The usual reminder that laser fusion research is done not for potential energy production, but for development of thermonuclear weapons: https://news.ycombinator.com/item?id=25210756
"The mission of the National Ignition Facility ib to achieve ignition and gain in ICF targets in the laboratory. The facility will be used for defense applications such as weapons physics and weapons effect testing, and for civilian application, such as fusion energy development and fundamental studies of matter at high temperatures and densities. This paper reviews the design, schedule and costs associated with the construction project."
That doesn't exactly follow. Knowledge of plasma has more applications than just nukes. Maybe it will help nuke design, and maybe that is even a goal of the project, but that doesn't mean that the knowledge gained is binned in "weapons" category and nothing else.
And most research fields are not practical at some point in their lifespans, that doesn't mean it will never be practical, or we won't (hopefully) learn things which make other energy technologies practical.
Since testing nuke behaviour is an explicitly stated goal, I don't see how your objection to the assertion that nuclear weapons research is the reason the NIF is getting funded by saying the results could have other uses?
Sure the results can have other uses, but that's not why LLNL gets millions each year for running the NIF.
Given that the last major technological generation of thermonuclear weapons in the US was the W88 (1988) nuclear warhead with a non-spherical primary I have a heard time imagining that spending tens of billions of dollars starting a decade later was the primary motivation of the NIF. The primary, of course, would have had nothing to do with fusion research. It is unclear what the direct application of research on a small target using massive lasers would be to current generation thermonuclear weapons.
Yes, they have stated that the nuclear research labs were contacted by the military; but this occurred after the reported test. Its a stretch to claim that the project is thus military.
China and Russia are or have modernized their weapons. A central issue with nuclear weapons is that they decay over time, their nuclear yield falls, inexorably, year after year. The US needs to build a new generation of warheads in order to remain a credible nuclear power, which is critical for a stable world order in the nuclear age.
There are some interesting issues though: prior generations of nuclear weapons production involved massive release of highly toxic byproducts. In addition to better control and containment, one would reason that processes with less toxic byproducts would also be desirable.
So how do you purify the plutonium cleanly? How do you cut and weld the plutonium to manufacture the parts, cleanly? And how much of this can you tolerate and still produce weapons-grade materials, and weapons-grade parts?
Given some modifications in design to accommodate the environmental restrictions, you need to model the reactions to make sure the new designs work. But even if you model everything, you need to validate the models. So you have to test something. So you need to produce high energy plasma for primary and secondary designers to validate their models and their designs.
The NIF is a weapons lab instrument for studying fusion reactions. The current hollraum is plutonium, designed to irradiate a deuterium-tritium target. If that doesn't sound like a primary acting on a secondary, I don't know what does.
All that is just a high-level counterargument. The real argument though is quite straight forward: LLNL (home of NIF) is a weapons lab. They are not in the nuclear power business.
Due to the moratorium on actually testing the nukes by setting them off, some other means of testing the physics involved is needed to predict how the current warheads behave as they age.
This is a super explicit part of all the NIF press releases, and these "stewardship" experiments is what they were doing most of the time last year and. Here's a quote from the LLNL press release as an illustrative example:
“This result is a historic step forward for inertial confinement fusion research, opening a fundamentally new regime for exploration and the advancement of our critical national security missions. It is also a testament to the innovation, ingenuity, commitment and grit of this team and the many researchers in this field over the decades who have steadfastly pursued this goal,” said LLNL Director Kim Budil.
I can’t speak to NIF but back in undergrad I got to tour a similar facility at the University of Rochester[0] called the laser energetics lab. It’s a similar enormous laser.
If I remember, and it’s surely not exactly the same thing, a lot of the issue is around pumping down vacuum spaces. Targets at the lab were suspended from handgrown spider silk (the lab employed a resident spider raiser/silk puller). And they tried to keep the large laser cavity pumped to vacuum constantly because pumping things to ultra low vacuum with things like veto pumps just takes forever. Stuff outgasses and creates militorr of pressure for annoying periods of time. The same issue shows up post maintenance in semiconductor manufacturing. Prep and recovery for maintenance can take 10x the time of the actual maintenance action itself.
