has caused a widespread open debate within the research community, which is currently trying to deal with the inconsistent and controversial findings, as well as attempting to overcome the lack of accurate theories and prediction models for the nanofluids in general. This review intends to present an extensive survey of the published scientific articles on the molten salt nanofluids. Other important realities concerning the development and thermal behavior of the molten salt nanofluids, such as the stability over time of the nanoparticles dispersed in the molten salts, latent heat, viscosity, and thermal conductivity, will be reviewed in the current work. Additionally, special focus will be given to concentrated solar power technology applications. Finally, the limitations and prospects of the molten salts nanofluids will be addressed and the main concluding remarks will be listed.

1. Introduction

In the nanotechnology area of research, the nanofluids upsurged as very promising fluids for heat transfer enhancement and energy storage. These fluids are composed of a base fluid where nano-scaled particles (less than 100 nm) are suspended in concentrations, in most of the cases, lower than 10%. The addition of nanoparticles into a thermal fluid produces a considerable heat transfer enhancement, with special emphasis on its thermal conductivity and specific heat. In contrast, the nanofluids have overcome the stability concern provoked by the Brownian motion shown by the nanoparticles suspended in the base fluid. The nanofluids were introduced by the researchers Choi and Eastman [1] and the initial experiments conducted by Eastman et al. [2] consisted of dispersing copper oxide nanoparticles in water at a concentration of 5% vol. The results revealed a 60% increase in the thermal conductivity as compared to that of the water itself. Additionally, the innovative thermal fluid offered superior suspension results, with no significant settling and accumulation of the nanoparticles. The thermal conductivity enhancement has also been investigated by several other researchers experimenting with different types of nanofluids. The carbon-based, ceramic, and metallic nanoparticles have been investigated as additives to base fluids like water and ethylene glycol, suspensions which can be recognized as conventional nanofluids. Nevertheless, diverse large-scale applications require the use of thermal fluids that operate at typical temperatures superior to those adequate for the traditional nanofluids. This is the case of the applications inherent to the concentrated solar power plants, given that the laboring of these plants depends greatly on the effective solar thermal energy-to-electricity conversion at temperatures beyond 300 °C. Additionally, the thermal energy transport is currently carried out by the usage of thermal oils, known as heat transfer fluids, with thermal stability up to 400 °C. The employment of molten salts as heat transfer base fluids has been a common practice in the last few years, since these fluids are able to operate in a wider temperature range, enhancing, consequently, the overall efficiency of the processes. Besides that, the use of molten salts as thermal energy storage materials has been the usual procedure in the concentrated solar power field of work [3]. The fundamental beneficial features of the molten salts used in this field are their cost-effectiveness and thermal stability up to higher temperatures in the order of 600 °C or more. However, their poor heat transfer rate limits their industrial implementation. Hence, the enhancement of the thermophysical properties of the molten salt nanofluids, such as the specific heat, latent heat, and thermal conductivity, should be the pathway to follow to improve the thermal energy storage systems and new heat transfer fluids in concentrated solar power facilities. With the purpose of fulfilling this need, an innovative nanofluid was developed by the authors Shin and Banerjee [4]. The authors used a eutectic salt Li₂CO₃-K₂CO₃ with a 62:38 molar ratio and silica nanoparticles. The specific heat of the nanofluids exhibited a maximum increase of 100% with respect to the specific heat of the base fluid alone, with only 1% wt. incorporated in the salt. The researchers also stated that the overall investment cost might be reduced by 50% because of the combined use of higher working temperatures and the reduction in the needed quantity of material to store an identical thermal energy amount. The synthesis method together with a set of suitable characterization techniques for these specific nanostructured materials were also introduced by Shin and Banerjee. In addition, many arrangements of salts and different types of nanoparticles can be currently found in the published literature. Nevertheless, the findings of the specific heat and other thermophysical properties are not consistent and, in some cases, are controversial. Apart from this, the intrinsic governing mechanisms of the interactions between the nanoparticles and the molten salt base fluid are still not yet totally understood. As mentioned before, the main purpose of the molten salt nanofluids is the thermal energy storage and heat transfer enhancement in concentrated solar power plants. These thermal fluids can be employed in this application according to three different routes: as sensible storage media, as heat transfer fluid, and as latent heat storage media recognized by nano-enhanced phase change materials. The benefits of the molten salt nanofluids can be summarized as follows:

(i)

Improved specific heat, latent heat, and thermal conductivity when compared to those of the base fluid alone.

(ii)

Stability over time of the nanoparticles in the molten salt.

(iii)

Thermal stability and evidence of no significant deterioration under thermal cycling conditions.

(iv)

Reduced increase of the viscosity caused by the inclusion of the nanoparticles, which allows for efficient pumping within the thermal management systems.

(v)

Chemical compatibility with the container materials and absence of erosion in the fluidic systems.

The diagram of Figure 1 illustrates the main concerns closely related with the synthesis and characterization of the molten salt nanofluids, together with their performance evaluation in flow boiling applications.

This article gives a comprehensive up to date summary of the materials involved, preparation methods, thermophysical and rheological properties, and potential energy storage applications of the molten salt nanofluids. The models explaining the augmentation of the specific heat and the molecular simulation results are also discussed. Additionally, the limitations and challenges closely linked with the molten salt nanofluids will be listed in Section 7. The innovation of the present work of review is the detailed analysis of the parameters influencing the specific heat capacity of the molten salt nanofluids, such as the concentration and morphology of the nanoparticles, the synthesis methods, and the nature of the used base salt. Furthermore, this work has reviewed in a first time published detailed manner the preparation methods, thermophysical properties, long term stability, corrosivity concern, applicability in concentrated solar power plants, and life cycle assessment of the molten salt nanofluids. In essence, this review intends to highlight the laboratory environment routes and potential large-scale ways to produce molten salt nanofluids and expose the factors that improve the stability and thermophysical properties of these fluids. This work plans to partially decrease the existing literature gap by congregating in only one work the comprehensive survey of all the fundamental parameters affecting the molten salt nanofluids as thermal energy storage media. According to the actual knowledge of the authors of this review, there is no evidence of published a survey with so many impacting factors interpreted in so much detail. Especially the specific heat capacity impacting factors inherent to the molten salt nanofluids and not to only the traditional nanofluids, as is commonly found in the published articles, and the extended detailed sections devoted to the potential thermal energy applications and associated environmental impact of the molten salt nanofluids are features that are capable of distinguishing it from the already published studies. Moreover, more than half of the cited references are bounded exclusively with the application of the molten salt nanofluids in concentrated solar power plants that also account for a considerable gathering of information closely related with the innovative thermal fluids applied in the renewable energy technology. Additionally, serious problems arising from the corrosivity concerns of the molten salts nanofluids and potential strategic solutions are also extensively reviewed. These are the factors that may lead to the conclusion that this review will, within real possibilities, fill the gap in the literature related with an extended interpretation of the factors and mechanisms directly connected with the development and implementation of the molten salt nanofluids.

2. Types of Molten Salt Nanofluids

According to the melting point value, the molten salt nanofluids can be divided in the three general categories presented in

Table 1. Molten salt nanofluids classification according to the melting point.

Class	Melting Point Range	Molten Salts	Applications	Benefits	Limitations	Authors/ Reference
Low Melting Point	70–200 °C	Hitec Salt	Thermal Energy Storage	Avoid Freezing in the Circuits Improved Thermal Energy Storage Smaller Dimensions of the Storage Tanks	Low Thermal Stability Upper Limit of the Nitrites Reduced Long-term Stability and Increased Viscosity with Calcium Nitrates	Isaza-Ruiz et al. [5]
		Hitec XL Salt				Chen et al. [6]
		LiNaKCaNO3 Salt				Peng et al. [7]
Medium Melting Point	200–350 °C	Solar Salt KNO3-NaNO3	Concentrated Solar Power	Improved Thermal Energy Storage Relative Low Cost Chemical Safety Relative Low Corrosivity	Temperature Range Restricted by Crystallization at 240 °C Maximum Operating Temperature of 565 °C	Bauer et al. [8]
High Melting Point	>350 °C	Chloride Mixtures	Nuclear Power Coolants	High Operating Temperatures Low Vapor Pressure Good Thermal Stability	High Cost of Lithium and Zinc Chlorides Very High Corrosivity	Du et al. [9]
		Fluoride Mixtures	Nuclear Power Coolants and Fuel Cells	Very Low Vapor Pressure Improved Heat Transfer Capability	Relative High Cost	Forsberg et al. [10]
		Carbonate Mixtures	Thermal Energy Storage	Improved Thermal Energy Storage	High Corrosivity	An et al. [11]

.

3. Preparation Methods

Almost all the published studies regarding the molten salts nanofluids employed the two-step method. Normally, the following main steps are carried out to finalize the laboratory preparation method:

(i)

Weighing and mixing of the salt and nanopowders.

(ii)

Water dissolution of the base salt.

(iii)

Stabilization of the nanoparticles by ultrasonication procedures.

(iv)

Dehydration by evaporation of the water content.

Moreover, the fundamental concern about the molten salts nanofluids manufacturing is the scalability of the process. The high consumption of energy needed for the elevated amount of water removal and for the stabilization by ultrasonication is the major obstacle to its large-scale production. In terms of the reproducibility of the methods, there are many factors that are difficult to control, including the power of ultrasonication, temperature and duration of the drying step, and concentration of the dispersion.

3.1. Liquid Dispersion

When following this method, many authors used commercially available dried nanopowders [12,13], whereas some researchers begin the synthesis procedure by using commercial water-based nanofluids with a homogeneous dispersion of the nanoparticles [14,15]. In the former approach, the inorganic salt and the nanoparticles are weighed and mixed in the solid state conforming to the intended fraction ratio to which a certain amount of distilled water was added. The most used solid/water mass ratio until now is 1:100 [16], although the quantity of water has been reduced in other published articles [17,18,19]. The reason for this reduction is directly associated with the different ultrasound stirring controlling parameters used by the different authors. The ultrasonication is conducted to obtain a homogeneous scattering of the nanoparticles in the base fluid. The ultrasonication stage is very useful for breaking up the clusters of the nanoparticles, generated and stabilized through the strong Van der Waals forces that exist between the clusters. Nonetheless, it has been found that an ideal sonication time can be achieved and if this time is exceeded, the agglomeration of the nanoparticles may happen, affecting the final characteristics of the nanofluids. In this sense, the authors Akanda and Shin [20] carried out a two-step method to evaluate the influence of the ultrasonication time on the final molten salt nanofluids for concentrated solar power purposes. In this sense, the researchers prepared the binary solar salt with the incorporation of 1% wt. of silica nanoparticles and tested the ultrasonication durations between one and a half hour to five hours. The thermal stability of the pure salt and nanofluids between 100 °C and 600 °C was investigated by thermogravimetric analysis. The authors concluded that the nanofluids that suffer an ultrasonication for 3 h were the ones that exhibited the highest specific heat, representing a 15.6% increase as compared to the specific heat capacity of the molten salt itself. Furthermore, the ultrasonication time can be considerably decreased if an ultrasonic probe is employed instead of an ultrasonic bath. The last step of the preparation method is the removal of the water through evaporation: typically, the aqueous suspension is poured into a Petri dish and after that heated on a hot plate. Additionally, the literature has reported diverse drying durations, ranging from two hours to complete nighttime, and heating temperatures from 60 °C to 200 °C. Nevertheless, in the drying stage, the boiling of the water must be prevented to hinder the undesirable agglomeration of the nanoparticles, given that the collisions between them provoke the degradation of the specific heat capacity [4].

3.2. Mechanical Dispersion

The direct blending of the nanoparticles with the base salt was also investigated, and for this purpose an in-house developed set up was proposed by Somani [21]. In his study, the nanoparticles were dispersed in acetone through ultrasonication, the resulting nanofluids were injected into the molten nitrates bulk and the nanoparticles stayed in the fluid, whilst the acetone vaporized. After that, an ultrasonication probe was used to obtain a uniform scattering of the nanoparticles. Moreover, the researchers Chieruzzi et al. [22] synthesized molten salt nanofluids by mixing silica, alumina, and silica/alumina nanoparticles with the solar salt NaNO₃-KNO₃ at a temperature of 300 °C with the aid of a micro-conical twin screw micro-compounder. Various mixing durations and screw speeds were applied to evenly mix the inorganic salts and nanoparticles. The technique of mixing the dry solid inorganic salt and the nanoparticles with the help of balls of steel was also used. The fundamental benefit of the technique was its potential scalability and on the prospective to prepare great quantities of molten salt nanofluids. Nevertheless, the main limitation could be a heterogeneous dispersion of the nanoparticles [23]. Additionally, the eventual contamination of the nanofluids by the milling steel balls can be taken as an extra back draw of this method. Furthermore, in the study conducted by Song et al. [24], a mechanical dispersion method was used to produce a phase change material with thermal energy storage purposes based on the nitrate salt NaNO₃ with the inclusion of silica, and silica/titania nanoparticles, and with a matrix of expanded graphite. The method involved the weighing of sodium nitrate and silica nanoparticles and the mechanical stirring of the components at 400 °C for 15 min and with a velocity of 750 r.p.m. Finally, a molten salt nanofluid with the inclusion of 1% wt. of silica nanoparticles was obtained. Moreover, the molten salt nanofluids with 0.1% wt. of silica nanoparticles and 0.9% wt. of titania nanoparticles were synthesized by the same mechanical dispersion method. The as-prepared molten salt nanofluid with only silica nanoparticles was placed in a salt bath at 400 °C, and different weight fractions of 7%, 10%, 15%, and 20% of expanded graphite were incorporated. After that, the mixture was stirred for one hour until the molten salt completely entered the pores of the expanded graphite. Then, the mixture was cooled and grinded in a desiccator, and hence the NaNO₃/SiO₂/EG phase change material was obtained. Consecutively, the nanopowder was pressed using a machine with 10 MPa of molding pressure. After that, the pressed nanopowder was placed inside a furnace for sintering, according to a heating rate of 2 °C per minute at temperature values from 30 °C to 150 °C. The slow heating rate was chosen to evaporate the water content completely in the original sample. Then, the heating rate was altered to 5 °C per minute between 150 °C and 400 °C and the phase change material was kept at 400 °C for 90 min. Finally, the sample was cooled in air until room temperature, and the definitive sintered NaNO3/SiO₂/EG composite phase change material block was obtained. Figure 2 schematically shows the mechanical dispersion method.

In the work of Sang et al. [25], a high-temperature mechanical mixing technique was applied to prepare the carbonate K₂CO₃–Li₂CO₃–Na₂CO₃ molten salt with a mass ratio of 4:4:2 and with the inclusion of nanoparticles of silica. The effects of the different stirring rates of 500 r.p.m., 750 r.p.m., and 1000 rpm and varying stirring times of 15 min, 30 min, and 60 min on the final thermal features of the nanofluids were evaluated. The synthesis method consisted of weighing a certain amount of the salt and heating it to 550 °C. Then, the salt was stirred with the aid of an electromechanical agitator for 40 min. After this procedure, an amount of silica nanoparticles was dispersed in the salt directly by the agitator. Hence, there were prepared nanofluids with different states of dispersion with the referred different stirring rates and periods. The obtained results exposed that a stirring rate of 750 rpm for 30 min was the most effective for the increase in the thermal conductivity and specific heat of the nanofluids. Additionally, the authors reported that when the stirring time was one hour, the number of formed nanostructures became reduced, which was caused by the re-aggregation process of the added nanoparticles, and with this stirring time the enhancement in the specific heat decreased. In addition, the decrease in the number of formed nanostructures was more pronounced under an agitation rate of 1000 r.p.m. when the agitation time was augmented from 15 min to 60 min. Nonetheless, the specific heat of the most effective nanofluids exhibited only a negligible degradation after being subjected to the thermostatic stability test at a temperature of 600 °C during 150 h and to the thermal shock stability test through 50 cycles from 600 °C to room temperature.

3.3. In Situ Production

Some works suggested the synthesis of the nanoparticles in-situ in the molten salt with the addition of precursors [26,27], being the preparation and dispersion of the nanoparticles carried out at the same time in the solvent phase. This one-step preparation method has the benefit of mitigating the agglomeration and settling effects of the nanoparticles. However, with this procedure, it is harder to control the size and shape of the nanoparticles since little alterations in the parameters of the synthesis procedure, such as temperature, duration, and feeding rate, among others, can appreciably modify the final characteristics of the nanofluids caused by the changes in the size distribution of the nanoparticles and long-term stability. The selection of the most suitable concentration value requires an enhanced knowledge of the mechanisms underlying the superior thermophysical characteristics of the molten salt nanofluids. The in-situ production can be carried out according to a physical route or, alternatively, to a chemical route. One example of the physical route is vapor condensation, where the metal-containing vapor is condensed to produce the nanoparticles in the base fluid by contacting the vapor flowing under low pressures within the fluid [1]. Other physical techniques include the synthesis by submerged arc spray [28] and the laser ablation [29]. Using these techniques, it is possible to obtain uniformly dispersed nanofluids but the methods require sophisticated apparatuses that strongly limit their use in large-scale industrial purposes. Moreover, in the wet chemical route, the precipitation of the additives or the thermal decomposition of the precursors occurs. This method can synthesize uniformly dispersed nanofluids with the tunable morphology of the nanoparticles. Nevertheless, the wet chemical route requires a base fluid different from the one of the finally targeted molten salt nanofluid, and the incorporation of additives entails the risk of contamination by different types of impurities. Hence, most of the molten salt nanofluids prepared by the wet chemical approach were still included in the two-step preparation methods class where the filtered precipitates were cleaned and re-dispersed into a fresh base fluid. In the one-step approach the water-assisted or wet mixing and the direct one-step method or dry mixing can be conducted. One example of the wet mixing method is schematically presented in Figure 3, in which the salt components were initially mixed with the precursor conforming to the intended weight ratio and after that were dissolved in deionized water. The solution is boiled on a hot plate for an extended period until the water completely evaporated from the mixture. Then, the solid mixture is placed inside a furnace for the in-situ production of the nanoparticles in the salts through the thermochemical decomposition of the chosen precursors. On the other hand, through the dry mixing methodology, which is schematically illustrated in Figure 4, the salt components were mixed with the precursors in the container under nanopowder form. After that, the nanopowder mixture was stirred and placed inside a furnace for an extended period. In the course of the thermal ramping and heating of the mixture, the crystal phase was gradually liquified, and the precursors decomposed in the liquid phase.

In the study performed by Lasfargues et al. [30], a molten salt nanofluid composed of the eutectic mixture of NaNO₃ and KNO₃ with a 60:40 molar ratio with the incorporation of titania nanoparticles and 3.0% wt. of TiOSO₄ as precursor was prepared. The goal of this study was to implement a low-cost alternative in-situ production method that is potentially scalable, given that the most employed available techniques are not entirely cost-effective for large-scale production. Figure 5 schematically illustrates the process of the in-situ production of the nanoparticles in the molten salt. Such a procedure assured a sudden temperature decrease and consequent crystallization of the salt. In addition, through thermogravimetric analysis and differential scanning calorimetry, the intended in-situ production of nanoparticles was analyzed and the specific heat was measured. The researchers confirmed an appreciably increased specific heat of the molten salt nanofluid, as compared to that of the binary nitrate eutectic mixture itself, by 5.4% at 390 °C and 7.5% at 445 °C, for a 3.0% weight fraction of precursors.

