102

Recently I read on Greenpeace's website, that the production of a 100 kWh battery, as in the Tesla Model S, produces as much CO2 as driving a regular car for 200,000 km.

Source: Tesla car battery production releases as much CO2 as 8 years of gasoline driving:

For every kilowatt hour of storage capacity in the battery generated emissions of 150 to 200 kilos of carbon dioxide already in the factory.

(Original Greenpeace link (in German))

Let's say you are driving a car with 8 L / 100 km. According to this calculator this would result in 16,000 liters of fuel burnt or 36,960 kg of CO2.

A 100 kWh battery takes about 20,000 kg of CO2 to produce. So the math seems to add up.

But is the 100 kg to 200kg of CO2 for 1 kWh battery realistic?

Also, isn't this calculation missing the fact that Tesla covers a part of its factory energy with its own solar panels and plans to use 100% renewable energy in the future?

  • 47
    The point of electric cars is that you can centralize energy production and make it greener there. So Definitely if Tesla starts using renewables (which I believe already does) then the CO2eq will drop significantly. You simply cannot do this with regular cars which require petrol to run. Also: does the 200Mm figure take into account CO2 emissions from building a normal car? It seems it doesn't so we are comparing apple to oranges... You need to get the equivalent distance for building a regular car and subtract them from the 200Mm. – Giacomo Alzetta Dec 11 '19 at 8:30
  • 1
    @Schmuddi I linked the WUWT article because it is in English and makes basically the same claim as the Greenpeace article. I am also only talking about high-powered batteries, like the one in a tesla, so I am not misrepresenting anything. I also provided a calculation. 8l / 100km is not even that economical of a car. – user1721135 Dec 11 '19 at 9:28
  • 1
    In addition to the other issues, is it fair to compare a Tesla S to a 'fuel efficient' gasoline or diesel engine car? You really have to compare like things. Compare the Model S to a more equivalent regular car.. in power, luxury, etc. – CGCampbell Dec 11 '19 at 15:24
  • 11
    There is also a cost, in terms of CO2, for transporting petrol around to petrol stations (everything from extracting it from the ground to huge lorries driving long distances to carry it). If you're going to consider costs you need to look at the whole picture, not just one area of production. – Lio Elbammalf Dec 11 '19 at 16:02
  • 3
    Just to emphasize, this article compares a gasoline car being driven 200'000 km to an electric car sitting still in a garage. Yes, electricity could be produced 100% from renewables, but it currently isn't. And then there's bio-diesel and ethanol which are renewables too. – Dmitry Grigoryev Dec 12 '19 at 13:06
112

tl;dr: The claimed range is 50% higher than the worst assumptions for battery production, and 500% higher than the best assumptions. But it's not an apples to apples comparison.


Carbon emissions from battery production

The range of values estimated for emissions from battery production varies widely in the literature:

  1. In an answer on Sustainability.SE I cited a 2011 study ("Life Cycle Environmental Assessment of Lithium-Ion and Nickel Metal Hydride Batteries for Plug-In Hybrid and Battery Electric Vehicles") giving the highest estimate I could find. Full-text of that article is no longer available, but I've included the values I cited previously.

  2. "Life cycle assessment of lithium-ion batteries for plug-in hybrid electric vehicles -- Critical issues", provides similar values, based on this figure:

Life cycle global warming impacts of two Li-ion batteries

  1. "Life Cycle Assessment of Greenhouse Gas Emissions from Plug-in Hybrid Vehicles: Implications for Policy" gives a much lower value (in the supplementary material).

  2. Commenters pointed out that the report the cited by Greenpeace has been updated, giving a range of values which are even lower.

Here's the values from all sources for comparison, showing the broad range:

[1] Li-ion (LFP-type):  0.250 kg CO2eq / Wh
[1] Li-ion (NCM-type):  0.200 kg CO2eq / Wh
[3] Li-ion:             0.120 kg CO2eq / Wh
[4] Li-ion (high end):  0.106 kg CO2eq / Wh
[4] Li-ion (low end):   0.061 kg CO2eq / Wh

Result for a Tesla Model S

Using these values, production of the 100 kWh battery for a Tesla Model S results in emissions of 6,100 to 25,000 kg CO2eq.

Comparison to ICE vehicle

When a liter of gasoline is burned, 2.3 kg of CO2 are released (source (PDF)). Thus production of a Tesla battery equates to burning about 2,700 to 10,900 liters of gasoline.

For a "regular car" (mentioned in the question) with fuel efficiency of 8 liters / 100 km, this amount of fuel equates to a driving range of 33,150 to 136,250 km.

This is well below the 200,000 km mentioned in the claim. The claim is 47% high for the worst-case battery production emissions, and 503% high in the best case.


What the claim (and this analysis) ignores

If we give the benefit of the doubt to Greenpeace, they're not really trying to compare EVs to ICEVs. They're pointing out that production of batteries is carbon intensive, and giving a handy reference to understand just how intensive it is.

But the comparison, as written, breaks down because of all that it ignores (thanks to the commenters for pointing out all these factors):

  • 8 L / 100 km is arbitrary. Some ICEVs are better/worse.
  • Production of gasoline also results in CO2 emissions, before it's even in the tank.
  • Production of all components of both ICEVs and EVs result in varying amounts of carbon emissions, and can be recycled at end-of-life to some extent.
  • The electric energy used in battery production could be sourced from renewables -- the cited studies assume an "average" mix, comprising some coal and natural gas.
| improve this answer | |
  • 1
    Comments are not for extended discussion; this conversation has been moved to chat. – Jamiec Dec 17 '19 at 13:23
68

In addition to all the other great answers there is one very important fact that's always ignored in these comparisons (Google translation, lightly corrected by me):

The total emissions of petrol and diesel are sugarcoated in this example.

