The Fusion Reactor Next Door
nytimes.com
nytimes.com
From my limited reading it sounds like this does a pretty great job of solving energy and climate to a large extent.
Why aren't more people excited about this? I feel like this can be a very politically viable solution that pleases most of the clout wielding sides of the political spectrum.
What am I missing?
https://www.forbes.com/sites/amorylovins/2014/09/07/fusion-p...
Thanks for this excellent piece.
https://www.greenbuildingadvisor.com/article/solar-thermal-i...
General Fusion - one of the first private fusion companies - has been going for 12 years without revenue and might need another ten. That’s a long time for venture capital.
The commercial stage ist decades away, there isn't even energy surplus generated in research. No reactor can run longer than some miliseconds, i think you are overly optimistic. Fusion is 20 years away since 50 years.
But, even if we had solved plasma problems for continuous operation with Q >> 1, the economics of fusion look even worse than fission. The lesser scientific problems of reaching breakeven and ignition are concealing the greater problems of making it worth doing at all.
The usual mistake people make in thinking about fusion is to focus on fuel cost and assume, incorrectly, that fuel cost determines the cost of power. The equipment has to be financed and paid for and maintained, and the cost of doing that will dominate.
Even in so-called "advanced fuel" reactors, burning a mixture other than DT, there will be enough neutrons to render the reactor structure too radioactive for hands-on maintenance. This will greatly complicate the problem of fixing the reactor if it breaks.
Research has to work toward broad goals and investigate new areas. In 1900 could you have made a 20 year plan to invent cell phones and all the technology needed for what we have today? No one would suggest that, but you can invest in basic science and engineering and you'll get some useful things along the way, radios, maybe repeaters, ever smaller ones and better batteries. But you don't necessarily get a specific thing.
The big one they claim to solve, cost, is an unproven assertion based on the assumption that nuclear power plant production can follow the "industrial learning curve" model of cost reduction. There is plenty of proof extant that this is a flawed assumption.
4. The power density of the Sun is surprisingly low, it's about 276 W/m3 [1]. If we would ever be able to create fusion similar to the one that's happening in the Sun, we'd need a few million m3 to get to the power generated by a typical large power plant (1 GW). Imagine what type of containment vessel you need to enclose a few million m3 of hot (100 million degrees) stuff. For comparison, the largest oil tankers in the world can hold about 300,000 m3 of oil.
3. Even assuming we can build such a humongous fusion based power plant, what happens if, you know, something goes unplanned? Anti-nuclear people point to Chernobyl and Fukushima, what could happen with a fusion power plant? You would not have radiation, that's true, but you would have an explosion that would make the Tsar Bomba look like firecrackers. Why? All that hydrogen that is supposed to undergo fusion needs to first be heated to about 100 MM degrees. That's a lot of energy. How much? Here's very rough back of the envelope: density of Sun = 1 kg/m3; let's say we are looking at 10 MM m3 of hydrogen (we account for some power conversion inefficiency), that's 10 Megatons of hydrogen. At about 15 kJ/kg.K specific heat, it takes about 1.5 billion kj to heat one kg of hydrogen, or 1500 GJ. This is about 0.3 T TNT equivalent [2]. Overall, about 3 Gigatons TNT equivalent. That's 60 times larger than the largest nuke ever tried (the Tsar bomba mentioned before). Or 200,000 larger than the Hiroshima bomb.
2. The accident discussed above assumes nothing goes really bad. It's simply a rapid release of some thermal energy. But what if instead of this, we have an accelerated fusion reaction? Then we have at hand many orders of magnitude more fusion fuel than all the thermonuclear bombs ever build. I don't want to do any calculations, but there's a chance that could be the end of the humans, or maybe all multicellular life on earth. 1. All this hope for clean nuclear energy means a lot of people don't want to even consider practical solutions for the fission-based nuclear energy, such as waste reprocessing or next generation nuclear reactors (like Bill Gates's Terrapower). Why would they, if clean nuclear energy is just around the corner? That kills what is probably our most realistic tool to reduce carbon emissions.
[1] https://physics.stackexchange.com/questions/370899/suns-powe... [2] https://en.wikipedia.org/wiki/TNT_equivalent
That's because the Sun is very different from an Earthbound fusion reactor. It runs at a very different temperature and density, using very different reactions. The analogy you are trying to draw is invalid.
> you would have an explosion that would make the Tsar Bomba look like firecrackers
No, you wouldn't, because only a very small amount of fuel is inside the reactor at any given time. The inside of a fusion reactor is very different from a nuclear bomb. The analogy you are trying to draw is invalid.
> what if instead of this, we have an accelerated fusion reaction?
