Ei

Aa

b

a

ARRAA

KNRTT

1

awwami4mlrtrmr

U

h0

Energy and Buildings 138 (2017) 641–647

Contents lists available at ScienceDirect

Energy and Buildings

j ourna l ho me page: www.elsev ier .com/ locate /enbui ld

nhanced solar spectral reflectance of thermal coatings throughnorganic additives

o Zhou a, Zechuan Yu a, Cheuk Lun Chow a, Denvid Lau a,b,∗

Department of Architecture and Civil Engineering, City University of Hong Kong, Hong Kong, ChinaDepartment of Civil and Environmental Engineering, Massachusetts Institute of Technology, Cambridge, MA 02139, USA

r t i c l e i n f o

rticle history:eceived 12 August 2016eceived in revised form 5 November 2016ccepted 9 December 2016vailable online 10 December 2016

eywords:umerical simulationeflective coatinghermal performanceitanium dioxide

a b s t r a c t

Space cooling of buildings is responsible for a substantial portion of energy consumption and greenhousegas emission. In order to reduce the building energy consumption, reflective coatings have been adoptedextensively since they can effectively minimize undesirable solar energy absorption. In this paper, func-tional additives (i.e. titanium dioxide and hollow glass beads) are used to improve the reflective andinsulation properties of thermal coating materials. The thermal impact upon the application of differ-ent reflective coatings on concrete panels is experimentally examined. The experimental results haveshown that the coating containing titanium dioxide and hollow glass beads leads to a drop in interiorsurface temperature, up to 3.5 ◦C, implying that such coating can effectively reduce heat absorption andcooling load for buildings. The finite-difference time-domain simulation (FDTD) which can simulate thepropagation of electromagnetic waves is used here to investigate the reflection of electromagnetic waves

and to explore the working principles of additives. The numerical simulations demonstrate the reflectivebehavior in the coating material, showing that embedded TiO2 nanoparticles can significantly improvethe reflective performance of coating films. It is envisioned that a nontoxic coating with high level ofreflectance and insulation can be characterized and the working principles of different additives can berevealed.

© 2016 Elsevier B.V. All rights reserved.

. Introduction

Traditional buildings require a lot of energy for space heatingnd cooling, resulting in huge energy consumption accompaniedith greenhouse gas emission. The indoor temperature increaseshen the exterior surfaces are exposed to solar radiation. A large

mount of energy is consumed by air-conditioning system toaintain indoor thermal comfort. In the United States of Amer-

ca (USA), the air-conditioning system accounts for more than0% of the building energy consumption according to the Depart-ent of Energy in the USA [1]. In order to reduce the cooling

oad and building energy consumption, actions should be taken toeduce undesirable absorption of solar radiation [2–7]. At present,hree thermal insulation methods are commonly used, i.e. inte-

ior insulating materials, energy storage system with phase change

aterials and reflective coatings [8]. The interior insulating mate-ials may cause a reduction in room size, loss of thermal mass and

∗ Corresponding author at: Department of Architecture and Civil Engineering, Cityniversity of Hong Kong, Hong Kong, China.

E-mail address: denvid@mit.edu (D. Lau).

ttp://dx.doi.org/10.1016/j.enbuild.2016.12.027378-7788/© 2016 Elsevier B.V. All rights reserved.

interstitial condensation problem. Some interior insulating materi-als can even pose health risks [9]. Implementation of energy storagesystem is relatively complicated. Phase change materials have tobe encapsulated or stabilized in the building, and encapsulationmay cause structural problems like leakage and corrosion. Further-more, phase change materials may release toxic substances whenexposed to fire [10,11]. In comparison with the two aforementionedmethods, the reflective coating method is more convenient and reli-able. Reflective coatings are generally non-combustible and can beconveniently applied to the surface of building envelope throughroller or sprayer without affecting the properties of original build-ing [12–14]. The reflective coatings should be featured with highreflectance throughout the entire solar spectrum and minimal heattransfer to the coated object. A schematic diagram exhibiting a com-parison of buildings with and without reflective coatings is shownin Fig. 1(a). During hot seasons, the reflective coatings on buildingenvelopes protect the building from being heated by solar radia-tion, preserving a relatively low indoor temperature. As a result,

the cooling load is decreased and the additional energy consumedby air-conditioning system can be reduced [15–19]. Furthermore,reflective coatings can also contribute to the decrease of outdoorair temperature and therefore reduce the heat island effect [20].

