Thermal Performance Test of a Phase-Change-Material Cool Roof System by a Scaled Model

General cool roof is effective for reduction of cooling load, but it has a problem of increasing heating load. ,erefore, the purpose of this study is to complement the disadvantages of the cool roof system by utilizing phase change characteristics of phase change material (PCM). ,e study was carried out to verify the thermal performance of the PCM cool roof system by measuring the temperature on the top and bottom of the PCM cool roof system by making a miniature model (600× 600× 600mm). PCM was inserted and not inserted, and the temperature difference according to the finish color (brown and white) was compared. As a result, the plate surface temperature using PCM was lower than that without PCM, and time-lag of temperature increase occurred. As a result of the comparison of temperature according to the finish color (brown and white), white showed a low temperature distribution up to 16.35°C. Even at room temperature, white maintained a low temperature distribution of 5.40°C than brown. ,e use of PCM cool roof system in roof finishes could lower the surface temperature and keep the room temperature low.


Introduction
A cool roof can increase the reflectivity and emissivity of a roof's surface, which results in a lower surface temperature compared to general roof finishes.Cool roofs are effective at cooling in the summer because they minimize the solar heat gain on the roof.However, they increase the heating load in winter [1].Various studies have been conducted to overcome this drawback.One solution is to apply phase-change materials (PCMs) to cool roofs [2][3][4][5][6][7][8][9].
A PCM can change phase from solid to liquid or liquid to solid at different temperatures.
is characteristic can be used to store and emit heat energy as a form of latent heat [10].A phase-change cool roof can prevent over-heating of the roof surface during summer and without increasing the heating load in winter.e decreased variation in the indoor temperature can make occupants feel more comfortable throughout the year.
is study demonstrates the improvements in energy savings and comfort using a PCM cool roof system.

Literature Review
e roof of a building accounts for approximately 14% of the total cooling load in a general residential house, which means that a large amount of energy could be saved through the roof [11].Chou et al. [12] evaluated the thermal performance of a general roof, an insulated roof, and an insulated PCM roof.ey measured the surface temperatures at the upper and lower sides of the roof and the indoor temperature.e insulated PCM roof exhibited the lowest temperature, whereas the general roof without insulation resulted in a higher indoor temperature.
Karlessi et al. [13] conducted a study on reducing the surface temperature by coating tiles with paint containing a PCM. e surface temperatures of the tiles with six different colors were analyzed, and the surface temperatures were reduced by 3-7 °C when the PCM was applied, depending on the color.Pasupathy and Velraj [14] installed PCM panels on the roof of a testing room with dimensions of 1.22 × 1.22 × 2.44 m 3 and measured the change in temperature at the ceiling surface.When PCM panels were not inserted, the indoor temperature of the ceiling surface was in the range of 23-30 °C.However, the range was 27-28 °C when the PCM panels were inserted, indicating a relatively constant temperature distribution.
Other studies on the application of PCMs to roof surfaces have also been conducted to reduce the occurrence of urban heat islands (UHIs).Roman et al. [15] combined roof nishes such as metal or concrete with PCMs to reduce the e ect of UHIs.Lu et al. [16] inserted a PCM into a PE-RT pipe to analyze the UHI reduction and the change in building energy e ciency in Tianjin, China.Chung [17] studied the reduction in surface temperature that resulted from applying according PCM tile, cool paint, green waterproof nish, and gray waterproof nish to a scale model.PCMs were sprayed over general tiles.Phase Change Energy Solutions in the USA applied "BioPCMatTM" [18] to the roofs of residential buildings.
ey reported total energy savings of approximately 30% during summer when BioPCMatTM was installed with a cool roof in a hollow layer.ese studies show that the temperature variation is stabilized when PCMs are applied to roof nishes.In particular, adding cool roof materials was more e ective for reducing the thermal load.

