Experimental Investigation on Graphene Oxide/SrCl2·6H2O Modified CaCl2·6H2O and the Resulting Thermal Performances

Although the inorganic salt hydrate phase change materials (PCMs) such as CaCl2·6H2O have promising potential for thermal energy storage in building application, the issue of supercooling has restricted their practical application. In this study, graphene oxide (GO) and SrCl2·6H2O as binary nucleation agents were used to modify CaCl2·6H2O and reduce its supercooling degree. Compared with pure CaCl2·6H2O, the incorporation of graphene oxide (GO)/SrCl2·6H2O reduced the supercooling degree to 0.3 °C significantly. In addition, the supercooling degree of modified CaCl2·6H2O after 200 thermal cycles was still much lower than that of non-modified CaCl2·6H2O. From the results of differential scanning calorimetry (DSC), the latent heat value and phase change temperature of the modified CaCl2·6H2O were 207.88 J/g and 27.6 °C, respectively. Aluminum capsules were used to encapsulate the modified PCM and placed inside the composite wallboard. The thermal performances of the composite wallboard with modified PCM were investigated using infrared thermography. Experimental results showed that the average temperature difference between the top and bottom surfaces of modified CaCl2·6H2O/wallboard composite after 1 h heating was kept around 15.8 °C, while it was 4.9 °C for the control wallboard. The above test results proved that the modified CaCl2·6H2O demonstrated good thermal performance and can be used in buildings to maintain thermal comfort.


Introduction
Inorganic phase change materials (PCMs) generally are non-flammable, readily available, and cheap, but the issues of corrosive to metal and supercooling have restricted their utilization in practical applications. Apparently, an inorganic PCM without the aforesaid issues can be seen as a promising candidate for building application [1,2]. In order to overcome the negative influence caused by supercooling, some scientific mechanisms and measures on inorganic PCMs have been reported. According to the current literature [3] in relation to supercooling reduction, nucleating agents are commonly used to mitigate the supercooling of inorganic PCMs. Some nucleating agents, such as copper foam (for thermal conductivity and matrix), carboxyl methyl cellulose (CMC, thickener), disodium hydrogen phosphate dodecahydrate (DHPD, nucleator), and sodium acetate trihydrate (SAT), were commonly used and studied [4,5], however the supercooling of inorganic PCM is still not completely resolved.
Recently, due to the rapid development of nanomaterials, some nanomaterials can be used as nucleating agents to stimulate the crystallization of inorganic PCMs and some of them can certainly change the properties of PCMs [4,6]. For instance, some researchers [7][8][9] found that adding graphene (0.02 wt %) could reduce the supercooling degree of pure water. On other studies, Li et al. [10,11] confirmed that SrCl 2 ·6H 2 O could reduce the supercooling of inorganic PCM. Therefore, it has been suggested that the usage of graphene oxide (GO) can be greatly reduced with the presence of SrCl 2 ·6H 2 O to have the effect on supercooling reduction. Furthermore, the thermal performance of PCMs can be enhanced due to high thermal conductivity of GO.
In order to enable PCMs to be used in buildings, they need to be encapsulated so as to avoid leakage likeliness. Many methods and measures have been proposed in the literature. For example, Cui et al. [12] and Fu et al. [13,14] respectively introduced interfacial polymerization method and vacuum-impregnation method to envelop PCM to be shape-stabilized PCM. However, the above methods could only store a low content of PCM inside the PCM composite materials. In this study, an aluminum foil capsule was proposed to carry a high volume of PCM in the building envelopes. Owing to its excellent corrosion resistance, the aluminum foil can also minimize the corrosion effects of hydrate salts.
In our previous study, the effect of individual graphene or SrCl 2 ·6H 2 O on supercooling of CaCl 2 ·6H 2 O had been investigated [15]. In this study, therefore, we proposed the use of GO/SrCl 2 ·6H 2 O to modify CaCl 2 ·6H 2 O and reduce its supercooling. It is believed that the modified CaCl 2 ·6H 2 O composites with high latent heat and supercooling-free would be very attractive for future use in energy-efficient buildings to achieve thermal comfort. In order to avoid deliquescence and metal corrosion, aluminum foil capsules were used to carry the modified CaCl 2 ·6H 2 O. The thermal performances of composite wallboard with modified CaCl 2 ·6H 2 O were investigated using infrared thermography.

