The Effects of Using Aluminum Oxide Nanoparticles as Heat Transfer Fillers on Morphology and Thermal Performances of Form-Stable Phase Change Fibrous Membranes Based on Capric–Palmitic–Stearic Acid Ternary Eutectic/Polyacrylonitrile Composite

In this paper, innovative capric–palmitic–stearic acid ternary eutectic/polyacrylonitrile/aluminum oxide (CA–PA–SA/PAN/Al2O3) form-stable phase change composite fibrous membranes (PCCFMs) with different mass ratios of Al2O3 nanoparticles were prepared for thermal energy storage. The influences of Al2O3 nanoparticles on morphology and thermal performances of the form-stable PCCFMs were investigated by scanning electron microscopy (SEM), differential scanning calorimetry (DSC), and measurement of melting and freezing times, respectively. The results showed that there was no apparent leakage trace from the SEM observation. The DSC analysis indicated that the addition of Al2O3 nanoparticles had no significant effect on phase transition temperatures and enthalpies of the CA–PA–SA/PAN/Al2O3 form-stable PCCFMs. The melting peak temperatures and melting enthalpies of form-stable PCCFMs were about 25 °C and 131–139 kJ/kg, respectively. The melting and freezing times of the CA–PA–SA/PAN/Al2O310 form-stable PCCFMs were shortened by approximately 21% and 23%, respectively, compared with those of the CA–PA–SA/PAN form-stable PCCFMs due to the addition of Al2O3 nanoparticles acting as heat transfer fillers.


Introduction
Thermal energy storage technologies for reducing energy consumption and improving energy efficiency have become increasingly important with the growth of population and the development of industry. Among the available strategies for energy storage, latent heat storage technology using phase change materials (PCMs) is considered as one of the most promising technologies. Currently, PCMs have been widely researched and successfully applied in numerous fields such as solar thermal collectors [1], heat recovery systems [2], residential and commercial buildings [3], refrigerated containers [4], greenhouse systems [5], Li-ion battery thermal management [6], thermal regulating fibers and textiles [7], and so forth.
Fatty acid eutectics, as a derivative of fatty acids, exhibit the same desirable properties as those of fatty acids, such as good thermal and chemical stabilities, high latent heat storage capacities, reversible phase change behavior, no supercooling and phase segregation, nontoxicity, noncorrosiveness, and cost effectiveness [8]. However, low thermal conductivity and liquid leakage problems limit their feasibility in thermal energy storage applications to a certain extent. According to the literature, the syringe with a stainless-steel needle whose inner diameter was 0.6 mm. The solution feed rate was set at 1 mL/h. A roller wrapped with aluminum foil was used as the collector, which was connected to the ground. A positive high voltage was applied to the needle during the electrospinning process. The electrospinning parameters, such as applied voltage, tip-to-collector distance, and rotating speed of the roller, were fixed at 18 kV, 18 cm, and 200 rpm, respectively. The obtained nanofibrous membranes were named as PAN, PAN/Al 2 O 3 5, and PAN/Al 2 O 3 10.

Fabrication of CA-PA-SA/PAN/Al 2 O 3 Form-Stable PCCFMs
The CA-PA-SA ternary eutectic with eutectic mass ratio of 83.82/10.19/5.99 was prepared as solid-liquid PCMs by heating-ultrasonic method, which has been reported by my previous published literatures [27,28]. The prepared PAN/Al 2 O 3 supporting membranes were immersed into the molten CA-PA-SA ternary eutectic for 1 h in the oven at 40 • C until the absorption was saturated. Subsequently, the PAN/Al 2 O 3 nanofibrous membranes absorbed with CA-PA-SA ternary eutectic were hung in the oven for 5 h to removal residual CA-PA-SA ternary eutectic on the surface of membranes. Finally, the prepared form-stable PCCFMs were termed as CA-PA-SA/PAN, CA-PA-SA/PAN/Al 2 O 3 5, and CA-PA-SA/PAN/Al 2 O 3 10. The similar preparation method has been reported in the literatures published by my research group [19,20,28].

