Saving Energy with Reflective Roof Coatings

Danny Parker and Steve Barkaszi are research scientists with the Florida Solar Energy Center in Cape Canaveral.

Reflective roof colors can save energy in hot climates and can help utilities in warm climates reduce peak demand. But how cost-effective are they?

Over the past two years, however, researchers in Florida and California have examined the impact of reflective roof coatings on air-conditioning energy use in retrofits of monitored homes. Simulation analysis suggests that a reflective roof color can cut a building's cooling load by 10-60%. The higher numbers are associated with uninsulated roofs.

White reflective coatings are increasingly being used for manufactured homes in the Southeast, based on homeowner reports that such coatings can reduce summer air conditioning costs. Until now, however, no investigation in a cooling-dominated climate examined the effect of reflective roofing on time-of-day air conditioning electrical demand in occupied residential buildings--important information for utilities where summertime peak demand is a concern.

One of the earliest whole-building studies that measured cooling-energy savings from reflective roof coatings was performed by the Mississippi Power Company. The utility monitored two identical side-by-side single-story commercial office buildings after the roof of one had been covered with a reflective white elastomeric coating. Both existing buildings had R-11 roof insulation. The results of the experiment? Summertime air conditioning was reduced by 22% in the building with the reflective roof coating.

More recently, researchers at Lawrence Berkeley Laboratory measured very significant cooling-energy savings from applying high-albedo1, or reflective, coatings to three buildings in central California (see Urban Heat Islands, p.16). At one site, energy demand for space cooling was nearly eliminated. But regardless of the potential of reflective roofing systems in California, Florida's higher humidity and nighttime temperatures make prospects for near elimination of space cooling energy use in that state very unlikely.

An Initial Experiment

In the summer of 1991 we conducted a preliminary experiment in Merritt Island, Florida. Our first test building (Site #0) was a 1,800 ft2 detached single-family, single-story home of conventional concrete-block construction. The pitched roof faced north-south, with plywood decking covered by green/gray asphalt shingles. The home's attic was well insulated with approximately two inches of fiberglass covered by an additional six inches of cellulose insulation, yielding a thermal resistance of about R-25. Air infiltration from the attic area into the conditioned interior (a common problem due to duct leakage), had been largely eliminated in a previous audit and retrofit. Beginning in May 1991, we submetered the home's air conditioner while maintaining a constant thermostat setting of 79deg.F. We also recorded the underside roof deck, attic air, and living room temperatures.

When we applied the reflective roof on September 5 of that year, the roof's reflectivity increased from 0.22 to 0.73.2 Spot measurements under full sun at midsummer had shown shingle surface temperatures of 160-170deg.F, prior to the roof treatment, compared to 110deg.F after we applied the coating. Analysis assuming an 81deg.F average summer temperature indicated that a reflective roof coating would reduce energy consumption by 10% (35 kWh versus 39 kWh per day).

Yet this test house probably understated the savings, since most existing Florida residences have fairly poor attic insulation and attic air frequently leaks into the conditioned interiors. Therefore, we obtained more typical residences for the detailed experiments we conducted the following year.

A Five-House Follow-up

To learn about how reflective roofs affect peak cooling demand we measured the 15-minute air-conditioning electricity demand in our follow-up study, along with meteorological conditions for three weeks before and after each home was retrofitted. We also used infrared thermography to examine how interior heat fluxes from the roof/ceiling were altered by the retrofit.

With equipment to instrument two buildings, we sought one residence with typical ceiling insulation levels (approximately R-11) and a second structure without any insulation at all. (Many homes built in Florida prior to 1965 have no attic insulation and were built with flat roofs that make retrofits difficult.) Data from Site #1 would be used to obtain results from a more-typical existing residential building, while Site #2 would help us define the maximum savings potential for reflective roof coatings in Florida. Experiments on three more houses in the summer of 1993 extended our sample size. Each house in the second round of experiments had unique characteristics that broadened our knowledge of how reflective surfaces reduce air-conditioning needs (see Table 1).

Results

Site #1

Site #1 was a fairly typical existing Florida home. The attic was insulated to approximately R-11, but the air conditioner was old and inefficient. Although pre- and post- application air temperatures and solar radiation were comparable, air-conditioning power demand was reduced by an average of 25% (from 40 to 30 kWh per day) after we applied the roof coating. The average electrical consumption of the air conditioning system during the utility coincident peak period (5-6 pm) was 2.4 kWh before the coating and 1.7 kWh afterward. This 700 W savings represents a 28% reduction in peak power demand attributable to the coating. Furthermore, average 24-hour attic air temperatures were reduced by 6deg.F, while peak attic temperatures between 2 pm and 6 pm fell by an average of 15deg.F.