642 A. Zhou et al. / Energy and Build

Fig. 1. (a) The right building envelope is covered by the reflective coating. The build-ing covered with reflective coating can reflect most of the solar energy and reducethe inward heat flux, exhibiting a lower surface temperature and cooler indoor tem-paa

Tb[ab[oahhsdmrme1dtc

ie

erature compared to those of uncovered building; (b) the photos of uncoated panelnd panel A corresponding to normal building scenario; (c) the photos of panel Bnd panel C corresponding to the building covered with reflective coatings.

he indoor temperature difference between coated and uncoateduildings has been observed up to 12 ◦C in hot and humid zones21,22]. Such distinct indoor temperature difference can achieven energy saving of 1.3 W/m2 according to a comparison studyetween medium-grey reflective coating and traditional painting23]. In another study, an electricity saving of 0.5 kWh/day has beenbserved after applying a reflective coating to a building envelopend improving its reflectance from 0.26 to 0.72 [24]. Due to theigh reflectance of coating, a concern that it may result in a highereating load of building in winter arises. A continuous two-yeartudy has been conducted in the subtropical area to investigate theynamic effect of reflective coatings on building thermal perfor-ance in summer and winter conditions. The results show that the

eflective coatings could reduce the external surface temperatureore than 10 ◦C in summer, while in winter, the temperature of

xternal surface covered with reflective coatings was only about◦C lower than that of uncovered surface [25]. This study hasemonstrated that the summer cooling saving is able to counteracthe winter heating penalty and the application of reflective coating

an still yield an energy saving effect.

Although the existing reflective coatings are effective in reduc-ng heat absorption, they show some limitations that impede theirxtensive applications [26]. Firstly, organic pigments in the exist-

ings 138 (2017) 641–647

ing reflective coatings are easily aged when exposed to ultravioletand the long-term durability is poor. Secondly, most inorganic pig-ments for reflective coatings contain hazardous metal elements(e.g. cobalt, chromium, cadmium), which are restricted by environ-mental regulations, for example, EU guideline No. 91/338/EEC [27].In order to overcome these limitations and promote the reflectivecoatings, it is necessary to characterize the thermal performanceof coatings with various inorganic additives which do not containhazardous metal elements and explore their working principles.

The ability to reflect radiation in infrared region is quantifiedby scattering rate, which is an important factor for the designof thermal coatings because the radiation in infrared region (i.e.wavelength longer than 720 nm) accounts for 49% of solar energyreceived by the Earth [28]. The scattering rate depends on the dif-ference in dielectric constants between additive and base material.Distinct dielectric constants between additive and base materialgive rise to a high scattering rate of coating. Thus, choosing anadditive with a high dielectric constant or a base material witha low dielectric constant is feasible to obtain a high scatteringrate. Normal base materials for thermal coating feature a dielec-tric constant around 3 (relative to vacuum), which is adequatelylow and has little room for further reduction. Therefore, the selec-tion of additives plays an important role in improving the scatteringrate of thermal coating. The titanium dioxide that exhibits a highdielectric constant makes an ideal candidate for thermal coatingadditive. In addition, heat insulation aggregates with low thermalconductivity can be used to improve the insulation capabilities ofcoatings. Because hollow glass beads possess a low thermal con-ductivity coefficient and a low density, they can act as an effectivethermal barrier at the outer surface of coating. With the above con-siderations, titanium dioxide and hollow glass beads are chosen asthe functional additives for reflective coating materials, which arethen evaluated in terms of their reflective and thermal insulationperformance.

The objective of this study is to characterize the reflective andinsulation performance of coatings with different additives andto explore the working principles of different additives. In thisresearch, two kinds of additives, i.e. titanium dioxide and hollowglass beads, are used. The thermal performances of pure acryliccoating, acrylic with titanium dioxide coating and acrylic withtitanium dioxide plus hollow glass beads coating are measuredand compared. The surface temperature and reflectance of variouscoatings are obtained from the experiment. To explore the work-ing principles, Finite-difference time-domain (FDTD) approach thatcan simulate the propagation of electromagnetic (EM) waves isemployed for different coating materials (i.e. pure acrylic, acrylicwith titanium dioxide), demonstrating how titanium dioxide par-ticles enhance the reflection. It is envisioned that a coating materialwith high level of reflectance and insulation can be characterizedand the working principles of different additives can be revealed,providing guiding instructions for picking suitable materials andgiving insightful implications to novel design of reflective coatingmaterials.

