A New Nuclear Technology Discussion Thread

[UserJoe]

I started this thread to discuss nuclear technology outside of the renewables vs. nuclear thread and the Fukushima thread. I would like this thread to stick to actual technology and stay away from economics and advocacy/detraction posts. Power reactors, research reactors, space reactors, and even fusion reactors or other nuclear technology are fair game. I'll start out with a 3 for 1 post.

Canada has been very interested in promoting Samll Modular Reactor (SMR) development. Another SMR has passed the first stage of the Canadian licensing process. This one is the U-Battery and is another 5 MWe gas cooled reactor. Reactors in this size range are known as micro-reactors and nuclear batteries. They generally have passive safety systems and long times between refueling. In some of the concepts you don't refuel. They take the old one away and put in a new one.

Westinghouse has a micro-reactor called the eVinci. It is cooled by sodium heat pipes. Herat pipes are passive heat removal systems with no moving parts driven by evaporation and condensation. Some other places where there are heat pipe cooling systems are satellites, PC cpu and gpu cooling systems, and space reactor designs. The eVinci is a fast reactor with low power density that can be carried on a truck and site installation is claimed to be less than 30 days. The reactor has a 10 year core life. I think this is one that they would take away at the end of the core life and not refuel it on-site.

It lookes like the US DOE is getting serious about building a new fast neutron flux spectrum test reactor. They just filed a notice of intent to prepare an Environmental Impact Statement for the Versatile Test Reactor (VTR). The VTR's capabilities are important for materials and fuel testing of advanced reactor concepts. The US hasn't had a fast flux test reactor since the Fast Flux Test Facility shut down in the early 90's.

 

[slowtech]

I know very little about nuclear power (compared to the posters here), but I check in on MIT's SPARC project from time to time.

It is fusion, not fission, but I don't think you excluded that.

There are a few videos on YouTube (I watched one that was over an hour, which I think I found here, but can't locate at the moment).

 

[UserJoe]

I know very little about nuclear power (compared to the posters here), but I check in on MIT's SPARC project from time to time.

It is fusion, not fission, but I don't think you excluded that.

There are a few videos on YouTube (I watched one that was over an hour, which I think I found here, but can't locate at the moment).

The SPARC project is interesting. MIT's tokamak research has always focused on compact high magnetic field Alcator family of machines. The approach was started by Bruno Coppi who was always proposing designs that pushed magnetic field and stress limits and were very aggressive from an engineering standpoint. SPARC is an extension of that approach. It's a big leap in plasma parameters to jump from Alcator C-mod to a Q of 3 device like SPARC. They also need to take some big technology and engineering leaps to get to the powerful magnets needed to get to the plasma parameters the are targeting. There is also a private company called Tokamak Energy in the UK using the same compact high magnetic field approach using a concept referred to as a spherical tokamak (not to be confused with a spheromak which is an entirely different confinement concept). They started up their latest tokamak last year and hope to demonstrate plasma temperatures high enough for fusion. They hope to build and operate a breakeven machine by 2030. Both projects hope that the compact machine approach will lead to an economical power plant design. They have both attracted a fair amount of private investment.

The competition for these small companies is the ITER project. ITER is a giant and expensive multi-national project that has taken a very conservative approach to virtually guarantee that they will be able to produce significant fusion power (500 MW and Q ~10) in their plasma. There are also some engineering testing objectives in the later stages of the project. The first plasma is scheduled for the second half of the 2020s and the plasmas generating significant fusion power won't happen until the 2030s.

 

[slowtech]

Cool, thanks! From what I gather, the big advance that made the SPARC project possible was the development of strong (superconducting?) magnetic tape, so they could create a strong enough field to contain the plasma.

Does that sound more or less correct to you?

 

[ramases]

The competition for these small companies is the ITER project. ITER is a giant and expensive multi-national project that has taken a very conservative approach to virtually guarantee that they will be able to produce significant fusion power (500 MW and Q ~10) in their plasma. There are also some engineering testing objectives in the later stages of the project. The first plasma is scheduled for the second half of the 2020s and the plasmas generating significant fusion power won't happen until the 2030s.

Part of that conservatism comes from that unless ITER achieves its objectives it will have been the last hurrah of a large Tokamak, so everybody's trying to make sure they don't bungle it through avoidable mistakes.

 

[UserJoe]

The competition for these small companies is the ITER project. ITER is a giant and expensive multi-national project that has taken a very conservative approach to virtually guarantee that they will be able to produce significant fusion power (500 MW and Q ~10) in their plasma. There are also some engineering testing objectives in the later stages of the project. The first plasma is scheduled for the second half of the 2020s and the plasmas generating significant fusion power won't happen until the 2030s.

