Keywords

1 Introduction

To utilize the exergy of solar and industrial exhaust heat, high-temperature thermal energy storage (TES) systems are required, which can effectively store the intermittent high-temperature heat of such sources. Recently, development of an advanced TES system with high heat storage capacity and high thermal responsivity was accelerated because conventional sensible heat storage (SHS) systems using ceramics, bricks, and liquid molten salts have low heat storage capacity and low thermal responsivity. Under these circumstances, latent heat storage (LHS) technology has attracted considerable attention. LHS is based on the storage or release of latent heat when a phase change material (PCM) undergoes phase transition from solid to liquid, or vice versa. LHS has three advantages: its latent heat provides high heat storage capacity compared to SHS, it can be a constant heat source at the phase-changing temperature during the phase transition, and the reversible phase-changing process allows for repeated use (Nomura et al. 2013).These advantages allow LHS to be used as an advanced TES system, instead of SHS. In this article, general LHS technology is outlined first; then recent advances in the use of LHS for high-temperature applications (over 100 °C) are discussed, with respect to each type of PCM (e.g., sugar alcohol (Kojima and Matsuda, 2004), molten salt, and alloys). In addition, the prospects for future LHS technology are discussed from the viewpoint of exergy.

Nomenclature

A

Energy level (−)

C P

Specific heat (J·g·K−1)

C P,l

Specific heat of liquid (J·g·K−1)

C P,s

Specific heat of solid (J·g·K−1)

\( \Delta {h}_{T_o}^0 \)

Enthalpy change of a chemical reaction under standard condition (J·mol−1)

L

Latent heat (J·g−1)

m

Mass (g)

Q

Quantity of heat (J)

\( \overrightarrow{Q} \)

Vector of heat source

\( \overrightarrow{R} \)

Vector of reaction

T

Temperature (°C)

T 0

Reference temperature (°C)

T m

Melting point (°C)

ε

Exergy (J·g−1)

ε L

Exergy of latent heat (J·g−1)

\( \Delta {\varepsilon}_{T_o}^0 \)

Exergy change of a chemical reaction under standard condition (J·mol−1)

Abbreviation

CSP

Concentrated solar power

DSG

Direct steam generator

HTF

Heat transfer fluid

LHS

Latent heat storage

PCM

Phase change material

SHS

Sensible heat storage

TES

Thermal energy storage

2 Principle of LHS

Figure 1 illustrates the differences between LHS using solid–liquid phase transition, and SHS. Here, the specific heat of PCM and SHS materials was constant. The cumulative heat storage capacity of SHS was

$$ Q={mC}_{\mathrm{P}}\left(T-{T}_{\mathrm{i}}\right) $$
(1)
Fig. 1
figure 1

Difference of cumulative heat storage capacity of LHS using solid–liquid phase transition and SHS

Here, Q is quantity of heat, m is weight of heat storage material, C p is a specific heat, T is temperature, and T i is initial temperature. In the SHS, the phases of the material do not change. On the other hand, Q of LHS was as follows:

$$ Q=m\left\{{C}_{\mathrm{P},\mathrm{l}}\left(T-{T}_{\mathrm{m}}\right)+L+{C}_{\mathrm{P},\mathrm{s}}\left({T}_{\mathrm{m}}-{T}_{\mathrm{i}}\right)\right\} $$
(2)

Here, C P,l is liquid sensible heat, C P,s is solid sensible heat, T m is melting point, and L is latent heat. LHS stores/releases thermal energy when it changes phase; therefore, Q drastically changes at T m. This simple principle provides high heat capacity, the function of use as an isothermal heat source, and repeatable use.

