Radiative Cooling Roofs: What They Could Mean in Romania

Home & Design14 min read

Radiative cooling can make a surface cooler than the surrounding air even when the sun is shining. It sounds contradictory: how can you cool a roof without switching on a compressor? The secret is not producing cold, but releasing heat as infrared radiation, under conditions that sufficiently limit heating from the sun and the surrounding environment.

The idea changes how we think about a home. Usually, we ask how to insulate it and which appliance to use for cooling. Here, a third question emerges: can the building’s surface send some of its heat towards the sky? Research suggests that it can. Yet there is a gap between a cool sample on a roof and a comfortable bedroom—a gap that advertising can conceal.

This article looks precisely at that gap: what has been demonstrated, what the figures mean, and what should be checked before a spectacular laboratory concept becomes a decision for a home in Romania.

For more information on advanced materials, see also Researchers made boron behave “wrongly”: Imma-B60 conducts electricity and withstands 23% plastic deformation, which provides complementary details on this subject.

At a glance: what to remember

A surface can lose heat through radiation, not only through contact with air.
Cooling below air temperature has been demonstrated experimentally, including during the day.
The temperature of a material, room temperature and electricity consumption are three different outcomes.
A benefit for a home should be assessed over the full year, not only on a summer afternoon.

The main image is an AI-generated conceptual illustration for InfoBuz. Infrared radiation is not visible to the naked eye; the house in the image does not depict a tested installation.

How radiative cooling works: two properties, not a magic paint

A surface designed for daytime cooling must solve two problems: absorb little solar energy and emit thermal energy efficiently. The first property is solar reflectance. The second is thermal emissivity. They are not the same thing, nor can they be fully inferred from the colour we see.

Lawrence Berkeley National Laboratory explains this distinction: solar radiation and the thermal radiation emitted by ordinary surfaces occupy different spectral ranges. A material can therefore reflect much of the energy coming from the sun while emitting well in thermal infrared. A white appearance is a visual clue, not a performance certificate.

For readers, a useful comparison is between two “doors”: one through which energy enters and another through which it leaves. A high-performing material seeks to keep the first door as closed as possible to the sun without closing the heat-release door as well. The image is simplified, but it explains why “very glossy” and “very good at cooling” are not synonyms.

The “window” to space and the seemingly impossible result

The atmosphere allows thermal radiation to pass through relatively well in a range known as the atmospheric window, approximately between 8 and 13 micrometres. Materials for radiative cooling can exploit this route for releasing energy. This does not mean that “cold from space” comes down onto the house: energy leaves the surface.

In a study published in Nature in 2014, the team led by Aaswath Raman and Shanhui Fan presented a photonic device that reflected 97% of incident sunlight. Under the experimental conditions, the surface reached 4.9°C below air temperature in direct sunlight. The authors separately reported a cooling power of 40.1 W/m², measured when the device was at ambient temperature.

These two findings should not be combined into a commercial promise. The minimum temperature and cooling power at ambient temperature describe different operating points. In addition, the experiment used a sample and a setup that limited unwanted heat exchange with the environment, rather than an entire house.

From photonic materials to coatings: three documented milestones

This is not a discovery that appeared this week. It is a field built through successive experiments, in which researchers have pursued not only a striking temperature result but also applicable materials and integration with real systems.

Milestone	What was tested	What it does not demonstrate on its own
Stanford, 2017	Panels that cooled circulating water in outdoor tests.	That a building’s entire system operates without electricity.
Columbia, 2018	A porous polymer coating tested in different environments.	That every white paint has the same performance.
Purdue, 2021	An ultra-reflective barium sulphate formulation.	That the percentage of reflected light is the percentage saved on an energy bill.

Cooled water and the difference between a test and a simulation

Stanford reported in 2017 on panels that kept water 3–5°C below air temperature during a three-day test. The team also used the data to model an office building in Las Vegas. The 21% reduction in electricity for cooling, cited in the release for the summer months, came from that simulation, not from measured bills for a Romanian home.

The distinction matters: the radiative heat-release phenomenon is passive, but integration into a circuit may involve pumps and other components. “Without electricity” must be understood in relation to the process described, not automatically extended to every part of the system.

A coating in which microscopic voids play an optical role

In 2018, researchers at Columbia presented a porous polymer whose air voids contributed to scattering sunlight. They reported temperatures approximately 6°C below ambient in Arizona and approximately 3°C below ambient in Bangladesh, under the respective test conditions.

What is interesting is the combination of material and test conditions. The results do not form a universal scale in which each city receives a fixed number of degrees. They show that a demonstration must be read alongside the location, weather and experimental configuration.

