Applied Energy xxx (xxxx) xxxx Contents lists available at ScienceDirect Applied Energy journal homepage: www.elsevier.com/locate/apenergy Inclusion of methyl stearate/diatomite composite in gypsum board ceiling for building energy conservation Md Jaynul Abdena, Zhong Taoa, , Zhu Pana, Laurel Georgeb, Richard Wuhrerb ⁎ a b Centre for Infrastructure Engineering, Western Sydney University, Penrith, NSW 2751, Australia Advanced Materials Characterisation Facility, Western Sydney University, Parramatta, NSW 2116, Australia H I GH L IG H T S stearate was incorporated into diatomite to develop form-stable PCM. • Methyl board ceiling containing 40% FSPCM was developed. • Gypsum thermal environment is improved with a saving of 16.2% in cooling energy. • Indoor • The payback period of FSPCM in gypsum board ceiling is estimated 1.7 years. A R T I C LE I N FO A B S T R A C T Keywords: Phase change materials Solar thermal energy Gypsum board False ceiling Energy savings Payback period Gypsum board with the advantages of low cost and ease of placement is widely used in buildings as ceiling and interior wall coverings. The interest in phase change material integrated gypsum board has been growing significantly over the last few years. In the present work, a composite form-stable phase change material (FSPCM) for use in gypsum board was developed based on methyl stearate and diatomite by a direct impregnation method. The material properties and thermal behaviour of the methyl stearate/diatomite composite were examined through different characterisation techniques to check its suitability for use in building materials, and particularly in gypsum board. Thermal cycling tests and thermogravimetric results confirmed that the FSPCM exhibited excellent thermal reliability and stability for long-term thermal management applications. The FSPCM was then integrated into gypsum board for potential use as building false ceiling for energy conservation. A small-scale test chamber with FSPCM-integrated gypsum board ceiling was prepared and modelled. The thermal/energy performance is evaluated and an economic analysis is performed against the conventional gypsum board without FSPCM in real environmental conditions. The results show that the use of FSPCM in gypsum board ceiling is economically feasible with cooling load savings of 16.2%. 1. Introduction With rapidly expanding global economies and growing populations, the demand for energy is also rapidly increasing. New energy production and storage technologies have therefore attracted great attention in recent years. Thermal energy storage (TES) is a new environmentallyfriendly energy saving technique which can be effectively used to minimise the gap between the need and provision of energy. Among many TES approaches, latent heat storage is a desirable one, which uses suitable materials that have a high energy storage density with a small temperature change during the process of storing and releasing heat [1]. Many phase change materials (PCMs) have the capacity to store solar thermal energy in the form of latent heat above their phase ⁎ transformation temperature, and then slowly release the heat during the reverse phase transition process over a narrow temperature range [2]. For this reason, they have captured prevalent interest in engineering applications, especially for the conservation of energy in buildings [3]. In particular, the high latent heat storage capacity of solid-liquid PCMs makes them one of the most promising materials for the use in TES systems. PCMs can be classified into organic and inorganic materials according to their chemical compositions. Compared to inorganic PCMs, organic ones, such as paraffin, fatty-acid and its derivatives, polyethylene glycol and polyalcohol, have gained extensive interest for storing solar thermal energy owing to their exceptional thermo-physical properties and chemical stability [4]. Among them, fatty acid esters are Corresponding author. E-mail address: z.tao@westernsydney.edu.au (Z. Tao). https://doi.org/10.1016/j.apenergy.2019.114113 Received 7 May 2019; Received in revised form 10 October 2019; Accepted 10 November 2019 0306-2619/ © 2019 Elsevier Ltd. All rights reserved. Please cite this article as: Md Jaynul Abden, et al., Applied Energy, https://doi.org/10.1016/j.apenergy.2019.114113 Applied Energy xxx (xxxx) xxxx M.J. Abden, et al. indoor air temperature during daytime and reduced heating load at night in Montreal, Canada. The test was conducted in a full-scale outdoor room using gypsum board containing 25% PCM as inside wall covering. The beneficial effects of using wallboard containing PCM on thermal comfort and energy saving were also reported in [21–23]. Zhou et al. [24] prepared two model rooms with sizes of 0.5 × 0.5 × 0.3 m, where gypsum board without PCM was used for one room and PCM gypsum board was used for the other. When the external temperature varied between 20 and 30 °C, a maximum temperature reduction of 5 °C was achieved for internal environment of the room with PCM gypsum board against 1.8 °C for the one with normal gypsum board. Schossig et al. [25] integrated PCM micro-capsules into gypsum wallboard and tested the performance in real office buildings. The measured data proved the capacity of PCM gypsum board in reducing the cooling demand and increasing the thermal comfort of these buildings. The above studies mainly concerned the performance of PCM gypsum wallboard. However, roof is also a main element in building envelope and significantly affects the indoor thermal environment, which is particularly true for single-storey houses. A full-scale experiment was conducted by Kosny et al. [26] from November 2009 to October 2010 in Oak Ridge, Tennessee, USA to evaluate the thermal performance of an attic roof with photovoltaic (PV) laminates and PCM macro-packed in plastic foil. Compared with a conventional asphalt shingle roof, the PV-PCM roof could have a reduction of 30% in heating loads generated by the roof in winter, and 55% in cooling loads during the summer period. Elarga et al. [27] installed hollow polycarbonate panels filled with PCM under a roof with clay tiles for a residential building located in Turin, Italy. Two types of PCM (RT28 and RT35) with nominal melting temperatures of 28 and 35 °C respectively were chosen. They found that the ongoing heat peak load could be reduced by 13–59% depending on the melting/solidification temperatures of the PCMs. Due to the mild summer temperatures in Turin, it was found that RT28 was more capable in reducing heat load. To maximise the benefits of using PCM, Elarga et al. [27] recommended the need of conducting numerical simulations in early design stages for selecting a suitable PCM. So far, very limited studies have been devoted to integration of form-stable PCM with building envelopes for energy conservation [28]. In particular, no study has been found in the literature to integrate FSPCM into gypsum board ceiling. The present study is a first attempt to develop a foamed gypsum board product with MeSA/diatomite composite FSPCM and use it for cooling load reduction of buildings in summer periods. A small-scale test chamber with FSPCM-integrated gypsum board ceiling was prepared accordingly and tested under real environmental conditions. The thermal/energy performance of the model room was then evaluated using empirical formulas. a rather new type of organic PCM, which exhibit high heat storage capacities (typically above 180 J/g) and melting congruency, and have the advantage of either easily impregnating, or directly incorporating, into porous supporting materials [5,6]. They show a phase transition within a small temperature range and are significantly less flammable compared to paraffinic PCMs. Fatty acid esters are manufactured by direct esterification of fatty acids separated from renewable sources of vegetable oils (such as soybean, coconut, or palm oils) and fats that are non-pollutant for the environment. These bio-based PCMs have the potential to absorb, store and release enormous amounts of heat energy with no or insignificant supercooling. Because they are fully hydrogenated, the bio PCMs remain stable for many years and many thousands of melt-freeze cycles free of oxidation threat [7]. It should be pointed out that leakage of molten PCMs and low thermal conductivity are the two main issues that