3 Heat Exchanger for LHS

Figure 2 shows the classification of LHS heat exchangers. Packed-bed, shell-and-tube, fluidized bed, and direct-contact type have been proposed. The packed-bed type and shell-and-tube type are classified as indirect-passive heat exchangers: heat exchange proceeds through a heat transfer wall between the PCM and the heat transfer fluid (HTF). A packed-bed heat exchanger using PCM capsules has advantages such as quick thermal response and high heat storage density, due to a large area of heat transfer. The advantages of the shell and tube type are that high-pressure steam can be used as HTF, only considering the high-pressure design of the heat transfer tubes. Fluidized bed designs are classified as indirect-active heat exchangers: heat is exchanged through a heat transfer wall between the PCM and the HTF. Here, not only the HTF but also the PCM is circulated in the heat storage system. The fluidized bed type can transport dense thermal energy from one process to another. Combination of the PCM with liquid (Yamagishi et al. 1999) or gas (Izquierdo-Barrientos et al. 2013) as the HTF was proposed. The other type is a direct-contact heat exchanger: the heat exchange proceeds between the PCM and the HTF without any heat transfer wall. Here, the PCM is required to be insoluble in the heat transfer medium, and the density of the PCM is required to be high enough to ensure that a phase separation is possible. The direct-contact heat exchanger has a very simple, light structure and a high thermal storage density. This is because only the inlet and outlet nozzles are required for the HTF, and a heat transfer pipe and capsule are not required. Therefore, this process can be carried out using not only a stationary system but also a heat transport system (Kaizawa et al. 2008a).

Fig. 2
figure 2

Classification of heat exchangers for LHS

4 Recent Advances of High-Temperature PCM

Figure 3 shows the relationship between T m and latent heat of common high-temperature PCM candidates. Among these high-temperature candidates are sugar alcohol, molten salt, and alloy. The T m of each material clearly involves a specific temperature range. The T m of sugar alcohols is under 200 °C, molten salt is mainly over 300 °C, and alloy is over 500 °C. Therefore, in this section, the recent advances in the use of LHS for high-temperature applications will be discussed for each type of PCM candidate: sugar alcohol, molten salt, and alloy.

Fig. 3
figure 3

Relationship between T m and latent heat of high-temperature PCM candidates

4.1 Sugar Alcohols (T m < 200 °C)

Sugar alcohols are promising PCM candidates for utilizing thermal energy at 100–200 °C. Table 1 lists the thermophysical properties of sugar alcohol PCMs (Hidaka et al. 2000; Kaizawa et al. 2008b). In this section, the advantages, problems, and prospects of sugar alcohol PCMs are presented.

Table 1 Thermophysical properties of sugar alcohol PCMs

Advantages

High heat storage density is a major advantage of sugar alcohols. As shown in Table 1, single component sugar alcohols have large latent heat (over 250 J·g−1). The latent heats of erythritol, mannitol, and galactitol are as large as the latent heat of the solid–liquid phase change of water. Other advantages of sugar alcohols are that they are nontoxic and inflammable. Typical applications of sugar alcohols involve foods and food additives. In addition, the sugar alcohols do not cause chemical corrosion of the metals used as the heat transfer wall.

Disadvantages

The low thermal conductivity of typical sugar alcohols is the most serious problem in using them as PCMs. For example, the thermal conductivity of the typical sugar alcohol PCM, erythritol (T m: 118 °C, latent heat: 314 J·g−1) (Kakiuchi et al. 1998) is 0.73 W·m−1·K−1 (Oya et al. 2013). Therefore, methods to enhance the effective thermal conductivity of the sugar alcohol PCMs have been widely studied (Liu et al. 2012). Another problem related to use of sugar alcohol PCMs, is that they have less thermal endurance. Sugar alcohols with T m over 150 °C have been reported, but their thermal endurance was too low to use for long periods (Nomura et al. 2015a).

Applications and Recent Studies

The LHS systems using sugar alcohol PCMs do have practical uses. Here, a latent heat transportation (LHT) system will be introduced. In this system, a mobile latent heat accumulator recovers industrial exhaust heat and distributes it over wide areas located at distances up to 35 km from the heat source. The industrial exhaust heat is compactly stored in the form of latent heat by melting the PCMs, which are then transported by truck to places such as office buildings, hospitals, and hotels, (Kaizawa et al. 2008a). Erythritol was used as PCM for the LHT system due to its large latent heat, and its suitability (T m 118 °C) for recovery of low-temperature industrial exhaust heat. In addition, this T m is high enough to operate a single effect absorption chiller that supplies cold water at about 5–7 °C. This heat transfer system has been applied commercially in Japan. A 20 t heat transfer container is used in the practical system, and each container can supply 1.3 MWh of thermal energy per cycle.