2. Experimental program

2.1. Materials

As one of most extensively used building materials, precastconcrete was adopted as substrate material in this experimentalprogram according to following advantages. Firstly, the shape and

dimensions of precast concrete can be standardized, eliminatingthe requirement of supporting formwork in situ. Secondly, the qual-ity of precast concrete can be standardized and is more reliable thanthat of concrete cast in situ.

Buildings 138 (2017) 641–647 643

dpacTspafTrrdtatt

2

pwtncgTifid(

2

rAs4a(entaaltisasssrstpcpmd−

A. Zhou et al. / Energy and

In this study, two functional additives were used, i.e. titaniumioxide and hollow glass beads. Three types of coatings were pre-ared with different components. Coating A was an ordinary whitecrylic emulsion; coating B was acrylic with titanium dioxide andoating C was acrylic with titanium dioxide and hollow glass beads.he preparation process for coating B is as follows. The acrylic emul-ion and titanium dioxide were added into the mixing tank with aortion of water. The appropriate dispersing agent, wetting agentnd antifoaming agent were also used to promote the coating per-ormance. The mixture was stirred at high speed to disperse evenly.hen, water was added to adjust the viscosity of coating and theeflective coating B with titanium dioxide was produced. The prepa-ation process for coating C was similar to that of coating B. The onlyifference was that the hollow glass beads were added to the mix-ure after pigment dispersion, then the entire mixture was stirredt low speed before adding water. All coatings possess the potentialo reflect solar radiation, reduce heat absorption, and thus reducehe cooling load.

.2. Specimens

Four concrete panels were used in this experiment, namelyanel N (reference uncoated concrete panel), panel A (covered withhite acrylic emulsion), panel B (covered with acrylic emulsion and

itanium dioxide) and panel C (covered with acrylic emulsion, tita-ium dioxide and hollow glass beads). Before the application ofoatings, the surfaces of concrete panels were cleaned using deter-ent and water so that no grease and ash exists on the surfaces.he coatings were applied to the concrete panels through brush-

ng. The panel samples were then put into laboratory environmentor 7 days curing. The photos of panels covered with different coat-ngs are shown in Fig. 1(b) and (c). It should be noted that theimensions of the concrete panel are 450 mm × 200 mm × 60 mmLength × Width × Height) and all concrete panels are identical.

.3. Instrumentation and data collection

Halogen lamp was used as heat radiation source to model solaradiation, irradiating the panels consistently in the experiment.ccording to the standard solar spectrum from ASTM G 173, 5% ofolar energy is distributed in the ultraviolet range (300–400 nm),6% of solar energy is distributed in the visible range (400–720 nm)nd 49% of solar energy is distributed in the near infrared range720–2500 nm) [29]. The spectrum of the halogen lamp used in thexperiment ranges from 380 to 3000 nm, covering the visible andear infrared regions, which are the major part of the solar spec-rum in terms of energy constitution. Meanwhile, the functionaldditives mainly improve the reflectance of coating in the visiblend the near infrared range. Based on above reasons, the halogenamp can be used to simulate the solar radiation for studying thehermal performance and the relative cooling effect of the coat-ngs containing different additives. The test environment for allpecimens was set to be identical for mimicking similar weathernd solar radiation condition in order to achieve a fair compari-on. The schematic diagram and photo of test instrumentation arehown in Fig. 2 and the halogen lamp is placed above the testedpecimens. The illuminated surface of panels in this experimentepresents exterior surface of building (outdoor) and unilluminatedurface stands for interior surface of building (indoor). During test,he radiation was kept on for 12 h continuously. The ambient tem-erature was kept the same as 25.6 ◦C throughout the test and it wasonsistent during the experiment for different samples. The tem-

erature of both illuminated surface and unilluminated surface waseasured at one-minute interval by thermocouples connected to

ata logger system. The range of temperature measurement is from10 ◦C to 200 ◦C and the accuracy is ±0.5% or ±0.5 ◦C (whichever is

Fig. 2. Experimental setup.

greater). After radiation, the halogen lamp was turned off and thepanel was cooled down until ambient air temperature. Then thepanel was replaced by another one for next test. The surface con-dition and surrounding environment were the same for all panels,and the heat transfer at the side surfaces of different panels wasconsistent for all measurements. The thickness of samples was thinenough so that the area of side surfaces was relatively small com-pared to that of the front surface. The heat transfer at the sideswould not affect the experimental findings. Through monitoringthe variation of illuminated and unilluminated surface tempera-ture, the thermal performance of panels covered with differentcoatings can be evaluated.

3. Experimental results and discussions

3.1. Surface temperature

The heat transfer of insulated surface under solar radiation canbe described by the following equation:

(1 − a)I = ε�(T4s − T4

sky) + h(Ts − Ta) (1)

where a is reflectance of surface; I is total radiation on the surface; εis emissivity; � is Stefan-Boltzmann constant; Ts is the equilibriumsurface temperature; Tsky is radiant sky temperature; h is convec-tive coefficient; Ta is the ambient air temperature [30]. When thesurface is under radiation, the surface reflectance of the objectiveis the governing parameter that affects the thermal performance ofspecimens.

The recorded temperature of illuminated surface and unillu-minated surface of specimens covered with different coatings isshown in Fig. 3 and Fig. 4, respectively. In the first 200 min,both the illuminated and unilluminated surface temperature rosequickly. After that, the increasing rate slowed down, implying thatthe thermal equilibrium between specimen and environment wasapproached gradually. Meanwhile, the illuminated surface tem-perature of uncoated panel N and panel A was close to eachother, which is far higher than the temperature of panel B andpanel C specimen. These results demonstrate that the ordinaryacrylic painting has little effect on reflecting radiation and reducingexterior surface temperature, while the coating B and C contain-ing titanium dioxide can reflect more radiation, reducing energyabsorption and the surface temperature of exterior wall signifi-cantly. The illuminated surface temperature difference betweenpanel B and panel C is only 0.4 ◦C, implying that the effect of hollowglass beads on reflective performance is not significant. When it

comes to the unilluminated surface temperature, an obvious tem-perature different between panel B and panel C can be observedfrom Fig. 4. Due to the low density of hollow glass beads, it canbe concluded that the hollow glass beads can effectively retard the

644 A. Zhou et al. / Energy and Buildings 138 (2017) 641–647

Fig. 3. The temperature variation against time (T-t curve) of illuminated surface indifferent samples.

Fd

ha

3

idclwsBlara

dwT

Table 1The physicochemical properties of coatings containing inorganic additives.

Item Qualification

Condition in container No agglomeration, uniformityafter stirring

Application property Brush twice barrier-freeFilm appearance NormalStability of low-temperaturestorage

No lumps, gel and separation

Temperature change resistance No changeWater resistance (96 h) No blister, crack, flakeScrub resistance (2000 times) No grinning effectTack-free time/h 1Alkali resistance (48 h) No blister, crack, flakeArtificial weathering resistance(400 h)

No blister, crack, flake Colorchange ≤ class 1

Table 2The spectral reflectance of coatings containing inorganic additives.

Reflectance Ultraviolet Visible Near infrared

Coating B 0.06 0.91 0.88Coating C 0.06 0.95 0.87

Table 3The temperature of specimens after 12 h radiation.

Specimens Temperature (◦C) Reflectance

Illuminated layer Unilluminated layer

N 48.5 40.1 0.21

ig. 4. The unilluminated surface temperature variation against time (T-t curve) ofifferent specimens.

eat transfer from the coating surface to concrete surface, forming thermal barrier for the exterior surface.

.2. Reflectance

The physicochemical and optical properties of the coatings withnorganic additions were tested following the China National Stan-ard (GB/T25261-2010) [31]. The physicochemical properties ofoatings containing titanium dioxide and hollow glass beads areisted in Table 1. The spectral reflectance of the coating B and C

hich ranges from around 250–2500 nm was measured throughpectrophotometric. The calculated spectral reflectance of coating

and C is shown in Table 2. It can be seen that in the ultravio-et range, the reflectance variation is small and coatings show highbsorption. In the visible and the near infrared range, the spectrumeflectance of reflective coatings is much higher than that of purecrylic emulsion, leading to an improved solar reflectance behavior.

According to the US standard MIL-E46136, the reflectance ofifferent coating materials can be calculated. A panel paintedith black enamel coating was used as a reference specimen.

he reference specimen was tested under the same instrumen-

A 47.6 37.5 0.26B 41.3 34.5 0.76C 40.9 34.0 0.81

tation until the surface temperature is stabilized. The stabilizedsurface temperature of reference temperature is 53.3 ◦C. The coat-ing reflectance can be calculated based on the following equation:

a = Tb − Tc

Tc − Tr(2)

where Tb is the temperature of black enamel specimen, Tc is thetemperature of coated panel and Tr is the room temperature [32].The ambient air temperature during testing is 25.6 ◦C. Based onthis formula, the reflectance of different panels can be calculated.The measured surface temperatures and the reflectance of differentcoatings are summarized in Table 3. The illuminated surface tem-perature after 12 h radiation could reach up to 48.5 ◦C for uncoatedconcrete panel N. The C specimen shows the lowest illuminatedsurface temperature, which is 7.6 ◦C lower than that of N specimen.Through observing these temperatures, the reflection performanceof different specimens can be evaluated. The illuminated surfacetemperature of panel A is only 0.9 ◦C lower than that of N panel,meaning that the ordinary white acrylic painting is not effective inreflecting the solar radiation. Through adding titanium dioxide, theilluminated surface temperature decreases 6.7 ◦C and reflectanceincreases from 0.21 to 0.76, which is observed in panel B, implyingthat coating titanium dioxide can reduce heat gain from radiationand improve the thermal reflection capability efficiently. The unil-luminated surface temperature difference between panel B and C is0.5 ◦C, implying that the hollow glass beads are efficient in imped-ing heat transfer from coating surface to concrete surface.

3.3. Energy saving effect

The unilluminated surface temperature is used to represent theindoor temperature in order to evaluate the thermal impact of dif-ferent coating materials on the indoor environment. The resultsdemonstrate that application of the reflective coating containing

A. Zhou et al. / Energy and Buildings 138 (2017) 641–647 645

Fig. 5. Snapshots of 2-d Meep models. (a) pure acrylic film (gray block) (b) acrylicetb

ttatibbwaHwbcwrtTc2iTwtt5

4

(pr

4

upsbpsccapa

mbedded with TiO2 particles (black dots). An emissive source is placed 3 �m abovehe block area, representing the incident EM wave. EM wave flux through the topoundary of the block is recorded for calculating radiation reflectance of the block.

itanium oxide and hollow glass beads can decrease the indooremperature and indoor temperature fluctuation significantly. Theverage unilluminated temperature of panel A is 3.5 ◦C higher thanhat of panel C. Based on experimental results, the energy sav-ng effect through replacing the ordinary white acrylic paintingy reflective coating containing titanium dioxide and hollow glasseads can be roughly estimated. The reflectance is now integratedith Transient Systems Simulation (TRNSYS) program to evalu-

te the cooling energy demand under annual weather condition atong Kong. A representative building is constructed. Simulationsith and without reflective coatings are performed. It is a 3-story

uilding with 10 × 15 m2 area and 3.5 m layer height. Whole-yearooling demand for the entire building is calculated under theeather condition of Hong Kong. According to the results, the

eflectance of solar radiation is set to 0.8 and 0.2 correspondingo the buildings with and without reflective coating, respectively.he cooling demand is defined as the energy required by air-onditioning system to maintain an internal temperature below4 ◦C. As Hong Kong is located in the subtropical area, space heat-

ng is not needed so the heating requirement is not considered here.he whole-year cooling demand for the building model with andithout reflective coating is 16694 kW·h and 15750 kW·h respec-

ively. The saved energy is 944 kW·h, occupying around 5.7% of theotal cooling energy demand. The result shows an approximately% energy save in terms of cooling demand.

. FDTD simulation

In order to figure out the working principles of TiO2, FDTDfinite-difference time-domain) method is adopted to simulate theropagation of EM waves at the surface of reflective coating mate-ials.

.1. Model construction and simulation setting

Finite-difference time-domain (FDTD) method is implementedsing Meep [33]. Two models are constructed for representing theure acrylic paint and the TiO2-embedded coating, as shown bychematic drawings in Fig. 5. The acrylic paint is represented bylock area with relative dielectric constant equal to 2.5. The TiO2articles are represented by spheres with relative dielectric con-tant equal to 9 and 200-nm diameter. Referring to the typical TiO2oncentration from existing experiments, the number of TiO2 parti-

les in a 40 × 40 �m2 is calculated as 3373 [32]. These 3373 particlesre randomly distributed inside the acrylic block. A dot source islaced 3 �m above the acrylic block. The top and bottom bound-ries of the simulation box are set to be a perfect matching layer

Fig. 6. Reflectance of acrylic paints with and without TiO2 particles calculated byFDTD method.

(PML) that can absorb electromagnetic (EM) waves. The left andright boundaries are set to be periodic, so the EM wave passingthrough the boundary from one side can re-enter the simulationbox from the other side. A flux counter is placed on the top surfaceof the block, recording the flux with a wavelength ranging from300 nm to 2500 nm. The reflectance of each model is calculated bydividing the reflected flux over the incident flux. The simulationkeeps running until the flux intensity decays by 1/1000 of its peak,permitting a convergent calculation.

