I sure hope this pans out, and works in space as well.
34 thoughts on “Lockheed Martin’s Fusion Reactor”
If this thing is intended for D-T, they will need a goodly amount more radiation shielding around it than they show in their patent drawings.
but capable of powering a Nimitz-class aircraft carrier or 80,000 homes, sometime in the next year or so
I think what’s going to be harder is dissipating the heat. They’re apparently speaking of a reactor in the 60-100 MW range (and a Nimitz-carrier fission reactor can achieve peak power of around 95 MW).
In space, you’re already speaking of a bad radiation environment, so you just put the reactor outside the areas that need to be shielded, then the shielding can serve double duty. Cooling of course means heat needs to be radiated into space and that is complicated.
On Earth, if you’re putting it on land, it can merely be buried to greatly reduce the radiation risk. Again, you need to figure out how to cool it. Not a hard problem, but harder than putting the reactor in a hole in the ground.
If you’re putting it in a ship, shielding and cooling are both relatively equivalent. You need to shield the reactor while at the same time, not making it so heavy that the ship can’t function. Cooling is easier because it’s in water and thus, it’s as easy a place as it’s going to get to dump heat. Big problem is going to be corrosion, particularly in salt water, but this is a well-known problem with a lot of solutions out there.
And of course, with the way the design is presented, it’s probably very close to a drop-in replacement for a fission reactor on a US fission-powered ship, meaning it can exploit existing shielding and cooling.
The power density of this thing will suck (maybe 0.5 MW/m^3). This is tiny compared to a fission reactor. In a fission reactor, the heat is transfered through the surface of fuel pins that are perhaps 1 cm in diameter. In a DT fusion reactor, the heat is transfered through the surface of a big fat plasma vessel ~1 m in diameter (or larger).
The low power density and complexity makes it completely unsuitable for use in naval vessels.
Umm, yeah, that’s a game changer. What size rocket would you need to boost one of these things to escape velocity?
Surely a better question would be: what size of rocket could this reactor boost to escape velocity?
A 100MW reactor that can run for a year on 25lbs of fuel hooked up to some kind of electric drive could be an interesting way of getting around the solar system.
“… what size of rocket could this reactor boost to escape velocity?”
I’m VERY skeptical that this thing can launch itself into earth orbit, leave alone escape velocity. Do you have numbers that show otherwise?
My thought is to send a group of them somewhere then build a colony around them.
Yes, I agree. But once you get it in LEO, it might have good enough mass-thrust characteristics.
I think the question was about escaping from orbit, not the earth’s surface.
Yes. I can’t think of any electrical drive that would have enough thrust to put itself into orbit, except perhaps that high-altitude airship someone was talking about a few years back.
Even once you’re in orbit, thrust-to-weight (to mass, really, but it’s a gravicentric culture we live in) very much matters.
It’s very easy to come up with an electric drive that would have you to (say) Mars in a month – if you could just ignore the six months of acceleration and deceleration on either end of that velocity.
It’s the “mouse-fart thrust” problem. To cut Mars transit times to a couple months, you need to accelerate/decelerate in a few weeks, and current electric drives _once you add power supply mass_ won’t do that.
To a first approximation, you need power supplies that’ll produce a kilowatt per kilogram to get around the inner solar system via electric drive usefully faster than current rockets. Current SOTA is roughly a tenth of that. And space reactors in general aren’t an easy answer here, between the additional masses needed for power conversion machinery and heat radiation.
Fusion is a lot more likely to be useful for near-term space propulsion if you can make the containment controllably leaky while injecting a bit of extra reaction mass, so you’re treating it as a very high-energy rocket.
Given this background, the PUFF work certainly does sound promising and, if it can be matured within the next decade, would enable dramatic improvements in our access to space-based resources… once we’ve achieved extremely reliable/frequent access to LEO!
I half-expected an April 1 dateline on the article. March 26, though.
Well, it is the 21st century…
I’ve been following this off and on, and was as skeptical as anyone about a quick route to magnetic confinement fusion (given the billions that have been spent on it for over 65 years). Mirror machines were the original idea, and I was surprised that this is just that. I had thought the earlier concepts used a “baseball stitching” magnet configuration. One thing I’ve never seen in a mirror machine were coils straddling the center coil that had reverse field orientation to the rest of the machine.