Scanning Electron Microscopy
Morphological structure of electrospun PAN and PAN/Al 2 O 3 nanofibrous membranes, as well as the CA-PA-SA/PAN/Al 2 O 3 form-stable PCCFMs with different amounts of Al 2 O 3 nanoparticles, were observed by scanning electron microscope (SEM, S-3400N, Hitachi, Tokyo, Japan).

Differential Scanning Calorimeter
Differential scanning calorimeter (DSC-Q200, TA Instruments-Waters LLC, Shanghai, China was used to investigate the thermal energy storage properties including peak onset temperatures (T o ), melting peak temperatures (T m ), freezing peak temperatures (T f ), peak end temperatures (T e ), melting enthalpies (∆H m ), and freezing enthalpies (∆H f ) of the CA-PA-SA ternary eutectic and the CA-PA-SA/PAN/Al 2 O 3 form-stable PCCFMs from −20 • C to 60 • C with the scanning rate of 8 • C/min under a nitrogen atmosphere. The flow rate of nitrogen was set at 50 mL/min for DSC measurement.
Moreover, good thermal stability over a number of thermal cycles are necessary for form-stable PCCFMs. Therefore, the accelerated thermal cycling experiment containing 100 heating and cooling thermal cycles was conducted to investigate the thermal stability of the CA-PA-SA/PAN/Al 2 O 3 10 form-stable PCCFMs in a climatic chamber (Binder MK56, BINDER GmbH, Tuttlingen, Germany) with the heating and cooling rates of 5 • C/min in the temperature range of −10-50 • C. Subsequently, the phase change temperatures and enthalpies of the CA-PA-SA/PAN/Al 2 O 3 10 form-stable PCCFMs after thermal cycling were also measured by DSC analysis under the same test conditions.

Measurement of Melting and Freezing Times
Improvement of heat transfer performances of the fabricated CA-PA-SA/PAN/Al 2 O 3 form-stable PCCFMs was evaluated by measurement of melting and freezing times. The CA-PA-SA/PAN/Al 2 O 3 form-stable PCCFMs were cut and put in the test bottle. The test bottle was placed into a water bath at 40 • C for melting measurement. Thereafter, the test bottle was rapidly transferred to a refrigerator at −10 • C for freezing measurement. A thermocouple positioned at the center of the bottle was employed to measure temperature variation of samples during the testing process. The temperature variation caused by thermal energy storage and release was automatically recorded by a computer with the temperature measuring accuracy of ±2 • C. The same analytical method has also been reported by my previous literatures [19,20,28]. Figure 1 shows the SEM images of electrospun PAN fibrous membrane, PAN/Al 2 O 3 5, and PAN/Al 2 O 3 10 composite fibrous membranes. As shown in Figure 1, electrospun PAN nanofibers exhibited the smooth surface and uniform fiber diameters along their lengths. The average fiber diameter was observed to be about 200 nm. It can be clearly seen from Figure 1 that electrospun PAN nanofibers were randomly deposited to form fibrous membranes with a three-dimensional porous network structure during the electrospinning process. Figure 1b Figure 2 that the CA-PA-SA ternary eutectic was uniformly adsorbed into the three-dimensional porous network structure of electrospun PAN/Al 2 O 3 fibrous membranes due to the high infiltrating ability of the CA-PA-SA ternary eutectic as well as capillary effect and surface tension force of fibrous membranes. Figure 2 indicated that the combination of Al 2 O 3 nanoparticles had no significant effect on the morphological structure of the CA-PA-SA/PAN/Al 2 O 3 form-stable PCCFMs. It is worthwhile to note that the melting temperature of the CA-PA-SA ternary eutectic was in the range of around 20-33 • C, and its melting peak temperature was about 25 • C (see Table 1). As we know, the working temperature of SEM characterization was also about 25 • C, which meant that the solid-liquid phase change behavior (i.e., melting process) of the CA-PA-SA ternary eutectic adsorbed into the form-stable PCCFMs was taking place during the process of SEM observation. However, the SEM images clearly demonstrated that the melted CA-PA-SA ternary eutectic was still encapsulated into the three-dimensional porous network structure of electrospun PAN/Al 2 O 3 fibrous membranes even when the ambient temperature was consistent with its melting peak temperature. In other words, the developed CA-PA-SA/PAN/Al 2 O 3 PCCFMs exhibited a form-stable structure because of the supporting effect of electrospun PAN/Al 2 O 3 nanofibrous membranes.  Figure 1 shows the SEM images of electrospun PAN fibrous membrane, PAN/Al2O35, and PAN/Al2O310 composite fibrous membranes. As shown in Figure 1, electrospun PAN nanofibers exhibited the smooth surface and uniform fiber diameters along their lengths. The average fiber diameter was observed to be about 200 nm. It can be clearly seen from Figure 1 that electrospun PAN nanofibers were randomly deposited to form fibrous membranes with a three-dimensional porous network structure during the electrospinning process. Figure 1b,c reveal that electrospun PAN/Al2O3 composite fibrous membranes with different amounts of Al2O3 nanoparticles also presented threedimensional porous network structures consisting of randomly distributed PAN/Al2O3 composite nanofibers with the average fiber diameter being in the range of approximately 218-274 nm. It was clear that electrospun PAN/Al2O3 composite nanofibers showed a relatively rough surface structure compared to that of the pure PAN nanofibers owing to the addition of Al2O3 nanoparticles. Moreover, Figure 2 presents the SEM images of the CA-PA-SA/PAN, CA-PA-SA/PAN/Al2O35, and CA-PA-SA/PAN/Al2O310 form-stable PCCFMs. It can be clearly found from Figure 2 that the CA-PA-SA ternary eutectic was uniformly adsorbed into the three-dimensional porous network structure of electrospun PAN/Al2O3 fibrous membranes due to the high infiltrating ability of the CA-PA-SA ternary eutectic as well as capillary effect and surface tension force of fibrous membranes. Figure 2 indicated that the combination of Al2O3 nanoparticles had no significant effect on the morphological structure of the CA-PA-SA/PAN/Al2O3 form-stable PCCFMs. It is worthwhile to note that the melting temperature of the CA-PA-SA ternary eutectic was in the range of around 20-33 °C, and its melting peak temperature was about 25 °C (see Table 1). As we know, the working temperature of SEM characterization was also about 25 °C, which meant that the solid-liquid phase change behavior (i.e., melting process) of the CA-PA-SA ternary eutectic adsorbed into the form-stable PCCFMs was taking place during the process of SEM observation. However, the SEM images clearly demonstrated that the melted CA-PA-SA ternary eutectic was still encapsulated into the three-dimensional porous network structure of electrospun PAN/Al2O3 fibrous membranes even when the ambient temperature was consistent with its melting peak temperature. In other words, the developed CA-PA-SA/PAN/Al2O3 PCCFMs exhibited a form-stable structure because of the supporting effect of electrospun PAN/Al2O3 nanofibrous membranes.        Table 1. It can be clearly seen from Figure 3 that no solid-liquid phase change peak (i.e., endothermic or exothermic peak) was observed from the DSC curves of electrospun PAN and PAN/Al2O310 nanofibrous membranes, which confirmed that electrospun PAN and PAN/Al2O3 nanofibrous membranes acting as supporting materials did not provide any phase change enthalpies within the temperature range of the DSC measurement. As shown in Table 1, the melting temperatures of the CA-PA-SA ternary eutectic were in the range of about 20-33 °C, and its melting peak temperature was approximately 25 °C. It is interesting to note that this temperature was generally considered as the most appropriate temperature for temperature regulation fibers and textiles as well as energy-saving building applications, thus the CA-PA-SA ternary eutectic was chosen as solid-liquid PCMs in this paper. Figure 3 presents that only one endothermic or exothermic peak was observed from the DSC curves of the CA-PA-SA/PAN/Al2O3 form-stable PCCFMs, which was similar to those of the pure CA-PA-SA ternary eutectic. DSC analysis also indicated that the phase change temperatures of the CA-PA-SA/PAN/Al2O3 form-stable PCCFMs were around 10-30 °C, as revealed in Table 1. Additionally, compared with those of pure CA-PA-SA ternary eutectic, the phase change enthalpies of the CA-PA-SA/PAN/Al2O3 form-stable PCCFMs had a varying degree of reduction because electrospun PAN/Al2O3 nanofibrous membranes acting as supporting materials accounted for a certain proportion of the form-stable PCCFMs. Meanwhile, Figure 3 and Table 1

Samples
Melting Process Freezing Process   Table 1. It can be clearly seen from Figure 3 that no solid-liquid phase change peak (i.e., endothermic or exothermic peak) was observed from the DSC curves of electrospun PAN and PAN/Al 2 O 3 10 nanofibrous membranes, which confirmed that electrospun PAN and PAN/Al 2 O 3 nanofibrous membranes acting as supporting materials did not provide any phase change enthalpies within the temperature range of the DSC measurement. As shown in Table 1, the melting temperatures of the CA-PA-SA ternary eutectic were in the range of about 20-33 • C, and its melting peak temperature was approximately 25 • C. It is interesting to note that this temperature was generally considered as the most appropriate temperature for temperature regulation fibers and textiles as well as energy-saving building applications, thus the CA-PA-SA ternary eutectic was chosen as solid-liquid PCMs in this paper. Figure 3 presents that only one endothermic or exothermic peak was observed from the DSC curves of the CA-PA-SA/PAN/Al 2 O 3 form-stable PCCFMs, which was similar to those of the pure CA-PA-SA ternary eutectic. DSC analysis also indicated that the phase change temperatures of the CA-PA-SA/PAN/Al 2 O 3 form-stable PCCFMs were around 10-30 • C, as revealed in Table 1. Additionally, compared with those of pure CA-PA-SA ternary eutectic, the phase change enthalpies of the CA-PA-SA/PAN/Al 2 O 3 form-stable PCCFMs had a varying degree of reduction because electrospun PAN/Al 2 O 3 nanofibrous membranes acting as supporting materials accounted for a certain proportion of the form-stable PCCFMs. Meanwhile, Figure 3 and Table 1 Table 2. It can be seen from Figure 4 that the endothermic and exothermic curves of the CA-PA-SA/PAN/Al2O310 form-stable PCCFMs after thermal cycling had no obvious variation compared with each other. It should be noted that the CA-PA-SA/PAN/Al2O310 form-stable PCCFMs were hung in the climatic chamber for the accelerated thermal cycling experiment. If the leakage behavior occurred during the measurement, the CA-PA-SA eutectic mixture leaking out of the form-stable PCCFMs would be removed from the membranes and dropped by gravity, which would lead to the decrease of phase change enthalpies. The data listed in Table 2 indicated that no significant change in phase change enthalpies of the CA-PA-SA/PAN/Al2O310 form-stable PCCFMs was observed, suggesting that the CA-PA-SA/PAN/Al2O310 form-stable PCCFMs had good thermal reliability and reusability in terms of thermal energy storage and release properties.   Table 2. It can be seen from Figure 4 that the endothermic and exothermic curves of the CA-PA-SA/PAN/Al 2 O 3 10 form-stable PCCFMs after thermal cycling had no obvious variation compared with each other. It should be noted that the CA-PA-SA/PAN/Al 2 O 3 10 form-stable PCCFMs were hung in the climatic chamber for the accelerated thermal cycling experiment. If the leakage behavior occurred during the measurement, the CA-PA-SA eutectic mixture leaking out of the form-stable PCCFMs would be removed from the membranes and dropped by gravity, which would lead to the decrease of phase change enthalpies. The data listed in Table 2    It is worthwhile to note that the phase change temperatures and enthalpies of form-stable PCCFMs are important thermal performance characteristics that could obviously affect whether they can be used in the specific practical engineering applications. For example, Xu and coworkers reported that the PCMs should have suitable phase change temperature of about 18-30 °C to maintain thermal comfort for building energy-saving applications [29]. Mondal reported that an important performance requirement of a PCM for temperature-regulating fibers and textiles application is that its melting temperature should be between 15 °C and 35 °C [30]. Table 3 summarizes the phase change peak temperatures and enthalpies of the CA-PA-SA/PAN/Al2O3 form-stable PCCFMs and their comparison with some form-stable PCMs reported in the literatures [9,10,13,14,16,17,20,24]. As shown in Table 3, the solid-liquid PCMs loaded in the form-stable PCMs published in the literatures included pure fatty acids (i.e., stearic acid [13] and lauric acid [16]), paraffin [9,14], fatty acid ester (i.e., glycerol monostearate [17]), fatty acid eutectics (i.e., capric-palmitic acid binary eutectic [10] and capric-myristic-stearic acid ternary eutectic [20]) and polyethylene glycol [24]. It was clearly found from Table 3 that the form-stable PCMs, including paraffin/EP [9], CA-PA/diatomite/EG [10], and CA-MA-SA/CA [20], exhibited similar phase change peak temperatures with those of the CA-PA-SA/PAN/Al2O3 form-stable PCCFMs reported in this paper (i.e., about 25 °C), but their phase change enthalpies were obviously lower than those of the CA-PA-SA/PAN/Al2O3 form-stable PCCFMs.