4.2. Reflectance of acrylic paint embedded with TiO2 particles

In the simulation, the acrylic film is treated as an ideal dielec-tric, non-magnetic material with dielectric constant equal to 2.5.In this case, the reflectance of the plain surface can be calculatedanalytically by Fresnel equations, which link the reflectance to therefractive index and incident angle

Rs =∣∣∣ n1 cos �i − n2

√1 −

(n1n2

sin �i

)2

n1 cos �i + n2

√1 −

(n1n2

sin �i

)2

∣∣∣2

, Rp =∣∣∣ n1

√1 −

(n1n2

sin �i

)2− n2 cos �i

n1

√1 −

(n1n2

sin �i

)2+ n2 cos �i

∣∣∣2

(3)

Eq. (3) calculates the reflectance of s- and p-polarization incidentfrom substance 1–2, with refractive index n1 and n2, respec-tively. Because solar radiation contains an equal mix of s- andp-polarizations and the incident angle ranges from 0 to 90 ◦, thetotal reflectance can be calculated by Eq. (4).

R = 12

(Rs + Rp

), Rtotal =

�2∫

0

Rd� (4)

With n1 = 1 (air) and n2 = 1.5811 (square root of the dielectricconstant of acrylic), a reflectance of 0.1634 is obtained. This resultis in agreement with the result from numerical simulations (theaverage reflectance is 0.1584) shown by Fig. 6, justifying the validityof the simulation. It is also shown that the reflectance of acryliccoating is significantly improved with embedded TiO2 particles. The

reflectance of the TiO2-embedded acrylic system shows fluctuatedvalues at different wavelengths. Such fluctuations could be relatedto the dispersive nature of the TiO2-partcile system. With detailedsimulation, we would further reveal how the TiO2 particles reflect

646 A. Zhou et al. / Energy and Build

Fig. 7. Schematic diagram of light beam traveling through a dispersive system ofTa

Ec

4

sEsaopbpmsfi

vibration of molecules can be a temporary storage for radiation

Fiar

iO2 particles. This phenomenon occurs when the macroscopic optical theory ispplicable.

M waves and demonstrate that the interval between particles is aritical parameter of the reflection.

.3. Light propagation in a dispersive system of TiO2 particles

Physical processes including reflection, refraction (at the macro-copic scale) and diffraction (at the microscopic scale) occur whenM wave meets the surface of a TiO2 particle. Depending on particleize and wavelength, either the macroscopic or microscopic opticsre applicable. When the size of the TiO2 particle is larger than halff the wavelength, reflection and refraction would be the majorhysical processes occurring. Light propagation can be describedy Snell’s Law. As shown in the schematic diagram in Fig. 7, theropagation path can be guided back to the outside. Following this

echanism, the radiation energy is partially reflected. However,

ince the light has already traveled for a distance inside the coatinglm and the energy is attenuated, the coating film is still heated. As a

ig. 8. (a)(b)(c) Evolution of energy distribution during the emission of the dot source in

n the figures are plotted proportionally to energy density (proportion of the total energyir. Aside from the one-time reflection at the air-coating interface, the diagrams show meferences to colour in this figure legend, the reader is referred to the web version of this

ings 138 (2017) 641–647

result, this process only provides limited improvement on thermalinsulating capabilities of coating materials.

The microscopic optical process, diffraction, occurs when thesize of the TiO2 particle is comparable to one-half the wavelength.In this case, light behaves more like a wave and the propagation canbe predicted by solving Maxwell’s equations in time domain overthe space. The diffracted lights interfere with each other, leadingto constructive or destructive interferences and redistributing theenergy unevenly. Such microscopic optical process can be simu-lated by the FDTD method. In this study, an emissive dot source isemployed to stimulate three systems including a free space, the air-acrylic interface and the dispersive particle matrix. The responsesof these systems are shown in Fig. 8.

From Fig. 8(b) it is noticed that the interface between two dis-similar materials causes a one-time reflection, corresponding tothe case of pure acrylic coating subjected to incident EM wave.On the other hand, multiple reflections are observed in the disper-sive particle system shown in Fig. 8(c), corresponding to the caseof TiO2-embedded acrylic coating. The multiple reflections resultfrom the dissimilar material property (dielectric constant) betweenthe acrylic and the TiO2. Lots of interfaces are generated betweenthe acrylic base and the TiO2 particles, leading to an enhancedreflectance. Such microscopic reflection patterns could explain howthe incorporated TiO2 particles improve the reflectance of theentire thermal coating film.

