OK, I admit it--my numbers in my previous posts on this topic dated November 20, 2011 were not quite right and need some revision (see: "How Clean Are Electric Cars" and "How Dirty are the Tar Sands" from 11/21/11)
Since that time, I've learned a few things. The basic argument that electric cars are a lot cleaner than internal combustion engine (ICE) vehicles still holds, but the numbers are changed enough to require revisiting. I think facts are critical, and I see a lot of estimates and guesswork on the internet, so I'm working hard to get this right. It's not all that simple, but I'll try to explain it as clearly as I can. The main corrections have to do with how much carbon dioxide (CO2) is produced by utilities in the generation of electricity (a little more than I had counted before because the previous numbers did not include the full life cycle of the process) and how much CO2 is generated in extracting and refining oil (a little less than I had counted before, as I'll explain below). The main effect of these corrections is that I can no longer say that an electric car is cleaner than a car getting 40 miles per gallon of gas if the electricity is generated solely by coal. So here is Take Two:
One criticism of electric cars is that they require fossil fuels to generate the electricity. Of course this is true, and that is why I support solar panels and renewable energy as a way to try to zero out the carbon footprint of our cars. However, my contention is that we don’t have to wait for our electric grid to be more renewable, or for our homes all to have solar panels, for electric cars to be superior to internal combustion engine (ICE) cars with respect to greenhouse gases (and in other ways as well).
Here are the updated figures:
For an average utility with a mix of coal, natural gas, nuclear power, hydroelectric, and some renewables, switching to an electric car would reduce CO2 emissions by 45% for an average car. This is less than the 69% I had estimated in my previous post, but it is still significant. Here in the San Francisco Bay Area, where PG&E uses almost no coal to generate electricity, the savings would be 70%. Even if your utility burns only coal to generate electricity, the reduction would be 12%. (See Table III, below, for further discussion about dirty oil--tar sands.)
Even compared to a car that gets 40 mpg, electric vehicles reduce CO2, unless the electricity is generated 100% by coal. The numbers here are:
Average power plant—electric car reduces CO2 by 12%
PG&E, non coal utility—electric car reduces CO2 by 52%
Utility using only coal—electric car increases CO2 by 40% (ouch!--see discussion below)
Calculations and the data (all numbers are subject to change, so please correct me if I’m wrong!):
Computing how much CO2 is created by both ICE vehicles and for electric cars is a bit of an adventure. The key numbers that are revised in this post are shown in Table I below. For the adventurous, I'll include how these numbers were calculated later in this very long post (I hope you enjoy numbers!).
Table I: Key data items for computing CO2 produced by electric vs. ICE cars
Data items that have changed | Current number | Previous (11/20/11) Number |
Pounds of CO2 per gallon of gasoline | 23.5 | 27.5 |
Grams of CO2 per mile for average ICE car | 501 | 583 |
Average mileage of ICE car (miles per gallon) | 23.4 | 21.5 |
Grams of CO2 per mile for car getting 40 miles per gallon | 314 | 312 |
Grams of CO2 per kwh for an average U.S. utility | 700 | 587 |
Grams of CO2 per kwh for a hypothetical all-coal power plant | 1200 | 1044 |
Grams of CO2 per kwh for PG&E (Northern California Utility where I live): | 310 | 238 |
Miles per kwh for our Chevy Volt | 3.08 | 3.3 |
Here are calculations showing where the numbers in the table above come from:
Pounds of CO2 per gallon of gasoline:
Carbon dioxide emissions from burning a gallon of gas include:
Feedstocks/Extraction— 1 pound of CO2*
Refining-- 3.1 pounds CO2* Burning-- 19.4 pounds per gallon (see for example: ehow.com)
Total-- 23.5 pounds of CO2 per gallon.
Note that the 23.5 pounds figure does not include deforestation, oil spills, or other sources of CO2, but I don’t know how to quantify those in terms of CO2.
*. The Argonne National Laboratories Greenhouse gases Regulated Emissions and Energy Use in Transportation (GREET) program provides a spsreadsheet that allows dozens of assumptions to be input--e.g. fuel mixes, composition of electricity generation sources, fuel economy, etc. GREET assumes an average fuel economy of 23.4 miles per gallon (mpg). Table II, below, shows the results of running the GREET model for a standard ICE car with their default assumptions. Note that 20 g/mile(highlighted in table) for feedstock x 23.4 mi/gal = 468 grams/mile = 1 pound/mile. Likewise 60 g/mile x 23.4 mi/gal = 1404 g/gallon = 3.1 pounds/gal. Also note that the total is 457 g CO2/mile. According to GREET the CO2 equivalent, which includes other greenhouse gases, is 478 g/mile. However, I'm using the CO2 figure to keep it simple.