Site #2

Site #2 was an ideal candidate for a reflective roof coating. The house had a flat roof and no space was available to insulate the ceiling assembly. Prior to the coating, the 2.5-ton air conditioner was unable to control the interior temperature adequately, running continuously each day from noon until 7 pm when the thermostat was finally satisfied.

Average air-conditioner electricity consumption dropped from 36 kWh to 20 kWh per day after the application--a 43% reduction. Savings would have been higher if the house had possessed a larger air conditioner, but the results did demonstrate the huge potential for gaining cooling-energy savings by whitening the roofs of homes without ceiling insulation.

The temperature reductions to the deck, deck airspace and ceilings were also striking, as was the change in the air conditioner's load profile. Before the retrofit, the daily interior temperature had ranged above the thermostat setpoint by 4deg.F or more. The average electrical demand of the air conditioning system during the utility coincident peak period (5-6 pm) was 2.2 kW before the coating and 1.4 kW after the application--a 38% reduction.

Site #3

Site #3 was a small house, cooled with a through-the-wall air conditioner. Since there was no attic duct system the site was of unique research value. The attic above the dropped ceiling contained no insulation, and the 1.5-ton air conditioner ran constantly prior to the coating (from 1-10 pm) unable to satisfy the thermostat. After the coating, the air conditioner cycled on and off during the same time period, maintaining improved interior comfort while reducing the utility coincident peak demand (5-6 pm) by nearly 960 W. Total daily air conditioning use was 11.9 kWh lower after the coating was applied--a reduction of 47% under peak-day conditions. After the retrofit, the average daily air conditioning savings totalled 5.6 kWh, or 25% during the summer period (see Figure 1) and peak demand savings averaged 30% (500 W).

Site #4

We selected Site #4 to see how whitening of a gravel roof (common in South Florida) might reduce energy use, and also because the household complained of high utility bills. The ceiling was well-insulated for a Miami home (R--11-R-19 blown fiberglass) and its 3-ton air conditioner was relatively efficient. But while auditing the home, we found a large duct-system supply leak in an inaccessible portion of the attic. (We found the leak with an infrared camera.) The leak was not repaired, but the roof was later coated with a white cementitious gravel roof coating. Although the percentage savings of air conditioning energy (15%) were lowest at Site 4, the absolute savings of 8.0 kWh per day were nevertheless significant.

Site #5

Site #5 had a tile roof, but the cement barrel tiles were old and stained a dark gray. The house also had relatively poor ceiling insulation and an inefficient air conditioner. The measured solar albedo was 20% before coating, but after being coated with a sprayed-on white elastomeric mastic, it was 64%. The absolute savings at this site were quite large at 11.6 kWh per day with a 988 W reduction in coincident peak-cooling demand.

Reflecting on Reflective Roofs

Reflective roofs can reduce space-cooling energy consumption and demand in Florida. Data collected so far suggest that air conditioning savings of 10-40% can be realized, with the larger reductions associated with poorly insulated roof assemblies or buildings with excessive attic air infiltration due to air handler return air leakage. Reflective coatings may be particularly suitable in existing residences where the roof structure makes it difficult to add insulation.

Average electricity consumption for central air conditioning in single family homes in Florida is approximately 4,400 kWh/year. Based on a savings level of 10-40%, reflective roofs can be expected to reduce household electricity use by 440 to 1,760 kWh per year--an annual savings of $35-$140 at current electricity rates (assuming 8cents per kWh). Obviously, the savings will vary depending upon the severity of the cooling season.

What About the Payback?

A frequent question concerns payback of reflective roofing. There are several angles on the answer, but generally speaking, reflective coatings are most appropriate when one is re-roofing. If the coating is applied to an existing roof that is in otherwise pristine condition, the cost equation is straightforward. The typical coverage of a white elastomeric coating is 25 ft2 per gallon, based on an application of two coats to a target thickness of 40 mils.

Cost for the material from vendors varies by 50% or more but averages about $60 per 5-gallon container when purchased in quantity. It is important to keep in mind that roof area is generally considered greater than building floor area, particularly with a steep roof pitch. For instance, a typical 1,500 ft2 home may have 2,200 ft2 of roof to be covered. The application then requires 90 gallons of coating material for a materials cost of approximately $1,100.

