Improving the performance of thermochromic coatings ...

6 downloads 0 Views 2MB Size Report
May 12, 2013 - and optical filters were used on thermochromic coatings applied on ...... Zhu B, Wu K. Preparation and solar reflectance spectra of chameleon-.
Improving the performance of thermochromic coatings with the use of UV and optical filters tested under accelerated aging conditions

..............................................................................................................................................................

Theoni Karlessi* and Mat Santamouris Section of Applied Physics, Physics Department, Group Building Environmental Studies, National and Kapodistrian University of Athens, Build. Physics 5, 15784 Athens, Greece

.............................................................................................................................................

Abstract

Keywords: thermochromic coatings; photodegradation; UV and optical filters; accelerated aging *Corresponding author: [email protected]

Received 3 January 2013; revised 13 March 2013; accepted 2 April 2013

................................................................................................................................................................................

1 INTRODUCTION Elevated temperatures in cities compared with surrounding rural or suburban areas provoke the heat island effect. It is mainly related to the high density of buildings and urban structures that absorb solar radiation, the use of highly absorbing materials, lack of green spaces, the characteristics of the urban canyons and the production of anthropogenic heat. Increasing the demand for energy and cooling loads, accelerating the formation of harmful smog and causing human thermal discomfort and health problems by intensifying heat waves over cities are the main effects of the phenomenon of urban heat island effect. Various mitigation techniques to fight heat island effect have been proposed. Besides the need for vegetation and green spaces, the role of materials used in the urban fabric is also very important for the microclimatic thermal balance. They absorb

solar and infrared radiation and dissipate part of the accumulated heat through convective and radiative processes into the atmosphere, increasing ambient temperature. Thus, the optical and thermal characteristics of the materials used determine to a high degree the energy consumption and comfort conditions of individual buildings as well as of open spaces [1– 6]. The main properties of a material that controls its surface temperature are the solar reflectance and the infrared emittance [7]. Increased values of reflectance and/or emittance result in lower surface temperatures. Regarding the building’s performance, lower surface temperatures decrease the heat penetrating into the building and, therefore, decrease the cooling loads in case of air-conditioned buildings, or create more comfortable thermal conditions in case of non-air-conditioned buildings. In an urban environment, it contributes to decrease in the ambient air temperature, mitigating the heat island effect [2, 4, 8, 9].

International Journal of Low-Carbon Technologies 2015, 10, 45–61 # The Author 2013. Published by Oxford University Press. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted reuse, distribution, and reproduction in any medium, provided the original work is properly cited. doi:10.1093/ijlct/ctt027 Advance Access Publication 12 May 2013

45

Downloaded from http://ijlct.oxfordjournals.org/ by guest on August 20, 2015

The role of the materials used in the urban fabric is highly important for the microclimate and the energy loads of the cities. This has led to the research and development of innovative solutions with advanced thermal and optical properties. Thermochromic coatings that are reflective in summertime and absorptive in wintertime can address the demand for lower surface temperatures and energy savings. The function, however, of thermochromic materials in outdoor spaces is complex as they degrade when interacting with solar radiation. Various techniques with UV absorbers and UV filters prove that ultraviolet radiation is not the only parameter that affects thermochromism. This study aims to go one step forward and investigate the factors that destroy the reversible thermochromic circle besides UV radiation. Combinations of UV and optical filters were used on thermochromic coatings applied on concrete tiles under accelerated aging conditions of a one-month period to isolate the parts of solar spectrum that cause the photodegradation. The measurements of reflectance and color prove that the use of UV filter did not improve significantly the dark phase of thermochromic effect, while at the white phase, SR was reduced by 5% and SRvis was reduced by 9.7% compared with the uncovered sample. Covering the sample with red filter, which cuts off wavelengths below 600nm, protects most efficiently the reversible color change of the thermochromic coating as the solar reflectance at the dark phase remains unaffected during the whole experimental period.

T. Karlessi and M. Santamouris

46

International Journal of Low-Carbon Technologies 2015, 10, 45–61

and mechanical properties and loss of the reversible thermochromic effect [17, 18]. During a 10-day experimental period in outdoor conditions of eleven thermochromic coatings, the colored phase faded and solar reflectance was increased, while the tone of the colorless phase became darker and solar reflectance was decreased [10]. Ma et al. indicated a significant change in the trichromatic coefficient of two thermochromic coatings after 400 h of aging. Various techniques have been tested to decrease the degradation of the thermochromic coatings and improve their outdoor performance. Experiments proved that when UV absorbers are incorporated in the thermochromic coatings, the optical efficiency does not improve and the aging problems remain. Efficiency is improved when the UV protectors are applied on the surface of the coatings but still the problem of degradation is important. UV filters with a transmittance at the UV part of the solar radiation close to zero have also been used for the photostabilization of the thermochromic coatings [6, 19–21]. Results, however, showed that although the optical performance improves considerably the problem remains. This indicates that not only the ultraviolet but also other parts of the solar radiation interact with the molecular bonds, having a negative effect on thermochromism. The advantages that can be derived from their colorchanging properties concerning energy efficiency in buildings, indoor air environment and urban microclimate encourage further investigation [10] and have inspired this work. The following study aims to investigate the optical performance of a thermochromic coating under accelerated aging conditions, using combinations of UV and optical filters to isolate the influence of different parts of the solar radiation.

