As far as I know laser fusion is not a practical technology for energy production due to the plasma instability effects, high pellets cost (several orders of magnitude more than necessary for economic viability), huge reliance on a very precise calibration, and low power capacity even in optimistic scenarios. Most of the funding this field receives is motivated by checking models and constants for thermonuclear weapons. Since nuclear tests are banned, it's the easiest way to study plasma in conditions similar to those of a detonated thermonuclear bomb.
"Putting aside for a moment the vexing question of whether a "big science" machine well into the construction phase should still be plagued with numerous unresolved technology development issues, there is considerable evidence that both the science and technology underlying the NIF are not sound."
and
"While a large majority of ICFAC voted in May 1994 to support proceeding with engineering design of the NIF, [20] their decision was once again based on non-peer-reviewed LASNEX code predictions that had been hastily generated in the weeks immediately prior to the meeting. These calculations purported to demonstrate ignition with novel gas-filled hohlraum targets -- predictions that were subsequently not borne out by actual experiments conducted after the meeting."
These big projects are great at eating budget from committees. Tons of meetings, delayed results for decades resulting in limited accountability, opportunities for everyone to be part of the picture and not point fingers at each other. Not to mention bragging rights for “leading” such large projects.
We often bemoan the state of the research job market on HN. I can’t help but wonder what moving funding from these facilities to smaller experiments and research labs would do.
Well I think the catch is that, to use the US as an example, "big science" projects such as the NIF, particle accelerators, synchrotron light sources, and supercomputers are generally all funded by the DOE, while small individual research labs are generally funded by NSF or (if biomedical) NIH.
Critically, the only reason that DOE can support the cost of these "big science" projects is that they are generally "dual use" so to speak, with both basic science and national security applications. A similar dynamic is arguably at play in the funding of the largest projects in other countries as well.
So it is not clear that cutting such "big science" projects would in practice actually open up more funding for small basic science grants, lacking the same national security applications.
Whatever is happening, we're not "padding the pockets" of professors. These people are actually being paid less than jobs at comparable levels in other fields.
> too speculative with an uncertain path to revenue (unlike Space).
It's really funny (and great) that there are actually people now that continue this phrase with the mention "unlike space". It used to be that "the best way to become a space millionaire, is to start life off as a billionaire."
The NIF was never intended to produce power, but it serves as a good science experiment for understanding the challenges of inertial confinement fusion. You need to make something that works before you can make something that works better.
The only fusion devices humanity has yet gotten to produce net power are thermonuclear bombs. Laser fusion tries to replicate those conditions as best as possible without detonating a fission primary. Yes, this means laser fusion is good for validating nuclear bomb models, but it's also the only route to fusion where we know for a fact that every issue is strictly an engineering problem.
ITER is magnetic confinement fusion - the plasma is heated to extreme temperatures but at low density. While theoretically it looks like a much more promising route to a practical fusion reactor, as far as we know there has never been a net-positive magnetically confined fusion reaction in the universe. Plasma physics is incredibly complicated and we've run into a long series of issues with plasma instabilities, we believe that these issues will be overcome but we still can't rule out some unknown physics which renders it impossible.
I personally believe tokamaks like ITER will be the route that leads to energy generation, but putting all our eggs in one basket is risky and inertial confinement experiments like NIF let us answer fundamentally different questions about reality.
The NIF's reactor model doesn't need plasma stability, right? The article points out they're intending to use a similar principle to an internal combustion engine: put in a tiny amount of fuel, blow it up, capture the energy, repeat.
No, it very much does. Laser fusion relies on creation of a spherically symmetric shock wave, which in turn momentarily creates huge pressure and temperature in the center of a pellet, thus igniting the fusion reaction. Any asymmetry (caused either by miscalibration, setup design, pellet flaw or even simple thermal noise) gets amplified by plasma instability effects and results in huge energy losses. The wiki article [0] briefly covers some of such effects.
How does the level of precision here compare to the precision needed to design LIGO?
I remember reading the precision requirements for LIGO and then reading about the efforts needed to achieve that precision and it seemed like magic. e.g. IIRC LIGO has a precision of PORTIONS of a wavelength of light at the end of arms that are kilometers long.
Is the precision for plasma stability that much higher?
(I understand that these are very different systems so asking more from an overall "what can we currently achieve" perspective).
raziel below has an excellent response. To add to their comment, I'll say the difference here is the precision needed for LIGO is with a more or less static equipment you can tweak. This requires plasma stability during a dynamic, very chaotic system. You do tweak the static parts, the laser and related apparatus, the target and what holds it in place, various diagnostics, but the actual process itself is a very dynamic explosion that it itself is hard to understand or predict fully a priori, and certainly not to the degree we understand gravitational waves at the point of detection at LIGO.
I mean, as I said elsewhere, the reason we don't have fusion for any fusion experiment is plasma instabilities, at this point that fact is like saying the reason ice cream melts in the sun is because it's hot. It's a fact of all fusion experiments. The only system we know of that can achieve net gain fusion we know of is gravitational confinement which uses a big mass to contain the hot plasma, also known as a star.
