Thorium reactors

[mr friendly guy]

I have found some stuff on Thorium reactors. Its obviously promoted by someone who thinks they are the next best thing since slice bread, but I thought it was interesting so I will share it with all the other people who are interested in nuclear.

16 minute video


This is a longer one, but you most probably only need the first five minutes to give you the summary.

[someone_else]

He looks like Kirk Sorensen, the guy that created Flibe Energy. (a company that intends to develop and sell throium reactors)

He is smart enough to not market them only for raw power (something where any other commercial reactor can do the same without research), but with a list of stuff that are byproducts or easy hacks, to offer something normal nuclear plants are unable to provide.

[mr friendly guy]

It is Kirk Sorenson. I am particularly impressed with his thoughts about using it to create liquid fuels by extracting the CO2 from the atmosphere, using the nuclear power plants to power the extraction process. The idea is that nukes provide so much power, we can afford to lose some via thermodynamics as we convert the fuel into one our vehicles can use. I had previously seen this idea in a science magazine.

The question is, does he get enough funding. If no one is interested, I suggest he sells his services to China. I hear they are interested in nuclear and thorium reactors to replace coal. I however haven't heard of them thinking about using reactors to suck out CO2 from the atmosphere to make fuel, since they prefer to use the coal liquefaction method.

[someone_else]

The idea is that nukes provide so much power, we can afford to lose some via thermodynamics as we convert the fuel into one our vehicles can use.

The idea is kinda old per-se, and it's something that makes any sense only with LFTRs. To do so you must have enough nuke plants to provide much more than base load (not necessarily 100% of the maximum load), when none needs that energy (at night for example) you redirect it to fuel production.

Most chemical production procedures do require significant heat that could be taken from the reactor directly, so it may be more efficient than converting heat to electrical to heat to chemical.

Anyway, it's something that makes sense only if making reactors is not a HORRIBLY EXPENSIVE AND COMPLEX process like is now for modern ones.
Theoretically LFTR should be much much much simpler since they don't need to keep water liquid at 400 or more degrees celsius (so you do without extremely heavy pressure vessels and piping).

The question is, does he get enough funding.

He did its best to attract investors by listing other tricks the LFTR has in its sleeve.
But his Master Plan (longish trascript, search for "military" to get to the right part) is developing and selling small-ish power plants to military, which is always looking for a good smallish reactor for true "base islanding".
He says that it is a very hot topic for army from personal experience, dunno if he is right or not.

This allows him to avoid the pain in the ass of making all the preparatory work and tests for a new nuclear reactor design to be accepted by civilian nuclear regulations (requires loads of money and is a show-stopper), and go with military nuclear regulations that are less demanding for a little while, then when he gathered enough money he plans to civilian-rate his stuff as well.

At the moment the plan looks somewhat sound to me (military is not afraid of nuclear like general population, and sounds something none else can really compete in), but the company is pretty new and announced its board of advisors less than a month ago.
I wish them good luck though.

[Rabid]

Haven't watched the video, just read Someone_Else's link :

Just so I understand, these reactor do not produce fission products that can be used to produce nuclear bombs, is that right ? If so, I think the IAEA and Areva are going to be really interested in it (you know, in order to sell atomic plants to brown people without having to fear them building Bombs).
In fact, wasn't India really interested in Thorium reactors ?

The only worry I have with them Thorium reactors is that they seem to produce a lot of fission products (in variety and in mass per unit of energy produced), which mean that, if you don't want to just dump all that shit in the nearest river, you need to have in place an infrastructure to collect them and to transform these "wastes" into "resources". A costly infrastructure that I doubt many countries are able to competently run : to take the example of India (or South Africa), I fear that you'd quickly end seeing dumps of barrels full of heavy metals, some of them radioactive, leaking into the water table because it would be otherwise too costly to process them (these dumps not necessarily in the country having produced the wastes...).
Is there a study somewhere that'd allow us to see if Thorium reactors worsen the Radioactive Waste problem or not ?

[Kryten]

Haven't watched the video, just read Someone_Else's link :
Just so I understand, these reactor do not produce fission products that can be used to produce nuclear bombs, is that right ?

They don't, but the U233 fuel can be used to make nuclear bombs.