Thermal Energy Storage Properties
Based on the consideration of phase change temperature, the developed CA-PA-SA/PAN/Al2O3 form-stable PCCFMs could be considered as promising form-stable PCMs for thermal energy storage applications such as temperature-regulating textiles and building energy conservation.  It is worthwhile to note that the phase change temperatures and enthalpies of form-stable PCCFMs are important thermal performance characteristics that could obviously affect whether they can be used in the specific practical engineering applications. For example, Xu and coworkers reported that the PCMs should have suitable phase change temperature of about 18-30 • C to maintain thermal comfort for building energy-saving applications [29]. Mondal reported that an important performance requirement of a PCM for temperature-regulating fibers and textiles application is that its melting temperature should be between 15 • C and 35 • C [30]. Table 3 summarizes the phase change peak temperatures and enthalpies of the CA-PA-SA/PAN/Al 2 O 3 form-stable PCCFMs and their comparison with some form-stable PCMs reported in the literatures [9,10,13,14,16,17,20,24]. As shown in Table 3, the solid-liquid PCMs loaded in the form-stable PCMs published in the literatures included pure fatty acids (i.e., stearic acid [13] and lauric acid [16]), paraffin [9,14], fatty acid ester (i.e., glycerol monostearate [17]), fatty acid eutectics (i.e., capric-palmitic acid binary eutectic [10] and capric-myristic-stearic acid ternary eutectic [20]) and polyethylene glycol [24]. It was clearly found from Table 3 that the form-stable PCMs, including paraffin/EP [9], CA-PA/diatomite/EG [10], and CA-MA-SA/CA [20], exhibited similar phase change peak temperatures with those of the CA-PA-SA/PAN/Al 2 O 3 form-stable PCCFMs reported in this paper (i.e., about 25 • C), but their phase change enthalpies were obviously lower than those of the CA-PA-SA/PAN/Al 2 O 3 form-stable PCCFMs.
Based on the consideration of phase change temperature, the developed CA-PA-SA/PAN/Al 2 O 3 form-stable PCCFMs could be considered as promising form-stable PCMs for thermal energy storage applications such as temperature-regulating textiles and building energy conservation.