It's kind of hard to compare, especially since I don't know what you mean by precision: the metrics on both projects are very different.
LIGO is all about minimizing losses as the laser light bounces back and forth the two mirrors. The losses here arise from the stack of materials that the mirrors are made out of, which are multilayers of different oxides plus/minus Si or Ge, I forget. So one metric of course is the surface roughness of the material, but then there are also energetic defects called two-level systems in which atoms can absorb a little bit of light to tunnel into another location and thus contribute to the losses by having absorbed laser energy. There's coefficients of thermal expansion, stresses, that all have to be taken into account and tested as you layer all these dissimilar materials that may behave ok at room temperature but not at cryogenic ones for instance.
So LIGO is a game of minimizing losses, because you're after detecting the faintest of signals: a gravitational wave. Their game is all about increasing the signal to noise ratio.
NIF is a monster of energy. There's the whole steering of an enormous laser pulse which is the addition of 192 beams that have to converge into a tiny area the size of a pencil eraser. There's the containment of all this energy into a steel capsule that looks like something out of a sci-fi piece. There's the manufacture of the 2 mm diameter capsule that starts as plastic but is then coated with diamond, beryllium or more plastic and that leads to inherent asymmetries because it's not easy to coat a non-planar geometry. Already this coating assymmetry is very likely to lead to the hydrodynamic instabilities (https://en.wikipedia.org/wiki/Rayleigh%E2%80%93Taylor_instab...) that are obstructing the path forward. The hole with which you fill up the capsule with DT is still a problem, but they try to sweep it under the rug. The surface roughness of the capsule is a problem, contaminants in the diamond or plastic is a problem, a few atomic percent is enough to significantly dampen the amount of X-ray absorption/transmission. Control of the material's density is another one. The work that goes on all of this is tricky because you can solve all of these and control them very well on a planar geometry, but the moment you want to take this onto a sphere it doesn't work as well, we cannot suspend something in Earth's gravity without using a string, which then introduces an asymmetry. So the best we use is we roll the sphere around, and it's not very good for the tolerances that the scientists think we need.
And NIF has a history of escalation. We are currently shooting 1.6 - 1.8 MJ, but the scientists' simulations had predicted kJ range shots in the beginning, and the estimate has continued to climb. You see it in the article itself that they're hoping to get funding to go to 3 MJ. But there are some studies (Halite-Centurion IIRC) from the 50s that showed you needed like 100 MJ to get inertial confinement fusion on a capsule, and I think we've been happily disregarding those results, because, well, politics, job-protection etc. It's a complicated story, shrouded in a lot of secrecy, so I'm glad to read that the NNSA is reviewing it.
So, precision in LIGO, I don't think translates to precision in NIF. LIGO feels more like golf, NIF feels, like some beasts fighting it out. Very different set of challenges, very different resource pools they can draw from (I would argue that NIF can draw more money but less talent because of security clearance requirements restricting employees to be US citizens).
"Capture the energy". They haven't even looked at that step.
Laser fusion is about the least promising, most impractical possible approach to building a power reactor, even if you do manage to get ignition. Since at least the early 1980s, the word on the street has been that the talk about power is just a politically palatable wrapper for the weapons application. And let's just say that some of the people I heard that from, back in the 1980s, were in enough of a position to know that they probably shouldn't have said anything to some random kid like me.
Not to say that there aren't a few benighted souls working on it who've managed to convince themselves it's a power source, but notice that nobody but Livermore has ever bothered to put much into inertial confinement... and Livermore is a weapons lab?
That's fair enough from a political perspective (after all, as they say in the article, >70% of NIF shots are going to [nuclear] stockpile stewardship now) -- but just to give NIF a bit of credit on the "basic science" front, it has turned out to be quite good for studying the properties of matter under extreme compression more generally.
Some mineral physicists I know have used this to write some cool papers about the properties of natural materials at conditions replicating those expected in the cores of giant exoplanets [1-3], for example.
Funny tidbit, there is a huge scramble and fight over these few shot days from different research groups across the US that can actually be used for non-nuke and non-fusion physics, they're called "Discovery Science" shots and there are only like a dozen or so each year. For example, last year 18 shot days were devoted to DS.[0]
Even if sustained fusion were achieved, it has issues with degradation of the vessel/equipment from high speed neutrons?
LFTR scaled down to a closet, so with "mass production" you could just replace the vessel at "reasonable cost", but these fusion facilities are massive and expensive.
You are correct on all points. The public story is energy, the real story is more about stockpile stewardship. Since we will never derive commercial power from fusion, it would be foolish of Congress to continue to fund this sector with that in mind:
Never really considered it before, but a cool by product of this research is the high-energy-density beamline development. This could have many interesting applications, including deep space internet and power transmission ;)
The L4n laser beamline of the P3-installation: Towards high-repetition rate high-energy density physics at ELI-Beamlines
ELI is the future of high power laser science! It's also a good example of how the future of this sort of physics is definitely in the EU and China, not in the US especially as people get impatient with NIF and other similar experiments.
