Momentary fusion breakthroughs face hard reality | Hacker News Reader

Momentary fusion breakthroughs face hard reality

spectrum.ieee.org

104 pointsmfiguiere68 comments

> Plenty of other obstacles remain than those noted above, too. Current calculations compare energy generated against the NIF laser’s output, but that brushes over the fact that the lasers draw more than 100 times the power from the grid than any fusion reaction yields. That means either energy gains or laser efficiency would need to improve by two orders of magnitude to break even in any practical sense. The NIF’s fuel pellets are also extremely expensive, says Kritcher, each one pricing in at an estimated $100,000. Then, producing a reasonable amount of power would mean dramatically increasing the frequency of NIF’s shots — a feat barely on the horizon for a reactor that requires months to load up the next nanosecond-long burst.

"months to load up" is even a euphemism - IIRC they have to repair parts of the lasers which are so overpowered that they damage their guidance optics with each blast. So the cost for one shot is even higher than the $100,000 for the fuel pellet and the cost of the energy consumed by the lasers. I wonder why everyone skirts around the obvious here: the NIF's task is weapons research, and for that, fusion reactions which last nanoseconds might be enough, but there is no way this method can be scaled up so it's actually a cost-effective way of producing electricity...

>I wonder why everyone skirts around the obvious here: the NIF's task is weapons research, and for that, fusion reactions which last nanoseconds might be enough, but there is no way this method can be scaled up so it's actually a cost-effective way of producing electricity...

We already have fusion weapons that work very well, there is some valuable data that can aid weapons design I don't believe that to be a primary goal here. Looks like they did shift to plutonium targets for a while about a decade ago focusing more on weapons research but then shifted back to trying to work toward controlled ignition for power generation.

Still a long way to go but the first step is getting more energy out then in on the main reaction without using fission bombs and destroying the device. Then figure how to scale up to get more energy out than in when considering the apparatuses efficiency, then getting more out than in when considering heat engine efficiency.

I think you may be slightly misinterpreting the parent comment. When they say the NIF's task is weapons research, I think they meant that the facility was built with that task in mind. When I read your comment it sounds more like you're talking about the current task of the people working there.

Both valid points of discussion, but different ones.

NIF uses inefficient lasers because they were cheap to build at the time and because NIF is a science experiment. Lasers have gotten better. And lasers aren't the only way ICF fusion could be carried out, it may be possible to use ion beams instead[0].

It does not matter if fusion reactions last microseconds if they generate more energy. Using optimistic, but not unrealistic assumptions, it appears feasible for electricity costs to reach $25 per MWh[1] with ICF. With the most important factors driving cost being achieving high gain and yield per shot

[0]https://ieeexplore.ieee.org/document/650904/

[1]https://royalsocietypublishing.org/doi/10.1098/rsta.2020.005...

Those numbers don't seem feasible. Near as I can tell, a steam generator with a magical source of never-ending heat would cost $15/MWh. There is no way a high-tech facility and the personnel required to run it are costing $10/MWh.

What’s your math on $15/MWh? As far as I can tell, the steam turbine is a pretty insignificant part of the price of coal energy generation

$15/MWh is 1.5 ¢/KWh, right? that's pretty darn cheap and means that the other 10~20 cents would be transmission and fuel, yes?

20 cents? I pay 13/kWh.

You're within the stated range, do you think you have the most expensive electricity in the world? 25c/kwh isn't unreasonable. Ireland pays 52c/kwh.

https://www.statista.com/statistics/263492/electricity-price...

Are you allowing for cooling towers, water treatment, pipework, and so on? Also the grid connection hardware, spare turbines, cooling pumps, etc., and O&M costs?

There are three routes to bypass this issue:

1. Move to supercritical Co2 turbines (lower mass —> lower cost), need RnD on corrosion resistant alloys

2. Move to thermoelectric or photovoltaic generators

3. Use aneutronic fusion

Directly use the charged particles

1. would have been done already if there were such alloys.

2. photovoltaic generators seem popular, but the don't require spending money on a fusion core.