4.4. Absorbance of infrared radiation by glass

In addition to optics, microscale mechanics can also be involvedin the interaction between solar radiation and substances. The

energy, which would later dissipate to ambient environment in theform of heat. Thus, the molecule acts as a barrier blocking the radia-tion. For example, water can absorb solar radiation specifically with

a free space, the air-coating interface and the dispersive particle system. Red colors) of each point. The blue lines indicate the boundary between thermal coating andultiple reflections generated by dispersive TiO2 particles. (For interpretation of the

article.)

Build

acoaasawrlcfrarti

5

iwticfsrpmcatttcn

A

tgCt

R

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[

[[

[

(2015) 220–225.[33] A.F. Oskooi, D. Roundy, M. Ibanescu, P. Bermel, J.D. Joannopoulos, S.G. Johnson,

MEEP: a flexible free-software package for electromagnetic simulations bythe FDTD method, Comput. Phys. Commun. 181 (3) (2010) 687–702.

[34] M. Iqbal, An Introduction to Solar Radiation, Elsevier, 2012.

A. Zhou et al. / Energy and

round 900-nm, 1100-nm, 1300-nm and 1800-nm wavelengths,orresponding to the frequencies of different vibration modesf water molecules. Such radiation absorbance can be observednd validated by comparing the solar spectrum with and withouttmospheric absorption [34]. Under cloudless skies, solar radiationhould be stronger than that under cloudy skies due to radiationbsorbance by water. Similarly, in hollow glass beads O H bond asell as other molecular bonds exist and can absorb near infrared

adiation, making a storage of solar radiation energy. Because hol-ow glass beads are lighter than other components and they oftenoncentrate near the upper surface of the coating, the energy gainedrom infrared radiation could dissipate through heat convection oradiation process to the air. The floating hollow glass beads can alsoct as a barrier blocking the heat conduction from ambient envi-onment to internal layer of the coated wall. These two effects fromhe hollow glass beads additive contribute to the improved thermalnsulating performance.

. Conclusion

In this work, experimental studies on thermal properties of var-ous coatings and numerical simulations that can simulate the EM

ave propagation have been conducted. The coating material con-aining titanium dioxide and hollow glass beads leads to a drop innterior surface temperature, up to 3.5 ◦C, implying the preparedoating can effectively reduce heat absorption and cooling loador buildings, and hence the energy consumed by air-conditioningystem can be saved significantly when the prepared coating mate-ial is applied to building envelope. Via numerical calculations, theropagation of EM wave in dispersive TiO2-particle system at theacroscopic scale, as well as the microscopic scale, has been suc-

essfully examined. By investigating the propagation of EM wavet the interface between the coating material and air, it is foundhat the embedded TiO2 nanoparticles can significantly improvehe reflective performance of coating films. This work shows thathe FDTD method is capable of predicting the optical properties ofoatings and can provide insightful information for the design ofovel thermal coating materials.

cknowledgements

The authors are grateful to the support from Croucher Founda-ion through the Start-up Allowance for Croucher Scholars throughrant number 9500012, and the support from the Research Grantsouncil (RGC) in Hong Kong through the Early Career Scheme (ECS)hrough grant number 139113.

eferences

[1] D&R International Ltd., 2010 Buildings Energy Data Book, US Department ofEnergy (2011).

[2] C.L. Chow, W.K. Chow, Assessing fire safety provisions for satisfying green orsustainable building design criteria: preliminary suggestions, Int. J. Archit. Sci.4 (3) (2003) 141–146.

[3] A. Zhou, K.-W. Wong, D. Lau, Thermal insulating concrete wall panel designfor sustainable built environment, Sci. World J. (2014).

[4] D. Karamanis, E. Kyritsi, S. Krimpalis, Well-ordered nanoporous materials forlow-temperature water phase changes and solar evaporative cooling, Sol.Energy Mater. Sol. Cells 139 (2015) 34–43.

[5] N. Skandalos, D. Karamanis, PV glazing technologies, Renew. Sustain. EnergyRev. 49 (2015) 306–322.

[6] A. Zhou, D. Lau, Thermal performance of concrete wall panel with advancedcoating materials: a sustainable design, in: 7th Int. Conf. SuDBE 2015, 27–29July, Reading, UK, 2015.

ings 138 (2017) 641–647 647

[7] K.W.D. Kalani, C. Dahanayake, C.L. Chow, The potential of energy savingthrough vertical greenery systems: using EnergyPlus simulation program,Energy Build. 138 (2017) 47–59.

[8] M. Santamouris, D. Kolokotsa, Passive cooling dissipation techniques forbuildings and other structures: the state of the art, Energy Build. 57 (2013)74–94.