Its a little strange that an article mentioning funding running out a few times never gets around to mentioning just how much funding this experiment has gotten over the years. Feels like a key piece of data that could help inform the reader.
In any case, hope they eventually get this to be net energy-positive.
I imagine the fact that it is a facility mostly for nuclear weapons research at a site that does mostly nuclear weapon work means the exact budget is somewhat opaque.
Am I wrong in thinking that the energy quantities here are shockingly small?
The 1.8 MJ laser array that's being used to trigger these reactions, that's like 400 calories, less energy than you'd get from a ham and cheese sandwich with mayo, and the output they're hoping to achieve, 100kJ, that's 2mL of gasoline.
1 MJ is the optimal energy scale for inertial confinement fusion. Going larger makes the fuel harder to compress, going smaller means there is less fuel that gets compressed; either way your ratio of input energy to output energy goes down.
Also while these energies are low, they are also for very short durations. 1.8 MJ in 1 ns is 1.8 PW.
If I understand correctly how this works, this is a laser pulse of 1.8MJ and that pulse's duration is in the 20ns range.
So it's actually massively powerful but very short.
If you were to burn gasoline at a rate of 2mL per 20ns you would be burning 100,000 L per second, which is a higher rate than Saturn 5 on takeoff so quite a burner...
Such comparisons are almost meaningless. For explosives, its not only about how much energy gets released, but also (almost more so) about how fast it gets released.
For example, per kilogram, TNT releases less than a tenth of the energy that burning gasoline does, and a sandwich stores more energy than the equivalent weight in dynamite (but both are cheating, as TNT and dynamite both include the oxygen needed to ‘burn’ them, whereas gasoline and sandwiches don’t)
And the forms of energy, deliverable size, etc., etc.
Ham sandwich prompt radiation, thermal pulse, blast, and long-term radiation effects are somewhat less than those of the Little Boy gadget. The last may depend on how much hot sauce is applied.
Both weight and volume exceeded the B-29 and would be better suited for a bulk-cargo carrier of about 30' beam, 40' depth, and 300' length.
The relationship between promptness and completeness of reaction is an interesting one, and not limited to TNT and petroleum. The WWII atomic weapons haad a pretty remarckably low efficiency --- most of the fissile material was dispersed before it could fully react. Nuclear power generation is far more efficient. And generally slower.
See related the Beruit 2020 ammonium nitrate explosion:
Wasn't all this effort involved just to prove Fusion is possible before going all into the tech? The machines are massive today but so were early computers.
That 1.8 Mj is the energy in the laser output. I haven’t looked at the numbers for a while, but it probably takes 10x that much energy to operate the laser facility. A commercial fusion power plant will never exist.
And it’s no accident that the article doesn’t mention the laser efficiency. NIF has a public relations office that has honed its game over decades. Their purpose is to influence Congress by selective reporting through press releases and by feeding propaganda into articles like this.
The design has a lot of similarities to NIF in some respects. I aim to create an ultra dense implosion of deuterium ions.
Thats where the similarity ends though. The key to viability in my eyes is to recover the energy of the ions that don't fuse to create another implosion.
In my device, ions leaving the focus are curved by a weak, uniform, magnetic field. Each ion travels along it's cyclotron trajectory to circle back to the focus at the right time for the next implosion.
The linked video and Kickstarter look more like a Farnsworth Hirsch inertial confinement reactor. Fusion reactions do occur in them, but a cold gas cloud forms at the center because of collisions with the physical electrode. What starts as a group of colliding and circulating beams quickly transforms into beam on target. The result is that ions only have a few chances to fuse before they lose too much energy.
My understanding of the Polywell is that it uses a trapped cloud of electrons to form the central cathode, which can help keep deuterium ions hot by reducing grid collisions.
The downfall I see is that you are still trying to "squeeze jelly with rubber bands".
The Brillouin limit describes the maximum steady state density possible in a non-neutral plasma.
In my device, the average density is well below this limit, but is briefly very very high. The goal is to have smooth and steady cyclotron trajectories most of the time, but all these trajectories intersect at the same point simultaneously.
Do you have a paper or technical document available for reference? Im mostly wondering about the standard fusion device questions such as: In the anticipated final device, i assume ion currents will have to be low enough to avoid interfering with the external field? How will charge balance be maintained between ion and electron species? What is the anticipated center of mass energy distrubution for the ions? Does the proposed focusing approach run afoul of the Liouville theorem?
It's been difficult for me to visualize the 6D phase space of the design. I'm starting to see the particles lie on something like a deforming 4D shell that maintains it's volume in 6 dimensions.
Edit (further description of the 4D shell under the Liouville theorem):
When the plasma has large spatial volume, the momentum of particles at any point creates a 1D line forming an empty 4D shell. When all particles enter the 1D spatial focus the particle velocities there are essentially random and have large variation in all 3 spatial dimensions. Essentially a 4D sphere with only 1 dimension in space.
The result is a morphing 4D shell that collapses all particles into a focus with 0 volume once per period.
Actually... On further visualization the plasma seems to just form a 4D sphere that rotates and translates in 6 dimensions. The flatness of the 2 "missing" dimensions corresponds to the quality of the focus and synchronization of the system, so they are not going to ever be completely missing.