3. The problem is hard, so let's replace it with a harder one?

NIF exists to research nuclear weapons without breaking the treaties. Their fusion energy research is a nice corollary but it is absolutely not their focus.

I think a lot of people who are really into fusion research know that the military isn't going to be doing the same kind of fusion research as the groups looking into fusion as an energy source.

Is what they did interesting sure, but alot of people who actually follow fusion development know that this was a PR meme that some news stretched.

Fusion has always been 20 years away its a literal meme....and yet we are making good advancements in the field but I think anyone with a basic understanding or those who realized this was so short lived results time wise realised it was an overhyped thing by media from this group.

Fusion is important to the government because they can theoretically squeeze more power out of nuclear weapons, but due to the difference in timescale zit just dosnt translate across domains of military to public energy that well

The US military is very much interested in fusion, as an energy source.

The Desert Storm conflict cost something like $1T (yes a Trillion dollars) just for gasoline.

There was on this site a description of a military effort to create a low-yield fusion generator that would fit on a military truck. It only had to fuel a mobile camp, not a city. If I recall, it was something like the Lockheed plan, to create a vortex (not a torus) to stuff the fuel through a constriction and get some kind of fusion yield. Wasteful, but not as wasteful as $1T

That seems wildly high? The figure widely quoted for the total cost of the Gulf War is $60B in 1991 dollars, or $137B today, so clearly the gasoline costs alone can't exceed that number. It's only the extremely prolonged conflicts like the War in Afghanistan that get into the trillions of dollars total.

$1T would be too high for Iraq War where the costs run $1T to $3T. The higher numbers are total long-term costs, not just the war fighting. There are lots of costs to war instead of just fuel; the people cost more.

Yeah way too high.

Here's an estimate of the costs of air conditioning tents in the mideast conflicts: https://www.npr.org/2011/06/25/137414737/among-the-costs-of-...

Setting aside the wildly suspect 1T claim, I’m not convinced by the argument. Fusion power plants are fixed large-scale facilities. In a war zone you need mobile power sources. I could see a fusion reactor powering a ship and maybe powering temporarily shore-based facilities. AFAIK the military is looking at solar for mobile power in forward deployments though which makes sense. Unless the argument is that fusion is used to generate transportable fuel (hydrogen, efuels etc). But then that argument is weird because the military would be doing that with fission today which actually exists and it has to expensive field experience with and supply chains supporting.

> Fusion power plants are fixed large-scale facilities.

Large scale facilities really are not all that interesting. They concentrate cost, danger and impose lots of logistical limitations.

Consider the Lockheed Skunk Works project - it's supposed to be portable.

I've read that the US military has looked into small fission generators. They expect logistics to be very difficult near the front if there is a war with a 'near-peer' (China or maybe Russia) and hope to reduce fuel logistics.

I've never heard of them looking into similar fusion reactors. Why would they? Nobody expects to build any kind of production fusion reactor for decades, much less one that could be operated by infantry soldiers and in very difficult conditions - e.g., explosions and people shooting at the reactor.

American military would love to squeeze nuclear reactor on every single ship. The conventional escort ships are much slower than aircraft carriers and need constant refueling, to the point they sometimes leave them behind and sail meet another fleet in destination port. It's not really feasible with fission but fusion would make it a lot safer.

So they should be researching LFTR like they did for the nuclear bomber.

> a low-yield fusion generator that would fit on a military truck.

another word for that is "drone target."

Everything in war is a target

True, but a small target that disables all the electronics on the base would have to be considered "high value."

Like, say, the generator they use now.

No, not like that.

Generators are cheap and replaceable, and they can keep spares in the depot.

A fusion reactor sounds like big bucks to me.

> a lot of people who are really into fusion research

Do you mean researchers and other experts, or people who follow it? And does that include you? Are you a researcher?

I think it would be more of people who follow the research in general but also have a good understand of the principals of modern nuclear weapons using both fusion and fission to increase yield %

I'm just a guy that like to follow the research in hopes of clean cheap energy...the cheaper the better

This basically the plot of the movie ”Real Genius”

Great film! https://www.youtube.com/watch?v=xuv7SIVNkx8

Oh man. I saw that movie as a kid, and remember thinking the hyperactive girl was really cute. I recently happened to see her scenes again, and unfortunately, the memories were better than the real movie.

