There is quite some misleading terminology being used in the field of nuclear fusion (see excellent overview here [1]). So I was genuinely curious what kind of Q is used in this paper, so here it is:
> In this regime, self-heating from α-particle deposition exceeds the external heating input into the DT8; this ratio is denoted Qα, where the self-heating is taken relative to the heating power to the plasma—for inertial fusion this is the PdV compressional work on the fuel and not the total laser energy (P, pressure, dV, volume change). Qα > 1 is a burning plasma.
So as far as I understand such Qα > 1 can be reached without sustaining neither pressure or volume, so one can think about it as an explosion rather that intuitive burning (somewhat continuous process).
To summarize: Q_total = total_energy_out / total_energy_in is what matters, not any Q related to plasma energy. Since to build a power plant we just care what it consumes in comparison with what it produces. For this particular case even though Qα > 1 you likely can extract just a tiny fraction of that energy stored in plasma, while spending huge amounts of energy to heat it in the first place.
Edit: I'm a big proponent of nuclear fission and fusion since both are the best (current and future) way of getting carbon neutral energy, which can cover the spikes in the grid and sun/wind-less days. It is just important to be careful with the terminology, which can be used to make research results more optimistic.
That's a great video. Her videos often offer a really refreshing sceptical balance to the dominant overly optimistic and downright incorrect pop-sci stuff out there.
So their denominator (energy in) is calculated in terms of the work that was usefully done in compressing the plasma, and not the total energy they used to make said work happen?
Any idea what multiple of PdV the input power was? (ie. energy used to power the lasers?) I'm curious what their total Q value would be in this experiment...
This type of laser is about 1% efficient (someone will correct me if I’m out of date on this, but it’s not likely to be better than 10%). Laser fusion really has nothing to do with energy production.
And, even if they got the actual Q up by two orders of magnitude, inertial D-D fusion can never be practical, because you can't fuse enough. And, even if you confined it magnetically, it costs so much to extract D-D energy that it can never compete even with fission. So it is really a dead end.
At the end of the video you linked, she accuses the author of The Star Builders of "only briefly mention[ing] the total energy gain and never giv[ing] a number".
I had just started the book when I watched her video, and I counted four separate times where he did mention total energy gain. In one chapter (pg. 142 in my kindle edition), he quotes from a source who lays out the distinction she claims people in fusion research want you to be confused about:
Like First Light Fusion, Tokamak Energy is much more interested in power production than energy breakeven. "Achieving a of one is a scientific goal," Jonathan Carling continues, "but it's nowhere near enough to produce commercial energy, which requires a Q [in the region] of tens." As mentioned previously, Q is the ratio of fusion power out to heating power in. His strong view is that unless other star builders have a credible plan to get to factors of twenty or thirty more power out than they put in, then they're in the science game and not the fusion energy game.
I don't think your quote from the book conflicts with Dr. Hossenfelder's criticisms at all. The quote vaguely says "Q = 1 is a scientific goal", then describes Qplasma (and calls it Q): "Q is the ratio of fusion power out to heating power in" (that sounds like Qplasma to me.)
The source in the book is basically saying "Q[plasma] needs to be twenty or thirty times higher than it is to get more _power_ out than you put in", but there's already a term for that: Qtotal. Hossenfelder's claim was that the book never mentions Qtotal and continues the misleading technique of using Qplasma as if it's an indicator of power production (and saying it needs to be 20-30x), rather than using a much more clear term Qtotal to describe "total" energy out vs energy in.
The response video you linked is even worse: They "correct" Dr. Hossenfelder by showing papers that spout Qplasma numbers as just "Q" and talk about how ITER achieves a Q=10. But that's not in contradiction at all with Dr. Hossenfelder; her qualm is how misleading it is to talk about "Q > 1" as if it's producing useful energy when you actually mean "Qplasma > 1" (which does not produce useful energy.)
The response video then goes on to talk about how useful Qplasma is, but that doesn't matter to Dr. Hossenfelder's point, which is that it is unethical to go around touting "We're going to get 10x the energy out that we put in" when you're just referring to Qplasma=10. That's all her video said, and the response is attacking a straw man.
Yes, that should be required reading/viewing for anyone inclined to take a press release about fusion from a national lab at face value—especially anything related to laser fusion.