The focus is like what you'd get if you were looking at slices of a solid ball. You'd see a point grow into a large circle, then shrink again to a point. It's basically the view on the left side of the simulation on my website.
The right side of the simulation video shows a different cross section. The x axis there is the one dimension of the 4D sphere that stays in 3 space.
There is no peer reviewed paper or technical document. I've got the patent, and various grant proposals, and my various attempts to explain the concept online.
You are right that the ion current will be small. I'm anticipating that the plasma will be a non-neutral deuterium plasma.
The energy distribution will be thermal targeting 15kev for the prototype.
I wasn't familiar with the Liouville theorem. I'm not quite sure how to reconcile this. Seems a bit paradoxical. I'm going to have to investigate further.
The goal they are trying to achieve is ignition, which is the point where the fusion reactions they directly cause with the lasers release enough energy to in turn lead to more fusion reactions. It's the same principle as using a spark plug to ignite gasoline in an engine: once some of the fuel ignites, that will cause the rest to burn. Getting that initial spark to light is hard, but going from ignition to net-power out is comparatively easy.
Hopefully, but the NIF should really stop pretending it's about anything other than developing nuclear weapons. This is after all a research project that has publicly failed to reach its basic science goal for a solid 45 years across four generations of experiments (to compare with the 2 and a bit for tokamak experiments).
I think they have the problem that if they come out and admit that they're going to invite a lot of protests and negative media, which they don't need at the moment since they're facing a demographic shift in the lab with a bunch of people retiring now. So they need to attract and hire a bunch of people.
I’m of the opinion that fusion is already done for classified military use. Exclusive availability of small scale fusion reactors would be a huge tactical logistical advantage.
I remember listening to a presentation by a quantum computing researcher in which he made a joke about the lab’s experience of receiving funding from the NSA (paraphrased): ‘getting funding from the NSA is great. They’re the only funding source that would say “we would prefer that you fail and this technology doesn’t exist - but if it does work, we want the first one”.
I don’t see why fusion research couldn’t be analogous.
This is actually the most plausible scenario for if/when fusion is achieved by any nation state. There is evidence that China is secretly working on fusion reactors and there's no reason to assume that the Russian and US militaries wouldn't do the same.
The first nation to achieve economically viable fusion will have a huge advantage over others. All of these open experiments like NIF are the equivalent of public ML research which is years behind what happens at secret research labs in Goog etc. From a game theory point of view, there is a huge incentive to work on fusion alone and in secret, and not tell anyone about it until long after the desired result is achieved.
Fusion is a similar technological leap to attaining nuclear weapons, and we're only now learning about the extreme levels of secrecy and investment that the US and other nations afforded to those projects.
Lockheed Martin is openly working on shipping container-sized fusion reactors, it does not take a stretch of the imagination to think that the results of such a program would be kept classified quite a while after success.
It would likewise not be surprising if there were other prime contractors working on the same technology.
I don't think they have the ability to keep a secret like that. A couple reasons:
1. The act of keeping the secret makes the tech useless. If it's not incorporated into equipment that could use it (e.g. a naval vessel, an isolated firebase) then what good is it? And if it is incorporated and there's a bunch of 20 year old techs working on it, it's going to leak.
2. I don't think they're that bad of people. If they could save us from global warming with fusion tech, not doing so would be a great tragedy. And historically at least the US military just isn't that selfish. For example, they opened up GPS even when they didn't have to. I think they get that a richer/stronger public means a richer/stronger military.
No, they wouldn't be issuing press releases if this was a black program. You're also not just taking Lockhead's claims at face value, which themselves have not be established with evidence, and going further to imagine that they're lying/slow playing their claims and actually are even further ahead. None of this is motivated by evidence.
Making a net energy producing fusion reactor container sized is a feat, making your first prototype container sized is just starting out. The first tokamak had a radius of 0.7m, so could easily fit inside a standard shipping container.
It's a conspiracy theory so obviously evidence would be nigh impossible to present. Doesn't mean he or she is wrong though. Doesn't it seem somewhat realistic that some technological advances would be kept secret? There is shitloads of less valuable information that is kept secret (only to be discovered later) so it stands to reason that if an extremely valuable advancement was discovered then it might be concealed.
I don't know how a basic understanding of how military technology is developed fits into "conspiracy theory".
I'm not projecting that a sinister cabal of technologists are hiding some great truth, just that the department of defense and defense contractors are obviously developing this technology and that new military tech is usually highly classified. There's no great mystery there.
I'm not suggesting a sinister cabal, I'm suggesting there is a literal conspiracy, a group coordinating in secret. And that group is the government. And the secret would be that highly classified information. In my mind we are seeing things the same, but I can see how the words I put to it might give a different impression.
I really don't like the current "conspiracy theory" understanding in modern usage. It completely screws over the literal usage of those two words together.
If you wish to, please reread my comment without the common connotation of "conspiracy theory" and see if we then agree.
There already are small scale fission reactors. USSR launched satellites powered by them (one even infamously fell on Canada). From military use, why would you need fusion?
