Not really. Stellar fusion is only a few watts per ton of hydrogen fuel. That would never be a practical energy source if replicated on earth, but it is why stars can burn for billions of years.
It's pretty easy to forget that the sun is a ball of gas. It not very dense (roughly 1/4 the density of the earth). So it's extremely unlikely that a specific hydrogen atom will undergo fusion during the lifetime of the sun. However the sun is also ENORMOUS. So in the aggregate there is a lot of fusion going on and a lot of energy is being produces. However its not super dense.
The large power output of the Sun is mainly due to the huge size [...] Theoretical models of the Sun's interior indicate a maximum power density, or energy production, of approximately 276.5 watts per cubic metre at the center of the core,[77] which is about the same power density inside a compost pile
Inside the sun's core, where the fusion occurs, the density is considerably higher - 150 g/cm3 vs. 5.5 g/cm3 (average) or 13 g/cm3 (inner core) for the earth.
So it's low in terms of "amount of energy in relation to amount of hydrogen in the star"? But wouldn't much of the un-fused hydrogen be eligible to fuse at some later time, if for instance convection streams pull it down into the deeper (=higher-pressure) zones of the star?
So it seems to me the relevant measurement ought to be the ratio of energy per hydrogen atoms fused, not hydrogen atoms total. (Certainly in a reactor, where presumably all or most of them should be.)
Not necessarily for humans. Stars get to be gravity-mass containment/fusion pressure automoderated, but there is nothing known even in blue sky to manipulate gravity beyond sheer mass. So humans have to deal with containing via electromagnetism and relatively little isolated surrounding matter, which means we have to deal with both ignition thresholds and the issue of neutrons. In stars neutrons provide heat/energy by sheer virtue of hitting stuff as they pass out from the core, and obviously radioactivity isn't a concern. But it's definitely an issue for a power plant, where neutrons mostly represent wasted energy since we don't have any good way to harness them for useful work, plus they'll activate surrounding materials over time which is also an irritation. So the ideal fusion may be aneutronic, which then faces tradeoffs of fuel availability vs ease of production.
For example terms of plain ease of engineering/compactness etc, helium-3 would probably be the best fuel. But helium 3 is also super ultra rare and effectively nonexistent naturally on Earth. We'd have to get it from space, or breed it from neutronic reactors which obviously isn't ideal (might be worth it as a compact fuel for far future spacecraft though). On the opposite side hydrogen-boron fusion requires much much higher ignition energy, greatly complicating actually building a working plant, but the fuels themselves are readily available. There are other potential ones between, or maybe we just suck it up and deal with neutron activation (still a lot less radioactivity than in fission plans), which could be moderated with more careful selection of construction materials.
At any rate, we're definitely not yet at the point where we can really say what the "best" fusion cycle will be, and it's perfectly possible we'll even end up with multiple (He3 for vehicle/more portable reactors, H-B or D-D for massive stationary ones say).
Hydrogen - Boron fusion is a really good thing, if it can be achieved. It requires high temperatures, but is aneutronic [1], only producing electrons which can easily be siphoned off to directly power an electrical grid:
Proton – Boron-11: 1p + 11B → 3 4He + 8.7 MeV
This means it doesn't create radioactive waste or have the metal brittling problems that neutronic (H-H or H-D) fusion has.
Some groups [2][3] think the high temperatures can be achieved in small plasma bubbles without large amounts of magnetic confinement, making for small and inexpensive reactors.
We see this is large blue stars. They burn through their fuel supply in millions of years and years and explode with the force of a billion suns. Meanwhile small stars can burn for billions and billions of years because they are conservative with their hydrogen supply.
One could think that all energy generated by nuclear fission is just energy stored from fusion. So it should be pretty clear fusion yields more as it is input for all fission.
Except it's not. The energy in fission comes from gravitational collapse, not from fusion (collapse is what allows free neutrons to be liberated to drive the r-process that makes uranium.)
Isn't it still nuclear fusion, just endothermic? The nuclei are formed from smaller particles colliding and fusing together. Or is it just lots of neutron capture?
> energy in fission comes from gravitational collapse
Do you have a source for gravity being in any way involved with fission’s energy release? I thought, and the article says, it’s from “the energy of the electromagnetic force when positively charged parts of the nucleus fly away from one another.” (EDIT: Ah, nvm.)