The field strengths are not that high: about 4 T is the highest I saw. They must have stumbled on something.
And, yes, space propulsion would be the very first application to which I would put this.
BTW, I think they’ll have trouble getting claims 6, 12, and 18 past the sharp-eyed folks at the USPTO.
This is not an old style mirror machine. It is a closed field line (except maybe the supports for the inside magnets) magnetic well machine.
FWIW I think “polywell” best describes the approach here. See my rant below…
Well closer to it. It’s not a polywell. The current term of the art is FRC design.
Get back to me when it reliably exceeds breakeven.
All of the neutrons produced from the fusion process would make it relatively easy to breed fissile plutonium 239 from fertile uranium 238 and fissile uranium 233 form fertile thorium 232. So this could dramatically increase the amount of fissile material for current commercial nuclear reactors– and for nuclear weapons.
In the case of tritium production this is already the case. As the only known *tangible* side effect of so-called cold fusion research. It has been known since about 1992 that tritium can be produced when high voltage and pulsed currents are passed through wires of palladium infused with deuterium gas. The initial investigation was written up in a paper from Thomas Claytor et al at Los Alamos. First in 1992 and later in 1998. No claim of over unity energy production was ever made. Just evidence of tritium being produced. After those two papers, the obvious implications of the technology seems to have precluded any further publications from Los Alamos on the topic. The experiments conducted by Claytor et al. were good out to 10 sigma according to their claim.
I give up. Maybe entering the …6T will work maybe not.
Search for “claytor palladium wires”, look for result “search for neutrons from deuterated palladium subject to high…” link should work, even if I can’t cut and paste it in word press.
Not sure why you believe there’s anything to a 24-year old paper that would have netted a Nobel Prize had it been correct.
Last I checked no one is handing out Nobel prizes for a cheap(er) way to make tritium. Rather than true fusion maybe it is muon catalyzed? Like I said, the process is NOT over unity and doesn’t appear to produce copious amounts of detectable neutrons.
I’ve seen no refutation to the Claytor paper nor a retraction. Have you?
Making tritium in that experiment would have contradicted the known laws of physics. That’s what it would have gotten the Nobel for, not a “cheaper way to make tritium”. And no, it could not have been muons.
That it has gone nowhere in 24 years should clue you in that the experiment wasn’t actually making tritium.
I would love to have seen a rebuttal paper published to this. The original was fairly detailed about the methods and measurement tools. Items one of the other national labs should have easily come by. I would assume that the error required for these false measurements would have to have been due to contamination. Not unheard of. The amounts of tritium produced were minuscule. I don’t think rebuttal papers are unsound science or a waste of time. Just the opposite. The U of Utah cold fusion claims were rebutted within months. I’m waiting, why 24 years? Where’s the beef?
FWIW & IIRC the wires essentially bursted beyond white hot in the experiment, they did not survive. In other words IIRC they obtained a plasma state in the experiment. This was nothing like LENR nor am I convinced this is anything like new physics. The only novelty here is if for some reason the distinct presence of palladium causes the D + D => T + H to be the preferred reaction. This is what I remember, but it has been 24 years after all… This would be interesting, but Nobel worthy? I suppose if you could prove palladium nuclei in a plasma can affect fusion branching ratios?
I’ve often wondered why neither palladium/silver alloys nor thorium have ever been used in “cold fusion” experiments. It’s long been known that an alloy of 40% silver, 60% palladium has a much higher affinity for hydrogen (all isotopes) than palladium. As for thorium, the affinity for hydrogen is off the charts. A pure palladium sample can hold about 1.9 hydrogen atoms per palladium atom. Thorium can hold 4 hydrogen atoms per thorium. If the theory is that it’s all about holding D2 nuclei close together, then I would seem that thorium is a much better choice.