2 THERMOCHROMIC COATINGS 2.1 Composition-production For the production of the thermochromic coating, organic water-based red thermochromic pigments of powder form and a color-changing temperature of 258C were used. The red thermochromic pigment (Kelly Chemical Corporation) is an odorless and non-hazardous powder composed of thermochromic colorant (8– 12%), melamin resin (2 – 8%) and temperature controller, usually an alcohol (80 – 90%). Pigments are microencapsulated with an average particle size of 1– 10 mm. Microencapsulation serves as a barrier between the thermochromic system and the chemicals around it, such as the paint base, protecting the system from weather conditions, oxidation, etc. [11, 16, 22]. An appropriate binder system that should not itself absorb infrared radiation was produced for the development of the thermochromic coatings. The system composed of water (32%), thermochromic pigment red (12%) and titanium dioxide (9%) to avoid transparency at the colorless state and other additives. With a temperature increase above 258C, the thermochromic coating becomes transparent, revealing the color of the substrate.

Downloaded from http://ijlct.oxfordjournals.org/ by guest on August 20, 2015

The intensive research that has been carried out in recent years has led to the development of new-generation materials and techniques that present advanced thermal characteristics, dynamic optical properties, increased thermal capacitance and a much higher heat island mitigation potential [6]. One main requirement of an innovative material is the ability to be more reflective and present lower surface temperatures during the cooling period, while being more absorptive and taking advantage of the solar gains during the heating period [10]. This property can be described and analyzed by thermochromism. The thermochromic effect can be defined as a change in the spectral properties of an organic or inorganic substance caused by heating or cooling. In intrinsically reversible organic thermochromic systems, heating above a defined temperature causes a change in color from darker to lighter tones. This transition is achieved by a thermally reversible transformation of the molecular structure of the pigments that produces a spectral change of visible color. When temperature decreases below the color-changing point, the system returns to its thermally stable state [11–16]. Toward this direction, thermochromic color-changing coatings have been developed and tested. Thermochromic pigments have been developed as three-component organic mixtures, and they were incorporated into common white coating [13, 14]. After an hour of exposure to solar radiation and for ambient temperatures below 208C, the thermochromic coating could absorb almost the same amount of solar energy as an ordinary colored coating, but when the temperature was .208C, it could reflect more solar energy, presenting 48C lower temperature than the ordinary colored coating [14]. Karlessi et al. developed eleven thermochromic coatings at a color-changing temperature of 308C by using thermochromic pigments into an appropriate binder system and tested their thermal and optical characteristics against color matching conventional and cool (highly reflective) coatings [10]. The results for the brown thermochromic coating indicate higher values for the solar reflectance (SR ¼ 0.55 for the colored and SR ¼ 0.76 for the colorless phase) compared with the common (SR ¼ 0.18) and cool coatings (SR ¼ 0.41) of the same color. Mean daily surface temperature during a hot summer was also measured, and the temperature difference between common and thermochromic coatings was DT ¼ 11.3K and that between cool and thermochromic coatings was DT ¼ 9.2K. The thermal effect of improved reversibly thermochromic cement-based material (white cement mixed with 10% black microcapsule, and its switching temperature was 248C) was tested [12]. At the lower temperature (108C), the material could absorb light and make the temperature higher than that of normal white Portland cement by 38C. At the higher temperature (308C), the material could reflect light and make the temperature the same as that of normal white Portland cement. These results reveal the potential of thermochromic materials to avoid overheating in summertime and absorb heat in wintertime when it is necessary. However, photodegradation is a major problem for thermochromic materials when exposed to an outdoor environment. Interaction with solar radiation causes the breaking and/or crosslinking of the polymer chains, leading to altered chemical

Improving the performance of thermochromic coatings

Figure 1. (a) Full coloration, (b) Intermediate phase and (c) Full decoloration.

2.2 Application

Downloaded from http://ijlct.oxfordjournals.org/ by guest on August 20, 2015

The thermochromic coating was applied on concrete tiles of size 6 cm  6 cm. In Figure 1, the thermochromic sample in three different thermal phases before the exposure to accelerated aging conditions is presented. Figure 1a depicts the sample in full coloration at a temperature of 108C, while in Figure 1c, the sample is completely decolorized at a temperature of 358C. An intermediate phase is presented in Figure 1b.

3 OPTICAL AND UV FILTERS To protect the samples from the exposure to accelerated aging conditions, the following optical and UV filters were used: † red filter (Hoya type R-60, Edmund Optics Inc.[23]) † yellow, blue and green filter (Kodak Wratten, Edmund Optics Inc.) † UV glass filter (Optivex, Pegasus Lighting [24]) The transmittance of the filters at a range of 300– 2500 nm and at the visible range of 380 – 760 nm is presented in Figure 2a and b, respectively. The appropriate selection of the filters was based on their transmittance range and the aim was to cover partially the whole wavelength range, providing thus the ability to isolate the influence of the aging conditions in each part of the spectrum and for each filter.