I simply do not have the time or energy to dispel all of the myths Sabine's words have conjured in the past half year all over again. Her argument is convincing laypeople that scientists and engineers who devote their lives to difficult, underpaying work are trying to cheat the public. As someone who has worked in fusion I can tell you there is exactly one reason why people work in fusion and it is to secure humanity's future. No level of cynicism from outsiders will change that fact, only make their lives more difficult.
The best I can do for now is point back to prior points at the cracks in the story.
Noble motives are neither necessary nor sufficient to achieve success. The laws of physics neither know nor care how altruistic and self-sacrificing fusion researchers are. If it doesn't work, it doesn't work.
I think the point he is making is that profitable fusion-based electricity generation isn’t a physics problem; it is an engineering problem. Engineering requires dealing with people in addition to physics.
Heavier than air flying machines? Epic failures across centuries led many to believe it was fundamentally impossible, despite Leonardo Da Vinci's math in the 16th century showing it was theoretically possible.
The incandescent light bulb? At one time considered to be a securities fraud! The British parliament assembled a team of investigators to probe Edison's claims.
History is full of inventions that seemed impossible until the dogged inventors made it work. Obviously not all fantastical inventions are possible, but fusion has plenty of evidence that points to its eventual possibility.
> The incandescent light bulb? At one time considered to be a securities fraud! The British parliament assembled a team of investigators to probe Edison's claims.
I don't know this story, but maybe they were investigating patent infringement? Edison's work drew heavily on the much earlier work of British (and other) inventors and was patented after Joseph Swan's similar patent. Joseph Swan had demonstrated his practical bulb about 10 months before Edison applied for his patent, and was lighting private homes (e.g. Cragside) and theatres at about the same time Edison got going.
Did anyone claim noble motives are sufficient to achieve success?
GP said that some claim it's all a scam, but GP disagrees and has seen noble motives. That's all. Your aggressively arguing against a claim that was never made.
Edit: on Q of ITER. The approximation is reasonable. ITER isn't a power plant. It's being built to study burning plasmas and blanket technologies. It has always been a stepping stone. Is the number not getting large enough fast enough to please the god emperor?
Which issue, exactly? She raises several flags then conflates them all.
The overarching theme is that scientists are misleading the public. There is no shortage of charlatans using the guise of fusion research to spend public goodwill. She is pushing the idea that even the real scientists are doing the same. They are not. Who are these figments "touting Q"? The pop science articles? Did she sit down and actually talk to a single researcher, ever? In the comments I link above I tear apart her primary sources as they are misrepresented.
She seems intent on claiming Qplasma is a bad metric and shouldn't be used. Well she isn't qualified to make that statement, certainly without some scientific basis for the claim. Astonishingly she never mentions triple product or empirical scaling laws, which are the actual metrics that real research machines use. Real machines are not nuclear, so Q is always zero. Does this mean fusion has made zero progress? This highlights her bias in the argument.
Edit: I will half-answer or rephrase my own question so as to make clear, there are no hidden intentions behind it:-)
Why did ITER decide to present Qplasma as the core metric without using the metric, most would consider the critical to decide closeness to commercial viability? Would you agree
with experts that claim that a minimum Q of 100 is the minimum to target for commercial viability?
You'll notice that the goal in question is the power output. Regardless, Qplasma is there because that is the plasma physicist's yardstick. Change it after 50 years because we're closer to ignition? For what purpose?
I think the wording towards the end of the first goal could be better. That feels like headline bait, but it isn't why the goal was set.
>most would consider the critical to decide closeness to commercial viability
The goal of plasma physicists is to straight-line achieve fusion on the grid, not put on a song and dance. The straight-line path has nothing to do with making a number go up, but everything to do with developing plasma models in new regimes, performing materials studies, and just plain old doing it. ITER wasn't conceived as a demonstration power plant.
Problem is plasma physicists are being subsidized on their very meritorious endeavors, at the cost of 21 billion dollars. A paragraph like the one below,
also from the reference you provided, is going to bite them bitterly:
"ITER is designed to produce a ten times return on invested energy: 500 MW of fusion power from 50 MW of input heating power (Q=10). It will be the first of all fusion experiments in history to produce net energy."
As ITER is supposed to start in 2025, and imagining a successful experiment running for 10 years. Adding to this the planning, and procurement process for a possible first device, one with the "real Q" above 1, then its clear it wont be here for at least another 50 years.
