The energy efficiency of gas-powered cars is worse than we thought
A quarter of the energy in gasoline is lost before it gets to the pump.
It’s well-known that cars are ridiculously inefficient; there is a limit on efficiency in any heat engine determined by the Carnot cycle. Then there is the silliness of moving around a steel box with a permanently installed suite of living room furniture and a storage closet on soft rubber wheels.
When you look at the Livermore Labs Sankey chart of energy use in the USA, transportation is the biggest end user at 28 quads, and the least efficient user, with only 20.96% or 5.87 quads converted to useful energy, with the rest lost to heat and carbon dioxide. I always thought that was bad enough, but it is actually much worse.
Writing on Linkedin, Ross Garside starts discussing the basic inefficiency of the car, and then suggests that “the electric power used to make your tank of gasoline EXCEEDS the electric power needed to charge an eVehicle of equivalent range.” I thought, really? Just the electricity used in the production of the fuel? Can this be true?
Actually, it’s not. It’s a misconception that started with Elon Musk in an interview in 2011:
“You have enough electricity to power all the cars in the country if you stop refining gasoline. You take an average of 5 kilowatt hours to refine [one gallon of] gasoline, something like the [Tesla] Model S can go 20 miles on 5 kilowatt hours.”
Elon said it, so it became the conventional wisdom, but it’s wrong; just because something is measured in kWh doesn’t mean it is electricity, it is a unit of energy.
According to various sources, refining one gallon of gasoline requires about 6 kWh of energy, but only about 15% of that is electricity. Most of it is from burning natural gas or other byproducts of the refining process. Someone (I think at the EPA, but so many of the links are dead) converted it all to the metric energy unit commonly used for electricity, which has caused the confusion ever since.
6 kWh of energy in electricity would push a typical electric car about 18 miles at 3 kWh/mile. As energy stored in gasoline is 33.7 kWh/gallon, 6 kWh worth of gasoline would push a car that gets 25 miles/gallon a total of 4.44 miles.
So while it is not true that the electric power used to make your tank of gasoline (0.2 to 0.9 kWh/gallon depending on source) would push a car further than the gasoline equivalent, the total energy used certainly would.
However, the amount of energy used to refine gasoline is still shocking. A full 15% of all the energy consumed for industrial purposes (4 quads using the Livermore numbers) is used for refining petroleum products. And that’s just the start.
Total everything else up along the way, and Stanford University estimates that we are getting 10 to 12 percent efficiency out of the fuel input that pushes our cars, and we haven’t even mentioned the energy and carbon emissions from building the car in the first place.
The ICE-powered automobile is the world’s most effective tool for converting energy into money for oil companies, governments, road builders and developers. The fact that it is only 10% efficient is a feature, not a bug; the more oil burned, the more money made. This is why the energy transition is so difficult and takes so long.
I wrote about this earlier:
Why we keep burning things: It drives the economy
In a post titled Vandalism, with a plan, Bill McKibben describes how the Trump administration is killing Energy Star, auto fuel economy programs, and doing just about everything they can to promote the burning of fossil fuels. He infers from all of this:
The worst thing about an electric car is that it is still a big steel box with permanently attached living room furniture inside. I like this image. I will keep riding my electric bike in preference to my smaller than average electric car - a Bolt. (I get an average of 4.7 miles per kilowatt hour.) The bike is powered by about 0.3 meters of solar panels (calculated from the actual production of the panels on my roof) carrying me more than 50kms on a charge (0.28 kwh), and I get exercise to boot. At the end of a long ride there is no measurable waste heat coming off the motor. What is not to love?
Nice piece, Lloyd. Along these lines, I’d be interesting in finding out if this is true: If we were to take all the land area in the U.S. that devoted to corn FOR ETHANOL and instead use it for photovoltaics (not that we would want to do that), we would generate twice as much electricity as the entire country uses—for EVERYTHING, not just for EVs. Is that true? (I’m not a fan of producing ethanol for vehicles using corn! The net energy of corn ethanol is negative, or close to it—meaning that it takes just about as much energy to produce a gallon of ethanol as that gallon of ethanol contains.)