Developing Composite Phase Change Material with Al–Si Base Microencapsulated Phase Change Material and Glass Frit for High Temperature Applications

Abstract

To achieve high energy efficiency and CO₂ reduction during iron- and steelmaking, thermal management is vital. Use of phase change material (PCMs) to store excess energy in the form of latent heat has the potential to realize excellent thermal management. Microencapsulated PCMs (MEPCMs) consisting of an alloy PCM core and an oxide coating have improved corrosion resistance and are easy to mix with other materials. Conventionally, composite PCM pellets are fabricated by mixing glass frit (to aid sintering) with Al–25 mass% Si MEPCM. However, this process has not yet been optimized. In this study, the optimal stoichiometry of composite PCMs prepared using Al–25 mass% Si MEPCM and glass frit was investigated. The pellets were prepared by mixing with glass frit at 60, 80 and 90 mass% of MEPCM, followed by molding and heat treatment. As a result, pellets were successfully fabricated with condition including 60 and 80 mass% of MEPCM. The latent heat capacity of the composite PCM was 146 J g^–1, which was at least 1.59 times higher than that of existing sensible heat storage (SHS) materials. Moreover, the composite PCMs withstood 300 melting and solidification cycles. In summary, composite PCMs with excellent latent heat capacity and durability were successfully prepared.

1. Introduction

Thermal management of iron- and steelmaking is important for improving the energy efficiency and reducing CO₂ emission of the process. Currently, production of iron and steel requires large amount of energy, which generates huge amount of CO₂. International Energy Agency (IEA) reported that iron and steel production contributes to 8% of global energy demand and 7% of total CO₂ emissions in the world.¹⁾ However, effective use of the energy released during the iron- and steelmaking processes is insufficient. During their production, part of the energy is released outside the system as exhaust heat. Therefore, effective utilization of the thermal energy could help improve efficiency and decrease CO₂ emissions. An example of thermal management is to use the heat from the gas generated in coke ovens to reheat the coke ovens. Studies have also been performed to recover the slag-derived heat from blast furnaces using dry granulation and heat exchangers.²⁾ Thermal management is also used to maintain the proper temperature for chemical reactions, such as methane reforming.³⁾ The introduction of thermal energy storage (TES) technology that can mitigate fluctuations in energy generation from heat sources is essential for the design of these thermal management systems. At present, sensible heat storage (SHS) is the mainstream TES technology, which is based on the specific heat of materials.^4,5) However, because the heat capacity of SHS is low, further development of TES technology with a high heat capacity, which can additionally reduce the size of the TES tank, is needed.

As the next development in TES technology, latent heat storage (LHS), which uses the melting or solidification of phase change material (PCMs), has promising applications in steelmaking because of its ability to store much higher energy.⁶⁾ Additional advantages of LHS include energy storage at a constant temperature, great thermal stability, and excellent reproducibility of the heat storage and release processes.⁷⁾ Molten salts or metal/alloys have been proposed as PCMs for LHS, which are proper when the temperature is above 300°C, are also relevant in iron- and steelmaking.⁸⁾ In particular, because metals/alloys have excellent thermal conductivity and fast heat exchange can be expected if they are introduced into TES systems as PCMs.⁹⁾

Many studies have investigated metals/alloys that have properties suitable for use as PCMs. Birchenall et al. measured the melting temperatures and latent heat capacities of binary and ternary eutectic alloys of Al, Cu, Mg, Si, and Zn using DSC or DTA.¹⁰⁾ In another study, Wang et al. investigated the latent heat capacities of Al–Si-based PCMs.¹¹⁾ The thermophysical properties of different alloy-based PCMs were also investigated by other researchers.^9,12) In addition to the thermophysical properties of PCM, it is important to investigate the corrosion resistance of the PCM to design a suitable material for the TES tank. Fukahori et al. investigated the corrosiveness of Al– and Al–Si-based PCMs in contact with ceramics and showed that Al₂O₃, AlN, and Si₃N₄ have high corrosion resistance.¹³⁾ Dindi et al. investigated the corrosive behavior of between Al–Si-based PCM and BN coated stainless steel, and reported that no reaction occurred between them even after 720 melting and solidification cycles.¹⁴⁾