The cost of labor for installation depends greatly on the roof surface, on whether the coating is to be rolled on or sprayed, and on labor rates. A typical labor cost might be approximately 50cents per ft2 for the required two applications. Thus the overall application would cost about $1 per ft2, or approximately $2,200 for a typical home. With annual energy savings in Florida of $35-$140, the payback times are long--usually lasting longer than the roof itself.

A completely different scenario emerges if the home is soon in need of re-roofing, however. Here the roof coating (which essentially creates a new weatherproof surface) might be seen as a way of extending the life of the roof by 5 to 10 years at half of the cost of re-roofing. The energy savings then become a side benefit.

For new homes, the situation is even more interesting. Here it is often possible to choose roofing types--such as metal roofing, tile roofing, or metal or ceramic shingles--that can be specified in a reflective white at no additional cost. Unfortunately, no truly reflective asphalt roofing shingles yet exist for the residential market, but this situation may change as researchers work with the roofing industry to develop new products and spread the word about the energy benefits to help create a market for the materials. For commercial buildings, a variety of reflective roofing materials are already available: Hypalon, white EPDM, and PVC single-ply membranes. Once such products are widely available for the residential market, the economics may be significantly altered as the cost of reflective roofing becomes inconsequential. n

Notes

1. Reflectivity or albedo is the hemispherical reflectivity integrated over a particular wavelength band of the electromagnetic spectrum. For the purposes of this article, the terms reflectivity and albedo are used interchangeably and refer to the wavelengths encompassing the range of solar irradiance from 0.28 to 2.8 microns.

2. Surface solar reflectivity is measured using a precision spectral pyranometer with the device alternately faced upward towards the sun and downward towards the roof to determine the ratio of incident to reflected solar radiation.

The above research is detailed in two reports available from FSEC: Measured Air Conditioning Electricity Savings from Reflective Roof Coatings Applied to Florida Residences, and Measured Cooling-Energy Savings from Reflective Roof Coatings in Florida: Phase II Report, FSEC, 300 State Road 401, Cape Canaveral, Florida 32920. Tel: (407) 783-0300.

Large cities typically contain darker surfaces and less vegetation than rural environments; these circumstances increase solar gain and thereby raise summertime cooling-energy demand. The dark surfaces and lack of vegetation also warm the summer air, leading to the creation of the urban heat island. In fact, the average temperature in a typical city on a clear afternoon can be 1deg.F-5deg.F hotter than that of the surrounding rural area. Researchers at Lawrence Berkeley Laboratory's (LBL) Heat Island Project estimate that the additional air-conditioning use caused by this urban air temperature increase is responsible for 5%-10% of urban peak electric demand, at an annual cost of several billion dollars.

The power needed to compensate for these higher temperatures requires additional generating capacity, which often contributes to urban air pollution. Moreover, the elevated temperatures themselves accelerate smog formation. According to researchers with LBL, the probability of smog increases by 2%-4% per deg.F increase in maximum daily temperature. But shade trees and light-colored surfaces can offset, and may even reverse the summer heat island effect.

In one experiment, LBL examined the savings due to reflective roofing systems installed on three buildings in Sacramento, California. One was an occupied residence with R-11 ceiling insulation under a composite shingle roof. The initial roof reflectivity was measured at 0.18, and this was altered to 0.78 by application of an elastomeric roof coating. Furthermore, the air-conditioning cooling load in the building was reduced by 69%, with a 28% reduction in peak electrical demand, and the seasonal energy savings amounted to a reduction of approximately 14 kWh per day and a 1 kW in peak power demand.

The second and third buildings were test bungalows. In both cases, the buildings' corrugated metal roof albedo was increased to approximately 70%, and measured air conditioning energy use was reduced by approximately 40%-50%.

Degradation of reflective roof coatings is of practical concern because their high-albedo property is primarily responsible for the cooling-energy savings. Elastomeric roof coatings may have good longevity when applied properly. For example, a five-year old swatch of an elastomeric coating applied to the cupola roof of the Florida Solar Energy Center's (FSEC) Passive Cooling Laboratory still showed a laboratory reflectance of 0.73--very close to the initial properties of such samples (0.70-0.79) The reflectivity of the roofs in FSEC's experimental homes was measured seven months after the coatings were applied. Some minor stains due to disintegrated leaves and dust were evident at Site #1, whereas no signs of degradation were in evidence on the flat roof at Site #2. The average of the measurements at Site #1 indicated a reflectance of 0.69 with greater variation in the readings over the roof surface. The tested reflectance at Site #2 was 0.73. Although both aged values were lower, a statistical test of the means revealed no significant differences in the data taken immediately after the coating were applied and those obtained seven months later. More recently, however, FSEC examined the roofs at Sites #1 and #2--18 months after they were coated. Although the flat roof at Site #2 still showed little sign of weathering, some staining was becoming apparent on the coated asphalt shingles at Site #1.

