Thermophysical characteristics and application of metallic–oxide based mono and hybrid nanocomposite phase change materials (NCPCMs) for thermal management systems.

This experimental study covers the chemical, physical, thermal characterization and application of novel nanocomposite phase change materials (NCPCMs) dispersed by TiO 2 , Al 2 O 3 , and CuO nanoparticles. A commercial-grade of paraﬃn, namely RT–35HC, was considered as a phase change material (PCM). The mono and hybrid NCPCMs were synthesized at a constant weight concentration of 1 . 0 wt.%. In the ﬁrst phase, various characterization techniques were used to explore the thermophysical properties and chemical interaction of mono and hybrid NCPCMs. In the second phase, the thermal cooling performance was investigated by ﬁlling the prepared NCPCMs in a heat sink at various input power levels. The results showed the uniform dispersion of TiO 2 , Al 2 O 3 , and CuO nanoparticles onto the surface of both mono and hybrid NCPCMs without altering the chemical structure of RT–35HC. The optimum latent–heat of fusion and highest thermal conductivity of 228 . 46 J/g and 0 . 328 W/m.K were obtained, respectively, of Al 2 O 3 +CuO dispersed hybrid NCPCM compared to pure RT–35HC. In comparison of RT–35HC, the increasing trend in speciﬁc heat capacity was observed of NCPCMs and 36 . 47% enhancement was obtained for hybrid NCPCM in solid–phase. The reduction in heat sink base temperature was achieved of 3 . 67%, 6 . 13%, 13 . 95% and 8 . 23% for NCPCM TiO 2 , NCPCM Al 2 O 3 , NCPCM CuO and NCPCM Al 2 O 3 + CuO , respectively, compared to RT-35HC. Further, no phase segregation, less subcooling, smaller phase transition temperature, higher chemical and thermal stability were observed with hybrid NCPCMs which can be used potentially for thermal management of electronic devices, Li–ion batteries and photovoltaic (PV) modules systems.


