Elsevier

Building and Environment

Volume 42, Issue 7, July 2007, Pages 2591-2605
Building and Environment

Cooler tile-roofed buildings with near-infrared-reflective non-white coatings

https://doi.org/10.1016/j.buildenv.2006.06.005Get rights and content

Abstract

Owners of homes with pitched roofs visible from ground level often prefer non-white roofing products for aesthetic considerations. Non-white, near-infrared-reflective architectural coatings can be applied in situ to pitched concrete or clay tile roofs to reduce tile temperature, building heat gain, and cooling power demand, while simultaneously improving the roof's appearance. Scale-model measurements of building temperature and heat-flux were combined with solar and cooling energy use data to estimate the effects of such cool-roof coatings in various California climates. Under typical conditions—e.g., 1kWm-2 summer afternoon insolation, R-11 attic insulation, no radiant barrier, and a 0.3 reduction in solar absorptance—absolute reductions in roof surface temperature, attic air temperature, and ceiling heat-flux are about 12, 6.2 K, and 3.7Wm-2, respectively. For a typical 1500ft2 (139m2) house with R-11 attic insulation and no radiant barrier, reducing roof absorptance by 0.3 yields whole-house peak power savings of 230 W in Fresno, 210 W in San Bernardino, and 210 W in San Diego. The corresponding absolute and fractional cooling energy savings are 92kWhyr-1 (5%), 67kWhyr-1 (6%), and 8kWhyr-1 (1%), respectively. These savings are about half those previously reported for houses with non-tile roofs. With these assumptions, the statewide peak cooling power and annual cooling energy reductions would be 240 MW and 63GWhyr-1, respectively. Statewide energy savings would reduce annual emissions from California power plants by 35 kton CO2, 1.1 ton NOx, and 0.86 ton SOx. The economic value of cooling energy savings is well below the cost of coating a tile roof, but the simple payback times for using cool pigments in a roof tile coating are modest (5–7 yr) in the warm climates of Fresno and San Bernardino.

Introduction

Owners of homes with pitched roofs visible from ground level often prefer non-white roofing products for aesthetic considerations. American Rooftile Coatings (Fullerton, CA) has developed non-white, near-infrared (NIR)-reflective architectural coatings that can be applied in situ to pitched concrete or clay tile roofs. These coatings can reduce tile temperature, building heat gain, and cooling power demand, while simultaneously improving the roof's appearance. Such NIR-reflective roof coatings can cool tiles while allowing choice of color by reflecting a significant fraction of the 52% of solar energy that arrives as invisible, NIR radiation1 [1], [2].

Widespread use of cool roofs can lower urban air temperatures, slowing the formation of smog and mitigating urban heat islands. This effect has been investigated in North America [3], [20], [21] as well as in Europe, Asia, and the Mediterranean [4].

Previous studies have measured and/or simulated reductions in cooling energy use and/or peak cooling power demand achieved by retrofitting dark-roofed homes with white roofs. For example, Parker et al. [5] measured energy savings averaging 19% for 11 Florida homes, while Akbari et al. [6] measured cool-roof energy savings exceeding 80% at a house in Sacramento. (The Sacramento house was unusual in that its walls were well-shaded by trees, making its roof the primary source of solar heat gain.) Konopacki and Akbari [7], Konopacki et al. [8] and Akbari and Konopacki [9] have simulated residential cooling energy and peak demand savings in many North American cities and estimate savings of about 10–20%, depending on the house construction and roof insulation. Cool-roof simulations have been used to develop the US Environmental Protection Agency ENERGY STAR® Roofing Comparison Calculator [10] and the US Department of Energy Cool Roof Calculator [11]. These online tools estimate cool-roof savings as a function of solar reflectance increase, building characteristics, cooling equipment, and climate.

Since cooling-power and energy savings are proportional to the increase in a roof's solar reflectance [8], white cool-roof savings can be scaled down to estimate savings achievable by (typically less reflective) non-white cool roofs. However, these savings are not directly applicable to tile-roofed homes, because the flow of heat from roof to building can be decreased by the (possibly ventilated) gap between the tile and the roof deck, and delayed by heat storage in the thermally manive tile [12], [13], [14].