Those space based reactors would have a hard time powering a single house, and the highly enriched fuels in a highly mobile platform would be a disaster. Those are about kilowatts.
This containerized fusion concept is planned to have the capacity for one unit to power a small city or a naval ship, many megawatts.
So you imply the DoD did a Manhattan project in peace time for years and nobody suspected a thing?
Anythings possible, sure, but that clearly is conspiracy theory category.
Unless they found a working cold fusion method by chance, it would require a HUGE budget and the best of the best scientists avaiable. Who all need to keep their mouth shut and pretend fusion is still not working. For years. Not likely. Not in peace times.
It would not need be manhattan project size, it’s a precision engineering problem not a fundamental physics problem mixed in with an industrial scale production problem.
There have been plenty of very expensive defense programs that were kept quite secret for a very long time.
I’ve known and worked with people with top secret clearances, they just didn’t talk about their work, it’s not that hard.
And everyone of these scientists is voluntarily giving up the opportunity to move a basic working fusion reactor into the private sphere? They could make billions and become the next GE or Tesla.
If just one of them had a startup bone in his body, we would see a small fusion reactor startup making consistent progress that seems amazing. It's not happening.
A scientist or engineer who left a classified project to do the same thing in a startup would very quickly find themselves in prison. Once you gain access to classified information, it is very very easy to convict you of very many crimes.
It's a situation that is apparently very difficult to understand for many people. You work on a classified program and you understand the situation. Nobody goes off and decides to turn their classified knowledge into a business.
Unless your assange, snowden, goldamns sachs ceo turned regulator or retired military general gone into the private sphere or fbi agent turned security consultant or blackwater pmc agent.
I get the value of the myth, people instinctively acheive what they believe is doable. It's just a myth though.
And until a working fusion reactor is build, it is indeed a fundamental physics problem, with all the energy involved, which is not at all completely understood since we are in quantum mechanic level down there.
I wonder if the push to divert funding away from clean energy and toward more nukes will continue under the Biden administration.
The article says the White House has recently been pushing to reallocate the funds, but that congress keeps blocking it. Is “recently” just the Trump administration, or were they pushing for the change under Obama?
It's unique to Trump. The only nuke related programs I'm aware of that were going during the Obama admin were the new fuse research program and the very beginnings of the ground based strategic deterrent program, which is going to be the replacement of the Minuteman III ICBMs.
On the other hand, Tillerson famously referred to Trump as a moron after Trump asked why the US wasn't working to expand our nuclear arsenal.
no, it specifically didn't. If it had ignited, it would be enourmous news because the age of fusion and clean energy has begun. It didn't ignite, it's "close to self-heating", which is progress, but not ignition, and they want funds to double their laser power to actually make it happen.
Do you know what the challenge in these sorts of thing are? For example, do we already know how to double the laser power if we just had the resources or would there need to be research on how to build the laser itself?
One of the biggest challenges with fusion is to be able to contain plasma as far as I understand. Magnets seem to be the way they want to do it but I am a layman so take what I say with a grain of salt.
Magnetic confinement is needed for the continuous-reaction donuts like French scientists are building.
The NIF strategy is more like an internal combustion engine: load fuel, start reaction, capture energy, remove byproducts, start over. The lasers are acting like spark plugs, triggering the fuel to release all its energy at the top of each cycle.
In other words, the French strategy is "build a miniature sun", the American strategy is "build a miniature hydrogen bomb". In theory, it should be easier, hydrogen bombs already exist after all, but to trigger the reaction, you need to find a way to squeeze a lot of energy into a really small space to make the hygrogen atoms kiss, and you can't use the traditional method of "put a plutonium bomb next to it" for obvious reasons.
Four decades, ICF has been predicted to reach break even at one kilojoule (kJ) of energy in 1972, then 5 kJ, 10kJ, 100kJ, 200kJ, and finally 1.8 MJ by 1979 and the construction of three generation of lasers that all proved the predictions wrong.
The NIF is the fourth shot at this, and has a target performance about an order of magnitude less power that what experimental tests (using nuclear weapons) indicate is needed to be directly supplied to the fuel.
Apparently most people think stuff just happens on the first run. I write code for a living and in most cases it doesn’t run as intended on the first run, I can’t imagine making a fucking fusion reactor “just work” by assembling it and pressing “start”.
And then your results are published online (either because you need funding or someone reports on them) and people criticize you for not accomplishing your goals. Humans... ¯\_(ツ)_/¯
> The facility is funded by the U.S. Department of Energy’s National Nuclear Security Administration (NNSA) and is a key element of NNSA’s science-based Stockpile Stewardship Program to maintain the reliability and safety of the U.S. nuclear deterrent without full-scale testing.
Most of the Department of Energy budget goes towards nuclear weapons, although they also do some work on civilian nuclear power and pure science. The name was always a bit of a euphemism.
Stockpile stewardship is explicitly one of DOEs purposes, and has been since the Manhattan project really.