Table II: GREET Results For Conventional Gasoline Car
Btu/mile or grams/mile | |||
Item | Feedstock | Fuel | Vehicle Operation |
Total Energy | 320 | 816 | 4,908 |
Fossil Fuels | 313 | 722 | 4,806 |
Coal | 35 | 47 | 0 |
Natural Gas | 209 | 387 | 0 |
Petroleum | 69 | 288 | 4,806 |
CO2 (w/ C in VOC & CO) | 20 | 60 | 377 |
CH4 | 0.444 | 0.207 | 0.015 |
N2O | 0.000 | 0.005 | 0.012 |
GHGs | 31 | 67 | 381 |
VOC: Total | 0.018 | 0.117 | 0.180 |
CO: Total | 0.029 | 0.033 | 3.745 |
NOx: Total | 0.135 | 0.099 | 0.141 |
PM10: Total | 0.013 | 0.024 | 0.029 |
PM2.5: Total | 0.008 | 0.011 | 0.015 |
SOx: Total | 0.056 | 0.068 | 0.006 |
VOC: Urban | 0.003 | 0.073 | 0.112 |
CO: Urban | 0.001 | 0.013 | 2.329 |
NOx: Urban | 0.006 | 0.033 | 0.088 |
PM10: Urban | 0.000 | 0.007 | 0.018 |
PM2.5: Urban | 0.000 | 0.004 | 0.009 |
SOx: Urban | 0.004 | 0.027 | 0.004 |
If the oil comes from tar sands, it adds 4.5 pounds CO2 per gallon. This estimate comes from: Department of Energy figures. Table 2.4 on page 12 of the DOE report says that a barrel of conventional crude oil produces 24.5 kg CO2 while Figure 2.4 on page 13 shows that a barrel of Canadian oil sands oil produces 111 kg CO2 for tar sands. This ratio of 111/24.5 = 4.5, multiplied by 1 pound per gallon for extraction of conventional oil, gives 4.5 pounds per gallon for tar sands oil. [update 3/15/2012: A better figure is 5.3 pounds per gallon equivalent CO2 since there is also methane released in the production of tar sands. Source: GREET) Table III below uses the 5.3 pounds figure as of 3/15/2012]
Grams of CO2 per mile for internal combustion engine (ICE) cars:
As noted above, and conveniently for arithmetic, the average mileage in the U.S. is now 23.4 miles per gallon according to GREET. This means that 23.5 pounds of CO2/gallon divided by 23.4 miles/gallon = 1 pound per mile = 454 grams CO2/mile. Add to this 47 g/mile for the production of the car. (National Academies study page 440). According to the National Academies study, the manufacture of an electric car produces 3 grams per mile more than an ICE car--i.e. 50 g/mile--presumably because producing the materials in the electric car--e.g. batteries-- produce more CO2 than the materials in an ICE car. Adding 454 g/mile for operation + 47 g/mile for the car's production gives a total CO2 for an ICE car of 501 grams/mile.
If a car gets 40 mpg, it produces 23.5 pounds per gallon/40 miles/gallon = 0.588 pounds of CO2/mile = 267 grams/mile. Adding 47 grams for production of the car gives 314 grams per mile.
If the oil is 100% tar sands, then there are 28.8 pounds of CO2 equivalent per gallon/23.4 mpg = 1.23 pounds per mile = 559 g/mile. Adding 47 gives 606 g/mile.
For a 40 mpg car with tar sands oil, there are 28.8/40 = 0.72 lbs/mile = 327 g/mile + 47 = 374 grams/mile.
As shown in the table below, if gasoline comes from tar sands, the reduction in CO2 is greater than for conventional oil. However, an electric car running on electricity from 100% coal, still produces 18% more CO2 than a car that gets 40 mpg, even if that car runs on tar sands oil.