[fnord]

Haven't watched the video, just read Someone_Else's link :

Just so I understand, these reactor do not produce fission products that can be used to produce nuclear bombs, is that right ?

Fission products are even less wanted in a device (which is a non-reusable reactor) than they are in a reusable reactor. Fissile material may be of interest, but it has to be annoyingly pure to reliably initiate with a predictable bang.

If so, I think the IAEA and Areva are going to be really interested in it (you know, in order to sell atomic plants to brown people without having to fear them building Bombs).
In fact, wasn't India really interested in Thorium reactors ?

India is interested in solid-fueled reactors, where Th gets fabricated into fairly conventional fuel rods. That's quite a significant difference from a reactor with a liquid fuel. The word "liquid" was repeated at least twice in the first paragraph.

The only worry I have with them Thorium reactors is that they seem to produce a lot of fission products (in variety and in mass per unit of energy produced),

Actually, it's almost the same fission product production per unit energy. FP distribution may differ slightly as U-233 is being fissioned instead of U-235 - a nuclear engineer like the Duchess could probably tell you exactly how it differs.

which mean that, if you don't want to just dump all that shit in the nearest river, you need to have in place an infrastructure to collect them and to transform these "wastes" into "resources".

How big an infrastructure are you thinking of? Massive, seperate, industrial facilities?

FP processing (in a MSR, not a full-blown LFTR) can be as simple as letting the gaseous FP bubble out when the fuel splashes/hits pump bowl/etc and pipe said FP into activated carbon beds and let the vast bulk decay. More large-bench scale (for a GW-class plant) than large-plant scale.

A costly infrastructure that I doubt many countries are able to competently run : to take the example of India (or South Africa), I fear that you'd quickly end seeing dumps of barrels full of heavy metals, some of them radioactive, leaking into the water table because it would be otherwise too costly to process them (these dumps not necessarily in the country having produced the wastes...).

As you can see in the 1960s era MSBR design a 100 MW-class plant is completely contained (core, processing cells, steam turbines, 30-year waste storage pit, etc) in a footprint of 520 feet x 220 feet. Everything but the air-cooling stack fits under 190 ft above grade.

Total annual arisings from the full-blown MSBR processing system appear to be 166-250 cubic feet/yr - up to 6.8 cubic metres/yr. Maybe double that to account for the plans to dump all discarded irradiated reactor/processing components there (core graphite, processing kit, etc) to bring it up to 14 cu m/yr.

By contrast. the Homebush Olympic pool (listed as the Competition Pool) used for the Sydney Olympics has a volume somewhere between 2575 and 3862 cubic metres. Total MSBR arisings would take somewhere between 183 and 275 years to completely fill such a volume.

Among the fission products discarded to the waste cell are boring stuff like xenon (arc lamps, reaction mass for ion thrusters), platinum (catalytic convertors), rare earths (electric motors, strong magnets) that are under the reactor operator's control, not whoever (eg China) has otherwise cornered the market, strontium and cesium (food/sewage irradiation).

I find it a little difficult to believe that none of those will be cheaply recoverable enough from older, cooled, discard salt to not constitute ores (except Sr and Cs - they are valuable because of their radioactivity, and mining them will reduce the overall half-life of the discard due to their 30-year half-lives). And if any of them are worth mining, well, that just slows down the time it takes to fill your storage pit.

It looks like that "costly infrastructure" you seem to be worrying about comes with the reactor itself.

Is there a study somewhere that'd allow us to see if Thorium reactors worsen the Radioactive Waste problem or not ?

What "waste problem"? The one the French have sorted to the point that their entire "spent" fuel arisings from their reactor fleet fit under a double squash court-sized area at Le Havre?

[Rabid]

@ Fnord : thanks for the interesting infos.

When I talk about the infrastructure, I don't mean the physical infrastructure in itself, but its human aspect, which is more easily fallible. At one moment or another, there's got to be humans in the loop, and they are the one that can take decisions as stupid as "What do I care about these wastes ? Just ship them to Somalia and let's be done with it !".

What "waste problem"? The one the French have sorted to the point that their entire "spent" fuel arisings from their reactor fleet fit under a double squash court-sized area at Le Havre?

