Fusion breakthrough?

[Roller]

Trump's reaction to the fusion announcement:

The Biden administration just announced this new thing. Fusion they call it. It’s not good for America. It’s when you mix Mexican and Chinese food. Fusion energy is better, but we did it first. The Democrats stole it, just like the election. We were going to shine it inside the body to kill the virus, but Birx and Fauci shut it down like dogs.

 

[Yoused]

So we’re getting waaay outside my knowledge base here but is the grid energy here just to get things started such that as long as the reaction is positive that grid energy isn’t needed to sustain the reaction and thus would be repaid … eventually (in this case very eventually and only in theory)?

Not exactly. The reaction involved a fuel pellet about the size of a grain of sand. The lasers were burning for less than a millisecond and the overall reaction lasted about another millisecond. It was essentially a 0.000000001 megaton neutron bomb.

The theory of Inertial Confinement is to create a series of net-positive minute nuclear explosions over and over and over again and harvest the energy gained from them. It has the particular advantage that you never have to worry about anything like a core meltdown situation because all you have to do is flip the off switch and it all stops right away. This is also true of magnetic confinement, as the reaction simply stops when the plasma is no longer being squeezed, which is one of the apealling aspects of fusion.

The reaction occured in a "holraum", which is a metal cylinder that provides x-ray feedback to sustain, if briefly, the reaction energy. These are also used, on a larger scale, in thermonuclear weapons, to contain the plutonium fission explosion that lights the fusion reaction. How exactly they would expect to use holraums to support power generation is a little baffling to me, as it seems as though it would not be all that cost-effective.

In the end, the practical feasibility of fusion power is questionable. The Sun, for example, produces volumetric net heat output mathematically equivalent to an active compost heap, but being so very large, it works out to a non-small amount. It is genuinely not obvious that fusion power on a comparably minute scale is realistically doable, or, honestly, even worthwhile. A viable fusion reactor requires arcane materials and construction that must be carefully designed to maintain the reaction, and these components are subject to intense radiation bombardment that must wear them out rather quickly, leading to decreasing efficiency and low-level nuclear waste.

So, it sounds cool and all, but I remain less than optimistic. The research should continue, because we might gain something of value from it, but right now there is not much to be truly hopeful about.

 

[theorist9]

I recommend nature.com for lay articles on scientific advances. They generally do a much better job than the popular press. Their article on this can be found here: https://www.nature.com/articles/d41586-022-04440-7

They explain that the lasers deposited 2.05 MJ into the target, and the reaction released 54% more energy (3.15 MJ), giving Q = 1.54. This is a significant breakthrough, since NIF's previous record was 1.35 MJ from 1.9 MJ of laser energy, for a Q of 0.7. [https://en.wikipedia.org/wiki/Fusion_energy_gain_factor]

On the other hand, the energy it took to charge the lasers so they could deliver this pulse was 322 MJ. I.e., these lasers are only about 2/322 = 0.6% efficient in converting electrical energy to EM output. Thus the Q of 1.54 is a highly qualified value. If we calculate a different Q, for output/total input, we have 3.15/322 = 0.01.

If you're thinking that puts this pretty far from being a prototype for fusion power, you're right. Indeed, the NIF wasn't designed with fusion power in mind. It was designed for nuclear weapons research—to study thermonuclear explosions. Reactors designed as fusion power prototypes, like ITER, have different designs:

“NIF was not designed to be efficient....It was designed to be the biggest laser we could possibly build to give us the data we need for the [nuclear] stockpile research programme.” [Mark Herrmann, deputy director for fundamental weapons physics at Lawrence Livermore National Laboratory.]

However, that's not to say the understanding they get from this won't be useful for fusion power research:
“The NIF experiments focused on fusion energy absolutely are valuable on the path to commercial fusion power” [Anne White, a plasma physicist at MIT.]

 

[theorist9]

In the end, the practical feasibility of fusion power is questionable. The Sun, for example, produces volumetric net heat output mathematically equivalent to an active compost heap, but being so very large, it works out to a non-small amount. It is genuinely not obvious that fusion power on a comparably minute scale is realistically doable, or, honestly, even worthwhile. A viable fusion reactor requires arcane materials and construction that must be carefully designed to maintain the reaction, and these components are subject to intense radiation bombardment that must wear them out rather quickly, leading to decreasing efficiency and low-level nuclear waste.

