[kragen]

Aha, thanks for clearing that up.

I think probably trains will have an easier time carrying batteries than electric cars do: a one-tonne diesel internal-combustion-engine car might get (in medieval units) 40 miles per gallon of fuel, while diesel freight trains routinely get 480 miles per tonne-gallon. Teslas need to recharge about every 500 miles, so we should expect battery-powered electric freight trains with the same battery mass fraction as a Tesla to need to swap batteries roughly every 6000 miles or 10000 km. A night train making it through the night isn't going to be a problem.

If that's true, then why haven't batteries already replaced diesel engines in diesel-electric locomotives? I suspect it's a matter of battery costs and network effects. A gallon of diesel is 146 MJ, so a tonne-mile on a freight train costs 300 kJ, or 189 kJ/tonne/km in non-medieval units. Lead-acid batteries only give you roughly 20 kJ/US$, and low-power lithium-ion batteries are usually more like 10 kJ/US$. You get a multiplier of about 3 because diesel engines are typically about 35% efficient and electric motors are about 95% efficient, so you only need 70 kJ/tonne/km. But 500 km of range would still cost you 175 grand of lead-acid batteries for every 100-tonne railroad car in the train, which more than doubles the cost of the train. If you use lithium-ion instead, it's twice that: US$350k a car. So, expect this to take a significant amount of investment, and therefore take a couple of decades—if it happens at all, because quite possibly it's all-around cheaper to use cheap solar energy to produce ammonia or hydrocarbons and burn those on the train.

The inability to detoxify waste is a concern because detoxifying is what we normally do with hazardous waste. Learning to handle hazardous waste in a different way is risky and will involve some accidents. I mean, it already has.