It's true there will be a PR reckoning for that wording. I am not under the impression that those that greenlit ITER were deceived about the target performance metrics, which is the implied scandal. The scientists and engineers gave pitches to committees, the committees asked questions, the scientists answered honestly. Not everyone on those committees is clueless about plasma physics, as indicated by their informed questions (see the DoE FES subcommittee transcripts). Is this how we should decide to allocate society's taxed income? I don't know, but I don't think the funding of existentially critical research that requires expensive machines that only produce incremental knowledge should be on the chopping bloc every year.
Regardless of whatever "well actually"s come about, the achievements of ITER should not be sidelined. If ITER is successful, it sets a very important bar for mega projects in the 2020s. It would be a valuable yardstick and a blueprint for a nuclear facility that has enormous subsystems for cooling and tritium processing. Thanks to HTS we might not need to make a machine much larger than ITER to hit the necessary performance metrics, but ITER is being made now, not in another two decades.
I had a conversation with him where I requested he make a Sabine response video in the comments of a previous video. I think he is a good communicator of plasma physics. His response didn't take the exact shape I would have taken, but I think his high level response will age better than my lower level points.
I shared this video on HN when it came out. This kind of content doesn't inspire as many emotions, so it doesn't take off in the same way. That's a problem nobody has figured out how to grapple yet. ;)
You're doing what I take issue with what she was doing: misrepresenting the arguments. There is a kernel to what she has said, however she pours gasoline onto it to expose some sort of villain that doesn't actually exist. It's a great story though.
I discredit her because she was never interested in finding truth but wears the guise of being someone who is. Now more than ever (even late 15th century) is it difficult to discern misinformation. The last thing we need are smart and capable people weaponizing misunderstanding.
The way I understand it, is that with Q>1 in a confined plasma you no longer need external heating and can just let the fusion heat power the plasma. I.e. you "ignite" it and it burns within confinement (self-sustaining). Whereas if Q<1 you need to continuously provide heating, and the efficiency of that heating is probably quite low so the "total" Q appears lower.
Basically, once you have ignition + confinement, Q no longer matters as you no longer have to heat the plasma to sustain it. Correct me if I'm wrong.
> once you have ignition + confinement, Q no longer matters as you no longer have to heat the plasma to sustain it
Assuming immutable containment and continuous fusion. In reality, combinations of heat dissipation, cycling (to permit material replacement and/or cooling) and even powering the computers and magnets subtracts from the theoretical power output.
In indirect-drive inertial confinement, the NIF drops a Hohlraum containing a fuel pellet and then does things to cause it to fuse. That's the burn. The pellet then falls, a new one is released, and the lasers pulse anew. The energy from the first burn may provide power for the second pulse, but in between is a lot of ground to be lost and thus purchase for Q.
Whatever symbols are used and implied meaning they have (or not), as a layperson I found that the article was very clear that they had not reached any total energy gain (they write something like 0.17 MJ out for 1.9 MJ in).
Just a couple hours ago, I've learned about this unfortunate (at best) terminology (as I fell for it too, though at least I knew that ITER wasn't ever supposed to produce electricity), straight from the website of the journalist that stumbled upon the issue (why no link from this blogpost to the original source ?), and was annoyed to see that it didn't seem to having been discussed in ITER discussions here...
Technically the NIF was funded to study nuclear explosions rather than just fusion so nominally so this definition is consistent with their mission. To the extent that one can create fusion events in this way helps all fusion research is also good.
The video should be entirely neglected. So yes, you can define your own variables, and in particular what you call Q_total would be important for a power plant. But nobody claims that current fusion experiments are power plants, and as a matter of fact they are not power plants. The Q_total variable is entirely meaningless when talking about them!
I still don't understand what the game plan is for laser-induced fusion. How do you collect the energy? By contrast: I have a vague idea of how one might do it for magnetic confinement, and I even remember one magnetically confined system where there was direct electrical potential generated and extracted from the system.
As far as I know, this kind of system is most useful for simulating nuclear explosions while remaining compliant with the Nuclear Test Ban Treaty of 1963.
It's a Department of Energy project, and DoE is primarily focused on nuclear weapons. It is undoubtedly used for nuclear weapons research. Discussion of the technology for use in any kind of power plant is just PR.
There is no such game plan. It is a giveaway to military contractors, corporate welfare.
Magnetically confined D-He3 or H-B fusion could, in principle, be viable, because there is no detour through collecting hot neutrons, but is harder to achieve.
It is research for nuclear stockpile maintenance, which depending on how you view nuclear weapons could be a bad thing but is absolutely not a giveaway to the MIC. You'll notice all the main authors of the paper work at national labs which are funded directly by the federal government.