Future Work

There is a strong need to develop and utilize new sugar alcohol PCMs with T m around 150 °C. This is because a PCM with T m around 150 °C could produce low-pressure steam and operate double or triple effect absorption chillers, of which the coefficient of performance is over 1.0. Although sugar alcohol PCM candidates with T m around 150 °C, such as mannitol have been reported, and pilot-scale heat exchange experiments were performed (Gil et al. 2013), they have low thermal stability (Sagara et al. 2014). Therefore, the enhancement of the thermal stability of sugar alcohol PCMs is expected.

4.2 Molten Salt (T m > 300 °C)

Molten salts (e.g., nitrates, carbonates, and chlorides) are the most common high-temperature PCMs for use of thermal energy sources at over 300 °C. Table 2 lists the thermophysical properties of the molten salt PCMs (Guillot et al. 2012; Nomura et al. 2010; Japan Society of Thermophysical Properties 2008). In this section, the advantages, problems, and prospects for the uses of molten salt PCMs, are presented.

Table 2 Thermophysical properties of molten salt PCMs

Advantages

High heat storage density is a great advantage of molten salt PCMs. Generally, molten salts are low cost; therefore, they have also high heat storage capacity per unit cost. As shown in Table 2, many molten salts can function as PCMs, therefore, molten salt PCMs can be used in a variety of systems.

Disadvantages

The molten salt PCMs have three serious disadvantages. First is low thermal conductivity: the thermal conductivity of typical molten salt PCMs is under 1 W m−1 K−1. Second is their high volume expansion ratio during phase transition: the volume of expansion makes it difficult to design PCM capsules or heat storage units because the thermal stress from the volume expansion breaks the heat transfer wall and shell of the PCM capsules. For example, the volume expansion ratio of NaCl (T m: 800 °C), the most often used molten salt PCM, is 0.26 (Japan Society of Thermophysical Properties 2008). Third is their high level of chemical corrosion: liquid molten salts generally exhibit high levels of chemical corrosion of metallic shell materials.

Applications and Recent Studies

TES plays a vital role in concentrated solar power (CSP) plants because it is subjected to the intermittencies of the source and to the difficulties of matching electrical needs with the availability of solar insolation (Guillot et al. 2012). Although SHS using liquid molten salt has been applied in some presently operating commercial CSP plants, a great deal of salt is required, for examples 28,500 t of KNO3-NaNO3 for generating only 50 MW × 7.5 h/day (Rellso and Delgado 2009). Therefore, recent development of an LHS system using nitric acid as the PCM for CSP facilities has been accelerating. In particular, the shell and tube type heat exchanger, which can serve as a direct steam generator (DSG), is a promising system (Bayón et al. 2010) for this use.

Bayón et al. (2010) reported a DSG demonstration experiment using a 100 kWth plant using a PCM composite of graphite/eutectic mixtures of KNO3-NaNO3. Graphite was mixed in to enhance the effective thermal conductivity of the KNO3-NaNO3. Laing et al. (2011) proposed a hybrid heat exchange system using concrete SHS and sodium salt LHS. In addition, a packed bed heat exchanger using PCM capsules, which could easily expand the heat transfer area, was also reported. Mathur et al. (2013) developed new PCM capsules of about 5 mm diameter.

Future Work

The work on the LHS system using nitric acids (T m about 300 °C) is in validation phase (Bayón et al. 2010). On the other hand, innovations in CSP solar receivers, from troughs to central towers, have increased the concentrated temperature. Therefore, development of an LHS system using a molten salt PCM with T m up to 1000 °C is expected.

4.3 Metal Alloys (T m > 500 °C)

Using metal alloys for high-temperature PCMs has many advantages. Table 3 lists the thermophysical properties of alloys as PCM candidates (Kenisarin 2010; Maruoka and Akiyama 2006). In this section, the advantages, problems, and prospects of metal alloy PCMs are presented.

Table 3 Thermophysical properties of metallic PCMs

Advantages

The heat storage capacity of alloys is as high as that of the molten salt PCMs. In addition, the latent heat, based on volume of alloy, is larger than that of molten salt, since the density of alloys is generally higher than that of molten salt. Therefore, more compact LHS alloy systems can be designed than for systems using molten salt PCM. The high thermal conductivity of alloys is their most critical advantage. For example, the thermal conductivity of Al is 400 times greater than that of NaNO3. The low volume expansion ratio of alloys during phase transition is also a great advantage since it makes the design of LHS easier.