4 EXPERIMENTAL PROCEDURE OF ACCELERATED AGING Accelerated aging of the samples was performed in an Accelerated Aging Xenon Test Chamber (Q-SUN, Xe-3HS) [25] for a one-month period in a 24-h basis continuously according to the specifications and requirements of ISO 11341: Paints and varnishes -Artificial weathering and exposure to artificial radiation- Exposure to filtered xenon arc radiation [26]. Xenon arc lamp test chamber tests materials for photostability by exposing them to ultraviolet, visible and infrared

Figure 2. (a) Transmittance of the filters at a range of 300– 2500 nm. (b) Transmittance of the filters at a range of 380 – 780 nm.

radiation. It produces the most realistic simulation of fullspectrum sunlight using filtered xenon arc light. With a nominal cut-on of 295 nm, the daylight filter used provides the most accurate spectral match with direct sunlight. The filter is recommended for the best correlation between the accelerated aging test chamber and natural outdoor exposures and conforms to the spectral requirements of ISO 4892, ISO 11341, ASTM G155, SAE J1960 and SAE J2527. The test chamber is equipped with a precision light control system which allows the choice of the desired level of irradiance. Irradiance is monitored and controlled at 340 nm. International Journal of Low-Carbon Technologies 2015, 10, 45– 61 47

T. Karlessi and M. Santamouris

Table 1. Specifications of the accelerated ageing. ISO 11341, cycle A, method 1 Filter: Daylight-Q

UV sensor: 420 nm

Black panel: insulated

Step

Function

Irradiance (W/m2)

BP temperature (8C)

Air temperature (8C)

Humidity (%)

Time

1 2 3

Light Light þ spray Go to step 1

0.51 0.51

65 65

38 38

50 50

1:42 0:18

† the geographical latitude of the exposure site (closer to the equator means more UV), † altitude (higher means more UV), † local geographical features, such as wind to dry the test samples, or the proximity of a body of water to promote dew formation, † random year-to-year variations in the weather, which can cause degradation to vary as much as 2:1 in successive years at the same location, † seasonal variations (i.e. winter exposure may be only 1/7th as severe as summer exposure), † orientation of the sample, † sample insulation, † operating cycle of the tester (hours of light and hours of wetness), † operating temperatures of the tester (hotter is faster), † the particular material tested. The spectral power distribution of the laboratory light source. Obviously, it is logically meaningless to talk about a conversion factor between hours of accelerated weathering and months of outdoor exposure. One is a constant condition, whereas the other is variable. In other words, weathering data are comparative and not absolute data. It is important to denote that any ‘rule of thumb’ that might have been developed by laboratories to convert tester hours into outdoor exposure hours is empirical from empirical comparisons of the laboratory’s own accelerated tests with their own outdoor exposures [27]. Furthermore, the rule of thumb conversions is valid only for the following: † the specific material tested. † the specific set of laboratory tester time cycles and temperature.

48

International Journal of Low-Carbon Technologies 2015, 10, 45–61

† the specific outdoor exposure site and sample mounting procedure. The field of accelerated aging of thermochromic materials is still rather unknown, and no empirical rule has been produced. The weathering procedure that has been followed meets the standard ISO 11341 simulating intense outdoor conditions with the specifications (Table 1) and the use of a daylight filter, which produces spectra approximately equivalent to direct, noon summer sunlight. The filters were placed on top of the samples covering the thermochromic surface as shown in Figure 3: a, sample with red and UV filter; b, sample with blue and UV filter; c, with green and UV filter; d, with yellow and UV filter. In total six thermochromic samples were submitted under testing: – – – – – –

sample without filters (uncovered), sample with UV filter, sample with UV and blue filters (UV þ B), sample with UV and red filters (UV þ R), sample with UV and green filters (UV þ G), sample with UV and yellow filters (UV þ Y).

Figure 4 depicts the accelerated aging test chamber with the samples.

5 MEASUREMENTS OF SOLAR REFLECTANCE AND COLOR The spectral reflectance, color coordinates and brightness were measured for each sample every five days at the colored (below the transition temperature) and the colorless (above the transition temperature) state. The solar reflectance and color difference DE were calculated. Comparative study of the values for every sample before the exposure and every five days after until the 30th day when the exposure was finished was carried out.

5.1 Measurements of spectral and calculation of solar reflectance The spectral reflectance of the samples was measured at a range of 300– 2500 nm of the solar spectrum that includes a part of the ultraviolet radiation (UV: 300 – 400 nm), the visible (VIS:400 – 700 nm) and the near-infrared part (NIR:

Downloaded from http://ijlct.oxfordjournals.org/ by guest on August 20, 2015

Temperature monitoring and control is performed by a black panel temperature sensor which controls the specimen’s surface temperature and simultaneously by the chamber air temperature control to give the ultimate determination of the specimen temperature. The effects of outdoor moisture are simulated by direct, pure water spray and by relative humidity control. Spraying cycle was skipped to isolate the effect of radiation. The relationship between tester exposure and outdoor exposure depends on a number of variables, including the following:

Improving the performance of thermochromic coatings

700– 2500 nm). A UV/VIS/NIR spectrophotometer (Varian Carry 5000) was used for measuring the spectral reflectance of the samples. The spectrophotometer is fitted with a 150 mm diameter integrating sphere (Labsphere DRA 2500) that collects both specular and diffuse radiation. The reference standard reflectance material used for the measurement was a PTFE plate (Labsphere). Measurements were conducted according to ASTM E903-96 standard [28]. Figures 5– 16 present the spectral reflectance of each sample at the dark and the white phases every five days of measurements. Tables 2 – 13 present the calculated values of solar, UV, VIS and NIR reflectance [29]. Figures 17 –20 present the solar reflectance SR and SRvis for the dark and the white phases of each sample on the 30th and last day of their exposure and the percentage (%) of the variation compared with the values before the exposure.

5.2 Measurements of color coordinates, brightness and color difference

Figure 4. Samples in the accelerated aging chamber.

The Carry Color Calculations application is a module that allows performing calculations on data collected by Varian Carry 5000. Calculations are performed for tristimulus values X, Y, Z, chromaticity coordinates x, y, z and color coordinates for CIE L*a*b*. Color difference DE*ab is also calculated. The set-up is for a selectable wavelength range of 380– 780 nm at an interval of 1nm, with a D65 illuminant and a CIE (1964) 108 observer. International Journal of Low-Carbon Technologies 2015, 10, 45– 61 49

Downloaded from http://ijlct.oxfordjournals.org/ by guest on August 20, 2015

Figure 3. (a) UV and red filter, (b) UV and blue filter, (c) UV and green filter and (d) UV and yellow filter.