Methodology
To test the performance of a PCM cool roof, a scale model, WPC, and packing-type PCMs were fabricated.e WPC was found to be appropriate for the PCM cool roof system based on comparison tests with general wood in previous studies [10].A PCM cool roof system was fabricated, followed by a scale-model test.e study method and scope for each phase are as follows: Phase 1: fabrication of packing-type PCM Phase 2: fabrication of the PCM cool roof system by inserting packing-type PCM into a wood plastic composite (WPC) Phase 3: applying the fabricated PCM cool roof system to a scale model Phase 4: performance test using di erent roof surface colors (white and brown) in the scale model

Fabrication of Scale Models.
e scale models of a roof cross section from a small general o ce building were fabricated with a scale of 1/10 (600 × 600 × 600 mm 3 ).To prevent heat exchange through the walls and oors, 40 mm of insulation was used in the nish.e basic structure of the roof was made from deck plate, mortar, insulation, and concrete, as shown in Figure 1. is structure was referenced from general roof structures used in South Korea.Temperature sensors were installed on the upper and lower sides of the plate and on the lower side of the concrete at the indoor side.
e sensors were used to measure the heat transfer from the roof surface to each of the structural layers.

Fabrication of WPC and Packing-Type PCM.
e WPC was rst utilized to apply the PCM to the roof layer.e WPC consisted of a lower bottom plate, an internal hollow layer containing PCM, and an upper surface cover used as an exterior material.
e plate was designed to have modular size for application in various areas.e thermal conductivity of the WPC plate was measured using a thermal conductivity analyzer (TCi), which measures the degree to which a given amount of heat is supplied to the object and its thermal energy raises the temperature.
e mean value calculated from 10 measurements with the hollow layer included was 0.53 W/m•K (Table 1).Since the PCM can be liquid or solid at di erent temperatures, a container is needed to store it.A packing-type PCM was fabricated and injected into a transparent pack made of nylon to maximize the thermal storage performance without causing damage due to PCM leakage.Each pack was 15 × 3 cm 2 , and the weight of the injected PCM was 35 ± 1 g (Figure 2).

Test according to PCM Melting Point (26 °C/44 °C) (Test 1).
In Test 1, the WPC was installed on the upper surface of the scale model.Two di erent PCMs with di erent phasechange temperatures were tested: BioPCM of Bio series (melting point: 26 °C) and n-Para n series RT-PCM of R ® (melting point: 44 °C).A comparative test was then conducted using a WPC plate without the PCM (Table 2).
e test was conducted at the rooftop of the building in Chung-Ang University located in Heukseok-dong, Dongjakgu, Seoul, from September 22 to 24, 2015.e weather conditions had no interference from solar radiation.e mean cloud cover was 4, the mean outdoor temperature was 23.4 °C, and the mean wind speed was 1.9 m/s (Table 3 and Figure 3).e plate without PCM had the highest surface temperature, followed by Bio 26 °C PCM and RT 44 °C PCM, with a di erence of up to 7 °C (Table 4 and Figure 4).

Internal Temperature of the Plate.
e highest internal temperature of the plate was measured for the plate without PCM (52.75 °C), followed by Bio 26 °C PCM (50.00 °C) and RT 44 °C PCM (42.90 °C).Bio 26 °C PCM maintained its temperature for about two and a half hours, while RT 44 °C PCM maintained it for about six hours due to the time lag resulting from the PCM content (Table 5 and Figure 5).

Lower-Side Temperature of the Plate.
e lower-side temperatures in the plate showed the same trend as the      e temperature was lower in the morning and 1-2 °C higher at night than the internal temperature of the plate (Table 6 and Figure 6).6 Advances in Civil Engineering into the WPC and applied to a rooftop.e performance of PCM was measured for di erent surface nish colors (brown and white).Brown was selected because it is generally used in existing WPC products, and white was selected because the nish color of the cool roof is white (Table 8 and Figure 8).Advances in Civil Engineering e test was conducted on the rooftop of a building at Chung-Ang University in Heukseok-dong, Dongjak-gu, Seoul, from October 20 to 25, 2015.e weather conditions had no interference from solar radiation.e mean cloud cover was 2.3, the mean outdoor temperature was 16.6 °C, and the mean wind speed was 1.6 m/s (Table 9 and Figure 9).11).

Lower-Side Temperature of the Plate.
e highest lower-side temperature of the plate for the brown model was 36.10 °C and that of the white model was 22.60 °C, resulting in a temperature difference of approximately 13.50 °C.ere was no significant difference in the lower side temperature between the RT 44 °C and Bio 26 °C PCMs for the white models (Table 12 and Figure 12).

