A company called Helion Energy thinks it can deliver that Holy Grail to Microsoft by 2028. It announced a power purchase agreement with Microsoft this morning that would see it plug in the world’s first commercial fusion generator to a power grid in Washington. The goal is to generate at least 50 megawatts of power — a small but significant amount and more than the 42MW that the US’s first two offshore wind farms have the capacity to generate today.
Surprising. A publicity gimmick?
Bwahahahahahahaha (gasp) hahahahahahahaha!
It will be built entirely out of gingerbread by specially trained flying monkeys wearing tiny saftey-fezes.
We already have a convenient fusion power source 93 million miles away. Which conveniently already delivers the power right to our doors. We just need to pick it up & stuff it into wires.
Microsoft is far from infallible, but these seems so outlandish I had to double check to see if the press release is dated 1 April.
I am convinced that the reason that commercial nuclear fusion is consistently 20-30 years away is because of something terrible that happened to space-time when that reactor was turned on in the 1960’s. That event is now receding from us into the future at the same rate we are moving through current time.
From that article:
“This is a binding agreement that has financial penalties if we can’t build a fusion system,” Helion founder and CEO David Kirtley tells The Verge . “We’ve committed to be able to build a system and sell it commercially to [Microsoft].”
It is unclear if Microsoft is providing any on-ramp funding but it seems like this agreement is mostly in service of attracting more venture capital investment; Helion Energy received ~US$500M in 2021 with another ~US$1.7B in funding commitments tied to demonstrated performance milestones. The article, which has numerous misapprehensions about fusion power generation in general, doesn’t go into any specifics about the approach that Helion is using but I believe they are using a field reversed configuration (here is a white paper on the topic by the Redmond Plasma Physics Laboratory, and a series of papers from the (sadly, now closed) University of Wisconsin Fusion Technology Institute). This is considered a promising approach versus many magnetic confinement and virtually all inertial confinement methods to achieve sustained fusion conditions, but there are still some extreme challenges to extracting energy and sustaining confinement conditions to achieve self-sustaining fusion, notwithstanding that they apparently plan to use the D-3He cycle that requires a triple product (the product of density, temperature, and confinement time, n·T·𝜏E) that is an order of magnitude higher and requires temperatures that are nearly an order of magnitude greater to achieve.
The article mentions “Helion, however, says that it has patented a process to make helium-3 itself by fusing deuterium atoms together in its plasma accelerator,” which is possible but misleading because what they really do is produce tritium (3H) and let it decay in to 3He, a decay that has a half-life of ~12.3 years, so it isn’t as if they can produce it in situ during the fusion process, and producing it by a separate process in a “plasma accelerator” is an enormous energy drain. It would seem far easier to produce tritium via neutron bombardment of 7Li or 10B, and since the D-T fusion process that will inevitably occur in this system (and likely a preferable reaction unless they can sustain temperatures exceeding ~104 million Kelvin) will produce a surfeit of energetic neutrons to the tune of 14.06 MeV per reaction. Quite frankly, none of this really makes any sense unless they have some super-efficient high temperature process that would seem to defy feasibility or else have an unlimited supply of magic pixie dust.
Stranger
Have there been any significant advancements in Muon-catalyzed fusion in the last 2 years?
Carl Sagan said:
The cosmos is within us. We are made of star-stuff. We are a way for the universe to know itself.
Joni Mitchell said “we are stardust.” And she said a decade earlier.
That happens. It’s just mildly ironic that it’s being quoted in a thread about nuclear fusion, which has been said to be ten years away every decade since before Woodstock.
OK, so the company is planning on absconding with the seed money and fleeing to someplace with no extradition treaty. Or more charitably, they just go bankrupt when the penalty fee comes due. Or at the absolute possible most charitable, they somehow have enough money to pay the penalty, and they’re just deluding themselves into paying a bunch of money to one of the biggest companies in the world for nothing.
In terms of advances leading to net energy production by macon-catalyzed fusion (μCF), I don’t think there has been anything notable in the last thirty-odd years. Even if you had a process that could inexpensively create muons upon demand, the α-sticking problem appears to be a fundamental limitation that will require new physics to resolve if that is possible at all.
I don’t see any reason to believe this is actual fraud; they are working on fusion technologies and have had independent audits of their progress by outside reviewers (presumably subject matter experts in the field although I don’t have any details). I suspect they just believe they’ve found the special sauce to make net energy production fusion power work, just as many other companies have previously convinced themselves and their investors, and maybe they have but even if they can get net output, delivering a production-ready working system by 2028 seems risible outside of a Marvel movie starring a drunken genius billionaire playboy philanthropist character who just had to be suitably motivated by having a hole blown in his chest to MacGyver-ing a fusion reactor out of a box of scraps.