Materials
In this study, the modified CaCl 2 ·6H 2 O composite was prepared from several raw materials. Anhydrous calcium chloride (CaCl 2 ) and strontium chloride hexahydrate (SrCl 2 ·6H 2 O) with purity >99% were supplied from Guangdong, China, Huada Chemical Co., Ltd. Deionized water was obtained from Guangdong, China, Hugke Water Treatment Equipment Co., Ltd. An industrial-grade graphene oxide nano-sheets (thickness < 5 nm, purity > 99%, size 2-8 µm) with an interlayer distance of 0.824 nm was supplied from Chengdu Organic Chemical Co. Ltd., Chinese Academy of Sciences. Other materials such as polystyrene foam plate (thickness = 45 mm), metal sheet and glass glue were obtained from the local building materials market in Shenzhen, China. The nucleating effects of GO were characterized by Fourier transform infrared (FT-IR, Perkin-Elmer, Waltham, MA, USA) spectroscopy and scanning electron microscopy (SEM, Quanta TM 250 FEG, Hillsboro, OR, USA).

Preparation of Modified CaCl 2 ·6H 2 O
To reduce the supercooling degree of CaCl 2 ·6H 2 O, GO, and SrCl 2 ·6H 2 O were added as binary nucleating agents. The procedures of preparation are as follows: Firstly, 0.02 wt % GO (weight for CaCl 2 ·6H 2 O) was initially mixed with 50 g deionized water and followed by sonication for about 30 minutes to ensure better dispersion results. The sonication was carried out by a rod-model ultrasonic machine (JY-92-IIN, Ningbo Xinzhi Biotech Co., Ltd., Ningbo, China, 25 kHz, power: 375 W). Then, the anhydrous CaCl 2 (50 g) was added to the GO suspension and agitated using a mechanical agitator for a total of 30 min in a water-bath (HH-2, Zhiborui Instrument Manufacturer, Changzhou, China) at a temperature of 40 • C. After ultrasonic dispersion, SrCl 2 ·6H 2 O (0.8 wt %) was added and finally the modified PCM composite was prepared.

Thermo-Physical Performance of Modified CaCl 2 ·6H 2 O
The thermo-physical performances of modified CaCl 2 ·6H 2 O were characterized by differential scanning calorimetry (DSC, DSC-200L, Nanjing Dazhan Electrical Technology Company, Nanjing, China) and T-history method to obtain the latent heat value and supercooling degrees, respectively. The supercooling degree of modified CaCl 2 ·6H 2 O under various thermal cycling conditions was studied. The samples were placed in a temperature and humidity programmable chamber (TEMI 300, Dongguan Bell Test Equipment Co., Ltd., Dongguan, China). The thermal cycle started when the sample was heated from 10 to 40 • C in 10 min (at a rate of 3 • C/min) and maintained at 40 • C for 30 min. Then, the temperature was reduced from 40 to 10 • C in 10 min (at a rate of −3 • C/min), and the temperature was held at 10 • C for 30 min, and then the cycle ended. A total of 200 thermal cycles were conducted where the humidity was remained constant for 50 RH %.

Encapsulation of Modified CaCl 2 ·6H 2 O
As shown in Figure 1, there was a thin piece of polyethylene (PE) sheet embedded inside the cap of each aluminum container (Diameter: 52 mm, Height: 20 mm) to prevent leakage. The procedures for encapsulation of modified CaCl 2 ·6H 2 O are as follows: Firstly, the modified CaCl 2 ·6H 2 O was packaged inside an aluminum capsule and then the capsule was stored in a refrigerator at −20 • C for 2h. Besides, a polypropylene (PP) bag was used to wrap each aluminum container for secondary protection and finally the bag was vacuumed and sealed.

Preparation of Modified CaCl 2 ·6H 2 O/Wallboard
The aluminum capsules containing modified CaCl 2 ·6H 2 O were used to form modified CaCl 2 ·6H 2 O/wallboard composite with a dimension of 650 mm × 550 mm × 30 mm. Each modified CaCl 2 ·6H 2 O/woodboard composite consists of three layers as shown in Figure 1a,b. The bottom layer is a thin metal board with a thickness of 4 mm. The metal board was used to resist the gravity of above layers. The layer in middle was a polyethylene sheet (thickness: 21 mm) embedded regularly with 56 aluminum containers wrapped with PP package. Glass glue was used to fill the gap between the capsules and polyethylene layer for better compactness and to avoid unwanted heat transition. The top layer is a thin woodboard with a thickness of 5 mm. A control wallboard without the modified CaCl 2 ·6H 2 O was used for comparison.

Thermo-Regulated Performance of Modified CaCl 2 ·6H 2 O
To provide better and distinctive observation of surface temperatures, infrared thermography technique was used to record the temperature variation and heat distribution of the top and bottom surfaces of wallboards after a prolonged heating time. In order to eradicate the environmental impact to the bottom surface temperature, a special room model made of polyethylene was used, as shown in Figure 2. The CaCl 2 ·6H 2 O /woodboard was used as roof panel on the room model with a dimension of 650 mm × 550 mm × 550 mm. The room models, except the roof, were all made of polyethylene (thickness: 45 mm). One of the two infrared cameras (Ti450, Fluke, Seattle, WA, USA) was installed at the bottom of room model and responsible for recording the internal (bottom) surface temperature of composite woodboard. Another camera was used to record the external (top) surface temperature at the same time. The heat lamp was placed one meter above the room model and operated for 7 h.

