Progress Studies Conference - I'll Defend Fusion

The progress studies conference was last month and Scott Alexander gave a rundown of it, including some comments on fusion energy:

Nobody wanted to defend fusion.

This surprised me, as I'm sure if they'd invited anyone from the fusion field they'd have heard a defense. Perhaps this defense would have been unconvincing, but at least then they could claim they gave fusion the opportunity.

They seem to not have invited anyone in fusion though, so I thus feel obligated to provide that possibly unconvincing defense.

Scott's summary goes on:

Fusion promises cheap clean limitless power if only we can solve difficult technological hurdles. But we already know how to produce cheap clean limitless power. The only delay is regulatory, and fusion doesn't solve this.

He's referring to nuclear fission power as the limitless power source we already have. I would quibble that nothing is really limitless, in fact the World Nuclear Association estimates about 90 years of fissile Uranium reserves available at current rates. This is in fact less than coal, which is estimated to have 139 years of reserves available.

(Total years of reserves often go up over time though, as we get better at prospecting and utilization.)

This isn't to say that economical fissile Uranium is likely to be a problem in the near future, more that if you're hoping to transition entirely to nuclear power and use 100X the energy you might indeed eventually run up against limitations.

Fusion has the advantage of using isotopes of hydrogen (the most common element in the universe) as fuel, so is "limitless" in a way that other energy sources aren't.^[1] Even if the world's current energy needs were straightforwardly met by solar or nuclear fission power, humanity would still be working on fusion because it's upside is so unbounded.

Won't people be more willing to tolerate it because it's meltdown-free? No; there are already designs for meltdown-free nuclear reactors, and we don't use them.

Again, there are levels here. There's a difference between "meltdown-free" and "meltdown-inconceivable".

Nuclear fission reactors are self-sustaining chain reactions, so an analogy might be like getting heat from a burning candle. Until it runs out of wax, a candle will stay lit without outside intervention. In a well-controlled environment, it's not going to cause problems and you could even make cool new designs that self-regulate or automatically extinguish in an emergency.

Meanwhile, fusion reactors are like getting heat by rubbing your hands together. It requires constant, intentional effort or heat just immediately dissipates. The effect is not self-sustaining, and there's no way for it to get hot enough to do any real damage–your hands would break first. Similarly, an out-of-control fusion reactor would immediately break itself and destroy the delicate balance required for a continuous fusion reaction to happen anywhere outside of a star.

This means that if you somehow got a pebble-bed or molten salt fission reaction to run away, it would be so shocking that it would have to be gross negligence or outright sabatoge.^[2] However, if you got a fusion reactor to run away and meltdown, you're more likely to end up with a Nobel prize, because with fusion, meltdown is so inconceivable that it would be considered a scientific achievement.

This is important, because it protects fusion from bad faith and low-effort arguments. Saying "Chernobyl" or "Fukushima" can unfortunately work against nuclear power in a way that saying "The Sun" can't. (Though it hasn't stopped some from trying).

Also, fusion is not radiation-free..

Again true, but a question of scale. As nuclear activists like to point out, coal power plants release radiation too, but everyone seems fine with this.^[3]

Fusion, importantly, doesn't necessarily create long-lived, high-level waste. Combine the smaller accident risk, plus no need to indefinitely store waste (currently stored on-site at fission plants), and does it become plausible to regulate fusion reactors more like scientific or hospital equipment?

The conference says no:

...and the radiation alone would be enough excuse to regulate it to death. Is it theoretically possible that - since there aren't existing fusion-related regulations - we'd have a fresh chance to fight the same battle and maybe win this time? Maybe, but everyone there thought that solar or small modular nuclear reactors would be an easier sell.

But the Nuclear Regulatory Commission (NRC) says yes!

The US Nuclear Regulatory Commission (NRC) announced in a unanimous vote that fusion energy would be regulated in the United States under the same regulatory regime as particle accelerator

The first regulatory battle in the war that fission lost, has now been won by fusion. There's never any guarantees, but there's a reason to be hopeful.