Disadvantages

Liquid metal alloys generally exhibit high chemical corrosion of metallic shell materials. This is the most serious problem regarding the use of metallic PCMs. To use them, ceramics must be applied over the structural materials used to build the heat storage units and shells of capsules.

Recent Studies

The use of metallic PCM for LHS is promising for effective use of high-temperature thermal energy, but development of this technology has just started. Thus, no large-scale experiments have been conducted, as far as is known. The development of materials and heat exchanger designs has been reported. The development of high-temperature LHS technologies based on Al-Si series PCMs, in particular, is accelerating. Nomura et al. (2014) have proposed a high-temperature PCM base on an Al-Si series alloy (T m over 500 °C). The heat storage capacity of this Al-Si series PCM was about 1.5-times higher than that of MgCl2 (a typical molten salt PCM). Kotze et al. (2014) proposed a shell-and-tube-type heat exchanger, in which Al-12wt%Si was used as a PCM, and NaK was used as the HTF. In addition, a super-high-temperature PCM (above 1000 °C) has also been reported. Maruoka and Akiyama (2006) proposed a core-shell-type PCM macrocapsule using a Cu core and Ni shell, achieved using an electroplating method. This PCM capsule has hybrid functions: thermal energy storage and catalyst. This encapsulation method is being refined (Zhang et al. 2014).

5 Discussion

In this section, the prospects for future LHS technology are discussed from the viewpoint of exergy.

Figure 4 shows the relationship between energy level (indicating energy quality), and exergy (latent heat) of the high-temperature PCM candidates. Here, the energy levels of PCMs (at T m) and their exergy are defined as follows:

$$ A=\frac{\left({T}_{\mathrm{m}}-{T}_0\right)}{T_{\mathrm{m}}} $$
(3)

and

$$ {\varepsilon}_{\mathrm{L}}= AL $$
(4)
Fig. 4
figure 4

Relationship between energy level and exergy of latent heat of these materials

Here, A is energy level, ε L is exergy of latent heat, and T 0 is environmental temperature (298.15 K).

By considering the relationship between A and ε L, the PCMs that should be focused upon can be determined. The line in Fig. 4 was a least square fitting of the plots. Interestingly, the PCMs that are used commercially, or near practical use (e.g., CH3COONa∙3H2O, threitol, erythritol, mannitol, and molten salt of NaOH-series) were almost on the line. Therefore, the PCMs on the line could be classified as good quality PCMs (adequate for practical use). The PCMs under the line were lower quality PCMs, and therefore, should not be selected, except for special reasons. From this viewpoint, NaNO3 or NaNO3-KNO3 should not be selected as PCM. However, they have been used in many studies (Acem et al. 2010; Pacheco et al. 2002; Pincemin et al. 2008) about LHS systems. This is because only the NaNO3 does not corrode structural materials, among the molten salt PCMs, for the temperature range of about 300 °C. The PCM candidates above the line, including the Al-Si series and NaCl, were classified as high-quality PCMs. Although there are few studies about these high-quality PCMs, the future development of LHS, with respect to exergy, should be focused upon them.

Focusing on the application of LHS, conversion of PCM thermal energy to chemical energy (e.g., hydrogen) using endothermic reactions has advantages since thermal exergy at low energy levels can be recuperated to chemical exergy at high energy levels (Tsutumi and Yoshida 1999). Here, the exergy recuperation (Tsutsumi 2004) concept will be introduced using the method called ‘thermodynamic compass’ (Ishida 2002). This is a graphic method developed by M. Ishida (2002) that can express every process and synthesis as vectors in two-dimensional space with X-axis of enthalpy change ΔH, and Y-axis of exergy change Δε. In evaluating a system thermodynamically, summations of ΔH and Δε in each process are restricted by the first and second laws of thermodynamics, as expressed in the following equations:

First law of thermodynamics:

$$ \sum \Delta H=0 $$
(5)

Second law of thermodynamics:

$$ \sum \Delta \varepsilon =0 $$
(6)

For each vector in a process, the ratio of Δε to ΔH gives the slope, called the energy level, A, indicating energy quality.