limit their practical applications. Form-stable or shape-stabilised PCMs (FSPCMs), made by encapsulating PCMs into porous supporting materials, have therefore been developed to prevent leakage issues while also increasing their thermal properties [5,6]. Fatty acid esters possess excellent surface tension in their liquid state (27–32 mN/m), which is enough to allow them to be retained in porous carrier material pores [8]. There have been limited studies, however, on the use of esters to prepare composites for thermal energy storage applications [5,6]. Different solid–liquid fatty acid esters have different melting temperatures and latent heats. As a type of fatty acid ester, methyl stearate (MeSA) is a favourable candidate for TES applications at mild temperatures due to its moderate phase transition temperature (36.8 °C) and high heat storage capacity (217.7 J/g). Moreover, methyl esters are manufactured with reduced heat energy exhaustion than fatty acids and have superior properties in contrast to parent fatty acids [9]. This ester compound can therefore be considered as a suitable PCM in the fabrication of formstable PCM composites for use in TES applications, such as gypsum board ceilings. Diatomite is a sedimentary rock mineral that has been used as carrier material to prepare form-stable PCMs by many researchers. Diatomite has unique properties including large specific area, highly porous three-dimensional structure (up to 80–90% porosity), low bulk density (128–320 kg/m3), fairly low price, superior thermal stability and chemical inertness even at elevated temperatures [10,11]. Xu and Li [12] prepared paraffin/diatomite composite with a maximum PCM load of 47.4 wt%. The measured melting temperature and latent heat were found to be 41.1 °C and 70.5 J/g, respectively. Li et al. [13] studied the effect of particle size of diatomite on thermal properties of PCM composites. They absorbed paraffin into the pores of three different grades of diatomite particles to produce form-stable PCM composites. The PCM amounts retained by the diatomite through the direct incorporation technique were found to be 50%, 38% and 35% for particle sizes of 11.7, 1097.7 and 1295.7 μm, respectively. Karaman et al. [14] developed form-stable composite PCMs through a vacuum impregnation technique, utilising polyethylene glycol (PEG) as a PCM and diatomite as the carrier substance. Their approach allowed a mass of 50% PCM to be confined by diatomite and the composite is free of PCM leaking. The latent heat of the PEG/diatomite composite reached 87.1 J/g. Wen et al. [15] prepared fatty acid eutectics/diatomite composites as shape-stabilised PCMs based on eutectic capric-lauric acid (CA-LA) blends. They found that the latent heat of the composite material impregnated with a mass fraction of 53.6% of CA-LA blends was 87.3 J/g. It should be noted that FSPCMs have also been developed by using other porous carrier materials, such as perlite, expanded graphite, gelator, vermiculite and kaolinite, etc. [16]. PCMs have been integrated into envelope components to improve thermal performance and energy efficiency of buildings. Researchers found that the thermal performance of PCM-integrated envelope depends on many factors, such as the PCM melting temperature [17], heat storage and transfer performance of the PCM [18] and local climate [19]. Athienitis et al. [20] reported a maximum reduction of 4 °C in the 2. Experimental procedure 2.1. Materials for preparing FSPCM The model room test was conducted in Penrith, Australia. Penrith, a suburb of Sydney, is frequently hit by heat waves and has a temperature record of 47.3 °C reached in January of 2018. It has been reported that the temperature in a roof attic could be even 15–20 °C higher than the outdoor air temperature [29,30]. Therefore, methyl stearate (MeSA) with a relatively high phase transition temperature of 36.8 °C was selected as PCM. MeSA (≥96% pure) was obtained from Sigma Aldrich Pty. Ltd., Australia. The diatomite sample used was sourced from Mount Sylvia Pty. Ltd., Queensland, Australia. Instead of being directly used as a MeSA carrier material, the diatomite particles were heat-treated at 120 °C for 24 h to remove all moisture. 2.2. Preparation of FSPCM To prepare the MeSA/diatomite composite, a certain amount of 2 Applied Energy xxx (xxxx) xxxx M.J. Abden, et al. 35 MeSA FSPCM Specific heat capacity (J/g.K) 30 25 20 15 10 5 0 0 10 20 30 40 Temperature (°C) 50 60 Fig. 1. Specific heat capacity of MeSA and FSPCM. Fig. 2. Test setup for thermal performance measurement. MeSA was first placed in a glass beaker and heated at 60 °C with continuous stirring for 30 min to decrease its viscosity. Then, the diatomite was added gradually in portions of 5% weight of MeSA at a time until no MeSA trace was viewed on the surface of the beaker. The obtained form-stable PCM was used for further characterisation. It was found that the densities of MeSA and FSPCM are 813.8 and 416.3 kg/m3, respectively. Tests were also conducted to measure their specific heat capacity by using a differential scanning calorimeter (DSC) and the results are shown in Fig. 1. It was found that their specific heat capacity at ambient temperature (25 °C) are 1.84 and 1.32 J/g K, respectively. 2.4. Accelerated thermal cycle test The thermal reliability of the prepared FSPCM was explored with regard to phase transition behaviour after a 50-cycle thermal test. The thermal cycle comprised of exposing the composite to a consecutive melting and solidifying step through a thermal cycler (TC-25/H, Bioer). The FTIR and DSC experiments before and after cycling give a demonstration of chemical stability and thermal reliability of the FSPCM during the charging and discharging processes. 2.5. Heat storage and release performance test 2.3. Characterisation of FSPCM To inspect the thermal performance of the prepared FSPCM in practical applications, the heat storage and release periods of the pure MeSA and FSPCM were examined by measuring the temperature with respect to time during the melting and solidifying processes. The experimental setup is depicted in Fig. 2. Two glass tubes were loaded with 15 g of MeSA and MeSA/diatomite composite, respectively. A thermometer with a temperature accuracy of ± 0.1 °C was located in the centre of each tube. The glass tubes were then put into a water-filled glass beaker (Fig. 2) and the temperature was recorded once every 1 min during the melting process from 25.5 to 45 °C, and then during the subsequent solidification process back to room temperature. The external surface morphology and microstructure of pure diatomite and MeSA/diatomite composite were examined by scanning electron microscopy (SEM) JEOL 6510LV, which was also equipped with an energy-dispersive X-ray spectroscopy (EDS) detector for elemental analysis. The SEM was operated in a low vacuum mode at 30 Pa and 15 kV. The chemical composition of the diatomite determined by EDS is given in Table 1, which shows that the diatomite particle is mainly composed of SiO2 (92.5%) and a small amount of Al2O3 (4.8%). This indicates that the diatomite can potentially combine with inorganic construction materials such as cementitious materials well after absorbing PCM. The specific surface area and pore structure of the raw diatomite were examined by a physisorption analyser (ASAP 2020, Micromeritics). The chemical identity of the FSPCM was gauged by Fourier transform infrared (FTIR) spectroscopy Bruker Vertex 70 between the wave lengths of 4000 and 360 cm−1 and X-ray diffraction (XRD) Bruker AXS with a Cu-Kα (1.5406 Å) radiation. Differential scanning calorimeter (DSC) was used to measure the thermal properties of the samples. Three repeated DSC measurements were conducted for each sample in an inert Ar gas atmosphere at 25 mL/min flow and 5 °C/ min heating–cooling rates. Thermogravimetric analysis (TGA) was done using a Netzsch STA 449C Jupiter instrument under a constant argon gas stream [31]. A hot-disc thermal constant analyser (TPS 2500 S) was used to determine the thermal conductivity of the MeSA and FSPCM at room temperature. 