The current study uses measurements from scale-model buildings to quantify the reductions in roof surface temperature, attic air temperature, and ceiling heat flux that are achieved by finishing roof tiles with NIR-reflective coatings. These data are then used to predict the energy savings that would accrue in various climates to full-scale homes roofed with NIR-reflective tiles.

We have monitored the interior and exterior temperatures of four adjacent, 1:10 scale, air-conditioned model houses sited in the hot, inland Southern California city of Riverside. The four buildings were roofed with concrete tiles finished with black, white, cool color (NIR reflecting), and standard color (NIR absorbing) coatings, respectively. The thermal performances of the buildings with matching cool and standard color roofs were compared to each other and to those of the black- and white-roofed reference buildings. From this comparison we characterized the thermal performance of the cool-roof tile coating, and extrapolated the results to estimate potential reductions in the peak cooling power demand and annual cooling energy consumption of homes in various California climates.

Section snippets

Reductions in roof surface temperature, attic air temperature, and ceiling heat flux

Neglecting thermal storage and linearizing the exchange of thermal radiation between the roof and its environment, reductions in roof surface temperature Ts (K), attic air temperature Ta (K), and ceiling heat flux q (Wm-2) are proportional to the reduction Δα in the solar absorptance of the roof, α. That is,Δx=kx×IΔα,where I is insolation (Wm-2); x is a thermal building property (Ts,Ta, or q); and the property-specific sensitivity (coefficient) kx depends on the thermal resistances (hereafter,

Construction, installation, and instrumentation of the scale models

We tested four identical, 1:10 scale, single-room wooden houses modeled after desert homes. Each building had

  • a concrete tile roof mounted on a pitched deck;

  • a naturally ventilated attic;

  • R-11 (1.9m2KW-1) foam-board insulation constituting the room's ceiling and lining the inside of the room's floor and walls;

  • an aluminum-foil facing on the top of the ceiling;

  • white exterior walls;

  • a thermoelectric air conditioner with a cooling output of 120 W (400BTUh-1), controlled by an electronic thermostat

Measured reductions in temperature and heat flux

The performance of each cool-roof tile coating was gauged by measured reductions (standard color value—matching cool color value) in the daily peak values of roof surface temperature Ts, attic air temperature Ta, interior air temperature Ti of an unconditioned building, and heat flux q into the ceiling of an air-conditioned building (Table 2). The coatings reduced roof surface temperature by 5–14 K and ceiling heat flux by 13–21%.

The ceiling heat flux was computed assuming a ceiling undersurface

Discussion

Power savings (per unit ceiling area) were about half those reported for Florida houses with R-11 attic insulation [5]. This may result from

  • the high thermal resistance between the roof tiles and the roof deck, which adds about R-2 to the roof-to-interior thermal resistance (Appendix);

  • convective cooling of the roof deck by air flowing between tile and deck; and/or

  • natural attic ventilation in the model houses.

It should be also noted that there is additional uncertainty in the energy savings

Conclusions

Application of NIR-reflective roof tile coatings yielded measurable reductions in roof surface temperature, attic air temperature, unconditioned interior air temperature, and ceiling heat flux. The coatings reduced peak roof surface temperature by 5–14 K and peak ceiling heat flux by 13–21%.

The coatings are predicted to save about 230 W (1.7Wm-2) in peak cooling power and 92kWhyr-1 (0.7kWhm-2yr-1) in cooling energy for a 1500ft2 (139m2) Fresno house with R-11 attic insulation. The economic value

Acknowledgements

This work was supported by the California Energy Commission (CEC) through its Public Interest Energy Research Program (PIER)/Energy Innovation Small Grant Program (EISG), and by the Assistant Secretary for Renewable Energy under Contract no. DE-AC03-76SF00098.

We would like to acknowledge the valuable assistance of David Wright, Atoya Mendez, Steven Lafund and especially Klaus Meister of Riverside Public Utilities; Dennis DeBartolomeo and Paul Berdahl of Lawrence Berkeley National Laboratory;

References (21)

There are more references available in the full text version of this article.

Cited by (201)

View all citing articles on Scopus
View full text