Introduction
Energy, Economy and Environment (3E) have the most powerful and urgency importance in current global issues and challenges [1].Carbon dioxide (CO 2 ) emissions, both natural and human sources, has been significantly increasing and contributing a major role in greenhouse and global climate change [2].Around 87% of CO 2 emissions are due to the human sources come from the utilization of fossil fuels coal, natural gas and oil [3].Nonetheless, the fossil fuels are still the dominant energy sources for power generation with their share increasing 66% in 2005 to 77% in 2050.Therefore, CO 2 emissions would increase by 61% over 2011 levels by 2050 [4].To minimize the CO 2 emissions, there is the critical need of alternative and sustainable energy resources to reduce the emerging demand of fossil fuels.
With the development of nanotechnology, the heat and mass transfer in nanoconned systems such as nanofluid [5,6] and nano-composites brings novel insight for energy conversion and storage [7,8] which make a contribution in thermal management applications.The thermal energy storage (TES) systems based on heat storage materials have been used extensively for energy conversion, storage and transportation in thermal management applications of electronic devices, photovoltaic (PV) modules, Li-ion batteries systems etc. [9,10,11].
Phase change materials (PCMs) as heat storage materials meet the series of requirements with higher energy storage density with suitable phase transition temperature range, good stability and low supercooling.For effective heat storage and heat transfer using PCMs, the latent-heat and thermal conductivity are the key thermal properties which improve the thermal management efficiency.Organic PCMs exhibit the best heat storage properties but they have low thermal conductivity which reduces the overall thermal response factor, heat transfer rate and efficiency.Thus, a PCM with higher latent-heat and thermal conductivity with stable chemical structure is the most suitable.To overcome this issue, researchers have introduced several heat transfer enhancement techniques including extruded metal-fins [12,13,14,15,16,17], metal-foam and porous materials [18,19], nanomaterials [20,21,22] encapsulated micro/nano-capsules [23,24,25,26].
Many researchers are working on different mechanisms for the thermal controllability of structures.Moreover, in addition to PCM there are some other materials that can be used for thermal management system as well as for energy harvesting systems having applications in mechanical and aerospace engineering [27,28].Nanomaterials as a supportive materials into the PCMs have taken keen attention in current era due their higher thermal properties and enhance the chemical and thermal stability.For instance, Babapoor et al. [29,30] synthesised the composite PCMs using SiO 2 , Al 2 O 3 , Fe 2 O 3 , ZnO nanoparticles and measured the thermophysical properties.They achieved the highest thermal conductivity with 8 wt.% of Fe 2 O 3 nanoparticles.However, the authors suggested that Al 2 O 3 nanoparticles were preferable for thermal management applications.Nourani and his co-authors [31,32] prepared NCPCMs using Al 2 O 3 nanoparticles of weight concentrations of 0.5, 5, 7.5, and 10 wt.% and SSL used as a surfactant to enhance the stability of Al 2 O 3 nanoparticles in paraffin wax.The effective thermal conductivity enhancement ratios were found of 31% and 13% in solid and liquid phases, respectively, at 10.0 wt.%.Chieruzzi et al. [33] synthesised the mono and hybrid composite PCMs of potassium nitrate (KNO 3 ), used as a PCM, by adding SiO 2 , Al 2 O 3 and hybrid of SiO 2 /Al 2 O 3 having 1.0 wt.%.The best results were obtained with SiO 2 nanoparticles with a decrease of 2 − 3 • C in onset temperatures and 9.5% increase in specific heat.Sharma et al. [34] used the TiO 2 nanoparticles of weight concentrations of 0.5, 1.0, 3.0, and 5.0 wt.% and palmitic acid as a PCM to prepare the NCPCMs.The highest thermal conductivity and lowest heat of fusion were obtained of 0.35 W/m.K and 180.03 kJ/kg, respectively, at 5.0 wt.% of TiO 2 .Yang et al. [35] prepared the paraffin and palygorskite mixed composite PCM and revealed the melting temperature and latent heat of 54-56 • C and 132.18J/g, respectively.Li et al. [36] synthesised the NCPCMs using calcium chloride hexahydrate (CaCl 2 6H 2 O) as a PCM with γ−Al 2 O 3 and measured the phase-change thermal properties.The maximum thermal conductivity was obtained of 1.373 W/m.K at 2.0 wt.% of γ−Al 2 O 3 nanoparticles.Sami and Etesami [37] prepared the NCPCMs of using TiO 2 and paraffin with mass fractions of 0.5, 0.7, 1, 2, 3 and 4 wt.% with and without SSL used as a surfactant.Results explored the optimum concentration of 3 wt.% of TiO 2 nanoparticles with the 47.8% increase in thermal conductivity with SSL compared to the pure paraffin.Further, authors suggested that addition of SSL in NCPCMs showed the better dispersion and thermal stability compared with pure paraffin and TiO 2 /paraffin composite.Bashar and Siddiqui [38] prepared the NCPCMs using various types of nanoparticles, Ag, CuO, Al 2 O 3 and MWCNTs and added in paraffin wax.
A higher heat transfer coefficient of 18% and 14% was obtained for CuO and Ag nanoparticles, respectively, compared with pure paraffin wax.Praveen and Suresh [39] prepared the composite PCMs using solid-solid neopentyl glycol (NPG) and CuO nanoparticles for thermal management application.It was found that maximum thermal conductivity and phase change enthalpy of 0.61 W/m.K and 112.4 kJ/kg, respectively, were obtained at 3.0 wt.% of CuO.Recently, Li et al. [40] synthesized MgO dispersed in eutectic carbonate salt of NaLiCO 3 and studied the thermal properties of light and heavy weight MgO nanoparticles.The results found that light weight MgO nanoparticles of particle size 3 − 5 µm had the maximum thermal conductivity of ∼ 5.5 W/m.K compared to heavy weight MgO nanoparticles of similar size.
The literature reveals that most of the studies focused on the development of NCPCMs used for the high temperature TES applications.The current study focuses on synthesising of new NCPCMs which will be the most suitable for thermal management of electronic devices, PV modules and Li-ion batteries, working under the safe and reliable operating temperature range of 30 − 40 • C [9,41].In addition, the literature indicates that a very little work has been performed on mono and hybrid NCPCMs.Therefore, the present study aims to improve on the thermophysical properties of a commercially available PCM to enhance the thermal performance.The microstructure, chemical and thermal properties are investigated of TiO 2 , Al 2 O 3 , and CuO nanoparticles dispersed mono and hybrid NCPCMs.Further, various characterization techniques were adopted to analysis the surface morphology, chemical and physical interaction, thermal properties including phase-change temperature, melting/solidifying latent-heat enthalpies, specific heat capacity, thermal conductivity, thermal stability and reliability.The newly developed NCPCMs were then poured in a heat sink to investigated the transient temperature thermal response to explore the thermal cooling performance.

Materials
In present study, the RT-35HC (a commercial grade of paraffin) was used as a PCM with phase-transition temperature of 34 − 36 • C, purchased from Rubitherm GmbH, Germany.
All materials were used as they received without further chemical treatment.

