Nuclear fusion: Ignition!


Two notes of caution:

  1. Lawrence Livermore National Laboratory, in the fine print, insists that it is still analyzing the data to “confirm” the “apparent” findings. I think DOE maybe outrunning its punt coverage making the announcement before the data is confirmed.

  2. Fusion has been searching for 2 threshholds: (a) excess energy production, and (b) Ignition. This announcement, if confirmed is only (a). That is, by expending 2.1 megajoules, the process produced 2.5 megajoules. Positive, if true. The Ignition standard is the fusion process produces enough energy to be both (i) self-sustaining (no outside energy needed to run the process) AND then produce energy in excess of that used to create and maintain the fusion.

Today’s announcement is nowhere near Ignition. Neither is it commercially viable in its present form. I hope they make further progress. Practically speaking, fusion is still decades away from commercial viability.


After further checking, they finally say that the ratio is 3.15:2.05 MJ. Nice!

In order to overcome some of the difficulties, a german company (Marvel Fusion) plans to use a femtoseconds laser:

(they want to get a demonstrator by 2026, and commercial power plants by 2030 (wishfull thinking, but let’s hope so! :slightly_smiling_face:))

They managed to do this with one shot, I mean how hard can it be to scale this up to a thousand shots per day to reach the typical output of 1GW of a conventional nuclear plant? (answer: pretty hard, probably) Still very exciting science/engineering.

Using magnets to induce fusion is still considered to be the most likely to reach commercial viability. ITER is currently planned to be ready by 2025 and results from JET are promising.

That said, even that is considered at the very least a decade or more away from commercial viability. And what the NIF have achieved is very impressive.


ITER is beeing delayed due to stress corrosion of some components actually, unfortunately… :slightly_frowning_face:

(first plasmas will be delayed to 2030 due to that… :frowning_face:)

As @RT545 points out, I’m not sure the materials science is up creating hardware capable of doing a thousand sequential shots model. More importantly, the grid is built on continuous voltage at a steady hertz level. A “pulsing” generator would require massive capacitors and step transformers (read as “double the cost to deploy”) to feed the grid at the stable levels needed for computers and sensitive manufacturing.

That’s why large scale, central station power generation converts chemical energy into mechanical energy (creates heat to boil water, creating steam used to drive turbines) with predictable and manageable outputs to power thousands of homes and businesses simultaneously for days at a time.


Guess who is investing in nuclear fusion?


Microsoft just made a huge, far-from-certain bet on nuclear fusion - The Verge

Too much spare cash. I wouldn’t be surprised if some investor activists didn’t try to start replacing board members to demand an increase in dividends rather than things like this.