2.6. Preparation and performance evaluation of FSPCM gypsum board The FSPCM described as above was incorporated into foamed gypsum board for latent heat storage. Gypsum is mainly composed of calcium sulphate hemihydrate (CSH). To reduce the density of gypsum board, foam bubbles were introduced into CSH slurry in the laboratory. The mix proportions for gypsum board with or without PCM are given in Table 2. A slurry was prepared by combining CSH, 0.225% accelerator (potassium sulfate), 0.65% glass fibres, 0.015% retardant (polyacrylic acid), 0.065% foaming agent (alkyl sulfate oligomers) and sufficient water to produce gypsum board with a water/CSH ratio of 0.86. It should be noted that all additives were measured in solid form as a percentage of the weight of CSH. The experimental process started by weighing the CSH and FSPCM (if any) in a beaker, and the water and retardant were then added and weighed in a mixer cup. The CSH/ FSPCM was then poured into the mixer cup and blended for 30 s to produce a gypsum slurry. Then, foam was generated separately by the inclusion of potassium sulfate before adding into the gypsum slurry. The mixture was blended for 30 s to produce foamed gypsum. Finally, the foamed gypsum was poured into a mould of 250 × 200 × 13 mm and cured at 90 °C for 48 h. The produced gypsum board without PCM is shown in Fig. 3a and the one with FSPCM is shown in Fig. 3b. Table 1 Chemical composition of diatomite. Compound SiO2 Al2O3 MgO Fe2O3 CaO Na2O TiO2 Ratio (%) 92.5 4.8 0.7 1.2 0.5 0.2 0.1 3 Applied Energy xxx (xxxx) xxxx M.J. Abden, et al. Table 2 Mix proportions of gypsum board with or without PCM. Sample Gypsum board FSPCM gypsum board a b c CSH (g) 715 511 FSPCM (g) − 204 Water (ml) Non-water components (g) Gauge water Foam water Accelerator 257 184 358 256 1.61 1.15 a Glass fibres Retardant 4.6 3.3 0.11 0.08 b Foaming agent c 0.46 0.33 Potassium sulfate. Polyacrylic acid. Alkyl sulfate oligomers. were used to measure the temperature variations at two different locations in a chamber (internal surface of the gypsum board and centre of the chamber), and the temperatures were recorded every 10 s by using a data logger (PCE-T 1200). Fig. 3c and d shows the test setup used to evaluate the thermal/ energy performance of gypsum board. The experimental design consists of two similar chambers with an external dimension of 290 × 240 × 20 mm and an internal dimension of 230 × 180 × 165 mm. The walls and floors of the chambers were made from expanded polystyrene insulating material. The gypsum board with PCM was used as the cover of one chamber to check its thermal regulating performance as a ceiling. For comparison purposes, the other gypsum board without PCM was used to cover the other chamber. The two chambers were placed outside from 26 to 28 January 2019 (summer season in Sydney) and exposed to natural solar heating and cooling processes during day and night time. Two T-type thermocouples 3. Results and discussion 3.1. Characterisation of diatomite material 3.1.1. SEM analysis SEM images of typical morphology of diatomite are presented in Fig. 4. As shown in Fig. 4a–d, the isolated diatomite particles, in the Fig. 3. Produced gypsum board and test setup for thermal/energy performance evaluation: (a) gypsum board without PCM; (b) gypsum board with FSPCM; (c) photo of test setup; and (d) sketch of test setup. 4 Applied Energy xxx (xxxx) xxxx M.J. Abden, et al. Fig. 4. SEM morphologies: (a-c) cylindrical shape, and (d) disc shape. order of microns, are mainly in disc and cylindrical forms with extremely porous and hollow centres. Numerous macropores (300–400 nm in diameter) are uniformly distributed, signifying the large specific surface area of diatomite as expected. They also have two different 3D cylindrical porous structures: one is full frustule with both ends entirely closed (Fig. 4b) and the other is half frustule with one end half open and the other end closed (Fig. 4c). The length of a typical diatomite particle is found to be in the range of 8–12 μm with a diameter of 5–8 μm. In general, the diatomite has smooth surface with various open pores, cavities and channels. All diatomite particles show a similar hierarchical porosity characteristic. The open pores in diatomite are mainly responsible for absorbing liquid PCM. As the used diatomite has high purity, only a few pores are blocked by visible impurities. The more pores are accessible, the more PCM can be loaded in the diatomite. (a) 0.07 Adsorption Desorption Differential pore volume (cm3/g) Volume adsorbed (cm3/g) 40 3.1.2. Surface properties The surface area and porosity properties of the thermally treated diatomite were analysed by nitrogen adsorption measurements. Fig. 5a shows the resulting adsorption–desorption isotherm and Fig. 5b demonstrates the corresponding density functional theory (DFT) pore size distribution graph of the thermally treated material. The isotherms feature an adsorption–desorption hysteresis loop in the relative pressure range of 0.45–0.96. It can be observed from Fig. 5a that the isotherm of the diatomite sample resembles a type II isotherm, as there is no plateau when it approaches the p/p0 ratio of 1, where p and p0 are the condensation pressure and saturated pressure of the bulk fluid, respectively. According to IUPAC nomenclature [32], macropores are defined as pores having diameters greater than 50 nm, mesopores 2–50 nm, and micropores less than 2 nm, respectively. The presence of macropores has been confirmed by the pore sizes observed in the SEM images shown in Fig. 4. The existence, however, of a small amount of uptake at low relative pressure could suggest the filling of micropores. 30 20 10 0 (b) 0.06 0.05 0.04 0.03 0.02 0.01 0 0 0.2 0.4 0.6 0.8 Relative pressure (p/po) 1 0 20 40 60 80 Pore diameter (nm) Fig. 5. Pure diatomite: (a) N2 adsorption-desorption isotherms, and (b) DFT pore size distribution. 5 100 120 Applied Energy xxx (xxxx) xxxx M.J. Abden, et al. interactions between MeSA and the supporting diatomite material. The physical interactions induce capillary and surface tension forces, arresting the liquid MeSA from leaking out during the phase transformation process [33]. The FTIR results shown in Table 3 indicate excellent chemical compatibility between MeSA and the diatomite matrix. To further characterise the diatomite, pure MeSA and MeSA/diatomite composite, X-ray diffraction (XRD) investigations were performed and the spectra are shown in Fig. 6b. The broad diffraction peak found in the diatomite spectrum spanning a range of approximately 20–25° can be ascribed to the non-crystalline silica (SiO2), whereas the strong peak centred at 26.7° can be assigned to the (1 0 1) reflection of quartz crystal. In the MeSA spectrum, the appearance of peaks at 11.1, 14.8, 19.9, 20.6, 21.8 and 24.1° confirms the formation of highly crystalline MeSA with a monoclinic system. It can be seen that the main diffraction peak of MeSA appears in the XRD pattern of MeSA/diatomite composite without appreciable shifting of the 2θ angle. Meanwhile, there is no new peak appearance. This supports that the MeSA segment of the composite has a similar crystal structure as the pure MeSA, without chemical interactions taking place between the MeSA and diatomite during the impregnation process. Also present is a hysteresis loop (0.5 < p/p0 < 0.9) which seems to have attributes similar to both an H3 hysteresis (the absorption capacity rises steeply at a higher relative pressure > 0.95) and H4 hysteresis (shows some micropore filling at low relative pressures). These hysteresis types could indicate the presence of macropores which have not been fully filled with pore condensate, or the existence of some mesopores. It is highly likely that this porous diatomite material possesses a microstructure with an array of pore sizes including micro-, meso- and macropores. The higher specific surface areas attributable to the mesopores are projected to influence interfacial processes. The BET surface area and pore structure characteristics are key factors to select porous carrier materials for wrapping PCM. The diatomite has a BET surface area of 22.69 m2/g and DFT pore volume of 0.04362 cm3/g, respectively. Due to the nature of the experiment, it should be noted that this pore volume only includes pores with diameters up to 120 nm (i.e. those shown in the DFT plot in Fig. 5b). The large surface area and pore volume explains the high absorption capacity of diatomite owing to capillary and surface tension forces, making it suitable for encapsulation of MeSA. 3.2. Characterisation of MeSA/diatomite composite 3.2.2. Morphology and thermal