5 ms·

Surely they mean per kwh of capacity, since they really can’t mean 31kg of carbon per kwh discharged (which would make batteries 60x worse than natural gas) >

Surely they mean per kwh of capacity, since they really can’t mean 31kg of carbon per kwh discharged (which would make batteries 60x worse than natural gas) > lithium-ion cell … 30.9 kilograms of emitted carbon dioxide per kilowatt-hour generated This still suggests that you don’t breakeven on carbon until 60 full charge-discharge cycles

papercrane 3y ago
What the study actually says: > current Li-ion cell, primary data were used in this study resulting in 50,8 kg CO2eq/kWh in the WCS and 30,9 kg CO2eq/kWh in the BCS. From the context in the study it's clear they're talking about capacity.
thsksbd 3y ago
Which is what? About 25 000 km? Doesn't sound unreasonable. Ive seen studies suggesting it might be an extra order of magnitude. If you assume battery capacity as a limiting factor, EV don't even break even compared to hybrids. Cars, all cars, are terrible. I say this as a car guy who doesn't give a rats butt about GW. The externality of cars are far more damaging than CO2 emissions.
PaulKeeble 3y ago
A battery lasts 1000-3000+ cycles. That takes the emissions of a car down to 6-2% of an equivalent ICE for the life of that battery. Most cars fall apart before the battery reaches that point so recycling of the cells will likely occur but still that is a monumental saving if you can charge that car from clean power. All of those emissions can be replaced with electrical replacements, from the machines that extract and move the material to the processing and making the batteries. So as electrical vehicles take over that number will drop to nothing. There are already replacements for a lot of the big machinery that extracts from mines that can run all the time in the US and other places in the world. It will obviously take an amount of CO2 emissions to transition but the emissions aren't intrinsic to the process but rather the power we are using to make it happen. A common complaint about Solar panels is they take quite a bit of power to make, which is true but if you also look at any of the solar manufacturing plants around the world they are not surprisingly getting their power from their own solar panels as well. When we study the emissions of these things we need to think more about what can not be easily replaced yet, like steel manufacture. Oil also has similar externalities in emissions too from all the extraction and moving of barrels of it around the globe, its one of the largest industries in the world for emissions.
- ProllyInfamous 3y ago
  >A battery lasts 1000-3000+ cycles. Modern lithium typically wears out due to time, not cycle count (when used outside of extremes).
  - marcosdumay 3y ago
    Still, if recharged once a day, that's 3 to 10 years, what's completely in line with the time durability. The time degradation will just bias the lifetime into the short end of that interval. If you have an electric car, and don't use it often, then your numbers will be worse. But if you use it weekly (what is normal for lightly used cars), you need about an year and half to reach the OP's number and become better than the next best alternative.
    - wlesieutre 3y ago
      A charge cycle is one full cycling of capacity, such as 0% to 100% once, 30% to 80% twice, or 70% to 80% ten times. I'd be surprised if there are any EV drivers consistently doing a charge cycle every day, outside of commercial vehicles where the battery size was picked to not be any bigger than its daily route.
      - jandrese 3y ago
        It's not that simple. Going from 100% to 0% is far more damaging to the battery than going from 80% to 70%. This is why most EVs are so conservative with their battery usage, they're trying to avoid needing replacement batteries before the rest of the car has fallen apart.
        wlesieutre 3y ago
        Right, just generalizing on how charge cycles are counted, but I'd probably file regular 0-100s into grandparent comment's "(when used outside of extremes)" The average American drives around 13500 miles per year[1], so supposing a car with 250 mile range that's 54 charge cycles, just about one charge cycle per week. The vast majority of that will mostly be with under 20% of a cycle per day, plus a handful of outlier days every year with longer trips. 1000 cycles at 54 cycles per year would be 18.5 years, and 3000 cycles would be 55.5 years. I'm no expert on lithium ion battery aging, but that sounds like it's well into the range where age is more of a problem than cycle count for an average driver. [1] https://www.caranddriver.com/auto-loans/a32880477/average-mileage-per-year/ https://www.caranddriver.com/auto-loans/a32880477/average-mi...
tzs 3y ago
They say that is a "cradle-to-gate" analysis, which means it covers the cost from gathering resources through manufacturing up to the point it leaves the factory. Consider an EV with a 75 kWh battery pack. Using their numbers that would be 2318 kg of CO₂ to make that battery. If the battery lasts 1000 full cycles that would be 75000 kWh of energy produced. Amortizing that 2300 kg of CO₂ over that gives 0.0303 kg CO₂/kWh discharged. A typical EV gets about 4 miles per kWh, so that's about 0.0076 kg of CO₂/mile driven. A gallon of gasoline releases 8.8 kg of CO₂ when burned. Given an ICE car that gets 40 miles/gallon that's 0.22 kg CO₂/mile. That's 29x as much CO₂ per mile as the EV. Instead of amortizing, and assuming clean electricity, the EV costs 2318 kg CO₂ upfront for its equivalent of a gas tank but has 0 CO₂ operating cost. The ICE has negligible CO₂ cost for its gas tank up front, but costs 0.22 kg CO₂/mile operating cost. Breakeven for the EV vs the ICE is at 10300 miles. For the average American that's about 9 months worth of driving. PS: I took a look at non-clean electricity. Electricity from coal or petroleum burning power plants is about 1 kg CO₂/kWh. An EV gets about 4 miles/kWh so that's about 0.25 kg CO₂/mi if you charge your EV with 100% coal or petroleum based electricity. Slightly worse than a 40 mpg ICE car. The US average is around 0.4 kg CO₂/kWh electricity, which would give 0.11 kg CO₂/mi to operate an EV. That's about 2x the cost of the 40 mpg ICE. Note that the above is ignoring the CO₂ costs of producing the gasoline. Brief Googling suggests that is somewhere from 1.5-3 kg CO₂/gallon from well to gas pump, which would raise the cost of buring a gallon of gas from 10.3-11.8 kg CO₂/gallon, or by 17-34%. That would make the EV slightly better than the ICE when powered from coal or petroleum plant electricity, and 2.3-2.7x better when powered by average US electricty, and 34-39x better when powered by clean electricity.
- seanmcdirmid 3y ago
  Does refining oil into gas have a CO2 cost?
  - tzs 3y ago
    Almost certainly. A more thorough comparison would need to take that into account. It would also want to consider electricity generation because not every place has 100% clean electricity. But if all one is trying to determine is whether 30.3 kg/kWh cradle-to-gate cost of batteries is enough to bring EV cars up to anywhere near the CO₂ cost of ICE cars for a typical use case the ICE car uses way more without taking making the gas costs into account so there is no need to go farther. I've updated the original to include a look at the case of non-clean electricity and to consider the CO₂ cost of producing gasoline.

