A review on supercooling of Phase Change Materials in thermal energy storage systems

Renewable and Sustainable Energy Reviews (xxxx) xxxx–xxxx

Contents lists available at ScienceDirect

Renewable and Sustainable Energy Reviews journal homepage: www.elsevier.com/locate/rser

A review on supercooling of Phase Change Materials in thermal energy storage systems ⁎

A. Safaria, R. Saidurb,c, , F.A. Sulaimanb, Yan Xua, Joe Donga a b c

School of Electrical and Information Engineering, The University of Sydney, NSW 2006, Australia Center of Research excellence in Renewable Energy (CoRE-RE), King Fahd University of Petroleum & Minerals (KFUPM), Dhahran, 31261, Saudi Arabia Faculty of Science and Technology, Sunway University, No. 5, Jalan Universiti, Bandar Sunway, 47500, Petaling Jaya, Malaysia

A R T I C L E I N F O

A BS T RAC T

Keywords: Supercooling Phase-Change Material Thermal energy storage Nanofluids Smart material Solar thermal Industrial applications

Thermal energy storage is at the height of its popularity to harvest, store, and save energy for short-term or long-term use in new energy generation systems. It is forecasted that the global thermal energy storage market for 2015–2019 will cross US$1,300 million in revenue, where the highest growth is expected to be in Europe, Middle East, and Africa followed by Asia-Pacific region. Thermal energy storage has become an inevitable component of fluctuant renewable energy systems due to their significant role in increasing efficiency and Quality of Service (QoS). Currently, one major research stream in such systems is improving the efficiency of heat exchangers and heat carriers. Hence, studying thermal behavior and thermophysical properties of heat storages is of great importance. In this study, we review a common but not very well-known problem of supercooling of Phase Change Materials (PCM). Supercooling is a thermophysical property of PCMs that is problematic in thermal storage applications. This review looks at supercooling from another point of view and investigates applications (such as specialized thermal storage applications) that can put supercooling into operation. To achieve this, development of techniques to increase state stability and designing reliable and stable supercooled heat storage systems will be investigated. The study will look at the thermal energy storage of supercooled liquids, degree and measurement of supercooling. Furthermore, factors that influence degree of supercooling and their effect on output capacity will be discussed. It looks at the supercooled material in four major categories and looks into the mechanisms for triggering crystallization in supercooled liquids. Applications including solar thermal storage will be the discussed in details. From the results discussed in this review researchers will identify and gain insight into supercooling control techniques, which are necessary for developing efficient heat exchangers, and also essential for promoting adoption of sustainable renewable energies.

1. Introduction Thermal energy storage systems are becoming particularly important for enhancing system reliability and Quality of Service (QoS) in new energy generation systems. Storing heat from renewable energy sources in thermal energy devices and providing it in the case of energy shortage or voltage sags enhances the overall system efficiency and promotes share of renewables in the energy mix. Use of solar energy as an energy source for PCM devices was started dates back to 1942 [1]. Current pervasive and entrenched use of thermal storage is well reflected in the report of the Global Thermal Energy Storage Market 2015–2019, which forecasts it will cross US$1,300 million in revenues by 2019 where Europe, the Middle East and Africa are expected to

witness the highest growth followed by Asia-Pacific region [2]. The thermal energy storage systems store thermal energy for consumption at a later time for heating or cooling applications or even power generation. They use sensible heat, latent heat or heat from thermo-chemical processes. Examples of sensible heat storage are liquid and air based systems, which use water and rock bed for heat storage, respectively [3,4]. Another example is using refractory bricks for heat storage in load levelling applications [5]. These heavy bulky bricks are also known as night storage heaters because they heat space during the day from the stored heat during the night. The sensible heat storages have low energy density and variable discharging temperatures. Hence, they are not efficient when compared with storage devices that involve latent or thermo-chemical heat storage process [6].

Abbreviations: PCM, Phase change material; QoS, Quality of Service; PV, Photovoltaic; SAT, Sodium acetate trihydrate; PEG, Polyethylene glycol; PCMMC, PCM microcapsules; MF, Melamine Formaldehyde; DSC, Differential Scanning Calorimetry; HVAC, Heating, ventilation, and air conditioning; COP, Coefficient of Performance ⁎ Corresponding author. E-mail addresses: [email protected], [email protected] (R. Saidur). http://dx.doi.org/10.1016/j.rser.2016.11.272 Received 2 November 2015; Received in revised form 21 July 2016; Accepted 30 November 2016 1364-0321/ © 2016 Elsevier Ltd. All rights reserved.

Please cite this article as: Safari, A., Renewable and Sustainable Energy Reviews (2016), http://dx.doi.org/10.1016/j.rser.2016.11.272

Renewable and Sustainable Energy Reviews (xxxx) xxxx–xxxx

A. Safari et al.

cycles. The heat transfer rate, effect of the supercooling level on heat recovery, effect of the rate of cooling on the degree of supercooling, and energy saving benefits with supressed supercooling material will be presented subsequently. In addition, we will present how supressing supercooling will help developers to incorporate PCMs into heating/ cooling systems, and how the improvements will affect the final cost analysis in real life examples. Publications have used undercooling [10], subcooling [5] and supercooling [15] alternatively. We review all these works and consider these phrases (i.e. undercooling, subcooling and supercooling) as the same. It is expected that this article will fill the research gap in reviewing supercooling of PCM along with providing new future directions and potential applications on the topic.

Table 1 Performance comparison of sensible, latent, and thermo-chemical thermal storages [9]. Thermal system

Capacity (kW h/t)

Power (MW)

Efficiency (%)

Storage period (h, d, m)

Cost (€/kW h)

Sensible heat

10–50

50–90

d/m

0.1–10

Latent heat Thermo-chemical heat

50–150 120–250

0.001– 10 0.001–1 0.01–1

75–90 75–100

h/m h/d

10–50 8–100

Thermal energy is also found in supercooled liquids where the material is in thermal equilibrium with its surroundings. The stored latent heat of fusion is released by triggering the crystallization of the supercooled liquid. Currently, latent heat storage which can be absorbed or released during the melting and solidifying process of Phase Change Materials (PCMs) is known as the most efficient way to store cold energy [7] or for heat recovery [8]. To show the advantages of the latent heat over the sensible heat, Hauer [9] conducted a comparison between the sensible heat storage of water, the latent heat storage and the heat from thermo-chemicals. Table 1 shows a comparison of capacity, power, efficiency, storage period and costs between latent, sensible, and thermos-chemical based systems. Based on the table, latent heat provides higher energy storage density compared to sensible heat based systems and less cost compared to techno-chemical based devices. Having said all advantages of latent heat storage, poor stability of supercooled material have caused key problems in their applications. For instance, poor stability leads to degradation in thermal properties in long term cycling. Corrosion between supercooled liquid and its container [10], change in density [11], experiencing phase segregation are other leading problems. Supercooling is “the delay in the start of solidification” and takes place whenever a PCM undergoes a phase change from liquid to solid [10]. It is a state where liquid PCM does not solidify immediately upon cooling below the freezing temperature, but start crystallization only after a temperature well below the melting temperature is reached. Thus, it is necessary to reduce the temperature below the phase change temperature to start crystallization and to release the latent heat. The latent heat is the largest proportion of charged thermal energy and would only release below the supercooled temperature [10]. During supercooling, sensible heat would be lost but latent heat is released instantly on crystallization. Supercooling and crystallization rate are categorised as kinetic properties of PCM [12]. Supercooling leads to reduced crystallization temperatures; thus the latent heat will be released at a lower temperature (wider temperature range) [13]. As a result, large temperature difference between charging and discharging is needed to fully utilize the latent heat, which is undesirable for efficient energy storage applications. Thus, supercooling is a key figure and a critical issue from the practical point of view and understanding the factors and methods to control supercooling is fundamental to advance thermal energy research and technology. Presently, research on materials and product development of latent heat storages is carrying out by several developers. Broad review articles on latent heat transfer [14], and quite sizeable amount of works are reported on PCM applications [5]. Among these problems, there is no critically analyzed review article that deal with supercooling and related techniques to control partial melting process, adding nucleating agents or discussing unpredictable character of crystallization. We will look into the thermophysical property data, simulations and experimental studies to identify short-term and long-term behavior of supercooled material through their useful life and thermal

2. Thermal energy storage of supercooled liquids Supercooling is a metastable state of PCMs in which they remain in liquid phase when cooled below their melting point temperature. For instance, pure water can be cooled down to −41 °C at atmospheric pressure in the laboratory without taking place of transition into solid phase [16]. Fig. 1 shows a supercooled liquid and its solid state. A supercooled liquid requires additional energy to release the stored latent heat. Hence, using supercooled liquid is not an efficient method in short-term applications. Nonetheless, supercooling is an interesting characteristic in long-term applications because supercooled liquid can be kept in a thermal equilibrium with the ambient temperature for a long period of time. During the initial cooling process, releasing the heat is inevitable but the remaining phase transition energy can be stored for extended periods of time without further loss of heat [15]. Supercooling is one of the very common phenomena in nature and technology processes but still new to researchers and developers. Water is the best known PCM when studying supercooling occurrences in nature. Small insects and fishes survive from freezing during the cold season by controlling crystallization in their system and taking benefit of supercooling. Examples of using supercooled water in technology are in thermal energy storage in solar systems or commercial heat pads for cold climates [8]. Details of research on supercooling in water is welldiscussed in the literature [18]. Metastable supercooled liquids are extremely vulnerable to impurities and external disturbances. They are favourably disposed towards stable condition through forming a new phase. Thus, capability of a reliable nucleation when releasing the latent heat is very important [19]. In addition, supercooled material should have a large amount of activation energy for nucleation to avoid a spontaneous crystallization

Fig. 1. A supercooled liquid and its solid state [17].