I think this was talking about "energy" in the same very loose sense as the original post, which was using the term energy for talking about the provenance of the fuel. The heavy elements used in fission are not created in the almost eternal smoldering suns but in the relatively short events that follow.
You're missing the point, the energy released in fission is energy stored in the uranium billions of years ago when it was formed by a star collapsing, that collapse was powered by gravity.
Good explanation, but it kind of buried the lede. The answer is way down at the bottom:
> Fission releases the energy of the electromagnetic force when positively charged parts of the nucleus fly away from one another. Fusion releases the energy of the strong force (much stronger at short distances than the EM force) when the small pieces are captured and held into one nucleus.
Fusion of tritium and deuterium
Input:
- 1 deuterium atom: 1 proton and 1 neutron
- 1 tritium atom: 1 proton and 2 neutrons
Output:
- 1 helium atom: 2 protons and 2 neutrons
Overall: 5 particles become 4, 20% become free to hit other atoms or turn into energy.
Decay of uranium-235 into barium and krypton
Input:
- 1 atom of uranium: 235 particles
Output:
- 1 atom of barium: 144 particles
- 1 atom of krypton: 89 particles
Overall: 235 particles become 233, less than 1% become free to hit other atoms or turn into energy.
Fission is much more similar to radioactive decay: big atoms turn into smaller atoms releasing a few free particles. Fusion releases basically the same amount of particles with much smaller atoms.
Of course, this is an extreme simplification of the process. Energy required to start each reaction is ignored, but the idea is mostly correct, I think.
EDIT: fixed tritium number of protons and neutrons and grammar.
This is not quite right. In no cases are nucleons converted into energy. D+T fusion releases a He4 atom and a free neutron - 5 in, 5 out. U235 absorbs a neutron, then decays into Ba144, Kr89, and 3 neutrons - 236 in, 236 out.
In the fission case, the neutrons escape to hit other U235 atoms, which causes the well-known chain reaction. In the fusion case, neutrons are not an input to the process.
The released energy comes from the binding energy of the nuclei. The nucleons in a U-235 atom are held together weakly.
My lay understanding which is probably wrong, is that the protons are trying to repel each other electromagnetically but are held together by the strong force, so splitting the atom means less strong force is "required" to hold the two new nuclei together.
I don't have a solid understanding of exactly how the binding energy is lower in a He4 nuclei, and would love someone to explain!
There's a balance of forces which results in binding energy per nucleon depending on the number of nucleons, but overall it's something like 8 MeV per nucleon, so He-4 ends up with ~28.3 MeV of binding energy, while U-235 goes up to 1.8 GeV due to the sheer number of nucleons. Note the downwards slope of the binding energy per nucleon in the graph, this kinda indicates how stability goes down. There are however "magic numbers" and iirc there is a hypothesis that you get a sort of "island of stability" way into not-yet-existing super-heavy trans-uranic elements.
I was watching "Undecided with Matt Ferrell" on it Dr. Martin Greenwald said that 0.1g deuterium and 0.3 lithium fused in a fusion power plant would release enough energy to power a typical US home for a year. Actually he said energy needs for an American so he may have meant more than electrical power.
From what I can see the typical US home uses 10,715kWh/year which is 38,574,000,000 Joules.
The bottleneck of nuclear fusion in all stars is the proton-proton chain reaction, turning two protons into deuterium. Only deuterium can further fuse, and usually deuterium doesn't survive in the sun for long. This reaction is caused by the weak force, so it's extremely slow. This is what's slowing the fusion in all the stars.
"The average proton in the core of the Sun waits 9 billion years before it successfully fuses with another proton. It has not been possible to measure the cross-section of this reaction experimentally because it is so low"
Because it's a non sequitur. The metric of "energy/mass of fuel" has nothing to do with why any actual consumer would consider the source to be better. Metrics like cost, safety, pollution, etc. are relevant; that metric is not.
And, thus far, fusion yields exactly zero joules of useful energy (unless you count solar).
It hasn't even vaporized anybody yet, never mind anybody you might have wanted to have vaporized.
(Tokamak) fusion will never produce commercially competitive energy. Every plausible system design would be overwhelmingly more expensive to build and operate than a fission plant of similar rating. And, fission is already not competitive, and gets less so every day.
(I still hold out hope for FRC De-He3 fusion for space probe propulsion--4 years to Pluto--but Tokamak eats all the research dollars, so who knows if we will ever find out?)
Well, Helion did just get $500M in funding, and is very close to that design space.
