Solar roadways - a good or a bad idea?

At first blush, this looks like a fabulous idea:

http://www.infrastructurist.com/2009/09/21/dubious-green-schemes-the-solar-roadway/

Turn roadways into an enormous solar cell and get lots of other advantages like better new infrastructure.  A long while ago, I looked at making solar roadways (and parking lots and driveways and footpaths and….) under contract when I was at Squid-Labs : http://www.squid-labs.com/projects/cc.html

The very difficult thing about making a road is making it structurally sound enough to carry vehicles.  That means it has to take very high loads, and be very durable for up to 50 years.

So putting a solar cell there is completely possible, but then you’ll need to put some protective material on top that has some texture (so the roads are not slippery) and enough resilience to last a long time.  The problem is that that protective material uses A LOT of energy to produce.  Perhaps the company pitching this idea has some magic special sauce up their sleeve (in which case I wish the media would talk about that), but I suspect it’s more about good PR than reality.

There are 3.95 million miles of roads in the US: http://www.fhwa.dot.gov/ohim/onh00/table5.htm – I’ve previously heard 4.1 million, but I suspect that depends on whether you count private roads, forestry roads, and other smaller things.

This book estimates the road area at 15 million acres, or 0.6% of the US land area:

http://books.google.com/books/Road Ecology

At 4046 m^2 per acre, this gives you around 60 billion square meters of road. – around 10m wide for the average road.

If we got 1000W of insolation (noon, clear skies) at all of them, with 15% efficient solar cells, the peak generation capacity would be around 9500 GW or 9.5TW.  That sounds like a GREAT idea.

If we only get 250W of averaged insolation day in day out, and we have 15% efficient solar cells (remember this is highly optimistic) that would mean we get something like 2400 GW or 2.4TW of averaged power.  Again, this sounds like a great idea, and enough to provide nearly all of the US’s primary energy supply (around 3.6TW).

ok, so what’s the bad news ?

Well, back to the fact we have to protect these cells with something strong.  Let’s choose polycarbonate, it’s strong, it’s clear (transparent) and cheap, because it’s a commodity plastic.  Let’s assume that a 5mm thick (almost a quarter of an inch) layer is enough to protect the cell.  How much energy would it take to make that protective coating?

Here’s the numbers for 1 m^2:

http://www.wattzon.com/stuff/items/k12zlqv75xptrope9j4r32mda7

Or a rougher estimate here:

http://spreadsheets.google.com/pub?key=tRYasGG3ZuIT4wtjVrJ_0wA&single=true&gid=3&output=html

Which would be 1.06 * 10^27 Joules of energy to produce the protective covering alone.

How long would it take to get that energy back?

3.34*10^20/2.4*10^12 = 140million seconds, or around 4.5 years.

And this is optimistic.  If the coating is twice as thick, that’s 9 years.  If it takes the same amount of energy to make the solar cell, the copper, and all the other pieces, it’ll take 9 or 10 years.

And when we tested solar roads, they lost 20% of their efficiency due to dust, scratches, and dirt, very quickly, as much as 50% over time.  Add to this snow cover and other losses, and it looks very unlikely that these roadways will ever get the energy invested into them back in their lifetime.  That basically means we might be digging the carbon hole deeper.

Or as Eric Wilhelm (instructables.com) who worked on the project said:

“All the functions are great, it’s just the cost of doing it.  We might do better to pay $35 Trillion / 300 M Americans = $100,000 per person to give up their cars and bike everywhere.   I bet most people would do it for 5 years at least!”

Oh, $35 trillion – their quoted price….

The market price for wind turbines is around $2 per installed Watt (like a peak value) which means around $5-6 per installed average year-round Watt.  Solar is about the same or a little more.  $35*10^12 / $6 = around 6 TW of power.  That’s close to double the existing power demand of the US. – in other words, this one could be cheaper….

On the bright side, there is about the same amount of area for the edges of roads as there are for roads, perhaps the company is going to use those?  That would be a much better idea, a lot less plastic to put down and an easier technology to make work and deploy.