Efficiency of equivalent lasers has improved by about 40X since NIF was built, and they seem to have nonlinear scaling on fusion output. NIF definitely isn't a practical energy system, but it is an experiment that helps move us in that direction.

Here's a NYTimes article[1] about various companies attempting laser fusion for power production. One of them has an approach very similar to NIF's, and a couple others are similar but do away with the need for a hohlraum by using a different type of laser. They also say:

"A decade ago, a report[2] by the National Academy of Sciences found much to like in the energy potential of laser fusion but recommended that the United States hold off major investments until ignition was achieved. That time is now."

[1] https://www.nytimes.com/2023/11/13/science/laser-fusion-ener...

[2] https://nap.nationalacademies.org/catalog/18289/an-assessmen...

Check out General Fusion who are actually building a scalable fusion reactor.

> I wonder why everyone skirts around the obvious here: the NIF's task is weapons research, and for that, fusion reactions which last nanoseconds might be enough, but there is no way this method can be scaled up so it's actually a cost-effective way of producing electricity...

I think you answer your own question.

NIF does a lot more than weapons. Ignition is also a different problem than sustaining a fusion reaction, but both are important aspects of the fusion energy generation problem. You can think of this like the difference between static friction and kinetic friction. You would never try to generate power by performing tons and tons of ignitions. You generate power by doing the ignition and then sustaining the reaction. There's even people working on setups just like this. You can see how ITER works here[0]. You can modify this procedure to have laser based ignition (or some other form) and then feed this into the toroidal reactor. There's a bunch of ideas out there but I think this is the most straightforward.

As to a more systematic approach to answering "why" and why this type of language is pervasive regardless of the domain, is because a lot of people have a difficult time differentiating TRLs[1]. Or maybe even more bluntly, research from products. I think it won't be hard to find high rated HN comments on research works that come out of universities who's critique is that the process is expensive and doesn't beat current industrial processes, but this is a grave misunderstanding of how technological progress happens. It's overly dismissive. Technology often progresses by iterative discontinuous S-curves, which at a zoomed out level looks more smooth. So new technology often starts off way worse but the key part of this is the theoretical maximum. A good example of this might be in batteries or solar panels. People are highly dismissive of any technology that is yet to become a product (TRL 6-7-ish). Which it is good to not jump to assertions that a low TRL result means new products, but it's the same type of error, just in the other direction. Either way, critical context is being ignored.

I think if we more appropriately contextualized technological development then we wouldn't have these issues. I don't like them either tbh. I'd rather be honest. But truth has bounded simplicity and lies don't. Since a lot of people ignore nuance it creates an incentive system to exaggerate or leave out important details. Frustrating, but understandable. Hopefully understanding can help us to disrupt this systematic incentive system and instead promote more nuanced and honest discussions. But I think we need to recognize expertise does not transfer between domains well and discourage claims and comments that lack nuance. But nuanced and accurate comments require more words, so we might just be fucked and this is just a pipedream. Some people think we can't rely on humans being nuanced and intelligent creatures but I'm not convinced and more convinced these are just products of environmental pressures. I guess we'll see.

[0] https://www.youtube.com/watch?v=5tH4obUsY64

[1] https://www.nasa.gov/directorates/somd/space-communications-...

There's a really good series on what's needed to build a working fusion reactor on the YouTube channel "Improbable Matter".

How nuclear fusion works (1) - fusors, thermonuclear reactions, lattice fusion: https://www.youtube.com/watch?v=2DzKXN1pcwY

How nuclear fusion works (2) - confinement, stars, nukes, inertial fusion energy: https://www.youtube.com/watch?v=mxmxZI2Ltvs

How nuclear fusion works (3) - magnetic confinement, tokamaks, stellarators: https://www.youtube.com/watch?v=gwOrbr8KWDs

How nuclear fusion (maybe) works (4) - reactor practicalities: https://www.youtube.com/watch?v=ZHmHBMaS6Sw

Strongly recommended!