2

Renewable and Sustainable Energy Reviews (xxxx) xxxx–xxxx

A. Safari et al.

Fig. 2. Supercooling curves [20].

since crystallization kinetics are critical and not the rate of heat transfer. It is also interesting to compare Fig. 2(a) and (b). Although both have the same extent of supercooling, the supercooled liquid of Fig. 2(b) attains temperature Tm very quickly indicating a higher thermal diffusivity. The supercooled liquid in Fig. 2(d) begins to solidify at Tm indicating no nucleation problems. However, the liquid and solid undergo supercooling during freezing which could be due to a very high rate of heat removal. The temperature takes a long time to drop to Tm in the case of supercooled liquid of curve Fig. 2(e) implying low thermal diffusivity in the liquid. That is the temperature at the location of the sensor is higher than Tm when freezing is taking place at another location [20].

in supercooled condition [8]. Although several developers have provided respectable theories of metastable liquids, the calculation of thermophysical properties still remains an open problem and controlling behavior or supercooled liquids is a challenge in technical applications. Fig. 2 shows supercooling curves and thermal behavior of a supercooled liquid before solidification begins. Figs. 2(a) and (b) depict supercooling due to poor nucleation. The temperature rises and stabilizes at the melting point (Tm ) immediately upon crystallization. Fig. 2(c) shows supercooling due to poor rate of crystal growth. When nucleation begins, the temperature rises and then stabilizes indicating that the heat removal equals the latent heat of solidification. However, the temperature stabilizes at the temperature lower than Tm due to poor crystallization kinetics. Such a process is called “kinetic controlled” 3

Renewable and Sustainable Energy Reviews (xxxx) xxxx–xxxx

A. Safari et al.

In this method, the temperature of the nanofluid in the glass breaker is measured by the thermocouple which is set in the middle of it. A precision thermometer ( ± 0.2 °C) calibrates the thermocouple and the temperature of the ethylene glycol solution is controlled at −15 °C by the thermostatic water bath during the freezing process. The average relative error is reported to be less than 3% when the temperature changed from −15 °C to 15 °C. Another supercooling measuring method is the “temperature history” method. The resultant enthalpy–temperature diagrams of this method are useful in identifying the useful temperature ranges of operation, temperature rise during crystallization, available enthalpy as a function of supercooling, and the fraction of the total heat of fusion lost in the initial supercooling process [23]. A variation of this method has been used in [15] to illustrate enthalpy–temperature curves of supercooling salt hydrates at different levels of supercooling. A refined version of the “ice fog technique” for controlling the nucleation and thereby the degree of supercooling was used in [24]. The nucleation of samples were facilitated at the temperature of interest by adding cold nitrogen gas into the chamber to form an “ice fog”. This technique successfully controlled the ice nucleation temperature of the solutions within 1 °C.

3. Degree of supercooling Degree of supercooling is the temperature difference between the melting and freezing (crystallization) points [13]. It can be shown as a hysteresis equation [21]:

∆Ts = Tm − Tf

(1)

where ∆Ts denotes the degree of supercooling, Tm is the melting temperature and Tf is the freezing temperature. Increasing the effective heat capacity of supercooled liquid is viable via reducing the interval of the melting and freezing temperature range (Tm − Tf ) which is the degree of supercooling. There are three methods to investigate the heat transfer during phase change numerically: the enthalpy method, the heat transforming method, and the effective heat capacity method. Enthalpy-temperature curves are commonly used to determine energy storage capacity over a given temperature range, while the effective heat capacity method is used to calculate the effective heat capacity (ceff), which is directly proportional to the stored energy and it is released during the phase change transition. The effective heat capacity is negatively correlated with the degree of supercooling and can be defined as [22]

ceff =

hf Tm − Tf

+ cs

4. Factors influence degree of supercooling (2) Factors influencing degree of supercooling have not been clarified in the literature; hence, there is no a uniform standard to select a specific additive or a method for the sake of reducing the degree of supecooling. Some of the features that have been addressed as the main factors influencing the degree of supercooling are listed below:

where h f is the sensible heat and cs is the solid phase change. Hence, the degree of supercooling should be small for less energy consumption and high heat capacity outcome. A large degree of supercooling means more energy is released during the initial supercooling process. Hence, the stored energy available during the solidification process decreases and, consequently, the maximum temperature is reduced which is not desirable [22]. Therefore, in this research area reducing the degree of supercooling is the main objective. A relevant research investigating supercooling behavior of three PCMs including two well-known salt hydrates, disodium hydrogen phosphate dodecahydrate, Na2HPO4.12H2O (DSP) and sodium acetate trihydrate NaCH3COO.3H2O (SAT), and also for the commercial product STL-47 is presented in [15]. Enthalpy-temperature curves of supercooled PCMs in metastable phase in ambient temperature showed a prevention in spontaneous formation of the thermodynamically stable solid phase by nucleation barrier. Sharp temperature increments were depicted due to the manual seeding after three hours of running the experiment [15]. Degree of supercooling of water-based graphene oxide nanofluid PCM was studied in [7]. Experiments were conducted on deionized water without any dispersants mixed by a small fraction of graphene oxide nanosheets. The experiments successfully decreased the supercooling degree of nanofluids by 69.1% and duration of the nucleation by 90.7%. Ice crystal nucleus did not grow on the side (thickness) surface of the graphene oxide nanosheet but grew on the top or bottom surface of the nanosheet only when the supercooling degree (ΔT) and the nanosheets size (D) satisfied the inequality, D.ΔT≥4.2×108. The results showed that for a certain supercooling degree, only when D is bigger than the critical edge length of the nanosheets (DC), ice crystal nucleus could form and grow on the nanosheets.

4.1. Heterogeneous nucleation Heterogeneous nucleation refers to the nucleation at the surface of foreign bodies such as the surface of heat exchangers and containers, suspended particles, impurities, and microscopic bubbles. Avoiding heterogeneous nucleation is an important criteria when studying degree of supercooling [25]. 4.2. Homogenous nucleation Homogenous nucleation is a rare behavior in phase transition process. Through this, pure supercooled liquid experiences spontaneous density or composition fluctuations forming embryos of a new phase. Thus, contrary to heterogeneous nucleation, solid nuclei forms without the help of a foreign solid surface. In homogenous nucleation, the new phase is formed spontaneously as soon as embryos grow to the critical size. Homogenous nucleation is important when studying supercooling since it determines the practical attainable limits of supercooled liquid [26]. 4.3. Rate of cooling The degree of supercooling is also dependent on the rate of cooling. There is controversy reports in the literature that explains the relation in different ways. It has been shown that a higher cooling rate increases the degree of supercooling. However, in [8] a report was made based on “general feeling” that a higher rate of cooling causes a lower degree of supercooling and easier crystallization. It may be the case that these variations are due to using different supercooled material. Also in [8], relevant experiments were not conducted and the report on supercooled SAT crystallize in the tank brush heat exchanger experiments were based on general observation. Fig. 4 shows Differential Scanning Calorimetry (DSC) curves of mC18 synthesized with stirring rate of 9000 rpm under various heating and cooling rates. The curves show that the degree of supercooling decreases as the cooling rate decreases. The degree of supercooling decreased quickly when the heating and

3.1. Measurement of supercooling There are various supercooling degree test systems which measure the melting and freezing point of a supercooled liquid with an acceptable accuracy. One of the common measuring systems is depicted in Fig. 3 [7]. It consists of a data collector, a computer, a Ttype copper-constantan thermocouple and a thermostatic water bath. The nanofluid is in a glass breaker placed in the ethylene glycol solution. The thermostatic water bath has an open top, hence, the experimental pressure is the atmospheric pressure. 4

Renewable and Sustainable Energy Reviews (xxxx) xxxx–xxxx

A. Safari et al.

Fig. 3. Schematic diagram of supercooling degree test system [7].

h l, Tsc = xh s, Tm +(1−x )h l, Tm

(3)

where hl,Tsc is the enthalpy of supercooled liquid at supercooled temperature, hs,Tm and hl,Tm are the enthalpies of solid and supercooled liquid at melting temperature. The fraction of formed solid (x) will be [8]: T

x=

h l, Tm − h l, Tsc h l, Tm − h s, Tm

∫ Cp, ldT =

Tsc

l (Tm )

(4)

where cp,l is the heat capacity of supercooled liquid and l(Tm) the latent heat at the melting temperature. This equation can be approximated if heat capacity is nearly constant in the temperature range as [8]:

x≈

Cp, l ∆Ts lSAT

(5)

where ΔTs is the degree of supercooling. Eq. (5) was used in [8] in studying supercooling pure SAT (latent heat=260 kJ/kg, specific heat capacity=of 3 kJ/kg) at 28.7 °C with testing conditions of ΔTs=86.7 °C and Tm=58 °C and crystallize all liquid SAT to solid. The effect of the degree of supercooling on output capacity can be measured by varying the crystallization onset temperature while water flow rate and water temperature are fixed. Authors in [8] studied the effect of the degree of supercooling of SAT on the output capacity, and found that a difference of 5 °C in the degree of supercooling affects only slightly the average output capacity. Results showed a larger effect when the crystallization onset temperature was 40 °C. They concluded that the degree of supercooling has a lesser influence on the output capacity but a stronger influence on the recovered heat. In other words, a small degree of supercooling helps to recover more latent heat but the effect on the output capacity is minor.