Table 2. Main characteristics and findings of the main preparation methods of the molten salt nanofluids.

Method	Liquid Dispersion	Mechanical Dispersion	In-Situ Production
Involved Techniques	Solid Mixing of the Salt and Nanoparticles Dissolution Ultrasonication Drying by Evaporation	Dry Mixing and Milling with Mechanical Aid Mechanical Stirring Ultrasonication	Physical Route: Vapor Condensation, Arc Spray Analysis, Laser Ablation Chemical Route: Precipitation, Thermal Decomposition of Precursors Wet Mixing Dry Mixing
Benefits	Use of Commercial Nanopowders	Scalability to Produce Large Amounts of Molten Salt Nanofluids	Mitigation of the Agglomeration and Sedimentation Effects of the Nanoparticles Improved Stability
Limitations	Possible Agglomeration of the Nanoparticles	Eventual Heterogeneous Dispersion of the Nanoparticles Possible Contamination from the Milling Equipment	Requires Base Fluids with Low Vapor Pressure and Expensive Equipment Possible Residual Reactants Hinder the Positive Effect of the Nanoparticles Health Risks Arising from the High-Temperature Molten Salt Manipulation Poor Scalability to Industrial Scale
Main Findings	30.6% Specific Heat Increase	38.5% Specific Heat Increase 50% Thermal Conductivity Increase	7.5% Specific Heat Increase
Authors/Reference	Schuller et al. [31]	Sang et al. [25]	Lasfargues et al. [30]

summarizes the involved techniques, benefits, and limitations of the main preparation methods of the molten salt nanofluids.

4. Thermophysical Properties

4.1. Thermal Conductivity and Diffusivity

The thermal conductivity of the molten salt nanofluids has already been investigated [32,33]. The elevated melting point and corrosive potential of the molten salts and their nanofluids usually make the accurate measurement of the thermal conductivity difficult. Nevertheless, many experimental works have measured the thermal conductivity of the molten salt nanofluids in the solid state and the results were compared with the thermal conductivity of the salts themselves. Additionally, Shin [34] used the Laser Flash Analysis to measure the thermal diffusivity of the binary salt Li₂CO₃-K₂CO₃ with a 62:38 molar ratio and with 1% wt. of silica nanoparticles. The thermal conductivity was determined through the following equation:

Kαρc = p

(1)

where k is the thermal conductivity expressed in W·m⁻¹·k⁻¹, α is the thermal diffusivity in m²s⁻¹, ρ is the density in kg·m⁻³, and cp is the specific heat in J·kg⁻¹·k⁻¹. An appreciable increase in the thermal conductivity was obtained regardless of the temperature, obtaining increments of 47%, 36%, and 37%, as compared with that of the carbonate salts themselves, respectively. On the other hand, no increment was observed in the thermal conductivity in the solar salt NaNO₃-KNO₃ with a molar ratio of 50:50 and alumina nanoparticles in different concentrations [19]. The thermal diffusivity of the salt and nanoparticles was determined by laser flash analysis at the different temperature values of 65 °C, 85 °C, 105 °C, 125 °C, and 145 °C. A decrement in the thermal conductivity at the temperature of 145 °C was reported. A reduction of the thermal conductivity as compared with that of the salt was reported, regardless of the amount of added nanoparticles. In the study performed by the researchers Ueki et al. [35], the thermal conductivity was measured of the molten salt nanofluid composed of the Heat Transfer Salt 40% wt. NaNO₂-7% wt. NaNO₃-53% wt. KNO₃ as the base fluid with the inclusion of silicon carbide nanoparticles. The researchers reported that at a concentration of 0.72% vol. of the nanoparticles, the thermal conductivity of the molten salt was increased by 13% at 200 °C. The investigation team hypothesized that the fractal-like fluid nanostructure present in the molten salt improved the thermal conductivity of the final nanofluid. In the study carried out by the authors Li et al. [36], molecular dynamics simulations to investigate the thermal conductivity variation of the solar salt with the addition of alumina nanoparticles were employed. The obtained results showed that the inclusion of nanoparticles improved the heat conduction of the solar salt. Additionally, adding nanoparticles caused the heat flux concentration in the regions of the nanoparticle, which promoted the heat transfer. The decomposition of heat flux discloses that the thermal conductivity of the nanofluid was mostly contributed to by the alumina nanoparticles. Moreover, in the cases where the concentration of the alumina nanoparticles was inferior to 6% wt., the thermal conductivity was mainly contributed to by collision. On the other hand, the potential energy and collision contributed to the heat conduction of the molten salt nanofluids in the cases where the concentration of the alumina nanoparticles was higher than 6% wt. In the work conducted by Lu et al. [37], magnesia nanoparticles with sizes ranging from 20 nm to 100 nm and weight fractions from 0.5% to 2.0% were dispersed in the ternary LiNO₃-NaNO₃-KNO₃ nitrate molten salt. The authors investigated the heat transfer of the involved system and found that the thermal diffusivity was increased through the addition of the magnesia nanoparticles to the base salt, and it increased with increasing temperature. The thermal diffusivity increments with the different nanoparticle sizes at temperature values between 220 °C and 340 °C were from 5.3% to 13.7%. Particularly, when the 40 nm-sized nanoparticles had a concentration of 1%, the thermal diffusivity reached a 13.7% enhancement with respect to that of the molten salt. Moreover, the thermal conductivity also increased with the addition of nanoparticles, being 21.2% higher than the one of the molten salt itself using 1% wt. of 40 nm nanoparticles. The Brownian motion of the nanoparticles transports thermal energy at elevated temperatures and moves fast toward areas having lower temperatures. With the temperature increment, the diffusivity and perturbations of the heat flow due to the Brownian motion will be greater. Hence, when the nanoparticles with the larger dimension of 100 nm were added, the enhancement on the thermal conductivity was reduced by 11%. The authors interpreted this fact based on when the same concentration of nanoparticles was added with increasing nanoparticle size, the number of dispersed nanoparticles will be smaller, and a weaker thermal disturbance will result. This weaker disturbance will induce a deterioration in the heat transfer characteristics of the system. In the study performed by Zhang et al. [38] there were synthesized nanofluids of the ternary salt Li₂CO₃–Na₂CO₃–K₂CO₃ with the inclusion of alumina nanoparticles with 20 nm, 50 nm, and 80 nm of size, and in mass fractions of 0.2%, 0.4%, 0.8%, 1.0%, 1.4%, and 2.0%. According with the specific heat enhancement criterion, the weight fraction and size of the alumina nanoparticles were optimized and the nanofluids containing 1.0% of 20 nm-sized alumina nanoparticles, 1.0% of 50 nm-sized ones, and 0.8% of 80 nm-sized ones had been selected. The authors reported that the thermal conductivities of the nanofluids were 2.1 W·m⁻¹·K⁻¹, 2.14 W·m⁻¹.K⁻¹, and 2.2 W·m⁻¹·K⁻¹, respectively, and the enhancements as compared to the base salt itself were 23.3%, 28.3%, and 30.9%, respectively. Additionally, the obtained results revealed that the thermal conductivity augmentation rose with a growing concentration of the nanoparticles. The investigation team also stated that the smaller nanoparticles with larger surface energy tend to agglomerate and precipitate, which hindered the formation of an effective heat transfer network. The heat transfer network will exhibit a partially missing or incomplete structure in the cases where the agglomeration or precipitation of the alumina nanoparticles happens. These sites will reduce the efficiency of the system, leading to a decreasing propensity of the heat transfer capability with high amounts of smaller alumina particles. As a result, the molten salt nanofluids having 80 nm-sized alumina nanoparticles achieved the highest thermal conductivity of the tested nanofluids. The researchers Wei et al. [39] prepared solar salt nanofluids with the addition of weight fractions of 2.5%, 3.5%, 4.5%, 5.0%, and 10.0% of magnesia nanoparticles, and the corresponding thermal conductivity over 220 °C has been investigated. Additionally, the thermal conductivity of the molten salt nanofluids was determined by the product of the corresponding thermal diffusivity, specific heat capacity, and density. The thermal diffusivity was determined by laser flash analysis. The obtained average thermal conductivity from 0 to 5.0% wt. was between 0.37 and 0.60 W/(m²·K) at a temperature of 375 °C, and the increment of the thermal conductivity when compared with the base solar salt itself reached 62.1%, which increased the Nusselt number with enhanced thermal performance. Additionally, the nanofluids exhibited higher thermal diffusivities than the those of the base salts with rises of 5.5%, 24.2%, 39.1%, 45.3% and 46.9% for fractions of the nanoparticles of 2.5%, 3.5%, 4.5%, 5.0%, and 10%, respectively. Moreover, the thermal conductivity of the nanofluids was raised considerably with growing magnesia content. The obtained enhancement was between 5.4% and 62.1% when the concentration of the nanoparticles was augmented from 2.5% to 5.0%. Additionally, the thermal conductivity experimentally obtained values of the nanofluids were compared to those calculated by the Maxwell equation [40]. For low concentrations of the nanoparticles, the experimental values were consistent with the calculated ones, but when the concentration of the nanoparticles was higher than 2.5% the calculated values were appreciably lower than those obtained by experiment. The discrepancy can be justified by the fact that the Maxwell model accounts for stationary nanoparticles in the base molten salt, but the nanoparticles indeed move through random Brownian motion. Thus, the increase of the temperature increased the Brownian motion of the nanoparticles. Additionally, micro-scaled perturbations were generated inside the base fluids due to the Brownian motion of the nanoparticles. Hence, the energy transfer capability between the nanoparticles and the solar salt fluid became improved. The researchers Qiao et al. [41] synthesized molten nitrate salt nanofluids, adding silica nanoparticles into NaNO₃, LiNO₃, and KNO₃ single salts, and the impacts of the size of the nanoparticles between 15 nm and 5 µm, weight fractions between 0.5% and 4%, and operating temperatures between 200 °C and 380 °C were evaluated. Regarding the NaNO₃ nitrate salt, the addition of 0.5% of silica nanoparticles decreased the thermal conductivity with respect to the one of the inorganic salts itself. With the increasing of the concentration of the nanoparticles from 0.5% to 4%, the thermal conductivity of the nanofluids first decreased and then increased. Additionally, the inclusion of silica nanoparticles decreased the thermal conductivity, as compared with the pure NaNO₃ and LiNO₃ salts. Additionally, the authors verified that increasing the temperature decreased the thermal conductivity. In conclusion, the inclusion of silica nanoparticles in the nitrate salts was expected to possess improved thermal conductivity, given that the silica is more thermally conductive than the tested nitrate salts. However, the carried-out experiments showed the opposite trend, and the researchers interpreted the facts based on the existence of an interfacial layer between the nanoparticles and the fluid molten salt, which offers an increased thermal resistance, hindering the heat transfer and, consequently, decreasing the thermal conductivity. Moreover, the authors Cui et al. [42] studied the heat transfer enhancement of the solar salt with the incorporation of alumina nanoparticles by molecular dynamics simulations. The results exposed that the incorporation of alumina nanoparticles improved the thermal conductivity of the solar salt, but the verified increase was not as significant as that with silica nanoparticles. The authors did not interpret the fact based on the Brownian motion of the nanoparticles, microconvention of the base fluid, ordered layer of the base fluid, and potential energy. Instead, it was interpreted as the consequence of material components and heat flux fluctuation modes. The researchers found that the thermal conductivity was contributed to by the motion of the atoms and interaction between atoms in the alumina nanoparticles, rather than in the base fluid. Additionally, the existence of a local heat flux revealed that the inclusion of nanoparticles caused the heat flux concentration in the nanoparticles nearby regions, promoting the heat transfer this way. Moreover, when the mass fraction of the alumina nanoparticles was inferior to 6% wt., the main contributor for the thermal conductivity was the collision work, whereas the potential energy and collision work contributed together to the thermal conductivity of the nanofluids, when the concentration of the alumina nanoparticles was superior to 6% wt.

Table 3. Main thermal conductivity/diffusivity enhancements of the molten salt nanofluids.

Molten Salt	Nanoparticles	Best Concentration % wt.	Thermal Conductivity/Diffusivity Enhancement (%)	Authors/Reference
NaNO3-KNO3	Silica	1.0	50.0 (Diffusivity)	Li et al. [43]
NaNO3-KNO3	Silica	1.0	60.9	Yu et al. [44]
NaNO3-KNO3	SiO2@Al2O3 Core-shell	1.0	19.0	Nithiyanantham et al. [45]
NaNO3-KNO3	Alumina NPs/Alumina NRs	1.0	16.0/12.0	Nithiyanantham et al. [46]
NaNO3-KNO3	Multi-Walled Carbon Nanotubes	0.3	293.0	Wu et al. [47]
MgCl2-KCl-NaCl	Alumina	0.7	48.0	Han et al. [48]
NaCl-CaCl2-MgCl2	Expanded Graphite	1.0	78.0	Tian et al. [49]
Li2CO3-K2CO3	Single-Walled Carbon Nanotubes	1.5	57.0	Tao et al. [50]
Li2CO3-K2CO3-Na2CO3	Carbon Nanotubes	1.0	149.2	Sang et al. [51]
Li2CO3-K2CO3-Na2CO3	Magnesium Oxide	10.0	155.9	Wei et al. [52]

summarizes the main findings regarding the thermal conductivity/diffusivity enhancements of the molten salt nanofluids.

4.2. Viscosity

The molten salt nanofluid should ideally exhibit both enhanced thermal properties and relatively low viscosity to prevent an excessive increment of the pressure drop and the needed pumping power. Efforts should be made to strongly mitigate the rise in viscosity provoked by the incorporation of nanoparticles into the molten salt. Moreover, the research on the rheological properties of the molten salt nanofluids is still rather scarce and the available research focused primarily in inferring the dependence of the dynamic viscosity on the concentration of the nanoparticles on the molten salts, temperature, and shear rate. Additionally, the authors Lasfargues et al. [53] deduced accurately a correlation between the viscosity and the temperature for the solar salt NaNO₃-KNO₃ with a 50:50 molar ratio. There were also 0.1% wt. of nanoparticles of copper oxide added at temperature values ranging from 250 °C to 450 °C and under shear rates varying from 100 s⁻¹ to 1000 s⁻¹. The authors declared that both the solar salt and copper oxide nanofluids exhibited Newtonian behavior, which can be defined by a viscosity kept constant with growing shear rates. They also reported that the copper oxide nanoparticles enhanced the dynamic viscosity of the solar salt. The researchers argued that an extra amount of energy was required to set the nanoparticles in motion because of the existence of a vortex on their surface that may justify the dynamic viscosity increase. The applied shear rate, the type, morphology, and incorporated amount of the nanoparticles, the type of the base fluid, and the temperature may have direct influence on the viscosity of the final nanofluid. Lasfargues [53] also reported that the shear rates superior to 100 s⁻¹ are responsible for the breaking of the clusters into smaller nanoparticles. Furthermore, Jo and Banerjee [54] observed an increment of the viscosity of the nanofluids based on the Li₂CO₃-K₂CO₃ inorganic salt with a 62:38 molar ratio with the inclusion of multi-walled carbon nanotubes in different concentrations. The authors reported that the incorporation of 5% wt. of multi-walled carbon nanotubes led to an increase in the viscosity of up to 130% at a shear rate of 1000 s⁻¹, the nanoparticles clustering being the main appointed reason behind that fact. Additionally, it was found that the nanofluids exhibited non-Newtonian behavior (shear thinning fluid) under low shear rates, and the non-Newtonian feature was extended to the high-shear rate region by enhancing the concentration of the nanotubes. The viscosity increased considerably with the inclusion of the nanotubes where the aggregation of the nanotubes was more likely to happen. It was also found that the Krieger–Dougherty model provided a good prediction for the results. In the Krieger–Dougherty model for the viscosity prediction [55], the maximum concentration that still enables the flow is 0.6, and the intrinsic viscosity is 2.5 for monodispersed systems [56]. The volume fraction of the aggregates $\phi$_a can be expressed as a function of the ratio of the aggregate radii to that of the primary nanoparticles a_a/a as follows:

$${\phi }_a=\varphi {\left(\frac{a_a}{a}\right)}^{3-D}$$

(2)

where φ is the volume fraction of the nanoparticles and D is the fractal index that was assumed to be 1.8. The measurements better matched the predictions from the theoretical model for a ratio a_a/a equal to 4.7, using both nanotube weight fractions of 2% and 5%. Additionally, under a shear rate of 1000 s⁻¹ where the asymptotic value for the viscosity values was observed, the increase in the viscosity was 11%, 93%, and 1130% with respect to the viscosity of the salt itself for the concentrations of nanotubes of 1%, 2%, and 5%, respectively. Furthermore, it was verified that the viscosity of the nanofluids prepared without gallic acid as a surfactant was 18% higher than the viscosity of the nanofluids with the inclusion of the surfactant. In addition, with the aid of a rheometer, Xiao et al. [57] studied the viscosities of a HITEC salt nanofluid with the incorporation of 1% wt. and 2% wt. of graphene nanosheets at a temperature range of 200–450 °C, and solar salt with the incorporation of alumina nanoparticles at a temperature range of 250–500 °C. The viscosity measurements were carried out under the shear rates between 1 s⁻¹ and 250 s⁻¹. For the HITEC salt having graphene, a slight decrease was noted in the viscosity with the increasing shear rate. The reason behind this fact could be that the graphene structure might be separated under elevated shear rates, which decreased the viscosity. Additionally, the viscosity was found to decrease with increasing temperature values almost linearly when those were superior to 350 °C, but the viscosity decrease was non-uniform when the temperature values were lower than 350 °C, i.e., the largest viscosity of HITEC salt seeded with embedded graphene nanosheets occurred at 300–350 °C at 2% wt. of graphene. The authors explain this fact based on the low-temperature uncertainty since the sliced graphene strongly affects the flow of the molten salt, and the addition of graphene into the HITEC salt induces a non-Newtonian behavior of the nanofluids and, consequently, the melting characteristics of HITEC salt might be changed with the graphene addition. On the other hand, the viscosity of the nanofluids containing 2% wt. of graphene nanosheets was greater than the viscosity reported for 1 wt.% graphene. It was found that the viscosity of solar salt with alumina nanoparticles was kept constant under various shear rates, and slightly increased with increasing shear rates in the cases where these ones were superior to 100 s⁻¹. The eventual justification for this fact should be like that appointed for the HITEC inorganic salt, which included the potential agglomeration of nanoparticles and the associated uncertainty of the measurements. The addition of alumina nanoparticles led to small alterations in the viscosity values of the pure salt, being within the range of −35.4–8.1% for the HITEC-based nanofluids and within the range of −9.2–68.1% for the solar salt-based nanofluids. It should be stated that the decreased values of the viscosity were mainly due to the uncertainty associated with the measurements. However, the incorporation of 2% wt. graphene nanosheets into the HITEC salt led to the appreciable increases in the viscosity of 77.0%, 232.5%, 751.4%, 987.3%, 528.2%, and 550.2% at 200 °C, 250 °C, 300 °C, 350 °C, 400 °C, and 450 °C, respectively. The investigation team interpreted the difference based on the diverse morphologies of the added particles. The larger size of the graphene nanosheets restricted the flow of the molten salt and, consequently, increased the viscosity of the nanofluid. The viscosity of the molten salt nanofluids operating as heat transfer fluid in a concentrated solar power plant can be directly linked with the required pumping power, since the addition of nanoparticles may produce a high dynamic viscosity, which are probable practical contributors to the overall cost increase of the pumping power system.