Preparation of nanocomposite PCMs
The schematic diagram of preparation of NCPCMs is shown in Figure 1.A well-precise two-step preparation method was carried out to synthesize the mono and hybrid NCPCMs, which has been used extensively in preparation of nanofluids [42].The RT-35HC was used a base PCM and TiO 2 , Al 2 O 3 and CuO nanoparticles were used as an additives.The purpose of adding the nanoparticles is to enhance the thermal properties.In first part of NCPCMs preparation, the RT-35HC was melted at a temperature of 70

Microstructural and chemical interaction analysis
An environmental scanning electron microscopy (ESEM, FEI Quanta-650) attached with the energy-dispersive X-ray spectroscopy (EDX) system was employed to examine the surface morphology, microstructure and to visualize the surface elemental distribution of nanoparticles in NCPCMs.Fourier-transform infrared spectroscope (FTIR, Bruker Tensor-

Thermophysical properties analysis
The TGA/DSC (SDT-Q600 TA instrument Inc., UK) (error ±0.1 µg) [44] was used to perform the thermogravimetric analysis (TGA) and derivative thermogravimetry (DTG) analysis of NCPCMs to measure the rate of weight loss and decomposition to address the thermal stability of NCPCMs.The aluminium pan was used filled with sample mass of 4 − 7 mg and heating rate was kept at 10 was used to perform the infrared thermography (IRT) tests to obtain the thermal images of samples.The samples were filled immersed in a water pot and constant temperature was provided through the hot-plate.

Thermal cooling performance analysis
The experimental setup to perform the thermal cooling performance analysis was carried out using a passive cooling thermal management system, as shown in Figure 3.The schematic diagram shows that thermal management system consists mainly the heat sink, data acquisition system, DC power supply and the laptop.The heat sink assembly consists of plate type silicon rubber heater (OMEGALUX), K-type thermocouples (OMEGA), rubber mat for insulation, perspex sheet to visualize the physical melt fractions of RT-35HC and NCPCMs.In the experimental set up, the heat sink assembly was perfectly insulated using a rubber pad, having very low thermal conductivity.The rubber pad was placed at the bottom and around the heat sink with a 50 mm extended width/thickness of the actual dimensions of heat sink.A constant temperature was measured at the outside of the insulation during the experiment which was the same with environment room temperature.
The top surface of the heat sink was covered with acrylic sheet having a thickness of 5 mm by keeping silicon gasket which prevent the PCM leakage, to visualize the physical melt font of molten PCM.A copper made heat sink was manufactured using computer numerical control (CNC) machine.Pure RT-35HC and NCPCMs were poured inside the heat sink.To ensure the proper stability, reduce the aggregation and uniform distribution of nanoparticles, each sample of NCPCMs was sonicated before pouring into the heat sink.
A DC power supply made of Keysight Technologies (N6700C, 0.5 A/100 V) was used to provide different power levels of 3-5 W at ambient temperature, to mimic heat generation.
A PC based digital data acquisition system (Agilent 34972A, USA) was used to record the temperature variation through thermocouples inserted at different points on the heat sink.
The data acquisition system was connected to the laptop and Agilent Technologies, Inc.
(Taft Ave.Loveland, USA) software is used to record the temperature at interval of 5 s during the experimentation.The uncertainty of the measured quantities is associated with the least count of the corresponding measurement devices.The uncertainty in temperature measurement according to according to ASTM standard [47] in a temperature range of 0- The uncertainty in heat input are calculated using Kline and McClintock [48] technique and minimum and maximum uncertainties are obtained of 1.68% and 1.91%, respectively.

ESEM and EDX analysis
The microstructure and surface morphology of mono and hybrid NCPCMs analysed by the ESEM are presented in embedded enough to show their presence in NCPCMs.Besides, the homogenous and heterogeneous percolation of nanoparticles in RT-35HC can be observed from LFD and BSED images.From Figure 4, the presence of zones characterized by the heterogeneous aggregates of nanoparticles and RT-35HC can be observed clearly.Similar observations of nanoparticles percolation in NCPCMs have been reported in Ref. [29,30,33].While solid-liquid phase transition of NCPCMs, the solvation, "Brownian motion" and "electrostatic repulsion" cause the nanoparticles to separate individually.However, this collision becomes more prominent between the nanoparticles by increasing the loading of the nanoparticles and Van der Waals force becomes stronger at a short range resulting in the formation of aggregates and sedimentation of nanoparticles at high concentration.Therefore, the uniform dispersion and percolation of nanoparticles depend on the size of the nanoparticles.
The EDX maps of NCPCM T iO 2 and NCPCM Al 2 O 3 +CuO are shown in Figure 5 and 6, respectively.Figure 5a, 5b and 5c show the distribution of carbon (C), oxygen (O) and titanium (Ti) elements, respectively, present in NCPCM T iO 2 .The presence of all elements C, Ti and O in different colours can be seen in single image of TiO 2 dispersed NCPCM T iO 2 in Figure 5d.
The EDX mapping of carbon (C), oxygen (O), aluminium (Al), and copper (Cu) elements of hybrid NCPCM Al 2 O 3 +CuO can be seen in Figure 6a, 6b, 6c and 6d, respectively.Figure 6e shows the presence of all C, Al, Cu and O elements in various colours in NCPCM Al 2 O 3 +CuO .
The ESEM and EDX images show the smaller zones of agglomeration of nanoparticles.The similar observations have been reported in previous studies [33,34].Consequently, it is recommended that for more better and uniform dispersion of the nanoparticles in PCM, the smaller size, low density and higher surface area nanoparticles are preferable [49,50].

FT-IR analysis
The chemical composition and structural interaction of TiO 2 , Al 2 O 3 and CuO dispersed NCPCMs were characterized by FT-IR spectroscopy.Figure 7  Furthermore, there is a small peak can be observed at 729 cm −1 which indicates the standard adsorption spectra pattern of the amorphous structure or disordered defects of γ−Al 2 O 3 .It can be revealed that both mono and hybrid NCPCMs presented the consistent peaks with that of the pure RT-35HC which means that NCPCMs contained similar characteristics of pure RT-35HC without emergence of new peak or any peak shift.This means that there no considerable chemical interaction occurred between the TiO 2 , Al 2 O 3 and CuO and RT-35HC and there is only physical interaction which indicate that prepared mono and hybrid NCPCMs are chemical stable.

XRD analysis
The crystallography analysis of TiO 2 , Al 2 O 3 , CuO, RT-35HC and NCPCMs were identified through XRD patterns to study the crystalline phase and dimensions of unit cell, , as shown in Figure 8.The diffraction peaks of 2θ at 25.3    Figure 10e illustrates the experimental and calculated latent-heat of fusion for NCPCMs, calculated using Equation 2 [56]:

TGA and DTG analysis
where ∆L N CP CM and ∆L P CM calculated and experimental latent-heat of fusion of NCPCM and PCM, respectively, ω and ϕ are the mass fraction of PCM and nanoparticles, respectively.It can be revealed that experimental values of latent-heat of fusion of mono and hybrid NCPCMs are lower than the calculated values.The deviation in results is due the type of the nanoparticles, since the loading of all nanoparticles in NCPCMs is constant.
The relative error (RE) of experimental and calculated values has been obtained 6 ∼ 9% for all NCPCMs, given in I n t e n s i t y ( a .u .)    L a t e n t h a e t ( J / g ) L a t e n t h e a t ( E x p . ) L a t e n t h e a t ( C a l .)    Figure 11b, as given by Equation 3: where, k N CP CM and k P CM are the thermal a conductivity of NCPCM and pure PCM, respectively.
Figure 11b presents an increasing trend in enhancement thermal conductivity are observed with change of temperature.A relative enhancements of 55.0%, 54.0% and 53.7% T e m p e r a t u r e ( °C ) nanoparticles have the smaller size and density and higher surface area compared to CuO nanoparticles.Therefore, they have more stronger degree of homogenization and dispersion rate into the RT-35HC.Thenceforth, the combination of Al 2 O 3 +CuO hybrid nanoparticles having different thermophysical properties reveal the finer dispersion stability.Further more, the boundary layer of thermal resistance of nanoparticles and lattice molecules of PCM contribute a major factor in variation of thermal conductivity results [58].Generally, there exists two major factors that cause the enhanced phenomenon of thermal conductance within the NCPCMs.Firstly, higher thermal conductivity of the nanoparticles and secondly the motion of the nanoparticles in NCPCMs in liquid-phase which causes a quasi-convection phenomenon [29,8].

Specific heat capacity analysis
The specific heat capacity results of RT-35HC and NCPCMs are shown in Figure 12 for solid and liquid phases as a function of temperature from 10 • C to 60 • C. Since the specific heat capacity has the less effect in total thermal energy storage due to low thermal energy storage density in sensible heat storage phase.Regardless, specific heat capacity influences additional effects in the total heat stored by a PCM within the storage temperature range of storage which is calculated using Equation 4 as follows: where, Q (J) is the quantity of total heat stored, T i ( • C), T m ( • C) and T f ( • C) are the initial, melting and final temperatures, respectively, m (kg) mass of the heat storage medium, C p (J/kg.K) is the specific heat capacity,α m is the faction of melted mass and ∆H m (J/kg) is the heat of fusion per unit mass.From Equation 4, it is important to investigate the specific heat capacity variations with temperature to analysis heat transfer and heat storage performance.
The specific heat capacity of NCPCMs increases gradually with temperature between 14 • C to 30 • C in solid-phase (Figure 12a) while it remains constant in liquid-phase (Figure 12b).The presented results of specific heat capacity of both phases showed the good agreement with the previous studies [33,59,60].The specific heat capacities of RT-35HC for solid and liquid phases are obtained of 1.88 and 1.77 J/g • C, respectively.It can be revealed that with addition of the nanoparticles, the specific heat capacity of NCPCMs increases both in solid and liquid phases.The higher values of specific heat capacities are obtained of 2.61 and  4. 398 Figure 13 represents the comparison of the comparison of specific heat capacities for RT-