Fig. 4. DSC curves of mC18 synthesized with stirring rate of 9000 rpm with various heating and cooling rates [8]: (A) 0.1, (B) 0.4, (C) 0.8, (D) 1.3, (E) 2.5, (F) 5.0, (G) 10.0, and (H) 20.0 °C/min.

cooling rate was lower than 5.0 °C/min. When the heating and cooling rate was 0.1 °C/min, the degree of supercooling was approximately 0.2 °C. They highlighted the significance of the research on the prevention of supercooling in microencapsulated n-alkanes for the energy storage applications due to the release of the latent heat at a lower temperature. 4.4. Roughness of a metallic surface Another factor that influence the supercooling degree is the roughness of the container surface. In [27], the impact of the roughness of a metallic surface on the magnitude of the supercooling during freezing of an aqueous solution was studied. It was found that the higher is the roughness, the lower is the supercooling degree.

6. Supercooled material The PCMs are the primary materials that experience supercooling during their freezing process. From the production point of view, PCMs are either organic or inorganic. Fig. 5 shows the classification of PCMs, and subsequently Table 2 shows the advantages and disadvantages of each group. A thorough comparison between the organic and non-organic compounds is presented in [5,10]. The comparison of the volume and mass of the storage unit required for storing a certain amount of heat showed that the inorganic compounds, such as hydrated salts,

5. Effect of supercooling degree on output capacity Degree of supercooling is important to determine the fraction of the supercooled liquid that would be crystalized [8]. As an adiabatic process (a thermodynamic process that occurs without transfer of heat), the enthalpy of supercooled liquid can be calculated as [8]: 5

Renewable and Sustainable Energy Reviews (xxxx) xxxx–xxxx

A. Safari et al.

pure components with PEG blends is suggested to overcome the morphological constraints and changing the temperature range associated with melting or freezing. It was reported that this method positively affects the entanglements in inter-lamellar regions and reduces the tendency of higher-molecular-weight PEGs towards crystallization. Facts and features of crystallization and prevention of supercooling of microencapsulated n-alkanes can be found in [13].

Supercooling Material

Organic

Single temperature Eutetics

Inorganic

Single temperature Eutetics

Commercial Paraffin waxes

Hydrated salts

6.2. Supercooling fatty acids Fatty acids are another group of attractive PCMs for thermal energy storage applications. They show reproducible melting and freezing behavior over thermal cycles, and do not experience supercooling during freezing [12]. The melting range and latent heat transition of the fatty acids vary between 30 and 65 °C, and between 153 and 182 kJ/kg, respectively. These properties are superior in designing latent heat thermal energy storage systems. Contrary to paraffin waxes, the solid-liquid transition in fatty acids falls within a narrow temperature range [29]. Capric, lauric, palmitic, and stearic acids are commonly used fatty acids in space heating applications. Binary mixtures and thermal properties of these fatty acids have been analyzed, and a mixture of capric and lauric fatty acids for a low temperature storage have been evaluated in [5]. The melting point of the mixture was reported to be about 14 °C while the latent heat of melting was between 113 and 133 kJ/kg depending on the mixture composition. Fatty acids have been also used to evaluate the performance of the storage units in testing facilities. For example, dimethyl-sulfoxide (melting point of 16.5 °C and latent heat of only 86 kJ/kg), and maleic anhydride (melting temperature of 52 °C and a latent heat of 145 kJ/ kg) [5]. These fatty acids are not useful in real applications.

Fatty acids Fig. 5. Organic and inorganic PCMs.

have a volumetric thermal storage density higher than that of most organic compounds due to their higher latent heat and density. The most common PCMs are Paraffin waxes, fatty acids, hydrated salts and eutectics. 6.1. Supercooling paraffin waxes Paraffin waxes are one of the most common material that is used in supercooling applications. As an organic compound, pure paraffin waxes suffer from having a low thermal conductivity and being costly. Various approaches have been undertaken to increase the thermal conductivity of the paraffin waxes. Among them are metallic fillers, metal matrix structures, finned tubes, and aluminum shavings, which revealed significant improvement. Also, in order to decrease the cost of pure paraffin waxes, they are used in technical grade qualities called “commercial paraffin waxes” for thermal applications [5]. Commercial paraffin waxes have comparably moderate thermal storage densities, are cheap, do not experience phase segregation when melting and show stable chemical properties. They also endure negligible supercooling. Nonetheless, their low thermal conductivity 200 kJ (~ kg or 150 MJ/m3 ) limits their applications [8]. For example, a commercial paraffin wax, P-116 which is used by a large number of investigators. It has a melting temperature of about 47 °C and the latent heat of melting is about 210 kJ . Supercooling of paraffin waxes

6.3. Supercooling salt hydrates Hydrated salts are inorganic compounds with high volumetric heat storage density (~350 MJ/m3), relatively high thermal conductivity (~0.5 W/m K) and cheaper as compared to paraffin waxes. Hydrated salts are superior choices in thermal heat applications because they have a large energy storage density and a high thermal conductivity. However, they experience phase segregation, supercooling, corrosiveness, low density compared to paraffin waxes, and instability on thermal cycling, which consequently limited their applications [8]. To make the system more stable with thermal cycles and prevent formation of the heavy anhydrous salt, using extra water was recommended. However, this recommendation reduced the storage density and resulted in a system that operated with a large temperature swing [5].

kg

with melting point around 55 °C is also reported in [5]. Three commercial waxes with melting temperatures 44, 53, and 64 °C; and latent heat of 167, 200, and 210 kJ/kg, respectively, were used in [5] to improve the efficiency of the heat storage units using paraffin waxes. Large supercooling of Polyethylene glycol (PEG) paraffin waxes is considered as a reason of the blend's freezing in [28]. Hence, replacing Table 2 Advantages and disadvantages of thermo-physical properties of organic and inorganic PCMs. PCM Organic (e.g. Commercial paraffin waxes)

Inorganic (e.g. Hydrated salts)

Advantages

Disadvantages

• Moderate thermal energy storage density • Non-corrosive and thermally stable • Chemically with little or no supercooling • Crystalize with most building materials • Compatible range of melting temperatures • Wide latent heat per unit weight • High vapour pressure • Low cheap prices • Commercial latent heat per unit volume • High energy storage density • Large thermal conductivity • High phase change enthalpy • Greater • Non-flammable than organic compounds • Cheaper • Fusion heat do not degrade with cycling 6

200 kJ (~ kg

3

or 150 MJ/m

thermal conductivity (~0.2 W/m °C) • Low phase change enthalpy • Low changes in volume on phase change • High large surface area • Require • Inflammable • No segregation

to most metals • Corrosive from decomposition • Suffer tendency to supercool • Strong segregation • Phase of thermal stability • Lack • Need nucleating and thickening agents

Renewable and Sustainable Energy Reviews (xxxx) xxxx–xxxx

A. Safari et al.

Supercooling is a critical problem in development of salt hydrates applications. Most salt hydrates suffer from poor nucleating properties and undergo supercooling before crystallization. The rate of nucleation is generally very low at the fusion temperature. Another related problem in salt hydrates is that proper heat extraction can be accomplished by just a few degrees of supercooling while 5–10 °C supercooling prevents it entirely. Therefore, controlling the degree of supercooling is very important in extracting stored heat from slat hydrates. One solution to achieve a reasonable rate of nucleation in salt hydrates is to reducing the supercooling degree. Hence, energy is discharged at a lower temperature instead of being discharged at a fusion temperature. Possible approaches to reduce the supercooling degree in salt hydrates are (i) helping formation of crystals by adding nucleating agent which has similar crystal structure, (ii) maintain some crystals in a small cold region called “cold finger” to serve as nuclei, (iii) using suitable thickening agent to prevent incongruent melting, and (iv) promoting heterogeneous nucleation using containers and heat exchangers with rough surfaces [20]. Examples of early studies on hydrated salts are Glauber salt (Na2SO4.H2O) which contains 44% Na2 SO4 and 56% H2O by weight in 1952 [5]. It is one of the cheapest materials that can be used for thermal energy storage with melting temperature of about 32.4 °C and high latent heat of 254 kJ/kg (377 MJ/m3). Borax was suggested as a nucleating agent to minimize supercooling [5]. This process requires some thickening agent to prevent settling of the high density borax. Thickening agent has been also suggested to interfere with the phase segregation. Bentonite clay with the Glauber salt is an example of this kind. Using thickening agent reduces the thermal conductivity of the mixture and results in a lower rate of crystallization and heat transfer to the salt. Reducing supercooling salt hydrates is also reported more recently in [30–32] where a graphite-compound-material was made of the PCM embedded inside a graphite matrix. Through this experiment the heat conductivity in the PCM increased without much reduction in energy storage and decrease of volume change in paraffin. Examples of supercooling salt hydrates are disodium hydrogen phosphate dodecahydrate, sodium acetate trihydrate (SAT) and STL-47 [15]. Fig. 6 illustrates the supercooling behavior of these salt hydrates as the temperature decreases below the freezing points of the hydrates.