Who makes "Improbably Matter"?

Imagine a second Manhattan Project, except this time, it's to produce fusion energy and try to save the world from climate disasters instead of making a weapon.

Who would you tap to participate in/lead this?

Elon Musk and his ability to gather financing and top engineers

Better get him to rehab first.

Oppenheimer again

Why do you think finishing the fusion research (just another 10 years? again?), and then designing commercial reactors, and then building out all the fusion generation capacity we need, could possibly take less time/money/effort than just building out all the wind/solar/battery capacity instead, which we already know how to make?

I think it's a good strategy to have alternate energy sources.

We use energy in so many ways, we should also generate it in many ways too.

We should develop a portfolio of energy sources. Yes, the top ways might be solar, wind, batteries.

But what if we need energy at night during cold weather without wind? We will continue to burn giant oil tankers full of fossil fuels.

Looking back, the outlook for Solar PV in president carter era seemed just about as unsatisfying as fusion is now.

I also think fission should get some predictable sustained research funding, and political will to make it a part of the picture, but safe and modern.

Things to ponder if you're having these conversations.

> a portfolio of energy sources. Yes, the top ways might be solar, wind, batteries.

Batteries are energy storage, not a source. Other storage methods are worth finding out about.

> what if we need energy at night during cold weather without wind? We will continue to burn giant oil tankers full of fossil fuels.

We already burn vast amounts of fossil fuels (how much per day globally | by various countries? it's worth a look).

The key thing here is how much less would be used if a country is only "topping up" energy storage and delivery during renewable down times?

What are the differences between a coal fired power station that drives a turbine and a gas powered power station that drives a turbine (ramping up | always "on" times, etc).

> I also think fission should get some predictable sustained research funding

In which country though? South Korea and China are steadily building fission power reactors, the USofA has had ongoing fission research as part of being a nuclear nation since the Manhatten project, it's DoE straddles nuclear for weapons and nuclear for power, they had nuclear submarines, etc.

We already have the ITER project, a well-funded international (~35 nations) attempt at building a fusion reactor prototype.

Also we probably shouldn't count on a miracle to save us from climate change.

The work from Rochester mentioned briefly is probably the second best performing fusion experiment and this was done with a laser that's at least 29 years old.

What the article doesn't mention is that the approach demonstrated could decrease the cost of fuel pellets. In NIF, the laser energy is converted to X-rays which compress the fuel pellet using expensive elements like gold. What they demonstrated is that the fuel pellet can be imploded directly with the lasers.

..using the laser approach. Other approaches don't have that issue:

https://www.iaea.org/newscenter/news/tokamaks-stellarators-l...

I guess that's why they're using "Momentary Fusion" in the title? Of course, it wouldn't have hurt to mention in the article that other approaches which might be more feasible also exist...

Well those of us who have never heard of "momentary fusion" read the title to mean "momentary breakthroughs in fusion"

I never heard the term "Momentary Fusion". There are other pulsed fusion methods, like Dense Plasma Focus[1] with it's very hot but very short lived plasmoids. Also still far from being economically viable, but I would say has a much better chance than that weapon research on NIF[2].

[1] https://en.wikipedia.org/wiki/Dense_plasma_focus

[2] https://spectrum.ieee.org/startup-lppfusion-embraces-instabi...

My bets are on Wendelstein 7-X / Stellarators

OT but are there any links to resources about using fission/fusion bombs directly for energy production?

The concept was explored in the United States as Project PACER:

https://en.wikipedia.org/wiki/Project_PACER

excellent, thanks

Why go through the expense of engineering and manufacturing a bomb, and then use it for a different purpose? If you want an electricity generator, engineer and manufacture for that. Or are you thinking of using excess bombs?

I want to say that fusion bombs require more energy input than they output, because until recently nobody has been able to create fusion reactions otherwise (and now only in a very limited way). But is that true? Fusion bombs release an incredible amount of energy. Maybe it's that they release it all at once, but still, how does that equation work out?