399
35HC and NCPCMs for 25  The enhancement in specific heat capacity is associated with the following reasons: (i) an enhanced anharmonicity of the atomic interaction due to their volume expansion, (ii)  [61], (iii) the grain boundaries of nanosized materials which possess an excess volume with respect to the perfect crystal lattice and (iv) the high specific surface energies related to the high surface area of the nanoparticles per unit volume [62,63].In addition, the morphology of the nanoparticles also influences significantly, since the smaller size nanoparticles provide the larger interfacial surface area per unit mass between solid nanoparticles and surrounding material, thus increase the contribution of interfacial effects in the suspension [64,65,66].The interfacial interaction of solid-solid or solid-liquid may alter phonon spectrum or phonon vibration mode of nanoparticles near the surface area and therefore, change the specific heat capacity of NCPCM [67].The high specific surficial interface area of nanoparticles can adsorb liquid molecules to its surface which form the liquid layers.These liquid layers constrain the nanoparticles and alter their free-boundary surface atoms into the non-free interior atoms [62,67].Therefore, the variation in specific heat capacity of NCPCMs is because of the varied Gibbs free energy of the nanoparticles and liquid layers.

IRT analysis
The IR thermographic images of mono and hybrid NCPCMs are shown in Figure 14 which presents the surface temperature distribution during the melting process at different time steps.The melting stages of NCPCM Al 2 O 3 , NCPCM CuO and NCPCM Al 2 O 3 +CuO can be seen clearly at each time step, labelled as (X), (Y) and (Z), respectively, as shown in Figure 14a.The change in temperature of mono and hybrid NCPCMs can be observed from the temperature scale which indicates the lower to higher temperature variations from blue to red colours, respectively.The uniform melting of each sample is observed while heating at constant temperature.The quick flow away and shrinkage is observed in NCPCM CuO , shown in 14f.This phenomenon during phase change from solid-to-liquid causes to lose its shape-stability completely.In case of hybrid NCPCM Al 2 O 3 +CuO , uniform phase transition is observed with slight shrinkage and flow away due to the heating-impact effect.Since, the NCPCMs only contain the 1 wt.% of Al 2 O 3 +CuO, therefore only slight shrinkage is observed.It has been reported that the flow away and shrinkage level decrease with the increase of the loading of nanoparticles.The melting process of NCPCM Al 2 O 3 +CuO shows that it does not melt completely till at 35 min and surface temperature of hybrid NCPCM is lower than the mono NCPCM.This reveals that the best heat transfer performance due to thermal properties and thermal stability.The enhancement in hybrid NCPCM Al 2 O 3 +CuO is due to the uniform and homogeneous dispersion of Al 2 O 3 and CuO nanoparticles in RT-35HC, which increase the viscosity and shape-stability.Moreover, the thermographic images of NCPCM Al 2 O 3 +CuO shows the enhanced latent-heat phase duration and uniform natural convection heat transfer which can be used in thermal management systems for electronics.

Transient thermal analysis
The transient thermal analysis was carried out by investigating the melting and cooling processes of pure RT-35HC and NCPCMs, as shown in Figure 15 at a power level of 4 W.All samples were poured into a copper made heat sink after sonication to ensure the homogeneous dispersion of nanoparticles.The purpose of using the heat sink in current study, was to investigate the real time thermal response of NCPCMs compared to pure PCM.Since the heat sink has been widely used as heat exchanger that transfers the heat generated inside the electronics towards the ambient through a fluid medium.RT-35HC and NCPCMs were used as fluid medium inside the heat sink as passive cooling medium.Heat transfer occurs in sensible heat transfer mode because of temperature gradient and specific heat capacity of the samples before melting starts.In addition, the heat transfer depends on the thermal conductivity of the samples.The quicker heating rate was observed for NCPCMs because of higher thermal conductivity.Since the loading content is constant in each sample, thus the variation in average heat sink base temperature is very small.Specific heat capacity of each sample also contribute a major role in total heat transfer rate.
In second stage, NCPCMs started melting and a large amount of heat was absorbed because of latent-heat of fusion.Thus, the heat sink base temperature increased slowly which eventually enhanced the thermal performance of NCPCMs based heat sink, used for passive  The lower temperature rise of heat sink base is beneficial for the thermal management of electronic devices, Li-ion battery and PV modules.ing process was longer than the melting process.Also, phase change process of PCM and NCPCMs was more prolonged while cooling process than the melting process.The reason behind of such irregular melting and cooling phenomenon depends on the heating/cooling rate.While cooling process, the melting rate was higher than the cooling process which took the less time to change the solid-liquid phase than the cooling process while phase transition from liquid to solid phase.The results revealed that the addition of nanoparticles improve the cooling rate of PCM result in decrease the average temperature heat sink base.
Thus, NCPCMs filled heat sink improves the cooling efficiency and enhance the reliability of electronic devices.