Indoor Ceiling Temperature.
e highest indoor ceiling temperature for the brown model was 29.40 °C and that of the white model was 24.00 °C, resulting in a temperature difference of approximately 5.40 °C.ere was no significant difference between RT 44 °C and Bio 26 °C PCMs for the white models (Table 13 and Figure 13).

Discussion
In this study, performance tests were conducted in which PCMs were combined with a cool roof to reduce the cooling load in summer and the heating load in winter.In the literature, the volume of the PCM inside the roof finish was large compared to the thickness of the roof.However, this study was unable to measure the thickness of the roof accurately due to the scale-model test.us, the amount of PCM could not be calculated precisely compared to the roof thickness, depending on the scale; thermal conductance changes in the linear manner and thermal mass changes in a cubic manner; and the miniature scale-model with specifications is not detailed enough to be applied to real buildings.However, the results of the scale model test clearly verified the thermal storage performance.When the WPC with PCM was fabricated, its thermal conductivity was measured as 0.55 W/m•K on average through a TCi analysis.
In the future, products made from stronger materials with lower thermal conductivity will be applied to reduce the surface temperature even further.

Conclusion
In this study, packing-type PCMs were inserted into WPC, and a scale-model test was conducted using different colors.e results are summarized as follows: (1) e PCM plates lowered the surface temperature more than the plate without PCM and exhibited a longer time lag.RT 44 °C PCM was more effective at reducing the surface temperature than Bio 26 °C PCM.(2) e white finish resulted in a lower temperature distribution by 13.50 °C to 16.35 °C than brown.Furthermore, the white finish maintained a 5.40 °C lower indoor temperature distribution than brown due to the difference in surface temperature.us, the scale-model test of the PCM cool roof demonstrated that cooling and heating loads of buildings could be decreased because the roof surface temperature can be reduced.

Figure 1 :
Figure 1: Scale-model cross section and temperature sensor locations.

Figure 6 :
Figure 6: Plate bottom temperature graph of Test 1.

Figure 12 :
Figure 12: Plate bottom temperature graph of Test 2.

Table 1 :
WPC thermal conductivity measurement results including the hollow layer.
4.1.1.Surface Temperature of the Plate.e surface temperatures of the plate without PCM was 54.40 °C, that with Bio 26 °C PCM was 51.75 °C, and that with RT 44 °C PCM was 47.40 °C.

Table 3 :
Average weather data of Test 1.
4.1.4.Indoor Ceiling Temperature.eindoorceilingtemperatureswere38.00 °C for Bio 26 °C PCM, 36.50 °C for RT 44 °C PCM, and 39.10 °C for the plate without PCM.e time when the maximum temperature of the indoor ceiling surface occurred showed that the plate without PCM was 16: 40, Bio 26 °C PCM was 17:20, and RT 44 °C PCM was 17:31.atis, a time lag phenomenon occurred in the case of the Bio 26 °C and RT 44 °C PCMs during cooling after sunset compared to the other specimens (Table7and Figure7).
°C PCM exhibited a lower surface temperature than the Bio 26 °C PCM. is was due to heat loss on the surface as a result of heat acquisition by the internal PCM. is result indicated that RT 4.2.Test by Finish Color: Brown and White (Test 2).In Test 2, the packing-type PCMs (Bio 26 °C/RT 44 °C) were lled

Table 9 :
Average weather data of Test 2.
4.2.1.Surface Temperature of the Plate.e highest surface temperature for the brown model was 38.85 °C and that of the white model was 23.70 °C, resulting in a temperature difference of approximately 15.15 °C.No significant difference in surface temperature was revealed between RT 44 °C and Bio 26 °C PCMs for white models (Table 10 and Figure 10).4.2.2.Internal Temperature of the Plate.e highest internal plate temperature for the brown finish was 38.15 °C and that for the white finish was 22.80 °C, resulting in a temperature difference of approximately 15.35 °C.e internal temperature of the white RT 44 °C model was 4.60 °C lower than that of the white Bio 26 °C model on average (Table 11 and Figure For the white models, the upper side temperature of RT 44 °C was 5.00 °C lower than that of Bio 26 °C, but there was no significant difference in the ceiling temperature.