A vector and energy level for a chemical reaction can be expressed by the following equation:

$$ \overrightarrow{R}=\left(\Delta H,\Delta \varepsilon \right)=\left(\Delta {h}_{T_{\mathrm{o}}}^0,\Delta {\varepsilon}_{T_{\mathrm{o}}}^0\right) $$
(7)
$$ A=\frac{\Delta {\varepsilon}_{T_{\mathrm{o}}}^0}{\Delta {h}_{T_{\mathrm{o}}}^0} $$
(8)

Here, \( \overrightarrow{R} \), \( \Delta {h}_{T_{\mathrm{o}}}^0 \), and \( \Delta {\varepsilon}_{T_{\mathrm{o}}}^0 \) mean vector, enthalpy change, and exergy change of a chemical reaction at standard temperature T 0 (298.15 K) and standard pressure (1 atm). A vector and energy level for a heat source at T, with enthalpy of Q, can be expressed by the following equation:

$$ \overrightarrow{Q}=\left(\Delta H,\Delta \varepsilon \right)=\left(Q, AQ\right) $$
(9)
$$ A=\frac{T-{T}_0}{T} $$
(10)

Here, \( \overrightarrow{Q} \) is a vector of the heat source.

To promote the efficient utilization of energy, the system should have a shorter resultant vector.

Figure 5 shows synthesis of a vector of dry reforming of CH4 to produce 2 mol of syngas (CO:H2 = 1:1 mol) and a basic heat source on the thermodynamics compass. The reaction equation, vector, and energy level of dry reforming of CH4, which is an endothermic reaction, is expressed as

Fig. 5
figure 5

Synthesis of a vector for dry reforming of CH4 to produce 2 mol of syngas and a basic heat source

$$ \overrightarrow{R_{\mathrm{reforming}}} $$
$$ {\mathrm{CH}}_4+{\mathrm{CO}}_2\to 2{\mathrm{H}}_2+2\mathrm{CO} $$
(11)
$$ \left(\Delta H,\left.\Delta \varepsilon \right)\right.=\left(0.247,\left.0.171\right)\right. $$
(12)
$$ A=0.69 $$
(13)

Here, \( {\overrightarrow{R}}_{\mathrm{reforming}} \) is a vector for dry reforming of CH4.

Due to thermodynamic restrictions, a heat source at 690 °C must be provided in a support process.

$$ \overrightarrow{Q_{\mathrm{reforming}}} $$
$$ \left(\Delta H,\left.\Delta \varepsilon \right)\right.=\left(-0.247,\left.-0.171\right)\right. $$
(14)
$$ A=0.69,\kern0.5em T={690}^{\circ}\mathrm{C} $$
(15)

Here, \( {\overrightarrow{Q}}_{\mathrm{reforming}} \) is a vector of the heat source for dry reforming. As results, the following relationship was established:

$$ \overrightarrow{Q_{\mathrm{reforming}}}=\overrightarrow{-{R}_{\mathrm{reforming}}} $$
(16)

As discussed before, the dry reforming of CH4 produces 2 mol of syngas by the combination of 1 mol of CH4 and heat source at 690 °C. The chemical energy of CH4 and syngas can be expressed on the thermodynamic compass as a perfect combustion reaction of each material:

$$ \overrightarrow{R_{\mathrm{CH}4}} $$
$$ {\mathrm{CH}}_4+2{\mathrm{O}}_2\to 2{\mathrm{H}}_2\mathrm{O}+{\mathrm{CO}}_2 $$
(17)
$$ \left(\Delta H,\left.\Delta \varepsilon \right)\right.=\left(-0.891,\left.-0.818\right)\right. $$
(18)
$$ A=0.92 $$
(19)
$$ \overrightarrow{R_{\mathrm{syngas}}} $$
$$ 2{\mathrm{H}}_2+2\mathrm{CO}+2{\mathrm{O}}_2\to 2{\mathrm{H}}_2O+2{\mathrm{CO}}_2 $$
(20)
$$ \left(\Delta H,\left.\Delta \varepsilon \right)\right.=\left(-1.137,\left.-0.989\right)\right. $$
(21)
$$ A=0.86 $$
(22)