T. Karlessi and M. Santamouris

Figure 6. Spectral reflectance of the uncovered sample at the white phase.

Figure 7. Spectral reflectance of the sample covered with UV filter at the dark phase.

50

International Journal of Low-Carbon Technologies 2015, 10, 45–61

Downloaded from http://ijlct.oxfordjournals.org/ by guest on August 20, 2015

Figure 5. Spectral reflectance of the uncovered sample at the dark phase.

Improving the performance of thermochromic coatings

Figure 8. Spectral reflectance of the sample covered with UV filter at the white phase

Figure 10. Spectral reflectance of the sample covered with UV þ green filter at the white phase. International Journal of Low-Carbon Technologies 2015, 10, 45– 61 51

Downloaded from http://ijlct.oxfordjournals.org/ by guest on August 20, 2015

Figure 9. Spectral reflectance of the sample covered with UV þ green filter at the dark phase.

T. Karlessi and M. Santamouris

Figure 12. Spectral reflectance of the sample covered with UV þ red filter at the white phase.

Figure 13. Spectral reflectance of the sample covered with UV þ blue filter at the dark phase. 52

International Journal of Low-Carbon Technologies 2015, 10, 45–61

Downloaded from http://ijlct.oxfordjournals.org/ by guest on August 20, 2015

Figure 11. Spectral reflectance of the sample covered with UV þ red filter at the dark phase.

Improving the performance of thermochromic coatings

Figure 15. Spectral reflectance of the sample covered with UV þ yellow filter at the dark phase.

Figure 16. Spectral reflectance of the sample covered with UV þ yellow filter at the white phase.

International Journal of Low-Carbon Technologies 2015, 10, 45– 61 53

Downloaded from http://ijlct.oxfordjournals.org/ by guest on August 20, 2015

Figure 14. Spectral reflectance of the sample covered with UV þ blue filter at the white phase.

T. Karlessi and M. Santamouris

Table 2. Solar reflectance (%) of the uncovered sample at the dark phase. Uncovered

Before exposure

5d

15 d

30 d

SR SRir SRvis SRuv

68 86 45 7

74 83 64 8

79 84 77 9

80 84 80 7

Table 3. Solar reflectance (%) of the uncovered sample at the white phase. Uncovered

Before exposure

5d

15 d

30 d

SR SRir SRvis SRuv

84 85 88 7

77 83 73 8

80 84 80 9

75 83 68 5

UV filter

Before exposure

5d

15 d

30 d

SR SRir SRvis SRuv

68 86 45 7

71 86 53 7

79 86 75 8

81 86 78 6

Table 5. Solar reflectance (%) of the sample covered with UV filter at the white phase. UV filter

Before exposure

5d

15 d

30 d

SR SRir SRvis SRuv

84 85 88 7

82 86 81 7

80 86 77 7

81 86 79 6

UV þ red filter

Before exposure

5d

15 d

30 d

SR SRir SRvis SRuv

68 86 45 7

67 85 45 7

68 85 45 7

67 85 45 6

Table 9. Solar reflectance (%) of the sample covered with UV þ red filter at the white phase. UV þ red filter

Before exposure

5d

15 d

30 d

SR SRir SRvis SRuv

84 85 88 7

84 85 89 8

84 85 88 8

83 84 85 6

Table 10. Solar reflectance (%) of the sample covered with UV þ blue filter at the dark phase. UV þ blue filter

Before exposure

5d

15 d

30 d

SR SRir SRvis SRuv

68 86 45 7

67 84 45 7

69 84 51 7

73 84 61 6

Table 11. Solar reflectance (%) of the sample covered with UV þ blue filter at the white phase. UV þ blue filter

Before exposure

5d

15 d

30 d

SR SRir SRvis SRuv

84 85 88 7

83 84 88 7

81 84 80 7

80 84 80 7

Table 12. Solar reflectance (%) of the sample covered with UV þ yellow filter at the dark phase.

Table 6. Solar reflectance (%) of the sample covered with UV þ green filter at the dark phase. UV þ green filter

Before exposure

5d

15 d

30 d

SR SRir SRvis SRuv

68 86 45 7

68 86 45 7

69 86 46 7

69 86 48 6

UV þ yellow

Before exposure

5d

15 d

30 d

SR SRir SRvis SRuv

68 86 45 7

68 86 45 7

69 86 47 7

69 85 48 6

Table 13. Solar reflectance (%) of the sample covered with UV þ yellow filter at the white phase.

Table 7. Solar reflectance (%) of the sample covered with UV þ green filter at the white phase. UV þ green filter

Before exposure

5d

15 d

30 d

UV þ yellow

Before exposure

5d

15 d

30 d

SR SRir SRvis SRuv

84 85 88 7

85 86 89 7

85 86 88 7

83 86 83 6

SR SRir SRvis SRuv

84 85 88 7

85 85 89 8

84 86 87 7

82 85 81 6

54

International Journal of Low-Carbon Technologies 2015, 10, 45–61

Downloaded from http://ijlct.oxfordjournals.org/ by guest on August 20, 2015

Table 4. Solar reflectance (%) of the sample covered with UV filter at the dark phase.

Table 8. Solar reflectance (%) of the sample covered with UV þ red filter at the dark phase.