The facility itself was fascinating. Banks upon banks of capacitors in the basement that apparently sound like a gunshot going off when it’s fired. Giggawatts of power for picoseconds.
When I was in school at the U of R, I lived with a woman who worked at the LLE. It was fascinating to hear her stories. She showed me a storage building there that had stacks and stacks of Unix workstations piled up collecting dust, because they had had classified info on them and so weren't allowed to be sold or thrown away even though they'd had all their storage removed. That was painful, all I wanted was one. Nobody would notice. She said no. :( But I did manage to get remote access to their Cray-2 by agreeing to do some free sysadmin work for them.
U of R has some pretty sketchy nuclear past history though, as the hospital ran some tests in the 1940's of injecting some unwitting patients with plutonium, uranium, and polonium to see what would happen. There's an almost abandoned tunnel (actually, it may well be abandoned at this point) that goes underneath Elmwood Ave between the medical center annex and the hospital proper which has a very old, very locked door labelled "isotope storage chamber". Creepy.
> Does anyone know if they still have to wait for months to open the chamber after experiments to wait for the radioactivity to cool down?
I don’t think this has ever been the case. Note that I have been involved with experiments on the NIF, but only from the academic side. The NIF typically fires 1-3 time per day, and I believe that this is mostly an optics constraint. We weren’t producing neutrons though, so YMMV.
Why did they have to do that? If there's radioactive gasses to worry about contaminating something important surely they could pump those out, gotta be a good vacuum system for that chamber anyway...
In magnetic confinement reactors, the runaway neutrons collide with the reactor's shielding and can make it radioactive by neutron activation. I don't know if it's the same for inertial confinement, but I suppose it could be an explanation.
> The results from the experiment on 8 August indicate an energy output of over one mega-joule, which marks the threshold agreed for the onset of 'ignition' and is six times the previous highest energy achieved.
They got 1Mj output from > 500TW input - this is a "major milestone" but only a small step on the path to fusion.
NIF is fascinating, and has been promising for the past decade. The idea of kick-starting fusion with lasers sounds awesome, and then more awesome when you realize just how much energy you need for fusion. They're shooting terawatts of laser at a target the size of a hollow grain of rice, and they're not there yet!
Why are you comparing energy and power numbers directly? Without a time component I can’t tell whether those numbers are ridiculous or impressive.
If your 500TW laser only fires for 1 nanosecond then you’re only putting in 500kJ so you’re getting 200% yield. If you’re firing your 500TW laser for an hour, well that obviously isn’t as impressive.
NIF, remember, is funded by the DoD. It's not a civilian reactor research program: it's a device for testing fusion reactions in the lab now that the Comprehensive Test Ban Treaty means that they're not allowed to test H-bombs out in Nevada.
This is not to say that it isn't fascinating and there won't be spin-off findings of use to the development of civil fusion reactors, but actual reactor research (e.g. JET, ITER, Wendelstein-7X) is all about magnetic containment for protracted periods (tens to thousands of seconds). And ITER, running on tritium and deuterium, is supposed to use 50 MW of heating power to create a plasma of 500 MW (thermal) for periods of 400 to 600 seconds. No need for a 500TW peak power input (actually much, much lower when you look at it in terms of megajoules delivered in a few nanoseconds).
>NIF, remember, is funded by the DoD. It's not a civilian reactor research program: it's a device for testing fusion reactions in the lab now that the Comprehensive Test Ban Treaty means that they're not allowed to test H-bombs out in Nevada.
*DOE’. There is a quasi-separation of powers, where DOE ‘owns’ the weapons and the DOD ‘uses’ them.
The DOE is a fascinating organization btw…anyone who is so nonchalant about public oversight and security practices that the DOD gets angry at them has to be fun at parties.
This article mentions they used 1.5MJ of energy for the fusion process, it seems to be related to this news, on wikipedia you can find a cited NYTimes article stating 80% of the energy input was released.