Part of that conservatism comes from that unless ITER achieves its objectives it will have been the last hurrah of a large Tokamak, so everybody's trying to make sure they don't bungle it through avoidable mistakes.

You are right. ITER is the end of the line for the "this one didn't work, let's build a bigger one" approach in tokamak fusion research. After not getting close to the projected Qs for TFTR and JET they are not taking any chances this time. The design pushed all the design parameters to the upper end of the uncertainty range and that also pushed the size and cost way up.

 

[UserJoe]

Cool, thanks! From what I gather, the big advance that made the SPARC project possible was the development of strong (superconducting?) magnetic tape, so they could create a strong enough field to contain the plasma.

Does that sound more or less correct to you?

Both the SPARK and Tokamak Energy designers are relying on high temperature superconductors to enable their compact tokamak designs. The magnet R&D is the primary focus for SPARC development since it will be a go/ no go point for the SPARC project. It will remain to be seen what the degradation time is for the superconducting magnets with DT fusion neutrons flying around. A compact tokamak would also have higher first wall and diverter heat flux loads.

 

[SpocksBeer]

Small reactors aren't a new thing - subs, warships and ice breakers all use them - can anyone explain why they've not been commercialised for grid-tied energy generation? (or is this a topic you wanted to steer clear of?)

 

[Dan Homerick]

Small reactors aren't a new thing - subs, warships and ice breakers all use them - can anyone explain why they've not been commercialised for grid-tied energy generation? (or is this a topic you wanted to steer clear of?)

Exactly what he wanted to steer clear of. The answer to "why hasn't X been commercialized?" usually comes down to $$$, which is out of scope here ("stay away from economics").

Sometimes you just want to talk about the technology, not the business case. In the Observatory that desire should be respected, IMO.

 

[shread]

Thanks for starting this UserJoe, it's a breath of fresh air, ie, science and engineering-laden air

 

[Quarthinos]

Small reactors aren't a new thing - subs, warships and ice breakers all use them - can anyone explain why they've not been commercialised for grid-tied energy generation? (or is this a topic you wanted to steer clear of?)

Exactly what he wanted to steer clear of. The answer to "why hasn't X been commercialized?" usually comes down to $$$, which is out of scope here ("stay away from economics").

Sometimes you just want to talk about the technology, not the business case. In the Observatory that desire should be respected, IMO.

If I remember, the nuclear icebreakers were (are?) owned by the Soviet/Russian gov't. I know subs and warships are. I think it's more about government secrets than $$$. And the tech that built them certainly isn't available from any public source.

 

[Not_an_IT_guy]

IIRC a small high power reactor like that takes a degree of fuel enrichment that makes it troublesome for civilian use.

 

[continuum]

using the same compact high magnetic field approach using a concept referred to as a spherical tokamak (not to be confused with a spheromak which is an entirely different confinement concept).

Woah, not familiar with spherical tokamak or spheromak...

I saw the Wikipedia links on both, are they any good to start for a newbie? Or are there better guides/introductions?
https://en.wikipedia.org/wiki/Spherical_tokamak

Seems light on how it actually is managed/a spheromak is built?
https://en.wikipedia.org/wiki/Spheromak

 

[UserJoe]

using the same compact high magnetic field approach using a concept referred to as a spherical tokamak (not to be confused with a spheromak which is an entirely different confinement concept).

Woah, not familiar with spherical tokamak or spheromak...

I saw the Wikipedia links on both, are they any good to start for a newbie? Or are there better guides/introductions?
https://en.wikipedia.org/wiki/Spherical_tokamak

Seems light on how it actually is managed/a spheromak is built?
https://en.wikipedia.org/wiki/Spheromak

The Wiki articles are mostly accurate but they really aren't good for an introduction. Here's a better link that describes spheromaks and spherical tokamaks. It has some figures to illustrate what the text is describing. Spheromaks are interesting objects. You can form one by dumping plasma with magnetic helicity (magnetic flux linkage) into a conducting can and they spontaneously form. Here's a video of a 3D MHD simulation showing spheromak formation. Since spheromaks don't have coils through the middle of the magnetic donut they have been looked at as targets for magnetized target fusion in which a magnetized plasma is compressed to get it to ignition conditions. It lies in the regime between inertial confinement fusion and magnetic confinement fusion.

 

[MilleniX]

Does Russia's erstwhile nuclear-thermal cruise missile count as new nuclear technology?

 

[Quarthinos]

Does Russia's erstwhile nuclear-thermal cruise missile count as new nuclear technology?

I don't think it's anything different than the US X-6. Ars' article leads me to believe the Russians at least added a heat exchanger to keep some of the radiation out of the atmosphere, but otherwise it sounds like any other nuclear thermal design.