properties analysis The morphology of the MeSA/diatomite composite PCM is shown in Fig. 7. As shown in Fig. 7a, the composite PCM has a light grey colour. Compared with the empty pores observed in the SEM images of the pure diatomite (Fig. 4), it is clear that MeSA has been absorbed into the pores of the diatomite in the composite, as shown in Fig. 7b. For form-stable composite PCM, it is important to ensure the highest amount of PCM absorption at a minimum loss of the latent heat, since carrier materials do not undergo phase change. The phase change properties of MeSA and MeSA/diatomite composite were determined by DSC tests. The DSC curves of MeSA and composite are presented in Fig. 8a and the resultant data of thermal properties are summarised in Table 4. Both the raw MeSA and composite PCM display only one peak in the endothermic and exothermic DSC curves, corresponding to the latent heat storage of MeSA during melting and solidifying processes. As shown in Fig. 8a, the raw MeSA displays sharp and strong endothermic and exothermic peaks with an onset melting temperature (Tm) of 36.8 °C and an onset solidifying temperature (Ts) of 32.7 °C, giving a total temperature range of approximately 26–46 °C. Conversely, the MeSA/diatomite composite PCM has broader and blunt endothermic and exothermic peaks with a negligible change of phase change temperatures (Tm of 36.5 °C and Ts of 33.1 °C) in comparison with MeSA. The phase change latent heat of the MeSA either in its pristine form, or contained in the composite material, is calculated by the area under the endothermic and exothermic peaks. The melting (ΔHm) and solidifying (ΔHs) latent heats of pure MeSA are determined to be 217.7 and 3.2.1. Chemical compatibility analysis The phase analysis of MeSA and diatomite was examined via FTIR and XRD analyses. The FT-IR spectra of different samples (raw materials and the composite) are demonstrated in Fig. 6a. The raw diatomite possesses three strong absorption peaks at 454, 780 and 1062 cm−1. The characteristic peak at 454 cm−1 belongs to the SieO bending vibration. The band at 780 cm−1 ascribes to the silanol (SiOeH) groups and 1062 cm−1 signifies the siloxane (SieOeSi) stretching vibration group. All these characteristic peaks belong to SiO2. In the pristine MeSA spectrum, the absorption bands at 2848 and 2915 cm−1 are attributed to the symmetric and asymmetric stretching vibrations of –CH2 group. The stretching vibration of –CH3 is found at 2951 cm−1. The stretching vibration of C]O group is detected at 1739 cm−1. The peaks at 1169 and 1107 cm−1 can be assigned to CeO asymmetric and symmetric vibrations. In addition, the peaks at 1462 cm−1 and 723 cm−1 are caused by the CH2 or CH3 deformation vibration and (–CH2–)n (n ≥ 4) rocking vibration. In the MeSA/diatomite composite spectrum, all typical absorption bands of both MeSA and diatomite are still visible as expected. Moreover, no new peaks have appeared in the composite PCM spectrum, suggesting that no chemical interaction takes place between the pore confined MeSA and diatomite. Table 3 shows the characteristic FTIR peaks of the diatomite, MeSA, and MeSA/diatomite composite. It seems that some FTIR peaks of the composite PCM have shifted slightly in comparison with those of MeSA and diatomite. These shifts may be because of the weak physical MeSA/diatomite * MeSA Diatomite Quartz * (b) MeSA/diatomite Intensity (a.u.) Transmittance (%) (a) * 15 20 25 30 MeSA Diatomite 3860 3360 2860 2360 1860 1360 860 Wavenumber (cm−1) 10 360 20 30 40 2 (degree) 50 Fig. 6. Diatomite, MeSA and MeSA/diatomite composite: (a) FTIR spectra, and (b) XRD patterns. 6 60 Applied Energy xxx (xxxx) xxxx M.J. Abden, et al. Table 3 FTIR characteristic peaks of diatomite, MeSA and MeSA/diatomite composite. Sample SieO –CH2 SiOeH SieOeSi CeO Diatomite 454 − 780 1062 − MeSA MeSA/diatomite − 452 723 723 − 779 − 1056 1107 1106 ∆Hm,com × 100% ∆Hm,MeSA ∆Hcom × 100% ∆HMeSA β − − − − − − 1169 1165 1739 1740 1462 1462 2848 2848 2915 2916 2951 2951 ∆T = Tm − Ts (3) where ∆T is the degree of supercooling; and Tm and Ts denote the onset melting and solidification temperatures, respectively. The degree of supercooling of the FSPCM is displayed in Fig. 8b, where the temperatures of Tm and Ts are given in Table 4. The phase transformation temperature of MeSA in the MeSA/diatomite composite is influenced by the supporting diatomite material. As shown in Fig. 8b, the Tm value of the FSPCM decreases, but on the contrary Ts increases, hence minimising the degree of supercooling of the FSPCM to a great extent. This result is consistent with previously reported studies on polyethylene glycol/polymethyl methacrylate composite PCMs [36]. The heterogeneous nucleation effect of diatomite pores is liable for the decline of Tm. However, the crystallisation-promoting effect of the diatomite pores would cause an increase of Ts. As seen in Table 4, the supercooling degree of the FSPCM decreases by 0.7 °C (~17%) compared with that of pure MeSA. This result indicates that employing diatomite as a supporting material has a favourable influence on reducing the supercooling degree of MeSA. Table 5 presents the comparison of thermal properties between the MeSA/diatomite FSPCM prepared in the present study and other composite PCMs reported in the literature [14,15,37–43]. It is worth noting that the current FSPCM has significantly higher phase change latent heat (> 110 J/g) compared with other composite PCMs (60–90 J/g). Hence, the developed MeSA/ diatomite composite is a promising PCM material to be used in solar heat storage systems. (1) where ∆Hm,MeSA and ∆Hm,com represent the melting latent heats of the raw MeSA and the MeSA contained in the composite material, respectively. The β-ratio of MeSA in the FSPCM is found to be 51.3%, which is very close to the maximum ratio reported in the literature (see Table 5). This value is also close to that measured by TGA (discussed in the following section). However, the reduction in latent heat for the MeSA/ diatomite composite PCM in comparison with pure MeSA might not be ascribed only to the lower fraction of MeSA in the composite material. The interaction between the MeSA and porous diatomite could be another reason leading to the reduction in phase change latent heat of the FSPCM. These interactions could be reflected by crystallisation fraction (Fc) in the composite [35]. The parameter Fc is represented by Eq. (2): Fc = –CH2 and –CH3 the control of supercooling degree of a PCM is fundamental for TES applications. The degree of supercooling (ΔT) can be evaluated by Eq. (3): 219.6 J/g, respectively. The melting and solidifying latent heats for MeSA contained in the composite are 111.8 and 110.6 J/g, respectively. Evidently, both the melting and solidifying latent heats of the MeSA decrease in the presence of a carrier material. Therefore, the mass fraction of MeSA is important for understanding the thermal performance of the prepared MeSA/diatomite material. The mass fraction (β) of MeSA in the composite can be calculated by Eq. (1) [34]: β (MeSA%) = C]O (2) where ∆HMeSA and ∆Hcom are the crystallisation latent heats of the MeSA and MeSA/diatomite composite, respectively. The variable Fc corrects the mass depletion effect owing to the addition of porous diatomite and may assist to evaluate the interaction between the MeSA and the carrier material. A greater value of Fc implies higher conservation of crystalline phase and thus signifies a weaker interaction. In the current study, a high Fc value of ~ 98% for the MeSA/diatomite composite suggests high preservation of crystallinity and a very weak interaction between the MeSA and diatomite. It is well known that supercooling of PCMs is a key issue from the point of practical applications. A high supercooling degree is associated with low effective heat capacity. As a result, the stored energy during the crystallisation process reduces, which is not desirable. Therefore, 3.2.3. Thermal stability analysis Thermal stability is critical for a FSPCM to be used in thermal regulation. TGA and derivative TG (DTG) analyses were conducted to investigate the thermal stability of MeSA and MeSA/diatomite composite, and the results are given in Fig. 9. Both pure MeSA and composite PCM show only one-step degradation (Fig. 9a). The sharp mass loss of the pure MeSA and the MeSA/diatomite composite (Fig. 9b) occurred at around 346 and 318 °C, respectively. The mass loss can be attributed to the decomposition of organic matter, namely the breaking Fig. 7. Prepared MeSA/diatomite composite PCM: (a) normal appearance, (b) SEM micrograph. 