Table 4. Main findings of the viscosity of the molten salt nanofluids.

Molten Salt	Nanoparticles	Concentration % wt.	Viscosity Findings	Authors/Reference
NaNO3-KNO3	Magnesium Oxide	2.5, 3.5, 4.5, 5.0 and 10.0	5.1 cp-2.4 cp Nearly the Same Viscosity of the Molten Salt	Wei et al. [39]
NaNO3-KNO3	Silica and Alumina	0.5–1.5	4.94 mPa.s at 300 °C	Munoz-Sanchez et al. [58]
NaNO3-KNO3 and LiNO3-NaNO3-KNO3 and LiNO3-NaNO3-KNO3-Ca(NO3)2	Silica	0.5 and 1.0	100.89–188.85% Increase for the Binary Salt, 18.75–71.12% Increase for the Ternary Salt, and 29.02–53.61% Increase for the Quaternary Salt	Jiang et al. [59]
NaNO2-NaO3-KNO3 and NaNO3-KNO3	Alumina and Graphene	1.0 Alumina and 2.0 Graphene	35.4~8.1% Increase for the Ternary Salt and −9.2~68.1% Increase or the Binary Salt with Alumina; Remarkable Increase with Graphene (e.g., 987.3%)	Xiao et al. [57]
Li2CO3-K2CO3	Silica	1.0	34.0% to 94.4% Increase	El Far et al. [60]
Ca(NO3)2-KNO3-NaNO3-LiNO3	Silica	0.5	0.72–2.20 mPa·s in the Temperature Range 150–450 °C	Chen et al. [61]
K2CO3-Na2CO3-Li2CO3	Alumina	1.0	35% Increase	Grosu et al. [62]

summarizes the main findings regarding the viscosity of the molten salt nanofluids.

4.3. Latent Heat

The latent heat of the molten salt nanofluids generally exhibits two tendencies. On one hand, the latent heat was reduced with the increasing concentration of the nanoparticles [63]. All the investigated nanoparticles followed this trend, exhibiting low latent heat enhancements with a maximum in the order of 5%, or even a decrement. Moreover, a parabolic correlation between the latent heat and the incorporated fraction of nanoparticles was found [64]. The nanofluids, having 1% wt. of nanoparticles, presented the maximum augmentation of the latent heat, regardless of the nature of the nanoparticles. Those nanofluids also presented the highest increases in the specific heat, and some authors proposed that these facts might be closely linked. Additionally, the researchers Chieruzzi et al. [12] proposed that the increment in the latent heat may be caused by the layers of small agglomerates at which the nanoparticles would be confined, given that the salt fraction trapped on the agglomerates might require higher degrees of energy to be melted, increasing the latent heat capacity. The authors Lee and Jo [65] investigated the latent heat of melting of the carbonate molten salt Li₂CO₃-K₂CO₃, and nitrate molten salt NaNO₃-KNO₃, with the inclusion of graphite nanoparticles. The authors stated that the latent heat of the nanofluids was ameliorated with the incorporation of nanoparticles and was also affected by the molar ratio of the inorganic salt mixtures. The results showed that for the carbonate molten salt, the latent heat was raised by up to 28.8%, whereas with nitrate salt mixtures, the latent heat was not considerably altered when using a 0.1% wt. of graphite nanoparticles, but it was increased by up to 9.9% at a 1.0% wt., and the magnitude of this increase was proportional to the molar fraction of NaNO₃. Likewise, the latent heat of the nanofluids was dependent on the molten salt composition. Hence, the latent heat of the molten salt nanofluids depends on the concentration of the nanoparticles and on the composition of the inorganic salts that are closely linked with the thermal energy storage capability. Furthermore, Qiao et al. [41], as already mentioned, prepared molten salt nanofluids based on the single salts NaNO₃, LiNO₃, and KNO₃, with the incorporation of silica nanoparticles. This time, the influences of the nanoparticle’s size modification from 15 nm to 5 µm, weight fractions between 0.5% and 4%, and temperature values ranging from 200 °C to 380 °C in the latent heat of the nanofluids were investigated. With the incorporation of 15–20 nm and 20–30 nm nanoparticles, the latent heat was reduced with the increasing concentration of the nanoparticles. The NaNO₃ with silica nanoparticles showed a lower latent heat than that of the pure salt. At 4% wt., the latent heat of the NaNO₃ with 15 nm to 20 nm and 20 nm to 30 nm nanoparticles was 167.2 and 162.1 kJ/kg, respectively. With the addition of 60 nm to 70 nm nanoparticles, only negligible alterations were verified on the latent heat with the inclusion of 0.5%, 1%, and 2% of silica nanoparticles. Nevertheless, the latent heat decreased to 166.7kJ/kg with the incorporation of 4% of silica nanoparticles. The addition of 1–5 µm nanoparticles did not appreciably affect the latent heat. The addition of 15 to 20 nm nanoparticles decreased the latent heat of LiNO₃, and the extent of this reduction was dependent on the concentration of the nanoparticles. With a concentration of 0.5% of nanoparticles, the latent heat was reduced to 368.8 kJ/kg and decreased to 351.6 kJ/kg at 4% of nanoparticles. For the 20 nm to 30 nm silica nanoparticles, the addition of 0.5% or 1% enhanced the latent heat, while an increase to 2% and 4% exhibited a decrement in the latent heat in reference to that of the base salt itself. The highest value of the latent heat of 377 kJ/kg was at 1% wt. When the concentration was augmented from 0.5% to 1%, the latent heat increased gradually; when the increase was from 1% to 4%, the latent heat decreased appreciably. Concerning the KNO₃ salt, the addition of 15 nm to 20 nm silica nanoparticles led to changes to the latent heat and the extent of the changes depended on the concentration of the nanoparticles. The largest latent heat of around 99.9 kJ/kg was achieved at 0.5% wt. The incorporation of 20 nm to 30 nm silica nanoparticles into the KNO₃ gradually decreased the latent heat as the concentration of the nanoparticles was increased from 0.5% wt. to 4% wt. The largest latent heat of around 100.3 kJ/kg was obtained with 0.5% of silica nanoparticles. For the 60–70 nm silica nanoparticles, the latent heat increased gradually to near 100.8 kJ/kg as the concentration of the silica nanoparticles increased to 1% and then decreased with a further increasing concentration of the nanoparticles. Additionally, the addition of 1 nm to 5 nm nanoparticles had little effect on the latent heat as the concentration of the silica nanoparticles was of 1% or 2%. However, with 0.5% and 4% of silica nanoparticles, a lower latent heat was observed, as compared with that of pure salt. Hence, it can be stated that the latent heat of the NaNO₃ salt was decreased or kept stable with the inclusion of silica nanoparticles, whilst for the LiNO₃ and KNO₃ cases, their latent heat increased to around 377 kJ/kg with the addition of 20 nm to 30 nm nanoparticles at 1% and to near 100.8 kJ/kg with 60 nm to 70 nm nanoparticles at 1%, respectively. In sum, it should be emphasized that there is no general trend to report considering the latent heat enhancement of the molten salt nanofluids. Taking into account the published findings about the latent heat, few authors found a parabolic correlation of the latent heat increase with the addition of nanoparticles, some authors found a linear increase and others even reported a decrease with the augmentation of the concentration of the nanoparticles.

4.4. Thermal Stability

There is a lack of knowledge about the impact of the inclusion of nanoparticles into a molten salt nanofluid on the thermal stability and kinetics associated with the decomposition of the base fluid, which are relevant to the use of these nanofluids in thermal energy storage applications in concentrated solar power plants usually operating from 250 °C to 400 °C. The published results showed that the addition of nanoparticles help to improve the thermal stability of the molten salt nanofluids [66], but on the contrary, may decrease it [67]. Even no significant alterations were reported by Myers et al. [68] with the inclusion of copper oxide nanoparticles into molten salt base fluids. Additionally, the thermal stability of solar salts is directly linked with the nitrate-nitrite conversion. The monitorization of the decomposition rate of the nitrates is essential to determine the stability of the nanofluids, given that the nitrite formation promotes the decomposition of the molten salt, and to enhance the corrosion rates. Furthermore, the work carried out by Salim and Rahman [69] attempted to evaluate the effect of the concentration of the nanoparticles on the thermal stability of lithium and potassium binary carbonates, having a 62:38 molar ratio and the addition 2%, 4%, and 6% wt. of graphite nanoparticles. The different curves from the initial thermal cycle exhibited diverse mass loss magnitudes of up to 100 °C, and the mass fraction was not considerably varied with the temperature values for the remaining thermal cycles from 150 °C to 560 °C. At 2% wt., the mass loss of up to 100 °C was considerable, indicating that the nanoparticles retained a higher moisture content than that of the molten salt itself, which in turn negatively affected the thermal energy storage capability of the nanofluid. Additionally, a decrement of the retained water vapor was noticeable with the increasing concentration of the nanoparticles, indicating that this increase triggered the mechanism that hindered the absorption of moisture by the nanoparticles. Hence, the authors argued that the percolated network observed in the SEM images prohibited the water from penetrating the nanoparticles through the formation of a layer on their outer surface. Consequently, higher amounts of percolated network resulted in lower moisture contents, ensuring the low-pressure superior performance of the nanofluids as thermal energy storage fluids. Moreover, the authors Song et al. [24] prepared NaNO₃ salts with the addition of silica, silica/titania, and expanded graphite nanoparticles. For the thermal stability of NaNO₃(SiO₂ + TiO₂), graphite was evaluated with weight fractions of 1% SiO₂, 0.9% TiO₂ + 0.1% SiO₂, and of 7%, 10%, 15%, and 20% of graphite nanoparticles with the aid of a furnace heated up to a temperature of 400 °C at a heating ramping of 10 K per minute to complete the phase transformation endothermic process. Then, the molten salt nanofluids were kept at 400 °C for 10 min and after that were cooled to finish the phase change process. The thermal cycling stability was determined through 300 of the above-mentioned heating and cooling thermal cycles. The mass loss was larger in the first 50 thermal cycles, changed much less in between 50 and 150 cycles, and kept constant after 200 cycles. Along the thermal cycling, the mass loss rates of the NaNO₃/(SiO₂ + TiO₂)/EG system were up to 15.9%. The main reason for the obtained mass loss rates was that a small quantity of molten salt on the surface of the sample was easy to leak under high-temperature conditions, but with an increasing number of thermal cycles after the leaking of a small quantity of sodium nitrate on the surface, the inside was adsorbed capillary forces induced by the graphite pores, and the mass remained unchanged, ensuring a good thermal stability. After the referred 300 heating and cooling cycles, the phase change temperature of the NaNO₃/(SiO₂ + TiO₂)/EG composite remained practically unchanged, the latent heat decreased by 7.4%, and the specific heat before and after the phase change was found to be reduced by 6.1% and 6.0%, respectively. Therefore, the molten salt nanofluid studied in this described work presented an improved thermal stability.

Table 5. Main findings of the thermal stability of the molten salt nanofluids.

Molten Salt	Nanoparticles	Concentration % wt.	Thermal Stability Findings	Authors/Reference
NaNO3-KNO3	Hexagonal Boron Nitride	0.5, 1.0 and 1.5	16% Increase at 1.5% wt.	Aslfattahi et al. [70]
NaNO3-KNO3	Silica	1.0	The heat treatments, both exposure to constant high temperature and low-high temperature circulation, can decrease the thermophysical properties of samples significantly. With equal operation time, the decrease rate of the cycled sample is lower than that of the sample exposed to constant high temperature	Li et al. [43]
Ca(NO3)2-KNO3-NaNO3-LiNO3	Silica	1.0	Good thermal stability under the same conditions. Moreover, the change in specific heat was minimal after 2000 h, which was less than 5%.	Chen et al. [61]
LiNO3-KNO3-Ca(NO3)2	Molybdenum Disulfide and Copper Oxide	0.5, 1.0, and 2.0	Good thermal stability up to 400 °C The thermal stability was lower with the molybdenum disulfide and increased with the copper oxide, since the pure MoS2 decomposes around 400 °C	Madathil et al. [71]
K2CO3-Li2CO3-Na2CO3	Silica	1.0	Good thermal stability and no considerable deterioration in the specific heat after the thermostatic at 600 °C for 150 h and thermal shock of 50 cycles	Sang et al. [51]
Li2CO3Na2CO3, Li2CO3-K2CO3, and Li2CO3-Na2CO3	Alumina	1.0	Good thermal stability up to 600 °C. However, the Li2CO3-Na2CO3 exhibited poor thermal stability since its nanofluids decomposed after 470 °C	Rizvi and Shin [67]

summarizes the main findings concerning the thermal stability of the molten salt nanofluids.