 

[dio82]

IIRC a small high power reactor like that takes a degree of fuel enrichment that makes it troublesome for civilian use.

++
Yes. They use highly enriched fuel, which does make it very troublesome for civilian use. There is also one hell of a lot more to a nuclear reactor than just the pressure vessel. It is the entire balance of plant that really pushes the costs of small nuclear reactors.

One of the tricks to make the costs of small nukes better is to build highly integrated SMRs, such as NuScale. The Reactor pressure vessel has integrated into itself the entire primary circuit including pressurizer and steam generators. This means that all things nuclear can be assembled in a factory and then shipped as one piece to the site and errected within a matter of a handfull of days. Sure beats traditional stick building with a construction time just for the nuclear island on the order of 3-4 years.

I have heared other (old guard) nuclear experts bemoan this concept as "impossibly difficult for maintenance work", but seriously who cares? But this is also only in relation to existing nuclear power stations. And their problems do not arise from too long refuelling pauses. Their problems are structural and highly integrated SMRs address at least some of them.

 

[UserJoe]

This post isn't about new nuclear technology but it's a new development in existing nuclear power plants in the US. Almost all of the US plants have received license extensions to extend their license to 60 years. Plants have started applying for their second 20 extensions to extend their operating life to 80 years. Here's a story about Peach Bottom applying for their second 20 year license extension. They are the third power plant (5th and 6th reactors) to apply to extend their life to 80 years.

 

[UserJoe]

Here's some recent articles about the use of nuclear technology in space. The first one is about the Kilopower reactor that DOE/Los Alamos National Laboratory is developing for NASA. They think it will be ready to use in 2022 if needed. The reactor output would be in the range of 1 to 10 kWe. It is cooled using heat pipes and uses a Stirling cycle to convert the heat to electricity. It could be used for power on the moon, Mars, or a deep space robotic mission.

Speaking of Mars, they started to fuel the power source that will power the Mars 2020 rover. The power source is a radioisotope thermoelectric generator (RTG) powered by the decay heat from Pu238. The RTG will provide 110 Watts of electricity and the heat can also keep the systems warm enough to function.

 

[Idleness]

Very interested in that Kilopower reactor they are developing. Nuclear power seems like an obvious win for larger installations and human presence on other worlds. Most likely even large stations. Solar + battery + fuel cell still can provide power for different things but we simply need gobs of power to keep folks alive for any decent length of time away from earth. Asking Google about power on the ISS I get this:

The 75 to 90 kilowatts of power needed by the ISS is supplied by this acre of solar panels. Eight miles of wire connects the electrical power system. Altogether, the four sets of arrays are capable of generating 84 to 120 kilowatts of electricity – enough to provide power more than 40 homes on Earth.

Solar + batteries + fuel cells already do a lot for generating power but carrying acres of solar around, with their associated infrastructure requirements, comes with some restraints that the Kilopower reactors might not. There's something to be said for shadow, dust, and ongoing maintenance that a 2 tonne brick (basically what they describe in the article) doesn't have to worry about.

 

[Dmytry]

Nuclear in space is hardly a brick... in addition to the brick you have a sizable panel, much like the solar, except it is black and radiating heat. Smaller in the surface area, assuming the cold side is quite hot, but with coolant channels running through it.

And at that point, in space, you are competing most directly against replacing your brick part with a thin film mirror (concentrated solar heat instead of nuclear heat). And further out from the sun you aren't just competing against r^2 sized solar panels, up to a point you are competing against the same size solar panels plus a thin film mirror concentrating the sunlight on them (not useful at earth orbit because that'd overheat the solar panels, lowering efficiency).

Near Earth you're getting ~1.3 kW/m^2 of heat from the sun, so any kind of low powered nuclear heat source can be replaced with a fairly modest sized mirror.

My understanding is that the current hierarchy would have photovoltaic solar beating concentrated solar and concentrated solar beating nuclear.

Obviously, on planets, situation is different, there's a better use case for nuclear.

 

[Megalodon]

Obviously, on planets, situation is different, there's a better use case for nuclear.

As they appear to be talking about moon/mars I don't think any of your objections apply. A lunar night is two weeks.

Switching to reactors is also very tempting for outer solar system missions as RTG isotopes are a pretty serious bottleneck, whereas a reactor probably just needs a few kilos of HEU.

 

[Dmytry]

Well when it comes to planets we also have a certain habit of going there planting flags and then not going there again (Addressing the "human presence" plans calling for decades of power).

edit: also nuclear is not without huge issues for crewed missions. You can't maintain jack shit anywhere near it. Solar has the advantage that it can be repaired.

edit2: agreed on the RTGs for robotic missions though. I'd rather they use low powered reactors than RTGs, because a reactor core that has never been used (on launch) is far far less toxic than an RTG.

