Let's assume a good solar panel yields 15 watts per square foot under direct sunlight. The DOT estimates about 1.5 million acres of interstate in the US.
Without getting too complicated, let's assume we produce at our ideal wattage for 3 hours a day. Given our ideal solar panel, this is 2.94 billion kWh per day.
Getting ideal conditions on a highway surface is unrealistic, so let's assume our fictional extremely rugged solar panel can only yield 0.25 watts per square foot. Now we're sitting at 49.01 million kWh per day.
To put this into perspective, the US produces roughly 11.45 billion kWh per day. So turning all of our interstates into inefficient solar panels covers less than a half a percent of our energy production.
Caveat: this is shitty napkin math and omits all other paved surfaces in the US of which I'm sure non-federal roads and parking lots make up a lot of, but I couldn't find good sources for those.
That puts your estimate out by a factor of maybe 25 or more (depending on road width) if all roads are converted to solar. And that's before the other paved areas like parking lots.
I'm not advocating solar roads. They're a strange idea with many, many flaws, but the inefficiency aspect isn't one of them. If you want to generate lots of electricity with solar you can have one very efficient array, or lots of inefficient arrays. Solar roads are the inefficient one.
Also, I would be preocupied by the ways to carry this hypothetically produced energy.
Right now we have either small "local" productions with "local " consumption (that works well as there is little distance to be covered) or "large" production concentrated on one site and with tension raised to thousands of volts to allow delivvery to the final user.
A "solar" road would probably make sense only in urban areas (where there would be other issues, like - say - shadow from buildings, less time of road free of vehicles, etc.) to avoid the issue of transporting the energy for long distances to the final users (or have a huge loss in the process).
Why the 60x output reduction in normal vs. road solar?
With say a probably still generous 6x derating, it would mean 2.94/6/11.45 = 4% of electricity production.
If you manage to avoid road solar tax (say by putting it above the road), it gets to about 25% of electricity production. Assuming a less pessimistic capacity factor of say 16% [1] gives 32% of electric production. Build a 3x wide solar cover over the interstates and .. use your imagination.
Let's assume a good solar panel yields 15 watts per square foot under direct sunlight. The DOT estimates about 1.5 million acres of interstate in the US.
Without getting too complicated, let's assume we produce at our ideal wattage for 3 hours a day. Given our ideal solar panel, this is 2.94 billion kWh per day.
Getting ideal conditions on a highway surface is unrealistic, so let's assume our fictional extremely rugged solar panel can only yield 0.25 watts per square foot. Now we're sitting at 49.01 million kWh per day.
To put this into perspective, the US produces roughly 11.45 billion kWh per day. So turning all of our interstates into inefficient solar panels covers less than a half a percent of our energy production.
Caveat: this is shitty napkin math and omits all other paved surfaces in the US of which I'm sure non-federal roads and parking lots make up a lot of, but I couldn't find good sources for those.