I’m not going into the speculative area of cold fusion or LENR. I think it is an interesting area of study, but unfortunately tainted. A great place for retired physicists with made careers and nothing left to prove. In the long back and forth on this subject it’s been all over the map. Not just about hydrogen affinity, but also surface characteristics of the metal. That palladium, unlike other elemental metals, provide the proper “nucleation” sites, etc. Thus surface working of the metal makes a difference etc, etc. Statistically the positive signals are barely above the error bars. Some of the best work I’ve seen in this area were by Ed Storms. And even his work did not produce any significant amounts of energy, let alone the missing neutrons. It was a shame that such a renown electro-chemist as Martin Fleischmann got caught up in this. I believe there were honest mistakes made. A cautionary tale about publishing via press release.
They could also be used to drive a reactor fueled with non-fissile radioactive waste from fission plants, producing energy and getting rid of waste at the same time.
My issue with fusion research has always been fundamentally an engineering one not necessarily scientific. I am continually amazed how often proof of concept seems to be ignored in the rush to get an over unity device. Now granted not the case with ITER but why does the SCALE of the device always have to be so large? Even for a prototype? If the scale could be reduced so would the expense of the device. Does anyone seriously think ITER in its current incantation is a viable fusion reactor for over unity production, even if it *can* be shown to work? I’m very, very skeptical. Why not go for a 100 watt demonstrator of the principle? Or even 10 or 1? Back in the early days of atomic fission, it was Glenn Seaborg along with Burris Cunningham and Louis Werner who produced the first one millionth of a gram of plutonium in 1942. And even at that scale it was measurable progress. Yeah, tell me all about the difficulties of plasma physics and the need for large (re. expensive) systems. Or maybe you just need a better experiment? This may be that experiment, but without measurable incremental progress, who knows? It seems borderline pathological science to me. Maybe I’m ignorant of those increments. If so, please inform me. I’ll read just about anything…
What’s worse, fusion reactors have negative economies of scale. They are limited by the power/area the first wall can withstand, so as they are made larger their volumetric power density must decline (square/cube law).
The fusion power/plasma volume of ITER is supposed to be 0.6 MW/m^3 (the power/volume of reactor is much lower). In contrast, the power/volume of a PWR reactor core (just the part with the fuel rods undergoing the nuclear reaction) is 100 MW/m^3. Anything resembling ITER cannot possibly become economically competitive with fission, never mind all the non-nuclear energy sources that are currently cheaper than fission.
This is a generic problem with DT fusion, btw, regardless of the confinement scheme. Higher beta might let the reactor be smaller, but the tritium breeding blanket still had to be thick, so the volumetric power density will not be high enough to compete.
IMO, the only hope for fusion is to go to some sort of advanced fuel that allows energy to be extracted directly rather than going through a thermal cycle. This might allow relatively expensive turbines and generators to be avoided. D+3He is the easiest of these. I recently became aware of an interesting D+3He reactor concept from Princeton based on FRCs (field reversed configurations, similar to Helion and Tri-Alpha), but using some novel tricks to suppress DD reactions so neutrons carry less than 1% of the energy output. They have a notional Pluto orbiter vehicle based on a 1 MW versions of this that could be launched on a Delta IV (or Falcon Heavy) and reach Pluto orbit in four years.
And the obvious advantage that when used in space all that Tritium can sent out the exhaust rather than contained.
The Farnsworth Fusor is another example of a small scale experiment, that unfortunately, didn’t pan out. But it was all said and done and fairly well characterized in about 12 years time….
If this thing is intended for D-T, they will need a goodly amount more radiation shielding around it than they show in their patent drawings.
but capable of powering a Nimitz-class aircraft carrier or 80,000 homes, sometime in the next year or so
I think what’s going to be harder is dissipating the heat. They’re apparently speaking of a reactor in the 60-100 MW range (and a Nimitz-carrier fission reactor can achieve peak power of around 95 MW).
In space, you’re already speaking of a bad radiation environment, so you just put the reactor outside the areas that need to be shielded, then the shielding can serve double duty. Cooling of course means heat needs to be radiated into space and that is complicated.
On Earth, if you’re putting it on land, it can merely be buried to greatly reduce the radiation risk. Again, you need to figure out how to cool it. Not a hard problem, but harder than putting the reactor in a hole in the ground.