Furthermore, encapsulation of PCMs is being considered to improve the corrosion resistance. Zhang et al. coated the surface of millimeter-sized Cu PCMs with Cr–Ni, which showed no leakage of Cu after more than 1000 melting and solidification cycles.¹⁵⁾ Fukahori et al. fabricated centimeter-sized capsules of Al–25 mass% Si sealed with Al₂O₃ and found no leakage of the alloy or failure of the capsules after 100 melting and solidification cycles.¹⁶⁾ Encapsulation of metal/alloy PCMs helps also improve their handling. Further, miniaturization of the capsule size is expected to improve properties such as heat transfer area.

Previously, we had reported on the development of a microencapsulated PCM (MEPCM) with a core of Al–25 mass% Si microparticles and the surface covered with stable oxides such as Al₂O₃.^17,18,19,20) MEPCM was fabricated by the boehmite treatment of alloy microparticles in boiling water to form a metal hydroxide coating, followed by oxidation at high temperature to obtain an oxide coating. In other studies, Al–12 mass% Si,^21,22) Al,^23,24,25) and Zn–30 mass% Al²⁶⁾ were considered as core materials for MEPCMs.

An advantage of microencapsulation of PCMs is that, because the surface is an oxide such as Al₂O₃, it can be handled as ceramic particles and thus can be combined with other materials. Takahashi et al. reported the use of a Ni catalyst on the surface of Al–25 mass% Si MEPCM and that the temperature increase due to heat generation during methanation reaction was suppressed by MEPCM.²⁷⁾ This result shows the feasibility of microscale thermal control of catalytic reactions with the use of MEPCM. Another advantage of MEPCMs is that they can be molded into different shapes according to the type of heat exchanger. Sakai et al. prepared composite PCM with pellet structures by sintering a mixture of Al–25 mass% Si (melting point: 577°C) MEPCM and alumina or glass frit as a sintering aid.^28,29) The composite PCM had a latent heat capacity of 122 J g⁻¹.²⁸⁾ However, scope for improvement in the performance of composite PCMs by further optimizing the fabrication conditions still exists.

In this study, the fabrication conditions of a composite PCM with Al–25 mass% Si MEPCM were investigated. In particular, the proper mixing ratio of MEPCM and glass frit was explored. The composition, thermophysical properties, and cyclic durability of the fabricated composite PCMs were evaluated.

2. Experimental

2.1. Materials

Al–25 mass% Si MEPCM (mode diameter: 42.6 μm, apparent density: 2.89 g cm⁻³, latent heat: 192.56 J g⁻¹, melting point: 576°C) was used as the raw material to fabricate the composite PCM. Details of the preparation method have been reported elsewhere.^19,20) Amorphous glass frit (mode diameter: 30.0 μm, apparent density: 2.60 g cm⁻³, thermal expansion coefficient: 45 × 10⁻⁷°C⁻¹, glass transition temperature: 740°C, softening point: 945°C, AGC Inc., HHR0704) was used as the sintering agent. The glass frit contained SiO₂, B₂O₃, ZnO, BaO, MgO, CaO, Al₂O₃, and Na₂O. Figure 1 shows the SEM images of the Al–25 mass% Si MEPCM and glass frit.

Fig. 1.

SEM images of A) Al–25 mass% Si MEPCM and B) glass frit.

2.2. Preparation of the Composite PCM

A powder consisting of the MEPCM and glass frit was prepared by hand mixing. The powder contained 60 mass% (57.4 vol%), 80 mass% (78.3 vol%), and 90 mass% (89.0 vol%) of the MEPCM. The preparation process of the composite PCMs was based on previous studies.²⁸⁾ A small amount of the powder (0.5 g) was then added to a die with a diameter of 1 cm and pressed for 1 min at 20 MPa and 25°C. Cylindrical pellets with a height of approximately 4 mm obtained after pressing were heated from 25°C to 950°C at a rate of 5°C min⁻¹ after which they were kept for a duration at the same temperature for 1 h in air. Hereafter the pellets will be denoted as ME60, ME80, and ME90, which is based on the MEPCM content.