4.5. Corrosivity

A complete understanding of the effect of the incorporation of nanoparticles on the corrosivity of the molten salt nanofluids has not been achieved yet. To obtain an improved knowledge on the matter, Nithiyanantham et al. [72] evaluated the impact of the inclusion of nanoparticles in molten salt nanofluids on the corrosion rates of carbon steel. The researchers stated that if the undesirable air bubbles trapped in between the nanoparticles were not the predominant effect, it was possible to achieve diminished corrosion rates because of the inclusion of the nanoparticles into the oxidation layer. In their experiments the authors dispersed alumina and silica nanoparticles in the eutectic NaNO₃-KNO₃ salt having a 51:49 mass ratio and evaluated their impact on the corrosion of the A516.Gr70 carbon steel, usually applied in the construction of thermal energy storage units. Additionally, it can be proposed that the addition of nanoparticles into the corrosion layer may decrease the corrosion rate by stabilizing the oxidation scale, whist the microbubble formation favors the corrosion process due to the enhanced oxygen amount in the involved system. To determine the evolution of the air microbubbles in the case of the eutectic molten salt, tests with the alumina nanofluid at temperature values of 310 °C and 390 °C were carried out. In the latter, the air escaped completely from the salt after only two hours, whereas in the former, even after 3 days of an experiment period, a few microbubbles were still observable. To verify the influence of the bubbles of air on the rates of corrosion, corrosion resistance tests were carried out by immersion with the eutectic salt and alumina and silica nanoparticles at a temperature value of 310 °C. It was verified that at a temperature of 310 °C, the corrosion layer, composed of Fe₂O₃ and Fe₃O_4, was larger for the nanofluids as compared to that of the salt itself. It was noticeable that at 310 °C, the nanoparticles dispersed in the molten salts caused a 2-fold enhancement of the corrosion rate in reference to that of the pure salt itself. On the contrary, at a temperature of 390 °C, the inclusion of nanoparticles resulted in smaller corrosion rates. In face of these results, the researchers hypothesized that the adverse effect of the air microbubbles was predominant at the temperature value of 310 °C, whereas such an effect becomes secondary at 390 °C due to the faster evacuation of the bubbles. The authors also stated that the nanoparticles decreased the corrosion rates regardless of the temperature of the corrosion experiments. Additionally, any chemical reaction between the nanoparticles and the salt was not reported, which indicates that the nanoparticles did not alter the electrochemical potential of the inorganic salt. After the corrosion tests, 2 µm-sized agglomerates were formed, which were much larger than the ones of 200 nm observed in the as prepared nanofluids. The obtained results suggested that even though, because of the agglomeration effect in the course of the corrosion tests, the mass fraction of the nanoparticles may be different to the initial mass fraction of 1%, the surface of the carbon steel was always exposed to the silica and alumina nanoparticles, enabling physical and chemical interactions. Yet, it should be referred that the agglomeration reduced the fraction of the elements of the nanoparticles that will chemically react. Therefore, the agglomeration may reduce the counter corrosion role of the nanoparticles and should be averted. Additionally, adding nanoparticles of alumina and silica strongly reduced the corrosion rate of the metal at a temperature of 390 °C. After 1500 h of the immersion experiment, the corrosion layer suffered a 3-fold reduction in reference to the same layer of the molten salt alone. Additionally, an heterogeneous oxidation having defects was found in the pure nitrate salt by direct observation of the surface of the steel after the corrosion experiments. Oppositely, the uniform oxidation of the surface of the metal was revealed after the corrosion tests with the nanofluids. Moreover, in the study performed by Yang et al. [73], the corrosion features of the carbonate salt Li₂CO₃:Na₂CO₃:K₂CO₃ were inferred, having a 31:34:35 mass ratio with the addition of zinc oxide nanoparticles. The molten salt nanofluids exhibited reduced corrosivity with a corrosion mass gain peak and corrosion rate of a austenitic stainless-steel of 1.72 mg.cm⁻² and 9.6 μm.y⁻¹, decreasing by 27.4% and 25.8% in comparison to the base salt itself, respectively. In the corrosion tests, the molten salt nanofluids were placed inside crucibles, and samples of 201, 304, and 316L stainless steels were immersed into the molten salt nanofluids. After this, the crucibles were heated in a furnace that was kept at a temperature of 550 °C for different periods. The authors observed that the mass of the metal samples increased under the corrosion of carbonate, and the mass enhancement augmented with increasing corrosion time. On one hand, using the pure salt, the mass gain of the 201, 304, and 316L stainless-steels after five days of corrosion were 2.51, 2.36, and 1.71 mg.cm⁻², respectively. On the other hand, employing the molten salt nanofluids, the mass gain of the stainless-steels after five days of corrosion were 2.03, 1.72, and 1.58 mg.cm⁻², respectively, decreasing by 19.1%, 27.4%, and 7% in comparison with that of the carbonate salt itself. In terms of corrosion rates, using the pure carbonate salt, the corrosion rates of the stainless-steels were 13.8, 12.9, and 9.2 μm.y⁻¹, respectively, while in the molten salt nanofluids, the corrosion rates of the stainless-steels were, respectively, 11.2, 9.6, and 8.6 μm·y⁻¹, which decreased by 18.7%, 25.8%, and 6.5% compared with the pure carbonate. As concluding remarks, the corrosion of the stainless steels in the carbonate salt nanofluid was considerably reduced as compared to that of the salt. The authors interpreted this evidence based on the migration of a fraction of the nanoparticles from the molten salt nanofluid to the corrosion layer, which contributed to the corrosion chemical reaction and produced oxides that hindered the contact between the molten salt and the steel, slowing down the corrosion rate. Furthermore, the corrosion of the AISI 304L stainless-steel in solar salt nanofluids was studied at 565 °C for 432 h by Ma et al. [74]. The corrosion experiments were carried out through the coupon bomb configuration, mimicking the walls of the vessels in concentrated solar power plants. According to this configuration, eight stainless-steel AISI 304L coupons were immersed in the solar salt inside a stainless-steel AISI 316 bomb. In the testing bombs set, two were controls and not filled with salt, two more were filled with the solar salt itself, the other two were filled with the solar salt-based nanofluid with silica nanoparticles, and the remaining duo was filled with the solar salt-based nanofluid with alumina nanoparticles. Furthermore, all the coupons were wrapped with a ceramic wire for being electrically isolated, in this way, ensuring that there was no contact between coupons and with the container since this contact might resulted in galvanic corrosion, amplifying the corrosion of the metal in the molten salt. A quantified analysis of the corrosion magnitude of the stainless-steel coupons in the molten salt was done by recording the initial mass for each coupon and determining the mass loss after 432 h of stage in a furnace. After the test completion, all the coupons were taken out from the bomb by fusing the salt at a temperature of 300 °C in a furnace, and after that, each coupon was descaled through a suitable procedure for the corrosion layer removal. Moreover, the final mass of the coupon was noted for calculating the corrosivity of the diverse samples of molten salt. Hence, the corroded depths were determined through the division of the mass loss per unit area by the density of the coupons. The results showed that the mass loss per unit area of the coupons in air, solar salt alone, solar salt-based nanofluid with silica, and the solar salt-based nanofluid having alumina were around 0.9 mg/cm², 1.3 mg/cm², 0.3 mg/cm², and 0.4 mg/cm², respectively. The results revealed a similar mass loss in air and pure salt, suggesting that the oxidation rate between the coupons and the nitrate ion at a temperature of 565 °C was similar to the oxidation rate between the coupons and air. Additionally, it was confirmed that the incorporation of silica and alumina nanoparticles prevented the corrosion of the stainless-steel by the action of the molten salt, given that the mass loss was decreased by more than 50%. The authors also noted that the procedure of descaling exhibited only little effect on the mass loss for the coupons heated in air and for pure solar salt samples. Taking into account that the descaling protocol with hydrochloric acid solution produced a surface lustrous finish, it can be stated that the oxidation reaction between the hydrochloric acid and the surface of the steel is much more vigorous than the oxidation reaction between the steel and the KMnO₄ and the NaOH descaling solutions. As the verified mass loss values were only in the order of milligrams, even very small mass losses derived from the descaling solutions would induce considerable uncertainties in the measured mass loss values and consequent misinterpretation of the corrosion results. Therefore, it can be concluded that the descaling procedure is a relevant influencing factor of the corrosion tests. Moreover, the chemical composition of the corroded layer on the steel in air and pure solar salt might be different from the one produced with the solar salt nanofluids. For instance, a few mixed oxides may surge in the layer between the coupons and the nanofluids, giving different results after being submitted to the descaling procedures. The coupons immersed in the solar salt itself exhibited alike superficial features with the ones detected in the air-heated coupons, in which the randomly existing clusters and defects were dispersed at some isolated points. The coupons immersed in the solar salt silica nanofluid showed a high density of precipitated silica nanoparticles on the coupons surface. These nanoparticles played the role of a passivation layer that prevented the corrosion of the coupons. For the coupons immersed in solar salt-based alumina nanofluids synthesized from one-step method, a precipitation with a lower density area was verified to be present on the surface of the coupons. Additionally, a significant fraction of the surface area of the coupon was not completely covered by the alumina layer. This may be because of the employed one-step protocol did not produce dispersed nanoparticles, but instead interconnected network nanostructures. These possess a lower propensity to spread homogeneously on the surface because of their much larger size, as compared to the precipitation of the silica nanoparticles. Additionally, the inclusion of nanoparticles in the molten salt has already been revealed to affect the corrosion rates evolution diversely. Additionally, the nanoparticles in the Solar salt-based alumina nanofluids were produced in-situ and, consequently, did not include any trapped air, hence, minimizing the adverse impact from the trapped air bubbles. Another counter corrosion mechanism was already confirmed [75] that is the inclusion of stable oxide nanoparticles in the corrosion layer, which averts the molten salt diffusion. The presence of external oxides in the corrosion layer can lead to the production of mixed oxides that offers augmented resistance to the nitrate and nitrite ion penetration. In addition, superficial adsorption may also be of relevance in the corrosion rates of the nanofluids. The inorganic solar salt when in a liquid state is composed of free Na⁺, K⁺, and NO^3- ions. Because of the exposed crystal facets and the existence of defects on the surface of the oxide nanoparticles, these free ions can be preferentially adsorbed in function of the dissimilar affinities of each ion to be adsorbed in a particular surface. The surface charge distribution can be modified through the preferential adsorption of ions phenomenon. Considering the alumina nanoparticle, the NO³⁻ anion is chemisorbed to the metal cation present in the surface of the nanoparticles with three different bonding structures [76]. The preferential adsorption of NO³⁻ on the surface of the alumina nanoparticles caused a gradual buildup of a negatively charged surface potential. This layer acts as a galvanic cell, given that the surface of the nanoparticles acquired an electric potential with respect to the stainless-steel surface and the molten salt, leading to the protection of the coupon from being oxidized by the molten salt. Hence, even in the cases where the included nanoparticles do not build a homogeneous coating on the coupons, the nanoparticle are still capable of reducing the corrosion rates within their area of influence for all the precipitated nanoparticles on the coupons. Hence, the precipitation of nanoparticles with a non-homogeneous covering of the surface could decrease the corrosion rates in the coupons. The coupons soaked in the molten salt revealed that even after the descaling process, there were remaining superficial residues on the surface coming from the molten salt crystals, which suggests that the descaling method was indeed efficient in removing the corroded areas without provoking any harm to the base metal. Thus, the mass loss analysis accurately offers the corrosion degree of the coupons in the molten salts. Finally, it was reported that the molten salt nanofluid prepared by the one-step method obtained the same counter corrosion degree and improvement in the thermophysical properties of those confirmed in the nanofluid prepared by the two-step method. In essence, a marked attention should be addressed to the corrosivity effect when doping the molten salts with nanoparticles. This issue is notably important to decrease the investment, operation, and maintenance costs. Nevertheless, the actual knowledge concerning the compatibility between the molten salt-based nanofluids and the materials of the pipping and storage tanks of thermal management fluidic systems is still in its infancy. The results are controversial, demonstrating reduced corrosion rates and enhanced material compatibility and increased corrosion rates with the implementation of certain options. Additionally, there is a pronounced interest in the corrosion effects of the pure molten salts. Nevertheless, further studies evaluating the effect of the addition of nanoparticles and possible corrosion inhibitors are needed, considering well-defined dynamic and static corrosion situations.

5. Specific Heat Enhancement Mechanisms

The specific heat of the molten salts may be taken as their thermophysical property with paramount importance, being the focus of the scientific community of the field. It was already found that the specific heat was enhanced with the use of molten salt nanofluids. However, some authors have noticed a decrease [17,77] in the specific heat, instead of an increase. Additionally, the inconsistences in the published articles between the results obtained for the same nanofluid are usual and the fact is most likely related to the differences on the used salts and nanoparticles, preparation methods, and characterization techniques employed to determine the specific heat. The following sub-sections describe the main influencing parameters and their impact on the specific heat of the molten salt nanofluids.

5.1. Salt Composition Ratio

In the study carried out by the authors Sang et al. [78], the carbonate salt K₂CO₃-Li₂CO₃-Na₂CO₃ was synthesized, having a 4:4:2 mass ratio with the incorporation of silica nanoparticles at 1% wt. The synthesis procedure involved the evaporation temperatures of 160 °C, 180 °C, 200 °C, and 240 °C. The authors intended to study the impact on the specific heat of the molten salt nanofluid of the composition ratio of the salt. It was revealed that the composition ratio of the ternary carbonate was altered during the evaporation in an electrothermal drier. Apart from a difference in the water solubility of the carbonates, it was also confirmed that the mode of heating affected the composition ratio of the salt. When evaporating at 180 °C in an electrothermal drier, the increase of the specific heat of the nanofluid was between 79.9% and 113.7% in the temperature range from 500 °C to 540 °C, as compared to the salt prepared by the direct mixing method. On the one hand, the temperature of evaporation played a relevant role on the formation of nanostructures in the nanofluids. On the other hand, the marked enhancement of the specific heat can also be attributed to the alteration in the mass fraction of Li₂CO₃ that was enhanced by more than 32%. The composition of the molten salts with an evaporation temperature of 180 °C was different from that of the salt synthesized by the direct mixing method. Furthermore, the different heating procedures in the evaporation process using, for instance, a hot plate or an electrothermal drier, can also lead to strong alterations in the composition ratio of the resultant salts. In addition, the authors Jo and Banerjee [79] investigated the impact of the composition of the carbonate salt Li₂CO₃-K₂CO₃ mixed in different proportions. The authors synthesized the nanofluids by adding 1% wt. of multi-walled carbon nanotubes to diverse carbonate mixtures. The researchers demonstrated that the Li₂CO₃-K₂CO₃ molten salt with a 62:38 molar ratio induced a minimum in the specific heat. The molten salt with 1% wt. of silica nanoparticles exhibited the opposite tendency since the specific heat was augmented close to the eutectic composition of the mixture. Additionally, the authors reported that the nanofluids having a high/low fraction of Li₂CO₃ exhibited a very small increment, or even in some cases a decrement, in the specific heat. Finally, it can be concluded that the composition ratio of the molten salt nanofluids affects their specific heat capacity enhancement and the changes in the composition ratio can be determined by different heating modes in the evaporation process of the synthesis method. Different composition ratios may lead to the production of different nano-scaled structures in the nanofluids that, in turn, directly affect their final specific heat capacity.

5.2. Size and Shape of the Nanoparticles

Many published studies have dealt with the impact of the size of the nanoparticles on the specific heat of molten salt nanofluids. In this direction, the authors Tiznobaik and Shin [80] reported that the specific heat of a molten salt-based silicon dioxide nanofluid increased by up to 25% when compared with the specific heat value of the molten salt itself for any size of the incorporated silicon-dioxide nanoparticles, having diameters between 5 nm and 60 nm. The researchers Jung and Banerjee [81] observed that the specific heat of nanofluids having carbon nanotubes suspended in the Li₂CO₃-K₂CO₃ molten salt with a 62:38 molar ratio was reduced with increasing dimensions of the nanotubes. Only a negligible enhancement in the specific heat of the nanofluids for diameters superior to 20 nm was found. Given that the specific surface area increased with decreasing nanotube dimensions, the nanotubes may exhibit superior thermal properties than those of the bulk liquid, which may contribute more to the specific heat improvement of the molten salt nanofluids. In addition, LiNO₃-based nanofluids with the incorporation of alumina nanoparticles of 10 and 200 nm of diameter were investigated by Thoms [77]. The author reported a decrement in the specific heat for the nanofluid with the 10 nm nanoparticles, and a negligible increment for the one having 200 nm nanoparticles. It was stated that the impact of the absorbed layer on the bigger particles was almost negligible because of their reduced interfacial area in comparison to a similar fraction of smaller particles. The researchers found that the salt specific heat decreased with the addition of the alumina nanoparticles, and the decrement in the specific heat was greater with smaller nanoparticles. Only in the cases where very big nanoparticles were added was a specific heat increase of only inferior to 4% as compared with the specific heat of base fluid induced. Furthermore, the studies performed with silica nanoparticles of different sizes revealed an increase of the specific heat of the molten salt with the incorporation of nanoparticles. Moreover, the researchers Dudda and Shin [82] included 1% wt. of silica nanoparticles, having sizes between 5 nm and 60 nm, in the solar salt. There were reported increases in the specific heat with the increasing size of the nanoparticles: a 10% increase with the smallest 5 nm nanoparticles and a 28% increase with the biggest 60 nm nanoparticles. Additionally, the presence of nanostructures having improved thermal characteristics was confirmed, and their quantity increased with the increasing size of the nanoparticles. The authors suggested that the smaller nanoparticles had more constraints to be uniformly dispersed and tended to agglomerate. The effect was responsible for the reduction of the effective heat transfer surface of the nanoparticles, hindering the production of the referred nanostructures. Furthermore, Riazi et al. [83] performed an investigation to address the impact of the shape of the nanoparticles on the specific heat of the molten salt nanofluids. With this purpose, silica nanoparticles with different morphologies were mixed into the solar salt. It was found that the specific heat of the nanofluids clearly depended on the morphology of the nanoparticles, and the maximum specific heat increase of 17.6% was observed for the more spherical and smaller nanoparticles with dimensions between 19 nm and 92 nm. On the other hand, the specific heat did not suffer any alteration, as compared to that of the solar salt itself when the nanoparticles were considerably agglomerated, giving form to larger clusters. Most of the published articles reported an improved specific heat in the cases where higher amounts of nanoparticles were added to the base molten salt. However, the opposite trend has already been confirmed [84] and the absence of close correlation between the size of the nanoparticles and the specific heat improvement was even reported [85,86]. By way of conclusion, it can be stated that less attention was offered to the impact of the size and shape of the added nanoparticles on the specific heat of the molten salt nanofluids by the researchers, as compared, for instance, with the impact of the nanoparticle concentration. Additionally, non-consistent findings were reported regarding the impact of the morphology of the nanoparticles on the specific heat of the molten salt nanofluids. Some studies claimed only a negligible effect on the specific heat by changing the dimensions of the nanoparticles and others observed a near 18% increment with the tested smaller nanoparticles and a near 28% increase with the larger tested nanoparticles. Hence, more experimental works should be carried out to evaluate the influence of the morphology of the incorporated nanoparticles in the molten salt base fluids and, consequently, to precisely interpret the anomalous behavior of the molten salt nanofluids. In addition, new correlations extracted from experimentally obtained results should be developed involving more factors than only the operating temperature.

5.3. Concentration of the Nanoparticles

The alterations on the specific heat of the molten salt nanofluids induced by the inclusion of different fractions of nanoparticles is the most widely studied issue through the available literature on the field. The goal is to determine the concentration of nanoparticles that maximizes the specific heat of an arrangement of salts and nanoparticles. In this direction, the solar salt NaNO₃-KNO₃ having a 50:50 molar ratio with the incorporation of different fractions of silica nanoparticles has been widely studied. The authors Mondragón et al. [85] achieved an enhancement of 31.1% in the specific heat with 0.5% wt. of silica nanoparticles. The authors proved that this concentration enabled a homogeneous dispersion within the matrix of the molten inorganic salt, thus enhancing the interfacial area to interact with the inorganic molten salt. A lower concentration of nanoparticles led to much dispersed and isolated nanoparticles and agglomeration promotion. In the study performed by Ma et al. [87], solar salts with alumina nanoparticles at various weight fractions were prepared by the one-step method where the nanoparticles were produced in the salt melt directly from the precursors. A heterogeneous enhancement in the specific heat of the molten salt nanofluid was verified with mass fractions of alumina nanoparticles between 0.5% and 1.5%. An increase peak of 38.7% in the specific heat with a concentration of 1% wt. was found. Through the SEM imaging observation of the morphological characteristics of the nanofluid, the authors claimed that the chemisorption of the free ions on the nanoparticles surface and associated extended ionic response could provoke the production of secondary nano-scaled structures in the molten salt nanofluid. These network-connected secondary structures are considered by some authors to be the major contributor to the specific heat improvement and are an extension of the compressed liquid layer formed in the nanoparticle nearby region. Additionally, these structures usually occupy a great part of the available total volume of the nanofluid, even with low fractions of added nanoparticles. Hence, the most important conclusion of the mentioned work is that the contribution of the secondary nanostructures in the nanofluid system accounting with the bulk solvent, the nanoparticles, and the compressed liquid layer on the surface of the nanoparticles, is an issue to account within the specific heat improvement. The investigation team also showed that for a fixed concentration and size of the incorporated nanoparticles, the volumetric fraction of the secondary nanostructures producing the network of percolation between neighboring particles can initially increase and then decrease with a rising concentration of the nanoparticles. Moreover, the non-consistent findings reported by the diverse researchers were also found in the solar salt with titanium oxide nanoparticles nanofluids. The published studies demonstrated that the inclusion of 2.0%, 5.8%, and 9.6% weight fractions of alumina nanoparticles decreased the specific heat of the base molten salt [17] and the inclusion of 1% wt. of titanium oxide nanoparticles led to a specific heat increment of 30.5% in comparison to that of the inorganic salt itself [88]. Additionally, a mixture of silica-alumina having a 82:18 mass ratio was incorporated in the referred molten salt, enhancing its specific heat by 22.5% with a weight fraction of 1% in the liquid state. The researchers confirmed the presence of a semi-solid nanolayer close to the surface of the nanoparticles as the major reason behind that increment. The SEM imaging observation revealed the uniform scattering of the nanoparticles, and any clustering for the 1% wt. concentration was not found. Additionally, other nanofluids were prepared by the addition of multi-walled carbon nanotubes in weight fractions of 0.05%, 0.1%, 0.5%, and 1.0% to the Li₂CO₃-K₂CO₃ molten salt with a 62:38 molar ratio [89]. The increase of the specific heat in the nanofluids with concentrations inferior to 0.1% wt. was of 9%, whereas the nanofluids with concentrations superior to 0.5% wt. exhibited a 17% increment in the specific heat. The authors stated that the ideal weight fraction considering these nanofluid was superior to 1.0%. In the experimental work performed by the researchers Schuller et al. [90], different weight fractions of alumina nanoparticles were scattered in the sodium nitrate and potassium nitrate eutectic mixture, having a 60:40 molar fraction. The results exposed that it was possible to establish a parabolic correlation between the specific heat and mass fraction of the added nanoparticles. Additionally, the nanofluid exhibited the maximum enhancement equal to 30.6% in the specific heat with 0.78% wt. of alumina when compared with the eutectic mixture alone. Moreover, the researchers also studied the stability of the specific heat values and found that the nanoparticle concentration with the highest specific heat was transferred from 0.78% to 0.3% when the nanofluids were re-tested after one and two months. The results revealed that a difference existed between the specific heat initial values and repeated ones, as the maximum difference was up to 9.75%. Unlike the original test, having a maximum specific heat of 1.92 J/g K at 0.78% of alumina nanoparticles, the maximum specific heats from the two repeated tests had maximized values at 0.30% of alumina nanoparticles of 1.90 J/g K and 1.89 J/g K. The investigation team highlighted that there were competing mechanisms affecting the specific heat of the nanofluids, including the concentration and possible agglomeration of the added particles, which in turn suggested that there shall be an ideal nanoparticle concentration for maximizing the specific heat. Furthermore, the authors Kim and Jo [91] prepared binary carbonate molten salts with the inclusion of graphite nanoparticles and investigated the impact of the concentration of the nanoparticles on the specific heat. With this purpose, the weight fraction of the nanoparticles was varied from 0.025% to 1.0% and the specific heat of the nanofluids was measured at temperatures between 150 °C and 560 °C. The authors noted that the specific heat grew with the increasing weight fraction of the nanoparticles and a considerable enhancement was verified at a weight fraction of 0.025% with a base salt composed of carbonates of lithium and potassium with a 75:25 molar ratio. Nevertheless, at nanoparticle concentrations higher than 0.1% wt., the enhancement was almost negligible. Additionally, the authors Ho and Pan [92] studied the ideal weight fraction of alumina nanoparticles incorporated in the Hitec molten salt for the specific heat enhancement of the molten salt nanofluid. The researchers found an ideal concentration of 0.063% wt. yielding the 19.9% maximum enhancement in the specific heat. Additionally, with the incorporation of 2% wt. of nanoparticles, the negative impact derived from the nanoparticles on the specific heat was notorious at all temperature ranges. The authors also stated that the ideal weight fraction of the nanoparticles was close to the one at which the contributions to the specific heat of the single nanoparticles and clusters (0.2 µm–0.6 µm long) were similar. As it will be discussed in the following sub-section, the preparation method and respective equipment impact the specific heat enhancement of the molten salt nanofluids very much and, considering this, the authors developed in this work an innovative sampling and processing apparatus, of which a schematic representation is shown in Figure 6, to prepare the molten Hitec nanofluids and to hinder an eventual precipitation of the added nanoparticles.