If you’re putting it in a ship, shielding and cooling are both relatively equivalent. You need to shield the reactor while at the same time, not making it so heavy that the ship can’t function. Cooling is easier because it’s in water and thus, it’s as easy a place as it’s going to get to dump heat. Big problem is going to be corrosion, particularly in salt water, but this is a well-known problem with a lot of solutions out there.
And of course, with the way the design is presented, it’s probably very close to a drop-in replacement for a fission reactor on a US fission-powered ship, meaning it can exploit existing shielding and cooling.
The power density of this thing will suck (maybe 0.5 MW/m^3). This is tiny compared to a fission reactor. In a fission reactor, the heat is transfered through the surface of fuel pins that are perhaps 1 cm in diameter. In a DT fusion reactor, the heat is transfered through the surface of a big fat plasma vessel ~1 m in diameter (or larger).
The low power density and complexity makes it completely unsuitable for use in naval vessels.
Umm, yeah, that’s a game changer. What size rocket would you need to boost one of these things to escape velocity?
Surely a better question would be: what size of rocket could this reactor boost to escape velocity?
A 100MW reactor that can run for a year on 25lbs of fuel hooked up to some kind of electric drive could be an interesting way of getting around the solar system.
“… what size of rocket could this reactor boost to escape velocity?”
I’m VERY skeptical that this thing can launch itself into earth orbit, leave alone escape velocity. Do you have numbers that show otherwise?
My thought is to send a group of them somewhere then build a colony around them.
Yes, I agree. But once you get it in LEO, it might have good enough mass-thrust characteristics.
I think the question was about escaping from orbit, not the earth’s surface.
Yes. I can’t think of any electrical drive that would have enough thrust to put itself into orbit, except perhaps that high-altitude airship someone was talking about a few years back.
Even once you’re in orbit, thrust-to-weight (to mass, really, but it’s a gravicentric culture we live in) very much matters.
It’s very easy to come up with an electric drive that would have you to (say) Mars in a month – if you could just ignore the six months of acceleration and deceleration on either end of that velocity.
It’s the “mouse-fart thrust” problem. To cut Mars transit times to a couple months, you need to accelerate/decelerate in a few weeks, and current electric drives _once you add power supply mass_ won’t do that.
To a first approximation, you need power supplies that’ll produce a kilowatt per kilogram to get around the inner solar system via electric drive usefully faster than current rockets. Current SOTA is roughly a tenth of that. And space reactors in general aren’t an easy answer here, between the additional masses needed for power conversion machinery and heat radiation.
Fusion is a lot more likely to be useful for near-term space propulsion if you can make the containment controllably leaky while injecting a bit of extra reaction mass, so you’re treating it as a very high-energy rocket.
Something like this has promise…
https://www.nextbigfuture.com/2018/04/pulsed-fission-fusion-puff-propulsion-concept-gets-phase-2-niac-funding.html
I was kicking a similar concept around with others back in the eighties – we called it “micro-Orion” which I still like better as a name…
The key propulsion performance requirements for rapid deep space missions were characterized my Moeckel back in 1972…
https://ntrs.nasa.gov/search.jsp?R=19720024129
…while your “micro-Orion” sounds very similar to ‘Mini-Mag Orion’ that AST were studying almost 20 years ago…
https://en.wikipedia.org/wiki/Mini-Mag_Orion
Given this background, the PUFF work certainly does sound promising and, if it can be matured within the next decade, would enable dramatic improvements in our access to space-based resources… once we’ve achieved extremely reliable/frequent access to LEO!
P.S. I note that the Lockheed-Martin’s fusion reactor may not be quite as ‘compact’ as originally envisaged…
https://www.nextbigfuture.com/2017/05/lockheed-compact-fusion-reactor-design-about-100-times-larger-than-first-plans.html
I half-expected an April 1 dateline on the article. March 26, though.