You are aware, are you not, that the national labs in question are operated by military contractors, literally under contract? And that the people nominally working for these labs are actually, literally, employees of those contractors?
You propose a distinction where there is no difference.
In the U.S., nuclear weapons get a couple orders of magnitude more government money than fusion, and nobody blinks an eye. So it seems more likely that weapons work is a backdoor to get some fusion research done.
Simple, heat something up and run a turbine. With D-T (the easiest fuel), the energy output is 80% neutron radiation so a heat cycle is your only option regardless of reactor design.
To avoid a turbine, you need a fuel whose output is mainly fast-moving charged particles.
For anything remotely resembling a practical reactor, the cooling system for the reactor vessel would have a boatload of heat to get rid of. So maybe the very hot coolant from there gets pumped through a steam generator, which then feeds a steam turbine, which drives a generator.
Or, you might leave a (carefully-shaped) hole in one side of your reactor, and call it a nuclear rocket.
There are two ways to do a nuclear rocket. The "clean" way, closed loop. And the "dirty" way. Open loop.
Closed loop is actually not all that different from using a reactor as a generator. But instead of using your cooling system to power a turbine via a head exchanger, you use it to heat a reaction mass such as H2.
Open loop is closer to what you describe. Instead of indirectly heating your reaction mass, you run it directly through the hot reactor. It's simpler and lighter but your exhaust is extremely radioactive. Your reaction mass essentially doubles as the cooling system. Another downside to this approach is you cant run the reactor without making thrust. So lights out if you are coasting I guess.
This is just fusion research. No one plans on using this as a practical way to set up stable fusion. Still you need to know what happens practically and not just on paper.
Absolutely none. The laser itself is absurdly inefficient, and you can't really collect the energy efficiently. It will essentially never be efficient.
It's most realistic application would be for space travel.
There are higher-energy reactions, like p+B11, that generate very few neutrons (it's dubbed "aneutronic" for the main reaction, if you can get there cleanly), and instead yield high energy charged (alpha) particles and photons. These could maybe be directly converted to electrical energy by a number of proposed mechanisms[2].
Highly challenging, but if possible, a very nice energy source, and we're finally out of the steam age!
To the best of my knowledge, there is no design for a laser induced fusion power plant. However, at the current stage all fusion experiments are precisely that, experiments and laser induced fusion is currently the only way to generate a plasma similar to the ones that would exist in a future fusion plant, so this is to study plasma properties, not a prototype for a power plant.
(And of course, that is also similar to the plasmas in a H-bomb, so there is a weapons aspect, but it's not like that we don't already know that H-bombs work.)
68 comments
[ 0.64 ms ] story [ 139 ms ] thread> In this regime, self-heating from α-particle deposition exceeds the external heating input into the DT8; this ratio is denoted Qα, where the self-heating is taken relative to the heating power to the plasma—for inertial fusion this is the PdV compressional work on the fuel and not the total laser energy (P, pressure, dV, volume change). Qα > 1 is a burning plasma.
So as far as I understand such Qα > 1 can be reached without sustaining neither pressure or volume, so one can think about it as an explosion rather that intuitive burning (somewhat continuous process).
To summarize: Q_total = total_energy_out / total_energy_in is what matters, not any Q related to plasma energy. Since to build a power plant we just care what it consumes in comparison with what it produces. For this particular case even though Qα > 1 you likely can extract just a tiny fraction of that energy stored in plasma, while spending huge amounts of energy to heat it in the first place.
1. https://www.youtube.com/watch?v=LJ4W1g-6JiY or as an article http://backreaction.blogspot.com/2021/10/how-close-is-nuclea...
Edit: I'm a big proponent of nuclear fission and fusion since both are the best (current and future) way of getting carbon neutral energy, which can cover the spikes in the grid and sun/wind-less days. It is just important to be careful with the terminology, which can be used to make research results more optimistic.
Any idea what multiple of PdV the input power was? (ie. energy used to power the lasers?) I'm curious what their total Q value would be in this experiment...
I had just started the book when I watched her video, and I counted four separate times where he did mention total energy gain. In one chapter (pg. 142 in my kindle edition), he quotes from a source who lays out the distinction she claims people in fusion research want you to be confused about:
Like First Light Fusion, Tokamak Energy is much more interested in power production than energy breakeven. "Achieving a of one is a scientific goal," Jonathan Carling continues, "but it's nowhere near enough to produce commercial energy, which requires a Q [in the region] of tens." As mentioned previously, Q is the ratio of fusion power out to heating power in. His strong view is that unless other star builders have a credible plan to get to factors of twenty or thirty more power out than they put in, then they're in the science game and not the fusion energy game.