Improving the performance of thermochromic coatings

Downloaded from http://ijlct.oxfordjournals.org/ by guest on August 20, 2015

Figure 17. Solar reflectance (SR) for each sample on the 30th day for the dark phase.

Figure 18. Solar reflectance (SR) for each sample on the 30th day for the white phase.

The CIE tristimulus values X, Y, Z are three imaginary primaries that were developed to overcome the problem of having a negative contribution of one of the traditional primaries for any given color. A particular color will always have a positive component of each of these three primaries. As it is difficult to draw a three-dimensional diagram on which to graphically display the tristimulus values, these values are reduced to a twodimensional graph via a transformation to produce the chromaticity values, x, y, z. The x and y coordinates evaluate the quality of a color (chromaticity) and z reports the spectral luminance (lightness) of the sample. Another color space, CIELAB or CIE L*a*b*, has been developed by the CIE in 1976 on the basis of the differences of three elementary color pairs: red – green, yellow– blue and black– white. Hue and chroma are defined by the coordinates a* and b*, which can have both positive and negative values. The

third characteristic, brightness, is designated by L* with scale values ranging from 0 (black) to 100 (white). Thus, in the approximately uniform CIE L*a*b* space, color differences can be calculated as a third dimension distance between the two samples [30 – 32]. Tristimulus coordinates, CIELAB L*a*b* parameters and color difference DE for each sample were measured. The results are presented in Table 13 for every sample for the dark and the white phases every five days of accelerated aging until the end of the procedure. The first value of DE column refers to the total color change before the exposure and the 30th day. Figures 22 – 25 depict the variation of brightness L* for four samples for the dark and the white phases during the experimental period. The color change DE of the last day of exposure compared with the state of each thermochromic sample before the

International Journal of Low-Carbon Technologies 2015, 10, 45– 61 55

T. Karlessi and M. Santamouris

Figure 20. Solar reflectance at the visible part (SRvis) for each sample on the 30th day for the white phase.

exposure is presented in Figure 26, while in Figures 27 and 28, the variation of DE for each sample for the dark and the white phases during the experimental period is depicted.

6 ANALYSIS OF THE RESULTS The analysis of the solar reflectance values of the thermochromic samples during the experimental period reveals that in every case SR increases in the dark phase, while in the white phase

56

International Journal of Low-Carbon Technologies 2015, 10, 45–61

decreases. As it derives from Figures 5– 16 and Tables 2– 13, the variation of solar reflectance is mainly in the visible part of the spectrum. The uncovered sample 30 d after the exposure to accelerated aging conditions has increased the SR value by 8% and SRvis by 16% at the dark phase compared with its condition before exposure. At the white phase, the reduction of SR is 7.6% and of SRvis is 17.6%. The use of UV filter did not improve significantly the dark phase of thermochromic effect, while at the white phase, SR was reduced by 5% and SRvis was reduced by 9.7% compared with the uncovered sample. Thus, the cut-off of

Downloaded from http://ijlct.oxfordjournals.org/ by guest on August 20, 2015

Figure 19. Solar reflectance at the visible part (SRvis) for each sample on the 30th day for the dark phase.

Improving the performance of thermochromic coatings

Figure 22. Variation of brightness L* for the uncovered sample.

Figure 23. Variation of brightness L* for the sample covered with UV filter.

the ultraviolet radiation does not ensure the improvement of the thermochromic behavior of the samples. The results of the use of optical filters prove that besides the UV part of the radiation, there are parts of the visible that also affect and degrade the thermochromic properties. Combined use of UV and optical filters further protects the thermochromic coatings depending on the parts of the visible spectrum that are transparent in each

case. Particularly, the greatest degradation is presented to the sample with the blue filter, presenting an increased SR value by 3.4% and SRvis by 7% at the white phase. Samples covered with the green and yellow filter show an increased SR value by 0.7% and SRvis value by 1.4% for the dark phase, while for the white phase, the percentage of reduction of the SR is 0.8% and of the SRvis is 4.4% for the green filter. For the yellow filter, the

International Journal of Low-Carbon Technologies 2015, 10, 45– 61 57

Downloaded from http://ijlct.oxfordjournals.org/ by guest on August 20, 2015

Figure 21. (a) Samples on the 5th day of exposure at the dark phase, (b) samples on the 5th day of exposure at the white phase, (c) samples on the 30th day of exposure at the dark phase, (d) samples on the 30th day of exposure at the white phase.

T. Karlessi and M. Santamouris

Figure 24. Variation of brightness L* for the sample covered with UV þ blue filter.

Figure 28. Variation of color difference for each sample at the white phase.

Table 14. Position of samples in Figure 21. UV

UV þ red

UV þ green

UV þ yellow

Table 15. Color values of the uncovered sample at the dark phase.

Figure 25. Variation of brightness L* for the sample covered with UV þ red filter.

Sample

Dark phase

Uncovered

Tristimulus

Days

X

Y

Z

L*

a*

b*

Before exposure 5 15 30

40.87 60.76 73.55 76.64

31.28 60.25 76.14 80.29

29.94 59.75 77.64 80.31

62.74 81.97 89.92 91.81

38.30 8.77 2.85 1.04

5.06 4.37 3.07 4.30

CIELAB

DE

47.3 35.2 3 1

Table 16. Color values of the sample covered with UV filter at the dark phase.

Figure 26. Color Difference (DE) on the 30th day for each sample.

Sample

Dark phase

UV filter

Tristimulus

Days

X

Y

Z

L*

a*

b*

Before exposure 5 15 30

40.87 51.00 71.15 75.12

31.28 44.03 73.15 78.64

29.94 45.65 73.56 76.81

62.74 72.25 88.52 91.07

38.30 26.23 3.85 1.17

5.06 1.73 3.85 5.69

CIELAB

DE

46.7 15.7 5.28 1.2

Table 17. Color values of the sample covered with UV þ green filter at the dark phase.

Figure 27. Variation of color difference for each sample at the dark phase.

58

International Journal of Low-Carbon Technologies 2015, 10, 45–61

Sample

Dark phase

UV þ green filter

Tristimulus

Days

X

Y

Z

L*

a*

b*

Before exposure 5 15 30

40.87 42.16 42.95 44.81

31.28 32.41 33.43 36.36

29.94 31.00 32.21 35.60

62.74 63.68 64.50 66.80

38.30 38.17 36.99 32.60

5.06 5.16 4.89 4.30

CIELAB

DE

7 0.95 0.96 3.2

Downloaded from http://ijlct.oxfordjournals.org/ by guest on August 20, 2015

UV þ blue

Uncovered

Improving the performance of thermochromic coatings

Table 18. Color values of the sample covered with UV þ red filter at the dark phase.

Table 22. Color values of the sample covered with UV filter at the white phase.

Sample

Dark phase

Sample

White phase

UV þ red filter

Tristimulus

UV filter

Tristimulus

DAYS

X

Y

Z

L*

a*

b*

Days

X

Y

Z

L*

a*

b*

Before exposure 5 15 30

40.87 41.97 42.30 42.51

31.28 32.21 32.58 33.21

29.94 30.66 31.06 31.28

62.74 63.51 63.82 64.33

38.30 38.32 38.02 36.41

5.06 5.35 5.31 5.90

Before exposure 5 15 30

85.82 79.27 74.63 76.46

88.61 79.45 76.46 79.80

96.16 81.96 74.46 76.76

95.42 91.44 90.07 91.59

3.42 7.94 4.45 1.63

20.72 2.41 5.82 6.63

CIELAB

DE

2.61 0.82 0.53 1.05

CIELAB

DE

8.48 6.79 2.93 1.12

Table 23. Color values of the sample covered with UV þ green filter at the white phase.

Sample

Dark phase

Sample

White phase

UV þ blue filter

Tristimulus

UV þ green filter

Tristimulus

DAYS

X

Y

Z

L*

a*

b*

DAYS

X

Y

Z

L*

a*

b*

Before exposure 5 15 30

40.87 42.38 48.00 58.34

31.28 32.74 40.36 56.92

29.94 31.52 42.07 56.65

62.74 63.95 69.72 80.13

38.30 37.69 29.00 10.91

5.06 4.88 1.42 4.10

Before exposure 5 15 30

85.82 86.56 85.01 81.64

88.61 89.95 88.55 84.69

96.16 96.67 91.98 84.54

95.42 95.97 95.39 93.75

3.42 2.39 2.00 2.63

20.72 20.09 2.07 4.50

CIELAB

DE

32.4 1.36 8.71 4.63

CIELAB

DE

5.55 1.32 0.84 1.01

Table 20. Color values of the sample covered with UV þ yellow filter at the dark phase.

Table 24. Color values of the sample covered with UV þ red filter at the white phase.

Sample

Dark phase

Sample

White phase

UV þ yellow filter

Tristimulus

UV þ red filter

Tristimulus

DAYS

X

Y

Z

L*

a*

b*

DAYS

X

Y

Z

L*

a*

b*

Before exposure 5 15 30

40.87 42.40 43.69 45.40

31.28 32.62 34.32 37.31

29.94 31.24 33.36 36.67

62.74 63.85 65.21 67.51

38.30 38.15 36.14 31.22

5.06 5.12 4.54 4.15

Before exposure 5 15 30

85.82 85.93 85.39 82.82

88.61 89.27 88.97 86.40

96.16 96.25 94.01 89.32

95.42 95.69 95.52 94.48

3.42 2.43 1.91 1.75

20.72 20.30 0.55 2.35

CIELAB

DE

8.58 1.12 1.72 2.79

CIELAB

DE

3.62 1.10 0.35 0.33

Table 21. Color values of the uncovered sample at the white phase.

Table 25. Color values of the sample covered with UV þ blue filter at the white phase.

Sample

White phase

Sample

White phase

Uncovered

Tristimulus

UV þ blue filter

Tristimulus

Days

X

Y

Z

L*

a*

b*

DAYS

X

Y

Z

L*

a*

b*

Before exposure 5 15 30

85.82 71.74 77.22 66.81

88.61 72.97 79.99 66.13

96.16 67.93 79.17 62.62

95.42 88.43 91.68 85.06

3.42 4.84 2.80 9.30

20.72 8.32 4.93 7.11

Before exposure 5 15 30

85.82 84.65 77.13 77.80

88.61 88.03 77.71 81.44

96.16 94.83 80.04 80.42

95.42 95.17 90.64 92.33

3.42 2.26 7.06 1.17

20.72 20.24 2.49 5.10

CIELAB

DE

6.97 11.52 1.5 66.81

CIELAB

DE

14.26 1.27 6.06 77.80

International Journal of Low-Carbon Technologies 2015, 10, 45– 61 59

Downloaded from http://ijlct.oxfordjournals.org/ by guest on August 20, 2015

Table 19. Color values of the sample covered with UV þ blue filter at the dark phase.