On the other hand even with robots there is still an issue of irradiating your electronics once the reactor is turned on, it is not just a drop in replacement by any means. Picture it being at the end of a much longer stick. I dunno how it works out for a rover where a long stick may be impractical, you'll probably need to carry far more shielding, which eats at the weight advantage.

edit3: also speaking of RTGs, Voyager's RTGs https://en.wikipedia.org/wiki/MHW-RTG were producing 2.4 KW thermal from 4.5kg of Pu-238 .

Aside from environmental concerns with isotope sources in a launch, if you're talking of single digit tens of kilowatts it's not some super easy slam dunk "reactor will be lighter" situation, because most of your weight won't be your heat source. The reason RTG electrical output is so low is largely the "thermoelectric" part.

Of course, there's still a plenty of other reasons from launch safety to having to subsidize Russian nukes any time you want an isotope heat source.

 

[Megalodon]

Well when it comes to planets we also have a certain habit of going there planting flags and then not going there again

You thinking we won't do something seems off topic to the point of reactors being the appropriate power source for it.

edit: also nuclear is not without huge issues for crewed missions. You can't maintain jack shit anywhere near it. Solar has the advantage that it can be repaired.

Not sure I agree. If heat is transported with eg a heat pipe you'd probably be able to do repairs on the eg stirling generator on the cold side of the heat pipe.

edit3: also speaking of RTGs, Voyager's RTGs https://en.wikipedia.org/wiki/MHW-RTG were producing 2.4 KW thermal from 4.5kg of Pu-238.

Total inventory is about 35 kg, of which 14 kg has already been allocated to missions. Total global production is substantially less than 1 kg/year. I'm not sure how much is needed to keep up with decay but it it's potentially below that.

Aside from environmental concerns with isotope sources in a launch, if you're talking of single digit tens of kilowatts it's not some super easy slam dunk "reactor will be lighter" situation, because most of your weight won't be your heat source. The reason RTG electrical output is so low is largely the "thermoelectric" part.

Agreed most of the weight wouldn't be the heat source, I just don't think the isotope supplies are there. Especially Pu-238, which doesn't happen in the normal course of existing fuel cycles. Americium-241 has lower power density but occurs in commercial waste so that's one option, and switching to a stirling generator would increase usable electric power, but I question the availability of needed isotopes.

Of course, there's still a plenty of other reasons from launch safety to having to subsidize Russian nukes any time you want an isotope heat source.

There's apparently a plan to irradiate special Neptunium rods in Canadian reactors to start making Pu-238, but to put this in perspective the hopes for that is to generate 1.5 kg/year by 2025. No one is questioning Pu-238 is a wonderful isotope but the scarcity is likely to be a bottleneck for anything beyond single digit kilowatts on a handful of missions per decade.

 

[.劉煒]

Concentrated solar might be interesting if it could be manufactured offplanet. For equation changing, I mean.

Aside from that, is anyone proposing projection of the https://en.wikipedia.org/wiki/Inertial_ ... ent_fusion type?

 

[Dmytry]

Not sure I agree. If heat is transported with eg a heat pipe you'd probably be able to do repairs on the eg stirling generator on the cold side of the heat pipe.

Maybe... I was going from their picture which has this umbrella radiator right over the reactor and assuming radiator or piping potentially risking need of repair (weight optimized piping, all the seals, etc). Stirling generator would probably be sealed and not repairable, but with 1 moving part maybe it can be reliable given enough testing.

Speaking of Russia and nukes in space, there's bes-5 and topaz/topaz 2, which are in about that class using a rather inefficient thermionic converter. You probably could get ~30kW out of those with a Stirling generator.

edit:

Aside from that, is anyone proposing projection of the https://en.wikipedia.org/wiki/Inertial_ ... ent_fusion type?

I recently went down the rabbit hole of reading Wikipedia about Russian/USSR nukes (because of that failed rocket test) and found out that USSR supposedly made weird small thermonuclear bombs with ~15 kiloton yield of which only 0.3 was from fission (of course claiming an entirely peaceful purpose of nuking gas wells shut and so on). US made something similar to attack tanks, although with not as extremely small fission yield (although USSR's figure could be bullshit). That is something I never knew about, seems related to nuclear propulsion in space (Project Orion style) - the possibility of using, not exactly clean, but much cleaner nukes. If someone could only come up with a way to do this without any fission at all.

 

[Megalodon]

Not sure I agree. If heat is transported with eg a heat pipe you'd probably be able to do repairs on the eg stirling generator on the cold side of the heat pipe.

Maybe... I was going from their picture which has this umbrella radiator right over the reactor and assuming radiator or piping potentially risking need of repair (weight optimized piping, all the seals, etc). Stirling generator would probably be sealed and not repairable, but with 1 moving part maybe it can be reliable given enough testing.

I think you're making assumptions not supported by available information.

The design proposes to remove heat from the core with heat pipes. Think of that as the primary loop. The stirling generator gets its heat from the heat pipe, and it's the stirling generator that rejects heat via the big radiator. That's the secondary loop, totally isolated from the core and sodium coolant. Should be safe to work on. I agree it's probably sealed but you can probably have anything up to an entire spare generator/radiator and swap out the entire thing.

Now you'd expect neutron activation of the sodium in the heat pipe but if you shut the reactor down the sodium condenses and collects at the bottom of the heat pipe so that seems reasonable. They also mention it can be buried to reduce shielding requirements. So you're potentially dealing with just the top of a heat pipe poking up out of the dirt with the generator/radiator being accessible and safe to work on.

 

[slowtech]

This is a nice summary of MIT's SPARC project, the technology that makes it possible, and the challenges involved in making it work.

What sets SPARC apart from ITER, JET, and other previous fusion tokamaks will be its use of a new type of high-temperature superconductor (HTS), yttrium barium copper oxide (YBCO). Current-carrying tapes made from YBCO remain superconducting at considerably higher magnetic fields than is possible with older superconductors. This is valuable because higher magnetic fields improve the thermal insulation of the plasma and thus allow for considerable improvement over the performance of previous tokamaks.

...

The prototype HTS coil will demonstrate the operation of the magnet system and test the integrated cryogenic coolant system used to cool the superconductors to 20 K – the operating temperature required to generate the high magnetic fields in SPARC. Maintaining this temperature in SPARC will be challenging, since the compact nature of SPARC leads to higher power density and thus high heat fluxes generated by the fusion reactions in the core of the machine. Additionally, the properties of the mechanical structure around the conductor in the magnet depend strongly on the temperature of the materials, and therefore this structure must be kept cold in order to withstand the stresses generated in the magnet. The model HTS coil will therefore experimentally validate the heat removal capabilities in HTS magnet systems, and the lessons we learn from it will be critical in moving the SPARC project forward.

In addition to addressing structural and cooling challenges, one of the primary missions of the prototype HTS coil will be to investigate our ability to detect and mitigate a “quench” event, which is a sudden loss of superconductivity due to external events such as reaching critically high temperatures, current, or strain levels. A quench can lead to rapid, localized heating of the superconductor, potentially causing damage if the magnet is improperly designed or operated. Understanding how to detect and prevent quench events is of particular importance for the robustness and reliability of the magnet and fusion system.

 

[Megalodon]

^^^^

Yeah those new superconductors are interesting. They're capable of higher magnetic fields but they can also be cooled by non-helium coolants. Not sure what they're using for SPARC but I believe there's a 30T superconducting magnet that can use liquid neon - which you can get from distillation of ambient air.

 

[Dmytry]

^^^^

Yeah those new superconductors are interesting. They're capable of higher magnetic fields but they can also be cooled by non-helium coolants. Not sure what they're using for SPARC but I believe there's a 30T superconducting magnet that can use liquid neon - which you can get from distillation of ambient air.

That's great long term because neon will always be around, but as far as I could find neon as a gas is more expensive than helium as a liquid, so for the time being it's not really a money saver.

 

[UserJoe]

The linked article is about preparations being made to dismantle the Chernobyl sarcophagus. The article has a picture from inside the Chernobyl New Safe Confinement (NSC) which is the largest moveable structure ever built. The NSC was built to enable dismantling the sarcophagus under controlled conditions and has 2 cranes with a 50 ton capacity.

 

[UserJoe]

The Russian floating nuclear plant is has set sail for its final destination. The reactors are based on Russian icebreaker reactors and can go 3-5 years between refueling.

 

[dmsilev]

^^^^

Yeah those new superconductors are interesting. They're capable of higher magnetic fields but they can also be cooled by non-helium coolants. Not sure what they're using for SPARC but I believe there's a 30T superconducting magnet that can use liquid neon - which you can get from distillation of ambient air.

Are you thinking of this magnet?. That's helium-cooled, because only the innermost core is YBCO. There's a middle layer of Nb3Sn and an outer layer of NbTi; both of these need liquid helium temperatures to become superconducting. It's worse than that; to maximize the field in the magnet, you want to be as cold as possible, and you can get another ~20% of headroom by pumping on your liquid helium bath and cooling it down from 4.3 to around 2 Kelvin. That page doesn't say one way or the other, but I believe that magnet is a 2 K system. Certainly the 45 T hybrid magnet at the same site has its superconducting portion cooled to 2 K (and then there's a big-ass water-cooled copper-alloy resistive magnet inside the SC coil. The whole thing is about 2 stories tall and is a frightening monster).