Table 3 List of most promising organic and inorganic eutectics. Material

Composition (wt %)

Melting point (°C)

Latent heat (kJ/kg)

Triethylolethane+water +urea Triethylolethane+urea CH3COONa.3H2O +NH2CONH2 Mg(NO3)3.6H2O+NH4NO3 Mg(NO3)3.6H2O +MgCl2.6H2O Mg(NO3)3.6H2O +MgBr2.6H2O

38.5+3105+30

13.4

160

62.5+37.5 40+60

29.8 30

218 200.5

61.5+38.5 58.7+41.3

52 59

125.5 132.2

59+41

66

168

eutectics and their compositions, melting points and latent heat is presented in Table 3 [12]. Eutectic mixtures of capric acid and lauric acid can be applied in building wallboards for heat energy storage under low temperatures. These eutectic mixture wallboards have thermal stability of melting temperature. They have stable thermal properties and, therefore, are suitable options for latent heat storage in building energy conservation [33]. Thermal properties and thermal reliability of eutectic mixtures of some fatty acids for low temperature solar heating applications is investigated in [34,35]. A selection of eutectic mixtures of lauric acid– myristic acid (LA–MA), lauric acid–palmitic acid (LA–PA), and myristic acid–stearic acid (MA–SA) was used under repeated melt/ freeze cycles. The thermal properties and thermal reliability of the selected eutectics leaded to the conclusion that they are capable to be used in four-year energy storage periods or 1460 thermal cycles. Having a close look into the current literature, focus of the current naïve literature on eutectic mixtures is on fatty acids [33–37] indicating a long way to fill up the research gap in this area. 7. Mechanisms for triggering crystallization in supercooled liquids The main stream of research in developing supercooled heat storage is to control and automate crystallization at desired temperatures. Various control strategies have been proposed including automated or manual triggering of crystallization in supercooled liquid below a certain temperature. In automated approaches, a control unit is used to trigger the crystallization using an electrical signal. Physical mechanisms used to trigger crystallization can be classified into manual seeding and dynamic nucleation. Manual seeding is the most effective way to induce crystallization. In this mechanism seed particles of supercooled materials are added to supercooled the liquid. Subsequently, a crystalline structure starts to grow immediately on their surfaces and continue to grow as long as the temperature remains below the equilibrium temperature and there is an available liquid phase materials. Dynamic nucleation, on the other hand, uses external stimuli to induce the nucleation of crystals. Dynamic stimulation can be agitation, friction, shock waves or ultrasonic vibration [26]. Different mechanisms for triggering crystallization in supercooled liquids are reported to reduce or even decrease it to zero. The most common mechanisms are:

6.4. Eutectic mixtures The supercooled eutectics are mixtures of fixed proportion of supercooled materials that melt or freeze at a single temperature which is lower than the melting points of each constituents or of any other mixture of them. A list of most promising organic and inorganic

7.1. Adding nucleating agents Adding “nucleating agents” is the most widespread and leading method that researchers adopted to reduce the supercooling degree. Nucleating agents are solid particles or crystals of materials in structure and lattice parameters, which initiate crystallization but do not dissolve at the operational temperature [20]. Adding nucleating agents and impurities influences positively the

Fig. 6. Measured temperatures as the PCM supercool. Seeding after 3 h result in sharp temperature increase [15].

7

Renewable and Sustainable Energy Reviews (xxxx) xxxx–xxxx

A. Safari et al.

strated that the possibility of the emergence of ice particles was nearly 100% when the temperature reached −4.0 °C [22]. Besides that, the TiO2–BaCl2–H2O nanofluids revealed a much lower supercooling degree as compared to the BaCl2–H2O nanofluids, which implies that different additives have different nucleation effects [22]. The derivatives of n-alkanes are effective as nucleating agents to cancel the supercooling too [13]. Another method to improve the supercooling reliability of SAT is to add purified water to make the solution of the molar fraction (1:4) (salt:water) [15]. 7.2. Microencapsulation Encapsulation is an approach when a large heat transfer area is required in supercooled material, such as organic compounds. It reduces the PCMs’ reactivity towards the outside environment and controls the changes in the volume of the storage materials as phase change occurs. On the other hand, when organic PCMs are encapsulated in microcapsules, they tend to supercool severely, most likely due to the absence of nuclei in such small space. Fig. 8 shows a latent heat storage unit using flat containers for encapsulation. To control the supercooling temperature of microencapsulated supercooled liquids, [38,39] used various polymer shells as PCM microcapsules to lower crystallization temperature. Lowering the crystallization temperature was also reported by employing poly (divinylbenzene) and different copolymers as shell materials using suspension polymerization method [40–42]. Based on the results, the crystallization temperature of the encapsulated PCMs decreased about 10 °C lower than that of non-encapsulated PCM. The melting temperature was not significantly affected though. The same method was used in [43,44] but with poly methyl methacrylate instead of poly (divinylbenzene) as a polymer shell. The supercooling of paraffin octadecane microcapsules can be as large as 13.6 °C. Using shell-induced nucleation of the triclinic phase and the metastable rotator phase with an optimised shell composition and structure, the homogeneous nucleation can be mediated during the melt crystallization into the thermodynamically stable triclinic phase [21].

Fig. 7. Influence of increasing impurity on freezing curves [20].

Table 4 Effect of adding nucleating agents to PCMs on reducing the degree of supercooling. PCM

Na2SO4.10H2O Na2HPO4.12H2O

CH3COONa.3H2O

Na2S2O3.5H2O

Nucleating agent (size, µm)

Borax (20×50–200×250) Borax (20×50–200×250) Carbon (1.5–6.7) TiO2(2–200) Copper (1.5–2.5) Aluminum (8.5–20) Na2SO4 SrSO4 Carbon (1.5–6.7) K2SO4 Na2P2O7.10H2O

Supercooling (°C) without nucleator

with nucleator

15–18 20

3–4 6–9 0–1 0–1 0.5–1 3–10 4–6 0–2 4–7 0–3 0–2

20

30

7.3. Adding nucleating agent or metal additives to the PCM prior to encapsulation

freezing curves of the PCMs. In a contaminated PCM, however, only the pure melt transfers into crystallization. As the crystallization progresses, the concentration of impurities in the liquid phase increases and the melting temperature reduces. Fig. 7 shows the effect of impurity on the freezing curves. Effect of adding nucleating agents to PCMs on reducing the degree of supercooling is also shown in Table 4 [5]. Reducing the degree of supercooling in the ice formation process was studied in [22]. Results showed that the degree of supercooling was affected by adding different nanoparticles into the pure water. Hence, it was concluded that the change in the degree of supercooling was due to the wettability and the crystal structure similarity between the ice and the additives. Adding nucleation additives in pure water also demon-

The PCM microcapsules (PCMMC) must not experience any supercooling, otherwise cannot be used in any latent heat applications. The effect of adding a nucleating agent to PCM prior encapsulation on the crystallization temperatures were investigated in [13,45]. The experiments were carried out on the melamine-formaldehyde (MF)/ndodecane microcapsules and gelatin/n-tetradecane microcapsules, and different types of nucleating agents, such as sodium chloride, 1octadecnol and paraffin. The results revealed that adding 2 wt% of 1tetradecanol (nucleating agent) effectively reduced the degree of supercooling. Adding about 6 wt% of sodium chloride to the emulsion or 9 wt% of 1-octadecanol in the core material prevented supercooling but

Fig. 8. Schematic of a latent heat storage unit using flat containers for encapsulation [5].

8

Renewable and Sustainable Energy Reviews (xxxx) xxxx–xxxx

A. Safari et al.

7.4. Controlling the size of microcapsules

Table 5 Effect of microencapsulation on pure n-alkanes. Pure n-alkanes

C12

C14

Degree of supercooling (°C) Microencapsulation (°C) Degree of supercooling (°C) after adding 3 wt% of C36 Reduction in enthalpies in microencapsulation containing C36 (%)

4.1 13.2 5.8 14.3

2.9 8 2.9 6.4

The latest technologies attempt to produce PCM microcapsules with a smaller diameter to reduce the degree of supercooling [49]. The crystal growth is controlled by the size of liquid droplets or PCMMCs after nucleation [50]. It has been found that the size of the microcapsules has a direct effect on the crystallization temperature of melamine-formaldehyde (MF)/n-dodecane microcapsules [4]. The degree of supercooling between microcapsule with 5–100 μm and 100–1000 mm diameters was compared in [4]. The results showed that the degree of supercooling were same for capsules with 100– 1000 mm diameters but it increased as the microcapsules sizes decreased in the range of 5–100 mm [13]. Enlarging the size of microcapsules are negatively correlated with their functionality in building applications. Hence, the diameter of capsules should not be very large since they often rupture during mixing [51–53]. To overcome this problem, a nucleation agent is usually mixed with the PCM prior to encapsulation.

caused a very rough surface of the microcapsules and poor stable aqueous dispersion. To solve these problems, 20 wt% paraffin (melting point 62–65 °C) was added into the core material. It eliminated supercoiling of n-octadecane and had no effect on morphology and aqueous dispersion of microcapsules. The only downside of this solution was that adding a large quantity of non-PCM material resulted in reducing latent heat of encapsulated liquid [4]. Effect of silica fume and tetradecanol as nucleating agents on supercooling in n-tetradecane microcapsules were investigated by [46,47]. Based on their findings, very small amount of silica fume (0.2 wt%) was ineffective in suppressing the supercooling of microencapsulated n-tetradecane. In another experiment, using 1–4 wt% silica fume in tetradecanol have greatly suppressed the supercooling. Nucleating agents n-dodecane (C12) and n-tetradecane (C14) were used in melamineformaldehyde microcapsules through “in situ polymerization” method in [47]. To manage the supercooling problem, a small amount of n-hexatriacontane (C36) was introduced as an organic gelator into the core of microcapsules. Table 5 lists the effect of microencapsulation on pure n-alkanes C12 and C14 using DSC analyses. Microencapsulation increased the degree of supercooling but adding 3 wt% of C36, which reduced the degree of supercooling. Authors in [48] fabricated silver nanocomposite PCMMCs to suppress supercooling. As a result the stored energy in PCMMC were released in two stages of phase change in cooling process. There was two peaks: first peak was exothermic peak (peak alpha) which corresponded to a heterogeneously nucleated liquid-rotator phase transition, and the second exothermic peak (peak beta) which corresponded to the homogeneously nucleated rotatorecrystal transition. The supercooling supressed to about 1–3 °C for the silver nanocomposite PCM microcapsules. Thus, silver nanoparticles in the core has acted as a foreign nucleus in crystallization stage. The effects of adding different types and concentrations of nucleating agents prior to encapsulation to the PCM on their thermal properties, shell permeability and surface morphology were studied in [5]. The study assessed the quality of the microcapsules, extensive testing of these PCM microcapsules using an accelerated thermal cycling test, mass loss test, beside the FT-IR, DSC, TGA, and SEM tests, which were rarely shown in previous publications. Elimination of supercooling in microencapsulated PCM was also discussed in [4]. The authors used Rubitherm®RT58 and1-octadecanol nucleating agents to supress supercooling in PCM microcapsules. Suspension polymerization method and Fourier transformed infrared (FT-IR) measurements were used to encapsulate the RT21. Precise DSC measurements confirmed that supercooling reduced the onset of crystallization temperature of encapsulated RT21 10 °C lower than that of the non-encapsulated RT21. Furthermore, using either RT58 or 1octadecanol reduces the degree of supercooling remarkably. However, this method have a disadvantage: adding 1-octadecanol negatively affected the PCM thermal behavior by spreading the solidification over a wider temperature range. The nucleation probability of supercooled water inside cylindrical capsules with or without nucleators during a cold storage process was experimentally investigated in [18]. Adding nucleating additives to supress supercooling have been recently replaced by new approaches that do not require any nucleating agents.

7.5. Optimizing the composition and structure of the capsule shells Optimizing shell composition and structure of microencapsulated PCMs is a new method of supercooling suppression as compared to adding the nucleation agent approaches. Supercooling suppression techniques in microcapsules is presented in Table 6. An experiment in [21] showed that with an optimised shell composition and structure, the need for nucleating additives is eliminated. A homogeneous nucleation was successfully mediated by shell-induced nucleation of the triclinic phase and the metastable rotator phase. Three cooling peaks in the octadecane microcapsules have been observed on the DSC cooling curves. These peaks were corresponding to the shell-induced liquid-rotator and liquidetriclinic transitions, rotatorecrystal transition, and homogeneously nucleated liquidetriclinic transition. The freezing temperature lifted from 21.3 to 23.4 °C at mass ratio of formaldehyde to melamine 1.25 and a pH of 8.5. 7.6. Using nanofluid PCM to eliminate supercooling Nanofluids are perfect materials with reduced supercooling degree and rapid nucleation behavior for cold energy storage applications. Supercooling degree of graphene oxide nanofluids is shown in Table 7. where To is the onset nucleation temperature (K), Tp is the phase change temperature, (∆Td ) is the supercooling degree of deionized water, and (∆Tn ) is the supercooling degree of nanofluids. The supercooling degree of nanofluids is negatively correlated with the edge length of nanosheets. In order to reduce the supercooling degree of nanofluids, the edge length of the nanosheets should be increased. Table 8 presents the critical edge length of the nanosheet (Dc ) at different supercooling degrees. Through experiments in [7], a new nanofluid PCM was developed by adding a small amount of graphene oxide nanosheets in a deionized water. Thermal performance of nanofluid PCMs was also investigated in [55]. Evidences showed that nanofluid PCM had 25% less freezing time and subcooling was eliminated as compared to that of deionized water. In another study, [56] tested Al2O3.H2O nanofluids and recorded a reduction of 70.9% in the supercooling degree of 0.2 wt% nanofluids and a reduction of 32.9% freezing time, respectively. The Cu-H2O nanofluids were experimented in [57]. Based on the results, the heat release rate was higher than that of pure water at the temperature of thermostatic water bath of 13 °C and flow of water pump 81 min−1. These are favourable properties for thermal energy storage applications. Thermal properties of CuO-oleic acid nanofluids were investigated in [58]. Results recommended the CuO-oleic acid nanofluids performed better when comparing with traditional PCMs. Hence, they are more 9

Renewable and Sustainable Energy Reviews (xxxx) xxxx–xxxx

A. Safari et al.

Table 6 Supercooling suppression techniques in microcapsules. Supercooling suppression technique Adding a nucleating agent to PCM prior encapsulation on the crystallization temperatures of melamine-formaldehyde (MF)/ndodecane microcapsules and gelatin/n-tetradecane microcapsules adding nucleating agent or metal additives to the PCM prior to encapsulation Using different types of nucleating agents, such as sodium chloride, 1octadecnol and paraffin

Using silica fume and tetradecanol as nucleating agents in n-tetradecane microcapsules Fabrication of silver nano composite microcapsules (using silver nanoparticles in the core as a foreign nucleus) Preparing microcapsules using poly (divinylbenzene) and different copolymers as shell materials by means of suspension polymerization Controlling the size of the microcapsules In situ polymerization of melamineformaldehyde microcapsules containing either n-dodecane (C12) or n-tetradecane (C14), and introducing a small amount of n-hexatriacontane (C36) an organic gelator into the core of microcapsules

Results on supercooling degree and thermal properties

Year

Ref.

2 wt% of 1-tetradecanol (nucleating agent) reduced the degree • Using of supercooling dramatically

1996

[54]

2005

[13]

2005, 2008, 2004

[13,45,48]

2006

[46]

2008

[48]

2011

[40]

2012

[52]

2012

[47]

2014

[21]

2015

[4]

The degree of supercooling of microencapsulated n-octadecane was • decreased by adding 10.0 wt% of 1-octadecanol as a nucleating agent about 6 wt% of sodium chloride to the emulsion or 9 wt% of • Adding 1-octadecanol in the core material prevented supercooling approximately 20 wt% paraffin (melting point 62–65 °C) into • Adding the core material eliminated supercoiling of n-octadecane heat of the PCM present in the microcapsules was reduced by • Latent the addition of a large quantity of non-PCM material fume with 0.2 wt% was ineffective in suppressing supercooling • ofSilica microencapsulated n-tetradecane wt% of tetradecanol had greatly suppressed supercooling • 1–4 supercooling was supressed to about 1–3 °C for the silver nano • The composite PCM microcapsules crystallization temperature of the microcapsules reduced by • The 10 °C melting temperature was very slightly affected • The recommended for building applications because microcapsules • Not rupture during mixing pure n-alkanes crystalized at a higher temperature • The degree of supercooling was 4.1 and 2.9 °C for C12 and C14, • The respectively degree of supercooling was aggravated to 13.2 and 8 °C for C12 • The and C14 microcapsules, respectively degree of supercooling was reduced to 5.8 and 2.9 °C for C12 and • The C14 microcapsules respectively when 3 wt% of C36 was mixed with PCM

Optimizing the shell composition and structure without nucleating additives

enthalpies of microencapsulated C12 and C14 containing C36 • The were reduced by 14.3% and 6.4%, respectively homogeneous nucleation was mediated by shell induced • The nucleation of the triclinic phase and the metastable rotator phase

Adding different types and concentrations of nucleating agents (Rubitherm®RT58 and1-octadecanol) to the PCM prior to encapsulation

onset of freezing temperature shifted from 21.3 to 23.4 °C when • The the mass ratio of formaldehyde to melamine was 1.25 at a pH of 8.5 onset crystallization temperature of microcapsule was • The approximately 10 °C lower than that of the non-encapsulated RT21,

when the shell composition and structure were optimised

due to supercooling

degree of supercooling has been reduced dramatically when • The either RT58 or 1-octadecanol was used addition of 1-octadecanol negatively impacted on the PCM • The thermal behavior, which represented by spreading the solidification over wider range of temperature

the nanofluid PCMs and suggested them as an efficient and an effective alternative to the traditional PCMs for cooling energy storage applications. The effects of nanofluid on the supercooling degree was reported in [61]. The degree of supercooling was recorded for different concentrations of the TiO2-H2O. Solidification of carbon nanotube nanofluids was studied in [62]. Latest studies on the supercooling degree and solidification behavior of nanofluids is summarised in Table 9.