The "problem" I talk about is two-fold :

- First, the quantity of fission product being produced, which you have already adressed
- And secondly, the degree of radio-activity of these fission products, and their degree of radio- and chemical-toxicity if released in the environment.

(minor nitpick : it isn't "Le Havre", but "La Hague". Totally different cities.)

[Magis]

There is a difference between fission products and activation products. The fission products produced by a thorium reactor will be near identical to those by a U-235/U-238 reactor. However, fission products are not the most important constituents of waste materials because fission products tend to decay away the fastest.

The activation products (as in, higher actinides) form the bulk of the long-lived radioactive nuclides that demand careful storage for many thousands of years. The activation products yielded from thorium fuel cycle reactors will differ substantially to U-235/U-238 fueled reactors, and can offer some benefit with respect to waste management.

However, the primary benefit of thorium reactor fuel cycles is the expansion of nuclear fuel reserves away from solely uranium. They are not generally thought of as being particularly proliferation-resistant, and in fact, it will probably be challenging to prevent them from being more proliferation-prone than traditional uranium reactors. This point depends on the postulated thorium reactor design (of which there are several), but the closed-cycle designs involve a massive fissile inventory of between 2 and 20 tons of weapons-grade material to kick-start the reactor initially, and will subsequently require a large mass of weapons-grade U-233 to keep the cycle going.

Presently, the cost savings of thorium fuel vs. uranium is the only motivator any utility would have for implementing such a fuel cycle.

[fnord]

Rabid - my cockup on the city.

As for "waste" half-lives (given the ore potentials mentioned earlier), due to the complete (since U, Np and Pu fluorination can be repeated as desired, and assuming U-233 fuel, Pu and higher actinide production is cut tenfold per unit fuel consumption) removal of those actinides from the arisings, we are looking at 5/6 of the annual arisings stable within, IIRC, 15 years, and if Sr and Cs are not mined, the entire annual arisings decaying to being less radioactive than the original Th-bearing ore (assuming burning U-233) , with a shorter residual half-life to boot, within 300 years.

Per-pass recoveries of selected element via fluoride volatility: (p560, "Fluorinated Materials For Energy Conversion"):

U - 95-99.5%
Pu - 98-99.5%
Am,Cm - do not form volatile fluorides
Np - 60-70%

As for the social infrastructure, I'm not sure. Wouldn't be a case of ensuring that operator and society interests align ?

Magis - how big a unit are you thinking of to require a 2t (let alone 20t) startup charge? From what Sorensen has said, MSR operate in the thermal spectrum, where fission cross sections are significantly greater than in fast spectra, meaning less fuel needed for the same fission rate, ceteris paribus. Here suggests a specific inventory of 300 kg fissile per GWe, although that looks like it may refer to an isobreeder (CR=1).

Activation products are the result of non-actinide materials undergoing neutron activation, rather than being fission products - although it's quite possible for some nuclides to come from both processes.

Assuming a full up LFTR (conversion ratio >=1, and thorium being bred into U-233), there would logically be no need for fissile addition beyond the start charge to maintain operations, as at least one unit replacement fissile would be generated per such unit consumed. The only thing added to the reactor, feed-wise, would be the fertile thorium and top-up salt to replace processing losses.

So where does your requirement for a "large mass of weapons-grade U-233" come from, especially since isotopes themselves are fissile, fissionable, fertile, or none of the above. It's the isotopic mix of a given element that has been politically defined as whatever-grade (reactor, fuel and weapons all come to mind).

[Rabid]

As for the social infrastructure, I'm not sure. Wouldn't be a case of ensuring that operator and society interests align ?

Well, if the money gained from mining the useful elements offset the cost of the treatment, I don't see why people wouldn't do it ; even more so if they make a benefit in the process. But if it don't cover enough of the costs, there's the very real possibility that corrupt operators would just say "Fuck it, we'll just dump it in the ocean". Even in developed countries (Europe, at least) it is not an unknown practice for the companies tasked with waste management (like Suez Environment) to cut costs by simply shipping their (normal, non radioactive) toxic wastes to India, China or Somalia, where they'll be dumped in the environment.

If the cost of recycling the wastes is bigger than the cost of just dumping them somewhere, for-profit private companies are likely to dump them if they can get away with it.