On the other hand, in the experimental work performed by the authors Liu et al. [93], silica nanoparticles having different sizes and silicon nitride and silicon carbide nanoparticles were scattered in the solar salt at weight fractions between 0.5% and 7%. The specific heat of the molten salt nanofluids, solar salt alone, and nanoparticles was measured at temperature values from 300 °C to 400 °C. The obtained results showed that the silicon nitride and silicon carbide particles had little impact on the specific heat of the solar salt at weight fractions between 0.5% wt. and 3 wt.%. Additionally, the specific heat of the 3% wt. silica nanofluids were enhanced by 4.3% to 9.7%, as compared with that of the Solar Salt itself. However, when 5 wt.% and 7 wt.% of silica with 30 μm were included, the specific heat was reduced by 10.6% and 9%, respectively. In addition, the techniques of characterization of the molten salt nanofluids demonstrated the production of sodium silicate caused by the reaction of silica and sodium nitrate that is interpreted by the researchers as being closely related to the reported reduction of the specific heat. In conclusion, it can be inferred that the already reported results for the specific heat capacity of the molten salt nanofluids are controversial in terms of the influence of the incorporated nanoparticles in this thermophysical property. Some studies claimed a considerable improvement in the specific heat of the molten salt nanofluids as compared with that of the base fluid mixture using small weight fractions of nanoparticles (e.g., 1% wt.). Oppositely, other studies clearly stated that the specific heat capacity of the molten salt nanofluids was reduced with an increasing concentration of nanoparticles.

5.4. Preparation Methods

It is noteworthy to highlight that the various preparation methods also impact the specific heat of the final molten salt nanofluid, given that the selection of one of the synthesis routes strongly affects the dispersion of the nanoparticles within the molten salt matrix. The synthesis procedures also have a great impact on the agglomeration or clustering of the added nanoparticles, and thus on the final size of the nanoparticles. Both factors define the specific heat magnitude of the nanofluid. Additionally, Banerjee [94] included 0.1% wt. of graphite nanoparticles to the Li₂CO₃-K₂CO₃ carbonate molten salt with a 62:38 molar ratio and the drying procedure was carried out within two different containers. The first method dried the dispersion in a glass vial and the second one in a Petri dish. The dried carbonate nanofluids using the Petri dish with more surface obtained a greater increment in the specific heat capacity. The researchers interpreted the fact based on the smaller drying surface of the glass vial that required an extended drying time to dehydrate the nanofluid. Hence, the nanoparticles had the tendency to be more agglomerated in the aqueous solution when using this type of container. The uniformity of the nanofluids was low and the specific heat was lower than that of the nanofluids dried on the Petri Dish. In addition, the authors Muñoz-Sánchez et al. [95] studied the impact of the volume of water required to dissolve the salts and scatter the nanoparticles. In this direction, the investigation team synthesized nanofluids of solar salts with the incorporation of 1% wt. of silica nanoparticles using 31.6 mL, 6.5 mL, and 0.97 ml of water per gram of dry material. It was reported that the specific heat of the nanofluids with the addition of the highest and the lowest water volume were 1.61 J·kg⁻¹·K⁻¹ and 1.58 J·kg⁻¹·K⁻¹, respectively, whilst the intermediate molten salt nanofluid presented the lowest specific heat value of 1.38 J·kg⁻¹·K⁻¹. The authors interpreted these differences based on the agglomeration level of the nanoparticles on the different nanofluids. Moreover, Riazi et al. [83] investigated the impact of the power of sonication on the clustering of the nanoparticles using the solar salt with the inclusion of two types of silica nanoparticles at 1% wt. The research team reported an ideal ultra-sound energy that minimized the size of the nanoparticles of 4000 J·m·L⁻¹ for one type and 3000 J·m·L⁻¹ for the other. It was also reported that a considerable enhancement of the clustering effect happened in the cases where the ultrasonication energy was beyond 7000 J·m·L⁻¹ on the two different nanofluids. On the other hand, in the work performed by Chieruzzi et al. [22], the results from an innovative method and the results from the hot-plate procedure were compared. The authors prepared the solar salt with silica, alumina, and silica/alumina nanoparticles through extrusion at high temperature. This technique promoted a decrement in the specific heat of the nanofluids, as compared to the solar salt alone. Oppositely, the specific heat was enhanced in the case where the nanofluids were prepared by the common method, regardless of the concentration and type of the nanoparticles. The researchers explained this behavior based on the poor scattering of the nanoparticles produced by the extrusion technique. Additionally, the authors Tiznobaik and Shin [80] inferred the potential impact of the presence of impurities in the specific heat. With this purpose, a Li₂CO₃-K₂CO₃ molten salt with a 62:38 molar ratio and 1% wt. of silica nanoparticles nanofluid was synthesized and exposed a 26% enhancement in the specific heat with respect to that of the inorganic salt itself. Nonetheless, the mentioned enhancement vanished totally in the cases where 0.02% wt. of NaOH was added to the nanofluid. The authors explained the occurrence by the excessive presence of OH^- ions, which might limit the interactions between the suspended nanoparticles and the base molten salt. Such an effect might provoke the absence of nanostructures with superlative thermal properties and, consequently, this explains the similarity of the specific heat of the nanofluid to that of the salt alone. In essence, it can be highlighted that the different synthesis approaches of the molten salt nanofluids led to different specific heat enhancements. Additionally, varying parameters in the same preparation method, such as the different heating/cooling rates and different dehydration process durations, may also influence the final specific heat of the nanofluids. There were already reported specific heat increases, as compared with the pure molten salts of near 31% [31] with the liquid dispersion method, of around 39% [25] using the mechanical dispersion method, and of 7.5% [30] with the in-situ production method. As already mentioned, the different preparation methods may lead to different dispersion degrees of the nanoparticles and, consequently, affect the agglomeration of the suspended nanoparticles and their final morphology, which in turn directly affect the specific heat capacity of the molten salt nanofluids.

5.5. Interfacial Thermal Resistance

The interfacial thermal resistance is increased between the nanoparticles and the molecules of the fluid because of the high surface area per unit volume of the nanoparticles. In other words, the specific heat of nanofluid is increased when the nanoparticles were added in the liquid solution because of the enlarged surface area. According to a previous study [96], the enhanced interfacial thermal resistance acts as the additional thermal storage between the nanoparticles and base fluid molecules. Since the interfacial interaction between interfacial molecules and nanoparticle atoms is increased, the interfacial thermal resistance is increased between interfaces. Therefore, this phenomenon may be responsible for the enhanced specific heat of the nanoparticles and base fluid. Regarding, for instance, the multi-walled carbon nanotubes, it can be found in the literature that the influencing features of the carbon nanotubes molten salt nanofluids thermal performance include the degree of dispersion of the carbon nanotubes, the concentration and morphology of the carbon nanotubes, and the interface thermal resistance of the carbon nanotube/molten salt. Additionally, of all these factors, the interface thermal resistance can be taken as the most influencing one, since it may block off the heat flow and induce the weakening of the interface interaction. Additionally, the interfacial thermal resistance rejects the effect of the heat transfer between the incorporated nanoparticles and the salt molecules, which makes the heat conduction rate slower and, consequently, the specific heat results enhanced. Moreover, two main mechanisms for the heat transfer enhancement in the molten salt nanofluids can be eligible: one is the nanotube-molten salt heat transfer and the other is the nanotube-nanotube heat transfer. These mechanisms are basically the interfacial thermal resistance and the thermal contact resistance. In the cases where the concentration of the nanotubes is low, the interfacial thermal resistance is the main influence factor of the nanofluids, and when their concentration is high, the nanotube-nanotube heat transfer is enhanced and the interfacial thermal resistance is deteriorated, causing a decrement in the specific heat of the nanofluids. Additionally, it was found that the concentration of the carbon nanotubes did not exhibit a linear relationship with the specific heat, but it can be stated that there is an ideal concentration range for the nanofluids. Hence, an excessive or insufficient concentration of the nanotubes are both disadvantageous for the interfacial thermal resistance. On the other hand, an excessive moisture content and inefficient ultrasonication may lead to the agglomeration of the carbon nanotubes, which are then unable to produce an effective interfacial thermal resistance and, hence, the salt molecules absorb a greater heat charge in the same temperature range, leading to the specific heat reduction. Concerning the addition of gold nanoparticles, they have stable physical and chemical properties, which enhanced the specific heat of the nanofluids. As compared to the carbon nanotubes, the gold nanoparticles are spherical, making their surface area large enough to scatter phonons. Therefore, a nanofluid incorporating gold nanoparticles has a very high interfacial thermal resistance and increases in the specific heat. It was already found in the literature that the larger nanoparticles on one hand can increase the density of the low-frequency vibrational phonon modes, but on the other hand reduce the interfacial thermal resistance and coupling loss of heat conduction from the interior of the nanoparticles to the nearby salt molecules, having phonons of different vibration frequencies. Hence, the smaller nanoparticles are advantageous for enhancing the specific heat. The amelioration of the interfacial thermal resistance of the particles and the salt molecules formed an effective two-phase interface, and a large amount of heat is absorbed by the nanoparticles and the heat transfer capability of the molecules of the salt is reduced. This causes the nanofluid to absorb or release heat with reduced temperature fluctuations, leading to a considerable increase in the specific heat capacity of the system. In conclusion, the existence of a considerable interfacial thermal resistance (also known as Kapitza resistance) between the incorporated nanoparticles and the ions of the molten salts can improve the specific heat and therefore the thermal energy storage capacity of the molten salt nanofluids. The interfacial thermal resistance is caused by the available high surface area per unit mass that depends on the type, size, and shape of the nanoparticles.

5.6. Compressed Liquid Layer

In the study carried out by the authors Tiznobaik and Shin [80], the specific heat enhancement was 3% in one performed experiment and it was 25% in the other. The difference between experiments derived from the additional hydroxide amount included in the first experiment to impede the formation of a compressed liquid layer, as it was theorized that the surface of the nanoparticles will be slightly charged, provoking the production of a nanostructure at the solid-liquid interface. It was verified that no nanostructure was formed in the first experiment, whilst it was present in the second experiment, indicating that the formation of a compressed liquid layer was the main promoting feature for the specific heat enhancement of the molten salt nanofluids. Moreover, the specific heat enhancement caused by the compressed layer depends on its volume fraction that in turn depends on the morphology of the nanoparticles and on the thickness of the liquid layer. Furthermore, the smaller nanoparticles contribute appreciably more to the compressed layer than the larger ones. Such evidence can be explained based on the increased specific surface area of the nanoparticles with decreasing particle size, enabling the formation of nanostructures. Nonetheless, with the decreasing nanoparticle size for the same concentration, the clustering effect also increases since the distance between the nanoparticles decreases. Additionally, the clustering of the nanoparticles act oppositely to the buildup of the compressed layer, given that the agglomerated nanoparticles possess a lower specific surface area in respect to the uniformly dispersed nanoparticles. On the other hand, Jo and Banerjee carried out molecular dynamic simulations of different mole fractions of Li₂CO₃-K₂CO₃ with the inclusion of graphite nanoparticles [97]. The authors found that with a mole fraction of 74.6 of Li₂CO₃, the compressed layer thickness was 1.05 nm and 1.35 nm with a mole fraction of 34 of Li₂CO₃. In addition, it was verified that with different mole fractions, the chemical composition of the compressed layer was also modified, creating localized concentration gradients close to the nanoparticles. Additionally, the mole fraction of potassium in the compressed layer was greater than the average mole fraction, which was probably caused by the higher adhesion force existing between the potassium fraction and the graphite nanoparticles. Accordingly, the molten salt nanofluid with the higher potassium mole fraction possessed a denser and thicker compressed layer that promoted a more pronounced specific heat increment. Moreover, the influence of the compressed liquid layer can be evaluated by an energy-explaining molecular level. The predominance of the forces of adhesion over the forces of cohesion in the compressed liquid layer induced a smaller packaging of the molecules of the solvent surrounding the nanoparticles, which in turn resulted in a higher mass density and potential energy. Additionally, if the total internal energy is the sum of the kinematic energy from the translational motion of the molecules and potential energy from the relative position of the molecules, the ratio between the potential and kinematic energies for the molecules confined in the compressed layer may suffer a 10-fold increase in reference to the ratio encountered in the molecules in the liquid bulk. Consequently, an inputted total energy greater amount is required for increasing the kinematic energy of the molecules in the compressed phase. Since temperature is an indication of the kinematic energy, this analysis justifies why the molecules in the compressed layer exhibit higher values of specific heat in comparison to the specific heat of the liquid bulk. Expressed in a different way, it can be stated that it is harder for the molecules in the compressed liquid layer to get enough impulse, except for the cases where the energy input is sufficient to beat the potential energy barrier of the compressed liquid layer. These conditions are like those encountered in melting, in which a considerable amount of inputted energy is applied in overcoming the intermolecular bonds in the solid phase, whilst a small amount of the total energy is employed in the enhancement of the kinematic energy of the molecules in the solid phase as it changes to the liquid phase. Thus, if it is assumed that the intermolecular structure of the compressed liquid layer is like that of the molten salt at the solid phase to liquid phase transition, the specific heat inherent to the compressed layer can be predicted to be like the one of the near melting point solid phase. The specific heat of the compressed layer is around 10 times the specific heat of the liquid phase. In the case where the specific heat of the compressed layer adsorbed on the surface of the nanoparticles is to be separately incorporated in the specific heat of the molten salt nanofluids, their specific heat can be determined with the specific heat mass fraction weighted values of the nanoparticles, compressed liquid layer, and liquid bulk by Equation (3):

$$C_{total}=\frac{\left[M_x{C}_n\right]+\left[\frac{m_s}{m_n}{M}_x{C}_s\right]+\left[\left(M-M_x-\frac{m_s}{m_n}{M}_x\right)C_l\right]}{M}$$

(3)

where M represents the total mass of the mixture, M_x is the mass fraction of the nanoparticles, (m_s/m_n) is the mass fraction ratio between the compressed layer and the nanoparticles, C_n is the specific heat of the nanoparticles, C_l is the specific heat of the base fluid, and C_s is the specific heat of the compressed layer. The morphology of the included nanoparticles aids in the determination of this ratio. Considering δ as the thickness of the compressed layer around a spherical nanoparticle having D_np as diameter, then the Equation (3) can be re-written as:

$$C_{total}=\left[x\right]C_n+\left\{\left(x\frac{{\rho }_s}{{\rho }_n}\right)\left[{\left(1+\frac{2\delta }{D_{np}}\right)}^3-1\right]\right\}{C}_s+\left\{1-x-\left(x\frac{{\rho }_s}{{\rho }_n}\right)\left[{\left(1+\frac{2\delta }{D_{np}}\right)}^3-1\right]\right\}{C}_l$$

(4)

As estimated by Equation (4), when the diameter of the nanoparticle is far greater than the compressed liquid layer thickness, the volume fraction of this layer on the surface of the nanoparticles will be relatively low and, hence, its contribution to the specific heat will be little or almost negligible. The specific heat of the eutectic solar salt NaNO₃-KNO₃ was determined to be 1500 J/(kg·K) [97], and the specific heat of the alumina nanoparticles was estimated to be 880 J/(kg·K). Replacing the values in Equation (4), the specific heat of the molten salt-based nanofluid having diverse fractions of nanoparticles can be predicted. The specific heat improvement depends on the size and concentration of the added nanoparticles. Particularly, the specific heat is considerably enhanced in the cases where the concentration of the nanoparticles is high enough, and the average size of the nanoparticles is small enough. Nonetheless, in this study, the predictions only showed a change inferior to 1% wt. in the specific heat and the contribution to the specific heat of the compressed layer was almost negligible when the size of the nanoparticles was superior to 25 nm. Therefore, this prediction was inconsistent with the obtained results, as a considerable enhancement in the specific heat was confirmed for the solar salt nanofluids with 1% wt. of alumina particles having diameters ranging within 10 nm–50 nm. Figure 7 schematically illustrates the different ions that contribute to the formation of the “semi-solid” compressed liquid layer.

In conclusion, the base fluid molecules phenomenon of adhering on the surface of the nanoparticles produces a compressed liquid layer. The thickness of such an adhesion layer depends on the surface energy of the incorporated nanoparticles and these layers possess superior thermal characteristics to those of the bulk liquid and, hence, promote the specific heat enhancement of the molten salt nanofluids. Additionally, for a given concentration of nanoparticles, there should be an ideal size of the nanoparticle to maximize the mass fraction of the adhered molecules forming the compressed liquid and, consequently, maximize the specific heat capacity increase. In addition, the optimum fraction of the incorporated nanoparticles can be a function of the nanoparticle nominal size or the size distribution of the nanoparticles.

