The road out of Tourouvre-au-Perche looked almost ordinary at first glance.
The surface ran a shade darker than the tarmac around it, and slightly glossy.
Under two lanes of routine French traffic, about 30,000 square feet of photovoltaic tiles lay glued flat to the asphalt.
They were rated to feed the local grid with enough power for the street lighting of a town of 5,000, in a commune of roughly 3,400 people.
When the meters were read, the gap raised a plain question: can a road surface do the job a solar cell needs?
Why a road surface is almost the worst place to put a solar cell
The problem starts with geometry. A rooftop panel in Normandy can be tilted to face the sun at the angle that captures the most light across the year. A road surface cannot tilt: it lies flat and faces straight up, which gives up a meaningful share of the annual harvest compared with even a modest roof pitch. In northern France, that penalty lands before a single car has crossed the panel.
Traffic adds more. A driving surface needs a tough, textured covering to grip tires, and that resin layer sits between the sun and the silicon. Vehicles shade the cells as they pass, dirt and fallen leaves settle on the glass, and tiles bonded into the pavement lack the airflow that keeps roof modules cool.
Then there is mechanics. The joints between adjacent tiles flex under axle loads, and on this stretch those joints broke up while tiles glued to the asphalt peeled away under the shear forces of heavy farm vehicles. Storms also knocked out parts of the electrical circuit.
What the road was built to do, and what it actually cost
French authorities opened what they described as the world’s first full scale solar road on the departmental road D5 outside Tourouvre-au-Perche, in the Orne, built on years of research with the national solar energy institute. The site was an ordinary rural route carrying roughly 2,000 vehicles a day, not a car park. Its purpose was to prove that solar road panels could supply a real grid connection from a working carriageway.
The price for that one kilometer, roughly 0.62 miles, was €5 million, about 5.2 million dollars, paid out of state funds. That bought the tiles, their resin coated protective surface, the wiring routed beside the road, and the grid connection that sent the output to the national distribution network.
The published expectation was roughly 280 MWh a year, an average of about 767 kilowatt hours a day with summer peaks near 1,500, a figure the environment ministry and the builder both put into circulation.
The numbers engineers actually measured
Across its first full year of operation, the stretch produced 149,459 kWh, a little more than half the projection. The builder argued that counting only the tiles that had run without interruption, about 83 percent of the target was met. Output then fell sharply: 78,397 kWh in the second year, and roughly 37,900 kWh in the first half of the year after that alone.
Independent analysis put the installation’s rated peak at 420 kW and its capacity factor at about 4 percent, against roughly 14 percent for a conventional angled solar plant near Bordeaux, at about a tenth of the cost per installed kilowatt. The shortfall was a stack of compounding losses: flat orientation, shading, soiling, and sections dropping out as joints degraded.
The road also showed structural wear far earlier than the design had assumed. About 330 feet of surface was lifted out because the deterioration could not be repaired, and the rough coating raised rolling noise enough that the speed limit on the stretch was cut.
What the trial turned up, and where the technology went next
The experiment did not simply stop. A second phase ran on a shortened stretch of about a quarter mile, where two batches of 576 next generation tiles were laid between 2020 and 2021, without lifting production significantly. The developer says around ten different tile versions were tested over roughly six years, giving engineers something no laboratory rig could supply: years of real traffic on a real road in real weather.
The company’s director conceded that “Our system is not mature for interurban traffic,” and the product was redirected toward cycle paths, sidewalks, parking areas and small standalone packs powering cameras, bus shelters and bike chargers. Its current guidance points toward low speed surfaces carrying light traffic.
Measured output per dollar still trails conventional arrays by a wide margin, including the vertical formats used in solar noise barriers. Independent reviewers noted the capacity factor gap between road mounted and tilted installations as the central lesson of the trial.
What the Normandy road changed about the way solar panels are shaped
The verdict from Tourouvre-au-Perche was not that photovoltaics were worthless but that a road surface is the wrong place for the job a solar cell has to do. Local councillors voted in early 2024 to end the experiment, machines began tearing out the tiles that May, and the work finished the following month. The mayor noted the commune paid nothing, with the dismantling carried by the company.
Noise walls, raised arrays over working farmland, and canopies over car parks all share one logic: keep the panel out of the abrasion zone, let it tilt toward the sun, and seal its edges against standing water.
The Normandy stretch supplied the clearest available evidence for that principle, in the form of a measured gap between a 280 MWh promise and a curve that fell year after year on a wet piece of French tarmac. What remains is data rather than power: a departmental road now carries ordinary asphalt again, and the degradation record of 30,000 square feet of embedded panels sits in the files for the next engineer who wants to know what a solar cell looks like after a few winters under a truck.