papercrane 3y ago

What the study actually says: > current Li-ion cell, primary data were used in this study resulting in 50,8 kg CO2eq/kWh in the WCS and 30,9 kg CO2eq/kWh in the BCS. From the context in the study it's clear they're talking about capacity.

thsksbd 3y ago

Which is what? About 25 000 km? Doesn't sound unreasonable. Ive seen studies suggesting it might be an extra order of magnitude. If you assume battery capacity as a limiting factor, EV don't even break even compared to hybrids. Cars, all cars, are terrible. I say this as a car guy who doesn't give a rats butt about GW. The externality of cars are far more damaging than CO2 emissions.

PaulKeeble 3y ago

A battery lasts 1000-3000+ cycles. That takes the emissions of a car down to 6-2% of an equivalent ICE for the life of that battery. Most cars fall apart before the battery reaches that point so recycling of the cells will likely occur but still that is a monumental saving if you can charge that car from clean power. All of those emissions can be replaced with electrical replacements, from the machines that extract and move the material to the processing and making the batteries. So as electrical vehicles take over that number will drop to nothing. There are already replacements for a lot of the big machinery that extracts from mines that can run all the time in the US and other places in the world. It will obviously take an amount of CO2 emissions to transition but the emissions aren't intrinsic to the process but rather the power we are using to make it happen. A common complaint about Solar panels is they take quite a bit of power to make, which is true but if you also look at any of the solar manufacturing plants around the world they are not surprisingly getting their power from their own solar panels as well. When we study the emissions of these things we need to think more about what can not be easily replaced yet, like steel manufacture. Oil also has similar externalities in emissions too from all the extraction and moving of barrels of it around the globe, its one of the largest industries in the world for emissions.