I have doubts about Helion being able to do it, but if I had to invest in fusion it would be them (and just possibly Zap). Going beyond DT and recovering energy electromagnetically addresses the engineering/economic showstoppers that make things like tokamaks burning DT seem like a dead end.
Helion is using a FRC (Field Reversed Configuration), which was what I was responding to.
Also: Helion is taking a riskier physics approach in order to reduce engineering difficulty. I consider this a very good idea.
Ordinary DT reactors make heat (80% of the energy comes off in neutrons, the energy of which gets turned to heat in a blanket.) Fission reactors should be able to make heat more easily and cheaply, so what is the point?
Helion's approach, on the other hand, has plasma energy converted directly to electrical energy by expanding against a magnetic field. If this can be made to work, it means the non-nuclear side of a fusion power plant can be made much smaller (ideally, turbines/heat exchangers/cooling towers could be avoided entirely; realistically they could be made much smaller for a given plant power output). Their fuel choice (DD, D3He), while much more difficult to burn, subjects the materials of the reactor to much less radiation damage. There is a realistic chance the first wall of a D3He reactor could last the life of the power plant.
Either the fusion fuel gets mixed with so much inert material that the mass of the fuel doesn't matter, or the reactor is so massive compared to the fuel that again the fuel mass doesn't much matter.
Yes. The difference between fossil fuels and fission is so gigantic that compared to fossil and fusion it really doesn't matter.
Fission actually has the cheaper more abundance fuel source.
Fission is technically much, much easier.
I really don't understand the money we waste on fusion, with that money we could have build viable fission thermal breeders that could have solved all energy our problems long ago.
Unless your trying to build fusion rockets for interstellar probes, fission is preferable in almost every situation.
49 comments
[ 5.6 ms ] story [ 102 ms ] threadFrom wikipedia: https://en.wikipedia.org/wiki/Sun#cite_ref-power_production_...
The large power output of the Sun is mainly due to the huge size [...] Theoretical models of the Sun's interior indicate a maximum power density, or energy production, of approximately 276.5 watts per cubic metre at the center of the core,[77] which is about the same power density inside a compost pile
The Sun Is A Mass Of Incandescent Gas https://g.co/kgs/g8rKyh
A follow up song, correcting the incorrect physics of the first one:
Why Does the Sun Really Shine? https://g.co/kgs/2QKBdw
So it seems to me the relevant measurement ought to be the ratio of energy per hydrogen atoms fused, not hydrogen atoms total. (Certainly in a reactor, where presumably all or most of them should be.)
Not necessarily for humans. Stars get to be gravity-mass containment/fusion pressure automoderated, but there is nothing known even in blue sky to manipulate gravity beyond sheer mass. So humans have to deal with containing via electromagnetism and relatively little isolated surrounding matter, which means we have to deal with both ignition thresholds and the issue of neutrons. In stars neutrons provide heat/energy by sheer virtue of hitting stuff as they pass out from the core, and obviously radioactivity isn't a concern. But it's definitely an issue for a power plant, where neutrons mostly represent wasted energy since we don't have any good way to harness them for useful work, plus they'll activate surrounding materials over time which is also an irritation. So the ideal fusion may be aneutronic, which then faces tradeoffs of fuel availability vs ease of production.
For example terms of plain ease of engineering/compactness etc, helium-3 would probably be the best fuel. But helium 3 is also super ultra rare and effectively nonexistent naturally on Earth. We'd have to get it from space, or breed it from neutronic reactors which obviously isn't ideal (might be worth it as a compact fuel for far future spacecraft though). On the opposite side hydrogen-boron fusion requires much much higher ignition energy, greatly complicating actually building a working plant, but the fuels themselves are readily available. There are other potential ones between, or maybe we just suck it up and deal with neutron activation (still a lot less radioactivity than in fission plans), which could be moderated with more careful selection of construction materials.
At any rate, we're definitely not yet at the point where we can really say what the "best" fusion cycle will be, and it's perfectly possible we'll even end up with multiple (He3 for vehicle/more portable reactors, H-B or D-D for massive stationary ones say).
Proton – Boron-11: 1p + 11B → 3 4He + 8.7 MeV
This means it doesn't create radioactive waste or have the metal brittling problems that neutronic (H-H or H-D) fusion has.
Some groups [2][3] think the high temperatures can be achieved in small plasma bubbles without large amounts of magnetic confinement, making for small and inexpensive reactors.