Conclusions
The current experimental study explores the physical, chemical and thermal properties The current study finally concludes that newly synthesized metallic oxide nanoparticles based hybrid NCPCMs have the better improvement in the thermal energy storage properties compared to the mono NCPCMs, which ultimately enhance the rate of heat transfer in conjugate heat transfer mode.Thereby, hybrid NCPCMs can be successfully used in passive thermal management systems for electronic devices, Li-ion batteries, photovoltaics modules and direct utilization in solar thermal energy storage and distribution in natural environment.
100 • C with maximum variation of ±0.1 • C. The uncertainty in data logger is ±0.1 • C. The uncertainties in voltage and current are measured of ±0.01 V and ±0.001 A, respectively.
presents the FT-IR spectrum of nanoparticles, RT-35HC and NCPCMs and absorption peaks were observed.For instance, the three absorption peaks at 2955 cm −1 , 2913 cm −1 , 2849 cm −1 represent the medium C-H symmetrical and anti-symmetrical stretching vibration of methyl (-CH 3 ) and methylene (-CH 2 -) group.The peak at 1472 cm −1 identifies the medium C-H scissoring of -CH 2 -and deformation of antisymmetric stretching vibration -CH 3 group in RT-35HC.The FT-IR spectrum also shows a series of absorption peaks at 1371, 1125, and 891 cm −1 which are due to the C-H bending and scissoring vibrations.The peak at 715 cm −1 represents to the weak rocking vibration of C-H in long-chain methyl group.The FT-IR spectrum of TiO 2 shows the two peaks around 510 and 526 cm −1 which are attributed to the Ti-O stretching vibrations in TiO 2 .In case of CuO, there are two stretching bands at 880 cm −1and 1432 cm −1 which represent the characteristics bands of Cu-OH functional group[51,52].

Figure 9
Figure 9 presents the thermal gravimetric analysis of RT-35HC and hybrid NCPCMs to investigate their thermal stability.The thermal stability of NCPCMs is estimated on the onset temperature of PCM degradation and rate of weight loss at which is occurs.The TGA and DTG curves show the one-step and two-step thermal degradation process, respectively.The TGA curves of RT-35HC and NCPCMs, shown in Figure 9a, represent that there is no distinguishable weight loss in any sample up to ∼120 • C. The rate of weight loss increases with the increase of temperature and reaches at the maximum degradation temperature with constant residual.For RT-35HC, the initial disintegration starts at 217.70 • C and maximum degradation temperature is 255.01 • C with 1.85% charred residue.While decomposition process, caused by the evaporation of pure RT-35HC, the hydrocarbon chains breakdown into monomers (CH 3 CH:CH+CH 4 ).The onset decomposition and maximum degradation temperatures of all samples are summarized in Table 2 and in case of hybrid NCPCM Al 2 O 3 +CuO , the charred reside was 3.62% at 262.78 • C and remained constant subsequently.In NCPCMs, the nanoparticles create the protective layer on the surface of RT-35HC which delays the vaporization of RT-35HC during the thermal degradation.The DTG curves of RT-35HC and NCPCMs are shown in Figure 9b and revealed that addition of nanoparticles have improved the thermal stability, because no decompositions was observed in samples until ∼150 • C. Thus, hybrid NCPCM Al 2 O 3 +CuO can effectively be used for thermal management