T. Karlessi and M. Santamouris

Table 26. Color values of the sample covered with UV þ yellow filter at the white phase. Sample

White phase

UV þ yellow filter

Tristimulus

DAYS

X

Y

Z

L*

a*

b*

Before exposure 5 15 30

85.82 86.33 84.71 79.85

88.61 89.70 88.08 82.33

96.16 96.33 90.96 79.81

95.42 95.87 95.19 92.72

3.42 2.42 2.30 3.56

20.72 20.05 2.42 6.24

CIELAB

DE

7.47 1.29 1.57 1.83

60

International Journal of Low-Carbon Technologies 2015, 10, 45–61

7 CONCLUSIONS Decreasing energy loads and improving the urban microclimate is a major concern that has put across a challenge of evolving techniques and materials with advanced properties. Thermochromic systems can function as energy-saving systems. For high temperatures, during summertime thermochromic coatings have the ability to reflect solar energy, reducing the surface’s temperature, while in wintertime absorb solar energy, increasing the surface’ temperature as reversible color change takes place. However, photodegradation is a major problem for thermochromic materials when exposed to outdoor environments. Various methods have been tested by applying different UV absorbers with different techniques in the thermochromic coatings, to photostabilize the color-changing effect of the material. The results, however, show that the performance of the thermochromic material was not improved and the degradation problems remain. This indicates that not only the ultraviolet but also other parts of the solar radiation interact with the molecular bonds, having a negative effect on thermochromism. The scope of this work is to detect the parts of solar radiation that damage thermochromism. For this reason, six samples of red thermochromic coating applied on concrete tiles were prepared and covered with different combinations of UV and optical filters. The samples were subjected to accelerated aging conditions for a one-month period and their optical characteristics (reflectance, color coordinates and color difference) were periodically measured. Variation of SR is mainly detected at the visible part of the spectrum. The uncovered sample, 30 d after the exposure to accelerated aging conditions, has increased SRvis value by 16% at the dark phase compared with its condition before exposure. At the white phase, the reduction of SRvis is 17.6%. The use of UV filter did not improve significantly the dark phase of thermochromic effect, while at the white phase, SRvis was reduced by 9.7% compared with the uncovered sample. The degradation of thermochromic properties was observed mostly at the samples where the filters are transparent at the parts of the visible spectrum closer to the ultraviolet, mainly in blue, green and yellow, while the sample with red filters remained unaffected at the dark phase and changed slightly at the white. The results of color change DE are in accordance with the reflectance results. The higher values until the last day of exposure are noticed for the uncovered sample and the sample with the UV filter, equal to 47 in both samples at the dark phase. The results for the combination of UV and red filter present the lowest DE equal to 2.6 at the dark phase and 3.6 at the white phase. The variation of DE every 5 d of exposure to accelerated aging conditions confirms that in the first 15 d, the greatest color change occurs in the uncovered sample and the sample

Downloaded from http://ijlct.oxfordjournals.org/ by guest on August 20, 2015

reduction is 1.7% for SR and 6.2% for SRvis. Covering the sample with a red filter, which cuts off wavelengths below 600 nm, protects most efficiently the reversible color change of the thermochromic coating as the solar reflectance at the dark phase remains unaffected during the whole experimental period. The white phase presents the lowest change compared with the other samples, 0.8% for SR and 2.6% for SRvis. The degradation of thermochromic properties was observed first at the samples where the filters are transparent at the parts of the visible spectrum closer to the ultraviolet, mainly in blue, green and yellow, while the sample with red filters remained unaffected at the dark phase and changed slightly at the white. The results are confirmed by the color measurements as presented in Tables 15 – 26, which show the color coordinates, CIELAB L*a*b* parameters and color difference DE. Brightness L* at the dark phase increases, while at the white decreases (Figures 23 – 26). For the uncovered sample and the sample with the UV filter, the biggest change in brightness is noticed during the first 10 d of exposure. Afterward, the brightness in both phases matches as the thermochromic properties are lost and the color change is irreversible. The samples that are covered with combinations of UV and optical filters and especially with yellow, green and red filter present smaller changes in brightness and distinct separation between the dark and the white phases during the whole experimental period. More precisely, at the sample with the red filter that provides the best protection of thermochromic characteristics, the change of brightness is 2.5% for the dark and 1% for the white phase. The results of color change DE are in accordance with the previous results. As observed in Figure 27, the higher values until the last day of exposure are noticed for the uncovered sample and the sample with the UV filter, equal to 47 in both samples at the dark phase and 7 and 8.5 in the white phase, respectively. The results for the combination of UV and red filter present the lowest DE equal to 2.6 at the dark phase and 3.6 at the white. The variation of DE every 5 d of exposure to accelerated aging conditions (Figures 30, 31) confirms that in the first 15 d the greatest color change occurs in the uncovered and the sample with the UV filter in both phases. Significant color change DE also occurs in the sample with the blue filter, which is transparent at the range of 390 –490 nm, which is closer to the UV part of the spectrum. Consequently, this sample is affected more

from the aging procedure compared with the other samples. The variation of DE at the sample with the red filter is minor.

Improving the performance of thermochromic coatings

with the UV filter in both phases. Significant color change DE also occurs in the sample with the blue filter, which is transparent at the range of 390 –490 nm, which is closer to the UV part of the spectrum. Consequently, this sample is affected more from the aging procedure compared with the other samples. The variation of DE at the sample with the red filter is minor. The stabilization of one thermochromic sample under intense and prolonged weathering conditions that is accomplished in this work is a breakthrough in the field of thermochromism. Considering also the advantages that can be derived from the color-changing properties concerning energy efficiency in buildings, indoor air environment and urban microclimate encourage further investigation. Future work including nanotechnology and thin-film techniques to deviate the incident radiation through multiple refractive layers and conduct heat can further improve the photostabilization of the thermochromic coatings preserving at the same time the thermochromic effect.

