Over the past several years there have been a number of mainstream media reports hyping the potential for solar roads. These roadways are intended to have solar panels embedded within them that can generate electricity to power lighting and heating systems (also placed within the road surface), as well as send any surplus energy to the electrical grid. The DOE has invested at least $850,000 since 2009 into a firm called Solar Roadways to pursue this concept. But before anyone gets too excited, these proposals need some rigorous examination.

We'll start with the price tag. The estimated cost of such solar roadways is U.S. $4.4 million per mile (although it is not clear how accurate this value is). In 2009, the United States had a total public road length of 4,050,717 miles. This translates into an estimated solar road infrastructure cost of $18 trillion, or about 125% of the United States' current annual gross domestic product.

To this author, there are a number of basic pass-fail tests such concepts should be exposed to before solar roads are given any serious consideration (or receive additional public funding).

The absolute first test is to confirm that the proposed solar road surface can withstand common traffic conditions. Namely, before worrying about the electronics and solar panel components sandwiched inside the solar road surface, one has to ensure that whatever surface materials are being used will not fail under normal traffic conditions. Based on the information this author has seen, the surface of solar roads will apparently be covered in glass that is textured to provide all-weather traction. Of course, the surface must not only be textured for traction, it must not produce any glare that affects drivers. One wonders if such surface modifications are possible (particularly anti-glare requirements during wet roadway conditions on sunny days) without significantly reducing light penetration through the surface layer to the underlying solar panels?

Regardless, test #1 is to prove that a glass surface will not fail under normal driving conditions. The test is simple. Place the proposed surface over a suitable substrate and drive a large vehicle over the material. If the glass surface does not break, move on to test #2. What is the problem? The answer is that this author has not yet seen any available drive-tests of a proposed solar road glass surface, yet he has seen claims of complete solar road panel prototypes. Such a situation seems like the cart is well before the horse. Placing solar panels on a horizontal surface (such as a roadway), connecting them together, and embedding lighting systems and microprocessors around the solar panels is a valid concept. The issue is whether or not we can make a reasonably priced light transparent cover for the solar panels and electronics that can withstand all possible traffic. No proof of this basic design requirement appears to have yet been made public. Why?

But it is not good enough to have solar road surfaces that are merely capable of handling a heavy traffic load (which includes maximally loaded semi-trailers and other pieces of heavy equipment, including military equipment with metal tracks [i.e., 70-ton M1 Abrams tanks] -- we cannot have roadways that restrict any movement of heavy equipment on them in the event of national emergencies, or that require immense repair costs following such uses), we also need road surfaces capable of withstanding major accidents. Thus, test #2 is to take the proposed solar road glass surface, place it over a suitable substrate, and then intentionally roll a large piece of heavy equipment (e.g., semi-trailer, tractor, etc.) over onto its side or roof, preferably at highway speeds, on this surface. If the surface survives this basic type of crash test -- which asphalt and concrete surfaces will -- then perhaps the concept is worth studying. Again, this author has not seen any such basic design requirement tests for the proposed solar road glass surfaces. Given the inevitably very high costs of solar road surfaces, we cannot afford to replace sections of the road surface each time there is an accident.

In response to questions about the possible impacts of earthquakes on solar roads, the Solar Roadways founder has this to say: "While we haven't had a chance to test it yet, we understand that an earthquake can be catastrophic for a road of any type. Basically, any such force that could destroy an asphalt or concrete road would have a similar result with a Solar Roadway." Asphalt roads can be patched and/or recycled in-place. The solar roads do not appear to have these lower-cost repair options available.

In reviewing discussions about the use of glass for a solar road surface, this author has seen some advocates describe how glass is as hard as steel, and thus favorable as a highway surfacing material. Yes, this hardness fact is generally true. But the real issue with glass is that it is typically brittle, whereas many steels are not. Hardness is not the sole criteria for a road surface. As a demonstration, solar road proponents can -- if they wish -- throw a glass cup against a cement wall, and then throw a stainless steel cup against a cement wall. Then throw a lump of asphalt against the same wall, and a ball of glass. See which one comes out in better shape for each test. Odds are the glass cup will be destroyed, whereby the stainless steel cup may have some damage, but will be usable. The lump of asphalt may be relatively fine, while the glass ball will probably shatter. Another example -- why don't we build cars entirely out of glass? or boats? Brittleness of glass is a major design problem depending on the end product use.

Having presumably passed basic design criteria tests #1 and #2 (and one could imagine many other basic design criteria) -- which, as yet, this author has never seen being substantially discussed by solar road advocates -- we need other questions to be clearly answered prior to us engaging in such endeavors at public expense. What about thermal expansion and contraction of the roadways? As ambient temperatures of the ground surface around and under the solar road panels change, particularly when moisture is present (e.g., frost heaving), we see significant movements of road surfaces horizontally and vertically. Won't this disrupt (i.e., break) the solar panels and/or their interconnections?

How will dust, leaves, oil and grease, residues from vehicle tires, etc., all impact the solar panel performance? Roads accumulate significant debris on them from normal use. This is not a major issue for asphalt and concrete, where a thin surface layer of debris does not substantially impact performance, but these issues seem to be fatal flaws for a solar road surface that requires transparency above the underlying solar panels to ensure concept functionality. Will we need to have the equivalent of window cleaning machinery continuously scouring the millions of miles of roads just to keep them sufficiently clear so that the solar panels are operational? What will this cost?