4. 5 .
DSC analysisDSC analysis presents the phase-transition temperatures and latent-heat of melting/solidification of NCPCMs, as shown in Figure10.Figure10aand 10b present the endothermic and exothermic curves, respectively, of RT-35HC and NCPCMs and measured values of phasetransition thermal properties are summarized in Table3.It can be seen that the addition of nanoparticles have slightly changed the thermal properties of RT-35HC after synthesizing.The latent-heat of melting/solidification of NCPMCs were decreased with the addition of nanoparticles, as expected.A single endothermic peak is observed of RT-35HC and NCPCMs during melting, indicating an isomorphous crystalline form of RT-35HC either in a pristine state or in the NCPCMs one.Contrary, a bimodal crystallization behaviour is predicted of RT-35HC and NCPCMs by presenting a main exothermic peak temperature along with an additional peak of higher temperature during solidification process.This twophase transitions phenomenon is ascribed by the appearance of a metastable rotator phase prior to completing the complete crystallization as a results of the heterogeneous nucleation during the cooling process[8,54,55].The comparison of the melting (∆Hm ) and solidification (∆H s ) enthalpies of RT-35HC and NCPCMs are presented in Figure 10c.The ∆H m and ∆H s of RT-35HC are determined of 255.88 and 260.06 J/g, respectively.The reduction in ∆H m of NCPCM T iO 2 , NCPCM Al 2 O 3 , NCPCM CuO and NCPCM Al 2 O 3 +CuO has been obtained of 7.83%, 8.24%, 8.59% and 10.72% as compared to the pure RT-35HC.Similarly, the maximum reduction in ∆H s latent-heats are 7.98%, 7.08%, 9.47%, and 11.81% for NCPCM T iO 2 , NCPCM Al 2 O 3 , NCPCM CuO and NCPCM Al 2 O 3 +CuO , respectively.The reduction in latent-heats are due to the addition of nanoparticles and mass fraction (ω) of crystallized RT-35HC in NCPCMs is determined by the Equation 1: N CP CM and ∆H P CM are the endothermic latent-heat of NCPCM and pure PCM, respectively.While synthesising of NCPCMs, the nanoparticles replace the RT-35HC molecules resulting in reduce the latent-heat capacity of the NCPCMs which lead to the absorption/release of more energy.The degree of super-cooling (∆T ) of NCPCMs is presented in Figure 10d.The peak melting temperature (T m ) and crystallization temperature (T c ) of RT-35HC are determined as 36.09• C and 31.71• C, respectively.The maximum deviations in T m are determined as −0.30%, −1.14%, −0.97% and −0.08% for NCPCM T iO 2 , NCPCM Al 2 O 3 , NCPCM CuO and NCPCM Al 2 O 3 +CuO , respectively, compared to the RT-35HC.The slight variations are observed in T c which are due the crystallization confinement of the nanoparticles surface layers within the NCPCMs.This interfacial surface layers cause to form the imperfect RT-35HC resulting in a slight variation in T m .These two factors result in an increase in ∆T .The reduction in ∆T for NCPCM Al 2 O 3 +CuO is obtained of 4.32 • C, which shows the better significance hybrid nanoparticles as nucleating agents compared to the mono nanoparticles in terms of its surface adsorption and effective homogeneous nucleation.

20
t f l o w ( J / g ) T e m p e r a t u r e ( °C ) C M A l 2 O 3 + C u O N C P C M C u O N C P C M A l 2 O 3 N C P C M T i O 2∆H m∆H s P h a s e -c h a n g e e n t h a l p y ( J / g )

Table 3 :
Phase-change thermal properties of TiO , Al O , CuO and RT-35HC based mono and hybrid NCPCMs * .

Figure 11 presents
Figure 11 presents the results of thermal conductivity (k) and thermal conductivity enhancement factor (η) of RT-35HC and NCPCMs.It can be seen that by dispersing the nanoparticles thermal conductivity generally increases because of the higher thermal conductivity of nanoparticles.The average thermal conductivity of RT-35HC and NCPCMs is measured for solid-phase (20-30 • C), phase-change (∼ 35 • C) and liquid-phase (40-55 • C) regions with an increment of 5 • C. It can be evidenced that the constant trend is predicted in enhancement in thermal conductivity in solid and liquid phases.However, a sharp increasing and decreasing behaviour was predicted when the temperature increased from 30 • C to 35 • C and 35 • C to 40 • C, respectively.This behaviour reveals that thermal conductivity is strongly depended on the temperature.The phenomenon of sudden increase and decrease in thermal conductivity was due to the random motion of molecules within a disordered microstructure of lattice in liquid-phase.The maximum thermal conductivities of 0.340, 0.464, 0.461, 0.469 and 0.472 W/m.K were obtained at phase-change temperature (∼ 35 • C) for RT-35HC, NCPCM T iO 2 , NCPCM Al 2 O 3 , NCPCM CuO and NCPCM Al 2 O 3 +CuO , respectively, shown in Figure 11a.However, the maximum thermal conductivities for solid phase at 20 • C were obtained of 0.214, 0.320, 0.282, 0.331 and 0.328 W/m.K for RT-35HC, NCPCM T iO 2 , NCPCM Al 2 O 3 , NCPCM CuO and NCPCM Al 2 O 3 +CuO , respectively.The higher thermal conductivities at phase-change temperature (∼ 35 • C) is because of the phase-change temperature range 34-36 • C at which the NCPCM is in metastable condition.At phase-change temperature, the crystalline structure of RT-35HC becomes unstable and the increase in temperature accelerates the molecular vibration of in the lattice, thus thermal conductivity of RT-35HC and NCPCMs increases dramatically at the melting point (∼ 35 • C) [57].It can be clearly observed that NCPCM Al 2 O 3 +CuO has the highest thermal conductivity in all three phases.Furthermore, the percentage of η of RT-35HC and NCPCMs is shown in

Figure 11 :
Figure 11: The results of (a) thermal conductivity and (b) thermal conductivity enhancement factor of TiO 2 , Al 2 O 3 , CuO and RT-35HC based mono and hybrid NCPCMs as a function of temperature.