5.7. Ionic Exchange Capacity

The authors Mondragón et al. [98] found that when the nanoparticles are dispersed in nitrate molten salts, they interact with the nitrate ions present. The main studied issue was the ionic exchange capacity of the added nanoparticles, tests of which exhibited the ion H⁺ exchange by a functional group. Additionally, the adsorption of the nitrate groups onto the nanoparticles was verified in this work by infrared analysis. For a low ionic exchange capacity, the functionalization level of the silica and alumina nanoparticles was not sufficient to impact the specific heat of the molten salt nanofluids. However, when the ionic exchange capability was beyond a certain threshold value, the increase in the specific heat was more intense because of the contribution of the functionalized nanoparticles. The obtained results confirmed the existence of an ionic exchange between the nanoparticle and the molten salt during the preparation method, leading to the adsorption of the ions of the nitrate onto the surface of the nanoparticles. Hence, the researchers concluded that the level of functionalization of the nanoparticles improved the specific heat of the molten salt nanofluids. In this sense, to confirm that the increment of the specific heat depended on the ionic exchange capacity of the nanoparticles, a molten salt composed of KNO₃ with the incorporation of non-protonated silica nanoparticles was synthesized. With such a system, only the low value of 0.36 mmol/g for the ionic exchange capacity was achieved and, consequently, the increment in the specific heat was solely 1.3%, but on the contrary, the specific heat enhancement obtained for the protonated nanoparticles was 10.1%. Additionally, the impact of the concentration of the included nanoparticles on the ionic exchange capacity was also inferred for KNO₃ with 3.73% wt. of protonated silica nanoparticles. In this case, with the increased concentration, the clustering of the nanoparticles occurred, leading to an augmented particle size and to a lesser specific heat transfer surface. Therefore, the amount of H⁺ ions adsorbed in the course of the protonation process with hydrochloric acid was reduced and the final ionic exchange capacity and specific heat augmentation decreased, as compared with those at 1.92% wt. of nanoparticles. Finally, the authors stated that when evaluating the impact of the amount of the nanoparticles, two opposite results should be considered: the concentration increase and consequent agglomeration effect of the nanoparticles led to a decrement in their degree of functionalization, but a higher number of nanoparticles led to a greater quantity of functionalized nanoparticles that contribute to the specific heat enhancement. Hence, a further commitment between the ionic exchange capacity and the included fraction of nanoparticles should be established. Additionally, the type and composition of the salt, morphology of the nanoparticles, and temperature affect the ionic exchange capacity of the nanoparticles and should be optimized to achieve better specific heat enhancements. In short, the referred study demonstrated that during the synthesis procedure of the molten salt nanofluids, an ionic exchange process occurs between the nanoparticles and the molten salt, leading to a new composition of the surface of the nanoparticles caused by the adsorption of salt ions. The improvement in the specific heat is determined by the degree of functionalization of the nanoparticles, and the latter is closely linked with the concentration of the nanoparticles. A high fraction usually leads to agglomeration that reduces the degree of functionalization of the nanoparticles. However, the same fraction leads to a larger quantity of functionalized nanoparticles, enhancing the specific heat of the molten salt nanofluids.

5.8. Secondary Nanostructures

The specific heat can be improved by the enlarged interfacial area of the so-called fractal-like secondary nanostructures. It was already proposed that the dispersion of nanoparticles in a molten salt promote the preferential surface adsorption of the chemical species in the salt mixture, in this way assisting the nucleation of the molten salt species on the surface of the nanoparticles. The elements of the salt adsorbed onto the surface induces the nucleation and growth of a dendritic-shaped layer of secondary long-range nano-scaled structures. Moreover, in the work performed by the researchers Tiznobaik et al. [99], there were scattered silica, alumina, and magnesia nanoparticles in a molten salt and the observed specific heat enhancements were linked with the production of dendrite-shaped structures in the molten salt nanofluids. The inclusion of nanoparticles in the working fluids promotes the nucleation and growth of a semi-solid phase that consists of the components of the salt. The growth of this phase emanating from the nanoparticles surface induces the generation of secondary structures with dimensions from 100 nm to a few microns. This semi-solid phase can usually exhibit a crystalline structure similar to the structure present on the surface of the nanoparticles. The nanoparticles can induce the preferential surface adsorption of the different chemical species present in the molten salt, which is caused by different molecular affinities. The separation of the components of the salt mixture near the nanoparticles can lead to different properties of the adsorbed layer than the bulk properties of the salt mixture. Hence, the adsorbed layer onto the nanoparticles can exhibit a higher melting point than that of the eutectic salt mixture. This semi-solid layer, once nucleated onto the nanoparticles, can grow, emanating from the surface of the nanoparticles, and form dendritic secondary nanostructures, which are considerably larger than the nanoparticles. These dendritic-shaped nanostructures provide an extended specific surface area, which is the surface area per unit mass of the structures, that augments the contribution of the surface energy to the specific heat capacity. The specific heat for the molten eutectic salt Li₂CO₃-K₂CO₃ with a 62:38 molar ratio was 1.62 kJ/kg K. All the nanofluids showed very close specific heat enhancements independently of the nature of the added nanoparticles. The enhancements in the specific heat were of 27%, 33%, and 22%, as compared with the molten salt itself for the silica, alumina, and magnesia nanofluids, respectively. This implies that the enhancement in the specific heat had low sensitiveness to the type of the nanoparticles. For all the tested nanofluids, the existence of fractal-shaped dendritic long-range secondary nanostructures was found. Since these nanostructures were taken as the main contributors to the increase in the specific heat, the observed enhancement was almost the same for the silica, alumina, and magnesia nanofluids. The structures in the magnesia nanofluids had the largest size and, consequently, the lowest specific surface area and the lowest improvement in the specific heat of 22% among all the nanofluids. Additionally, the reason for the different sizes and area fractions of the dendrites in all the nanofluids can be assumed to be the different affinity between the nanoparticles and the components of the molten salt. The ridge-shaped nanostructures volume fraction was greater than the concentration of the nanoparticles in the molten salt. Such evidence suggested that the nanoparticles induced a long-range effect in the molten salt, promoting the generation of the secondary nanostructures. Moreover, the preferential adsorption of ions onto the surface of the nanoparticles modifies their charge distribution, leading to an accumulation of surface charges. Considering the alumina nanoparticles, the preferential adsorption of the NO⁻ anion onto their surface causes a slow growth of a negatively charged surface potential with respect to the bulk phase. This phenomenon will induce strong electrostatic attractive/repulsive forces around the nanoparticles and between the nanoparticles. Under the influence of the electrostatic forces, the ionic liquid layer may extend to up to 1 µm, where the adsorbed molecules in the compressed layer align and mimic the epitaxial structure of the surface of the nanoparticles. Furthermore, it was inferred that the interactions in the system produce ordered layers, which will extend from the surface of a single nanoparticle and form secondary nanostructures interconnecting the nearby nanoparticles, resulting in foam-like percolation networks. The estimation of the specific heat of the molten salt nanofluids, including the influence of the long-range nanostructures, can be expressed by Equation (5):

$$C_{total=}\frac{\left[M_x{C}_n\right]+\left[\frac{m_s}{m_n}{M}_x{C}_s\right]+\left[\frac{m_f}{m_n}{M}_x{C}_f\right]+\left[\left(M-M_x-\frac{m_s}{m_n}{M}_x-\frac{m_f}{m_n}{M}_x\right)C_l\right]}{M}$$

(5)

where m_f/m_n represents the mass ratio between the nanostructures and the nanoparticles, whereas C_f is the specific heat of the nanostructures. Hence, any attempt to determine the specific heat of the molten salt nanofluids should consider the properties of the material, as well as with the structure of the molecules of the solvent in the secondary nanostructures. Furthermore, the researchers Sang and Liu [100] investigated the influence of diverse nanoparticles on the 40% wt. K₂CO₃-40% wt. Li₂CO₃-20% wt. Na₂CO₃ carbonate salt and observed maximum enhancements of 116.8%, 73.9%, 56.5%, and 66.5% in the specific heat by adding silica, copper oxide, titania, and alumina nanoparticles. These enhancements were explained based on the dispersion degree of the nanoparticles and the number of formed needle-like secondary nanostructures. To conclude, the chemisorption of free ions onto the surface of the incorporated nanoparticles and the inherent long-range ionic effect can provoke the generation of secondary nanostructures in the solvent phase of the molten salt nanofluids. Such nanostructures may determine the specific heat improvement of the molten salt nanofluids. Possessing a significant overall volume fraction, these secondary network nanostructures can be considered as an extension of the compressed liquid layer formed near the surface of the nanoparticles. Hence, apart form the contributions for the specific heat enhancement from the bulk fluid, nanoparticles, and compressed layer, the secondary nanostructures should be taken seriously into account when considering the specific heat improvement of the molten salt nanofluids.

5.9. Cloud Nuclei

In the thermal fluids composed by molten salts with the addition of nanoparticles, the ions of the salt approach the surface of the nanoparticles driven by the interaction of the Van der Waals forces between the nanoparticles and the ions of the salt and the electrostatic force between the hydrogen bonds and the ions of the salt. During this process, the salt ions were adsorbed onto the surface of the nanoparticles in a layer-by-layer fashion, forming a cloud nucleus. In time, immense cloud nuclei centered at the nanoparticles and surrounded by the ions of the salt were produced until the interaction between the ions of the salt outermost localized and the nearby nanoparticles reached the equilibrium [101]. In the preparation method, a too short duration of the agitation process will maximize the amount of formed cloud nuclei and some nanoparticles may not generate saturated cloud nuclei, whereas an excessive agitation duration would induce a destroying action on cloud nuclei, provoking a notorious agglomeration of the cloud nuclei. Furthermore, in the cases where the molten salt nanofluids achieve their melting point, the ions of the salt in the vicinity of the cloud nuclei will gradually fuse. As the temperature of nanofluid further increases, the ions of the salt inside the cloud nuclei progressively melt from the edge to the center of the cloud nuclei. With the temperature of the nanofluids rising above their melting temperature, the ions adsorbed by the Van der Waals forces within the cloud nuclei completely melt. At this stage, the water continues to be adsorbed by the hydrogen bonds between the hydroxyl groups and the water molecules and the Van der Waals force between the nanoparticles and the molecules of water. As a result, the water molecules leaving the nanoparticle surface is still a difficult process. Nevertheless, when the temperature of the nanofluid goes up to a higher level, the hydrogen bonds will start to breakup, and since the hydrogen bonds inside the cloud nuclei are broken, the adsorbed water molecules will be detached from the cloud nuclei and evaporate, whilst the hydroxyl groups will form again on the surface of the nanoparticles. Nonetheless, since the temperature required for the hydroxyl group detachment of the nanoparticles is much greater than the service temperature of the nanofluid, the hydroxyl groups will continue to be adsorbed onto the surface of the nanoparticles. Additionally, with the further increase in the nanofluid temperature, the distance between the nanoparticles and the ions of the salt within the cloud nuclei is extended. This larger distance will make the molten salt ions depart from the cloud nuclei and the diameter of the cloud nuclei will diminish. On the other hand, during this process, more heat is required, explaining the specific heat improvement when using molten salt nanofluids. Due to the cloud core microstructure generation, the specific heat of the carbonate nanofluids synthesized by the aqueous solution dissolution method increased rapidly to 122% [102] with respect to that of the molten carbonate salt alone. The authors Yaxuan et al. [103] synthesized carbonate molten salt nanofluids with the inclusion of silica and magnesia nanoparticles and attempted to interpret their specific heat enhancement based on a molecular perspective. It was reported that the specific heat of the nanofluids was enhanced by 20.7% and 34.1% with the inclusion of silica and magnesia nanoparticles, respectively. Additionally, the authors stated that the formation of different nanostructures was observed in the silica and magnesia nanofluids due to the varying size and structure of the cloud nuclei, which promoted various thermal energy storage capability improvements. Moreover, the motion of the molecules of the salt departing or moving toward the cloud nuclei improved the specific heat, while the chains of the cloud nuclei enhanced the thermal conductivity of the nanofluids. Figure 8 schematically presents the cloud nuclei process formation in the added silica nanoparticles.

In the study conducted by Xiong et al. [104], there were prepared potassium nitrate, sodium nitrate, and respective binary mixtures nanofluids with the incorporation of 20 nm silica nanoparticles. The authors found that for the potassium nitrate nanofluid, its specific heat was increased by 34.1% at 1.0% wt. of silica nanoparticles. At a similar fraction of nanoparticles, the specific heat of the sodium nitrate was enhanced by 11.9%. For the binary nitrate salt, the specific heat was enhanced by up to 15.9% at 0.5% wt. of silica nanoparticles. Additionally, at 0.7% wt., the specific heat of the molten salt nanofluids possessed a similar value to that of the base salts themselves at a temperature value lower than 275 °C. Furthermore, the authors stated that the higher specific surface energy of the silica particles adsorbed the surrounding molecules of the salt in a layer-by-layer wise and formed cloud nuclei, improving in this way the specific heat of the nanofluids. Such cloud nuclei were centered on the nanoparticles and were surrounded by the molten salt molecules. Considering the same salt, the dimensions and number of the cloud nuclei define the specific heat of the molten salt nanofluids. The incorporation of silica nanoparticles provokes the formation of the various crystalline microstructures of the molten salts. Such modifications can be caused by the interactions between the silica particles and the different base salts. The base salts have different physicochemical characteristics, whereas the alkali cations present different electronegativities. These characteristics lead to different interactions between the silica nanoparticles and different base salt molecules. As aforementioned, coral-like and sugar-like microstructures and micro-scaled crystals were formed in different molten salt nanofluids. The cloud nuclei dimension depends on the size of the nanoparticles, nature of the salt molecules and degree of mixing. Furthermore, the neighboring cloud nuclei can interact with each other, forming extended nuclei that will be crystalized by the molten salt-based nanofluids. Therefore, these microcrystals are formed mainly by physical adsorption. Expressed in a different way, the layer of salt molecules around the surface of the nanoparticles is produced by chemical adsorption. From the second layer on, the molecules of the salt are layer-by-layer physically adsorbed onto the nanoparticles. Finally, the potential generation of diverse structures in the molten salt nanofluids depends on the nature of the base salts. With further heating, a rising amount of salt molecules continuously depart from the cloud nuclei until the melting is completed. In sum, it was demonstrated that the added nanoparticles (e.g., silica) absorb the surrounding salt molecules in a layer-by-layer fashion and form cloud nuclei. These cloud nuclei enhance the overall specific heat capacity of the molten salt nanofluids. It was found that the extension and number of the cloud nuclei define the specific heat improvement of the molten salt based nanofluids in a solid/liquid state.

6. Agglomeration and Sedimentation Over Time

Some researchers have performed stability over time inferring experiments with molten salt nanofluids in a liquid state [105] and as phase change materials [12]. The nanoparticles dispersed within the molten salt may collide into each other because of the Brownian motion, and the impacts usually lead to the formation of randomly located agglomerates having different morphologies and sizes. When a critical size of the aggregates is reached, they will settle down and only a minor part of the nanoparticles will remain suspended in the molten salt. The strong heterogeneity caused by the clustering and consequent settlement of the nanoparticles will reduce the heat transfer capability of the nanofluids. Hence, great efforts should be made to overcome this limitation and maintain the improved thermophysical properties of the nanofluids in the long term, averting their settlement. In the cases where the nanofluids are taken as phase change materials, the nucleation and crystallization stages should be added to the required stability over time in the liquid phase to evaluate the stability of the thermal fluids over time. These phenomena may change the settling of the nanoparticles, given that, in this case, the Brownian motion is constrained by the nucleation and growth of the salt crystals. Additionally, it should be noted that the stability in the domain of time evaluation is commonly investigated by the initial and terminal specific heat comparison. The salt is maintained in the liquid state for a well-defined period [21] or it is thermally cycled between two points, suffering a phase change in the temperature interval [12] or without a phase change [15]. Moreover, Mondragón et al. [86] maintained a nanofluid composed of solar salt and 0.5% wt. of silica nanoparticles at 500 °C for 12 h. The authors found a decrement in the specific heat on the upper part of the nanofluid after the referred treatment, whereas the bottom part of the nanofluid remained with unchanged specific heat. The observed decrement was explained based on the settling of the added nanoparticles. The authors observed that the nanofluid was not stable when the temperature went beyond 400 °C, exhibiting a slow decrease in the specific heat after each thermal cycle. On the contrary, the same authors found that the inclusion of alumina nanoparticles into solar salt was stable, and the specific heat remained constant after all the conducted experiments. Moreover, Lasfargues [105] proposed three different preparation routes to enhance the stability over time of the solar salt with the addition of 0.1% wt. of copper oxide nanoparticles at 300 °C in liquid state: bubbling, forced circulation, and vigorous mechanical stirring. After 840 h under static conditions, the accumulation of nanoparticles was observed at the bottom of the container. The specific heat of the nanofluid suffered an appreciably decrement as compared to the initial value. The mechanical stirring was found to provide a good and uniform dispersion of the nanoparticles and the specific heat remained at the initial value and increased with the experiment increasing time. The bubbling and forced convection were found to be less effective on maintaining the initial specific heat, melting point, and latent heat of fusion of the nanofluids. In the work performed by Munoz-Sánchez et al. [106] the solar salt 60% wt. NaNO₃ + 40% wt. KNO₃ with the inclusion of 1% wt. of alumina nanoparticles was synthesized. It was inferred that the tendency of the nanoparticles was to agglomerate and settle through their size and concentration over time. The authors stated that the higher the specific surface of the added nanoparticles, the higher their interaction with the nearby salt ions and, consequently, the lower their tendency to agglomerate and sediment. The evolution in the domain of time of the particle size was irregular since it increased in the first hour and decreased after 5 h. This could be proof of the process of agglomeration and the consequent sedimentation that takes place in the nanofluid, where the nanoparticles clustered and then fell to the bottom of the container. Moreover, it should be noted that the count rate in kilo counts per second is a factor closely linked with the measurement time. The lower the count rate, the longer the duration will be, and this parameter was affected by the amount of alumina nanoparticles in the nanofluid. The lower the concentration of alumina nanoparticles, the lower the count rate was, and a count rate of 38.7 after 5 h revealed a very low concentration of alumina nanoparticles. Hence, it can be stated that the agglomeration and sedimentation of the nanoparticles was faster than expected. The measured sizes of the alumina nanoparticles by dynamic light scattering were due to only a small fraction of nanoparticles that remained suspended within the molten salt after 30 min. Additionally, no increment on the specific heat was found on the solar salt nanofluid with reference to the base salt after 30 min of the experiment. Because of the minimal content of alumina nanoparticles remaining in the nanofluid, there was practically no effect of the inclusion of the nanoparticles in the specific heat of the nanofluids. Finally, it was also stated that the base salts always contain impurities that also precipitate with time, and these impurities may affect the stability of the nanoparticles. Hence, the white layer that was observed in the molten salt nanofluids was most likely a mixture of agglomerated nanoparticles and insoluble impurities from the base salt. Furthermore, the researchers Navarrete et al. [107] developed a new experimental setup composed of a dynamic light scattering system and a cuvette to determine the distribution of the size of the suspended nanoparticles at 500 °C. The authors prepared a Solar Salt with silica nanoparticles and aluminum/copper nanoencapsulated phase change material and the colloidal stability of the system was measured. The impact of the base fluid chemical composition and nanoparticles was checked, exhibiting a pronounced instability of the nanoparticles in the molten salt nanofluids. Nevertheless, the aluminum/copper nanoencapsulated phase change material displayed the highest stability, given that the initial nanoparticle size distribution was recovered immediately after a mechanically driven redispersion after 4 hours in a static condition.