Stranger
It looks like MacGyver might be the only one able to make fusion a viable energy source out of a hydrogen balloon and a magnet. Do you think the current advancements in fusion technology are steps to a viable energy resource, or just a justification for research money?
If they had a workable reactor design today would it be operable in 2028? What’s the construction time on a commercial nuclear power plant, including getting all government permits?
The news just last month was that NRC has decided to not regulate fusion like fission. For now. But I don’t know the details of what they did settle on.
Also, I’m told that “fusion reactor” is out, in favor of “fusion energy system”. Because reactor = fission. Or something. All the chemEs in the room found this puzzling.
Even a new coal, gas, or hydro plant I’m sure would involve years of environmental impact studies, fights with landowners, etc.
Heck you often can’t put in power lines without a fight. Although this is small, only 50 MW. So I’d expect more along the lines of whatever it takes to put in a peaker plant. Plus a little bit of radioactive stuff.
The gold standard is modular reactors energy systems. Build them in a factory or shipyard somewhere, and then ship them to where the power is needed.
Building the facility on site is generally going to be much more involved.
It should be understood that in a conventional nuclear fission power plant, the actual reactor is just once small (albeit critical) part of the system, which basically consists of a pool, fuel rods, neutron moderator, monitoring and control systems, and the inner coolant loop, but most of the physical hardware and systems are concerned with the outer coolant loop, steam generators, condensers and cooling systems, the steam turbines that generate electrical power, electricity control management, and of course all of the fueling systems and waste fuel management infrastructure. So just having a working ‘reactor’ would be one piece in that overall puzzle.
For a fusion reactor, the same is true although depending on the specifics of the system it may be even more complex, especially if it requires the breading of tritium or the use of neutron-absorbing fissionable materials to feed into a supplementary or hybrid fission system. All of these systems are crucial to the reliability of an operating power production facility so they all need to be proven out at scale. Going from a net power production fuel cycle today (and it can’t be just net plasma energy output but a significant net over unity the total system input) to a fully production ready power generation system is easily the work of a decade even at a highly directed development effort.
I’m not sure what that means but I strongly doubt that the Nuclear Regulatory Commission would just be in abeyance in regulating fusion power generation. The framework for the regulation will depend upon the system but any foreseeable system of nuclear fusion is going to have radioactive or proliferation-sensitive material inputs (tritium, lithium-7, boron-10, thorium-232, depleted uranium, et cetera) as well as radioactive intermediate and waste materials including elements of the reactor itself that will be transmuted by neutron bombardment. Regardless, it would be criminally negligent to allow companies to bring fusion power generation systems online at public expense without significant regulatory oversight because even if the ‘nuclear’ part is as clean as the World Nuclear Association would like you to believe, to place a significant portion of energy demand upon fusion power generation would require high confidence in the reliability of these systems, which again even if demonstrating to proof of concept will require an enormous amount of development to achieve a production-ready status.
Stranger
There is no way you can build enough battery backup to make a solar powered electrical grid. Not even close. Pumped hydro storage could, with massive effort and cost, help distribute daytime power to nighttime, but there’s no way we can build enough storage to replace solar when it goes offline for days/weeks, which it does routinely. Storage also cannot fix the difference between winter and summer, which in the populous northern regions can vary by 3:1 or 4:1.
Solar is only a partial answer. It has to be coupled with baseload energy and peaking power. Peaking power might be provided by batteries (shifting load by minutes to an hour or so to avoid price spikes). Solar will never be more than 25-30% of our power in the northern latitudes (i.e. everywhere above California).
As for fusion, it’s important to remember that we don’t have a power shortage: We have an economical power shortage. Fusion has not reached breakeven yet in terms of input power to output power. ITER might get there. The recent news about breakeven was only from the output of the lasers, not the input. About 10X as much energy went into that system than what came out.
So let’s say we get to breakeven. Then we have to start generating enough excess power that the grid can use it. Then we have to figure out how to do this for a price that’s lower than nuclear or renewables+storage. No one will care about unlimited fusion power if it costs $2/kWh.
We have no idea at this point if fusion breakeven will happen. If it does happen, we have no idea if we can make it cheap enough to sell profitably. If we can, my wild-ass guess would be that economic breakeven won’t be reached for at least 30-40 years, if ever.
Not for the “they” in the post you responded to; they’re not making a thermal power station.
It means they can get by with materials licenses, which are vastly simpler.
Bringing up irrelevant source materials constitutes fear-mongering. They’ll need a license for their tritium inventory. There’s already a framework for byproduct materials.
They’re not “irrelevant source materials” depending on why kind of system is being used, and next time you baselessly accuse me of “fear-mongering” we’ll have a different kind of discussion about what a disingenuous asshole you are.
Stranger
The posts you responded to were about Helion and their claims. There are many reasons to be skeptical of their claims. But steam loops and thorium aren’t on that list.