2.3. Characterization

The diameters and heights of the composite PCMs were measured, and their volumes, including the porosity, were calculated to obtain the bulk density. The apparent density was measured using an ultrapycnometer (Quantachrome Instruments, Ultrapycnometer 1000). The phases present in the composite PCMs were identified by X-ray diffraction (XRD; Rigaku Miniflex600, Cu Kα). The composite PCMs were embedded in a resin and polished with emery paper. The polished cross-section was observed by scanning electron microscopy (SEM; JEOL, JSM-7001FA), and elemental analysis was performed using energy dispersive spectroscopy (EDS). The latent heat capacities of the composite PCMs were measured using thermogravimetric analysis and differential scanning calorimetry (TGA-DSC; Mettler Toledo, TGA-DSC-3). The powder obtained by grinding the composite PCM in a mortar and pestle was heated and cooled at a rate of 5°C min⁻¹ in Ar. Thermal expansion for the composite PCM was analyzed by a thermomechanical analyzer (TMA; TMA7300, Hitachi) from 20°C to 500°C with a heating rate of 10°C min⁻¹.

2.4. Thermal Cycle Tests

The as-prepared composite PCMs were subjected to 300 melting and solidification cycles to investigate their thermal cyclic durability. The tests were conducted in a tube furnace with two mobile heaters that could be automatically moved along the tube via an air compressor. The two heaters were set at 650°C and 450°C. The composite PCMs were placed in the air-filled tube, and the heaters were sequentially moved to alternately melt and solidify the Al–25 mass% Si alloy PCM in the composite PCM. The melting and solidification times were six and nine minutes, respectively.

3. Results

3.1. Morphology and Composition

Figure 2 shows the ME60 and ME80 before and after the heat treatment. Herein it can be observed that even though the pellets became darker after the heat treatment, their shapes remained the same. However, only ME90 is just grabbing it with tweezers causing the particles comprising the PCM composite to fall. ME90 has the lowest amount of sintering aid present, and sufficient sintering could not be achieved. From the above, the results of ME60 and ME80, which has enough strength to be used as composite PCM, are described in this paper. Figure 3 shows the apparent, bulk, and theoretical densities of ME60 and ME80. The theoretical density was calculated based on the apparent density and mixing ratio of each particle constituent of the composite PCMs. The apparent densities of both ME60 and ME80 were 99% of their respective theoretical densities. Moreover, the bulk densities of the ME60 and ME80 pellets were 56% and 53% of their theoretical density, respectively.

Fig. 2.

Overview of ME60, ME80, and ME90 before and after heat treatment. (Online version in color.)

Fig. 3.

Apparent, bulk, and theoretical densities of ME60 and ME80. (Online version in color.)

Figure 4 shows the XRD spectra of ME60, ME80, and the glass frit pellet. The preparation condition for the glass-frit pellet was the same as that of the composite PCMs. Sharp peaks corresponding to Al, Si, and α-Al₂O₃ were detected in both the composite PCMs. In contrast, a broad peak between 20° and 30° (and no sharp peak) was observed in the XRD profile of the glass frit. Thus, the peaks of Al, Si and α-Al₂O₃ correspond only to the alloy or shell of the MEPCM in the composite PCM because glass frit is not detected in the XRD spectra.

Fig. 4.

XRD spectra of ME60, ME80, and glass frit. (Online version in color.)

Figure 5 shows SEM images and EDS mappings of the polished cross sections of ME60 and ME80. Two distinct gray regions can be observed in the SEM images. Moreover, it can be observed that the lighter gray region is composed of two kinds of phases—one circular and the other irregularly shaped. The circular phase is approximately 40 μm in diameter, and the irregularly shaped phase is several tens to hundreds of micrometers in size. In addition, pores with size less than 10 μm can be observed within the circular phases. From the EDS maps, Al and Si were detected in clearly separate regions within the circular phases. Meanwhile, O was detected only in the irregularly shaped phase. Based on the shape, size, elemental content, and the XRD spectrum (shown in Fig. 4) of each phase, it was inferred that the circular and irregularly shaped phases were MEPCM and glass frit, respectively. In contrast, C was detected mostly in the darker gray region (Figure S1, Supporting Information), which is the resin used for polishing. Detection of the resin in the SEM images suggests the presence of open pores in the composite PCMs.

Fig. 5.

SEM images and EDS maps of the cross sections of ME60 and ME80. (Online version in color.)

3.2. Thermal Cyclic Durability and Thermal Property

Figure 6 shows the optical images of the composite PCMs after the thermal cycle. Compared to that before the thermal cycle (shown in Fig. 2), shapes of the composite PCMs did not change after the thermal cycle. Figure 7 shows the DSC curves of the composite PCMs before and after the thermal cycle. Both the composite PCMs exhibited an endothermic peak at approximately 575°C during heating and an exothermic peak at approximately 550°C during cooling. Because the melting temperature of the Al–25 mass% Si alloy PCM (which is the core material of the MEPCM) is 577°C, the endothermic and exothermic peaks are attributed to the melting and solidification of the Al–25 mass% Si alloy PCM, respectively. The transition of the exothermic peak to a lower temperature during cooling is due to the undercooling necessary for solidification. The latent heat capacities of the composite PCMs calculated from the area of the endothermic peak prior to the thermal cycle were −109.26 J g⁻¹ for ME60 and −146.00 J g⁻¹ for ME80. The latent heat capacity of ME80 was higher than that of ME60 because of the higher MEPCM content in the former. Furthermore, the latent heat capacity of ME80 is 1.2 times higher than that obtained in a previous study of the composite PCM.²⁸⁾ The latent heat capacities of the pellets calculated from the endothermic peaks of the DSC curves after the thermal cycle were −112.59 J g^–1 for ME60 and −142.54 J g^–1 for ME80. These values were 103% and 98% of those of the composite PCMs before the thermal cycle, respectively. Because the measurement error of the latent heat capacity from the TG-DSC was ±5%, the above deviations were well within the error range. Thus, the composite PCMs exhibited high latent heat capacity and excellent durability for 300 thermal cycles.

Fig. 6.

Optical images of A) ME60 and B) ME80 after 300 thermal cycles. (Online version in color.)

Fig. 7.

DSC curves of the composite PCMs A) before and B) after the thermal cycle.

Figure 8 shows the thermal expansion (ΔL/L) of ME60 and ME80. Almost linear positive expansion was observed for each composite PCM. In addition, thermal expansion was pronounced in ME80. The coefficient of thermal expansion of Al, which is the main component of MEPCM, is 237 × 10⁻⁷°C⁻¹.³⁰⁾ This value is higher than that of glass frit, resulting in larger thermal expansion with ME80, which contains more MEPCM. Table 1 shows the coefficient of thermal expansion of ME60, ME80, rock, and concrete.^31,32) In the temperature range from 100°C to 500°C, ME60 and ME80 have similar or lower coefficients of thermal expansion than rocks and concrete used as the SHS materials Therefore, the composite PCM is relatively stable to rapid heating and cooling and can be applied to a wide range of heat utilization.

Fig. 8.

Thermal expansion (ΔL/L) of ME60 and ME80.

Table 1. Coefficient of thermal expansion of each material. The coefficient of thermal expansion for ME60 and ME80 were calculated in 50°C increments.

Material	ME60	ME80	Rock31)	Concrete32)
Coefficient of thermal expansion (10−7°C−1)	70–80 (at 100–500°C)	77–98 (at 100–500°C)	100–500 (at 100–500°C)	116 (at 350°C)

4. Discussions

4.1. Porosity of the Composite PCMs

As previously mentioned, the respective bulk and apparent densities of the composite PCMs were lower than and similar to their theoretical density, respectively. This indicates that even though the number of closed pores was negligible, open pores were present in the composite PCMs. Although the presence of open pores decreases the amount of heat per volume that could be stored in the composite PCM, its effective surface area increases. Therefore, it is expected to provide other advantages, such as achieving efficient catalytic reactions and thermal control by supporting a catalyst on the surface of the composite PCM. However, since a reduction of pores in the PCM increases the amount of heat per volume that can be stored, this should be considered as well.

4.2. Factors Involved in the Thermal Cyclic Durability of the Composite PCMs

As shown earlier, the composite PCMs maintained their shapes and latent heat capacities even after 300 thermal cycles. One of the factors that determine the thermal cyclic durability of the composite PCMs is the durability of the constituent MEPCMs. Referring back to the SEM images shown in Fig. 5, pores were also observed inside the MEPCM. During the heat treatment for making the MEPCM, the Al–25 mass% Si alloy PCM was micro-encapsulated at its maximum volume due to melting and thermal expansion. This resulted in the formation of pores inside the MEPCM when unmelted PCM. The pores act as a buffer for volumetric expansion during the solid–liquid transformation of the alloy, leading to improved durability of the capsule.¹⁸⁾ Another factor that contributed to the improved durability of the composite PCM is that the glass frit may have acted as an additional shell for the MEPCM. This needs to be investigated in detail in the future.

4.3. Comparison of Heat Storage Capacity

A comparison of heat capacities of the prepared composite PCMs with conventional SHS materials is shown in Fig. 9.^33,34) Herein, the term heat capacity implies latent heat capacity for the composite PCMs, whereas, for rock and concrete, it implies their sensible heat capacities at ΔT = 100 K. It can be observed in Fig. 8 that the heat capacities of ME80 were 1.47 and 1.59 times of that of rock and concrete, respectively. The heat capacities of ME60 were 1.10 or 1.19 times of that of SHS materials. Moreover, as the heat stored by ME80 and ME60 is in the form of latent heat, their temperature remains constant during heat storage. This implies that, compared to the SHS materials based on specific heat, the composite PCMs can store heat at high temperature intact from heat source. Thus, the composite PCMs can be used to design highly performing TES systems that can aid in the thermal management of iron- and steelmaking processes.

Fig. 9.

Comparison of the heat capacities of ME60, ME80, and conventional SHS materials.

5. Conclusions

In this study, composite PCMs were fabricated from Al–25 mass% Si MEPCM and glass frit, and their preparation conditions were investigated. As a result, composite PCMs were successfully prepared in conditions containing 60 or 80 mass% MEPCM. In addition, the composite PCM with 80 mass% MEPCM had the highest heat capacity of 146 J g⁻¹. This value is more than 1.59 times higher than that of existing SHS materials (at ΔT = 100 K) and is also higher than that of the composite PCMs reported in earlier studies. In addition, the composite PCMs retained its shape and latent heat capacity even after 300 thermal cycles. Owing to its superior thermal properties and cyclic durability, the prepared composite PCM can contribute to further improving the performance of TES systems. To expand the range of applications of the prepared composite PCM, its performance needs to be improved by further optimizing the manufacturing process.

Supporting Information

EDS map of a cross section of the PCM showing C present only in the resin.

This material is available on the Journal website at https://doi.org/10.2355/isijinternational.ISIJINT-2022-109.

Acknowledgement

The authors greatly appreciate the financial support from the New Energy and Industrial Technology Development Organization (NEDO). A part of this work was conducted at Hokkaido University, supported by the “Nanotechnology Platform” Program of the Ministry of Education, Culture, Sports, Science and Technology (MEXT), Japan.

References

• 1)  IEA: Iron and Steel Technology Roadmap Towards More Sustainable Steelmaking, IEA, Paris, (2020), 3. https://aceroplatea.es/docs/Iron_and_Steel_Technology_Roadmap_IEA.pdf, (accessed 2022-01-12).
• 2)   A.  Carpenter: IEA Clean Coal Centre, Report No. CCC/193, United States Energy Association, Washington, D. C., (2012), 46. https://usea.org/publication/co2-abatement-iron-and-steel-industry-ccc193, (accessed 2022-01-12).
• 3)   W.  Uribe-Soto,  J.-F.  Portha,  J.-M.  Commenge and  L.  Falk: Renew. Sustain. Energy Rev., 74 (2017), 809. https://doi.org/10.1016/j.rser.2017.03.008
• 4)   A.  Dinker,  M.  Agarwal and  G.  Agarwal: J. Energy Inst., 90 (2017), 1. https://doi.org/10.1016/j.joei.2015.10.002
• 5)   R.  Tiskatine,  R.  Oaddi,  R. A.  El Cadi,  A.  Bazgaou,  L.  Bouirden,  A.  Aharoune and  A.  Ihlal: Sol. Energy Mater. Sol. Cells, 169 (2017), 245. https://doi.org/10.1016/j.solmat.2017.05.033
• 6)   V.  Dwivedi,  P.  Tiwari and  S.  Tiwari: 2016 Int. Conf. on Emerging Trends in Electrical Electronics and Sustainable Energy Systems (ICETEESES), IEEE, New York, (2016), 42. https://doi.org/10.1109/ICETEESES.2016.7581349
• 7)   M. K.  Rathod and  J.  Banerjee: Renew. Sustain. Energy Rev., 18 (2013), 246. https://doi.org/10.1016/j.rser.2012.10.022
• 8)   M.  Liu,  E. S.  Omaraa,  J.  Qi,  P.  Haseli,  J.  Ibrahim,  D.  Sergeev,  M.  Müller,  F.  Bruno and  P.  Majewski: Renew. Sustain. Energy Rev., 150 (2021), 111528. https://doi.org/10.1016/j.rser.2021.111528
• 9)   A. I.  Fernández,  C.  Barreneche,  M.  Belusko,  M.  Segarra,  F.  Bruno and  L. F.  Cabeza: Sol. Energy Mater. Sol. Cells, 171 (2017), 275. https://doi.org/10.1016/j.solmat.2017.06.054
• 10)   C. E.  Birchenall and  A. F.  Riechman: Metall. Trans. A, 11 (1980), 1415. https://doi.org/10.1007/BF02653497
• 11)   X.  Wang,  J.  Liu,  Y.  Zhang,  H.  Di and  Y.  Jiang: Energy Convers. Manag., 47 (2006), 2211. https://doi.org/10.1016/j.enconman.2005.12.004
• 12)   P. J.  Shamberger and  N. M.  Bruno: Appl. Energy, 258 (2020), 113955. https://doi.org/10.1016/j.apenergy.2019.113955
• 13)   R.  Fukahori,  T.  Nomura,  C.  Zhu,  N.  Sheng,  N.  Okinaka and  T.  Akiyama: Appl. Energy, 163 (2016), 1. https://doi.org/10.1016/j.apenergy.2015.10.164
• 14)   A.  Dindi,  N.  Lopez Ferber,  D.  Gloss,  E.  Rilby and  N.  Calvet: J. Energy Storage, 32 (2020), 101961. https://doi.org/10.1016/j.est.2020.101961
• 15)   G.  Zhang,  J.  Li,  Y.  Chen,  H.  Xiang,  B.  Ma,  Z.  Xu and  X.  Ma: Sol. Energy Mater. Sol. Cells, 128 (2014), 131. https://doi.org/10.1016/j.solmat.2014.05.012
• 16)   R.  Fukahori,  T.  Nomura,  C.  Zhu,  N.  Sheng,  N.  Okinaka and  T.  Akiyama: Appl. Energy, 170 (2016), 324. https://doi.org/10.1016/j.apenergy.2016.02.106
