Here’s what it would cost for the United States to go 100% solar
What would it take to replace all energy with green energy? How much would it cost? How soon could we get it done?
Today, we’re at only 12% renewable energy:
Solar makes up 1.4% of the total. We’d have to build 70 times as many solar panels to go 100% solar.
How many solar panels would it take to power America? Try to take a guess. Would we need to cover one city with solar panels? One whole state?
First I’ll do the tedious math, and then give a simple graphic.
Let’s imagine we put all the panels in the southwest, because that gives the most sunlight, especially during winter.
We use 100 quadrillion BTU’s of energy in a year.
That’s 29 trillion kilowatt hours per year.
That’s 80 billion kilowatt hours per day.
A square meter of solar panels makes 1 kWh of energy per day (in Arizona).
We need 80 billion square meters of panels.
That’s 31,000 square miles.
That’s a square, 180 miles by 180 miles.
If you put all the solar panels in Arizona, they would take up this much space:
We wouldn’t really bulldoze Phoenix to build it. But, split those panels across deserts in Arizona, New Mexico, Southern California, and a few other states and you could probably find enough unused land.
You could also put the panels across the country, but you’d have to build more panels since it would be less reliable in the winter. You could also put panels on every rooftop. That would shrink the size of the power plant, but it would increase the cost — retrofitting roofs with solar panels is much more expensive than building a utility scale solar farm.
The solar array would have 160 billion panels, each making 100 Watts.
Buy those panels on Amazon and each one would cost $80.
If we’re buying at scale, maybe it’s cheaper. One government report says that utility solar costs $40 per panel, but the full cost of building a power plant and hooking them up is more like $100 per panel.
The whole solar array would cost 16 trillion dollars if we built it today.
The same report thinks that prices could go down another 50% in the next 10 years. So, maybe we could even get it done for 8 trillion.
That’s not an impossibly large project. The war in Iraq cost 1 trillion dollars. The US government spent 7 trillion on its covid response.
We could build that much solar power, if we made it a national priority.
Here’s the real problem: that only gives us power while the sun is shining, so we would still need batteries, and those cost a lot more.
The cost of a national battery
I did that calculation based on the amount of sunlight in December, in the southwest. So we’d have just enough power in the winter and too much in the summer.
But I just used the average winter day. Some days are cloudier than others. Sometimes it’s cloudy for a week. Sometimes Arizona even gets a snowstorm, and all power generation would stop until the snow melts:
We need to store power in reserve. We need at least half a day’s worth of storage, just to get through every night. Let’s arbitrarily say we need 3 days of storage to get us through some bad weather. The actual worst case number could be higher.
3 days of power for the United States is 238 billion kilowatt hours.
Imagine we supplied that with a giant collection of cheap 12 volt car batteries. Here’s a $70 car battery that stores 420 watt hours of energy.
We’d need 566 billion of these. The total cost would be 39 trillion dollars just for the batteries, plus even more to hook them up.
Now, that’s a bit out of reach, even for the government. The national debt is 30 trillion dollars. All the real estate in America is only worth about 44 trillion.
And, just like a cheap car battery, it would only last a few years and then we’d need to replace it. We can’t just ask Canada for a jumpstart when it fails.
Are there better or cheaper batteries out there?
Tesla sells something called a Powerwall to store energy for your house. That holds 13.5 kWh, for $8500 (plus installation).
We would need 18 billion Powerwalls, for a cost of 150 trillion dollars.
Tesla also sells a Megapack, an industrial size battery that holds 3100 kWh. Each one costs 1.6 million dollars.
We’d still need 77 million of those. For a cost of 123 trillion dollars.
The problem isn’t that Tesla’s prices are too high. An EIA report says that all large scale batteries come with similar costs.
Things might get cheaper with economies of scale. But I also ignored the cost of installation, the cost of land, the cost of hooking these up to a new power grid.
Any way you look at it, the cost is too high, by maybe a factor of 10.
The green new deal comes with the same disclaimer as old toy commercials:
“Batteries Not Included”.
The biggest mining operation in history
Suppose that money weren’t an issue and we still wanted to build a nation sized battery. We might still run into a problem just finding enough materials to build a battery that large.
Assume first that we made it with lead-acid batteries, like the kind used for cheap car batteries.
These need 30 pounds of lead for each kilowatt-hour of storage. The national battery would need 3.5 billion tons of lead.
All the lead mines in the world dig up about 5 million tons of lead per year. So, we’d need to dig for 700 years at current rates to build that battery.