7 Applied Energy xxx (xxxx) xxxx M.J. Abden, et al. 6 38 (a) (b) Melting point Freezing point 37 Temperature ( C) Heat flow (W/g) 4 Heating 2 0 -2 Cooling MeSA MeSA/diatomite -4 20 30 40 Temperature ( C) 50 The extent of supercooling (°C) 35 4.1 3.4 MeSA FSPCM 34 33 -6 10 36 32 60 Fig. 8. Thermal properties of the phase change composite: (a) DSC curves, (b) phase transition temperature and degree of super-cooling of composite FSPCM. 0.57% and 1.3%, respectively, whereas the latent heats for the melting and solidification processes changed by 0.18% and −0.36%, respectively (Table 6). The phase change temperatures and latent heat values for both melting and solidification processes of FSPCM after thermal cycling exhibited a negligible change, suggesting very good thermal reliability. The XRD pattern after thermal cycles is presented in Fig. 10b. It is observed that the peak position of MeSA in the MeSA/diatomite composite remains similar to that before 50 melting/solidifying cycles. The β (mass fraction of MeSA) value is also nearly the same as the original value (Table 6), which indicates a retention of crystalline phase of MeSA in the FSPCM after numerous melting-freezing cycles. Besides, no leakage of the MeSA is found by weighing the mass fraction in the composite (Table 6) before and after 50 thermal cycles, signifying excellent recyclability and thermal stability. It seems that the multiporous structure of the diatomite arrested the seepage of melted MeSA from the FSPCM by capillary action and surface tension forces. In addition, the FTIR spectra (Fig. 10c) of the MeSA/diatomite composite confirm that no changes of the shape and frequency values of major peaks occurred and no new peaks appeared after thermal cycles. These results again confirm that after 50 melting/solidification cycles, the chemical structure of the FSPCM is not affected. The thermal reliability and chemical stability of the prepared FSPCM thus meet the latent heat storage application requirements. Table 4 Thermal properties of MeSA and diatomite/MeSA composite. Sample MeSA MeSA/diatomite Heating cycle Cooling cycle Tm (°C) ΔHm (J/g) Ts (°C) ΔHs (J/g) Extent of supercooling (∆T = Tm − Ts ) (°C) 36.8 36.5 217.7 111.8 32.7 33.1 219.6 110.6 4.1 3.4 of the MeSA chains. The earlier mass loss in the composite PCM compared to that of the pure MeSA may be due to the change of physical behaviour of the pore-confined MeSA. But no decomposition and mass loss are detected for MeSA/diatomite composite PCM at temperatures below 220 °C, suggesting that the thermal stability of this prepared FSPCM is high in its temperature range of phase change. The total mass loss of MeSA/diatomite composite is 52% (Fig. 9a). This mass fraction of MeSA in the composite generally agrees with the measured mass fraction value of 51.3% found from previous DSC analysis. 3.2.4. Thermal reliability analysis Thermal reliability is a key parameter for evaluating the durability of a FSPCM. To verify the reversibility of FSPCM, solid–liquid phase change cycling tests were performed. Samples were characterised before and after 50 cycle tests. The resulting DSC curves in Fig. 10a show that the phase transition temperatures for melting and solidification after 50 thermal cycles are 36.7 and 33.5 °C respectively, with the corresponding latent heats found to be 112.0 and 110.2 J/g, respectively. As compared to those values measured before the thermal cycles, the melting and solidifying temperatures of the FSPCM were shifted by 3.2.5. Thermal behaviour of FSPCM composite Thermal storage and release rates are also important for practical applications of PCM. It is favourable for PCM to have relatively high thermal storage/release rates. Thermal storage and release processes of MeSA and MeSA/diatomite composite were studied by the same Table 5 Comparison of thermal properties between the current FSPCM and those reported in the literature. Composite PCM Paraffin/diatomite Capric-stearic acid/perlite Dodecanol/bentonite PEG/diatomite Paraffin/vermiculite Paraffin/perlite Capric-palmitic acid/pumice Lauric acid/kaolinite Capric-lauric acid/diatomite a Methyl stearate/diatomite a Maximum PCM ratio (%) 47.3 50 32 50 38.5 − 35 48.0 53.6 51.3 Melting process Freezing process Reference Tm (°C) Hm (J/g) Ts (°C) Hs (J/g) 27.1 29.6 22.6 27.7 27.0 27.6 23.1 43.7 23.6 36.5 89.4 82.1 67.6 87.1 77.6 67.1 56.5 72.5 87.3 111.8 26.5 17.4 21.1 − 25.1 23.6 21.7 39.3 22.5 33.1 89.9 82.6 62.3 − 71.5 69.1 55.4 70.9 86.9 110.6 Vacuum impregnation method 8 [37] [39] [42] [14] [40] [43] [41] [38] [15] This paper Applied Energy xxx (xxxx) xxxx M.J. Abden, et al. 120 10 (a) (b) Deriv. weight (%/ C) 100 Weight (%) 80 52% 60 40 20 MeSA MeSA/diatomite Diatomite 0 0 -10 318 °C -20 MeSA MeSA/diatomite Diatomite 346 °C -30 -40 -20 0 100 200 300 400 Temperature (°C) 500 0 600 200 400 Temperature (°C) 600 Fig. 9. Diatomite, MeSA and MeSA/diatomite composite: (a) TGA and (b) DTG curves. custom-designed experimental system shown in Fig. 2. Fig. 11 shows the curves of temperature versus time of the MeSA and composite samples during the heating and cooling processes. It is notable that both MeSA and MeSA/diatomite composite have an obvious temperature plateau during the heating and cooling processes because of phase transition. However, the composite PCM displays much higher heat storage/release rates than those of the MeSA. In the melting process (Fig. 11a), for example, the times required to raise the temperature from 37 to 39 °C are 1815 and 1550 s for the Table 6 DSC data of the FSPCM before and after 50 thermal cycles. Sample PCM (%) Before cycles After cycles 51.3 51.4 Heating cycle Cooling cycle Tm (°C) ΔHm (J/g) Ts (°C) ΔHs (J/g) 36.50 36.71 111.8 112.0 33.10 33.54 110.6 110.2 2 (a) 1.5 (b) Before cycle test After cycle test 0.5 Intensity (a.u.) Heat flow (W/g) 1 0 -0.5 -1 -1.5 Before cycle test After cycle test -2 -2.5 10 20 30 40 Temperature (°C) 50 60 10 20 30 40 2 (degree) 50 60 Transmittance (%) (c) Before cycle test After cycle test 3360 2360 1360 Wavenumber (cm−1) 360 Fig. 10. The MeSA/diatomite composite after 50 thermal cycles: (a) DSC curves, (b) XRD patterns, and (c) FTIR spectra. 9 Applied Energy xxx (xxxx) xxxx M.J. Abden, et al. 50 50 (a) (b) MeSA MeSA/diatomite 45 40 Temperature (°C) Temperature (°C) 45 MeSA MeSA/diatomite 35 30 25 40 35 30 25 20 20 0 1000 2000 3000 Time (s) 4000 5000 0 500 1000 1500 Time (s) 2000 2500 Fig. 11. Temperature versus time curves of MeSA and MeSA/diatomite composite: (a) melting process and (b) freezing process. temperature of 40 or 70 °C, which is above the melting temperature of the MeSA. Any leakage of MeSA would leave oily stains on the filter paper. After the testing, it was found that PCM leakage occurred for the sample containing pure MeSA, as shown in Fig. 13f. The PCM leakage is more obvious at 70 °C. To clearly show the PCM leakage, some stained areas were marked in the subfigure. In contrast, no visible trace of PCM leakage was found for the sample with FSPCM. It is also confirmed that there is no weight difference of the filter paper before and after the test. From the experiments, it can be concluded that the use of diatomite can eliminate leakage of MeSA in gypsum board. MeSA and MeSA/diatomite composite, respectively. In the freezing process, times of 1140 and 470 s are needed to drop the temperature from 37 to 35 °C (Fig. 11b) for the MeSA and composite, respectively. The results indicate that the melting and solidification times of FSPCM decrease by 14.6% and 58.8%, respectively, in comparison with those of the pure MeSA. These results demonstrate the beneficial effect of the diatomite on the MeSA’s heat storage/release performance for practical applications. The enhanced heat storage/release rates can be attributed to the higher thermal conductivity of the MeSA/diatomite composite compared to the pure MeSA, which can be seen from the comparison of thermal conductivity in Fig. 12. In comparison with the pure MeSA, the thermal conductivity of the FSPCM is considerably enhanced, with an increase as high as 63.7%. This suggests that the use of diatomite could enhance the thermal conductivity of MeSA, which is beneficial for thermal management applications. 