The paint that reflects 98.1% of sunlight

Purdue announced in 2021 a barium sulphate-based formulation with solar reflectance of up to 98.1%. The distribution of particle sizes contributed to light scattering. Tests also showed cooling below ambient temperature, not merely less heating than on a dark surface.

The figure of 98.1% describes the light reflected by the material under the measurement conditions. It does not describe the efficiency of a device, does not mean a 98.1% fall in temperature, and cannot be converted directly into lei saved.

The calculation that explains why the last percentage points matter so much

Illustrative example, calculated for this article: assume that 1,000 W of solar energy reaches one square metre and compare three opaque surfaces. In this simplified calculation, the portion that is not reflected is absorbed. The 90% and 95% values are hypothetical scenarios, not characteristics attributed to any particular products.

Assumed reflectance	Reflected energy	Absorbed solar energy
90%	900 W/m²	100 W/m²
95%	950 W/m²	50 W/m²
98.1%	981 W/m²	19 W/m²

The difference between 90% and 98.1% may appear small when looking only at the percentage reflected. But in this example, absorbed solar energy falls from 100 to 19 W/m²: by 81%. This is why a few percentage points can matter greatly in a heat balance.

The calculation does not represent net cooling power. It does not include incoming thermal radiation, surface emission or heat exchange with the air and substrate. Nor does it show how much the room cools. It is an arithmetic demonstration of solar-energy input, not a complete building model.

So the useful formula when reading an advertisement is simple: first ask “percentage of what?” The percentage of light, the percentage of cooling load and the percentage of the total bill have different denominators. Without that clarification, the same figure can create three entirely different impressions.

The roof, the room and the bill: three outcomes that must not be confused

Suppose, purely as an example, that the air is at 30°C and a sample reaches 25°C. That would demonstrate a result for the sample in that setup. It would not demonstrate that a room beneath it will be 25°C, nor that cooling that room no longer consumes energy.

For assessment, we propose three separate questions: how cool is the surface, what indoor comfort do you achieve, and what consumption do you measure? An offer that answers only the first question is incomplete for someone interested in the latter two.

The U.S. Department of Energy states that the benefit of a reflective roof depends on climate, insulation, roof type and building systems. It also warns of reduced useful solar gain in winter. That is why a serious estimate starts with the specific building.

For a new home, the question should be considered together with the design of the building envelope. We explained this approach in the InfoBuz guide to designing an nZEB house. This article adds one option to assess; it does not replace the home’s energy design.

Would radiative cooling make sense for a home in Romania?

It is a legitimate design question, not a saving that can be calculated from a press release. We do not present measurements carried out by InfoBuz in Constanța, Buzău or Bucharest. Using international examples as proof of an exact local saving would go beyond the available information.

A useful comparison would begin with three versions of the same building: the existing solution, a well-documented reflective solution and, if there is a verifiable offer for that project, a specialised radiative-cooling system. The geometry, room use and desired level of comfort must remain comparable. Otherwise, the comparison risks being mostly between scenarios rather than materials.

For a flat that overheats in the afternoon, do not assume that the roof is the main problem. Ask for the contribution of windows and shading to be analysed as well. The guide to windows, glazing, shading and installation provides context for this discussion.

A useful result for the client would show separately summer comfort, annual cooling energy, heating energy and the cost of the intervention. A single number, such as “the roof is X degrees cooler”, cannot replace this comparison.

The winter issue and materials that change their behaviour

If the aim in summer is to remove heat, the aim in winter may be different. This very tension has led to research into adaptive materials. In 2021, Berkeley Lab presented a coating called TARC, designed to change its thermal emission according to temperature.

The team combined sample experiments with simulations for buildings in U.S. climate zones. The idea was to favour heat release when it is warm and limit it when it is cold. This is a research result on regulating material properties, not evidence that a home in Romania can be heated and cooled free of charge throughout the year.

The direction is fascinating precisely because it shifts the question from “which material cools the most?” to “which material behaves appropriately at different times?” For a home, performance on a single day can be impressive; the annual balance is more relevant to the decision.

How to read a data sheet without being misled by a large number

The Cool Roof Rating Council explains the difference between reflectance, emissivity and the SRI index. The latter is a value calculated from a product’s radiative properties. It can exceed 100: an SRI of 110 does not mean that 110% of solar energy is reflected, nor does it mean 110% system efficiency.

For readers, the practical rule is not to treat all numbers as interchangeable percentages. Ask for the name of the quantity, the unit, the testing method and the condition of the tested product. A value measured at the outset and one measured after weather exposure answer different questions.