Identified lead reserves in the entire world are about 2 billion tons. So we’d need to find twice as much as we currently know of and rapidly dig it up.
Okay, how about Tesla’s lithium batteries?
Lithium-ion batteries are much lighter than lead-acid batteries. It only takes 160 grams of lithium per kilowatt-hour.
For the nation-sized battery, we’d need 42 million tons of Lithium.
World lithium reserves are 89 million tons. So, there is enough lithium identified in the world to complete this project. Unless other countries also want clean power. Then we won’t have enough for Europe and Asia to do the same thing, we’ll have to find more or we’ll end up fighting for limited resources, just like we fight wars over oil.
To some degree, this is already happening. The biggest known lithium reserves are in Bolivia:
Today’s batteries are both too expensive and rely too much on rare materials. We don’t have a better alternative ready yet. Even pumping water uphill for energy storage is still expensive and unlikely to scale to solve this huge problem.
Wind energy won’t save us
We could avoid some of the battery storage problem by building both lots of solar panels and lots of wind turbines. It could make the grid more reliable to have two inputs and to overbuild both of them. It could reduce the number of batteries needed for the worst case scenario.
This doesn’t solve everything, there will still be times when the sun isn’t shining and the wind isn’t blowing.
Wind turbines can also fail when it snows.
Still, it might help.
Suppose we powered the entire nation with wind.
What does your intuition say? Will wind power take up more or less space than all those solar panels?
The US uses 29 trillion kWh per year of energy.
That’s 3.3 million megawatts of production.
Each wind turbine makes about 2 megawatts.
We would need to build 1.5 million of those, if the wind blew all the time.
The wind doesn’t blow all the time, figure it’s maybe 35%.
So we need more like 4.3 million wind turbines.
Wind turbines cost 1.3 million dollars per megawatt.
The total cost would be 12 trillion dollars.
It would be cheaper to build solar+wind than to build solar+batteries.
How much land would this wind farm take up?
Wind turbines take up about 50 acres per megawatt. Some other sources say you can get away with half as much land per windmill. But it’s not obvious how low can can go — the power output goes down if you pack them too close together. Let’s go with 50 acres.
In total, we need 430 million acres. That looks like this:
That area would need to be completely covered with windmills.
I chose those states because the highest wind speeds are in middle of the country, from North Dakota down to Texas:
Covering most of Kansas with windmills is not impossible — most of the 50 acres around each one aren’t used, you could put them in the middle of every farm. In practice, it’s not entirely safe. If they ice up, the ice can get thrown off at high speeds. If a rotor ever came off, that would be deadly. Also, some people just won’t want to live with windmills everywhere they look.
I could much sooner imagine the US converting deserts into solar farms than putting windmills in every farm in multiple states.
Even if we built solar and wind, we’d still need a lot of batteries.
Natural gas helps, but it isn’t green
Suppose we just ignore the battery problem. We could build some solar and wind power and then burn natural gas at night and during bad weather.
This would work fine. It’ll still be costly, because we need to overbuild—we need enough solar and wind to power the country, and then we also need a huge number of gas turbines to be ready to run, as a backup. We need a lot of gas storage for those to be ready. Unlike batteries, we already have a lot of this infrastructure in place.
Maybe we get to a point where the mix is 30% solar, 20% wind, and 50% gas. That’s still better than today, it would be 50% renewable energy instead of 12%. Also, gas produces half as much CO2 as coal, to make the same amount of energy. So it’s more than a 50% reduction in carbon.
You might suggest that the other 50% could come from nuclear power instead of gas. In practice, that doesn’t work very well. You can’t turn nuclear power plants on and off quickly, so they can’t fill in when wind and solar falter. If we were going to go 50% nuclear, we might as well go closer to 100% nuclear.
Massive investments in wind and solar could slow down global warming, and maybe that buys us a few more decades to figure out batteries, or some other solution.
Nuclear power could also solve climate change. The two solutions don’t complement each other well. Either one would be the biggest infrastructure replacement in history. It would be good to have a plan before we start down one path or the other.
We still can’t support future growth
Suppose we did pull this off, somehow. We cover the southwest in solar panels, we build as much wind as we can, we invent and build new batteries that are much cheaper, the United States goes fully green.
Capitalism still demands growth. Let’s say the economy grows at a gradual 2.3% per year, and energy use goes up with it. In 100 years, we’d be consuming 10 times as much energy. Where would we find enough land to put 10 times as many solar panels or windmills?
Solving green energy isn’t going to be easy, it will require new technology to store energy. It’s not going to support growth like we had in the past. It might not even support the same standard of living we have today.
Read More:
