Lighting (and computing / electronics) are something of outliers.
Electric motors haven't become remarkably more efficient --- the starting point was already quite high.
Thermal resistance heating (water, space, cooking) operates at a 1:1 ratio.
There's been some increase in refrigeration, air conditioning, and heat pump operations, but that's largely a marginal improvement, certainly nowhere near that of microprocessors and LEDs.
You want to look at net total per-capita consumption. And the good news is that electricity usage in the US does appear to have largely flattned out since 2000. Though I can remember a great deal of efficiency discussion in the decades preceeding that, which ... seem to have had little effect. Consumption has roughly doubled since the late 1960s.
Per capita consumption of electricity in the US went up by nearly a factor of 4 from 1950 to 1973. The increase from there to the peak (early 2000s) was less than a factor of 2. The sudden slowdown in percentage growth of consumption of electricity after 1973 (the start of the energy crisis) was a major factor in the end of the nuclear build out in the US (and the bankruptcy of WPPSS, which had bet too much on the growth continuing.)
Among the first sets of Sankey (energy flow) charts that LLNL did were backcasts to the 1950s and forecasts to 1990. Projected usage for 1990 was nearly 150 quad --- fifty percent more than we're using 30 years later.
> Electric motors haven't become remarkably more efficient --- the starting point was already quite high.
Batteries have, though. And if we're talking about personal transport, the passenger mass to total mass ratio in today's cars leaves room for huge efficiency gains in principle.
The minimum energy use for transportation as dictated by the laws of physics is zero (for round trips). All kinetic energy can in principle be recovered, and the change in potential energy for a round trip is zero. In practice, of course, there are inefficiencies, but it's not entirely obvious how large the losses must be.
There's been phenomenal improvement. Much has been in price, though power density and recharge cycles have also improved.
Ultimately, though, battery performance is bounded by chemistry. There's only so much electrical potential per unit mass and volume. It's good enough for automobiles, and lighter bikes. I'm highly dubious of trucks, though battery-assist plus some sort of catenary / third-rail feed might work. Shipping, high-capacity long-range air, and rail are likely out though.
My view is that the biggest shifts will be in land-use, and transportation and commerce patterns, as distance and transport become more expensive.
Entry-level cars were hitting the $5k price point or less (e.g., Tata, Chinese models). Tesla is clocking in at about 10x that.
Scooters, bicycles, electrfied bikes, public transit, increased walkability, seem more likely.
How is that you came to this conclusion? The larger and heavier your vehicle, the less energy/mass you need to make it move, so relatively batteries become smaller. If they work for cars, they will work for trucks with slack.
(And the correct conclusion is that no, they don't completely work for cars today. They only work in urban environments. But they shouldn't see any physical limit until they are only about 2 times as heavy as diesel (maybe they can improve further, I wouldn't know), and that limit is quite practical already.)
Boats, by their turn, have no restriction at all on batteries power density. They could use today's batteries without a problem. They are all about costs, and physics do not limit those a lot.
> and rail
And well, that one doesn't even need large batteries.
Trucks have greater total mass, but a much higher net:gross weight ratio.
Curb weight of Tesla Model 3 is 1.6 tonnes. This is about 300 kg (23%) heavier than a typical ICE automobile: Mazda 6: 1.3 tonnes, Honda Accord: 1.4 tonnes, Toyota Corolla, 1.3 tonnes.
Payload is 1.5 adult humans[1] at 0.08 tonnes each, total 0.12 tonnes. Payload:gross ratio is 7%.
Note that the additional mass of the Tesla is > 2x the payload mass. Total battery mass is 530 kg, so Telsa have lightweighted their vehicle by 230 kg and still added another 300 kg mass. Total energy storage weighs 4.4 times the human payload.
An "18-wheeler" tractor-trailer weighs about 35,000 lb empty (16 tonnes), and has a maximum loaded gross weight of 80,000 lb (36 tonnes).[2] The average loaded gross weight is closer to 20 tonnes, for a payload:gross ratio of 55%.
Personal automobiles have a vastly greater overhead of vehicle mass to payload, and can accommodate a comparatively larger batter storage pack. This added mass is partially offset by reduced motor, drivetrain, and structural elements.
Cargo vehicle mass is largely dictated by load requirements. I'd be surprised if lightweighting of electrified cargo vehicles, especially of trailers, could achieve much.
Tractor-trailer fuel capacity is typically 120--300 gallons, or 4.8 -- 12.1 MWh. Because electric motive conversion is more efficient than internal combustion, that can be reduced to about 1.6 -- 4.0 MWh.
The Tesla Model S battery stores 85 kWh in 530 kg, or 6.4 tonne/MWh. Multiplying out, we get 10.24 to 25.6 tonne battery to provide equivalent motive power, when factoring in the greater efficiency of electric traction.
Note too that fuel is burned off during travel, battery mass is not. (Air-metal batteries gain mass as charge is expended.)
Engine weight for a big rig is about a tonne.[3] I'm not finding a transmission weight, though I'll assume it's roughly comparable, and that electrification shaves 50% off this. So we lose 1 tonne electrifying the drivetrain.[4] That's ... fairly minimal.
So, our electrified truck starts with a 16 tonne dry mass, drops 1 tonne for drivetrain, and adds 10--25 tonnes of battery storage, for a total vehicle weight of 25--35 tonnes. Our maximum gross weight is 36 tonnes. We have from 1 to 11 tonnes cargo capacity, compared against 4 tonnes average and 20 tonnes maximum for a fuel-powered truck.
Increasing total vehicle mass tremendously increases roadbed deterioration, as well as traffic safety considerations. It is probably not an option.
Other alternatives are lighter loads, more trucks (and drivers, or driver automation), more frequent recharging or in-transit power supply, or battery swaps.
A separate "battery trailer" might permit reasonably swapping batteries. Otherwise, a battery integrated into navigation or trailer chassies doesn't seem reasonably swappable. I've suggested a similar option for passenger vehicles where daily vs. touring ranges could be supported through an additional "touring battery".
Alternatively, trucks could operate either entirely or significantly in a wire-fed configuration, where the vehicle is powered and/or charged from a grid-fed electrical infrastructure. For various reasons, this is probably best done at at lower speeds or when stopped --- the slower the vehicle during the charging phase, the more efficient charge-capacity per unit length of charging infrastructure. Given multiple independent ownership and maintenance of equipment, traffic-induced variances, and the need for physical contact with charging infrastructure (probably), odds are high that this will be a high-maintenance, and failure-prone option.
Other alternatives include far more efficient multimodal transport transistions, with rail used for long-haul segments and dedicated tracked trucks or trollies for local distribution. I've done some casual search of literature on rail technology developments, and in the freight realm have found virtually none. High-speed passenger rail, yes. Greater freight / cargo efficiencies, not so much.
The US freight rail system is tremendously efficient. It's still plauged with delays, congestion, theft (as noted in recent news), and other issues. It is an excellent option for large, massive, and time-insensitive loads. It's less suited to high-value or time-sensitive shipments, particularly of fresh (or even frozen) food. It's cumbersome to navigate for small-scale shippers. And it's an increasingly monopolistic / oligopolistic industry with almost all significant routes in a handful of rail lines. There are five Class I freight railroads in the US: BNSF Railway, Canadian National Railway, Canadian Pacific Railway, CSX Transportation, Kansas City Southern Railway, Norfolk Southern Railway, and the Union Pacific Railroad.[5]
Electric motors haven't become remarkably more efficient --- the starting point was already quite high.
Thermal resistance heating (water, space, cooking) operates at a 1:1 ratio.
There's been some increase in refrigeration, air conditioning, and heat pump operations, but that's largely a marginal improvement, certainly nowhere near that of microprocessors and LEDs.
You want to look at net total per-capita consumption. And the good news is that electricity usage in the US does appear to have largely flattned out since 2000. Though I can remember a great deal of efficiency discussion in the decades preceeding that, which ... seem to have had little effect. Consumption has roughly doubled since the late 1960s.
https://www.eia.gov/todayinenergy/detail.php?id=49036