Weapons research is one of the top level missions of LLNL [0], where the NIF is located. I guess it’s controversial to some, but it’s not like it’s supposed to be a secret or anything.
https://web.archive.org/web/20010209102303/http://www.nrdc.o...
Has nice quotes like:
"Putting aside for a moment the vexing question of whether a "big science" machine well into the construction phase should still be plagued with numerous unresolved technology development issues, there is considerable evidence that both the science and technology underlying the NIF are not sound."
and
"While a large majority of ICFAC voted in May 1994 to support proceeding with engineering design of the NIF, [20] their decision was once again based on non-peer-reviewed LASNEX code predictions that had been hastily generated in the weeks immediately prior to the meeting. These calculations purported to demonstrate ignition with novel gas-filled hohlraum targets -- predictions that were subsequently not borne out by actual experiments conducted after the meeting."
We often bemoan the state of the research job market on HN. I can’t help but wonder what moving funding from these facilities to smaller experiments and research labs would do.
Critically, the only reason that DOE can support the cost of these "big science" projects is that they are generally "dual use" so to speak, with both basic science and national security applications. A similar dynamic is arguably at play in the funding of the largest projects in other countries as well.
So it is not clear that cutting such "big science" projects would in practice actually open up more funding for small basic science grants, lacking the same national security applications.
This would have a far more salutary effect on releasing funds for research.
Although perhaps these projects are too speculative with an uncertain path to revenue (unlike Space).
It's really funny (and great) that there are actually people now that continue this phrase with the mention "unlike space". It used to be that "the best way to become a space millionaire, is to start life off as a billionaire."
The only fusion devices humanity has yet gotten to produce net power are thermonuclear bombs. Laser fusion tries to replicate those conditions as best as possible without detonating a fission primary. Yes, this means laser fusion is good for validating nuclear bomb models, but it's also the only route to fusion where we know for a fact that every issue is strictly an engineering problem.
What's different about ITER? If I remember correctly, they're ahead in terms of results towards energy generation.
I personally believe tokamaks like ITER will be the route that leads to energy generation, but putting all our eggs in one basket is risky and inertial confinement experiments like NIF let us answer fundamentally different questions about reality.
[0]: https://en.wikipedia.org/wiki/Inertial_confinement_fusion#Is...
I remember reading the precision requirements for LIGO and then reading about the efforts needed to achieve that precision and it seemed like magic. e.g. IIRC LIGO has a precision of PORTIONS of a wavelength of light at the end of arms that are kilometers long.
Is the precision for plasma stability that much higher?
(I understand that these are very different systems so asking more from an overall "what can we currently achieve" perspective).
I mean, as I said elsewhere, the reason we don't have fusion for any fusion experiment is plasma instabilities, at this point that fact is like saying the reason ice cream melts in the sun is because it's hot. It's a fact of all fusion experiments. The only system we know of that can achieve net gain fusion we know of is gravitational confinement which uses a big mass to contain the hot plasma, also known as a star.
LIGO is all about minimizing losses as the laser light bounces back and forth the two mirrors. The losses here arise from the stack of materials that the mirrors are made out of, which are multilayers of different oxides plus/minus Si or Ge, I forget. So one metric of course is the surface roughness of the material, but then there are also energetic defects called two-level systems in which atoms can absorb a little bit of light to tunnel into another location and thus contribute to the losses by having absorbed laser energy. There's coefficients of thermal expansion, stresses, that all have to be taken into account and tested as you layer all these dissimilar materials that may behave ok at room temperature but not at cryogenic ones for instance.
So LIGO is a game of minimizing losses, because you're after detecting the faintest of signals: a gravitational wave. Their game is all about increasing the signal to noise ratio.
NIF is a monster of energy. There's the whole steering of an enormous laser pulse which is the addition of 192 beams that have to converge into a tiny area the size of a pencil eraser. There's the containment of all this energy into a steel capsule that looks like something out of a sci-fi piece. There's the manufacture of the 2 mm diameter capsule that starts as plastic but is then coated with diamond, beryllium or more plastic and that leads to inherent asymmetries because it's not easy to coat a non-planar geometry. Already this coating assymmetry is very likely to lead to the hydrodynamic instabilities (https://en.wikipedia.org/wiki/Rayleigh%E2%80%93Taylor_instab...) that are obstructing the path forward. The hole with which you fill up the capsule with DT is still a problem, but they try to sweep it under the rug. The surface roughness of the capsule is a problem, contaminants in the diamond or plastic is a problem, a few atomic percent is enough to significantly dampen the amount of X-ray absorption/transmission. Control of the material's density is another one. The work that goes on all of this is tricky because you can solve all of these and control them very well on a planar geometry, but the moment you want to take this onto a sphere it doesn't work as well, we cannot suspend something in Earth's gravity without using a string, which then introduces an asymmetry. So the best we use is we roll the sphere around, and it's not very good for the tolerances that the scientists think we need.
