[norrsson]

Usual disclaimer regarding the energy density of gasoline: current gasoline engines have an efficiency of 10-30% while EVs are around 90%.

So to reach similar kWh/g we're looking at ~3k Wh/kg

[thot_experiment]

I think your point still stands but modern engines cluster toward the higher end of that range and some do exceed the 30% efficiency mark, hybrid drivetrains can approach 40%.

[fpoling]

Indeed. Tesla Model 3 consumes about 50MJ of battery energy per 100 km. Toyota Prius consumes about 4.5 liters of gasoline. That gives roughly 150 MJ. So a electrical car consumes 3 times less or about 33% of energy of one of the best hybrid drivetrain.

[bagels]

Not that any of the tech is practical for every day cars or how they're used, but F1 engines with the multiple energy recovery systems are up to 52% efficient if they are to be believed (we'll ignore the rules subterfuge around tricking the fuel sensors, injecting oil, and who knows what else).

[lukevp]

Is there a similar volumetric equivalent measurement or is it all about energy density by weight? Like, if the batteries are lightweight but massive, that would also be a bit of a problem since the structure to safely transport a large volume could be expensive and heavy.

[sn0wf1re]

Looks like the created cell is 614 Wh/L from the above comment. Gasoline is ~2.2kWh/L [0]. So my take is that even with the created cell the density is not going to be an issue with car or grid batteries -- only <4 times the size even at this non-theoretical cell. Who knows how the packs will be configured though as I am sure airflow will be a design consideration when making larger packs.

[0] This uses the 3kWh/kg that was provided above and a density of gasoline of .75g/mL

    units
    You have: 0.7429 g/mL * 3 kWh/kg
    You want: kWh/L
 * 2.2287

[adrian_b]

A lithium-air battery (in general all metal-air batteries) is likely to have lower efficiencies for a complete cycle than other lithium-based batteries, perhaps not much above 80%, if not even less. The lower efficiency is caused by one of the reactants being a gas, which causes certain thermodynamic constraints.

A fuel cell with hydrocarbons would have a slightly better efficiency than the best mobile thermal engines, e.g. of 60%, while the ideal energy per mass ratio is more than double for hydrocarbons in comparison with lithium-air batteries, so even with a better efficiency lithium can never match hydrocarbons in usable energy per mass, not even in lithium-air batteries.

The claim from the parent article is wrong and it is based on an incorrect method for computing the ideal energy per mass ratio for lithium-air batteries.

[gpm]

> A lithium-air battery (in general all metal-air batteries) is likely to have lower efficiencies for a complete cycle than other lithium-based batteries, perhaps not much above 80%, if not even less.

This paper directly contradicts this claim with actual measurements of efficiency.

> The energy efficiency of the first cycle was 92.7%, and it gradually dropped to 87.7% after 1000 cycles.

Which is centered just above the 90% mark the person you are replying to gave.

[grapesodaaaaa]

> The claim from the parent article is wrong and it is based on an incorrect method for computing the ideal energy per mass ratio for lithium-air batteries.

Can you elaborate for laypersons such as myself?

[andy_ppp]

Does the engine and drive train weight make these calculations even better for electric vehicles?

[TOMDM]

I feel like a good like to like comparison would be the density of battery + motors compared to the density of fuel tank, engine and drive train.