3.2.7. Thermal regulating performance of FSPCM gypsum board The thermal performance of the gypsum board with or without FSPCM was evaluated by examining the indoor air temperatures of the miniaturised test chambers for three consecutive days (26–28 January 2019). The environmental temperatures are depicted in Fig. 14a and the measurement started at 12:00 am (midnight) on 26 January 2019. As day 1 was clear, day 2 partly cloudy, and day 3 overcast, the measured peak temperature decreased from 46.3 °C in day 1 to 38.7 °C in day 3. The variations in the interior surface temperature of the ceiling and the indoor air temperature of the test chamber are presented in Fig. 14b and c, respectively. As can be seen from Fig. 14b, the peak temperatures of the inside surface of the FSPCM ceiling are significantly lower than those of the reference ceiling without PCM. The differences are 8.9, 8.8 and 3.4 °C in days 1, 2 and 3, respectively. Meanwhile, as shown in Fig. 14c, the peak air temperature inside the chamber with a FSPCM ceiling is also reduced by an average of 3.5 °C in three days and a maximum of 4.9 °C in day 1 when compared with the chamber with normal gypsum board ceiling. The results indicate that the use of FSPCM ceiling is advantageous for building space cooling, thus keeping a comfortable indoor thermal environment. This is consistent with previous studies on PCM gypsum board. For example, Sari et al. [44] examined the thermal response of gypsum board with eutectic mixture of capric acid and stearic acid in a laboratory scale test room, and reported that the peak indoor temperature was reduced by 1.3 °C. In another study, a reduction of 3.1 °C in the peak indoor temperature was reported by Karaipekli et al. [45]. Compared with test results reported in the literature, it seems the current gypsum board with FSPCM is more effective for building space cooling, which might be due to its high phase change latent heat (> 110 J/g) and the high environmental temperatures. It should be noted that the heat is mainly transferred from the internal surface of the ceiling to the indoor air of the test chamber through natural convection. The convective heat flux of the interior surface of the gypsum board ceiling can be calculated using Eq. (4) [46]: 3.2.6. Stability of MeSA and FSPCM in gypsum The stability of pure MeSA and FSPCM in gypsum was characterised through leakage tests, where MeSA (Fig. 13b) or FSPCM (Fig. 13c) was mixed with calcium sulphate hemihydrate (CSH) (Fig. 13a) and a suitable amount of water. In the fresh slurry, a layer of liquid MeSA appeared on the top of the mixture, as shown in Fig. 13d. However, no leakage of MeSA was found when FSPCM was used, as evidenced in Fig. 13e. MeSA leakage might also occur in hardened gypsum board during the phase change process. To prove this, hardened samples with pure MeSA (Fig. 13f) and FSPCM (Fig. 13g) were placed on filter paper and put in an oven. The samples were kept in the oven for 6 h at a Thermal conductivity (W/m.K) 0.5 0.403 0.4 63.7% 0.3 0.246 0.2 0.1 0 MeSA FSPCM Fig. 12. Thermal conductivities of MeSA and composite FSPCM. 10 Applied Energy xxx (xxxx) xxxx M.J. Abden, et al. Fig. 13. Stability of form-stable PCM composites in gypsum board: (a) CSH, (b) liquid MeSA, (c) FSPCM, (d) fresh slurry with pure MeSA, (e) fresh slurry with FSPCM, (f) hardened sample with pure MeSA, and (g) hardened sample with FSPCM. qconv = h natural (Ts − Ta ) indoor air temperature is kept at 26 °C, the daily energy consumption (in watts) of the test chamber can be calculated by Eq. (5) [50]. (4) 2 where qconv is the convective heat flux (W/m ); Ts and Ta (in °C) are the internal surface temperature of the gypsum board and the inside air temperature of the test chamber, respectively; and h natural is the convection heat transfer coefficient (CHTC) of the ceiling. The value of CHTC is highly dependent on the surface to air temperature difference [47]. An average CHTC value of 4.8 W/m2K was determined for ceilings with a surface to air temperature difference between 0.5 and 5 °C [48]. Therefore, the h natural value is taken as 4.8 W/m2K in this paper as the surface to air temperature difference of the chamber is in this range. The influence of FSPCM on the calculated convective heat flux of the interior surface of the gypsum board ceiling is shown in Fig. 14d. The FSPCM gypsum board has sharp negative and positive heat flux peaks while the normal board only has positive peaks. Meanwhile, a reduction in the positive heat flux fluctuation amplitude and a delay in the heat flux peak are evident for the FSPCM gypsum board. The results suggest that the FSPCM absorbed solar radiation and converted the radiation into thermal energy to improve the indoor thermal environment. Q= ∑ j=1 n (Ui Ai + 0.33NV ) 24 (Tj − Ti ) if Tj > Ti 1000 η (5) 2 where Ui and Ai are the heat transfer coefficient (W/m K) and area of the ith element of the building envelope (m2), respectively; N is the air changes per hour (ACH) in h−1, taking as 0.25 h−1; V is the volume of the house (m3), η is the efficiency of the cooling system, taking as 2.5 [46], Tj is the mean daily indoor temperature (°C) in day j and Ti is the designed indoor air temperature in summer, taking as 26°C. In the calculation, the U-values are taken as 0.86 W/m2K for the expanded polystyrene wall, 4.25 W/m2K for the gypsum board ceiling, and 3.89 W/m2K for the FSPCM gypsum board ceiling. These values are determined based on the material conductivities following the procedure presented in [50]. The energy saving fraction (ESF) can then be determined from Eq. (6). Q ESF (%) = ⎛1 − FSPCM ⎞ × 100 QRef ⎠ ⎝ ⎜ 3.2.8. Effect of FSPCM on cooling load The application of FSPCM gypsum board as ceiling of a building might reduce the energy consumption for cooling in hot summer periods. To determine the possible energy saving for using FSPCM gypsum board ceiling, the energy consumption to cool the test chamber is compared with that to cool the other chamber with reference gypsum board. The calculations of energy consumption are conducted using empirical formulas [49,50]. The evaluation is conducted based on the three-day weather conditions shown in Fig. 14a, as they represent hot and sunny days, hot and partly cloudy days and overcast days. If the ⎟ (6) where QRef and QFSPCM are the energy consumptions of the reference and FSPCM chambers, respectively. The saving in energy with respect to the space cooling of the test chambers can be clearly seen in Table 7. The total amount of energy consumed to cool the internal space of the reference chamber is 0.0790 kWh during the three days. However, the energy consumption is reduced to 0.0662 kWh if FSPCM gypsum board is used. Therefore, the net energy saving is 0.0043 kWh/day, or a saving of 16.2% in cooling energy for the use of FSPCM gypsum board ceiling. 11 Applied Energy xxx (xxxx) xxxx M.J. Abden, et al. 50 (a) Environment temperature Temperature (°C) 45 40 35 30 25 20 0 8 16 24 32 40 Time (h) 48 56 64 72 Fig. 14. Measured data for gypsum board roof ceiling: (a) environmental temperatures from 26–28 January 2019, (b) inner surface temperatures of gypsum board, (c) inside air temperatures of the test chamber, and (d) heat flux of inner surface of gypsum board ceiling. CFSPCM = Ce DQFSPCM Table 7 Energy consumption and energy saving fraction. Day 1st day 2nd day 3rd day Energy consumption (kWh) Reference chamber FSPCM chamber 0.0267 0.0312 0.0211 0.0214 0.0251 0.0197 Energy saving/day (kWh) Average energy saving/day (kWh) ESF (%) 0.0053 0.0061 0.0014 0.0043 16.2 for reference chamber (7b) where CRef and CFSPCM are the annual cooling costs of the reference and FSPCM chambers, respectively; Ce is the cost of electricity in AUD $/kWh, taking as 33.33 c/kWh [54]; D is the number of hot days every year to operate an air conditioning system for cooling a building interior, taking as 120 days in view of the climatic conditions in Penrith, Australia [55]; and QRef and QFSPCM are the daily energy consumptions of the reference and FSPCM chambers, respectively, determined by Eq. (5). To convert the total cooling cost over a life-time of n years into present value, the present worth factor (PWF) needs to be calculated using Eq. (8) [52]. 