It is not enough for a material to have an interesting optical property. Before application, ask a specialist to check compatibility with the substrate, waterproofing and operating conditions. This is a list of questions for an assessment, not a recommendation to experimentally coat an existing roof.

Seven questions before paying for a “cooling roof”

The following framework is an InfoBuz editorial proposal for discussing an offer. It is not a certification standard and does not assume that all technologies use the same installation method.

What exactly was measured? The temperature of the paint, substrate, water or room air? Ask for the answer in writing rather than leaving it to a suggestive thermal image.
What is the comparison against? Air, a black roof, a conventional white material or the building before the intervention? “Cooler” says too little without a reference point.
Under what conditions was the test conducted? Ask for the location, period, configuration and available weather data. A performance peak is not an annual average.
Are the savings measured or simulated? For simulations, ask for the assumptions. For measurements, ask for the period and how conditions were compared.
What happens in winter? Request a balance that includes heating, not only reduced cooling. Ask whether the result refers to energy or its cost.
What needs maintenance? Ask for cleaning instructions, warranty conditions and data on performance after ageing. Include these obligations in the comparison between offers.
Who is responsible for integration into the roof? Clarify the permitted substrate, application and responsibility for the assembly before installation. Do not sacrifice an essential roof function for a thermal promise.

What an honest comparison for your own home would look like

Before choosing a solution, write down the objective: do you want a more comfortable room, lower consumption or a lower surface temperature? These are compatible objectives in some projects, but achieving one should not be assumed to prove them all.

Then decide what result you would accept as evidence. For a professional assessment, a simple “before and after” photograph may be insufficient. Request a report explaining where measurements were taken, for how long, and what changed between the situations compared. If windows were shaded on one day and not another, an additional difference has been introduced into the comparison.

Finally, keep observation and conclusion distinct. “The surface had temperature X during period Y” is an observation. “The material will reduce my annual bill by Z” is a conclusion that requires more information. This discipline is useful regardless of how new or convincing the technology may seem.

We do not recommend going onto a roof or modifying its layers for an improvised experiment. Assessment and installation must be organised with professionals, without compromising safety or warranties.

What radiative cooling is not

It is not a photovoltaic panel. In the examples discussed, the aim is to release heat, not produce electricity. Nor does the term “passive” mean autonomy for the entire building. If you are pursuing independence from utility networks, the issue is much broader, as the analysis of an autonomous house designed starting from winter explains.

It is not a promise of a fixed temperature. An experimental sample is not equivalent to a thermostat guaranteeing the same temperature regardless of the weather. Nor does the label “radiative cooling” alone prove that the product being offered reproduces a result published by a university.

It is not a shortcut around design. For a home, the right question is not whether the phenomenon exists, but what justified role it can have in that project. A technology can be real and valuable without being the main solution for every home.

Frequently asked questions about radiative cooling

Does radiative cooling work during the day?

Yes, experimental demonstrations exist in direct sunlight. The outcome depends on the material’s properties, the setup and environmental conditions. It should not automatically be extrapolated from a sample to an entire building.

Does every white paint produce the same effect?

No. A white appearance does not certify reflectance across the full solar spectrum or thermal emissivity. Product data are needed for comparison, and a laboratory result must be connected to the tested formulation and conditions.

Can I give up air conditioning?

The studies presented do not justify this conclusion for a particular home. A possible role is reducing the cooling demand or supporting a system. Giving up equipment requires an assessment of the building and the desired comfort level.

How much would I save in a home in Romania?

This article does not provide a guaranteed local percentage. Data are needed on the building, climate, use, product and systems. A figure obtained in a U.S. simulation cannot be applied directly to a Romanian family’s bill.

Is an SRI index above 100 an error?

Not necessarily. SRI is a calculated index, not the percentage of energy reflected. Values above 100 are possible. Solar reflectance and emissivity must be read separately, with the testing method and conditions stated.

The “wow” idea is not a house without bills, but another route for heat

Radiative cooling is interesting because it adds an option to the way we think about comfort: sometimes, we can modify surfaces rather than only buy a more powerful appliance. But the value of a physical idea and the value of an investment for your home are two things that must be demonstrated separately.

Perhaps the most useful question this field leaves us with is not “how cool can a paint become?”, but “what evidence do I need to know that the solution works for me?” When a technology is real, it does not need impossible promises to be remarkable.

Sources and research methodology

Explanatory material based on the scientific paper and institutional sources below, consulted on 26 September 2026. The years mentioned indicate the publication of research, not the launch of products in Romania. The comparative calculation and the question framework are original editorial examples; they are not InfoBuz measurements, a product test or a technical project.