Today (3/15/2012) I just read that tar sands also have a net amount of CO2 added since the land cannot be restored to the way it was before mining. The peat bogs where the tar sands are located are good carbon sinks, but the land they are replaced with (hopefully replaced, that is) are not as effective. Thus there is an additional net increase due to tar sands mining. I'm not sure how much it would be, but it is substantial. See: Global Edmonton
Today (3/15/2012) I just read that tar sands also have a net amount of CO2 added since the land cannot be restored to the way it was before mining. The peat bogs where the tar sands are located are good carbon sinks, but the land they are replaced with (hopefully replaced, that is) are not as effective. Thus there is an additional net increase due to tar sands mining. I'm not sure how much it would be, but it is substantial. See: Global Edmonton
Table III: Reduction in CO2 Comparing Tar Sands and Conventional Oil
Tar Sands | Conventional Oil | |
Volt vs avg. ICE; std grid | -54% | -45% |
Volt vs avg. ICE; pg&e | -75% | -70% |
Volt vs avg. ICE; coal | -27% | -12% |
Volt vs 40mpg; std grid | -26% | -12% |
Volt vs 40 mpg; pg&e | -60% | -52% |
Volt vs 40mpg; coal | +18% | +40% |
Tar Sands Mining in Canada
Grams of CO2 per kilowatt hour
According to the Energy Information Agency (EIA), electricity production created 2,389 million metric tons of CO2 in 2010. The EIA also reports that there were 4,120 billion kilowatt hours of electricity produced (you need to scroll to the right in Table 8.2a below to find the 4120 number). Dividing these numbers gives 580 grams per kwh. (2,389 x 1012 grams of CO2 ÷ 4.12 x 1012kwh). The EPA reports an average of 545 grams per kwh for 2009, but I'm using the higher number to be conservative. 580 g/kwh does not include the cost of feedstock/extraction and transmission losses. The GREET program estimates that for the standard mix of coal/natural gas/nuclear/hydro/renewables, there is an additional 11% CO2 created in the feedstock process and there are 8% losses in transmission. Multiplying 580 g/kwh x 1.11 ÷ .92 = 700 g/kwh.
Below are tables and links used to provide these numbers (highlighted):
Table 8.2a Electricity Net Generation: Total (All Sectors), 1949-2010
(Sum of Tables 8.2b and 8.2d; Billion Kilowatthours) (scroll right for totals)
(Sum of Tables 8.2b and 8.2d; Billion Kilowatthours) (scroll right for totals)
Year | Fossil Fuels | Nuclear Electric Power | Hydro- electric Pumped Storage 5 | Renewable Energy | Other 10 | Total | |||||||||||||||
Coal 1 | Petroleum 2 | Natural Gas 3 | Other Gases 4 | Total | Conventional Hydroelectric Power 6 | Biomass | Geo- thermal | Solar/PV 9 | Wind | Total | |||||||||||
Wood | |||||||||||||||||||||
2009 | R 1,755.9 | R 38.9 | R 921.0 | R 10.6 | R 2,726.5 | R 798.9 | R -4.6 | R 273.4 | R 36.1 | R 18.4 | R 15.0 | R .9 | R 73.9 | R 417.7 | R 11.9 | R 3,950.3 | |||||
2010P | 1,850.7 | 36.9 | 981.8 | 11.2 | 2,880.7 | 807.0 | -4.1 | 257.1 | 38.0 | 18.6 | 15.7 | 1.3 | 94.6 | 425.2 | 11.3 | 4,120.0 | |||||
The following table is from Energy Information Agency Electric Power Annual 2010:
Electric Power Annual 2010 | |||||||||||||
Released: November 2011 | |||||||||||||
Next Update: November 2012 | |||||||||||||
Table 3.9. Emissions from Energy Consumption at Conventional Power Plants and Combined-Heat-and-Power Plants, 1999 through 2010 | |||||||||||||
(Thousand Metric Tons) | |||||||||||||
Emission | 2010 | 2009 | 2008 | 2007 | 2006 | 2005 | 2004 | 2003 | 2002 | 2001 | 2000 | 1999 | |
Carbon Dioxide (CO2) | 2,388,596 | 2,269,508 | 2,484,012 | 2,547,032 | 2,488,918 | 2,543,838 | 2,486,982 | 2,445,094 | 2,423,963 | 2,418,607 | 2,470,834 | 2,366,302 | |
Sulfur Dioxide (SO2) | 5,401 | 5,970 | 7,830 | 9,042 | 9,524 | 10,340 | 10,309 | 10,646 | 10,881 | 11,174 | 11,963 | 12,843 | |
Nitrogen Oxides (NOx) | 2,491 | 2,395 | 3,330 | 3,650 | 3,799 | 3,961 | 4,143 | 4,532 | 5,194 | 5,290 | 5,638 | 5,955 | |
How much CO2/kwh is produced by coal power plants?