7. Applications in Concentrated Solar Power Plants

7.1. Concentrated Solar Power Plants

The changing of the production of energy from fossil fuels to renewable clean energy sources involves dealing with the intermittency of some of these sources such as the solar one, which can result in uneven power generation availability. Hence, to mitigate these limitations, it is necessary to develop a suitable energy storage capability to keep operating the power production stations. Moreover, in the solar thermal energy plants named as concentrated solar power plants, the most employed mechanism of energy storage consists of using two tanks—one hot and one cold—filled with molten salts. As a consequence, the cold salt is heated by solar energy and is transferred to the hot tank. In the cases where the sunlight cannot be collected because of the weather or nighttime, the salts are transferred from the hot tank to the cold tank by means of a heat exchanger, which enhances the temperature of the heat transfer fluid responsible for keeping the plant working [108]. Figure 9 presents a schematic diagram of a typical concentrated solar power plant.

A concentrated solar power plant converts solar energy into electrical energy using concentrators such as mirrors and lenses, which target the collected solar energy to a receiver. The concentrated solar power plant configurations can be classified as parabolic trough, Fresnel collector, tower, and parabolic dish systems. The thermophysical characteristics, including the specific heat, thermal conductivity, thermal stability, and viscosity of the heat storage medium, are critical and directly affect the efficiency of the concentrated solar power plants. The solar field heats the heat transfer fluid and, with the aid of heat exchangers, the heated heat transfer fluid charges the thermal storage unit. Additionally, there are some possible concentrated solar power plant configurations according to the solar energy harvesting mode. In the line-focusing concentrated solar power plants with a parabolic trough or linear Fresnel reflectors, the temperature of harvesting is usually between 0 °C and 400 °C, while the point-focusing plants having a parabolic dish or tower collectors work at the temperature range of 290–560 °C [108]. These differences lead to the usage of different combinations of heat transfer fluids and thermal energy storage materials, which depend on the configuration of the plant. In line-focusing scenarios, the thermal oils play the role of heat transfer fluids and the thermal energy storage is achieved by including molten salt tanks that operate as an indirect storage system, given that the heat in excess is primarily transferred from the heat transfer fluid to the thermal energy storage material, and then back to the heat transfer fluid whenever needed by employing heat exchangers [109]. The replacement of the conventional thermal oils by molten salt nanofluids is a beneficial procedure if it is considered that the operating temperature of the solar power energy station can be increased by enhancing the efficiency of the power section [110].

7.2. Molten Salt Nanofluids in Concentrated Solar Power Plants

The employment of molten salts having a wide range of service temperatures is an option with great potential to be employed as thermal storage materials in concentrated solar power plants. One of the most used in the concentrated solar power field of technology is the solar salt 42 mol% KNO₃ and 58 mol% NaNO₃, particularly following a two-tank sensible energy storage design. However, the commonly explored nitrate molten salts decompose at temperature values close to 600 °C [111] and, consequently, the upper limit service temperature of the actual thermal energy storage systems is limited to 560 °C because of the maximum service temperature of the molten salt. At the same time, the usage of molten salt nanofluids would reduce the total investment cost of the systems by reducing the quantity of the required thermal energy storage materials and the dimensions of the storage tanks and reducing the levelized cost of electricity produced by the concentrated solar power facilities. Moreover, the molten carbonate salts have received great attention due to their improved thermophysical properties. These salts are also very useful for sensible and latent heat storage applications. Additionally, considering the thermophysical and corrosivity characteristics, carbonate salts are advantageous when compared to chloride, fluoride, and nitrate salts, which were also explored for the same ends. The carbonate salts were employed in the high-temperature thermal energy storage at concentrated solar power plants, where the molten salts were to be used at temperature values superior to 600 °C [112]. The molten carbonate salts have been investigated for future thermal energy storage purposes because of their advantageous features, such as the possibility to operate under high temperature values, enhanced specific heat, improved thermal stability, reduced vapor pressure, and decreased induced corrosion [113]. Meanwhile, it is worth mentioning that the following mechanisms of the thermal energy storage should be considered [107]:

(i)

Sensible heat storage that regards the heat that is stored through the addition of kinetic energy to a material, enhancing its temperature without any phase change. The energy is absorbed with the increase of the temperature of the material heat charge—and is released when the material cools down—heat discharge. The sensible heat can be determined by Equation (6):

Q_sensible = m⋅c_p⋅ΔT

(6)

where m is the mass of the material, c_p is the material specific heat, and ΔT it the temperature gradient.

(ii)

Latent heat storage that concerns the thermal energy storage through the phase change without any temperature variation. The operation of the latent heat materials is based on storing the thermal energy during the melting process (heat charge) and on releasing the thermal energy in the solidification process (heat discharge). The quantity of energy that is exchanged can be calculated by Equation (7):

Q_latent = m⋅ΔH

(7)

where ΔH is the phase change enthalpy.

(iii)

Thermochemical energy storage that deals with the energy harvested by the breaking-up of the chemical bonds during the reversible chemical reactions and sorption systems. It has the highest storage capacity but possesses some limitations that hinders its large-scale applicability, such as the chemical stability, durability, and the need for separate storage systems.

The sensible thermal energy storage is the most found in the concentrated solar power thermal energy storage. Furthermore, apart from the operation temperatures of the concentrated solar power plants, many other factors are to be considered in the selection process of the thermal energy storage material, including the volumetric storage capacity, thermal conductivity, thermal expansion and stability, chemical stability, environmental benevolence, availability, and corrosivity degree, together with the indispensable economic analysis. The most important factor is the increased thermal energy storage capacity aiming to reduce the facility volumetric dimensions. The molten salt aqueous mixtures can operate under a broad temperature range depending on the composition of the salts, with some nitrate-based salts, such as the already referred to solar salt, which is commonly employed in the solar power technologies for thermal energy storage [114] and has a melting point of 220 °C. Additionally, the Hitec XL salt Ca(NO₃)₂-KNO₃-NaNO₃ with a 33.8-51.5-14.7 molar fraction that can operate at only 140 °C, and the mixture of KNO₃-LiNO₃-Ca(NO₃)₂ having a melting temperature of 80 °C [115], are alternatives with great potential. Moreover, the carbonate molten salts have been identified as strong candidates to be applied in temperature sensible heat energy storage applications at high-temperature values ranging from 400 °C to 850 °C [110,116]. On the other hand, with the latent heat storage, they are commonly employed materials with high phase change enthalpies designated by phase change materials. For phase change materials with a solid-liquid transition, the most suitable material can be selected according to its melting temperature and phase change enthalpy that will accomplish the well-defined operating variables of the intended application. For instance, the carbonate-based phase change material prepared by using 43% wt. of Li₂CO₃ and 57% wt. of Na₂CO₃ exhibits an enthalpy of 348.5 J/g [90]. Similarly, the ternary carbonate Li₂CO₃-Na₂CO₃-K₂CO₃ with a 32.2-33.3-34.5 mass fraction was also investigated to be applied as a phase change material possessing a phase change enthalpy of 166 J/g [116]. Although some concentrated solar power plants already use molten salts as heat transfer fluids, the most common approach still recommends the use of thermal oils for the heat transfer and molten salts for the thermal energy storage based on their sensible heat when melted. However, the high flammability of the mineral oils can be the main problem associated with their safe use, whereas the synthetic oils are not cost-effective [117]. Oppositely, molten salts present improved thermal stability at high temperature and thermal conductivity, are non-flammable, and entail low environmental impacts, as compared to those associated with the use of the traditional thermal oils. Apart from these advantageous facts, reaching higher temperatures in concentrated solar power stations usually leads to an enhanced efficiency. Nonetheless, some other factors should be considered, such as the freezing point of the molten salts, since an extra heating auxiliar might be needed to avert the solidification of the salts in the pipping that may deteriorate the facilities, or corrosive interactions with the materials of the pipping. Additionally, many combinations of anhydrous inorganic salt mixtures have been proposed in the literature for their use in thermal energy storage and heat transfer applications. Among the different possibilities of molten salts used in thermal energy storage, the carbonates possess a higher service temperature range than the nitrates, enabling a greater thermal energy-to-electricity conversion. The carbonate molten salts are also more cost-effective than the fluoride molten salts and present less corrosive strength concerns than the chloride molten salts and are one of the most suitable options to further enhance the efficiency of the solar thermal plants and advanced energy conversion processes. Concerning the corrosion resistance issue, different corrosion mitigation approaches were experimented, such as the purification of the salts and the inclusion of corrosion inhibiting nanoparticles [117,118].

Table 6. Main Corrosion Strategies to be Applied in CSP plants.

Procedure	Techniques/Examples	Benefits	Limitations	Authors/Reference
Graphitization of Steel	Spray Graphitization of Steel	Corrosion Rate Reduction for Nitrate Molten Salts	Reduced Chloride Content in the Molten Salt Mixtures Humid Conditions May Still Provoke Severe Consequences The Addition of Nanoparticles May Entail Adverse Impacts on the Effectiveness of the Graphitization	Grosu et al. [119]
Addition of Graphite Nanoparticles	Addition of Graphite Nanoparticles in the Molten Salts	Mitigation of the Corrosive Behavior of the Nitrate Molten Salts	Eventual Formation of Microbubbles That Increase the Corrosion Layer Thickness Owing to the Higher Amount of Oxygen in the System Severe Corrosion Rates in Stainless-Steels Against Fluoride Molten Salts	Gonzalez et al. [120]
Addition of Magnesium Nanoparticles	Addition of Magnesium Nanoparticles into the Oxidation Layer of Chloride Molten Salts	Corrosion Inhibition by Reducing the Redox Potential of the Molten Salts Reduction of the Corrosion Rate by More Than 90%	Eventual Formation of Microbubbles That Increase the Corrosion Layer Thickness Due to the Higher Oxygen Amount in the System	Ding et al. [121]
Salt Purification	Physical, Chemical, and Electrochemical Purification of the Molten Salts	Strongly Minimizes the Corrosive Impurities (e.g., MgOH+) in the Molten Salt Cost-Effective Operation in CSP Plants	Pre-Implementation Process Designed Primarily for Laboratory Environment Lack of In-Situ Purification Methods High Flammability and Toxicity of Purge Gases	Ding et al. [122]
Use of Stainless-Steel	SS AISI 316 L, SS AISI 430, SS AISI 347, SS AISI 321H, Among Others	Reduced Corrosion Rates of Circuits and Containers Especially Against Nitrate Molten Salts	Relative High Cost in CSP Environment Some Stainless-Steels (e.g., SS AISI 316L) Exhibit Excessive Corrosion Rates for Industrial Applications	Gomes et al. [123]
Use of Special Alloys	Hastelloy C-276, Inconel 718, and Inconel 625	Reduced Corrosion Rates of Pipping and Vessels The Hastelloy Alloy Achieved the Corrosion Rate up To Industrial Implementations	High Cost in CSP Environment Short-Term Corrosion Protection Approach for Certain Alloys The Fluoride Molten Salts Together with an Increased Moisture level Provokes Intergranular Corrosion and Pitting	Liu et al. [124]
Use of Alumina Forming Austenitic Alloys	OC4 and HR224 Alloys	Reduced Corrosion Rates of Pipping and Vessels	Need to Carefully Control the Thickness, Uniformity, and Stability of the Layers Eventual Cracks in the Layers May Initiate Further Localized Corrosion	Fernandez et al. [125]
Use of Pre-Oxidized Alloys	Pre-oxidizing the Fe-Cr-Al Alloys Conducts to the Formation of an Alumina Scale	Improved Corrosion Resistance of Commercial Alloys Against Chloride Molten Salts	Present Better Corrosion Resistances in CO2 Atmosphere	Frangini et al. [126]
Metal and Metal Oxide Coatings	Nickel, Cobalt, Aluminum Nitride, and Alumina	Drastic Reduction of the Corrosive Rates of Fluoride and Chloride Molten Salts Remarkable Thermal Stability at High Temperatures	Complexity of the Coating Procedures Long-Term Material Compatibility	Zhu et al. [127]

summarizes the main followed corrosion mitigation routes.

Table 7. Main Molten Salts and Their Characteristics to be Applied in CSP plants.

Molten Salts	Nitrate	Chloride	Fluoride	Carbonate
Specific Heat Capacity	High Specific Heat Capacity	Relatively High but Lower Than That of the Other Molten Salts	Very High Specific Heat Capacity	High Specific Heat Capacity
Thermal Stability	Up to 500–600 °C	High Thermal Stability at T > 800 °C	High Thermal Stability at T > 700 °C	High Thermal Stability at 650–850 °C
Melting point	Low Melting Point LiNO3 and Ca(NO3)2 Reduce the Melting Point Near 100 °C	Moderate Melting Point at around 400 °C	Relatively High Melting Point	Moderate Melting Point at around 400 °C
Effect of Impurities	Affect the Thermal Stability Range	Aggravate the High Corrosivity of the Molten Salts	Aggravate the Corrosion Behavior	Do Not Require Salt Purification Procedures
Counter Corrosion Strategies	Graphitization, Use of Stainless-Steels, and Alumina Forming Austenitic Alloys	Anaerobic Atmosphere, Salt Electrochemical Purification, and Addition of Mg Inhibitor	Metal and Metal Oxide Coatings	CO2 Inert Atmosphere, Use of Alumina Forming Alloys, and Pre-Oxidized Alloys
Economic Feasibility	Should be Carefully Considered	Relative Low Cost	High Cost	High Cost due to Li2CO3 Addition

summarizes the importance, benefits, and limitations of the molten salt with potential to be used in CSP plants.

7.3. Modeling and Practical Case Studies

The researchers Karim et al. [128] investigated the application of the molten salt nanofluids as heat transfer fluids in direct absorption solar collector systems. With this purpose, a two-dimensional computer fluid dynamics model of a direct absorption high-temperature molten salt nanofluid concentrating solar receiver was proposed by the authors to evaluate the impact of the design and operating variables on the performance of the receiver, such as length and height, inlet velocity, and solar concentration. Additionally, it has already been reported that the Carnot efficiency is augmented with increasing length and height of the receiver, increasing solar concentration, and decreasing inlet velocity. In the cases where the receiver is connected to a power generation cycle, the overall efficiency of the system is commonly found to exceed 40% in the cases where the solar concentrations are greater than 100X. In the referred study, an adjusted Carnot efficiency has been employed in conjunction with the upper temperature limit of the nanofluid to give more emphasis to the receiver temperature increment. The adjusted total efficiency gave a peak in the efficiency for the solar concentration that was reduced with a decrease in the concentration, making sure that every possible configuration of the receiver has an optimal solar concentration. The used molten salt nanofluid was the binary nitrate NaNO₃-kNO₃ solar salt with the inclusion of graphite nanoparticles. The model considered the surface to ambient radiation loss, convective loss, and re-emission loss from the nanofluid. The model demonstrated that the efficiency of the receiver was augmented with increasing solar concentration, decreasing length of the receiver, decreasing concentration of the nanoparticles, and increasing height of the receiver. It also revealed that the temperature augment throughout the receiver enhances with the increasing length of the receiver, decreasing inlet velocity, and increasing solar concentration. A higher concentration receiver will result in a higher efficiency receiver, given that the increment in the efficiency of the receiver compensates for the decrement in the Carnot efficiency, since the efficiency of the receiver exceeded 90% in certain cases. The results also revealed that there is an ideal length of the receiver, the increase of which depends on the increasing incident solar concentration, and on the inlet velocity. Furthermore, the adjusted total efficiency resulted in a peak efficiency for solar concentration that diminished with decreasing concentration, making sure that each potential configuration of the receiver possesses an ideal solar radiation concentration. On the other hand, Karim et al. [129] proposed a model to better understand the influence of the involved parameters in the performance of a direct absorption concentrating solar receiver using a molten salt nanofluid. This time, the researchers developed a three-dimensional computational fluid dynamics model using the binary carbonate Li₂CO₃-K₂CO₃ molten salt nanofluid containing dispersed graphite nanoparticles. The spectral properties of the molten salt and nanoparticles were modeled regardless of the selected wavelength, and the absorption of the incident solar radiation was taken as a volumetric heat released in the heat transfer fluid. Under high working temperatures, the efficiency of the receiver decreased from near 0.8 to 0.6 and the total efficiency also decreased from around 0.5 to 0.4, whereas the Carnot efficiency increased slightly from around 0.6 to near 0.7 with increasing receiver length. An adjusted Carnot efficiency, i.e., considering the inlet temperature as the low temperature instead of the ambient temperature, revealed that the increase in the temperature of the heat transfer fluid had an exponential relationship with the length of the receiver until a maximum length value, beyond which the temperature increase was insufficient to compensate for the reduction in the efficiency of the receiver. Moreover, using the normal Carnot efficiency, it was found that a rise in the heat transfer fluid velocity resulted in practically no impact on the total efficiency of the receiver of η = 0.66, η = 0.6, and η = 0.4 at 1071 K, and resulted in a not significant improvement in the efficiency of the receiver, with a length of 1 m as the outlet temperature dropped down. Nevertheless, the adjusted total efficiency of the receiver decreased with increasing inlet velocity. Hence, the receiver with the highest efficiency happened to be the one with low thermal fluid velocity, as this resulted in an enhanced operating temperature until the upper temperature limit of the molten salt nanofluid was reached. Furthermore, volume fractions of the added nanoparticles ranging from 1 × 10⁻⁶ to 5 × 10⁻⁴ were analyzed with receivers of different lengths with the initial results showing that the receiver with the highest efficiency of η_tot ≈ 0.5 was the one having the lowest volume fraction at the shortest length. This was not an entirely accurate perspective, especially when considering the adjusted overall efficiency. It was verified that the efficiency of the receiver increased with the increasing volume fraction and length of the receiver, the latter to a certain extent, since the existence of an ideal length was clear. It was revealed that the higher the concentration of the nanoparticles, the shorter the optimal length of the receiver. A higher concentration resulted in an increased temperature at the outlet and an enhanced efficiency, but also entailed a higher sensitivity to the loss of heat to the environment. Additionally, the initial evaluation of the impact of the concentration of the solar radiation demonstrated that the Carnot efficiency from 0.6 to 0.7, overall efficiency from −0.6 to 0.2, and the adjusted Carnot efficiency and adjusted total efficiency were slowly augmented with increasing solar concentrations from 10× to 100×. Nevertheless, one tendency was verified in the cases where the difference between the receiver average and peak temperatures increased from 5 K to 41 K by augmenting the solar concentration from 30× to 100×. This evidence makes sure that if the temperature peak was kept constant by balancing both length of the receiver and inlet velocity, the average temperature would decrease with increasing solar concentration, resulting in decreases in both the Carnot efficiency and the efficiency of the receiver, contradicting the initially predicted tendencies. Maintaining the peak temperature constant, an ideal concentration of the solar radiation was indeed found in the cases where the adjusted Carnot efficiency was considered. Hence, for a receiver with 1 m of length and a concentration of 1 × 10⁻⁴, the ideal solar concentration was near 150×. Finally, a comparative analysis demonstrated that the solar salt presented a better performance than that of the carbonate salt caused by the efficiency enhancement of the collector. However, the LiCO₃-K₂CO₃ molten salt as a base fluid has a higher potential to be applied in thermal energy storage purposes, since a larger amount of energy can be stored with a smaller volume of fluid at high-temperature conditions. Furthermore, a large-scale case study worth mentioning for the practical applications of the molten salt nanofluids is the one performed by Liaqat et al. [130] in which the effect of using molten salt nanofluids for performance improvement of a concentrated solar power plant was studied. The study was conducted in the context of Pakistan, which is a country which has been dealing with energy crises for a long time, and the site was the Nara desert, more precisely in Nawabshah in the province of Sindh, which receives an annual direct normal irradiance of 1955 kWh/m². Additionally, a 100 MW Parabolic Trough Collector plant was modeled using the system advisor model that was validated by comparing the obtained results with those of existing commercial-scale plants. The plant was integrated with a thermal energy storage of up to 10 h. In addition, the nanoparticles with the highest share of the employment of parabolic trough collectors were selected, which are the copper oxide, alumina, and titania, having different concentrations, and the Hitec solar salt as a base fluid. A parametric optimization was conducted for each of the salt alternatives and the performance was compared based on the annual power generation, capacity factor, plant efficiency, and required thermal storage volume. The comparison of results showed that by using the molten salts nanofluids, the annual power generation, capacity factor, and efficiency were enhanced by 8.86%, 8.88%, and 8.9%, respectively. In terms of needed thermal storage volume, this operating factor was reduced by 36%. Additionally, the authors found that the ideal concentration of the nanoparticles was of 1% vol., 2% vol., and 5% vol. for the copper oxide, alumina, and titania nanoparticles, respectively.