For oil extraction, refinery and transport on tankers, in pipelines and trucks 44 kWh of energy was used for our 6.4 liters of diesel fuel. In other words, with this energy, an electric car would have driven 250 kilometers before the diesel fuel even reaches the tank.

The comparison was "per 100 km" so the electric car actually has an advantage of almost factor 3 in efficiency, unless you assume that fuel magically appears at the gas station. The transport of electricity, on the other hand, is almost completely free and lossless.

I only have a German-language source for this as well, unfortunately: https://www.wiwo.de/technologie/mobilitaet/hajeks-high-voltage-1-nachgerechnet-wann-elektroautos-sauberer-sind-als-verbrenner/25218614.html

Original quote:

Dabei sind die Gesamtemissionen bei Benziner und Diesel hier noch geschönt.

Denn für Ölförderung, Raffinade und Transport auf Tankern, in Pipelines und Lkws wurden 44 kWh Energie für unsere 6,4 Liter Diesel verbraucht. In anderen Worten: Mit dieser Energie wäre ein E-Auto bereits 250 Kilometer gefahren, ehe der Diesel-Kraftstoff auch nur den Tank erreicht.

.

Second of all, the comparison assumes battery production in Asia (as is the case for the Nissan Leaf, for example) with their average electricity mix that's heavy on coal. But automobile companies, especially Tesla, increasingly use renewable energy for the production of batteries and car components with the target to use 100% renewable energy generated from solar panels on top of every Gigafactory. Even without those on-site renewable sources, production of the battery in the US or EU has a higher percentage of renewable energy already from the grid.

See https://www.carbonbrief.org/factcheck-how-electric-vehicles-help-to-tackle-climate-change

So, in conclusion: If you compare a Tesla Model S against the most fuel-efficient Diesel car you can find, ignore production and transportation emission for conventional fuel, take the emissions from the battery production of a Nissan Leaf and assume they are the same for a Tesla battery, twist the numbers some more to support your bias, then you might end up with a 200,000 km advantage.

| improve this answer | |
  • Comments are not for extended discussion; this conversation has been moved to chat. – Jamiec Dec 16 '19 at 8:47
20

IVL, the source of the 150 to 200 kilograms of CO2 figure, recently published a new study that ended up with a much lower estimate: 61-106kg per kWh of battery capacity, depending on the energy sources and efficiencies of different manufacturing plants. "[This] puts it much more in line with other studies."1

Reasons for the difference:

  • The new study leaves out emissions associated with recycling the battery, which accounted for 15kg of CO2 in the 2017 study.1 1 Implies this makes for a fairer apples-to-apples comparison.
  • The chemistry of newer batteries is changing. As a large part of battery-related emissions comes from mining and processing the raw materials needed, this has an impact on the emissions. "There is a trend underway to increase nickel and decrease cobalt in the cathode chemistry being used."1
  • The new study took advantage of more recent data that measures emissions during critical steps in the manufacturing process.1 Battery production becomes more efficient as the process matures.

    • The main energy expenditure during the process happens when the cathodes and anodes of the batteries are made by mixing materials in a solvent (water or otherwise) and the solvent gets evaporated to leave a powder behind.1
      More recent measurements in operating plants gave a much lower figure than the 2017 study, "which estimated 1.6 time to 3 times greater energy use for drying."1
    • The new version also acknowledges that the electricity used in the manufacturing process is coming from cleaner sources and could potentially come entirely from renewables. That helps bring the low end of the estimated range down. Of the estimated 61-106kg of CO2 emissions per kilowatt-hour of battery capacity, 59kg comes from the raw materials used in the battery. Then, the manufacturing process accounts for 2-47kg, depending on the mix of energy sources used.1

    • The 2017 study used a slightly higher number for raw materials—60-70kg of CO2—but estimated manufacturing emissions at 70-110kg.1

To what degrees each step from the manufacturing process contributes to the CO2 emissions associated with battery production, from the 2019 IVL study:1 Energy Demands for a Lithium-Ion Battery Pack: Electronic Parts 9%, NMC111 Powder 37%, Cell Production 19%, Other Materials 35%

| improve this answer | |
2

No.

There's the calculator from Finnish climate change panel at https://www.ilmastopaneeli.fi/autokalkulaattori/

Unfortunately, the calculator is in Finnish, but the tool has quite good defaults. The vehicle 1 is a gasoline powered car (Bensiini), and the second vehicle is a pure electric vehicle (Sähkö) with a battery size of 42.1 kWh.

The life cycle emissions at 14000 km per year even out between 2 and 3 years. Let's say 2.5 years (although to be fair it's closer to 2 years than 3 years). So, about 35000 km and the additional battery production emissions are offset.

The electricity emissions, 0.137 kg / kWh, are approximately what the rather clean electricity production / imports of foreign electricity in Finland produce. Gasoline burning emissions (2.348 kg / liter) are real and I assume the indirect gasoline production emissions (0.655 kg / liter) are approximately correct.

The battery electric vehicle consumes 17 kWh / 100 km, and the gasoline vehicle consumes 7.1 liters / 100 km. Quite realistic figures in my opinion.

By pressing the "info" button, a rather detailed description of the theoretical basis and calculation formulas is displayed, again unfortunately only in Finnish.

If you find the tool interesting, please do send feedback to the authors at palaute@autokalkulaattori.fi and kindly ask an English language version of the tool.

| improve this answer | |
  • This is Finnish, but not the end! – YetiCGN Dec 18 '19 at 13:59

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .