For reasons I won’t go into (given the forum) I have taken a mild interest in TAE technologies a company that is working towards commercial fusion primarily based on their results outlined in the following paper appearing in nature communications (a very good journal) available on their website
As expected they are proposing a very rosy estimate of their chances of success (a reactor producing power connected to the grid by 2031) and I have other reasons to be skeptical (again left unsaid in this forum). But to give them a chance, to someone in the know are the results presented in this paper a giant leap, big step or tiny shuffle on the path towards viable commercial fusion.
I know less than nothing about this field, but that only makes my opinions bolder and uncorrupted by ‘facts’ and ‘coherence’.
This sounds like an experimental result, and they have to now verify it (peer review notwithstanding), work out commercial scaling requirements, get installation designs drawn up, find somewhere to build and get the requisite approvals, convince a utility to invest effort and capital into building and managing potential connection infrastructure, persuade the relevant state and national regulators that it is not the next Three Mile Island, find investors and negotiate returns.
Assuming they started at New Years, they have 6 years to hit the mark - that’s a bit more than 2,000 sleeps. Which suggests they’d not want to enjoy sleep too much to make this happen if it can actually happen.
I am on the same page as @Banksiaman. If they started building today I would be shocked if they could make that deadline they claim.
Construction like this takes a long time and unless they have already started (got the land purchased, plans drawn up and legal hurdles…hurdled) I can’t see hitting that optimistic deadline. That is no comment on the tech…maybe it works or maybe it doesn’t. I do not know.
While MUCH bigger ITER started construction something like in 2010. It is not planned to have operation till 2035 and full operation till 2039.
Granted, that is a whole other level of size and complexity but gives a notion of just how much effort there is in these things. Even a small test platform would take many years to build.
Sorry, mostly being coy, while also wanting to make it clear that I don’t want to get sidetracked by that side of things.
I am taking it as pretty much a given that that the 2031 date is unachievable, but I am still interested as to the extent to which their work indicates an actual advance in the field. It looks like had some sort of success with something related to confinement. But I don’t have enough knowledge in the area to know if this work is something that gives represents a real advance and possibly a leg up over the competition, or if its just something that was expected and anyone with knowledge of the field could have done if they put in the expense and effort.
One discussion of fusion research I’ve read is - so what? You get fusion. Now how do you capture the resultant energy, and turn it into electricity? (Without melting or damaging the containment vessel) Obviousy it can be done, somehow. But this experimental device does not demonstrate that aspect. Presumably, some sort of surrounding jacket that gets heated and then runs a steam turbine like every other power plant. Everyone touts their milestones toward achieving fusion, very few disuss this next step.
IIRC a criticim of cold fusion was that it was not happening because there was no detection of neutron emmission, which is a side effect of the deuterium fusion that was allegedly happening. That would be an added complication of a power plant such as this, capturing neutron radiation.
(The same issue applies to any device that achieves serious levels of fusion reaction).
If they expect to have a working, productive demostrator power generation system online in 2031 they’ve got a lot more work to do, very different than demonstrating fusion. .
[Posting in response to a private request on this issue]
Just to qualify my observations, I am not a high energy/plasma physicist (although I did several years of coursework toward a physics degree before switching to engineering) but have an interest in nuclear fusion power production from both a technical and energy policy aspect, and I have followed the nuclear fusion ‘industry’ (such as it is) for over thirty years in all different modes of achieving controlled nuclear fusion for power production with what I can only characterize is progressive disappointment and disillusionment about the potential for a revolution in energy production. This is not to say that there aren’t honest and devoted people working on the problem for their entire careers, but every leap in computational capacity, magnetic strength potential and plasma heading, et cetera ends up revealing even bigger and more difficult challenges to obtaining sufficient control and extracting useful energy such that the system is capable of net power output.
In general, all commercial research tends to be wildly overstated in terms of achieving technical thresholds and development milestones, and even the International Thermonuclear Experimental Reactor (ITER), a well funded effort by an international consortium supported by some of the most accomplished people in the nuclear fusion research field, is close to a couple of decades behind the original schedule with still several technical hurdles before even achieving first plasma, and is really just intended to be a grand subscale proof-of-concept experiment of a tokamak-type reactor using deuterium-tritium fuel. It is currently expected to achieve first plasma (i.e. a ‘burning’ plasma capable of sustaining nuclear fusion for some controlled period) sometime in the 2030s, and it is anyones’ guess as to when that will lead to a commercially viable fusion reactor costing on the order of US$10B or more per installation. The smaller commercial companies such as Commonwealth Fusion Systems, EMC2, Helion Energy, First Light Fusion, HB11, General Fusion, Tokamak Energy, and TAE Technologies, have all failed to actually demonstrated any kind of working proof of concept for their various approaches, and in many case are pursuing specific technologies to achieve fusion conditions or types of fusion that can be charitably referred to as “speculative”, and then there are a bunch of companies which were incorporated in last seven or eight years that are either transparently scams (which no lab facilities and only vague descriptions of how their approach is supposed to work) or are run by people who are deluding themselves into believing they have found the special sauce that no one in the last eighty years has thought about. As of the past few years, “AI” has been bantered around as a magic pixie dust solution to finding just the right set of parameters to achieve controlled nuclear fusion, because of course “AI” is the solution for everything, but while it is certainly plausible that certain kinds of machine learning systems could be invaluable in developing the algorithms to control highly transient magnetic fields required to stabilize a high energy plasma field, most of this is just complete handwaving by people highly motivated to sell large language model-based systems and transformer architectures as the fix-all to what ails you even though they can’t actually explain how it will solve any specific problem.
TAE Technologies in particular is employing a field reversed configuration (FRC) which ‘pinches’ the magnetic field and uses momentum of the plasma flow to form a virtual spheromak resulting in a compact toroid. This produces a fluctuating confinement field to achieve fusion temperature and pressure conditions in rapid, microsecond-scale timeframes. It is employing an ‘aneutronic’ proton-boron (p-11B) reaction to largely avoid production of neutrons (which are not charged and therefore difficult to directly extract energy from, and also damage materials and pose a hazard to people), instead producing charged alpha particles (essentially, an ionized helium nucleus) from which the momentum can be directly converted into electric current by passing the flow through an a loop or grid. This also avoids the need to breed tritium for the D-T reaction or inane planes to scrape traces amounts of 3He from Lunar regolith or whatever. I’ve read the linked paper which is ‘interesting’ but doesn’t prove viability. The ability to capture the majority of the output so directly is a distinct advantage of aneutronic fusion; however, the ‘triple product’ threshold to achieve conditions for sufficient fusion power output to be self-sustaining are orders of magnitude higher than D-T or even D-3He fusion, and despite more than seventy-five years of research into FRCs and spheromaks in general, no one has really gotten anywhere close to achieving even marginal fusion conditions or demonstrating how to achieve a quasi-stable state to extract energy.
It appears someone already alluded to the “Trump” connection, so to explicate that, late last year TAE Technologies announced a plan to merge with Trump Media & Technology Group, and if there isn’t a hallmark that more clearly identified desperation and a willingness to compromise all principles for survival and graft, I don’t know what it would be. I guess you could opine that they are just trying to get on the ‘right’ side of politics to assure funding and access to government resources, but then if they are really on the path to commercially viable controlled nuclear fusion, I don’t think they have to curry favor with a bunch of would-be Christo-fascists. YMMV.
Although there are some ‘interesting’ phenomena that have been observed as a part of “Low Energy Nuclear Reaction (LENR)” research (the modern term for the Pons & Fleischmann “cold fusion” scandal of 1989, not to be confused with the very real phenomenon of muon-catalyzed fusion which was previously referred to as ‘cold’ fusion), there is no indication of actual nuclear fusion going on at any level that would be useful even as a scientific observation. LENR research has continued (mostly in Japan) at a relatively low level of funding but with some respectable scientists and has a journal and annual conference, but to my knowledge even people working in the field are not under the illusion that it is any kind of path to nuclear fusion power production or any kind of practicable energy system. It is mostly a curiosity that falls at the boundary between solid state and (low) high energy nuclear physics, which may reveal new insights and future technologies but not in high energy systems.,
Youtube: “Fusion Master Class” (A deep survey into different approaches to nuclear fusion and the historical experience by the author of the above-referenced Fusion’s Promise.)