You can't. If anything goes wrong the injection of fuel into the reactor stops and the reactor shuts down.
No, the analogy he draws is the correct one: the sun actually works. "Earthbound" controlled nuclear fusion doesn't in any meaningful way relating to energy production. There's no reason physics should cooperate with us and make it possible; "wouldn't it be nice" wishful thinking isn't good enough, and the results thus far are all complete and total crap. It isn't that we haven't spent enough: we simply don't know how to do this -there is no reason to believe it is possible. All the results from inertial confinement, tokomaks, fusors, whatever indicates we really don't know WTF we're doing.
His other concerns are legit as well. If you can do controlled nuclear fusion, you have a really, really scary weapon.
Under, as I said, very different conditions from those which Earthbound fusion reactors are attempting. The only meaningful similarity between the two is that both are fusion reactions. You certainly cannot use numbers for the Sun to make any valid estimates about Earthbound fusion reactors, as the poster I responded to was attempting to do.
> Earthbound" controlled nuclear fusion doesn't in any meaningful way relating to energy production.
It hasn't yet, but that doesn't mean it never will. If it does, it certainly will not be by reproducing conditions similar to those that produce fusion in the Sun. And if your criticisms of fusion power turn out to be valid and it never works, then everything the poster I responded to said is irrelevant anyway, since he was assuming fusion power would work.
> If you can do controlled nuclear fusion, you have a really, really scary weapon.
No, you don't, because, as I said, only a very small amount of fuel is in the reactor at any given time--enough for a few seconds' worth of energy production. That equates to the energy equivalent of a few tons of TNT, not a few megatons. And since the entire system is designed to remove that amount of energy at that rate, concerns about even that few tons of TNT energy equivalent "exploding" are invalid--that amount of fuel undergoing complete fusion and liberating all that energy is what the reactor is designed for, not an abnormal condition. And, as I said, all you need to do to stop energy production is stop injecting fuel, and the reactor shuts down.
Again, a fusion reactor is very different from a fusion bomb.
If we do figure out how to do this, of course the possibility of a controlled fusion reactor can be used as an absurdly powerful weapon. You've basically invented something that can make gigawatts of energy appear in a small volume. Sure, a power-generating fusion reactor won't make as big a mess as a fission reactor if there is a fault. That's not the point. If you know how to do this, you can design a system where you can open a hole in one end and point it at things you don't like. And people most certainly will do this if they can. Making the non-proliferation aspects of such a postulated technology absurd. Imagine you could build a NIF type device that actually was sustainable, and you could fit into a truck. Just the neutron flux from the thing would be absurdly murderous, and even using NIF figures (~200MJ), being able to direct the energy of some fraction of a ton of TNT worth of energy at things you don't like, fueled by a tiny amount of hydrogen seems pretty dangerous to me.
No, it isn't. We think it's possible because we have extensive experimental data on fusion reactions from Earthbound experiments. From those experiments we know a wide range of possible conditions that can produce sustained, controlled fusion. The conditions inside the Sun are only a very narrow part of that range.
> You've basically invented something that can make gigawatts of energy appear in a small volume.
You do realize that any power plant that is making gigawatts of energy has the same amount of fuel energy content inside it as a fusion reactor, right? A coal plant that is making a gigawatt of energy has a ton or so of TNT's worth of fuel energy inside it at any given time, just like a fusion reactor. There is no difference between them in terms of weapon potential. It's true that the amount of fusion fuel needed to make a ton or so of TNT's worth of energy occupies a much smaller volume than the amount of coal or oil or natural gas needed to make that amount of energy, but so what? It's the energy content that matters, not the volume.
> Imagine you could build a NIF type device that actually was sustainable, and you could fit into a truck.
That means you're building a laser capable of short pulses of gigawatts of power that can fit into a truck. If you can build that, you don't need to include a pellet of fusion fuel with it to make it a weapon.
Also, nobody is talking about building gigawatt fusion reactors that can fit into trucks. They're large facilities, just like other power plants.
> being able to direct the energy of some fraction of a ton of TNT worth of energy at things you don't like
We already have weapons that can do this, and are much cheaper than building a fusion reactor with a hole in the side of it would be. They're called missiles.
There is zero evidence sustained, controlled, above break-even fusion works at the types of energy densities people talk about for power plants. None; keine, zero, zilch: there is arguably as much evidence for this as the existence of ghosts or leprechauns. The models don't work! I ain't saying we shouldn't try, but there are by now reasonable arguments to be made that it's a big waste of time and money better spent on something boring that we know will work, like thorium.
To make another technological analogy from nature: if the earth's gravitational potential well were larger, we wouldn't have been able to achieve orbit using chemical rockets. The specific impulse of fuels which exist in the periodic table of the elements would have been insufficient to achieve this. For example: https://arxiv.org/abs/1804.04727 -similarly, the properties of plasmas may make it impossible to achieve controlled, above break-even nuclear fusion in a conveniently small and dense volume. There is at present no evidence that it can be achieved. They've been working on it for 65 or 70 years now depending on how you count things, you'd think there would be some positive indications by now beyond optimistic press releases.
As for your wishful thinking that a controlled nuclear fusion technology couldn't be used as a weapon, well, there you go again with that wishful thinking. Controlled fusion is a non-proliferation nightmare. Should this actually come to pass and become routine enough to be a meaningful part of the power grid, it will be a hell of a lot easier to build a fusion howitzer than an ICBM with H bomb at the business end of it.
I mean, you're a nuclear physicist, man!
This is just another way of saying we haven't gotten it to work yet. That's true, we haven't. The same could have been said about controlled fission in 1941.
> there are by now reasonable arguments to be made that it's a big waste of time and money better spent on something boring that we know will work, like thorium.
I agree there are reasonable arguments to be made about this. But comparing controlled fusion to ghosts and leprechauns is not one of them.
> it will be a hell of a lot easier to build a fusion howitzer
I have no idea where you're getting this from. Again, I really don't think you've grasped the implications of the fact that the energy content of the fuel inside a reactor producing, say, a gigawatt of power, is the same whether the reactor is fueled by coal, oil, natural gas, or fusion. So a "fusion howitzer" based on a fusion reactor would have no more explosive power than a chemical explosive howitzer (which is not based on a coal, oil, or natural gas plant anyway, it's a very different thing).
Nuclear fission is a unique case here because of the historical accident of fission power developing in such a way that multiple years' worth of fuel, containing orders of magnitude more stored energy than was being burned at any given time, was all inside the reactor core at once, instead of the reactor having fuel continually injected just in time to be burned, like every other kind of reactor. All of the downsides of fission--the possibility of meltdown, decay heat removal, huge spent fuel pools containing lots of long-lived radioisotopes, even an accident like Chernobyl--are due to this unique attribute. Newer designs like pebble bed reactors have shown that this is not a necessary attribute of fission power. But it won't be an attribute of fusion power at all.
The difference between the energy content of a coal reactor and a hypothetical fusion reactor: you can store a lot more gigawatts in the back of a pickup truck for one of them, and that is the one that makes enough neutrons to sterilize a city block when you burn them up. I mean, how many pounds of fusion fuel does it take to level a city?
Huh? That milestone was reached some years ago.
https://www.livescience.com/43318-fusion-energy-reaches-mile...
What has not yet been achieved is ignition--a self-sustaining reaction that does not require continuous energy input.
> how many pounds of fusion fuel does it take to level a city?
We already know how to make a fusion reaction that will level a city. Having a working controlled fusion reactor would not affect our capability to level cities. So I really don't understand why you keep harping on this.
The NIF result created 35000 neutrons, and you should know that press release is the functional definition of a scientific fraud -I wrote a couple of articles on it which you can go find if you care about fraud.
Yes, we can already do this, it's called a fusion bomb.
> then level a city with it using the same technology has a hypothetical fusion reactor possible
No, a bomb is not the same as a fusion reactor. A bomb burns a megaton of TNT's worth of fusion fuel all at once. A fusion reactor outputting a gigawatt of power burns the same amount of fuel in about a year. It's not at all the same thing.
Google isn't showing me anything useful. Do you have links?
So basically you're talking about a fusion-powered one-gigawatt flamethrower?
The only thing we can do with such amounts of heat to produce the kind of energy we need and consume is build thermo-electric power plants to generate electricity. Here is the crunch:
> Thermoelectric power use has a significant impact on water resources and the power sector is highly dependent on these water resources; the United States Geological Survey (USGS) estimated on a national level that 41% of all freshwater withdrawals in the United States in 2005 were for thermoelectric power operations, primarily for cooling needs.[1]
[1] Operational water consumption and withdrawal factors for electricity generating technologies - https://iopscience.iop.org/article/10.1088/1748-9326/7/4/045...
I don't see how that observation is in any way absurd. It's actually quite helpful to counter the "fusion as a panacea" narrative with a dose of reality of how power plants actually work.
This is already regulated, see for example this discussion of monitoring systems for proving compliance:
https://www.power-eng.com/articles/2007/01/external-temperat...
[1] https://www.chicagotribune.com/news/ct-xpm-2012-08-20-ct-met...
However, it doesn't have to. And in particular, as water becomes more constrained by growing population and climate change, using water the same way that it has in the past is becoming untenable.