(really powerful magnets tend to be built this way because while Nb3Sn can go to higher fields than NbTi, it's a much more annoying material to work with. And YBCO is far more difficult than Nb3Sn. So only the core of the magnet, which is where the field is the highest, is made from the most obnoxious material...)

 

[Megalodon]

^^^^

Yeah those new superconductors are interesting. They're capable of higher magnetic fields but they can also be cooled by non-helium coolants. Not sure what they're using for SPARC but I believe there's a 30T superconducting magnet that can use liquid neon - which you can get from distillation of ambient air.

Are you thinking of this magnet?. That's helium-cooled, because only the innermost core is YBCO. There's a middle layer of Nb3Sn and an outer layer of NbTi; both of these need liquid helium temperatures to become superconducting. It's worse than that; to maximize the field in the magnet, you want to be as cold as possible, and you can get another ~20% of headroom by pumping on your liquid helium bath and cooling it down from 4.3 to around 2 Kelvin. That page doesn't say one way or the other, but I believe that magnet is a 2 K system. Certainly the 45 T hybrid magnet at the same site has its superconducting portion cooled to 2 K (and then there's a big-ass water-cooled copper-alloy resistive magnet inside the SC coil. The whole thing is about 2 stories tall and is a frightening monster).

(really powerful magnets tend to be built this way because while Nb3Sn can go to higher fields than NbTi, it's a much more annoying material to work with. And YBCO is far more difficult than Nb3Sn. So only the core of the magnet, which is where the field is the highest, is made from the most obnoxious material...)

Yeah I remembered wrong it seems.

-SPARC uses helium, which is probably good anyway because helium shuns neutrons almost as much as it does chemical bonds
-SPARC's peak magnetic field is 23T, compared to 12T for ITER (both are less in the plasma).
-I'm pretty sure the coils for SPARC are windings of YBCO tape? Whatever that tape is it appears it's easy enough for them to work with that they don't use anything else.
-Also apparently it's theoretically possible to distill helium out of air since it's present at ~5 ppm but the cost would go up about 4 orders of magnitude, seems like the most likely outcome would be party balloons get priced out of the market but fusion reactors can still afford it. That's probably a likely outcome with 1-2 orders of magnitude rather than 4.

I wonder if you'd ever see fusion reactors using liquid hydrogen as a coolant. An active industrial site can tolerate different safety assumptions than a hospital, it's probably acceptable to flare the hydrogen in the event of a quench.

 

[UserJoe]

The most advanced fusion startup is TAE Technologies. They use a confinement concept called a Field Reversed Configuration (FRC) which is a high beta compact toroid. The company used to be called Tri-Alpha Energy which came from the 3 alpha particles that are the reaction product of their targeted p-B fusion. They have fairly recently said that their next generation machine will hit D-T breakeven in less than 5 years. It doesn't require any exotic new technologies to build it. Their current generation machine is already of the scale previously only seen at national laboratories in government sponsored programs. The nice thing about FRCs is that the geometry makes it easier to engineer a commercial reactor. They are targeting a reactor in the 300 MWe range.

 

[Megalodon]

The most advanced fusion startup is TAE Technologies. They use a confinement concept called a Field Reversed Configuration (FRC) which is a high beta compact toroid. The company used to be called Tri-Alpha Energy which came from the 3 alpha particles that are the reaction product of their targeted p-B fusion. They have fairly recently said that their next generation machine will hit D-T breakeven in less than 5 years. It doesn't require any exotic new technologies to build it. Their current generation machine is already of the scale previously only seen at national laboratories in government sponsored programs. The nice thing about FRCs is that the geometry makes it easier to engineer a commercial reactor. They are targeting a reactor in the 300 MWe range.

It's really hard for anyone outside the field and I suspect even anyone inside the field to evaluate the likelihood any of these will work. The track record for previous claims in the field is consistently that various groups have had expectations too high, but on the other hand the physics suggests breakeven is within reach with close to current technology.

It's also really difficult to take it seriously when someone claims their technology will scale to extremely difficult reactions like p-B11 when even D-T takes huge international collaborations. I think I'm actually less likely to take it seriously when someone claims they can do p-B11 like TAE, rather than someone like MIT claiming better tokomak performance with newer magnets. That doesn't mean it's impossible everyone has overlooked favorable confinement strategies, but it's pretty hard to believe.

 

[demultiplexer]

Hi, I'm one of those people down on nuclear technology, looking at this thread and seeing a train wreck. How can you discuss any technology without actually discussing it? Uncritically listing things happening in the industry is just a way to get to become an industry press outlet. Let's fuck up this tranquil thread, shall we?