Thermal Energy Storage and Release Performance
The improvement of heat transfer efficiencies of the CA-PA-SA/PAN/Al 2 O 3 form-stable PCCFMs with different mass fractions of Al 2 O 3 nanoparticles was discussed through the comparison of melting and freezing times during phase change processes. The corresponding temperature-time curves acquired from thermal performance tests are shown in Figure 5. The interval time between the initial temperature (around −5 • C) and the melting peak temperature (around 25 • C) was defined as the melting time, and the interval time between the initial temperature (around 35 • C) and the freezing peak temperature (around 17 • C) was defined as the freezing time. As shown in Figure

Conclusions
The novel CA-PA-SA/PAN/Al2O3 form-stable PCCFMs were prepared by developing electrospun PAN/Al2O3 nanofibrous membranes with different mass ratios of Al2O3 nanoparticles as supporting materials to incorporate CA-PA-SA ternary eutectic via physical adsorption for thermal energy storage. The SEM images revealed that the CA-PA-SA ternary eutectic was successfully absorbed into the three-dimensional porous network structure of electrospun PAN/Al2O3 nanofibrous membranes owing to the high wetting ability of CA-PA-SA ternary eutectic, the capillary effect, and surface tension force of nanofibrous membranes. The melting peak temperatures and melting enthalpies of the CA-PA-SA/PAN/Al2O3 form-stable PCCFMs were approximately 25 °C and 131-139 kJ/kg, respectively. The absorption capacities were determined as about 90-95%. The results of thermal performance tests indicated that there was a significant improvement in heat transfer rates of the CA-PA-SA/PAN/Al2O3 form-stable PCCFMs as the Al2O3 nanoparticles loading increased. The melting and freezing times of the CA-PA-SA/PAN/Al2O310 form-stable PCCFMs were shortened by around 21% and 23%, respectively, in comparison with those of the CA-PA-SA/PAN form-stable PCCFMs. Therefore, based on these results, the developed CA-PA-SA/PAN/Al2O3 form-stable PCCFMs are concluded to have potential applications in the fields of temperature-regulating fibers and textiles, as well as building energy conservation.
Author Contributions: H.K. conceived and designed the experiments, did the experiments and data analysis, as well as wrote this article. Y.L. contributed reagents/materials/analysis tools, did the measurements and data analysis, as well as performed the literature study.

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

Conclusions
The novel CA-PA-SA/PAN/Al 2 O 3 form-stable PCCFMs were prepared by developing electrospun PAN/Al 2 O 3 nanofibrous membranes with different mass ratios of Al 2 O 3 nanoparticles as supporting materials to incorporate CA-PA-SA ternary eutectic via physical adsorption for thermal energy storage. The SEM images revealed that the CA-PA-SA ternary eutectic was successfully absorbed into the three-dimensional porous network structure of electrospun PAN/Al 2 O 3 nanofibrous membranes owing to the high wetting ability of CA-PA-SA ternary eutectic, the capillary effect, and surface tension force of nanofibrous membranes. The melting peak temperatures and melting enthalpies of the CA-PA-SA/PAN/Al 2 O 3 form-stable PCCFMs were approximately 25 • C and 131-139 kJ/kg, respectively. The absorption capacities were determined as about 90-95%. The results of thermal performance tests indicated that there was a significant improvement in heat transfer rates of the CA-PA-SA/PAN/Al 2 O 3 form-stable PCCFMs as the Al 2 O 3 nanoparticles loading increased. The melting and freezing times of the CA-PA-SA/PAN/Al 2 O 3 10 form-stable PCCFMs were shortened by around 21% and 23%, respectively, in comparison with those of the CA-PA-SA/PAN form-stable PCCFMs. Therefore, based on these results, the developed CA-PA-SA/PAN/Al 2 O 3 form-stable PCCFMs are concluded to have potential applications in the fields of temperature-regulating fibers and textiles, as well as building energy conservation.