[1] https://en.wikipedia.org/wiki/Aneutronic_fusion
[2] https://lppfusion.com/technology/dpf-device/
[3] https://hb11.energy/how-it-works/
We see this is large blue stars. They burn through their fuel supply in millions of years and years and explode with the force of a billion suns. Meanwhile small stars can burn for billions and billions of years because they are conservative with their hydrogen supply.
[1]: https://en.wikipedia.org/wiki/Tsar_Bomba
Edit: I see, it's neutron capture: https://en.m.wikipedia.org/wiki/R-process
Do you have a source for gravity being in any way involved with fission’s energy release? I thought, and the article says, it’s from “the energy of the electromagnetic force when positively charged parts of the nucleus fly away from one another.” (EDIT: Ah, nvm.)
> Fission releases the energy of the electromagnetic force when positively charged parts of the nucleus fly away from one another. Fusion releases the energy of the strong force (much stronger at short distances than the EM force) when the small pieces are captured and held into one nucleus.
Of course, this is an extreme simplification of the process. Energy required to start each reaction is ignored, but the idea is mostly correct, I think.
EDIT: fixed tritium number of protons and neutrons and grammar.
In the fission case, the neutrons escape to hit other U235 atoms, which causes the well-known chain reaction. In the fusion case, neutrons are not an input to the process.
The released energy comes from the binding energy of the nuclei. The nucleons in a U-235 atom are held together weakly.
My lay understanding which is probably wrong, is that the protons are trying to repel each other electromagnetically but are held together by the strong force, so splitting the atom means less strong force is "required" to hold the two new nuclei together.
I don't have a solid understanding of exactly how the binding energy is lower in a He4 nuclei, and would love someone to explain!
https://upload.wikimedia.org/wikipedia/commons/thumb/5/53/Bi...
https://www.youtube.com/watch?v=SgM2wxELF4U
From what I can see the typical US home uses 10,715kWh/year which is 38,574,000,000 Joules.
"The average proton in the core of the Sun waits 9 billion years before it successfully fuses with another proton. It has not been possible to measure the cross-section of this reaction experimentally because it is so low"
https://en.wikipedia.org/wiki/Proton%E2%80%93proton_chain
In nuclear reactors, we can start from deuterium, making the process much faster and power intensive than the sun.
This is a completely dumb argument.
(why do we think it's dumb again?)
It hasn't even vaporized anybody yet, never mind anybody you might have wanted to have vaporized.
(Tokamak) fusion will never produce commercially competitive energy. Every plausible system design would be overwhelmingly more expensive to build and operate than a fission plant of similar rating. And, fission is already not competitive, and gets less so every day.
(I still hold out hope for FRC De-He3 fusion for space probe propulsion--4 years to Pluto--but Tokamak eats all the research dollars, so who knows if we will ever find out?)
Well, Helion did just get $500M in funding, and is very close to that design space.
I have doubts about Helion being able to do it, but if I had to invest in fusion it would be them (and just possibly Zap). Going beyond DT and recovering energy electromagnetically addresses the engineering/economic showstoppers that make things like tokamaks burning DT seem like a dead end.
Also: Helion is taking a riskier physics approach in order to reduce engineering difficulty. I consider this a very good idea.
Ordinary DT reactors make heat (80% of the energy comes off in neutrons, the energy of which gets turned to heat in a blanket.) Fission reactors should be able to make heat more easily and cheaply, so what is the point?
Helion's approach, on the other hand, has plasma energy converted directly to electrical energy by expanding against a magnetic field. If this can be made to work, it means the non-nuclear side of a fusion power plant can be made much smaller (ideally, turbines/heat exchangers/cooling towers could be avoided entirely; realistically they could be made much smaller for a given plant power output). Their fuel choice (DD, D3He), while much more difficult to burn, subjects the materials of the reactor to much less radiation damage. There is a realistic chance the first wall of a D3He reactor could last the life of the power plant.
Fission actually has the cheaper more abundance fuel source.
Fission is technically much, much easier.
I really don't understand the money we waste on fusion, with that money we could have build viable fission thermal breeders that could have solved all energy our problems long ago.
Unless your trying to build fusion rockets for interstellar probes, fission is preferable in almost every situation.
ITER, if it could run continuously indefinitely, would take 300,000 years to fuse its own mass in fuel.
Beamed power makes more sense for that use case.