Well, it is the 21st century…
I’ve been following this off and on, and was as skeptical as anyone about a quick route to magnetic confinement fusion (given the billions that have been spent on it for over 65 years). Mirror machines were the original idea, and I was surprised that this is just that. I had thought the earlier concepts used a “baseball stitching” magnet configuration. One thing I’ve never seen in a mirror machine were coils straddling the center coil that had reverse field orientation to the rest of the machine.
The field strengths are not that high: about 4 T is the highest I saw. They must have stumbled on something.
And, yes, space propulsion would be the very first application to which I would put this.
BTW, I think they’ll have trouble getting claims 6, 12, and 18 past the sharp-eyed folks at the USPTO.
This is not an old style mirror machine. It is a closed field line (except maybe the supports for the inside magnets) magnetic well machine.
FWIW I think “polywell” best describes the approach here. See my rant below…
Well closer to it. It’s not a polywell. The current term of the art is FRC design.
Get back to me when it reliably exceeds breakeven.
All of the neutrons produced from the fusion process would make it relatively easy to breed fissile plutonium 239 from fertile uranium 238 and fissile uranium 233 form fertile thorium 232. So this could dramatically increase the amount of fissile material for current commercial nuclear reactors– and for nuclear weapons.
In the case of tritium production this is already the case. As the only known *tangible* side effect of so-called cold fusion research. It has been known since about 1992 that tritium can be produced when high voltage and pulsed currents are passed through wires of palladium infused with deuterium gas. The initial investigation was written up in a paper from Thomas Claytor et al at Los Alamos. First in 1992 and later in 1998. No claim of over unity energy production was ever made. Just evidence of tritium being produced. After those two papers, the obvious implications of the technology seems to have precluded any further publications from Los Alamos on the topic. The experiments conducted by Claytor et al. were good out to 10 sigma according to their claim.
https://www.researchgate.net/publication/253905890_Search_for_neutrons_from_deuterated_palladium_subject_to_high_electrical_currents
https://www.osti.gov/biblio/10128295
http://adsabs.harvard.edu/abs/1994cofu.conf….6T
http://adsabs.harvard.edu/cgi-bin/nph-data_query?bibcode=1994cofu.conf….6T&db_key=PHY&link_type=ABSTRACT&high=5ac29e304a14664
I give up. Maybe entering the …6T will work maybe not.
Search for “claytor palladium wires”, look for result “search for neutrons from deuterated palladium subject to high…” link should work, even if I can’t cut and paste it in word press.
Not sure why you believe there’s anything to a 24-year old paper that would have netted a Nobel Prize had it been correct.
Last I checked no one is handing out Nobel prizes for a cheap(er) way to make tritium. Rather than true fusion maybe it is muon catalyzed? Like I said, the process is NOT over unity and doesn’t appear to produce copious amounts of detectable neutrons.
I’ve seen no refutation to the Claytor paper nor a retraction. Have you?
Making tritium in that experiment would have contradicted the known laws of physics. That’s what it would have gotten the Nobel for, not a “cheaper way to make tritium”. And no, it could not have been muons.
That it has gone nowhere in 24 years should clue you in that the experiment wasn’t actually making tritium.
I would love to have seen a rebuttal paper published to this. The original was fairly detailed about the methods and measurement tools. Items one of the other national labs should have easily come by. I would assume that the error required for these false measurements would have to have been due to contamination. Not unheard of. The amounts of tritium produced were minuscule. I don’t think rebuttal papers are unsound science or a waste of time. Just the opposite. The U of Utah cold fusion claims were rebutted within months. I’m waiting, why 24 years? Where’s the beef?
FWIW & IIRC the wires essentially bursted beyond white hot in the experiment, they did not survive. In other words IIRC they obtained a plasma state in the experiment. This was nothing like LENR nor am I convinced this is anything like new physics. The only novelty here is if for some reason the distinct presence of palladium causes the D + D => T + H to be the preferred reaction. This is what I remember, but it has been 24 years after all… This would be interesting, but Nobel worthy? I suppose if you could prove palladium nuclei in a plasma can affect fusion branching ratios?