[Sarevok]

Can Thorium reactors be used for naval propulsion ? If so would it allow the production of "export" SSNs ?

[Magis]

Magis - how big a unit are you thinking of to require a 2t (let alone 20t) startup charge?

According to a report that I've seen recently, a 3 GW thermal power thorium reactor requires on the order of 1.5 to 3 metric tons of U233 (compared to 7-14 tons of fissile Pu239). In addition to thermal power, the fissile inventory will probably depend on end-of-life fuel burnup targets. Another paper I've seen, which was published in ICAAP '06, projects 13 metric tons of Pu239 along with 43 tons of initial thorium for a 1 GW-electric reactor, which is roughly equivalent to 3GW thermal.

From what Sorensen has said, MSR operate in the thermal spectrum, where fission cross sections are significantly greater than in fast spectra

That depends on the isotope in question.

As per your link, the fissile inventory requirements of liquid fuel reactors may differ substantially from solid fuel reactors, but will also depend heavily on end-of-life fuel burnup targets. Also, I don't think Dr. LeBlanc's proposed design has been subject to much rigorous simulation or safety analysis, so it's hard to say how well his projections will hold up.

Activation products are the result of non-actinide materials undergoing neutron activation,

Activation products are the result of any material undergoing activation. The activation products that are of concern w.r.t. waste management are higher actinides, generally arising from the capture of neutrons by U238 (which is an actinide).

Assuming a full up LFTR (conversion ratio >=1, and thorium being bred into U-233), there would logically be no need for fissile addition beyond the start charge to maintain operations, as at least one unit replacement fissile would be generated per such unit consumed. The only thing added to the reactor, feed-wise, would be the fertile thorium and top-up salt to replace processing losses. So where does your requirement for a "large mass of weapons-grade U-233" come from...

For liquid-fuel reactors I can buy the argument that it won't require any operational fissile feed aside from the initial inventory, assuming a C.R. of one or more.

However, for solid fuel reactors (which I think are more realistic an application for reasons I will give down below), this macroscopic analysis is insufficient. There are a host of safety and regulatory conditions that must be satisfied. For example, for a postulated solid-fuel thorium reactor, it's not enough to say that as a system it will achieve an adequate C.R. It also matters where the breeding is taking place in order to satisfy local behavioral requirements such as power density limits and linear element power ratings. That is essentially why most (if not all) solid-fuel thorium designs have the fissile inventory physically separated from the fertile inventory - for example, a fissile core surrounded by a fertile blanket. After sufficient irradiation, the blanket would be reprocessed, and the U233 extracted from the blanket would form the new fissile core, etc. This physical separation of fertile and fissile material leads to a proliferation risk.

Concerning liquid fuel reactors in general - I think they're dangerous. For starters, the reactivity of the system would be dependent in a very complex way on the fuel flowrate. A pump failure and stagnation of flow could therefore lead to a reactivity increase, for example. If we were to look at an accident scenario through Fukushima-tinted glasses, just imagine how much worse Fukushima could have been if, when the coolant leaked out of the reactor, it was also liquid fuel. As mentioned in the OP video, the gaseous fission products would evaporate from the liquid fuel and be released.

In more technical terms, liquid fuel reactors have two fewer barriers to radioactivity release (levels of containment) than solid fuel reactors. A typical PWR has five:
1) Fission products in solid fuel are trapped within the grains of the ceramic UO2. They can only escape if the fuel melts, or if they manage to slowly migrate to a grain boundary.

2) Solid fuels are encased in a sheath made from either zirconium alloys or stainless steel. Fission products that escape through ceramic grain boundaries are trapped by the sheath.

3) Upon sheath failure, fission products are trapped within the primary heat transport system.

4) Upon PHTS failure, the products are trapped within the concrete containment structure.

5) Upon containment structure failure, fission products are physically separated by inhabited areas by the facility exclusion zone.

Liquid fuel reactors, by definition, are missing barriers #1 and #2, and that automatically disqualifies them from receiving regulatory operating licenses in some jurisdictions.

[mr friendly guy]

Can Thorium reactors be used for naval propulsion ? If so would it allow the production of "export" SSNs ?

My understanding is that its technically difficult because of the corrosive nature of thorium salts. In any event, if more civillian nukes use thorium, that frees up uranium for more submarines for your own use.