Here, \( \overrightarrow{R_{\mathrm{CH}4}} \) and \( \overrightarrow{R_{\mathrm{syngas}}} \) indicate vectors of the perfect combustion reaction of CH4 and syngas. Figure 6 expresses the relationship of \( \overrightarrow{R_{\mathrm{CH}4}} \), \( \overrightarrow{R_{\mathrm{syngas}}} \), \( \overrightarrow{R_{\mathrm{reforming}}} \), and \( \overrightarrow{Q_{\mathrm{reforming}}} \) on the thermodynamic compass. Here, the following equation is established:

Fig. 6
figure 6

Relationship of \( \overrightarrow{R_{\mathrm{CH}4}} \), \( \overrightarrow{R_{\mathrm{syngas}}} \), \( \overrightarrow{R_{\mathrm{reforming}}} \), and \( \overrightarrow{Q_{\mathrm{reforming}}} \)

$$ \overrightarrow{R_{\mathrm{CH}4}}-\overrightarrow{R_{reforming}}=\overrightarrow{R_{\mathrm{syngas}}} $$
(23)

From Eq. (16) and Eq. (23),

$$ \overrightarrow{R_{\mathrm{CH}4}}+\overrightarrow{Q_{reforming}}=\overrightarrow{R_{\mathrm{syngas}}} $$
(24)

As discussed earlier, dry reforming means the conversion of the chemical energy of CH4 and thermal energy of a heat source to chemical energy of syngas. As shown in Fig. 6, the energy level of the heat source (0.69) is upgraded to that of the syngas (0.86). In other words, the exergy of thermal energy is recuperated in the form of syngas chemical energy. This exergy recuperation method can achieve a high degree of energy recovery.

Next, the application of PCM to the dry-reforming heat source will be considered. Figure 7 shows the thermodynamics compass of dry reforming as a target process, and the PCM heat source as a support process. PCM is a heat source at a constant temperature (T m); therefore, the energy level can be expressed by Eq. 3. In this study, two PCMs were assumed. PCM1 had T m at 800 °C and PCM2 had T m at 1400 °C. The vectors and energy levels can be expressed as follows:

$$ \overrightarrow{Q_{\mathrm{PCM}1}} $$
$$ \left(\Delta H,\left.\Delta \varepsilon \right)\right.=\left(-0.247,\left.-0.178\right)\right. $$
(25)
$$ {T}_{\mathrm{m}}={800}^{\circ}\mathrm{C},A=0.72 $$
(26)
$$ \overrightarrow{Q_{\mathrm{PCM}2}} $$
$$ \left(\Delta H,\left.\Delta \varepsilon \right)\right.=\left(-0.247,\left.-0.203\right)\right. $$
(27)
Fig. 7
figure 7

Thermodynamics compass of dry reforming as a target process and heat source using PCM as a support process

$$ {T}_{\mathrm{m}}={1400}^{\circ}\mathrm{C},\kern0.5em A=0.82 $$
(28)

As shown in Fig. 7, the exergy loss of the system can be reduced by the energy level of PCM being closer to that of dry reforming. This means that the PCM selected is quite important for the efficient use of exergy. In other words, to decrease exergy loss, the T m of the PCM should be as close as possible to the equilibrium temperature of the reaction, and two solutions should be considered to minimize the exergy loss. First is improvement of the heat transfer characteristic of the PCM. This will be lower than the temperature difference between PCM and reaction temperature, as the driving force of heat transfer. Because a marked increase in the heat transfer area is needed, a conventional shell-and-tube, or packed-bed type could not satisfy the need. Recently, microencapsulated PCM with diameter of about 40 μm, using Al-25 wt%-Si as PCM, was reported by Nomura et al. (2015b). The microencapsulated PCM has the potential to increase the heat transfer area remarkably. Second, is the development of catalytic PCM capsule. The reaction rate can be improved and the reaction temperature made close to the equilibrium temperature by using catalytic PCM. Catalytic PCM capsules with an Ni shell were reported by Maruoka and Akiyama (2006); however, there are few studies about these materials.

In summary, the development of catalytic microencapsulated PCM will be important for the development of future LHS technology.

6 Conclusions

This article described the recent advances in the use of LHS for high-temperature (over 100 °C) applications, with respect to each type of PCM (sugar alcohols, molten salt, and metal alloys). Table 4 gives a summary of the status of LHS development. The establishment of a high-temperature LHS technology is expected, because the temperatures of heat utilization systems are expected to increase.

Table 4 Status of LHS development