Amongst the several articles we reviewed that tackle the question: which emits more carbon dioxide, a gas-powered car or an electric vehicle charged from 100% coal power, many leave out an important concept, which is the last key road bump before we bring it all across the finish line. The electricity grid is not 100% efficient.
The kilowatt-hours (kWh, a unit of electricity) that leave the power plant are not the same as the kWh that arrive in your home. Transmission lines lose power. Electrical sub-stations lose power. Low voltage transformers – those grey cylinders that sit on your power line poles – lose power. The grid is good, but by no means 100% efficient. Which means that the carbon intensity of electricity from a coal power plant – which last time we determined was between 2.31 pounds of CO2 per kWh on the low end to 3.38 pounds of CO2 per kWh on the high end – actually should be a little higher.
If the grid is 90% efficient then we take our range – 2.31 to 3.38 – and divide by 90% giving us a new range: 2.57 – 3.76 pounds of CO2 per kWh of electricity that makes it into your home.
And now for the final lap.
• We know the carbon intensity of gasoline per mile: 0.65 pounds of CO2 per mile.
• We know the mileage efficiency of electric vehicles: 2.91 miles per kWh
• We know the carbon intensity of electricity that arrives in your house from 100% coal power plants; 2.57 – 3.76 pounds per kWh.
• All that’s left is to find pounds of CO2 per mile for the coal-powered electric vehicle.
Which is just a matter of dividing the carbon intensity per kWh (2.57 to 3.76) by the mileage efficiency (2.91 miles per kWh) or:
• 2.57 pounds CO2 per kWh / 2.91 miles per kWh = 0.88 pounds of CO2 per mile
• 3.76 pounds CO2 per kWh / 2.91 miles per kWh = 1.29 pounds of CO2 per mile
So the range for a 100% coal powered electric vehicle is 0.88 pounds CO2 per mile to 1.29 pounds of CO2 per mile. Which is greater than the carbon intensity of a gasoline powered vehicle at 0.65 pounds of CO2 per mile. Surprising!
But it’s important to realize that, as with any model, much depends upon the initial assumptions. For one, if we assume a gasoline powered car only gets 20 miles per gallon, then it’s carbon intensity jumps to 0.98 pounds of CO2 mile. Second, our high end coal estimate entailed using the estimate of 8,800 BTU released per pound of coal, but in many cases this number is over 10,000 BTU; the coal number is probably more often closer to the low end estimate (0.88 pounds CO2 per mile) than the high end (1.29 pounds CO2 per mile).
And then there is the reality that there are hardly any regions that have 100% coal power. In California, the carbon grid intensity is about 0.49 pounds of CO2 / kWh using EIA data summarized here.
Lowest Recent World Gas Prices (2008)
So assuming 90% grid efficiency, and 2.91 miles per kWh, the carbon intensity of driving an electric car in California is 0.19 pounds CO2 per mile. Almost 4x better than gasoline powered cars.
And even if you charge your vehicle in more coal-heavy Wyoming, where the grid intensity is 2.11 pounds CO2 per kWh, then your electric car is using 0.81 pounds of CO2 per kWh, right at parity with a 25 mpg gasoline car, and easily outpacing a typical sport utility vehicle or truck which get far worse mileage.
And thus far we’ve left out the financial component as well. Which is very exciting for electric car enthusiasts:
• If gasoline costs $3.50/gallon, then a 30 mpg gallon gasoline powered car pays 11.7 cents per mile, and a 20 mpg gas car pay 17.5 cents per mile.
• If electricity costs $0.10/kWh, which is the national average and closer to what electricity costs at night even in expensive areas, and a car goes 2.91 miles per kWh, then it costs 3.4 cents per mile to drive an electric car. Less than a third the price of driving a gasoline car.
• Plus, while gasoline prices are extremely volatile, nighttime electricity costs should remain relatively low and flat. It is daytime and peak energy costs that are the real driver of electricity price hikes.
We hope this series was educational. We certainly learned something. The impressive inefficiencies and carbon intensities of coal power plants leave even more to be desired than we thought before researching this issue.
But fortunately we already live in world where electric cars are no more carbon intensive than gasoline guzzling ones and usually much better. We live in a world where energy costs are on the side of electric vehicles and where battery research is speeding along (stay tuned for a post on this shortly). And most importantly, we live in world where there is a real way forward away from foreign oil dependence, away from distributed particulate and SOx and NOx pollution, away from the destabilizing geopolitics of oil, and away from energy intensive and destructive process of ripping up the earth for materials only to combust them into the atmosphere and start all over again. That is why we, and many others, are so excited about electric cars.