I’ve often wondered why neither palladium/silver alloys nor thorium have ever been used in “cold fusion” experiments. It’s long been known that an alloy of 40% silver, 60% palladium has a much higher affinity for hydrogen (all isotopes) than palladium. As for thorium, the affinity for hydrogen is off the charts. A pure palladium sample can hold about 1.9 hydrogen atoms per palladium atom. Thorium can hold 4 hydrogen atoms per thorium. If the theory is that it’s all about holding D2 nuclei close together, then I would seem that thorium is a much better choice.
I’m not going into the speculative area of cold fusion or LENR. I think it is an interesting area of study, but unfortunately tainted. A great place for retired physicists with made careers and nothing left to prove. In the long back and forth on this subject it’s been all over the map. Not just about hydrogen affinity, but also surface characteristics of the metal. That palladium, unlike other elemental metals, provide the proper “nucleation” sites, etc. Thus surface working of the metal makes a difference etc, etc. Statistically the positive signals are barely above the error bars. Some of the best work I’ve seen in this area were by Ed Storms. And even his work did not produce any significant amounts of energy, let alone the missing neutrons. It was a shame that such a renown electro-chemist as Martin Fleischmann got caught up in this. I believe there were honest mistakes made. A cautionary tale about publishing via press release.
They could also be used to drive a reactor fueled with non-fissile radioactive waste from fission plants, producing energy and getting rid of waste at the same time.
My issue with fusion research has always been fundamentally an engineering one not necessarily scientific. I am continually amazed how often proof of concept seems to be ignored in the rush to get an over unity device. Now granted not the case with ITER but why does the SCALE of the device always have to be so large? Even for a prototype? If the scale could be reduced so would the expense of the device. Does anyone seriously think ITER in its current incantation is a viable fusion reactor for over unity production, even if it *can* be shown to work? I’m very, very skeptical. Why not go for a 100 watt demonstrator of the principle? Or even 10 or 1? Back in the early days of atomic fission, it was Glenn Seaborg along with Burris Cunningham and Louis Werner who produced the first one millionth of a gram of plutonium in 1942. And even at that scale it was measurable progress. Yeah, tell me all about the difficulties of plasma physics and the need for large (re. expensive) systems. Or maybe you just need a better experiment? This may be that experiment, but without measurable incremental progress, who knows? It seems borderline pathological science to me. Maybe I’m ignorant of those increments. If so, please inform me. I’ll read just about anything…
What’s worse, fusion reactors have negative economies of scale. They are limited by the power/area the first wall can withstand, so as they are made larger their volumetric power density must decline (square/cube law).
The fusion power/plasma volume of ITER is supposed to be 0.6 MW/m^3 (the power/volume of reactor is much lower). In contrast, the power/volume of a PWR reactor core (just the part with the fuel rods undergoing the nuclear reaction) is 100 MW/m^3. Anything resembling ITER cannot possibly become economically competitive with fission, never mind all the non-nuclear energy sources that are currently cheaper than fission.
This is a generic problem with DT fusion, btw, regardless of the confinement scheme. Higher beta might let the reactor be smaller, but the tritium breeding blanket still had to be thick, so the volumetric power density will not be high enough to compete.
IMO, the only hope for fusion is to go to some sort of advanced fuel that allows energy to be extracted directly rather than going through a thermal cycle. This might allow relatively expensive turbines and generators to be avoided. D+3He is the easiest of these. I recently became aware of an interesting D+3He reactor concept from Princeton based on FRCs (field reversed configurations, similar to Helion and Tri-Alpha), but using some novel tricks to suppress DD reactions so neutrons carry less than 1% of the energy output. They have a notional Pluto orbiter vehicle based on a 1 MW versions of this that could be launched on a Delta IV (or Falcon Heavy) and reach Pluto orbit in four years.
https://w3.pppl.gov/ppst/docs/thomas2018niac.pdf
Correction: 2 MW
And the obvious advantage that when used in space all that Tritium can sent out the exhaust rather than contained.
The Farnsworth Fusor is another example of a small scale experiment, that unfortunately, didn’t pan out. But it was all said and done and fairly well characterized in about 12 years time….
http://www.farnovision.com/chronicles/fusion/vassilatos.html
I’m using this site for dates, not for its conclusions…