The nuclear reaction in the sun indeed would not be practical for generating fusion energy here on earth, which is why none of the fusion reactors employ it.

It starts by converting two protons to a deuterium nucleus, which means two protons have to be converted to a proton plus a neutron. The likelihood of this happening during any given proton-proton collision is extraordinarily small. Consequently (and fortunately for both the sun and us), even at the extraordinary densities at the center of the sun (~150 g/cm^3), fusion proceeds extremely slowly, giving us a sun whose fuel will last for billions of years. This is also why the energy generated per unit volume is, as you mentioned, so low.

None of this is relevant to fusion power, because no fusion prototype would attempt to use such an extraordinarily slow process, i.e., none use pure hydrogen-1 (protons only) as fuel. Instead, they use fuel that contains neutrons to start with. Here's a pic from Philip Ball's Nov. 2021 article in Nature (https://www.nature.com/immersive/d41586-021-03401-w/index.html) depicting typical fuel mixtures; the neutrons are shown in purple:

 

[AG_PhamD]

I believe another inertial confinement reactor had a >1 energy emitted per energy absorbed by the plasma (the Q factor) about a decade ago. Instead, it looks like what they achieved is that the energy *outputted* by the laser is less than the energy emitted by the plasma. The important distinction here: the lasers probably required more energy to run that the energy that was later outputted by the laser. So the reaction in the plasma had a Q >1, but the reactor still probably required more energy to run the experiment than it produced.

It's an important milestone, but all kinds of nuclear fusion reactor have been getting closer and closer to the Q factor of 1 for a while, there's nothing inherently special about the breakeven point. It's not really a leap in the progress towards commercial fusion, more like a milestone of years of steady progress.

I believe I read somewhere the laser used was 500MJ and not factored into the equation.

Fusion isn’t something I follow closely or have thorough understanding of, but it seems like for every achievement, it raises a whole new set of problems.

I’m under the impression they’re making a much bigger deal of this than is actually the case. Even if true net positive fusion was discovered in a lab setup, scaling it up into a practical and affordable power generation source is still a very long ways off, if even practically feasible.

As I understand there are many physicists who have little faith that fusion will ever be harnessed as an energy source. I hope they are wrong. But in the meantime, we should also be investing in nuclear fission. If we want to reduce CO2, the answer is staring us in the face the entire time.

 

[Andropov]

So we’re getting waaay outside my knowledge base here but is the grid energy here just to get things started such that as long as the reaction is positive that grid energy isn’t needed to sustain the reaction and thus would be repaid … eventually (in this case very eventually and only in theory)?

I don't know for certain how much energy is lost at the start vs maintaining the laser output. I'm not familiar with super high power pulsed lasers, but I don't think a "garden variety" laser (let's say something like ~600W) does not have a warm up period or anything like that (pulsed lasers need to charge for a very brief moment before each pulse, but that's all).

In any case, high-power lasers have significant efficiency losses (heavy refrigeration is needed), pulsed lasers probably even more so. You lose energy in the element used to pump energy into the laser (a diode, for example), thermal losses on the laser gain medium, photons being emitted off-axis...

On a laser you're basically pumping energy into the laser medium to excite the atoms into a specific excitation state. What you want is for most atoms in the medium to be in that excited state (something called population inversion) for the laser to amplify the source of energy. This excited atoms will emit a coherent photon when excited by another photon at a certain frequency, so once you've gotten most atoms into the excited state, you can start a sort of cascade of photons (a photon causes the next excited atom to emit an second photon, so you keep adding another photon for each excited atom hit).
But you get diminishing returns when pumping a lot of energy into the laser medium. Some atoms spontaneously start decaying without being hit by a photon (this energy is lost because is emitted off-axis), it becomes harder to excite the remaining atoms in the ground state when most atoms are in an excited state, so the energy is thermally absorbed by the medium... and all this losses are required just to maintain the laser (in a continuous laser) or to emit a new pulse (on a pulsed laser), those are not initial energy costs that can be offset by having the laser run for a long time.