ProllyInfamous 3y ago
>A battery lasts 1000-3000+ cycles. Modern lithium typically wears out due to time, not cycle count (when used outside of extremes).
- marcosdumay 3y ago
  Still, if recharged once a day, that's 3 to 10 years, what's completely in line with the time durability. The time degradation will just bias the lifetime into the short end of that interval. If you have an electric car, and don't use it often, then your numbers will be worse. But if you use it weekly (what is normal for lightly used cars), you need about an year and half to reach the OP's number and become better than the next best alternative.
  - wlesieutre 3y ago
    A charge cycle is one full cycling of capacity, such as 0% to 100% once, 30% to 80% twice, or 70% to 80% ten times. I'd be surprised if there are any EV drivers consistently doing a charge cycle every day, outside of commercial vehicles where the battery size was picked to not be any bigger than its daily route.
    - jandrese 3y ago
      It's not that simple. Going from 100% to 0% is far more damaging to the battery than going from 80% to 70%. This is why most EVs are so conservative with their battery usage, they're trying to avoid needing replacement batteries before the rest of the car has fallen apart.
      - wlesieutre 3y ago
        Right, just generalizing on how charge cycles are counted, but I'd probably file regular 0-100s into grandparent comment's "(when used outside of extremes)" The average American drives around 13500 miles per year[1], so supposing a car with 250 mile range that's 54 charge cycles, just about one charge cycle per week. The vast majority of that will mostly be with under 20% of a cycle per day, plus a handful of outlier days every year with longer trips. 1000 cycles at 54 cycles per year would be 18.5 years, and 3000 cycles would be 55.5 years. I'm no expert on lithium ion battery aging, but that sounds like it's well into the range where age is more of a problem than cycle count for an average driver. [1] https://www.caranddriver.com/auto-loans/a32880477/average-mileage-per-year/ https://www.caranddriver.com/auto-loans/a32880477/average-mi...

tzs 3y ago

They say that is a "cradle-to-gate" analysis, which means it covers the cost from gathering resources through manufacturing up to the point it leaves the factory. Consider an EV with a 75 kWh battery pack. Using their numbers that would be 2318 kg of CO₂ to make that battery. If the battery lasts 1000 full cycles that would be 75000 kWh of energy produced. Amortizing that 2300 kg of CO₂ over that gives 0.0303 kg CO₂/kWh discharged. A typical EV gets about 4 miles per kWh, so that's about 0.0076 kg of CO₂/mile driven. A gallon of gasoline releases 8.8 kg of CO₂ when burned. Given an ICE car that gets 40 miles/gallon that's 0.22 kg CO₂/mile. That's 29x as much CO₂ per mile as the EV. Instead of amortizing, and assuming clean electricity, the EV costs 2318 kg CO₂ upfront for its equivalent of a gas tank but has 0 CO₂ operating cost. The ICE has negligible CO₂ cost for its gas tank up front, but costs 0.22 kg CO₂/mile operating cost. Breakeven for the EV vs the ICE is at 10300 miles. For the average American that's about 9 months worth of driving. PS: I took a look at non-clean electricity. Electricity from coal or petroleum burning power plants is about 1 kg CO₂/kWh. An EV gets about 4 miles/kWh so that's about 0.25 kg CO₂/mi if you charge your EV with 100% coal or petroleum based electricity. Slightly worse than a 40 mpg ICE car. The US average is around 0.4 kg CO₂/kWh electricity, which would give 0.11 kg CO₂/mi to operate an EV. That's about 2x the cost of the 40 mpg ICE. Note that the above is ignoring the CO₂ costs of producing the gasoline. Brief Googling suggests that is somewhere from 1.5-3 kg CO₂/gallon from well to gas pump, which would raise the cost of buring a gallon of gas from 10.3-11.8 kg CO₂/gallon, or by 17-34%. That would make the EV slightly better than the ICE when powered from coal or petroleum plant electricity, and 2.3-2.7x better when powered by average US electricty, and 34-39x better when powered by clean electricity.

seanmcdirmid 3y ago
Does refining oil into gas have a CO2 cost?
- tzs 3y ago
  Almost certainly. A more thorough comparison would need to take that into account. It would also want to consider electricity generation because not every place has 100% clean electricity. But if all one is trying to determine is whether 30.3 kg/kWh cradle-to-gate cost of batteries is enough to bring EV cars up to anywhere near the CO₂ cost of ICE cars for a typical use case the ICE car uses way more without taking making the gas costs into account so there is no need to go farther. I've updated the original to include a look at the case of non-clean electricity and to consider the CO₂ cost of producing gasoline.