Figure 13 :
Figure 13: (a) Comparison of specific heat capacity, (b) specific heat capacity enhancement of TiO 2 , Al 2 O 3 , CuO and RT-35HC based mono and hybrid NCPCMs.

4. 9 . 1 .
Melting processThe melting curves of RT-35HC, NCPCM T iO 2 , NCPCM Al 2 O 3 , NCPCM CuO and NCPCM Al 2 O 3 +CuO are shown in Figures 15a.It can be seen from melting process that there is rapid increase of temperature linearly in all samples up to the first 20min without showing any phase change.

4. 9 . 2 .
Cooling processThe cooling curves of RT-35HC, NCPCM T iO 2 , NCPCM Al 2 O 3 , NCPCM CuO and NCPCM Al 2 O 3 +CuO are shown in Figure 15b.The transient temperature variation of heat sink base temperature can be divided into three different stages: pre-sensible cooling, latent-heat of solidification and post-sensible cooling.It can be observed closely that NCPCMs have the faster decrement in temperature compared to the pure PCM.The higher rate of temperature decrease reflects the higher thermal conductivity enhancement of NCPCMs due the addition of TiO 2 , Al 2 O 3 , CuO nanoparticles.During the cooling process, the base temperature decreased sharply and NCPCMs released their sensible heat to cool down up to their congealing temperature.Further, the NCPCMs started solidifying process due to their latentheat of solidification at ambient temperature through natural convection phenomenon.The minimum temperatures of 22.36 • C, 21.54 • C, 20.99 • C, 19.24 • C and 20.52 • C were recorded for RT-35HC, NCPCM T iO 2 , NCPCM Al 2 O 3 , NCPCM CuO and NCPCM Al 2 O 3 +CuO , respectively, after 150 min of cooling process.The reduction in heat sink base temperature was achieved of 3.67%, 6.13%, 13.95% and 8.23% for NCPCM T iO 2 , NCPCM Al 2 O 3 , NCPCM CuO and NCPCM Al 2 O 3 +CuO , respectively, compared to RT-35HC.It can be observed that cool-
This research is facilitated by the Faculty of Engineering, University of Nottingham, UK research infrastructure.The first author (Adeel Arshad) acknowledges the University of Nottingham for awarding him the Faculty of Engineering Research Excellence PhD Scholarship to pursue a Ph.D. research program.The authors acknowledge the use of facilities at Nanoscale and Microscale Research Centre of the University of Nottingham supported by Engineering and Physical Sciences Research Council [grant number EP/L022494/1].
NCPCMs were prepared and labelled as NCPCM T iO 2 , NCPCM Al 2 O 3 , NCPCM CuO and NCPCM Al 2 O 3 +CuO .Lastly, samples of NCPCMs were cooled to the ambient temperature.

Table 2 :
The temperature changes and residue of TiO 2 , Al 2 O 3 , CuO and RT-35HC based mono and hybrid NCPCMs.

Table 3
[36,8,56]ilar trend of deviations in experimental and calculated values latent-heat of fusion have reported in Refs.[36,8,56].The results reveal that NCPCM Al 2 O 3 +CuO has the optimum value of latent-heat which is preferable for thermal management systems.The discrepancies in latent-heat values are because of the surface morphology, structure, size and rate of dispersion of the nanoparticles while synthesising the NCPCMs.
• C and 50 • C for solid and liquid phases, respectively.Figure15a400 shows that variations in specific heat capacity are significantly depends on the type and spe-401 cific heat capacity of the nanoparticles.Additionally, the nanoparticles size, density, surface area and morphology contribute the most prominent effects in enhancement in specific heat capacity.The specific heat capacity enhancement factor (ζ) is calculated using Equation6to evaluate the effect of nanoparticles in enhancement of specific heat capacity: P N CP CM and C P P CM are the specific heat capacity of the NCPCM and PCM, respectively.The maximum enhancements are obtained of 38.62% is achieved for NCPCM Al 2 O 3 , whereas it is 36.47%for NCPCM Al 2 O 3 +CuO in solid phase compared to the RT-35HC.

Table 4 :
Coefficients of the third-order polynomials in solid-state, C p (J/g.• C).