Fusion isn’t something I follow closely or have thorough understanding of, but it seems like for every achievement, it raises a whole new set of problems.

I’m under the impression they’re making a much bigger deal of this than is actually the case. Even if true net positive fusion was discovered in a lab setup, scaling it up into a practical and affordable power generation source is still a very long ways off, if even practically feasible.

As I understand there are many physicists who have little faith that fusion will ever be harnessed as an energy source. I hope they are wrong. But in the meantime, we should also be investing in nuclear fission. If we want to reduce CO2, the answer is staring us in the face the entire time.

I feel like other types of fusion reactors (like tokamaks) are way, way closer to any commercial applications than an inertial confinement (laser-based or otherwise) reactor. Reactions can already be sustained for much longer. JET (a scaled-down version of ITER) had a 5-second reaction this year, for example, with a respectable Q of 0.33 (it was not designed to reach the breakeven point). The ITER (which is an order of magnitude bigger) is expected to reach a Q of ~10 eventually, and should at the very least break even soon after is built (2026?). This is no longer 15 years away, steady progress has been made over the last decades and most engineering problems have been solved, it's mostly a matter of funds and political interest now.

I absolutely agree with your last point though. Fission is right now the way to go. There are some super safe and efficient 3+ Generation reactor designs available right now. I hope the gas crisis in Europe works as a wake up call for nuclear reactors to start being built again. It worked, at least in France. But again, it's a matter of political will, mostly...

 

[NT1440]

Oooooooh, “when” this actually becomes viable, who’s getting the private contracts to enslave humanity to the capitalist model instead of bringing free energy to the world?

 

[Hrafn]

Oooooooh, “when” this actually becomes viable, who’s getting the private contracts to enslave humanity to the capitalist model instead of bringing free energy to the world?

Musk, silly. He'll get a sweetheart loan from the Trump foundation.

 

[AG_PhamD]

Musk, silly. He'll get a sweetheart loan from the Trump foundation.

Let me preface this by saying I am no fan of Elon Musk, before it was fashionable. I believe it was the Obama administration that gave Tesla half a billion dollars, allowing them to still exist today. Trump, who I’m also not a fan of, tried to cut finding to Tesla and other EV manufacturers. Biden just dealt out a bunch of money, including to Tesla, for battery production. The Obama administration also gave at least half a billion in direct grants to Solar City (a subsidiary of Tesla)- not to mention $750m from NY State. The Bush administration gave SpaceX a boost funding the Falcon 9 rocket, but under Obama the Commercial Orbital Transportation and Commercial Crew programs really pushed things along to where SpaceX is today.

I’m not sure where Trump and Musk stand these days, but I know there was a feud during the Trump administration because Trump was trying to cut federal benefits to Tesla.

Not to mention the insane stock valuation of Tesla, up until relatively recently. How was it that a company that struggles to produce (let alone deliver or service) 350,000 cars/year in 2019 is valued more than a company like VW, who makes that many cars in 2 weeks?

Between the Biden, Trump, and Obama administrations Musk entities have gained many billions in grants, subsidies, tax breaks, government/military contracts, etc. To a large extent he is a monster of our own creation. He 100% wouldn’t be here without government subsidies (which he conveniently is opposed to for other companies).

If Musk can bring a commercially viable fusion reactor online faster and at lower cost, I’m all for it regardless of his personality. I don’t think that’ll be the case though. I know you’re being factious but as it is, power plants are such extremely complex systems and usually rely on a multitude of companies to make the individual major components ie (one company makes the boiler, one company makes the generator, one company makes the control system, one company to make the cooling system, one company to make the transformers, etc).

 

[NT1440]

Let me preface this by saying I am no fan of Elon Musk, before it was fashionable. I believe it was the Obama administration that gave Tesla half a billion dollars, allowing them to still exist today. Trump, who I’m also not a fan of, tried to cut finding to Tesla and other EV manufacturers. Biden just dealt out a bunch of money, including to Tesla, for battery production. The Obama administration also gave at least half a billion in direct grants to Solar City (a subsidiary of Tesla)- not to mention $750m from NY State. The Bush administration gave SpaceX a boost funding the Falcon 9 rocket, but under Obama the Commercial Orbital Transportation and Commercial Crew programs really pushed things along to where SpaceX is today.