There are multiple big, government-funded fusion research projects out there. Even with the anemic state of government funding of science nowadays, these projects still have a lot more resources available, more consistently, for a longer time, than any startup private company could hope for. All of the best minds in the industry are going to be working for one of these big projects. While it’s conceivable that some small startup might make some revolutionary breakthrough that the big projects have missed, that’s definitely not the way to bet.
I’m aware of that, I just cited it as where I saw the countering logic of “what to expect if you do get fusion.” And at the time it was mentiond that fusion should be producing decent amounts of neutrons. (In Pons’ case, they alleged D-D fusion I assume and the evidence did not bear it out; in which case they tried pulling “aternate facts” about fusion processes out of a bodily orifice.).
You are saying the paper cited uses (tries to use) different fusion (pB) no neutrons. I guess that would solve the neutron capture problem.
As Stranger points out, this particular effort relies on a fusion cycle without neutrons, fortunately. But two issues - fusion generates a lot of heat - what is the mechanism for capture? This is the engineering detail glossed over - presumaby, along with the neutrons - that the fusion generation mechanism will be bombarded with high energy particles and this heat must be siphoned off without disruption to the reaction vessel and its function. I’m not too familar with nuclear details, but I suspect neutrons, known for deep penetration, could cause all sorts of problems with the containment.
Obviously this is something to be dealt with after actually achieving sustainable, break-even fusion, but I suspect it will not be trivial.
The joke goes that commerical fusion is just 30 years away… and has been for 30 years. I think I heard that joke 30 years ago.
I got a tour of the University of Rochester’s Laboratory of Laser Energetics not long after it opened, in the early 1970s. So really, more than 50 years.
The people there were very nice and very optimistic, extremely proud of the lab, as they should have been. They of course admitted that break-even was years off, but that future seemed in sight. Imagine how exciting it looked to an sf nerd before we knew what we know now.
I think @chronos, @stranger, and others have answered this pretty thoroughly. The hard part of fusion isn’t really creating a plasma with a high enough temperature to overcome the repulsion of two positively charged nuclei, it’s holding it together long enough that energy from the fusion reactions can keep the thing hot for more fusion, until you get more energy out than you put in. So far, the only place that’s been done is at the National Ignition Facility at Livermore National Lab in California, where they’ve harnessed about 2 MJ of laser energy to produce about 10 MJ of fusion energy.
Several catches there: First, it took about 400 MJ of energy to get 2 MJ into the laser. So they definitely need more efficient lasers! Next, they’ve done this about 20 times in the last few years, and a power plant would need to do it about 10 times per second in perpetuity. That’s the pitch of the most conservative fusion startup (Inertia).
It may be that a certain configuration of magnetic fields and a lower density that enables continuous operation is a more promising path, but plasmas are famously difficult to confine (hence ITER’s size and expense).
Every couple of months we see an article claiming a big breakthrough in design, and I’ll be excited to see what can happen with a bunch of venture capital to supplement the significant government funding that’s gone into fusion energy. But I can tell you that there are a lot of gates to get through, including the engineering challenges of energy capture without turning your first wall into a friable mess and the fundamental scientific challenge of understanding how material behaves (or doesn’t) when it’s in a high energy state only seen in stars, where gravity does the heavy lifting of confinement.
Thanks for your very detailed and instructive post. This was exactly the sort of information I was looking for. I did notice that they were using a different form of fusion rather than the Deuterium Tritium that others were, which they were plugging as cleaner, but I didn’t realize that it might have advantages in terms of capturing energy. However I also figured that if everyone else was doing Deuterium Tritium there was probably a very good reason.
This other reaction they are working with needs to be at about ten times the temperature and pressure. That’s why the real work is with Deuterium and Tritium.