And NIF has a history of escalation. We are currently shooting 1.6 - 1.8 MJ, but the scientists' simulations had predicted kJ range shots in the beginning, and the estimate has continued to climb. You see it in the article itself that they're hoping to get funding to go to 3 MJ. But there are some studies (Halite-Centurion IIRC) from the 50s that showed you needed like 100 MJ to get inertial confinement fusion on a capsule, and I think we've been happily disregarding those results, because, well, politics, job-protection etc. It's a complicated story, shrouded in a lot of secrecy, so I'm glad to read that the NNSA is reviewing it.
So, precision in LIGO, I don't think translates to precision in NIF. LIGO feels more like golf, NIF feels, like some beasts fighting it out. Very different set of challenges, very different resource pools they can draw from (I would argue that NIF can draw more money but less talent because of security clearance requirements restricting employees to be US citizens).
Laser fusion is about the least promising, most impractical possible approach to building a power reactor, even if you do manage to get ignition. Since at least the early 1980s, the word on the street has been that the talk about power is just a politically palatable wrapper for the weapons application. And let's just say that some of the people I heard that from, back in the 1980s, were in enough of a position to know that they probably shouldn't have said anything to some random kid like me.
Not to say that there aren't a few benighted souls working on it who've managed to convince themselves it's a power source, but notice that nobody but Livermore has ever bothered to put much into inertial confinement... and Livermore is a weapons lab?
Some mineral physicists I know have used this to write some cool papers about the properties of natural materials at conditions replicating those expected in the cores of giant exoplanets [1-3], for example.
[1] https://www.nature.com/articles/s41550-018-0437-9
[2] https://doi.org/10.3389/feart.2019.00023
[3] https://advances.sciencemag.org/content/4/4/eaao5864
[0] https://lasers.llnl.gov/news/next-10-discovery-science-exper...
All fusion experiments are not reaching ignition due to plasma instabilities. Not exaggerating, all of them.
The funding from NNSA is actually in this very article.
LFTR scaled down to a closet, so with "mass production" you could just replace the vessel at "reasonable cost", but these fusion facilities are massive and expensive.
http://progressive.org/op-eds/let-cut-our-losses-on-fusion-e...
The L4n laser beamline of the P3-installation: Towards high-repetition rate high-energy density physics at ELI-Beamlines
https://aip.scitation.org/doi/10.1063/5.0022120
The 1.8 MJ laser array that's being used to trigger these reactions, that's like 400 calories, less energy than you'd get from a ham and cheese sandwich with mayo, and the output they're hoping to achieve, 100kJ, that's 2mL of gasoline.
Also while these energies are low, they are also for very short durations. 1.8 MJ in 1 ns is 1.8 PW.
How much power it produces is the better question.
So it's actually massively powerful but very short.
If you were to burn gasoline at a rate of 2mL per 20ns you would be burning 100,000 L per second, which is a higher rate than Saturn 5 on takeoff so quite a burner...
For example, per kilogram, TNT releases less than a tenth of the energy that burning gasoline does, and a sandwich stores more energy than the equivalent weight in dynamite (but both are cheating, as TNT and dynamite both include the oxygen needed to ‘burn’ them, whereas gasoline and sandwiches don’t)
Ham sandwich prompt radiation, thermal pulse, blast, and long-term radiation effects are somewhat less than those of the Little Boy gadget. The last may depend on how much hot sauce is applied.
Both weight and volume exceeded the B-29 and would be better suited for a bulk-cargo carrier of about 30' beam, 40' depth, and 300' length.
The relationship between promptness and completeness of reaction is an interesting one, and not limited to TNT and petroleum. The WWII atomic weapons haad a pretty remarckably low efficiency --- most of the fissile material was dispersed before it could fully react. Nuclear power generation is far more efficient. And generally slower.
See related the Beruit 2020 ammonium nitrate explosion:
https://news.ycombinator.com/item?id=25154155
https://news.ycombinator.com/item?id=24064248
Even if they achieve 'burning plasma', it's still like a 100hp engine using an 1800hp spark plug.
And it’s no accident that the article doesn’t mention the laser efficiency. NIF has a public relations office that has honed its game over decades. Their purpose is to influence Congress by selective reporting through press releases and by feeding propaganda into articles like this.
http://www.DDproFusion.com
The design has a lot of similarities to NIF in some respects. I aim to create an ultra dense implosion of deuterium ions.
Thats where the similarity ends though. The key to viability in my eyes is to recover the energy of the ions that don't fuse to create another implosion.
In my device, ions leaving the focus are curved by a weak, uniform, magnetic field. Each ion travels along it's cyclotron trajectory to circle back to the focus at the right time for the next implosion.
https://www.kickstarter.com/projects/1992078142/building-the...
My understanding of the Polywell is that it uses a trapped cloud of electrons to form the central cathode, which can help keep deuterium ions hot by reducing grid collisions.
The downfall I see is that you are still trying to "squeeze jelly with rubber bands".
The Brillouin limit describes the maximum steady state density possible in a non-neutral plasma.
In my device, the average density is well below this limit, but is briefly very very high. The goal is to have smooth and steady cyclotron trajectories most of the time, but all these trajectories intersect at the same point simultaneously.
I'm trying to target 15kev ion energies with my prototype so I was going to aim for 60kv on the end electrodes, and 0.35T for the magnet.