Thermo-Physical Performance of Modified CaCl 2 ·6H 2 O
The supercooling degree of modified CaCl 2 ·6H 2 O is shown in Figure 3a. The cooling curve of the modified CaCl 2 ·6H 2 O displays a huge difference compared to the control (without GO and SrCl 2 ·6H 2 O). After the modification, the supercooling degree of CaCl 2 ·6H 2 O reduced significantly from 25.4 • C to 0.3 • C. Having a reduction of 98.8% supercooling degree, it can be concluded that the binary nucleation agents are super effective to stimulate the crystallization of CaCl 2 ·6H 2 O, and supercooling is no longer a concern in the modified PCM. The reduction of supercooling degree can be explained by the presence of oxygen-containing functional groups on the surface of GO which have been recognized as a potential platform in crystal attachment due to better wetting and adsorption [16]. According to the FT-IR results, a powerful absorption at 3413 cm −1 due to O-H stretching vibration could be seen in Figure 4a. C=O stretching was also observed at 1781 cm −1 , and the absorption around 1400 cm −1 was attributed to the tertiary C-OH groups. The peak occurred at 1149 cm −1 was stretching vibrations of C-O [17,18]. These functional groups indicated the oxidation process of GO was in place. Besides, as shown in Figure 4b, the GO sheet can be characterized as nano-scale thin with large surface area for the attachment of oxygen functional groups. Thus, better dispersion of GO by means of ultra-sonication can effectively modify CaCl 2 ·6H 2 O [19]. Besides, Figure 3b shows the DSC results of modified CaCl 2 ·6H 2 O. According the provisions of the International Confederation for Thermal Analysis (ICTA) standardization committee, the intersection of the front extension baseline and the maximum frontier slope of the peak represents the melting point, and the area of the peak represents the latent heat. Hence, the latent heat of modified CaCl 2 ·6H 2 O was around 207.88 J/g, which is higher than those reported in other studies (170 J/g and 191 J/g) [20]. The phase change temperature was 27.6 • C, which is within the operational temperature and suitable for building applications.    Figure 5 shows the thermal cycle test results of modified CaCl 2 ·6H 2 O. From the figure, it can be seen that the supercooling degree slightly increased with increasing thermal cycles, but it was still remained less than 1 • C after 200 thermal cycles. It is well known that supercooling is a major issue which hampers the long-term performance of PCM. In this study, the supercooling degree of modified CaCl 2 ·6H 2 O after 200 thermal cycles was still much lower than that of pure CaCl 2 ·6H 2 O. Thus, the newly modified CaCl 2 ·6H 2 O is more suitable for practical application. Further study about effect of thermal cycles more than 200 times on the supercooling degree of the modified CaCl 2 ·6H 2 O will be presented in future publication.

Conclusions
Based on the above, the following conclusions can be drawn: (1) In this study, graphene and SrCl 2 ·6H 2 O were utilized as nano nucleating agents and successfully reduced the supercooling of pure CaCl 2 ·6H 2 O to around 0.3 • C which was much lower than that of the original pure CaCl 2 ·6H 2 O (25.4 • C). The latent heat value and phase change temperature of modified CaCl 2 ·6H 2 O were 207.88 J/g and 27.6 • C, respectively.
(2) The supercooling degree of modified CaCl 2 ·6H 2 O after 200 thermal cycles was still much lower than that of non-modified PCM. (3) Aluminum capsules wrapped with polypropylene bag can effectively prevent leakage and deliquescence of PCM. (4) The infrared thermography showed that the temperature difference between the top and bottom layers of modified CaCl 2 ·6H 2 O/wallboard composite was 15.8 • C after heating for one hour, while it was 4.9 • C for control wallboard. The results further demonstrated the excellent thermal energy storage and thermal-regulated capacity of modified CaCl 2 ·6H 2 O.
Author Contributions: Z.J. did the experiments and data analysis. Y.T. did the experiments and data analysis. X.X. did the data analysis and wrote part of this paper. H.C. provided the original ideas and did the experiments and data analysis. W.T. wrote part of this paper. G.S. wrote part of this paper. Y.Y. wrote part of this paper.

Funding:
The work described in this paper was fully supported by grants from Natural Science Foundation of China (no. 51678367), Shenzhen Fund for Fundamental Research (no. JCYJ20160422092836654), and Australian Research Council Discovery Project (no. G1500225).

Conflicts of Interest:
The authors declare no conflict of interest.