The most significant research on the longevity of reflective roofing systems was performed recently at Lawrence Berkeley Laboratory. This research examined 26 spot measurements of aged high albedo roofs of various types and found that most of the weathering and reduction in solar reflectance occurred in the first year after application or even within the first two months. For a gravel coating the albedo was reduced by 8% over six years, but 6% of the drop occurred in the first year. However, other reflective roofing types experienced reductions to albedo of up to 24%. The LBL researchers also experimented with washing reflective roofing systems, and found that it was possible to restore roofs to 90% of their initial values but that it would not be cost-effective if someone had to be hired to perform the work.

Making a roof reflective doesn't always require application of a reflective coating. For new construction or re-roofing, there are types of roofing where a reflective white color can simply be specified over another color: for instance, a white galvalum metal roof or white tile can be chosen. The cost premium is usually zero, so payback is immediate. For commercial buildings, white EPDM, Hypalon, and PVC roofs are available. Unfortunately, no truly reflective asphalt shingle yet exists--even nominally white shingles have a solar reflectance no greater than 30%--but this may change. Researchers at LBL and FSEC are working with the roofing industry to help develop new products with greater solar reflectivity.

The Florida Solar Energy Center publishes a directory of suppliers of energy-efficient roofing products, including information on coatings, paints, and roofing types. The directory is updated annually. For a copy of BDAC Sources: Energy Efficient Products for Buildings-Roofing contact Kashif Hannani, Florida Solar Energy Center, Building Design Assistance Center, 300 State Road 401, Cape Canaveral, FL 32920 Tel:(407)783-0300 (Ext. 195). Copies are available for $5 each. The same information is also available free of charge via FSEC's computer bulletin board service. Tel:(407)730-BDAC.

Figure 1a. Air conditioner use and interior air temperature before and after a reflective roof coating is applied at site #3.

Figure 1b. Temperatures before a reflective roof coating is applied at site #3.

FSEC

Figure 1c. Temperatures after a reflective roof coating was applied at site #3.

Many considering the potential of reflective roofing are concerned about color. The Florida Solar Energy Center has evaluated the solar reflectance of some 37 different roofing materials, with the measured data showing that white roof materials generally exhibit the best performance. They are highly reflective across the solar spectral bandwidth, while being highly emissive in the far-infrared region--this is another way of saying they strongly reflect solar heat and any heat they absorb will readily re-emit to the cooler sky temperatures. It may seem a bit counter-intuitive, but silver reflective aluminum paints do not perform nearly as well as a simple white coating. This is because, although the aluminum flake paints have a high solar reflectance, they also have a low infrared emissivity--they tend to hold whatever heat they absorb--negating part of the reflectance properties.

Fortunately, for those who demand non-white roof colors, it appears possible to tailor paints and pigments so they are not so reflective in the visible solar range, but are very reflective in the invisible near infrared region. The Navy has conducted research in this area to help develop infrared reflective coatings. Paints have been created that are twice as reflective in the near infrared as in the visible region. Researchers with Lawrence Berkeley Laboratory are examining spectrally selective paints that offer the possibility of significantly increasing the solar reflectance of even darkly pigmented colors. Physics suggests green-colored pigments with large particle size may further enhance the performance of solar reflective non-white paints. Even so, such coatings will not likely perform better than materials that are uniformly very reflective access the solar spectrum--particularly since the energy intensity of solar radiation is greatest in the visible bandwidth. Regardless, such developments promise to provide improved roofing materials with high albedo, while still preserving the designer's pallette of colors.

It may also be possible to tailor the properties of white reflective coatings to create superior performance. An ideal coating would be very reflective across the entire solar spectrum, while being very emissive in the long infrared region so that heat is readily re-emitted. Research shows promise in this area. One specialty coating, used to coat astronomical observatory domes, has a 98% solar reflectivity--so high that the temperature of the material is only slightly higher than air temperature under moderate solar intensity. Most current coatings use titanium dioxide to provide the white pigment. However, research shows that introduction of barium sulfate into coating pigment may result in a similar solar reflectivity while possessing a greater long-wave emittance and greater cooling potential. Thus, it may be possible to tailor the composition of roof coatings to further optimize their performance.

Related Articles