> Why go through the expense of...

Good question. I should have asked strictly about fusion bombs. These being a known-working example of energy production by fusion, why not use them when we can't get 'controlled' fusion working?

From my recollection, even in a fusion bomb, most of the energy comes from the fission. I don't fully remember/understand the physics involved. But apparently the fission reaction triggers the fusion reaction (that part I get) and then that fusion reaction in turns boosts the fission reaction.

So my takeaway is that fission bombs are not very efficient to begin with. You're after all trying to extract as much energy as possible as quickly as possible from a system that is literally self destructing. We do not have that constraint with current fission plants, the plant is designed to run in a steady state for years, the fissile materials stay in place and my intuition is that this significantly helps efficiency.

Yes, the gist is that in a fusion device, you've got a chain reaction where the splitting uranium or plutonium is making neutrons, and those neutrons strike other uranium or plutonium, which makes more neutrons, etc. But you have a very short period of time to take advantage of this chain reaction, because before long, your device has blown apart, so a large percentage of the fuel in a conventional fission device never undergoes fission (the device has blown up before it has a chance). The job of the fusion reaction in a fission-fusion-fission device (how "modern" thermonuclear weapons work) is, in addition to producing energy of its own, to rapidly produce lots of lots of neutrons, which can very quickly fission a second quantity of fissile material.

One of the biggest challenges is that the fusion reaction can't be fully contained. There are particles without electric charge that simply go through your magnetic containment field and smash into the reactor wall, continuously destroying it. You don't really care about preserving the shell of your bomb. You care about preserving the shell of your reactor.

How do fusion researchers handle that problem?

Replacement parts! That may sound flippant but yeah, replace the components/structural members subject to neutron flux before they get too compromised, basically.

This is yet another reason that fusion research is expensive and time consuming, the cost involved in manufacturing these components and time lost in painstaking teardowns and overhauls between test runs.

Let me elaborate on thmsths's answer. The United States developed and tested large "clean" fusion bombs that produced only a very small fraction of their energy from fission. Those bombs could produce raw thermal energy at a cost below any fossil fuel or conventional nuclear reactor. For example, the Housatonic test shot of Operation Dominic in 1962 yielded about 10 megatons and 99.9% of the energy released came from fusion [1].

The problem is that explosions of that magnitude are not usable for controlled energy generation. Confining them underground is impractical, due to the scale of energy release. The largest underground US test was 5 megatons in 1971, and it was so disruptive (induced surface motion equivalent to an earthquake) that it had to be conducted at a remote Alaskan island instead of the usual Nevada test site [2].

For those reasons, Project PACER considered bombs of only up to 50 kilotons. At those sizes (0.5% the energy yield of Housatonic), there is no known design to get 90%+ of the bomb's energy yield from fusion. The efficient fusion designs don't gracefully scale down to that range. And at that much smaller scale where underground confinement was practical, bombs were not actually any cheaper as a source of raw thermal energy. Quoting Wikipedia about the project here, "In a 1975 review of the various Plowshares efforts, the Gulf University Research Consortium (GURC) considered the economics of the PACER concept. They demonstrated that the cost of the nuclear explosives would be the equivalent of fuelling a conventional light-water reactor with uranium fuel at a price of $328 per pound. Prices for yellowcake at that point were $27 a pound."

[1] "Ripple: An Investigation of the World’s Most Advanced High-Yield Thermonuclear Weapon Design" http://web.mit.edu/zoz/Public/jcws_a_01011.pdf

[2] https://en.wikipedia.org/wiki/Cannikin

Great answer, and from others too, thank you all.

A new fusion company name Blue Laser Fusion is working on inertial fusion with new laser technology. Does anyone have information on what they are doing, and what the prognosis is for that company's approach?

I don't think fusion will ever be cheaper than solar and wind

No, but it will be more reliable, and potentially portable. Both very important characteristics.

I do agree that distributed solar and wind (with battery storage) can take take care of grid base load, but that doesn't mean fusion is pointless.

Reply on news.ycombinator.com