This video by the Improbable Matter channel has a good response to Dr. Hossenfelder's general remarks: https://www.youtube.com/watch?v=KtqC8W0_Ups
The source in the book is basically saying "Q[plasma] needs to be twenty or thirty times higher than it is to get more _power_ out than you put in", but there's already a term for that: Qtotal. Hossenfelder's claim was that the book never mentions Qtotal and continues the misleading technique of using Qplasma as if it's an indicator of power production (and saying it needs to be 20-30x), rather than using a much more clear term Qtotal to describe "total" energy out vs energy in.
The response video you linked is even worse: They "correct" Dr. Hossenfelder by showing papers that spout Qplasma numbers as just "Q" and talk about how ITER achieves a Q=10. But that's not in contradiction at all with Dr. Hossenfelder; her qualm is how misleading it is to talk about "Q > 1" as if it's producing useful energy when you actually mean "Qplasma > 1" (which does not produce useful energy.)
The response video then goes on to talk about how useful Qplasma is, but that doesn't matter to Dr. Hossenfelder's point, which is that it is unethical to go around touting "We're going to get 10x the energy out that we put in" when you're just referring to Qplasma=10. That's all her video said, and the response is attacking a straw man.
Inertial fusion is an explosion. This milestone is about making these (laser induced) explosions energy positive.
The best I can do for now is point back to prior points at the cracks in the story.
https://news.ycombinator.com/item?id=28730707
https://news.ycombinator.com/item?id=28736859
Right. That's why we have such a thriving fusion power industry.
The incandescent light bulb? At one time considered to be a securities fraud! The British parliament assembled a team of investigators to probe Edison's claims.
History is full of inventions that seemed impossible until the dogged inventors made it work. Obviously not all fantastical inventions are possible, but fusion has plenty of evidence that points to its eventual possibility.
I don't know this story, but maybe they were investigating patent infringement? Edison's work drew heavily on the much earlier work of British (and other) inventors and was patented after Joseph Swan's similar patent. Joseph Swan had demonstrated his practical bulb about 10 months before Edison applied for his patent, and was lighting private homes (e.g. Cragside) and theatres at about the same time Edison got going.
GP said that some claim it's all a scam, but GP disagrees and has seen noble motives. That's all. Your aggressively arguing against a claim that was never made.
(At correct time) https://youtu.be/LJ4W1g-6JiY?t=329
Edit: More specifically, would you argue the approximate calculation here is incorrect?
(At correct time) https://youtu.be/LJ4W1g-6JiY?t=521
Which issue, exactly? She raises several flags then conflates them all.
The overarching theme is that scientists are misleading the public. There is no shortage of charlatans using the guise of fusion research to spend public goodwill. She is pushing the idea that even the real scientists are doing the same. They are not. Who are these figments "touting Q"? The pop science articles? Did she sit down and actually talk to a single researcher, ever? In the comments I link above I tear apart her primary sources as they are misrepresented.
She seems intent on claiming Qplasma is a bad metric and shouldn't be used. Well she isn't qualified to make that statement, certainly without some scientific basis for the claim. Astonishingly she never mentions triple product or empirical scaling laws, which are the actual metrics that real research machines use. Real machines are not nuclear, so Q is always zero. Does this mean fusion has made zero progress? This highlights her bias in the argument.
"ITER will generate 10 times more heat than used to start the fusion reaction." https://fusionforenergy.europa.eu/iter/
Edit: I will half-answer or rephrase my own question so as to make clear, there are no hidden intentions behind it:-)
Why did ITER decide to present Qplasma as the core metric without using the metric, most would consider the critical to decide closeness to commercial viability? Would you agree with experts that claim that a minimum Q of 100 is the minimum to target for commercial viability?
Also, that site belongs to some sort of EU bureaucracy, not a research institution.
This one is easy: they did not.
https://www.iter.org/sci/Goals
You'll notice that the goal in question is the power output. Regardless, Qplasma is there because that is the plasma physicist's yardstick. Change it after 50 years because we're closer to ignition? For what purpose?
I think the wording towards the end of the first goal could be better. That feels like headline bait, but it isn't why the goal was set.