7.4. Life Cycle Assessment

The Life Cycle Assessment [131] is a method to address the environmental impacts associated with the life-cycle stages of a product throughout the manufacturing, distribution, use, and recycling or waste disposal of the involved materials. Normally, the Life Cycle Analysis is performed according with the following steps:

(i)

definition of the involved system.

(ii)

collection of data of interest.

(iii)

risk characterization and impacts quantification.

(iv)

interpretation of the results.

The system addressed in the work performed by the authors Grosu et al. [132] was comprised of the manufacturing of the materials, preparation of nanoparticles, heating storage usage, working lifespan, dismantling of the elements, and management of the waste disposal. The materials and nanoparticles were considered in the life cycle analysis according to qualitative environmental and risks for human health evaluation. In the assessment of the nanofluids, the life cycle analysis was performed according to the ISO Standard 14,040 using the SimaPro and EcoInvent databases. In this work, a life cycle assessment analysis was performed for the (Li, Na, K)₂CO₃ and alumina nanofluid and compared with the molten salt itself, and the synthesis of the salts and alumina nanoparticles through the explosive wire method, addition of the nanoparticles to the molten salt, use in solar energy production systems, and waste disposal were considered. A 30-year lifespan was considered for the concentrated solar power plant, and to characterize the processes, the information of some existing concentrated solar power plants was gathered (GEMASOLAR [132], Andasol-1 [133], and Solana [109]). The characteristics of the salt and alumina nanofluids were based on the results of this work and serve as the input to the system. The addition of the nanoparticles resulted in the maximum enhancement of 12% in the heat capacity and, consequently, in the reduction of the mass of the salt by 12%, and of the volume of the storage vessels by 12%. An extra beneficial feature was the 2-fold decrease in the corrosion rates of the nanofluid as compared to the base molten salt alone. The decreased corrosion rates also appreciably reduced the quantities of the required construction materials. The quantification of the risk characterization and environmental impact were carried out by the IMPACT 2002+ methodology. The impacts arising from the molten salt utilization were mainly the non-renewable energy, global warming, ecotoxicity, respiratory risks, and non-carcinogens/carcinogens. These are associated mainly with the energy consumption issues during the manufacturing and operation phases. The risk characterization was carried out for the alumina nanoparticles [134], that this aspect must be considered, and adequate risk management should be undertaken in the phases of production, usage, and disposal. It should be stated that the assessment is affected by parameters such as the influence of the dynamic viscosity on the energy requirement and rate of erosion. Other relevant concerns are the logistic factors, synthesis, and handling of the molten salts and alumina nanoparticles. In addition, the management of the molten salts after their service life should be seriously considered. It can also be concluded that the fundamental environmental impact of the molten salt nanofluids derives from the base molten salt manufacturing. The results suggested that the preparation of the nanofluids and respective pumping operating system were the most prominent issues impacting global warming. Moreover, it can be stated that while the chloride salts provided lower environmental impacts, their thermophysical characteristics and corrosive potential makes them less appropriated when compared to the nanofluid employed in this analysis. The dynamic viscosity provoked a 10% rise in the overall impact, counterbalancing the benefits from the specific heat and corrosion rate. Consequently, the viscosity should be taken as one of the fundamental compromise characteristics for the enhanced specific heat. For the studied system, the viscosity increment should not be superior to 20% to be compensated by the ameliorated specific heat and corrosion resistance factors. Considering the uncertainties associated to the thermophysical properties, transportation, manufacturing of the base materials, waste disposal management, corrosion resistance, and required operating pumping power, the impact in the environment of the (Li, Na, K)₂CO₃ with alumina nanoparticles can be higher or lower compared to the impact of the molten salt itself. Particularly, the dynamic viscosity together with the synthesis protocol are both fundamental factors to evaluate the effect of the addition of nanoparticles on the life cycle analysis.

8. Potential Geothermal Applications

After the circulation of fluids through the ground loop system, geothermal fluids, such as mineral oil, become saturated with some minerals due to the interaction between the geothermal fluids and the earth surface underground. Hence, the oil recovery from shale bearing or hydrate bearing sediments is an issue that should be seriously considered and involves many benefits and limitations, such as the ones described in [135,136,137,138]. The enhanced oil recovery (EOR) involves thermal recovery methods that increase the temperature of the existing reservoirs to heat the crude oil and, at the same time, reduce its viscosity and vaporize a fraction of the oil, facilitating the recovery through the decrease of the mobility ratio of the oil. The thermal recovery methods include, among others, the steam flooding, the nanofluid flooding, in-situ combustion, and the electrical heating. Additionally, thermal recovery methods induce the oil expansion, which in turn will decrease the oil viscosity and increase its permeability. The molten salt nanofluids have great potential to be applied in nanofluid flooding procedures, given that the molten salt nanofluids can withstand the harsh environment of the earth surface underground of elevated temperatures, high shear stresses, pressures, and salinity, and remain stable in the reservoirs. The use of molten salt-based nanofluids with the addition of silica, or alternatively, hybrid nanoparticles (e.g., iron oxide, alumina, and silica) in the EOR technological field should be further addressed with different types of sediments. Moreover, the combined usage of molten salt nanofluids with the incorporation of nano-scaled catalysts seems to be a promising innovative oil recovery approach, which should be further assessed by accurate experimental works. On the other hand, the molten salt nanofluids playing the role of thermal storage medium may be potentially applied in the geothermal heat extraction and conversion from oil reservoirs by using geothermal heat exchangers or heat pipes. Nonetheless, this potential application should be further explored based on further experimental works and numerical simulations such as the one presented in [139].

9. Limitations and Future Prospects

The main obstacles of the molten salt-based nanofluids and their possible solution strategies can be summarized in the following points:

• The available data on the properties of the inorganic salts and the nanoparticles is still rather scarce. The specific information of the nanoparticles size and morphology and consequent effect on the final characteristics of the molten salt nanofluids should be addressed in-depth.
• The scalability and reproducibility of the preparation methods of the molten salt nanofluids should be further studied. The most used method to produce molten salts nanofluids usually requires the employment of large amounts of water and powered ultrasonication. This strongly hinders the large-scale industrial possible uses and the reproducibility of the preparation method. Hence, some innovative methods are required to tackle these limitations, such as the mixing and the uniform dispersion of the nanoparticles within the liquid salts under high-temperature scenarios.
• The underlying mechanisms for the strong enhancement of the specific heat capacity and thermal conductivity of the molten salt nanofluids should be better understood. For instance, the arrangement/packaging of the ions in the secondary nanostructures in the compressed layer as well as in the percolation network is of relevance. Additionally, the nucleation, growth and assembly of the secondary nanostructures are mechanisms that required further analysis. The correlations between the structure and the properties for these secondary nanostructures should also be numerically modeled and validated in a laboratory environment.
• The long-term stability of the molten salt nanofluids should be assessed deeper. Some studies revealed that the nanoparticles in a molten salt agglomerate and settle only after a few hours. A deeper understanding of the interaction mechanisms between the molten salts and nanoparticles at a molecular level is needed to proceed in the search of the most stable dispersions. The employment of auxiliary dispersing techniques, such as sonication or vigorous stirring to break up the agglomerates, may be the secure pathway to follow. Additionally, the use of surface modification functionalized nanoparticles may appreciably diminish their clustering into the molten salt.
• It is suggested to obtain a correlation between the ion adsorption onto the surface of the nanoparticles and the long-term stability of the molten salt nanofluids. For instance, in the case of nitrate salt nanofluids, the adsorption of the nitrate ions onto the surface of the nanoparticles induces a strong electrostatic repulsive force that reduces the possibility of agglomeration and settling.
• Innovative numerical models and experimental methods should be implemented to determine the behavior between the molten salt nanofluids and an electrically charged heating surface at high temperatures. In this direction, sophisticated equipment operating at temperatures superior to 500 °C will need to be developed to obtain improved knowledge of the involved transport mechanisms.
• The measuring methods of the thermophysical properties of the molten salts nanofluids should be standardized to allow a reliable comparison of results between the different research teams. The methods usually encountered in the literature of the field to measure the specific heat capacity, latent heat capacity, and dynamic viscosity of the molten salts nanofluids made any concluding comparison or tendency of the intrinsic properties of certain suspensions difficult.
• It is recommended to develop methods and equipment suitable to the high-temperature conditions and to the corrosive character of the molten salts. For instance, the current thermal conductivity measuring procedures are not appropriate for molten salts. A similar limitation occurs in measuring the size of the nanoparticles, given that the indirect measurements through scanning electronic microscopy and other techniques, or the implementation of in-house developed protocols and apparatuses have been the followed so far. However, these routes do not present a high level of accuracy and representativeness, hindering the comparison of results.
• The influence of the nanoparticle size in the specific heat of the molten salt nanofluids should be clarified. The published articles do not usually report the initial size and morphology of the nanoparticles, although their size or agglomerated clusters in the final nanofluids is often summarized. The size of the nanoparticles is commonly measured by SEM imaging observation or, alternatively, by DLS in the solid state, and not in the real state of the nanofluids, thus constraining the evaluation of the impact of the size of the nanoparticles on the specific heat of the molten salt nanofluids.
• The viscosity of the molten salt nanofluids should be further investigated, given that the rheological properties are of vital importance for possible large-scale ends. In this direction, some feasible strategies and solutions should be undertaken to prevent an excessive viscosity. The latter can provoke an increment in the pressured drop of the systems and, hence, an extra pumping power will be needed for the systems to operate.
• There should be more future experimental works on the corrosion concern of the molten salts nanofluids, since the corrosive nature of the salts can induce severe erosion of the heat pipes, heat exchangers, retention tanks, and other equipment with the continuous motion of the nanofluids. Furthermore, it is of relevance to achieve a better knowledge about the corrosion potential modification of the salts caused by the inclusion of the nanoparticles.
• Further specific studies should be carried out according to the potential application field that include corrosion resistance evaluation and inherent costs of the salt purification to conclude whether the carbonate salt nanofluids are or not more suitable than the binary nitrate and ternary chloride salts for the same intended applications.
• It is suggested to explore the applicability of the soft computing techniques for modelling the specific heat capacity and other thermophysical properties of the molten salt nanofluids. These techniques should consider the effect of the working temperature and concentration, size, and intrinsic specific heat of the nanoparticles. Additionally, the computing techniques should rely on the modern machine learning prediction models and algorithms such as the optimized feed forward back propagation neural network and its corresponding ANN algorithm.
• More turbulent thermal convection flow modelling with molten salt nanofluids should be carried out. For instance, in the work performed by Harish et al. [140] the melting and turbulent heat transfer of potassium nitrate phase change material dispersed with hybrid nanoparticles of alumina–silica, alumina–titania, and alumina–multi walled carbon nanotubes inside a differentially heated rectangular enclosure was numerically studied. The impact of the high Rayleigh number and concentration of the nanoparticles on the transient melting behavior and melt pool temperature distribution are compared with that of the pure phase change material. Additionally, similar numerical simulations to those reported in [141,142,143] should be further performed considering molten salt nanofluids.
• Efforts should be made to accurately evaluate the economic impact of the use of molten salt nanofluids, given that the eventuality of implementing these fluids depends on its final cost. For instance, the enhancement of the values of the thermal properties and the required quantity of nanoparticles to reach those values should be studied. It should be emphasized that the incorporation of nanoparticles in molten salts can appreciably enhance the specific heat capacity, but the overall system cost could be unacceptable.
• The further performing of Life Cycle Assessment analysis for the molten salt nanofluids is highly recommended. Considering the uncertainties associated with the properties, transportation, manufacturing of materials, disposal, corrosion, and pumping power requirements, the environmental impact of the nanofluids may differ very much. The dynamic viscosity and manufacturing procedures are among the most relevant factors to determine the impact of the incorporation of nanoparticles according to the Life Cycle Assessment analysis.
• It is of relevance to conduct further quantitative studies about the thickness of the compressed liquid layer needed for a more accurate evaluation of the specific heat and latent heat changes for single salt, multi-salt mixtures, and nanoparticles. Hence, additional measurements are required to better understand the underlying mechanisms that cause alterations in the latent heat of the molten salt nanofluids and, furthermore, to the validation of these nanofluids as suitable media for solar thermal applications and latent heat storage systems involving encapsulation.
• Other possible applications of the molten salt nanofluids as thermal energy storage materials should be explored, including the conventional power generation in coal and gas power generation plants, nuclear power industry, generation of power from geothermal resources, industrial process heating, and desalination stations, among others.
• It is also recommended to evaluate other possible applications of the molten salt nanofluids prepared by the one-step method, such as molten salt batteries for electrical energy storage, processing of chemicals, refining of metals, production of ceramic nanoparticles, bio, and medical purposes, among others.
• It is suggested to carry out further experiments in the enhanced oil recovery area of research with molten salt nanofluids, rather than only with aqueous and oil-based nanofluids. The nanofluid flooding with molten salt nanofluids with the addition of silica, carbon, and zirconia nanoparticles should be further tested in pilot projects and oil fields. The enhanced thermophysical properties are promising for shale oil recovery, but bringing some more light on the enhanced oil recovery impacting factors is needed, such as wettability modification, interfacial tension reduction, and recovery magnitude through both laboratory experiments and in-situ practical situations.
• The assessment should be further elaborated for the systems composed of carbonate salts and alumina nanoparticles with various characteristics to elucidate which one is the most suitable in terms of environmental benevolence and sustainability.

10. Conclusions

The fundamental conclusions of this review can be summarized in the following points:

• Nowadays, the molten salt nanofluids present a great potential for improving the heat transfer capability and energy storage of thermal management and conversion systems, with the main favored industry for the development of this solution being the solar thermal power one.
• Although many considerable advances took their place so far, the knowledge stage is still on its infancy and further coordinated efforts should be made by the scientific community to fully comprehend the features of the molten salt nanofluids. In the medium term, the fully development of these thermal fluids would seriously impact concentrated solar power science and technology.
• In terms of carbonate molten salts nanofluids, the synthesis methods that do not imply the water dissolution of the carbonates are more suitable than the wet methods, given that these ones may lead to the chemical instability of the carbonates and to a heterogeneous mixture in water because of the different solubility. Additionally, this lack of homogeneity may be misleading since it usually conducts to a non-eutectic composition, which may be mistakenly taken as a specific heat capacity enhancement.
• The stability over time of the nitrate and carbonate molten salts nanofluids is usually poor, and the verified thermophysical properties improvement decreased within 3 to 4 h, or even less. However, the mechanical redistribution of the nanoparticles and the consequent recovery of the improved specific heat value is viable. This evidence highlights the need for potential instability solution procedures, such as vigorous mechanical mixing, pumping, and stirring.
• The accurate measurement of the thermophysical characteristics of the molten salt nanofluids requires well-defined methods, statistical analysis, cross-verification, and reproducibility of methods and results.
• A two-fold corrosivity reduction of stainless-steel was found for the (Li,Na,K)₂CO₃ molten salt with alumina nanoparticles, as compared to the molten salt itself at 600 °C. This corrosion rate decrement is mainly caused by the interaction of the alumina nanoparticles with the stainless-steel and the production of a mixed oxide containing aluminum with a counter-corrosion effect. These findings are consistent with the previous corrosion resistance inferring studies on nitrate salts nanofluids and suggest that some nanoparticles will migrate from the nanofluid to the corrosion layer.
• The published studies on the nanoparticle induce erosion of the thermal storage systems argued that the added nanoparticles did not cause an appreciable deterioration in the equipment, even in these cases where agglomeration occurred.
• The working temperature using the thermal energy storage medium in the solid and liquid phases and the thermal energy storage method using the sensible heat mechanism only or, alternatively, along with the latent heat mechanism, must be taken into account in the selection of a thermal energy storage medium exhibiting maximized thermal energy storage capability.
• When the thermal energy storage capability is the goal, the performing of intense thermostatic and thermal shock stability tests should be seriously considered to infer if the specific heat of the molten salt nanofluids suffer any considerable deterioration after being subjected to those tests.
• A relevant concern about the addition of phase change materials in molten salt nanofluids is the non-congruent melting process that diminishes the phase change reversibility and, consequently, the thermal energy storage capability. Hence, before a nanomaterial can be used as a phase change material, the characterization of its thermal features should be carried out, such as the temperature at which the phase change process occurs, sub-cooling, the rate of nucleation, and the enthalpy trend.
• The ternary molten salts, including the Hitec and Hitec XL, were found to be very suitable for operating in the concentrated solar power facilities because these salts had low melting points, which decreased the thermal dispersions. The nanoparticles that gave the highest increments in the thermophysical properties of the molten salt nanofluids were found to be titania, silica, alumina, and magnesia. Nevertheless, the multi-walled carbon nanotubes and the single-walled carbon nanotubes were also widely used for increasing the thermal storage density.
• The addition of alumina and silica nanoparticles can decrease the rate of corrosion of the molten salt nanofluids by 50% or more. Additionally, another suitable counter-corrosion measure is the reduction of the chromium content in the molten salts.
• The usage of NaNO₃ and KNO₃ in a two-tank storage configuration was found to be the most employed thermal energy storage technological approach with enhanced efficiency.