I’m not sure where Trump and Musk stand these days, but I know there was a feud during the Trump administration because Trump was trying to cut federal benefits to Tesla.

Not to mention the insane stock valuation of Tesla, up until relatively recently. How was it that a company that struggles to produce (let alone deliver or service) 350,000 cars/year in 2019 is valued more than a company like VW, who makes that many cars in 2 weeks?

Between the Biden, Trump, and Obama administrations Musk entities have gained many billions in grants, subsidies, tax breaks, government/military contracts, etc. To a large extent he is a monster of our own creation. He 100% wouldn’t be here without government subsidies (which he conveniently is opposed to for other companies).

If Musk can bring a commercially viable fusion reactor online faster and at lower cost, I’m all for it regardless of his personality. I don’t think that’ll be the case though. I know you’re being factious but as it is, power plants are such extremely complex systems and usually rely on a multitude of companies to make the individual major components ie (one company makes the boiler, one company makes the generator, one company makes the control system, one company to make the cooling system, one company to make the transformers, etc).

It’s worse than that. TONS of retirement funds and (other car manufacturing, via being reliant on buying Tesla’s carbon offset assets) now are tied up with Tesla holdings. To the point that if that ridiculous company tanks at any point, Tesla is now a *systemic risk* in our economy.

We have to root for the company, because it’s been allowed to get get so deeply entwined in the economy that if it tanks a ton of other major companies go with it, along with millions of peoples 401Ks (which are a scam built to further entrench people’s retirement livelihoods into this corporate run economy).

I suspect Teslas *outrageous* valuation, given it’s actual production output, has been the economic insiders (hedge funds, other companies way over invested, etc) trying to keep it afloat given the systemic risk they are all aware of. It’s a house of cards.

Whoopsie.

 

[Yoused]

Oh Em Gee. They did a 60 Minutes story on this tonight, and the NIF scientist said that we could have working fusion power plants in just twenty years.

I mean, I have laughed harder, but not by much.

 

[KingOfPain]

In the 80s I read a non-fiction book by Arthur C. Clarke, where he described a few things that would show up in our future soon.
You guessed it, fusion was one of those...

 

[Yoused]

I would be interested in cold fusion research. Not the blistering hot Ponds/Fleischman cold fusion concept, but 30°K cold fusion. If you could generate a cold plasma, confining and compressing it would be much simpler than a zillion-degrees-hot plasma. If you squeeze the right fuel hard enough, you can get fusion to happen at cryotemps and harvest the fast neutrons in the outer jacket.

 

[Roller]

In the 80s I read a non-fiction book by Arthur C. Clarke, where he described a few things that would show up in our future soon.
You guessed it, fusion was one of those...

Was it Profiles of the Future? I read the original version in 1967 and was fascinated by Clarke's predictions.

 

[Yoused]

My HS math/Comp-Sci teacher essentially was predicted the iPhone in the early seventies. He did not say "phone" because that was not really on his radar, but he said that we would have computers the size of pocket calculators by te early two-thousands. There was no mention of GUI touch interface or any kind of internets. He just envisioned the smallification of computers.

 

[KingOfPain]

Was it Profiles of the Future? I read the original version in 1967 and was fascinated by Clarke's predictions.

Sorry, I unfortunately cannot recall the title. It was a German translation anyway.

 

[Roller]

Sorry, I unfortunately cannot recall the title. It was a German translation anyway.

Thanks. I bet that's what it was. As far as I know, he only wrote one non-fiction book with his tech predictions.

 

[KingOfPain]

I thought it might be in storage, but I actually found it on one of my bookshelves. The original title is:
Arthur C. Clarke‘s July 20, 2019: Life in The 21st Century

But I was wrong about when I read the book, because this German edition was printed in August 1990.
The weird title is a reference to the moon landing, since the first chapter deals with it.
Edit: Thinking about it, 2019 should also be the year in with the original Blade Runner movie is set. I‘m not sure if that‘s intentional or not, but one chapter title is roughly A Day in the Life of a Robot.