>most would consider the critical to decide closeness to commercial viability
The goal of plasma physicists is to straight-line achieve fusion on the grid, not put on a song and dance. The straight-line path has nothing to do with making a number go up, but everything to do with developing plasma models in new regimes, performing materials studies, and just plain old doing it. ITER wasn't conceived as a demonstration power plant.
"ITER is designed to produce a ten times return on invested energy: 500 MW of fusion power from 50 MW of input heating power (Q=10). It will be the first of all fusion experiments in history to produce net energy."
As ITER is supposed to start in 2025, and imagining a successful experiment running for 10 years. Adding to this the planning, and procurement process for a possible first device, one with the "real Q" above 1, then its clear it wont be here for at least another 50 years.
Regardless of whatever "well actually"s come about, the achievements of ITER should not be sidelined. If ITER is successful, it sets a very important bar for mega projects in the 2020s. It would be a valuable yardstick and a blueprint for a nuclear facility that has enormous subsystems for cooling and tritium processing. Thanks to HTS we might not need to make a machine much larger than ITER to hit the necessary performance metrics, but ITER is being made now, not in another two decades.
Watch SPARC for now.
"In defense of "Q-plasma" - a response to Sabine Hossenfelder" https://youtu.be/KtqC8W0_Ups
I shared this video on HN when it came out. This kind of content doesn't inspire as many emotions, so it doesn't take off in the same way. That's a problem nobody has figured out how to grapple yet. ;)
https://news.ycombinator.com/item?id=28964858
All responders who are trying to discredit her are just throwing even more diversions, jargons and ambiguity. Which proves her right even more.
If people wish to prove her wrong, it's simple – just openly and unambiguously discuss the Qtotal numbers.
I discredit her because she was never interested in finding truth but wears the guise of being someone who is. Now more than ever (even late 15th century) is it difficult to discern misinformation. The last thing we need are smart and capable people weaponizing misunderstanding.
Basically, once you have ignition + confinement, Q no longer matters as you no longer have to heat the plasma to sustain it. Correct me if I'm wrong.
Assuming immutable containment and continuous fusion. In reality, combinations of heat dissipation, cycling (to permit material replacement and/or cooling) and even powering the computers and magnets subtracts from the theoretical power output.
In indirect-drive inertial confinement, the NIF drops a Hohlraum containing a fuel pellet and then does things to cause it to fuse. That's the burn. The pellet then falls, a new one is released, and the lasers pulse anew. The energy from the first burn may provide power for the second pulse, but in between is a lot of ground to be lost and thus purchase for Q.
Just a couple hours ago, I've learned about this unfortunate (at best) terminology (as I fell for it too, though at least I knew that ITER wasn't ever supposed to produce electricity), straight from the website of the journalist that stumbled upon the issue (why no link from this blogpost to the original source ?), and was annoyed to see that it didn't seem to having been discussed in ITER discussions here...
http://news.newenergytimes.net/2017/10/06/the-iter-power-amp...
Magnetically confined D-He3 or H-B fusion could, in principle, be viable, because there is no detour through collecting hot neutrons, but is harder to achieve.
You propose a distinction where there is no difference.
To avoid a turbine, you need a fuel whose output is mainly fast-moving charged particles.
Or, you might leave a (carefully-shaped) hole in one side of your reactor, and call it a nuclear rocket.
Just don't point the exhaust at anything inhabited.
Closed loop is actually not all that different from using a reactor as a generator. But instead of using your cooling system to power a turbine via a head exchanger, you use it to heat a reaction mass such as H2.
Open loop is closer to what you describe. Instead of indirectly heating your reaction mass, you run it directly through the hot reactor. It's simpler and lighter but your exhaust is extremely radioactive. Your reaction mass essentially doubles as the cooling system. Another downside to this approach is you cant run the reactor without making thrust. So lights out if you are coasting I guess.
It's most realistic application would be for space travel.
Highly challenging, but if possible, a very nice energy source, and we're finally out of the steam age!
[1] https://en.wikipedia.org/wiki/Aneutronic_fusion#Boron
[2] https://en.wikipedia.org/wiki/Direct_energy_conversion
(And of course, that is also similar to the plasmas in a H-bomb, so there is a weapons aspect, but it's not like that we don't already know that H-bombs work.)
1. https://news.ycombinator.com/item?id=30081034
https://en.wikipedia.org/wiki/National_Ignition_Facility
https://scottlocklin.wordpress.com/2014/03/05/anatomy-of-a-f...