The solar road boosters claim that self-cleaning glass with photocatalysts embedded in the road surface will keep the roadway clean. Having a doctorate in mechanistic organic photochemistry and research experience on photocatalysis, I can say that this argument is deeply problematic. Photocatalysis requires light to reach the catalyst, and even if it does, it cannot break down all types of materials. Deposits of very difficult to photodegrade oil, grease, tire residues, etc., on road surfaces will likely shield the underlying photocatalysts from light -- thereby negating the cleaning option (never mind the issues of what happens on cloudy/rainy/snowy days as the recalcitrant road grime builds up). The proponents argue that oil spills can be cleaned up with "street sweepers (vehicles with large rotating brushes)." Unless those street sweepers are using an expensive detergent mix that will require subsequently rinsing off the road with clean water in order to prevent detergent residues from building up on the road surface and reducing light transmission to the underlying solar panels, this option also seems like a non-starter. In short, cleaning heavy oil and grease deposits off a textured surface with just water and a brush is very difficult.

Some solar road proponents claim that the solar panels in the roads may be able to generate electricity that can somehow be transferred to electric vehicles moving over the road surface (i.e., recharge as you drive). Leaving aside details regarding energy transfer from the photovoltaic panels to a moving vehicle travelling overhead, wouldn't such approaches deplete the road surface of the energy needed to keep it clear of snow and ice in winter? In other words, part of the proposition for solar roads is that they would use solar energy during wintertime to heat the road surface, keeping it free of snow and ice. But if vehicles are using the road's solar energy production for recharging, what energy is left in the roads for melting snow and ice? Or are we only talking about possible solar energy transfer from the road to overhead vehicles during periods of the year when the risk of snow/ice cover is negligible? If so, do we not then need a redundant system of electric vehicle recharging stations along our roads that will just be used in wintertime, because in the summertime the electric vehicles can be recharged by the road as you drive? What kind of a business model is this? Seasonally operated recharging stations along our road systems? So many questions, so few answers. The problem is, many of these questions -- and the general feasibility of the concept -- can be answered at the desktop level, but to date the questions remain apparently unanswered.

What about shading from roadside vegetation and buildings? More unanswered questions.

Finally, large portions of the United States are subject to winter snowfall. Even if these are rare events, any road system must be sufficiently robust to deal with them over an expected lifespan of many decades, if not centuries. Apparently, a "solar-heated highway would be permanently heating (during winter months) at 40 degrees Fahrenheit, meaning snow wouldn't accumulate in the first place." This is indeed quite the claim. We know how cold it can get, even on occasion in the southern states (check out the record lows by state). In the northern heartland, extended periods of temperatures below -40 F with windchills on the order of -60 F occur. A solar road can maintain a road surface temperature of +40 F throughout such conditions? This seems impossible. Where are the energy balance calculations by the solar road proponents?

Many regions see temperatures at which exposed skin can freeze in minutes, and often in seconds (despite our body's 98.6 F internal heat engine). Somehow, during -40 F blizzarding conditions where little winter solar insolation penetrates to the ground surface, such solar roads will maintain a uniform road surface temperature of +40 F for 24 hours per day for the entire route between Minneapolis and Billings, or between Chicago and Cheyenne? To say that this sounds scientifically impossible is a massive understatement.

We can perform some back-of-the-envelope calculations of our own. About half the U.S. land mass receives somewhere between 0 and 2 kWh/square meter/day in solar radiation on a flat surface during December. A mid-range value of 1 kWh/square meter/day converts to about 4 MJ/square meter/day. Standard reference texts provide the energy required to melt various forms of snow and ice. Thus, at 4 MJ/square meter/day of average daily solar insolation during December, we obtain the following daily maximum depths of snow and ice possible to melt assuming all incoming solar insolation is perfectly converted into heat for melting (i.e., these are unconservative upper estimates, as there will be substantial conversion inefficiencies and other losses in the roadway heating systems): snow (new, loose), 5.3"; snow (on ground), 1.5"; snow (drifted and compacted), 0.9"; ice (32 F), 0.5"; and ice (-40 F), 0.4".

A quick review of daily historical snowfall rates for much of the northern U.S. indicates these amounts of solar insolation are not likely able to maintain snow and ice free solar road surfaces throughout a typical winter. Furthermore, once snow or ice begins to accumulate on a solar road surface, the underlying solar energy production drops off very rapidly and severely, thereby allowing further snow and ice to accumulate quickly. In order to restore solar energy production, the roadway would need to be cleared of accumulated snow and ice by snowplows, or salt, or both. How much damage would snowplows do to the solar road surface? And wouldn't salt crystals left after a successful snow/ice melt event cover the road surface with a film that impedes light transmission to the underlying solar panels? And since the road surface is textured for vehicle traction, if snow and ice did accumulate and was subsequently scraped off with a snowplow, the snowplow wouldn't be able to remove the snow between the surface undulations (think of trying to perfectly clean a snow-covered surface that is dimpled -- perhaps like a golfball -- using a flat scraping device). How would this affect restoration of full performance? And if snow and ice did accumulate, and drivers who still needed to travel used studded tires or chains, wouldn't this damage the glass road surface?

Of course, many of these types of issues have already been thought through in terms of the design criteria for heated roadways. A classic document states that "[s]now and ice melting systems must provide sufficient heat to melt snow as well as to offset surface heat loss by evaporation, convection and radiation and heat loss from the slab into the ground." One presumes the solar road advocates have all these heat loss numbers worked through? If so, where are they? Such calculations are one of the first steps in assessing this technology, and are not looked into only after large amounts of taxpayer dollars are spent. They are critically examined beforehand.

Overall, a preliminary analysis suggests solar roads would be unable to work effectively throughout a typical winter in much of the nation. But, as noted above, this is just the tip of the iceberg for serious concerns over solar roads. Cost, durability, maintenance difficulties, etc., the problems seem insurmountable.

Sierra Rayne holds a Ph.D. in Chemistry and writes regularly on environment, energy, and national security topics. He can be found on Twitter at @rayne_sierra.