------
Referring to the linked video in the nextbigfuture article: https://www.youtube.com/watch?v=oHiyHBO-4pk

It's like a TED talk in a lot of ways. Let's scrutinize some stuff to see how they smell.

(...) so the notion that you hear Fusion is another 20 years away 30 years away 50 years away it's not true we're talking commercialization coming in the next five years for this technology and more importantly when you (...)

This is about four minutes in, the first few minutes were taken up with mostly investor talk; talking about the age of the company, the personal-ish relationship between interviewer and the size of the company financially. Obviously, this is an investor talk, not a technical or scientific talk, but this whole spiel is straight out of the investor scam's handbook. We are a company right on the brink of a massive discovery, we have a bunch of money and development time already behind us, you can trust us because our interviewer trusts us, give us more money so we can succeed and you'll get ROI in 5 years. Not saying the guy's peddling a scam per se, but a bad first impression.

Worse impression coming in 3... 2... 1...:

something I'm working on this year - this is a company that's going to really revolutionize the electric drivetrain. Think taking the current batteries and motors and all the components but put smarter software and smarter electronics together derived of the power power supply developments, and that's going to be enhance the drivig range of cars by about 30% give you another 40 percent of power and performance (...)

Just utter bullshit, I'm in this business, this is just nonsense. There is no 30% loss anywhere in the drivetrain of any EV. There are already plenty of ways to get way more than 40% higher performance out of any of the drivetrain components. We're 5 minutes in and I've already given up on this guy.

Now comes the juicy part.

with the goal now to ultimately get to something that we call hydrogen boron fuel that requires three billion degrees (...) so today we're continuously operating the machines that you saw somewhere in the neighborhood of 30 million and we're going to 50 million next year (...) around 100 million degrees is what we're pursuing with the next machine called Copernicus. And then by the late 20s we're gonna be at hydrogen boron

Right. So right now, you're saying, you are barely equalling JET, but somehow you're going to go to p-B temperatures and confinement - even though that literally requires orders of magnitude stronger confinement and higher energies than kilometer-sized particle accelerators - by the end of the 2020s? How?

Then, talking about their software which they reckon is all you need to do this instead of building expensive particle accelerators:

and you can do this with a few hundred people as opposed to tens of thousands of people and you don't need a big balance sheet to do that either

Of course, the software! We were too focused on solving all kinds of physics problems instead of focusing on making a decent version of Flappy Bird with a million-degree plasma interface which we can sell on the side for more profit instead of building those pesky unprofitable power plants! (sorry, you can't get all of this out of the quote I just put here, but this is hyperbolically what he is saying)

Here's a great one just a minute on

every time you fire the machine you are heating the plasma greater than the center of the Sun in this building (...) and the issue they had with the city and permits was the color of what your nitrogen tank (...) the Planning Commission in Lake Forest that didn't like a liquid nitrogen tank

Great tech talk, guys. If I would recount all my stupid planning, insurance, banking and supplier minutiae to my future investors, will they give me $700M as well? Because I have a lot more of those than actual technical explanations of my product!

I'm not picking on this guy for one off remark. He goes on for 5 solid minutes on this.

And... that's it. There is no more information in this video. You can skip everything after about 9 minutes, because it's just stories about him meeting people. He very shortly goes into talking about cold fusion (yes....), their rejection in journals and then their 'scrutinized' approach:

we put a science panel together of some of the brightest minds in the field and they come twice a year and they do basically a Ph.D exam on the company (...) and they write a report to the investors on the board

Not in a journal, no peer review, nothing actually scientific about this. It's just a whitepaper submitted to the board. To be fair they also talk about bringing in outsiders, but nothing that hints at any kind of scientific review. I mean, it says a lot if the guy says bullshit like this:

I'll just say that it's rare that's a practice from academia is actually helpful in the Entrepreneurship world

(facepalm)

(different guy now)

the whole reason even I joined as a graduate student this was that fusion is mainly a game of academic research today, and it's driven by the lowest-hanging physics. (...) this dictates where you go this unfortunately doesn't necessarily connect with an end point of economics and viability in the utility space.

[/quote]

Your scientific conclusions don't align with my personal beliefs, so we choose to reject science and go our own way.

Also, seriously? If you don't like doing academic research or you don't think fundamental research is useful enough, you're in the wrong space. Do engineering. I did that, you can basically freely choose whatever the fuck you think is a cool thing to work on and figure out for yourself if it makes for a good company. There are techcellerators all over engineering universities. Especially in the USA, but likewise just as much in the Netherlands where I studied. No shortage of great business-focused engineering fields.

------------

Let's switch gears and for a moment consider TAE separate from this obvious investor scam talk.