And I don't get where you get "couple order of magnitude too low" from, 500TW means nothing, it's a workload, not the total work the laser performed. Over a few nanoseconds, that checks out.
(1) That’s from 2018. (2) It does not say what you say it says. It does not state how much energy was “used”. The laser is about 1% efficient. (3) That 80%, which is the number that goes with this recent news, has nothing to do with the “energy input”. It is 80% of the energy that their models say was absorbed by the hohlraum. These are part of where my couple of orders of magnitude remark comes from.
This milestone of ignition is an important one - the energy released by the initial fusion was of sufficient quantity and sufficiently contained that it could in turn cause more fusion, which in turn could cause more fusion. It is analogous to holding a lighter to a piece of wood and getting the wood to continue burning after you take the lighter away. Hence the term ignition.
That said, while our fuel was alight for a little bit, it still extinguished itself pretty quickly. The energy released by the burning fuel was more than what the fuel recieved, but it was still only a fraction of the power output by the lasers (the lighter), which in turn is only a tiny fraction of the power it takes to make those lasers fire.
This reactor can do one of these shots every couple of months. A reactor of this type producing useful amounts of power would need to produce about 41,000 times as much energy per shot, and fire shots at about a 14,000 times higher rate to generate 300 MW of continuous thermal power, which could be converted to about 100 MW of electrical power.
To go further and make it practical, you need to do all that while also decreasing the price tag by about 90% and solving various technical hurdles such as fuel production and reactor maintenance.
On a scale from 1 to 10 where 1 is "might be physically possible" and 10 is "we are ready to build a bunch of these", the technology is approximately a 3.
I can recommend a great article answering that, however unfortunately it's in German [1].
But the gist is that there are two major unsolved challenges.
One is finding a material stable enough to let the reactor run for longer times (all experiments like ITER only plan to run the reaction for very short times, which is okay for research, but not useful for real energy).
The other is finding a way to produce enough tritium for the reaction. It is expected that ITER alone will consume all the tritium that is available and that is currently only produced in some special kind of nuclear reactor. The idea is that eventually fusion reactors will create this tritium themselve, but noone has figured out how to practically do that.
Tokamaks are inherently pulsed machines. No material will change this. Stellarators are very similar to tokamaks but do not use a central solenoid/transformer to induce the poloidal twist: they use the geometry of flux surface/coils/vessel. Stellarators are steady state.
Tritium breeding blankets have been the answer to question 2 for longer than I've been alive.
The real engineering challenges lie in divertor physics and strike point materials.
This sounds interesting. You sound like you work in nuclear physics. Please can you explain, or link to an explanation of, divertor (sic?) and strike point materials?
Im not an expert but someone pointed out yesterday [0] that this technology is a bit further from practical than implied by the article. This milestone of ‘ignition’ is declared because the laser energy making it into the fusion fuel is less than the energy produced by the fusion fuel.
But this is ignoring the fact that there is significant (85% if I remember correctly?) loss of energy as the laser travels through the housing around the fuel. And even more importantly, the laser itself is very inefficient and loses something like 99% of the energy that it pulls from wall power.
So even in lab conditions, we need to produce several times more energy to even call it ‘ignition’. And this is not mentioning the many and difficult issues with scaling from lab scar to real-world power plant scale.
We already have a fusion reactor... it achieved ignition some 5 billion years ago with enough fuel to sustain itself for about 10 billion years (5 down, 5 remaining). It's conveniently located at a safe distance suspended in empty space at the center of our solar system. All we have to do is harvest the energy it's producing, and we have the tech to do that, including a road-map for growing civilization by many orders of magnitude using that energy (up to a full Dyson sphere).
As for the notion that somehow you can harvest useful energy from a fusion reaction using anything less than the mass of a star... well, at the very best it isn't likely to have a very good EROEI (energy return on energy invested) because no matter how much progress we make it will always be expensive to ignite and then contain a mini-star. And it's quite possible, even likely, that the EROEI will always remain negative.