3.2.9. Economic feasibility of FSPCM gypsum board A life-cycle cost analysis (LCCA) is conducted to evaluate the economic feasibility of FSPCM gypsum board to be used as building false ceiling [51]. The analysis is carried out for the test chamber based on the measured environment temperatures in Fig. 14a. The interest rate (i), inflation rate (φ) and life-time (n) of gypsum board are considered in the analysis, where the present value is adjusted based on the method proposed by Cuce et al. [52] to account for the interest and inflation rates. The LCCA is conducted following the procedure adopted by Daouas et al. [53]. The annual cooling cost for a test chamber is calculated by Eq. (7). CRef = Ce DQRef for FSPCM chamber PWF = (1 + τ )n − 1 τ (1 + τ )n (8) where n is the lifetime of ceiling board, taking as 10 years; and τ is a factor related to the interest rate (i) and inflation rate (φ), taking as 1.5% and 2.25%, respectively [56]. According to Cuce et al. [52], the τ − factor is given by Eq. (9). τ= (7a) 12 i−φ (if i > φ) 1+φ (9a) Applied Energy xxx (xxxx) xxxx M.J. Abden, et al. Table 8 Results of life-cycle cost analysis. Item Capital cost (AUD$) Total life-cycle cost (AUD$) Net ERS (AUD$) Payback period (year) Reference gypsum board FSPCM gypsum board 0.20 0.45 10.33 8.93 1.4 1.7 τ= φ−i (if i ≤ φ) 1+i test results show that the peak air temperature inside the chamber with a FSPCM ceiling is reduced by a maximum of 4.9 °C in the first day and an average of 3.5 °C in three days when compared with the chamber with normal gypsum board ceiling. The use of FSPCM in gypsum board ceiling is economically feasible with cooling load savings of 16.2% and a payback period of 1.7 years. The developed gypsum board containing FSPCM can be easily used to replace normal building false ceiling for energy conservation in real buildings. It should be noted that this study only used a small-scale test chamber to evaluate the performance of FSPCM gypsum board. It is no doubt that the results from a test chamber provide useful information for evaluating the FSPCM’s performance in real buildings. However, successful use of the FSPCM in real buildings depends on many factors, such as properties and amount of FSPCM used, location of the FSPCM, building ventilation conditions, building design and orientation, local climate, and occupants' behaviour, etc. [58]. Further studies should be conducted to integrate the FSPCM into envelope of real buildings for energy conservation and evaluate the building performance under various conditions. Analysis can then be conducted to evaluate the lifecycle cost and energy savings of using FSPCM in real buildings. Direct measurement of energy saving should also be conducted in the future for using FSPCM. (9b) The life cycle cost of cooling (CT) for a chamber is given by Eq. (10). CT,Ref = Ci,Ref + CRef × PWF for reference chamber CT,FSPCM = Ci,FSPCM + CFSPCM × PWF for FSPCM chamber (10a) (10b) where Ci,Ref and Ci,FSPCM are the capital costs of gypsum board without and with FSPCM, respectively. In this study, Ci,Ref and Ci,FSPCM are taken as $4.00 and $9.00 per m2 respectively, which are estimated based on the wholesale material prices obtained from the suppliers. Hence, the energy-related savings (ERS) for using FSPCM gypsum board are calculated from Eq. (11). ERS = (CT,Ref − CT,FSPCM) (11) Finally, the payback period defined as the required time length to repay the cost of the FSPCM investment can be calculated from Eq. (12) [52]. Ci,FSPCM − Ci,Ref ⎞ Payback period = ⎛ × PWF ERS ⎝ ⎠ (12) A shorter payback period is usually more desirable for any investment, as it is the main determining factor for investment. The results of the life-cycle cost analysis are given in Table 8. According to the evaluation, the payback period by the inclusion of FSPCM is only 1.7 years. This is relatively short due to the effectiveness and low cost of the developed FSPCM. Li et al. [46] also conducted small chamber tests to assess the feasibility of incorporating PCM in glazed roof, where PCM (about 1.8 kg) was poured into the closed cavity between two panes of glass. They reported that the PCM melting temperature has a great influence on energy saving and payback period. Through an economic feasibility analysis, they found that the annual energy cost saving increases from CNY 5 to CNY 67 and the corresponding payback period decreases from 6.2 to 3.3 years, when the PCM melting temperature changes from 18 to 32 °C. As can be seen, the PCM payback period (a minimum of 3.3 years) in [46] is longer than the payback period (1.7 years) of the FSPCM developed in this study. Since the environmental temperatures and uses of PCM are different, it is difficult to make a direct comparison on payback period. However, two reasons may be primarily responsible for the lower payback period of the FSPCM gypsum board in this study: (1) The paraffin wax used in [46] is much more expensive (~AUD $6.43/kg) than most other types of PCM [57]. In contrast, MeSA used in this study is much cheaper, which is AUD $2.1/kg on a wholesale basis; and (2) MeSA has a higher melting temperature and larger latent heat than those of the PCM used in [46]. Declaration of Competing Interest The authors declare that there is no conflicts of interest. Acknowledgements The authors would like to acknowledge the support of Advanced Materials Characterisation Facility (AMCF) and the assistance of its staff for conducting the material tests. The authors would also like to thank Mount Sylvia Pty. Ltd. for the donation of diatomite used in this study. References [1] Yuan F, Li MJ, Ma Z, Jin B, Liu Z. Experimental study on thermal performance of high-temperature molten salt cascaded latent heat thermal energy storage system. Int J Heat Mass Transf 2018;118:997–1011. [2] Tang LS, Yang J, Bao RY, Liu ZY, Xie BH, Yang MB, et al. Polyethylene glycol/ graphene oxide aerogel shape-stabilized phase change materials for photo-tothermal energy conversion and storage via tuning the oxidation degree of graphene oxide. Energy Convers Manage 2017;146:253–64. [3] Lai CM. Virtual special issue-PCM applications in building environment. Build Environ 2016;96:303–4. [4] Pielichowska K, Pielichowski K. Phase change materials for thermal energy storage. Prog Mater Sci 2014;65:67–123. [5] Karaipekli A, Sarı A. Preparation and characterization of fatty acid ester/building material composites for thermal energy storage in buildings. Energy Build 2011;43(8):1952–9. [6] Sarı A, Biçer A. Thermal energy storage properties and thermal reliability of some fatty acid esters/building material composites as novel form-stable PCMs. Sol Energy Mater Sol Cells 2012;101:114–22. [7] Kosny J, Kossecka E, Brzezinski A, Tleoubaev A, Yarbrough D. Dynamic thermal performance analysis of fiber insulations containing bio-based phase change materials (PCMs). Energy Build 2012;52:122–31. [8] Doll KM, Moser BR, Erhan SZ. Surface tension studies of alkyl esters and epoxidized alkyl esters relevant to oleochemically based fuel additives. Energy Fuels 2007;21:3044–8. [9] Abhat A. Low temperature latent heat thermal energy storage: heat storage materials. Sol Energy 1983;30:313–32. [10] Losic D, Mitchell JG, Voelcker NH. Diatomaceous lessons in nanotechnology and advanced materials. Adv Mater 2009;21(29):2947–58. [11] Sun XW, Zhang YX, Losic D. Diatom silica, an emerging biomaterial for energy conversion and storage. J Mater Chem A 2017;5(19):8847–59. 