Recent regulations are starting to end "once through" use of water this way, which is leading to the shut down of older plants that are close to the end of their lifespan anyway.
And then there are plants like this nuclear plant in Arizona which cools using treated sewage, which has almost zero impact:
https://en.m.wikipedia.org/wiki/Palo_Verde_Nuclear_Generatin...
We will need to use water resources better! There are options for doing that, but no way to make thermoelectric power plant use insignificant.
Edit: On second read, your point is that heat is a pollutant itself. My next sci-fi daydream will be about cooling down the planet by converting atmospheric heat.
2) It's not even totally accurate. Aneutronic fusion can create electricity directly from charged particles without going through the heat generation step[1]. It's an active area of research with the highest profile group working on it being TAE[2].
https://www.researchgate.net/profile/Wallace_Manheimer/publi...
The story reminds me of Transatomic. Venture capital doesn't necessarily evaluate science well.
[1]https://www.sciencedirect.com/science/article/pii/S037026931...
"None of these efforts has yet tested its device with hydrogen–boron fuel, so the anticipated performance is based on extrapolating from theory, experimental results with other fuels and from simulations."
Also, even with good proportion of energy contained in such particles to somehow tap efficiently in practice, the reactor itself is still going to need a great deal of cooling, just to stop it melting.
Also the article we are commenting on, does not seem to concern such a reactor.
"let’s assume that in 100 years or so, the world population is 10 billion, and everyone is living at a European standard of living, using 125 kWh per day derived from fossil sources, from nuclear power, or from mined geothermal power. The area of the earth per person would be 51 000 m^2. Dividing the power per person by the area per person, we find that the extra power contributed by human energy use would be 0.1 W/m^2. That’s one fortieth of the 4 W/m^2 that we’re currently fretting about, and a little smaller than the 0.25 W/m^2 effect of solar variations. So yes, under these assumptions, human power production would /just/ show up as a contributor to global climate change."
If the promise of a technology is that we can use several orders of magnitude more energy per person, then that .1W/m2 could become 10W/m2.
If we don't increase energy usage, sure, no big deal, but the entire point of pursuing fusion is to enable fantastically more energy usage, at least as I see it.
So we also need a second revolutionary technology, direct conversion of energy rather than thermal energy conversion.
Fusion will never be a cost winner if it's using thermal energy conversion, anyway. Solar and wind will soon undercut the steam turbine lifetime costs, so even if the fusion part were free, there's no way for heat engines to compete on cost.
Fiction, science fiction and storytelling generally are ideal for that sort of usage. They let people pre-consider a lot of the arguments and assumptions of the people they are talking to.
Depending on what type of energy you exactly have in mind.
It is totally possible to increase consumption of iee. electric energy and reduce overall energy needs. We burn most of the fuels and use them very inefficiently:
https://en.wikipedia.org/wiki/Primary_energy
In short: you need to burn coal with 3x as much energy to generate 1x of electricity. So when you switch from coal to nuclear (or renewable) and increase 2x your energy usage you still come up ahead in total primary energy consumption.
https://energypost.eu/iea-underreports-contribution-solar-wi...
No, it's to enable enough energy usage for, as MacKay says, a first world standard of living for the entire world's population, instead of just a small fraction of it. That's more than today, but not fantastically more.
Also, as technology advances, the energy usage required for a first world standard of living goes down. Our entire house now has LED bulbs; that cuts our energy usage for lighting to about 10% of what used to be the standard (incandescent bulbs), or about a third of the previous "low energy" standard (fluorescent bulbs--which also have hazmat issues since they contain mercury). Computers today use only a small fraction of the energy of older computers while still delivering more computing power. HVAC systems and refrigerators are more and more efficient. Cars are more efficient--I have a small SUV with a four cylinder engine (up until a few years ago I would never have even considered a four cylinder engine for an SUV, but this one has good enough performance) that gets better gas mileage than the 4-door sedan with a V6 engine that I had before it. The list goes on.
WOHA was a cursory and self acknowledged "back of the envelope" style reference, it is also now outdated.
Heat pollution is often not a minor concern locally, see introductions such as "Thermal pollution: an overview"[1]. Even globally heat pollution may have a significant contribution to climate, see studies such as "Global anthropogenic heat emissions from energy consumption" [2]
>Although the mean Antropogenic Heat Flux is small compared to the forcing of GHGs, it may exert quite distinctive effects on the climate and the environment because of the surface-based emissions and uneven geographical distribution.[2]
[1] https://www.sciencedirect.com/topics/earth-and-planetary-sci...
[2] https://link.springer.com/article/10.1007/s10584-017-2092-z
https://www.forbes.com/sites/amorylovins/2014/09/07/fusion-p...