From Table 8.2 shown above there were 1756billion kwh of electricity produced by coal in 2009. From the U.S. Environmental Protection Agency (EPA) "Emissions from Energy" (page 3-8) there were 1,747.6 trillion grams of CO2 emissions from coal. This gives 995 grams per kwh. Multiplying that by 1.11 for extraction/feedstock CO2 and dividing by 0.92 for transmission losses, gives 1200 grams/kwh CO2 for 100% coal plants. I had previously used the figure 2.3 pounds per kwh (1044 g/kwh) as provided by the Carbon Dioxide Information Analysis center. I'm not sure why their figure is lower than the 1200 g/kwh calculated above, but I'll use the higher number to be conservative.
Note also that older, dirtier coal power plants are being closed on a regular basis, so the coal numbers are coming down. The break even point for electric cars vs. 40 mpg cars comes when a power plant produces 822 g/kwh.
How much CO2/kwh is produced by PG&E (non-coal utility)?
I previously used the figure 238 g/kwh as reported by Pacific Gas & Electric(PG&E).
However, as mentioned, this does not include feedstocks and transmission losses. Unfortunately natural gas has significant methane leakage in production (see the film Gasland). GREET estimates 20% additional CO2 equivalent for this methane loss in their California mix, so I'll use this same estimate for PG&E. I've seen news reports that the methane leakage is greater than previously assumed, so stay tuned for updates.
The current assumptions give an estimate for PG&E of 238 x 1.2 ÷ .92 = 310 g/kwh
How much CO2/kwh is produced by solar panels?
Since solar panels do not create any CO2 while operating, this question has to do with the manufacture and recycling of used solar panels. According to the National Academies Study, solar panels result in 20 - 60 g of CO2 per kwh (Page 143). To be conservative I'll use 60 g/kwh.
How much CO2/mile does the Volt create?
That depends on how much CO2 is created in generating the electricity. Our Volt goes an average of 3.08 miles per kwh (at least that's what we got for the first year). This means that the Volt creates 700 g /kwh ÷ 3.08miles/kwh = 227 g/mile with the standard U.S. mix of energ sources. Add to this 50 g/mi for the production of the vehicle National Academies Study, page 440, gives 277 g/mile.
For PG&E with 310 g/kwh ÷ 3.08 = 101 g/mile + 50 g/mile for production = 151 g/mi.
For coal with 1200 g/kwh ÷ 3.08 = 390 g/mile + 50 g/mile for production = 440 g/mi.
For solar with 60 g/kwh÷ 3.08 = 19 g/mile + 50 g/mile for production = 69 g/mi.
Calculation of CO2 reduction of Volt vs. ICE car.
The ICE car generates 501 g of CO2 per mile, so the savings of the Volt compared to an ICE car is 501-277/501 = 45% for a standard grid.
If the ICE car runs on tar sands oil, it generates 590 g/mi, so the savings would be 590-277/590 = 53% as shown in Table III.
For PG&E, with 310g/kwh ÷ 3.08 mi/kwh = 101 g/mile. Add 50 g/mi for production of the car = 151 g/mi. Reduction = 501-151/501 = 70%
If only coal is used to generate power, it creates 1200 g CO2/kwh. For this a Chevy Volt would generate 1200/3.08 = 390 g CO2/mile. Add 50 g/mil for production of the car = 440 g/mi. This is 501-440/501 = 12% less CO2.
For electricity generated by solar the savings is 501-69/501 = 86%.
Comparison of Volt vs. 40 mpg car
As discussed above, If a car gets 40 miles per gallon, then it produces 314 g/mile. Compared to this 40mpg car the Volt saves 314-277/314 = 12%
Compared to PG&E electricity, the Volt saves 314-151/314 = 52%
However for a 40 mpg car in an all-coal area, the electric car creates 440-314/314 = 40% more CO2!
A 40 mpg car running on tar sands oil generates 365 g/mile. The Volt saves 365-277/365 = 24% given a standard grid.
For electricity generated by solar the reduction is 314-69/315 = 78%
Do solar and electric cars provide the answer to stopping global warming?