Edit (further description of the 4D shell under the Liouville theorem):
When the plasma has large spatial volume, the momentum of particles at any point creates a 1D line forming an empty 4D shell. When all particles enter the 1D spatial focus the particle velocities there are essentially random and have large variation in all 3 spatial dimensions. Essentially a 4D sphere with only 1 dimension in space.
The result is a morphing 4D shell that collapses all particles into a focus with 0 volume once per period.
The focus is like what you'd get if you were looking at slices of a solid ball. You'd see a point grow into a large circle, then shrink again to a point. It's basically the view on the left side of the simulation on my website.
The right side of the simulation video shows a different cross section. The x axis there is the one dimension of the 4D sphere that stays in 3 space.
You are right that the ion current will be small. I'm anticipating that the plasma will be a non-neutral deuterium plasma.
The energy distribution will be thermal targeting 15kev for the prototype.
I wasn't familiar with the Liouville theorem. I'm not quite sure how to reconcile this. Seems a bit paradoxical. I'm going to have to investigate further.
The organization I am most excited about is https://www.tokamakenergy.co.uk/
I don’t see why fusion research couldn’t be analogous.
The first nation to achieve economically viable fusion will have a huge advantage over others. All of these open experiments like NIF are the equivalent of public ML research which is years behind what happens at secret research labs in Goog etc. From a game theory point of view, there is a huge incentive to work on fusion alone and in secret, and not tell anyone about it until long after the desired result is achieved.
Fusion is a similar technological leap to attaining nuclear weapons, and we're only now learning about the extreme levels of secrecy and investment that the US and other nations afforded to those projects.
It would likewise not be surprising if there were other prime contractors working on the same technology.
https://lockheedmartin.com/en-us/products/compact-fusion.htm...
I'm not projecting that a sinister cabal of technologists are hiding some great truth, just that the department of defense and defense contractors are obviously developing this technology and that new military tech is usually highly classified. There's no great mystery there.
I really don't like the current "conspiracy theory" understanding in modern usage. It completely screws over the literal usage of those two words together.
If you wish to, please reread my comment without the common connotation of "conspiracy theory" and see if we then agree.
The US military accepted that climate change is a huge threat to stability across the world. They are motivated to prevent the worst of it.
This containerized fusion concept is planned to have the capacity for one unit to power a small city or a naval ship, many megawatts.
Anythings possible, sure, but that clearly is conspiracy theory category.
Unless they found a working cold fusion method by chance, it would require a HUGE budget and the best of the best scientists avaiable. Who all need to keep their mouth shut and pretend fusion is still not working. For years. Not likely. Not in peace times.
There have been plenty of very expensive defense programs that were kept quite secret for a very long time.
I’ve known and worked with people with top secret clearances, they just didn’t talk about their work, it’s not that hard.
If just one of them had a startup bone in his body, we would see a small fusion reactor startup making consistent progress that seems amazing. It's not happening.
It's a situation that is apparently very difficult to understand for many people. You work on a classified program and you understand the situation. Nobody goes off and decides to turn their classified knowledge into a business.
I get the value of the myth, people instinctively acheive what they believe is doable. It's just a myth though.
And until a working fusion reactor is build, it is indeed a fundamental physics problem, with all the energy involved, which is not at all completely understood since we are in quantum mechanic level down there.
“NIF researchers believe they are close to an important intermediate milestone known as “burning plasma”.
The article says the White House has recently been pushing to reallocate the funds, but that congress keeps blocking it. Is “recently” just the Trump administration, or were they pushing for the change under Obama?
On the other hand, Tillerson famously referred to Trump as a moron after Trump asked why the US wasn't working to expand our nuclear arsenal.
The NIF strategy is more like an internal combustion engine: load fuel, start reaction, capture energy, remove byproducts, start over. The lasers are acting like spark plugs, triggering the fuel to release all its energy at the top of each cycle.
In other words, the French strategy is "build a miniature sun", the American strategy is "build a miniature hydrogen bomb". In theory, it should be easier, hydrogen bombs already exist after all, but to trigger the reaction, you need to find a way to squeeze a lot of energy into a really small space to make the hygrogen atoms kiss, and you can't use the traditional method of "put a plutonium bomb next to it" for obvious reasons.
A decade and 3,000 failures and they are probably spinning some PR so they don’t get defunded
The NIF is the fourth shot at this, and has a target performance about an order of magnitude less power that what experimental tests (using nuclear weapons) indicate is needed to be directly supplied to the fuel.
> The facility is funded by the U.S. Department of Energy’s National Nuclear Security Administration (NNSA) and is a key element of NNSA’s science-based Stockpile Stewardship Program to maintain the reliability and safety of the U.S. nuclear deterrent without full-scale testing.
Most of the Department of Energy budget goes towards nuclear weapons, although they also do some work on civilian nuclear power and pure science. The name was always a bit of a euphemism.
Weapons research is one of the top level missions of LLNL [0], where the NIF is located. I guess it’s controversial to some, but it’s not like it’s supposed to be a secret or anything.
[0]: https://www.llnl.gov/missions