[fnord]

With great difficulty, I am got hither.

Friendly guy,

The MSRE corrosion problems were solved, IIRC, by modifying the nickel alloy composition (adding titanium and niobium) and salt redox state.

Magis -

I concede the lack of your first two listed FP barriers. Would not the safety case for a liquid-fuelled reactor be helped by the aggressive reduction, if not elimination of, the accident source term?

First reduction is the core pressure - operating at or near atmospheric pressure, compared to 100+ atmospheres, typical of a PWR (and 70-80 odd, typical of a BWR).
Second reduction is the online fuel processing - FP removed during previous running would have a hard time escaping the core due to a severe accident.

I''ll have to sleep on the flow/rate reactivity coupling - I'm just about dead on my feet right now

[someone_else]

Yeah, the fact they aren't so easy to blow up as a conventional reactor is an important safety feature nontheless.

The reduction in pressure also reduces drammatically the price and complexity of the reactor design and building stages. Dig a hole in the ground, make some strong concrete walls and you can place there the reactor, which will also be safe from airplanes and whatever a tornado throws around (something most other reactors need protection against).

If you make the thing around the reactor airtight you can make multiple shells to contain shit.

Can Thorium reactors be used for naval propulsion ? If so would it allow the production of "export" SSNs ?

The first and only molten salt reactors (wich is the closest thing to LFTRs, but using uranium instead of thorium) was designed to power a fucking airplane. And it actually flew a few times while in operation (although it was not powering the airplane, those were just tests)

Based on this I don't see why a naval version can't be done.

If you can make a nuke plant so compact to fly a fucking airplane (which is basically due to the lack of need to keep water liquid at 400+ celsius with overtly heavy steel plates since salts melts at those temps), moving a ship is a joke.

[fnord]

I could see reluctance for a naval (as opposed to merely maritime) version due to the odd tendency naval vessels have of deliberately sailing into harm's way. Namely, how would a MSR fare after taking battle damage?

[Sea Skimmer]

Yeah, the fact they aren't so easy to blow up as a conventional reactor is an important safety feature nontheless.

The reduction in pressure also reduces drammatically the price and complexity of the reactor design and building stages. Dig a hole in the ground, make some strong concrete walls and you can place there the reactor, which will also be safe from airplanes and whatever a tornado throws around (something most other reactors need protection against).

Right, and then the building will constantly face problems from water leakage; building highly complicated structures or really just about anything underground is a bad idea. Nuclear plants number one enemy is already corrosion, building underground will make it much worse unless you are lucky enough to be able to tunnel into dry rock formations.

If you can make a nuke plant so compact to fly a fucking airplane (which is basically due to the lack of need to keep water liquid at 400+ celsius with overtly heavy steel plates since salts melts at those temps), moving a ship is a joke.

And that's exactly why nobody is going to waste so much money and accept such a high technological and safety risk to put one on a ship. Molten salt was attractive for aircraft because it had very high power density, any warship going with nuclear power is going to be big enough that this simply isn't necessary nor worthwhile. Molten salt introduces all kinds of additional operating problems you simply don't face with a water moderated reactor.

[Sky Captain]

Wouldn't there be safety advantage in molten salt reactors with online fuel reprocessing that fission products are constantly removed (and presumambly stored in passively cooled arrangment) making the whole issue with decay heat removal much easier in case of emergency?

Zirconium encased fuel rods - is it really a safety feature? IIRC the whole Fukushima accident started when zirconium reacted with water wapor, produced hydrogen that blew the whole place into huge mess. Also in spent fuel pools zirconium casing caught fire and helped to spread radioactive materials.

[Magis]

Yeah, the fact they aren't so easy to blow up as a conventional reactor is an important safety feature nontheless.

The reduction in pressure also reduces drammatically the price and complexity of the reactor design and building stages.

A low pressure design is not some magic bullet. What happens during an accident that involves a power excursion (which has happened to reactors in the past)? All that liquid will heat up until it vaporizes (which has happened to reactors in the past), and the vaporization will cause an overpressure beyond the design limit and burst the pipe network. Oh, except with this thorium reactor, instead of just mildly radioactive coolant vapor bursting out, it's an incredibly lethal fume of nuclear fuel and tons of fission and activation products. That low operational design pressure isn't going to help anything in a power excursion scenario.