• 17)   T.  Nomura,  C.  Zhu,  N.  Sheng,  G.  Saito and  T.  Akiyama: Sci. Rep., 5 (2015), 9117. https://doi.org/10.1038/srep09117
• 18)   T.  Nomura,  N.  Sheng,  C.  Zhu,  G.  Saito,  D.  Hanzaki,  T.  Hiraki and  T.  Akiyama: Appl. Energy, 188 (2017), 9. https://doi.org/10.1016/j.apenergy.2016.11.025
• 19)   N.  Sheng,  C.  Zhu,  G.  Saito,  T.  Hiraki,  M.  Haka,  Y.  Hasegawa,  H.  Sakai,  T.  Akiyama and  T.  Nomura: J. Mater. Chem. A, 6 (2018), 18143. https://doi.org/10.1039/C8TA04708A
• 20)   N.  Sheng,  C.  Zhu,  H.  Sakai,  Y.  Hasegawa,  T.  Akiyama and  T.  Nomura: Sol. Energy Mater. Sol. Cells, 200 (2019), 109925. https://doi.org/10.1016/j.solmat.2019.109925
• 21)   T.  Nomura,  J.  Yoolerd,  N.  Sheng,  H.  Sakai,  Y.  Hasegawa,  M.  Haga,  G.  Saito and  T.  Akiyama: Sol. Energy Mater. Sol. Cells, 187 (2018), 255. https://doi.org/10.1016/j.solmat.2018.08.001
• 22)   C.  Han,  H.  Gu,  M.  Zhang,  A.  Huang,  Y.  Zhang and  Y.  Wang: Sol. Energy Mater. Sol. Cells, 217 (2020), 110697. https://doi.org/10.1016/j.solmat.2020.110697
• 23)   T.  Nomura,  J.  Yoolerd,  N.  Sheng,  H.  Sakai,  Y.  Hasegawa,  M.  Haga and  T.  Akiyama: Sol. Energy Mater. Sol. Cells, 193 (2019), 281. https://doi.org/10.1016/j.solmat.2018.12.023
• 24)   T.  Kawaguchi,  J.  Yoolerd,  H.  Sakai,  Y.  Shimizu,  A.  Kurniawan and  T.  Nomura: Sol. Energy Mater. Sol. Cells, 237 (2022), 111540. https://doi.org/10.1016/j.solmat.2021.111540
• 25)   K.  Li,  Z.  Gu,  X.  Zhu,  Y.  Wei and  H.  Wang: ACS Sustain. Chem. Eng., 6 (2018), 13226. https://doi.org/10.1021/acssuschemeng.8b02840
• 26)   T.  Kawaguchi,  H.  Sakai,  N.  Sheng,  A.  Kurniawan and  T.  Nomura: Appl. Energy, 276 (2020), 115487. https://doi.org/10.1016/j.apenergy.2020.115487
• 27)   T.  Takahashi,  H.  Koide,  H.  Sakai,  D.  Ajito,  A.  Kurniawan,  Y.  Kunisada and  T.  Nomura: Sci. Rep., 11 (2021), 7539. https://doi.org/10.1038/s41598-021-86117-1
• 28)   H.  Sakai,  N.  Sheng,  A.  Kurniawan,  T.  Akiyama and  T.  Nomura: Appl. Energy, 265 (2020), 114673. https://doi.org/10.1016/j.apenergy.2020.114673
• 29)   H.  Sakai,  A.  Kurniawan,  T.  Akiyama and  T.  Nomura: ISIJ Int., 60 (2020), 2152. https://doi.org/10.2355/isijinternational.ISIJINT-2019-827
• 30)  Japan Society of Thermophysical Properties: Thermophysical Properties Handbook, Yokendo, Tokyo, (2008), 20.
• 31)   H. W.  Fricker: Sol. Energy Mater., 24 (1991), 249. https://doi.org/10.1016/0165-1633(91)90065-S
• 32)   D.  Laing,  C.  Bahl,  T.  Bauer,  M.  Fiss,  N.  Breidenbach and  M.  Hempel: Proc. IEEE, 100 (2012), 516. https://doi.org/10.1109/jproc.2011.2154290
• 33)   S.  Hasnain: Energy Convers. Manag., 39 (1998), 1127. https://doi.org/10.1016/S0196-8904(98)00025-9
• 34)   D.  Laing,  W.-D.  Steinmann,  R.  Tamme and  C.  Richter: Sol. Energy, 80 (2006), 1283. https://doi.org/10.1016/j.solener.2006.06.003