As noted above, if all the electricity for a car is generated by solar, the savings is 86%. However, as I've argued in previous posts and in the paper I co-wrote in 2010, "What Will it Take to Stop Global Warming--The Case for Electric Cars", we need to cut CO2 down to one ton per person. If current usage in the U.S. is 18 tons per person, that means we need to cut CO2 by 17/18 = 94%. So solar panels and electric cars are getting there, but not quite. I think that if all the electricity to manufacture cars and solar panels is also generated by solar/wind and other renewables, and if all the mining is done with trucks and machinery running on low carbon biofuels, then we would achieve the 94%. So this is do-able, but we have a lot of work to do. Also, if we allow urban sprawl to continue, we will need ever more cars and solar panels, which will produce more CO2, so we need to work on city planning that does not increase vehicle miles traveled.
And, of course, the Volt uses gasoline after 36 miles--e.g. on road trips. So we still need a gasoline substitute or much stronger batteries to get away from fossil fuels.
Another correction--confusion about energy and electricity
In my previous post on the topic of how clean are electric cars, I argued that an electric car can go about as far on the electricity used to refine a gallon of gas as an ICE car can go on that gas. This argument was based on the fact that about 6 kwh of energy are lost during refining, and about 2 kwh during extraction/feedstock production. If this energy were electricity, that would be enough to go about 8 x 3.08 = 25 miles, i.e. more than an average car goes on a gallon. However, the 8 kwh are mostly not electricity. To produce 8 kwh of electricity would require about 20 or more kwh of energy. So it is not true that we can drive our electric cars as far on the electricity used to refine a gallon of gas as an ICE car can go on that gallon. I think a closer estimate is that we can drive about half as far on the electricity used to refine a gallon of gas--say about 10 miles, as a car can go on that gallon. That's still significant, but not as dramatic as my previous claim.
This same confusion between energy and electricity led me to overestimate how much CO2 is produced in extraction and refining of gasoline. If there were 6 kwh of electricity used to refine a gallon of gas, my estimate would have been OK. However, this 6 kwh figure is for all energy used in refining, not just electricity. Electricity creates more CO2 per kwh than natural gas, oil and hydrogen which are also used in refining oil, so the calculation was too high. I had estimated 6 pounds CO2 per gallon for refining; the updated estimate is 4.1 pounds per gallon as discussed previously.
National Academies Study/GREET Data The 2009 National Academies study has a lot of useful information. I used its estimate for how much CO2 is produced by solar panels
I also used its finding that an ICE car produces 47 g/mile due to its production and an electric car produces 50 g/mile.Page 440.
However, I found that the report, which relies on GREET data from 2009, overestimates emissions by electric vehicles because it uses older numbers for emissions by average utilities (628 g/kwh vs. 580 g/kwh, which I used), and it uses 2.4 miles/kwh for electric vehicle compared to the 3.08 mi/kwh actually measured for my Volt. Note that the GREET model is currently using 2.8 miles/kwh because they feel that this is appropriate for a mid-size electric car. Personally I think the Volt meets most of the needs of people who buy mid-size cars--smooth and comfortable ride, safety in crashes, and attractive appearance. It doesn't have room in the back for three people, however, so it will be interesting to see what true mid-size electric cars get in terms of miles per kwh. In this regard, I'm encouraged by recent reports that the energy density of batteries (how many kwh per pound) is likely to improve dramatically in the next few years. See for example: Oakland Tribune Report
What About Other Pollution from Coal?
There have been recent reports (e.g. TIME) arguing that electric cars put out more pollution than ICE cars in China, which relies heavily on coal. As discussed extensively above, this is true of CO2 for cars that get 40 miles per gallon. The real focus here, however, is all the additional pollution that coal plants create. I agree that this is a serious health problem for China, and for the U.S., and wherever coal plants exist. Fortunately in the U.S. coal plants are finally being regulated under the Clean Air Act. The Union of Concerned Scientists reports that "by 2016, the new standards will lower power plant emissions of mercury 90 percent and acid gases 88 percent". (Spring 2012 Earthwise newsletter) Hopefully China will follow suit and start cleaning up their coal plants. Even more important, however, is to close coal plants asap since they create so much more CO2 than natural gas. Natural gas is no ultimate solution either since it is still a fossil fuel.
Conclusions
I have two basic conclusions:
1. Electric cars are cleaner than most internal combustion engine cars, especially when the grid is powered by 50% or less coal. Electric cars are dramatically cleaner when powered by solar panels.
2. To even have a chance to stop the rise in CO2 that is causing global warming it will be necessary to switch to 100% electric cars powered by 100% renewable energy.
Whew! If you made it this far, you deserve a break. For a change of pace check out my song about global warming and CO2: "We Don't Want To Set the World on Fire"
No comments:
Post a Comment