Thanks for taking the time to do the analysis on this. I know if can vary significantly depending on geographic location, actual vehicle characteristics, driving habits, etc; but it is a great reference point to be able to quote average values from well-known and respected sources.
Now, perhaps you can expand on this with regards to hybrids… obviously better than 100% gasoline engines, and worse than 100% coal fired electric, but I am curious where it lies on the spectrum. And since you have already gone through the analysis, maybe you can make this update?
Side note: I drove a SULEV Honda Civic before moving over seas, and had the pleasure of enjoying both a relatively clean pollution making machine and relatively high fuel efficiency. And the technology is proven — in 2003 one of the big concerns was the battery life, and replacement expense. As of this weekend, the car – which is now owned by a good friend – has 87k miles, zero problems, over 42 mpg average fuel efficiency, and still no sign of deterioration of the batteries.
Our pleasure to run this analysis! It’s a surprisingly complex topic to model out.
Your question about non-plug-in hybrid electric vehicles is a very good one. As it turns out, the analysis for hybrids actually does not involve the electric grid. The reason for this is that non-plug-in hybrid vehicles (a Prius being the most famous example) are still 100% gasoline powered even though they have batteries in them. Among other things they acheive such high fuel efficiencies because they
1) are aerodynamically designed
2) have a very clever system for recharging the battery when you are braking
3) draw down the battery when you are idling in traffic which is where so much fuel is wasted
But at the end of the day, all the power comes from gasoline (and never from the grid). If you never plug the car into anything, you never draw power from another source. So analysis for a hybrid car would involve a calculation only of fuel efficiency, not fuel switching, and is easier to do. The cleanliness of biofuels or natural gas powered vehicles is yet another interesting topic.
The 2007 report by EPRI and the NRDC cited in Part I has some additional good details if you’re interested. I hope this helps.
So in Wyoming an EV is “right at parity with a 25 mpg gasoline car” – well, it seems that they should strongly prefer a Prius in Wyoming, for now, then.
(I wonder if a similarly-sized Versa would qualify as that 25 mpg gasoline car, or a bit better? If a smaller less versatile Leaf is worse in WY than a Versa, I’d say folks there should just stay away from EVs for a while… not that there is probably much risk of most of them doing otherwise anytime soon)
Thanks for doing this analysis, I had been meaning to do similar, but it looks like you’ve been more thorough than I could have managed….
For gasoline, you only looked at the tailpipe emissions. That is far from all the CO2 that gas causes. About 15kWh of electricity is used refining each gallon of gas. Refineries are among the biggest electricity consumers in Cali and Louisiana.
For gas you started with 19.4 pounds Carbon Dioxide per gallon. Using your 3 methods for electricity, this refining energy adds another 38 – 56 pounds of CO2 per gallon. This brings the CO2 for gas to 2.2 pounds per mile. Far higher than the electric vehicle.
Of course, the same argument that the refinery would not be sourced by 100% coal applies equally to the refinery as it does the EV.
These are both astute comments.
In power grid areas with high carbon intensities (e.g. Wyoming), non plug-in-hybrids like the Prius are a great choice because though they are still 100% powered by gasoline, their efficiencies are much higher. Every mile they go, they’re only using about half the gasoline and therefore emitting half the carbon.
This is not a full “well to wheels” analysis. The 2007 report by EPRI and the NRDC cited in Part I has good additional details, but it is not perfect, either. Our analysis starts from your local gas pump and your local coal-powered power plant. But substantial energy is required to bring the gas and coal to each of those places, and calculating this energy is no small task. Energy spent refining the gasoline is part of the calculation, as is: the energy of extracting coal and gasoline from the earth, the energy of transporting the fuels to power plants and refineries, and the energy of transporting gasoline from the refineries to the pumps. One must also ask questions like: what percentage of the coal is surface mined vs. underground mined? If it’s surface mined what percentage is area mined vs. contour mined vs. pit mined? What percentage of the coal is transported by barge vs. rail? Answers to these questions and many more would be required to develop a perfect understanding of the carbon emissions of the fuels we use.
For gasoline, http://www.fueleconomy.gov/feg/label/GHG-emissions.shtml says:
“If you want to compare total tailpipe plus fuel production GHG emissions for an electric or plug-in hybrid electric vehicle to those for a gasoline vehicle, you should multiply your gasoline vehicle tailpipe GHG emissions value on the Fuel Economy and Environment Label by 1.25 to reflect the fuel production GHG emissions for gasoline.”
It’s not clear how they did that calculation, but from that, 25% “extra” sounds like the right multiplier for total GHG for gasoline well-to-wheels.
You’re right though that to be fair, you’d have to try to do the same thing for coal.
One other argument against EVs that I’ve heard is that the fantastic efficiency of the electric motor ignores the INefficiency of the actual (fossil) fuel source. In other words, the gasoline engine is inefficient because the combustion occurs in the vehicle; for an EV, the combustion occurs upstream of that wonderful 90% efficiency, so actual fuel use is nowhere near that efficient for an EV.
But, as you’ve shown, if your goal is reducing g CO2 per mile, it’s really an orthogonal argument…
Great points, Eric, and thank you for sharing the fuel efficiency link.
The 90% efficiency of an electric motor ignores the efficiency of the fuel source, which is why these posts look so closely at coal power plant efficiency. But it’s important to note that fuel source efficiency varies widely based on power plant type. A modern combined cycle natural gas power plant will utilize fuel at about 55% efficiency while an old car engine may burn gasoline with 15% efficiency. Alternatively, an old coal plant may only have 25% efficiency, while a modern car may burn gasoline with 25+% efficiency. So many details, so little time….
One more link to share
http://www.afdc.energy.gov/afdc/vehicles/electric_emissions.php
In my region, an HEV has a lower footprint than an EV, or even a PHEV.
Damned coal mix …. OTOH, I buy wind energy, in theory. I’m not sure if that would absolve me or not!
Eric,
I’d like to add two points to this discussion.
1. The “mean” power source in the region may not be the relevant factor. As an example, in Switzerland, more than half of the electricity is hydro-power, and almost no fossil fuel powered plants exist; the average carbon emission intensity of Swiss electricity is virtually 0. Then people in my country argue, that they therefore can drive e-cars almost w/o creating emissions, they are absolutely wrong: the capacity of Swiss hydro-power plants is almost fixed (dams are already “everywhere where you can build some”): if we consume more electricity, additional power stations are built either in our country or in the surrounding countries from where we may import electricity. For the emissions attributable to my eventual e-car driving it is those from the typical additional plants built today that are relevant, at least if I really care about the environment.
2. Moreover, I consider it always important to point out the “Independence Fallacy”: If you want to do something good for the environment, you may indeed pay money to a company which promises to install wind energy for that money (from a provider which assures you that it constructs plants somewhere where no plants would have been built if its customers wouldn’t pay the extra money; else your money will not truly lead to additional green electricity to be generated as the wind plant may have built even without your contribution). But once you have the possibility to pay for that wind electricity, to be produced independently of your own electricity consumption, to any kWh you consume at home or with your e-car, is to be attributed the emissions of the standard marginal power plant that is to be built (if currently electricity consumption in your region has a rising tendency) resp. to be decommissioned (if currently electricity consumption in your region is decreasing, such that when regional consumption decreases, the typical corresponding change in the power generation park is that one of the stations is earlier shut down).
This implies – “unfortunately” in some sense for our good conscience – that no matter how much we pay for wind electricity, if we have a true preference for emission reductions, we can never drive even an e-car without being responsible for the amount of emissions corresponding to those that are occurring when standard power installations (or decommissions) are carried out in the region.
Thank you for the detailed analysis as well as the caveats!
What I find most exciting about electric vehicles is the unknown potential. Unlike internal combustion engines which have stalled in terms of efficiency, batteries and electric motors have a huge upshot. Faster charging and greater energy density will increase the range and flexibility of EV’s, and with numerous companies competing the price will surely drop. This would address the three main concerns of EV’s: cost, range, cost.
Taking a philosophical approach, EV’s allow multiple energy sources to compete for the same flexible energy currency (electricity). By reducing our energy system to one standardized energy currency, we can focus efforts on optimizing energy generation.
Hi Matt,
Thank you for the comments. That’s a great point that EVs allow greater competition between different energy sources, thank you for bringing that up.
Best,
Brenden
Average electricity production is not particularly relevant for EVs. Most EVs will be charged at night. Outside of the Northwest, coal provides the overwhelming majority of electricity at off peak times.
Gas and hydro, used for peaking, are shut down at night. Coal-based electricity cannot be readily reduced at night, so it (along with nuclear) produce almost all the electricity for EV charging.
Starting the 2012 model year (a few months from now), the CAFE standards will be 36 mpg, which should be the comparison: a new EV vs. a new ICE.
All in all, EVs are not cleaner than new ICEs.