And with liquid fuel reactors, the margin between normal operation and a prompt-critical power excursion is smaller than a solid fuel counterpart. This is inherent in the liquid fuel design in which fuel flows through a reaction chamber, fissions, and then flows to a heat exchanger. Some of the delayed fission neutrons will be emitted after the fuel has left the reaction chamber, and thus they will not contribute to the chain reaction within the critical assembly. This means that the chain reaction will be sustained by a lower effective delayed neutron fraction, and it is that fraction that provides a margin between normal operating conditions and an uncontrollable power excursion.

[Sea Skimmer]

Wouldn't there be safety advantage in molten salt reactors with online fuel reprocessing that fission products are constantly removed (and presumambly stored in passively cooled arrangment) making the whole issue with decay heat removal much easier in case of emergency?

I kind doubt you would gain any safety advantage from having so much more pipework directly attached to the reactor core, increasing the chance of a direct loss of coolant. Most emphasis on the latest reactors has been making them safer by making them simpler. Online fuel reprocessing will be incredibly complicated, and all of that machinery will have to fit inside the containment system to be acceptable, vastly increasing the size and thus cost of said containment.

Zirconium encased fuel rods - is it really a safety feature? IIRC the whole Fukushima accident started when zirconium reacted with water wapor, produced hydrogen that blew the whole place into huge mess. Also in spent fuel pools zirconium casing caught fire and helped to spread radioactive materials.

Its kind of vital to normal operations.The zirconium prevents direct contact between fuel and water, keeping the water from becoming highly contaminated with fission products during operations. Zirconium is used because it can withstand the heat, it doesn't corrode and it doesn't block neutrons. You don't have much other choice if you are going to use a solid nuclear fuel; and liquid nuclear fuels as some molten salt reactors would employ have the rather obvious problem that if you had a crack in the reactor as at Fukushima then all the fuel would drain out! No need to wait for a meltdown, it would just flow out of the core out of hand... great feature. You wouldn't have to wait for a meltdown for it to happen, even if you still had cooling power. Fukushima started when an earthquake massively damaged the internal piping of the plant, then a ocean death wave that killed far more people then the radiation probably ever will wiped out the fuel tanks for the generators. A response to the accident would be better emergency power, and simpler more rugged reactors. Also we've known the Mk1 BWR is a flawed design since 1972... This was compounded by TEPCO simply being incompetent and refusing to vent the poorly designed reactor cores until it was too late, as well as having failed to make certain improvements that were made to US Mk1 BWRs a long time ago. Zirconium did help make hydrogen to cause explosions, but that was way down the list of chain of failures. With improved safety and features like very large water cooling tanks found in newer designs it would have never gotten that bad, or at least several days would have been available to respond.

[Skgoa]

and liquid nuclear fuels as some molten salt reactors would employ have the rather obvious problem that if you had a crack in the reactor as at Fukushima then all the fuel would drain out! No need to wait for a meltdown, it would just flow out of the core out of hand... great feature. You wouldn't have to wait for a meltdown for it to happen, even if you still had cooling power.

And you can design your reactor building to cope with that. Molten salt at a few hundred degrees is something completely different from a core in meltdown melting through the bottom of the containment building. Oh and no steam explosion means no contamination outside the reactor building, but who cares about that?

[someone_else]

Besides, his design features a "kill switch" that if shit hits the fan drains the reactor and dumps the molten stuff in a place where it can cool off without risks. So if there is a crack they can drain everything before the entire reactor escapes. And being liquid they will find out easily that there is a hole somewhere.

A low pressure design is not some magic bullet. What happens during an accident that involves a power excursion (which has happened to reactors in the past)? All that liquid will heat up until it vaporizes (which has happened to reactors in the past), and the vaporization will cause an overpressure beyond the design limit and burst the pipe network.

Given that you start from an atmospheric pressure system, adding safety (which I think should take the form of heavier plumbing) for these kinds of events is much easier than a normal nuke plant where you have to keep everyting under very heavy pressure for it to work in the first place.

Besides, i wonder what's the vapourization temp for the salts used with thorium (and the temps you are talking about for power excursions). I also wonder how much will their volume increase (i.e. the pressure they give) when they get from liquid to gas, not everything expands as much as water.

Most designs I saw around work at 600 or so degrees celsius, but my google-fu is failing me on finding the boiling temp fo the salts used. Wikipedia throws around temps in excess of 1400 degrees celsius for VHTR which is just a pebble reactor using salts as coolant, so not really the same thing.

Oh, except with this thorium reactor, instead of just mildly radioactive coolant vapor bursting out, it's an incredibly lethal fume of nuclear fuel and tons of fission and activation products.

At a ludicrously lower pressure than anything coming out of a normal nuke plant. Which means it is ludicrously easier to contain or keep into the reactor with a bit stronger pipes. With nuke plants you have to vent it in the open because making everything strong enough to hold it would make the plant too expensive.

Once this gas escapes the reactor somehow, is cooled by contact with the fucking surrounding air (not anywhere close 100 degrees, go figure 1000+ degrees like required by the salt gas to fucking stay gaseous) to well below its liquid state temperature, then you have to deal with very stable salts that won't dissolve in water. And this well before it could manage to get close to a wall.

[Magis]

Oh, except with this thorium reactor, instead of just mildly radioactive coolant vapor bursting out, it's an incredibly lethal fume of nuclear fuel and tons of fission and activation products.

At a ludicrously lower pressure than anything coming out of a normal nuke plant. Which means it is ludicrously easier to contain or keep into the reactor with a bit stronger pipes. With nuke plants you have to vent it in the open because making everything strong enough to hold it would make the plant too expensive.

The low-pressure, low(ish)-temperature state of the liquid fuel during normal operation will not necessarily be maintained during any kind of accident. Sustaining its operating temperature requires control over both the fission reaction rate and cooling system. If the cooling system fails (hmm... where has that happened recently?), then the temperature of the liquid fuel will increase, and may increase catastrophically. Even when the cooling systems are working just fine, a power excursion could bring the fuel to thousands of degrees in an instant. Flash-boiling however many tons of fuel is going to breach the piping system - period. Not that boiling is even strictly necessary. Even in a liquid state, increased temperature will lead to a density (and thus pressure) change.

Once this gas escapes the reactor somehow, is cooled by contact with the fucking surrounding air (not anywhere close 100 degrees, go figure 1000+ degrees like required by the salt gas to fucking stay gaseous) to well below its liquid state temperature, then you have to deal with very stable salts that won't dissolve in water. And this well before it could manage to get close to a wall.

You have no grasp of how much cooling is necessary for nuclear fuel if you think that natural convection to air is going to get the job done. Assuming the liquid fuel that escapes into the building is in a non-critical formation, it will still be generating heat at a rate equivalent to 7% of full-power. So, assuming a 3000 MW-th core, you'd be looking at 210 MW of heat generated in the fuel. Good luck trying to keep that temperature down just through contact with air or other building materials. Not to mention that in the meanwhile, fission product gasses will be flooding the entire facility killing every operator/electrician/janitor in its path.

Here's an idea: instead of this batshit crazy liquid fuel design that has substantial safety issues and no immediately obvious benefits, how about the world continue to build solid-fuel reactors that, you know, work.

[Sea Skimmer]

And you can design your reactor building to cope with that. Molten salt at a few hundred degrees is something completely different from a core in meltdown melting through the bottom of the containment building. Oh and no steam explosion means no contamination outside the reactor building, but who cares about that?

Its not going to stay a few hundred degrees if the cooling system fails, and its going to be spewing out radioactive gases if it leaks which will get outside of the building, and paper designs for pipework to dumb the fuel are no guarantee against massive earthquake damage. Of course, since it is that cool during normal operations would also completely suck at making electrical power. But that doesn't matter right? Its not like a plant costing billions of dollars has to be economical or anything hun? You know no reason actually exists why U-235 powered nuclear plants must be hot as they are, they could perfectly well operate with just enough heat to boil water... but that would suck for making steam, and low low pressure steam would demand even more massive turbines and larger numbers of them with prohibitively high operating costs which is why nobody bothers. You can in any case design a reactor building to cope just fine with normal meltdown, as shown at Three Mile Island, while meanwhile Chernobyl and involved reactors designed and built decades ago and massively flawed from the onset.