4. Concluding remarks This study investigated the thermal/energy performance of gypsum board ceiling incorporating a composite form-stable phase change material (FSPCM) for cooling load reduction in summer periods. The FSPCM was prepared by directly impregnating 51.3 wt% methyl stearate into porous diatomite. The developed FSPCM has very good phase change properties and thermal stability. Foamed gypsum board was then produced by integrating 40 wt% FSPCM for potential use as building false ceiling. The efficiency of the FSPCM gypsum board for cooling load reduction was evaluated in a miniaturised test room. The 13 Applied Energy xxx (xxxx) xxxx M.J. Abden, et al. polyethylene glycol for heat storage. Phys Chem Chem Phys 2012;14(38):13233–8. [36] Zhang L, Zhu J, Zhou W, Wang J, Wang Y. Thermal and electrical conductivity enhancement of graphite nanoplatelets on form-stable polyethylene glycol/polymethyl methacrylate composite phase change materials. Energy 2012;39(1):294–302. [37] Jeong SG, Jeon J, Lee JH, Kim S. Optimal preparation of PCM/diatomite composites for enhancing thermal properties. Int J Heat Mass Transf 2013;62:711–7. [38] Song S, Dong L, Zhang Y, Chen S, Li Q, Guo Y, et al. Lauric acid/intercalated kaolinite as form-stable phase change material for thermal energy storage. Energy 2014;76:385–9. [39] Wei T, Zheng B, Liu J, Gao Y, Guo W. Structures and thermal properties of fatty acid/expanded perlite composites as form-stable phase change materials. Energy Build 2014;68:587–92. [40] Xu B, Ma H, Lu Z, Li Z. Paraffin/expanded vermiculite composite phase change material as aggregate for developing lightweight thermal energy storage cementbased composites. Appl Energy 2015;160:358–67. [41] Karaipekli A, Sarı A. Development and thermal performance of pumice/organic PCM/gypsum composite plasters for thermal energy storage in buildings. Sol Energy Mater Sol Cells 2016;149:19–28. [42] Sarı A. Thermal energy storage characteristics of bentonite-based composite PCMs with enhanced thermal conductivity as novel thermal storage building materials. Energy Convers Manage 2016;117:132–41. [43] Yao C, Kong X, Li Y, Du Y, Qi C. Numerical and experimental research of cold storage for a novel expanded perlite-based shape-stabilized phase change material wallboard used in building. Energy Convers Manage 2018;155:20–31. [44] Sarı A, Karaipekli A, Kaygusuz K. Capric acid and stearic acid mixture impregnated with gypsum wallboard for low-temperature latent heat thermal energy storage. Int J Energy Res 2008;32(2):154–60. [45] Karaipekli A, Sarı A, Biçer A. Thermal regulating performance of gypsum/ (C18–C24) composite phase change material (CPCM) for building energy storage applications. Appl Therm Eng 2016;107:55–62. [46] Li D, Wu Y, Zhang G, Arıcı M, Liu C, Wang F. Influence of glazed roof containing phase change material on indoor thermal environment and energy consumption. Appl Energy 2018;222:343–50. [47] Le Dréau J, Heiselberg P, Jensen RL. Experimental investigation of convective heat transfer during night cooling with different ventilation systems and surface emissivities. Energy Build 2013;61:308–17. [48] Delaforce SR, Hitchin ER, Watson DMT. Convective heat transfer at internal surfaces. Build Environ 1993;28:211–20. [49] Li J, Li X, Wang N, Hu Y, Feng R. Experimental research on indoor thermal environment of new rural residence with active solar water heating system and external wall insulation. Appl Therm Eng 2016;95:35–41. [50] Energy in buildings: calculating the total heat loss of a house. https://www.open. edu/openlearn/nature-environment/the-environment/energy-buildings/contentsection-2.4.1. (Accessed 14 November 2018). [51] Yu J, Yang C, Tian L, Liao D. A study on optimum insulation thicknesses of external walls in hot summer and cold winter zone of China. Appl Energy 2009;86(11):2520–9. [52] Cuce E, Cuce PM, Wood CJ, Riffat SB. Optimizing insulation thickness and analysing environmental impacts of aerogel-based thermal superinsulation in buildings. Energy Build 2014;77:28–39. [53] Daouas N, Hassen Z, Aissia HB. Analytical periodic solution for the study of thermal performance and optimum insulation thickness of building walls in Tunisia. Appl Therm Eng 2010;30(4):319–26. [54] Australian electricity prices; the cost of electricity in Australia per kWh. https:// gobulk.com.au/australian-electricity-prices/ (Accessed 04 June 2018). [55] Penrith climate, averages and extreme weather records. http://www. farmonlineweather.com.au/climate/station.jsp?lt=site&lc=67113. [56] Minutes of the monetary policy meeting of the reserve bank board. https://www. rba.gov.au/monetary-policy/rba-board-minutes/2018/2018-12-04.html. (Accessed 04 December 2018). [57] Waqas A, Ud Din Z. Phase change material (PCM) storage for free cooling of buildings—a review. Renew Sustain Energy Rev 2013;18:607–25. [58] Rodriguez-Ubinas E, Ruiz-Valero L, Vega S, Neila J. Applications of phase change material in highly energy-efficient houses. Energy Build 2012;50:49–62. [12] Xu B, Li Z. Paraffin/diatomite composite phase change material incorporated cement-based composite for thermal energy storage. Appl Energy 2013;105:229–37. [13] Li Xiangyu, Sanjayan Jay G, Wilson John L. Fabrication and stability of form-stable diatomite/paraffin phase change material composites. Energy Build 2014;76:284–94. https://doi.org/10.1016/j.enbuild.2014.02.082. [14] Karaman S, Karaipekli A, Sarı A, Biçer A. Polyethylene glycol (PEG)/diatomite composite as a novel form-stable phase change material for thermal energy storage. Sol Energy Mater Sol Cells 2011;95(7):1647–53. [15] Wen R, Zhang X, Huang Z, Fang M, Liu Y, Wu X, et al. Preparation and thermal properties of fatty acid/diatomite form-stable composite phase change material for thermal energy storage. Sol Energy Mater Sol Cells 2018;178:273–9. [16] Lv P, Liu C, Rao Z. Review on clay mineral-based form-stable phase change materials: preparation, characterization and applications. Renew Sustain Energy Rev 2017;68:707–26. [17] Bontemps A, Ahmad M, Johannès K, Sallée H. Experimental and modelling study of twin cells with latent heat storage walls. Energy Build 2011;43(9):2456–61. [18] Saffari M, de Gracia A, Fernández C, Cabeza LF. Simulation-based optimization of PCM melting temperature to improve the energy performance in buildings. Appl Energy 2017;202:420–34. [19] Soares N, Gaspar AR, Santos P, Costa JJ. Experimental evaluation of the heat transfer through small PCM-based thermal energy storage units for building applications. Energy Build 2016;116:18–34. [20] Athienitis AK, Liu C, Hawes D, Banu D, Feldman D. Investigation of the thermal performance of a passive solar test-room with wall latent heat storage. Build Environ 1997;32:405–10. [21] Kuznik F, Virgone J, Roux JJ. Energetic efficiency of room wall containing PCM wallboard: A full-scale experimental investigation. Energy Build 2008;40(2):148–56. [22] Kuznik F, Virgone J. Experimental assessment of a phase change material for wall building use. Appl Energy 2009;86(10):2038–46. [23] Kuznik F, Virgone J, Johannes K. In-situ study of thermal comfort enhancement in a renovated building equipped with phase change material wallboard. Renew Energy 2011;36(5):1458–62. [24] Zhou T, Darkwa J, Kokogiannakis G. Thermal evaluation of laminated composite phase change material gypsum board under dynamic conditions. Renew Energy 2015;78:448–56. [25] Schossig P, Henning H, Gschwander S, Haussmann T. Micro-encapsulated phasechange materials integrated into construction materials. Sol Energy Mater Sol Cells 2005;89(2–3):297–306. [26] Kośny J, Biswas K, Miller W, Kriner S. Field thermal performance of naturally ventilated solar roof with PCM heat sink. Sol Energy 2012;86(9):2504–14. [27] Elarga H, Fantucci S, Serra V, Zecchin R, Benini E. Experimental and numerical analyses on thermal performance of different typologies of PCMs integrated in the roof space. Energy Build 2017;150:546–57. [28] Zhou G, Zhang Y, Wang X, Lin K, Xiao W. An assessment of mixed type PCM-gypsum and shape-stabilized PCM plates in a building for passive solar heating. Sol Energy 2007;81(11):1351–60. [29] Fallahi A, Durschlag H, Elliott D, Hartsough J, Shukla N, Kosny J. Internal roof and attic thermal radiation control retrofit strategies for cooling-dominated climates. U.S Depart Energy 2013. [30] Nielsen A, Morelli M. Measured temperature and moisture conditions in the roof attic of a one and a half story house. Energy Procedia 2017;132:789–94. [31] Afroze JD, Abden MJ, Islam MA. An efficient method to prepare magnetic hydroxyapatite-functionalized multi-walled carbon nanotubes nanocomposite for bone defects. Mater Sci Eng C Mater Biol Appl 2018;86:95–102. [32] Sing KSW, Everett DH, Haul RAW, Moscou L, Pieroti RA, Rouquerol J, et al. Reporting physisorption data for gas/solid systems. Pure Appl Chem 1985;57:603–19. [33] Yang J, Qi GQ, Liu Y, Bao RY, Liu ZY, Yang W, et al. Hybrid graphene aerogels/ phase change material composites: thermal conductivity, shape-stabilization and light-to-thermal energy storage. Carbon 2016;100:693–702. [34] Liang W, Chen P, Sun H, Zhu Z, Li A. Innovative spongy attapulgite loaded with ncarboxylic acids as composite phase change materials for thermal energy storage. RSC Adv 2014;4(73). [35] Wang C, Feng L, Yang H, Xin G, Li W, Zheng J, et al. Graphene oxide stabilized 14