I don't mind giving this company the benefit of the doubt on many of their claims. They have presented at much better venues and so far they have have published a decent number of papers in actual peer-reviewed journals, with most of them also being peer-reviewed at some point. Not even bad journals, they've made Nature and a look down the list shows decent impact factor journals. I'd be much more sedate in my response if this was all they had.

But none of their papers actually talk about the juicy stuff. They expect p-B fusion to yield the correct daughter particles to do proper fusion, but this has been panned on multiple occasions that I know of as the energies involved require either confinement times orders of magnitude less than we can reliably create or the involvement of technologies at scales orders of magnitude higher than we currently have; say petawatt lasers on sub-picosecond timescales. Not impossible physically, but way more than 10 years off. Even D-T fusion at system-level breakeven is deemed wholly impossible with machines of the scale that TAE is proposing, even from basic physics principles. Yeah, the reactor core may be confined in a few dozen foot long machine, but the entire facility has to be huge, complex and massively expensive. You're just building JET at that point, but six times larger. I don't understand why this apparently 'doesnt need new exotic equipment'. It does! If it wouldn't, we'd have it in other projects already.

I've put about 2 hours into looking through the papers and haven't found anything talking about the actual thing they are actually doing - either their D-T tests and p-B11->aaa physics. Maybe it's diffusely spread over all the papers, maybe I haven't looked hard enough, maybe that's their trade secret. Help me out here, link me an actual paper and we can discuss it in more detail.

-----------

But what really irks me is the way this guy just talks bullshit and word salad on his own technology. I understand he's not a nuclear physicist and that others have done a better job representing the company's technology, but at the same time I would fire anybody purporting to shun academia whilst hiding behind the veneer of legitimacy that peer reviewed papers bring. I would be wary of anyone purporting to have conveniently built a bunch of startups in hot industries right now building off of their nuclear fusion work - startups that make no sense on a tech level. The only reason I'm even devoting this time to them is because they haven't used 'AI' and 'Blockchain' yet. Well, they did talk about machine learning... that's a red flag.

I can't promise this is my last post in this topic, but jeez guys, be a bit more critical.

 

[Megalodon]

Right. So right now, you're saying, you are barely equalling JET, but somehow you're going to go to p-B temperatures and confinement - even though that literally requires orders of magnitude stronger confinement and higher energies than kilometer-sized particle accelerators - by the end of the 2020s? How?

I agree skepticism is warranted especially for p-B11 but by far the worst train wreck I've seen in this thread is you getting your energies wrong by 6 orders of magnitude. I've noticed you making this error before and challenged you on it but I don't think I've ever received a satisfactory answer for why you're doing this.

Reaction rate for p-B11 peaks at about 600 KeV. Kilometer sized accelerators reach energies in the high hundreds of GeV. So you're off by about a factor of a million.

Commercially available proton beam therapy medical devices easily get into the hundreds of MeV, hundreds of times that necessary for even p-B11 fusion. While there is rightly skepticism they can reach breakeven just the basic temperature requirement is demonstrably not limited to kilometer sized accelerators.

Your discussion usually demonstrates a high degree of technical knowledge so I find such a glaring recurring error on your part baffling, especially after repeated attempts at correction.

I don't have any real expectation of a response this time, but would you care to elaborate on why you keep doing this?

 

[Dmytry]

TBH this looks like a typical BS startup trope. Startups frequently choose a more difficult goal than necessary, because that allows to do nothing for longer than a less difficult goal does. When you have no intention of attaining a goal, dream big. In this case a fairly extreme variation thereof (my understanding is that p-B has to be at least 10x hotter, regardless of bickering over the exact figure it's massively more difficult than the minimum viable product).

There is a lot of dumb money sloshing around, companies and investors that managed to get a ton of money primarily by getting lucky and the tremendous intrinsic monopolization advantage in tech due to customer lock in. The typical startup exists to take some of that dumb money. It is atypical for a startup to do anything other than to serve that financial function.

 

[Megalodon]

TBH this looks like a typical BS startup trope. Startups frequently choose a more difficult goal than necessary, because that allows to do nothing for longer than a less difficult goal does.

In this case I think they deserve a bit more credit than that as they have a proposal to generate useful energy from the Bremsstrahlung x-rays. That's quite significant.

(my understanding is that p-B has to be at least 10x hotter, regardless of bickering over the exact figure it's massively more difficult than the minimum viable product).

It's about 10x hotter than D+T, yes. I encourage you to check the numbers though because I think a ~6 magnitude error in demultiplexer's idea of the required energy is more than a minor quibble over the exact figure. The temperature he suggests is getting into top quark territory, it is ludicrously beyond what would be needed for any conceivable fusion reaction.

If the concept is to be dismissed out of hand we should do so with actual physics in mind.