[9] M. Jansen, H. Letschert, Inorganic yellow-red pigments without toxic metals,Nature 404 (6781) (2000) 980–982.

10] P. Sittisart, M.M. Farid, Fire retardants for phase change materials, Appl.Energ. 88 (9) (2011) 3140–3145.

11] A. de Gracia, L.F. Cabeza, Phase change materials and thermal energy storagefor buildings, Energy Build. 103 (2015) 414–419.

12] W.K. Chow, Fire safety in green or sustainable buildings: application of the fireengineering approach in Hong Kong, Archit. Sci. Rev. 46 (3) (2003) 297–303.

13] C.L. Chow, W.K. Chow, H. Yuan, A preliminary discussion on selecting activefire protection systems for atria in green or sustainable buildings, Archit. Sci.Rev. 47 (3) (2004) 229–236.

14] W.K. Chow, C.L. Chow, Evacuation with smoke control for atria in green andsustainable buildings, Build. Environ. 40 (2) (2005) 195–200.

15] Y. Huang, J.-l. Niu, T.-m. Chung, Study on performance of energy-efficientretrofitting measures on commercial building external walls incooling-dominant cities, Appl. Energ. 103 (2013) 97–108.

16] R. Yao, Design and Management of Sustainable Built Environments, Springer,2013.

17] M. Kolokotroni, B. Gowreesunker, R. Giridharan, Cool roof technology inLondon: an experimental and modelling study, Energy Build. 67 (2013)658–667.

18] A. Synnefa, M. Santamouris, I. Livada, A study of the thermal performance ofreflective coatings for the urban environment, Sol. Energy 80 (8) (2006)968–981.

19] H. Shen, H. Tan, A. Tzempelikos, The effect of reflective coatings on buildingsurface temperatures, indoor environment and energy consumption—anexperimental study, Energy Build. 43 (2) (2011) 573–580.

20] M. Kolokotroni, R. Giridharan, Urban heat island intensity in London: aninvestigation of the impact of physical characteristics on changes in outdoorair temperature during summer, Sol. Energy 82 (11) (2008) 986–998.

21] V. Cheng, E. Ng, B. Givoni, Effect of envelope colour and thermal mass onindoor temperatures in hot humid climate, Sol. Energy 78 (4) (2005) 528–534.

22] A. Synnefa, M. Santamouris, H. Akbari, Estimating the effect of using coolcoatings on energy loads and thermal comfort in residential buildings invarious climatic conditions, Energy Build. 39 (11) (2007) 1167–1174.

23] R. Levinson, P. Berdahl, H. Akbari, Solar spectral optical properties ofpigments—Part I: model for deriving scattering and absorption coefficientsfrom transmittance and reflectance measurements, Sol. Energy Mater. Sol.Cells 89 (4) (2005) 319–349.

24] C. Romeo, M. Zinzi, Impact of a cool roof application on the energy andcomfort performance in an existing non-residential building. A Sicilian casestudy, Energy Build. 67 (2013) 647–657.

25] A.L. Pisello, F. Rossi, F. Cotana, Summer and winter effect of innovative coolroof tiles on the dynamic thermal behavior of buildings, Energies 7 (4) (2014)2343–2361.

26] S.W. Lee, C.H. Lim, Reflective thermal insulation systems in building: a reviewon radiant barrier and reflective insulation, Renew. Sustain. Energy Rev. 65(2016) 643–661.

27] E. Directive, 91/338/EEC on the approximation of the laws regulations andadministrative provisions of the Member States relating to restrictions on themarketing and use of certain dangerous substances and preparations, Off. J.Eur. Uinon 186 (1991) 59–63.

28] X. Wang, C. Kendrick, R. Ogden, B. Baiche, N. Walliman, Thermal modelling ofan industrial building with solar reflective coatings on external surfaces: casestudies in China and Australia, J. Build. Perform. Simul. 5 (3) (2012) 199–207.

29] G.-. ASTM, Standard Tables for Reference Solar Spectral Irradiances: DirectNormal and Hemispherical on 37◦ Tilted Surface, ASTM, International, WestConshohocken, PA, 2012.

30] ASHRAE, Handbook, ASHRAE Fundamentals, Inc., Atlanta, GA USA, 2009.31] GB/T25261-2010, Solar heat reflecting insulation coatings for buildings, in,

SAC, Beijing, 2010.32] Z. Zhang, K. Wang, B. Mo, X. Li, X. Cui, Preparation and characterization of a

reflective and heat insulative coating based on geopolymers, Energy Build. 87