[1] Doulos L, Santamouris M, Livada I. Passive cooling of outdoor urban spaces. The role of materials. Sol Energy 2001;77:231– 49. [2] Synnefa A, Santamouris M, Apostolakis K. On the development, optical properties and thermal performance of cool colored coatings for the urban environment. J Sol Energy 2007a;81:488– 97. [3] Synnefa A, Santamouris M, Akbari H. Estimating the effect of using cool coatings on energy loads and thermal comfort in residential buildings in various climatic conditions. J Energy Build 2007b;39:1167 –74. [4] Akbari H, Davis S, Dorsano S, et al. Cooling our Communities—A Guidebook on Tree Planting and White Colored Surfacing. US Environmental Protection Agency, Office of Policy Analysis, Climate Change Division, 1992. [5] Santamouris M, Adnot J, Alvarez S, et al. Cooling the Cities. Eyrolees, Paris, France, 2004. ISBN: 2-911762-54-1. Ecole de Mines de Paris, Les Presses. [6] Santamouris M, Synnefa A, Karlessi T. Using advanced cool materials in the urban built environment to mitigate heat islands and improve thermal comfort conditions. Sol Energy 2011;85:3085 – 102. [7] Akbari H, Bretz S, Taha H, et al. Peak power and cooling energy savings of high albedo roofs. J Energy Build1997;25:117– 26. [8] Berdahl P, Bretz S. Preliminary survey of the solar reflectance of cool roofing materials. J Energy Build 1997;25:149 –58. [9] Bretz S, Akbari H, Rosenfeld A. Practical issues for using solar reflective materials to mitigate urban heat islands. J Atmos Environ 1997;32:95– 101. [10] Karlessi T, Santamouris M, Apostolakis K, et al. Development and testing of thermochromic coatings for buildings and urban structures. Sol Energy 2009;83:538 – 51.

International Journal of Low-Carbon Technologies 2015, 10, 45– 61 61

Downloaded from http://ijlct.oxfordjournals.org/ by guest on August 20, 2015

REFERENCES

[11] White MA, LeBlanc M. Thermochromism in commercial products. J Chem Educ 1999;76:1201– 05. [12] Ma Y, Zhu B. Research on the preparation of reversibly thermochromic cement based materials at normal temperature. Cem Concr Res 2009;39:90 –4. [13] Ma Y, Zhu B, Wu K. Preparation and solar reflectance spectra of chameleontype building coatings. J Sol Energy 2001;70:417– 22. [14] Ma Y, Zhang X, Zhu B, et al. Research on reversible effects and mechanism between the energy-absorbing and energy-reflecting states of chameleontype building coatings. J Sol Energy 2002;72:511– 20. [15] Lopes F, Neves J, Campos A, et al. Weathering of microencapsulated thermochromic pigments. Res J Text Apparel 2009;13:78– 89. [16] Bamfield P. Chromic phenomena, technological applications of color chemistry. R Soc Chem 2001;33 –41. [17] Pospisil J, Nespurek S. Photostabilization of coatings. Mechanisms and performance. Prog Polym Sci 2000;25:1261– 335. [18] Berdahl P, Akbari H, Levinson R, et al. Weathering of roofing materials—an overview. Constr Build Mater 2008;22:423– 33. [19] Pospısil J, Pilar J, Billingham NC, et al. Factors affecting accelerated testing of polymer photostability. Polym Degrad Stab 2006;91:417 – 22. [20] Oda H. Photostabilization of organic thermochromic pigments: action of benzotriazole type UV absorbers bearing an amphoteric counter-ion moiety on the light fastness of color formers. Dyes Pigm 2008;76:270– 6. [21] Oda H. Photostabilization of organic thermochromic pigments. Part 2: effect of hydroxyarylbenzotriazoles containing an amphoteric counter-ion moiety on the light fastness of color formers. Dyes Pigm 2008;76:400– 5. [22] Aitken D, Burkinshaw SM, Griffiths J, et al. Textile applications of thermochromic systems. J Rev Prog Color 1996;26:1 –8. [23] Edmund Optics Inc. http://www.edmundoptics.com (5 March 2013, date last accessed). [24] Pegasus Lighting. http://www.pegasuslighting.com/optivex-uv-filter.html (20 February 2013, date last accessed). [25] Q-Lab. http://www.q-lab.com/products/q-sun-xenon-arc-test-chambers/ q-sun-xe-3 (17 January 2013, date last accessed). [26] ISO. 11341: Paints and Varnishes -Artificial Weathering and Exposure to Artificial Radiation- Exposure to Filtered Xenon Arc Radiation. International Organization of Standarization, 2002. [27] Q-Lab TECHNICAL BULLETIN LU-0833. [28] ASTM Standard E903-96. Standard Test Method for Solar Absorptance, Reflectance and Transmittance of Materials using Integrating Spheres. [29] ASTM Standard G159-91. Standard Tables for References Solar Spectral Irradiance at Air Mass 1.5: Direct Normal and Hemispherical for a 378 Tilted Surface. [30] Billmeyer F, Fairman H. CIE method for calculating tristimulus values. Color res appl 1987;12:27– 36. [31] Ohta N. Correspondence between CIELAB and CIELUV color differences. Color res appl 1977;99999:178– 182. [32] X-Rite. A Guide to Understanding Color Communication, 2000.