And Uranium-235 about 1GWh/kg

Uranium-235 is around 24 GWh/kg [1] (24,000,000,000 Wh).

[1] https://www.euronuclear.org/glossary/fuel-comparison/

Whoa so… 100kwh is 40mg! 3grains of sand to run a Tesla

https://www.wolframalpha.com/input?i=100kwh%2F%282.5+gwh%2Fk...

Except a lot more is released at once so it will accelerate like a jet engine on every stop sign

Haha yes I did not think about that. There’s no throttling there

You need to account for all the weight required to turn the radiation into electricity.

That'll make the numbers ... a bit different.

Not necessarily:

https://en.wikipedia.org/wiki/Radioisotope_thermoelectric_ge...

The GPHS RTG contains 7.8 kilograms of plutonium 238 but masses 57 kg in total. It also generates only 300 watts from that 57 kg package:

https://en.wikipedia.org/wiki/GPHS-RTG

You'd need about 3 metric tons of them to power one Model 3 cruising at highway speed (assuming ~16 kilowatts continuous power draw).

Not to mention that you have to mine and refine a couple of tonnes of ore for every kilogram of refined uranium.

I wonder how close to mass production those intermediary technologies are.

Bumping the energy density closer to something like lithium-sulfur would probably make 95% of ICE-based technology scrap heap tech.

Iron-air batteries (1,200 Wh/kg), and in general metal-air [1], might bring a surprise after 2024: December 2022, "Form Energy will site first American iron-air battery manufacturing plant in Weirton, West Virginia" [2].

[1] 2017, Yanguang Li, Jun Lu, Metal–Air Batteries: Will They Be the Future Electrochemical Energy Storage Device of Choice? https://pubs.acs.org/doi/10.1021/acsenergylett.7b00119 Betteridge's law of headlines answers "no", but good overview.

[2] https://formenergy.com/west-virginia-governor-jim-justice-an...

The various "-air" batteries tend to have major downsides...

They tend to get heavier as they discharge. They usually aren't rechargeable (or if they are, only a few times or with much lower energy densities). They tend to self-discharge within a few weeks of non-use.

Yes, there are downsides, as always in engineering, it's a matter of managing the compromises for the current implementation and researching better solutions for the next iteration.

and I think they can't output as much current like the current batteries too.

its worse actually for ICE because you are probably only accounting for engine efficiency but there are also transmission losses to the wheel. Further all the 3000 or so component of ICE weight fair bit too. I have not seen any analysis on combine energy to the wheel/Kg comparison between ICE & EVs but I'd bet it gets significantly worse for IC cars even at 500wh/Kg.

Lithium ion is already over 250 Wh/kg so that calls into great suspect the rest of the numbers.

Not sure what over 250 Wh/kg means. Wikipedia mentions Specific energy 100–265 Wh/kg and Energy density 250–693 Wh/L quoting papers from 2010/1 and 2016/7 [1]. Other sources mention similar numbers (100-265 Wh/kg or 250-670 Wh/L) [2], although the research is ongoing: "Tesla’s new 4680 cells have an energy density of 272-296 Wh/kg and which is considered very high by current standards" [3].

[1] https://en.wikipedia.org/wiki/Lithium-ion_battery

[2] https://www.cei.washington.edu/education/science-of-solar/ba...

[3] https://thedriven.io/2023/04/03/scientists-hail-new-battery-...

Thanks for the sources, as said this shows the article is highly suspect if your "maximum theoretical energy density" figure already is beaten by in-the-market devices.

Doesn't gasoline habe to (typically) go through carnot efficiency limits though when it combusts?

IIRC, EV motors are 90% efficient with battery power --> road power conversion. A typical ICE engine is, what, 30% efficient and maybe a bit more with good turbo design. So practically gasoline is about 4000 Wh/kg?