Table 7 Supercooling degree of graphene oxide nanofluids [7]. Concentration of nanofluids

T0/°C

TP/°C

ΔT/°C

(ΔTd−ΔTn)/ΔTd (%)

Deionized water 10 mg/100 ml 20 mg/100 ml 30 mg/100 ml 40 mg/100 ml 50 mg/100 ml

−6.8 −4.0 −3.6 −3.1 −2.8 −2.1

0.00 0.00 0.00 0.00 0.00 0.00

6.8 4.0 3.6 3.1 2.8 2.1

– 41.2 47.1 54.4 58.8 69.1

7.7. The rolling cylinder method Note: ∆T =Tp − To ; the phase change temperature of all the samples is 0 °C.

The cylindrical PCM container configurations are classified into three models [14]: the pipe model, the tube-tube model and the tubeshell model. These classifications are based on the heat transfer method between the PCM and the heat transfer fluid in tubes. Supercooling is a common problem in these tubes; hence a “rolling cylinder method” was proposed by scientists of General Electric Research and Development, New York to overcome the problem of salt segregation and supercooling of salt hydrates [63]. This method uses a cylindrical vessel mounted horizontally with two sets of rollers. A rotation rate of 3 rpm produces motion in the solid content. This method is used to create effective chemical equilibrium, prevent nucleation of solid crystals on the walls, and for rapid attainment of axial equilibrium in long cylinders. Advantages of the rolling cylinder method are: (i) complete

Table 8 The critical edge length of the nanosheet at different supercooling degree. Concentration of nanofluids (mg/100 ml)

10

20

30

40

50

ΔT (K) Dc (nm)

4.0 10.5

3.6 11.7

3.1 13.5

2.8 15.0

2.1 20.0

favourable for cooling applications. They confirmed that nanofluid PCMs reduce the nucleation time and the degree of supercooling. The phase change performance of TiO2-BaCl2-H2O nanofluids was reported by [59,60]. They advised an excellent thermal performance in 10

11

Adding pseudomonas as nucleating agent

Graphene oxide (two dimensional material)

0.1 wt% multiwall carbon nanotubes in deionized water

Deionised water

Carbon nanotube nanofluids: L-MWNT-1030, L-MWNT4060, L-MWNT-60100

Deionized water

TiO2 nanoparticles

SiO2 α-Al2O3

Pure water (H2O) Ethylene glycol

Saturated BaCl2 aqueous (BaCl2-H2O)

TiO2 nanoparticles

Saturated BaCl2 aqueous (BaCl2-H2O)

CuO nanoparticles with sizes ranging from 1 to 80 nm

Cu nanoparticles (particle size 25 nm)

Deionized water (H2O)

Oleic acid

Nanoparticle

Base material (solution)

Table 9 Latest studies on the supercooling degree and solidification behavior of nanofluids.

2

• • • • • • • •

nanofluids with 0.5, 1.0, 1.5 and 2 wt% of CuO nanoparticles could be saved by 7.14, 14.28, 25% and 28.57% respectively, than the base fluid The thermal conductivities of nanofluids PCMs was enhanced by 12.76% at −5 °C with volume fraction of 1.130% The supercooling degree was reduced by 84.92% The latent heat and specific heat are slightly decreased with suspending nanoparticles The viscosity increased with the increasing volume fraction, which had no effect on the cool storage system The nanofluids were recommended for the industries low temperature energy storage Adding nucleating agent helps to eliminate supercooling in nanofluids Solidification time of was reduced by 25% using nanofluids • 50% of mass solidified in 25% of total freezing duration The supercooling degree was reduced by 69.1%, and the nucleation was started in advance, shorting the time by 90.7%

2

The heat release rate was higher than that of pure water when the flow of water pump was 8 l/min and the • temperature of thermostatic water bath keep at 13 °C Cu-H O nanofluids had remarkable lower supercooling degree than water • The the mass fraction of Cu, the freezing time was shorter than that of deionized water • ByTheincreasing change temperature of Cu-H O nanofluids was about 1 ℃ higher than that of the deionized water, • whichphase indicates its great potential for lowering cost and reducing consumption of energy thermal performance and potential to replace the traditional PCMs in cool storage applications • Excellent high thermal conductivities compared to the base material • Remarkably The supercooling degree of 0.2 wt% nanofluids was reduced by 70.9% and the beginning time of freezing was • ahead by 32.9% additives of different mass fraction showed different nucleation effects • Same nano-additives mixed with the ethylene glycol solution showed poor nucleation, lower temperature of the • The supercooling rupture, and larger energy consumption The wettability between the additives and the basic fluid plays an important role in crystallization, and the • •better wettability results in the better nucleation mass concentration of the surfactant and the size of the carbon nanotubes influence the supercooling degree • The and freezing time the L-MWNT-1030, the carbon nanotube nanofluids had lower supercooling degree and shorter beginning • For freezing time than the deionized water With the mass concentration of surfactant 0.10%, the finishing freezing time was equal to that of the deionized • water the L-MWNT-4060 and L-MWNT-60100 carbon nanotube nanofluids, the supercooling degree and freezing • For time were comparable with those of the deionized water mechanism of nucleation and stability of the carbon nanotube nanofluids also influence the solidification • The behavior of thermal conductivity of nanofluids with different mass fraction of CuO nanoparticles was found • Enhancement to be higher than oleic acid The CuO-eoleic acid nanofluids was recommended as better PCMs for cooling thermal energy storage • applications These research results showed that nanofluids as PCMs could reduce the supercooling degree and the nucleation • time solidification times of nanofluids with 0.5, 1.0, 1.5 and 2 wt% of CuO nanoparticles could be saved by • Complete 10.71, 16.07, 19.64% and 27.67% respectively, than the base fluid. Similarly, complete melting times of

Effect on supercooling and thermal properties

[62]

2010

2015

2014

2012

[7]

[55]

[60]

[58]

[56]

2009

2012

[59]

[57]

Ref.

2009

2009

Year

A. Safari et al.

Renewable and Sustainable Energy Reviews (xxxx) xxxx–xxxx

Renewable and Sustainable Energy Reviews (xxxx) xxxx–xxxx

A. Safari et al.

and in heat and cold storage units. Application of supercooled liquid to trigger and warm up the engine before ignition is reported in [8]. Before that, preheating a bus petrol engine from –10 °C to 20 °C in less than 7 min with 65 kg of sodium hydroxide hydrate (NaOH.H2O) was experimented by [64]. In all these applications, however, it is necessary that the melting temperature of PCMs lies within a certain range of operation, melt congruently with a minimum supercooling and to be chemically stable. Further, its cost, toxicity, and corrosiveness exert influence on its widespread applications. Authors in [10] have emphasised on “easily adjustable melting point” as the most important factor in selecting a supercooling material for passive solar applications.

phase change, (ii) experimental latent heat released in range of 90– 100% of the theoretical latent heat, (iii) repeatable performance over 200 cycles, (iv) high internal heat transfer rates, and (v) uniform freezing [12]. 8. Supercooling energy storage applications Materials that experience supercooling are used in a wide range of applications based on their melting and solidifying temperatures. When selecting an appropriate supercooled liquid for different applications, the major selection criteria is its melting temperature. Appropriate melting temperature of inorganic substances is between 0 for water and 897 °C for K2CO3 [10]. They also differ widely in fusion heat, thermal conductivity, and density. Authors in [10] have classified supercooling materials into three main categories based on melting temperature of different applications:

• • •

9. Supercooling heat for solar thermal storage Supercooling heat storage is reserved for seasonal heat storage applications due to its “heat on demand” characteristics. Long term storage for seasonal storage is desirable to maintain system reliability and shifting some of the peak load to the off-peak load period. They also reduce energy consumption, CO2 emissions, and costs since approximately half of the energy consumed in industry and buildings is in the form of thermal energy. Abundant solar radiation can be collected and stored in supercooled liquid without heat loss during summer, and be utilised for space heating or water heating during winter. Further, solar thermal storage is the ultimate solution to overcome high energy prices during high demand seasons when customers are experiencing energy shortage or voltage sags. Hence, heat storage devices have become one of the main components in new energy generation systems, which is using solar energy. Fig. 9 shows a typical micro grid incorporating the four main microgrid system components i.e. photovoltaics, grid power, storage device and DC load. Conventional thermal devices such as NAS batteries are made from sodium and sulphur and, therefore, need to be maintained at high temperatures. They have substantially a high energy density, a reasonable efficiency and do not suffer from self-discharging. These advantages have made it the choice in several projects in Japan [23] and the USA [24]. Another example is a storage system made of high purity graphite blocks made by RAPS Pty Ltd. The storage of heat is within graphite blocks which are capable of storing large amounts of thermal energy at high temperatures for relatively long periods of time with minimal loss of energy. The graphite blocks are heated up using energy received from the grid, renewable sources, or by direct heating. This energy may then be used to produce steam through embedded heat exchangers and converted back to electricity with steam turbine generators [25]. The storage capacity of high purity graphite blocks ranges between about 300 kW h (thermal) per ton at 750 °C and about 1000 kW h (thermal) per ton at 1800 °C [26]. There are pilot deployments in Silverwater and King Island, Australia, for example.

Materials that melt below 15 °C for air conditioning applications. Materials that melt above 90 °C for absorption refrigeration. Materials that melt between 15 °C and 90 °C for solar heating or heat load levelling applications.

Comprehensive list of most common PCM being used for latent heat storage, and a large number of possible compounds that cover a wide range of temperatures are reported in [5]. Applications of supercooled material in thermal systems are reported in [10]:

• • • • • • • • • • •

Thermal storage: passive solar thermal storage, thermal storage in spacecrafts. Cooling: cooling solar modules in solar power plants, ice banks, electric and combustion engines. Buildings and architectures: passive storage in bioclimatic buildings, thermoelectric refrigeration. Heating: hot water, passive solar-heated buildings, solar air heating systems. Safety: temperature maintenance in rooms with computers or electrical appliances, protection of electronic devices and appliances. Food: solar cookers, thermal food processing, vegetable cooling, food protection and transport. Agriculture: wine, milk products (absorbing peaks in demand), greenhouses. Medical: transportation of drugs and biological materials (specially in rural areas), sanitary hot water, operating tables, hot–cold therapies. Transportation: thermal comfort in vehicles, warm-up or cooling engines. Chemical reactions: softening of exothermic temperature peaks. Entertainment and sports: Avoiding hypothermia during scuba diving, heat pack hand warmers, ice cream scoop handle.

Currently, supercooled liquids are very attractive in the context of the building and construction industry. Two general means to incorporate thermal storage materials in construction elements are: (a) encapsulation of supercooled liquid in high-density polyethylene pellets mixed with a gypsum wallboard material, and (b) immersion of conventional wallboard in a supercooled liquid. Installing PCM impregnated wallboard elements instead of ordinary material has the same installation costs and users will benefit of 3–5 years energy savings payback [29]. Supercooling compounds are used in heating, ventilation, and air conditioning (HVAC) systems in heat and coolness storage devices. Advancement of nanotechnology has emerged a new application for supercooled-based systems in thermal storage of artificial satellites, and recently they found their way in textile industry in making protective clothing [28]. Architectural integration of supercooled liquids can be found in building walls and other building components,

Fig. 9. Microgrid with four main system components including photovoltaics, grid power, storage device and DC load.

12

Renewable and Sustainable Energy Reviews (xxxx) xxxx–xxxx

A. Safari et al.

hot water from a common array of solar thermal collectors. Short wavelength solar radiation is converted directly into electricity in the PV cells, while the long wavelength segment of the spectrum is used to produce moderate to high temperature thermal energy. It is mostly used for underfloor heating, hot water, and space heating. Solar water heaters can be designed to embed a layer of supercooled-filled capsules underneath. During the day, waters is heated up using solar heat and the stored latent heat is transferred to the supercooled liquid beneath it. Hence, the supercooled material will be in a liquid state due to the absorbing of the latent heat. During cold hours when there is no more solar irradiation, the hot water is withdrawn and is substituted by cold water, which gains energy from the supercooled liquid. The latent heat is released by liquid to solid phase transition. The only disadvantage of this method is the poor heat transfer between supercooled liquid and water. One of the proposed integrations is a composition of a cylindrical supercooled storage unit filled with a paraffin wax (p-116) and stearic acid in the closed loop with a flat plate solar collector. The absorber plate–container unit performs the function of both absorbing the solar energy and storing the supercooled compound. The solar energy is stored in supercooled paraffin wax which is used as the heat storage medium [70]. Inclusion of a supercooled modules in water tanks for domestic hotwater supply is a very promising technology [71]. In [72], a conventional open-loop passive solar water-heating system was combined with a sodium thiosulfate pentahydrate-phase change material. In solar air heating systems integrated with supercooled modules, the air-based system utilizing sodium sulfate decahydrate as a storage medium requires roughly one-fourth the storage volume of a pebble bed and one half the storage volume of a water tank. Also, the PCM should be selected based on its melting point rather than its latent heat [12].

Extensive efforts are made to use supercooled liquids to improve the stability and efficiency of solar cells [65]. Supercooled SAT has shown a great efficiency in seasonal heat storage system due to its ability to supercool in ambient temperatures [66]. Two fatty acid esters and three eutectic mixtures of some esters including methyl stearate, methyl palmitate, cetyl stearate, cetyl palmitate, and their binary mixtures were studied in [29] as new materials for solar thermal storage. They studied solid/liquid transitions in fatty acid esters by DSC measurements and observed promising results for using them as solar heat storage material. The results showed a sharp phase transition which indicates high enthalpy of melting/fusion reversible process. Also, the system demonstrated a stable thermal properties over a large number of thermal cycles during 18 months, and the temperature of the melting/solidification of pure or mixed components were in a suitable range. Solar thermal collectors collect solar irradiation in the form of heat and use a heat medium fluid to transport the heat to a buffer store. It is important to increase the efficiency of the heat medium and the storage liquid through decreasing the heat loss process. They can be installed separately, side-by-side PV modules, or even can be fully integrated as PV/thermal units, and convert the wasted heat into useful energy. Supercooled salt hydrates and paraffin waxes are used in solar collectors for water heating. It is suggested that using a supercooled liquid increased the performance efficiency of the latent heat storage system due to its ability in keeping the hot water in the required temperature range [67]. Fig. 10 shows (a) a solar air conditioner and (b) a solar heat engine using solar thermal storages [20]. A thorough review on the applications of solar thermal systems was conducted in [68]. In all of these applications, the use of a supercooled material in solar collectors is favourable with respond to the high energy density, no corrosion, low processing techniques, and less maintenance. However, a low thermal conductivity, large thermal expansion, reduced efficiency at high temperatures (non-selective), weathering and ageing under high temperatures and ultraviolet light have been accounted as disadvantages [69]. Supercooled based modules can also be integrated into solar combisystems which provide both solar space heating and cooling as well as

10. Future directions This review paper has laid significant groundwork for further investigation in designing efficient and reliable supercooled thermal energy storage systems. Supercooled liquids are very useful in applications that demand short-term or long-term thermal storage such as food processing, transportation, medical sciences and air conditioning

Fig. 10. (a) A solar air conditioner and (b) a solar heat engine using solar thermal storages [20].

13

Renewable and Sustainable Energy Reviews (xxxx) xxxx–xxxx

A. Safari et al.

of buildings and material constructions. Further contributions can be made in more applications by investigating thermal properties and quantification of energy of supercooled liquids, and dissecting environmental and financial benefits in real-life examples. It seems a good idea to design multi-storage supercooled-based systems. In this suggestion, each heat module uses a supercooled liquid with a complementary melting temperature, which would be very useful in supercooled-based systems that require large temperature range. Another approach is using supercooled liquids in sub-storages (or co-storages) along with traditional heat exchangers to improve efficiency of the whole thermal system. One can use supercooled heat exchangers to increase the Coefficient of Performance (COP) of the system, and proceed with further optimisation steps such as increasing surface and size of the heat exchangers. The issue of reducing the degree of supercooling is an intriguing one which can be usefully explored in further research. There are several ways to reduce the degree of supercooling. One particular supercooling degree reduction method can be further investigated by realising time dependent characteristics and useful thermal cycles of each material. Another recommended future research is development of a material with a high melting point which is desired from the viewpoint of refrigerator efficiency. This can be done by developing a new efficient method for the control of nucleation in supercooled liquids, and developing a method to force the uniform melting of a crystallised PCM.

[10] [11] [12] [13] [14] [15] [16] [17] [18]

[19] [20] [21]

[22] [23]

11. Conclusions

[24]

Using the latent heat of supercooled liquid is embodiment of the so called “heat on demand” approach in improving the stability and efficiency of new energy generation systems. Supercooled liquids are reliable substitutes to be considered as an innovative power source for heat or cool supply or even for applying them directly to a process in various applications such as refrigeration and air conditioning. They are comparably cheaper than conventional thermal systems when using commercial compounds, such as commercial paraffin waxes, as a heat carrier. Taking full advantage of supercooling depends on understanding, the thermo-dynamic properties and development techniques for appropriate integration of systems and proper design of heat exchangers. Feasibility of integrating supercooled PCM into industrial applications also depends on modifying existing energy systems, heating and cooling demand analyses. In-built and sectioned storage systems with large collector modules are economically more favourable due to the use of a small initial cost for a large heat recovery benefits. The economic outlook for these systems is more admirable when the system is operating in remote regions where there is no access to a reliable stable energy sources.

[25]

[26] [27]

[28] [29] [30]

[31] [32]

[33] [34]

[35]

References

[36]

[1] Mae HG. Device for Accumulating, Retaining, and Discharging Heat. Google Patents. 1942. [2] Reports R. Global Thermal Energy Storage Market 2015–2019. 2015.. Available from: 〈http://www.reportsnreports.com/reports/378346-global-thermal-energystorage-market-industry-analysis-2015-2019.html〉. [3] Kumaresan V, et al. Role of PCM based nanofluids for energy efficient cool thermal storage system. Int J Refrig 2013;36(6):1641–7. [4] Al-Shannaq R, et al. Supercooling elimination of phase change materials (PCMs) microcapsules. Energy 2015;87:654–62. [5] Farid MM, et al. A review on phase change energy storage: materials and applications. Energy Convers Manag 2004;45(9):1597–615. [6] Agency IEEnergy Conservation Through Energy Storage (ECES) Programme. 1 August 2015. Available from: 〈http://www.iea-eces.org/files/090525_broschuere_ eces.pdf〉. [7] Liu Y, et al. Study on the supercooling degree and nucleation behavior of waterbased graphene oxide nanofluids PCM. Int J Refrig 2015;50:80–6. [8] Keinänen M. Latent heat recovery from supercooled sodium acetate trihydrate using a brush heat exchanger. Espoo: Helsinki University of Technology; 2007. p. 104. [9] Hauer A. Storage technology issues and opportunities. In: Proceedings of the

[37]

[38]

[39]

[40] [41]

[42]

[43]

14

international low-carbon energy technology platform, strategic and cross-cutting workshop “Energy storage - issues and opportunities”. Committee on Energy Research and Technology (International Energy Agency), Paris. France; 15 February, 2011. Zalba B, et al. Review on thermal energy storage with phase change: materials, heat transfer analysis and applications. Appl Therm Eng 2003;23(3):251–83. Velraj R, et al. Heat transfer enhancement in a latent heat storage system. Sol Energy 1999;65(3):171–80. Sharma A, et al. Review on thermal energy storage with phase change materials and applications. Renew Sustain Energy Rev 2009;13(2):318–45. Zhang X-x, et al. Crystallization and prevention of supercooling of microencapsulated n-alkanes. J Colloid Interface Sci 2005;281(2):299–306. Liu S, Li Y, Zhang Y. Review on heat transfer mechanisms and characteristics in encapsulated PCMs. Heat Transf Eng 2015;36(10):880–901. Sandnes B, Rekstad J. Supercooling salt hydrates: stored enthalpy as a function of temperature. Sol Energy 2006;80(5):616–25. Angell C. Supercooled Water. DTIC Document. 1983. Frequently Asked Questions About Phase Change Materials. 2015. Available from: 〈http://www.puretemp.com/stories/understanding-pcms〉. Chen S-L, Wang P-P, Lee T-S. An experimental investigation of nucleation probability of supercooled water inside cylindrical capsules. Exp Therm Fluid Sci 1998;18(4):299–306. Chernov A, Pil'nik A, Islamov D. Initial stage of nucleation-mediated crystallization of a supercooled melt. J Cryst Growth 2016;450:45–9. Garg H, Mullick S, Bhargava VK. Solar thermal energy storage. Holand The Netherlands: Springer Science & Business Media; 2012. Cao F, Yang B. Supercooling suppression of microencapsulated phase change materials by optimizing shell composition and structure. Appl Energy 2014;113:1512–8. Zhang X, et al. Analysis of the nucleation of nanofluids in the ice formation process. Energy Convers Manag 2010;51(1):130–4. Marín JM, et al. Determination of enthalpy–temperature curves of phase change materials with the temperature-history method: improvement to temperature dependent properties. Meas Sci Technol 2003;14(2):184. Rambhatla S, et al. Heat and mass transfer scale-up issues during freeze drying: II. Control and characterization of the degree of supercooling. Aaps Pharmscitech 2004;5(4):54–62. Dannemand M, et al. Experimental investigations on prototype heat storage units utilizing stable supercooling of sodium acetate trihydrate mixtures. Appl Energy 2016;169:72–80. Mullin JW. Crystallization. Oxford UK: Butterworth-Heinemann; 2001. Faucheux M, et al. Influence of surface roughness on the supercooling degree: case of selected water/ethanol solutions frozen on aluminium surfaces. Int J Refrig 2006;29(7):1218–24. Mondal S. Phase change materials for smart textiles – an overview. Appl Therm Eng 2008;28(11):1536–50. Nikolić R, et al. New materials for solar thermal storage—solid/liquid transitions in fatty acid esters. Sol Energy Mater Sol Cells 2003;79(3):285–92. Mehling, H, Hiebler, S, Ziegler, F. Latent heat storage using a PCM-graphite composite material. In: Terrastock 2000––Proceedings of the 8th international conference on thermal energy storage. Stuttgart, Germany; 2000. Cabeza LF, et al. Heat transfer enhancement in water when used as PCM in thermal energy storage. Appl Therm Eng 2002;22(10):1141–51. Py X, Olives R, Mauran S. Paraffin/porous-graphite-matrix composite as a high and constant power thermal storage material. Int J Heat Mass Transf 2001;44(14):2727–37. Shilei L, Neng Z, Guohui F. Eutectic mixtures of capric acid and lauric acid applied in building wallboards for heat energy storage. Energy Build 2006;38(6):708–11. Sarı A. Eutectic mixtures of some fatty acids for low temperature solar heating applications: thermal properties and thermal reliability. Appl Therm Eng 2005;25(14):2100–7. Sarı A, Sarı H, Önal A. Thermal properties and thermal reliability of eutectic mixtures of some fatty acids as latent heat storage materials. Energy Convers Manag 2004;45(3):365–76. Sari A, Kaygusuz K. Thermal performance of a eutectic mixture of lauric and stearic acids as PCM encapsulated in the annulus of two concentric pipes. Sol Energy 2002;72(6):493–504. Karaipekli A, Sarı A. Preparation, thermal properties and thermal reliability of eutectic mixtures of fatty acids/expanded vermiculite as novel form-stable composites for energy storage. J Ind Eng Chem 2010;16(5):767–73. Qiu X, et al. Microencapsulated n-alkane with p (n-butyl methacrylate-comethacrylic acid) shell as phase change materials for thermal energy storage. Sol Energy 2013;91:212–20. Chaiyasat P, Islam MZ, Chaiyasat A. Preparation of poly (divinylbenzene) microencapsulated octadecane by microsuspension polymerization: oil droplets generated by phase inversion emulsification. RSC Adv 2013;3(26):10202–7. You M, et al. Microencapsulated n-Octadecane with styrene-divinybenzene copolymer shells. J Polym Res 2011;18(1):49–58. Li W, et al. Fabrication and morphological characterization of microencapsulated phase change materials (MicroPCMs) and macrocapsules containing MicroPCMs for thermal energy storage. Energy 2012;38(1):249–54. Chaiyasat P, et al. Preparation of divinylbenzene copolymer particles with encapsulated hexadecane for heat storage application. Colloid Polym Sci 2008;286(2):217–23. Qiu X, et al. Fabrication and characterization of microencapsulated n-octadecane with different crosslinked methylmethacrylate-based polymer shells. Sol Energy

Renewable and Sustainable Energy Reviews (xxxx) xxxx–xxxx

A. Safari et al. Mater Sol Cells 2012;98:283–93. [44] Qiu X, et al. Microencapsulated n-octadecane with different methylmethacrylatebased copolymer shells as phase change materials for thermal energy storage. Energy 2012;46(1):188–99. [45] Fan Y, et al. Super-cooling prevention of microencapsulated phase change material. Thermochim Acta 2004;413(1):1–6. [46] Alvarado J, et al. Characterization of supercooling suppression of microencapsulated phase change material by using DSC. J Therm Anal Calorim 2006;86(2):505–9. [47] Kong-ying Z, et al. Supercooling suppression of microencapsulated n-alkanes by introducing an organic gelator. Chem Res Chin UNIVERSITES 2012;28(3):539–41. [48] Song Q, L.Y. , Yuen M, Xing JJ. Supercooling of silver nanocomposite PCM microcapsules. Fiber Bioeng Inform 2008;1:239–48. [49] Smith MC. Microencapsulated phase change materials for thermal energy storage: development, evaluation, and application. ResearchSpace@Auckland; 2009. [50] El Rhafiki T, et al. Crystallization of PCMs inside an emulsion: supercooling phenomenon. Sol Energy Mater Sol Cells 2011;95(9):2588–97. [51] Sun G, Zhang Z. Mechanical properties of melamine-formaldehyde microcapsules. J Microencapsul 2000;18(5):593–602. [52] Rahman A, Dickinson ME, Farid MM. Microencapsulation of a PCM through membrane emulsification and nanocompression-based determination of microcapsule strength. Mater Renew Sustain Energy 2012;1(1):1–10. [53] Hu J, Chen H-Q, Zhang Z. Mechanical properties of melamine formaldehyde microcapsules for self-healing materials. Mater Chem Phys 2009;118(1):63–70. [54] Yamagishi, Y., et al. An evaluation of microencapsulated PCM for use in cold energy transportation medium. In: Energy conversion engineering conference, 1996. IECEC 96, Proceedings of the 31st Intersociety. 1996. IEEE. [55] Chandrasekaran P, et al. Solidification behavior of water based nanofluid phase change material with a nucleating agent for cool thermal storage system. Int J Refrig 2014;41:157–63. [56] WU P, et al. Nucleation characteristics of nano-additives in binary ice thermal storage system. J Zhejiang Univ: Eng Sci 2009;43(9):1668–71. [57] Li X, Zhu D. Experimental study of cold storage characteristics of nanofluids as the phase change material. Mater Rev 2009;23(3):11–6. [58] Harikrishnan S, Kalaiselvam S. Preparation and thermal characteristics of CuO–

[59] [60]

[61] [62] [63] [64] [65] [66]

[67]

[68] [69] [70] [71] [72]

15

oleic acid nanofluids as a phase change material. Thermochim Acta 2012;533:46–55. Liu Y-D, et al. Experimental study of thermal conductivity and phase change performance of nanofluids PCMs. Microfluid Nanofluidics 2009;7(4):579–84. He Q, et al. Experimental study on thermophysical properties of nanofluids as phase-change material (PCM) in low temperature cool storage. Energy Convers Manag 2012;64:199–205. Jia Ls, Peng L, Chen Y, Mo Sp, Li Z. Investigation on solidification and melting processes of nanofluids 2013. Eng Thermophys 2013;34:328–31. Mo SP, et al. Experimental study on solidification behavior of carbon nanotube nanofluid. In Advanced Materials Research. Switzerland: Trans Tech Publ; 2011. Garg H. Advances in solar energy technology: volume 1: collection and storage systems. New Delhi: Springer Science & Business Media; 2013. Vasiliev L, et al. Latent heat storage modules for preheating internal combustion engines: application to a bus petrol engine. Appl Therm Eng 2000;20(10):913–23. Fei Z, et al. A supercooled imidazolium iodide ionic liquid as a low-viscosity electrolyte for dye-sensitized solar cells. Inorg Chem 2006;45(26):10407–9. Desgrosseilliers L. et al. Thermodynamic evaluation of supercooled seasonal heat storage at the drake landing solar community. In: Eurotherm seminar# 99 advances in thermal Energy storage; 2014. Fieback, K, Linderberg, G. Advanced thermal energy storage through phase change materials and chemical reactions–feasibility studies and demonstration projects. In: Proceedings of the Eighth work shop and experts meeting of annex. 2005. Mekhilef S, Saidur R, Safari A. A review on solar energy use in industries. Renew Sustain Energy Rev 2011;15(4):1777–90. Sandnes B. Energy efficient production, storage and distribution of solar energy. Norway: University of Oslo; 2003. Mettawee E-BS, Assassa GM. Experimental study of a compact PCM solar collector. Energy 2006;31(14):2958–68. Cabeza LF, et al. Experimentation with a water tank including a PCM module. Sol Energy Mater Sol Cells 2006;90(9):1273–82. Canbazoğlu S, et al. Enhancement of solar thermal energy storage performance using sodium thiosulfate pentahydrate of a conventional solar water-heating system. Energy Build 2005;37(3):235–42.