Boiling fluoride thorium reactor

[dgjxqz]

Some polymeric intermetallic pentaflouride have boiling points around 200+°C. Can an LFTR be build like a low pressure BWR while maintaining comparable efficiency under lower operating temperature?

 

[Nik_2213]

Could you link to an example, please, as my google-Fu came up short.

How stable are such compounds to intense irradiation while in contact with fuel rods etc ??

IMHO, one problem with the potentially ionising environment is free-radical production, leading to volatile, reactive monomers and/or longer chain length, potentially intractable 'brown gunge' analogous to tholins.

 

[dgjxqz]

I do not have well characterized sources either. My thought was that intermetallics would revert to their highest possible oxidation state if surrounded by excess fluorine when cracked (given that structural parts of the reactor are less electropositive), but I could be wrong.
Apart from fluorides, other possible options for low boiling point coolant could be phosphorous or enriched ²⁰⁴Hg. Any other suggestions?

 

[anorlunda]

Some polymeric intermetallic pentaflouride have boiling points around 200+°C. Can an LFTR be build like a low pressure BWR while maintaining comparable efficiency under lower operating temperature?

Regardless of the fluid, the limits of Carnot Efficiency still apply. You want higher temperatures, not lower.

 

[dgjxqz]

Well IMHO, focusing on thermal efficiency is more important when fuel cost dominates such as in fossil fuel plants. In nuclear plants, higher temperature often translates to higher corrosion and material cost. Hence I am thinking along the line of lowering startup cost while maintaining acceptable efficiency. But again, I could be wrong ...

 

[anorlunda]

Well IMHO, focusing on thermal efficiency is more important when fuel cost dominates such as in fossil fuel plants. In nuclear plants, higher temperature often translates to higher corrosion and material cost. Hence I am thinking along the line of lowering startup cost while maintaining acceptable efficiency. But again, I could be wrong ...

That's fine, but the dominant efficiency is $/MW. If I spend $5B to build a plant, I want to maximize the MW power I can get out of it.

 

[mfb]

Everything that can produce free fluorine atoms or molecules (e.g. when exposed to ionizing radiation!) sounds like a nightmare to work with.

 

[dgjxqz]

Perhaps non metals can be used as structural materials at low temperature ...

The coolant need not be fluorides, but need to at least have:
- low boiling point
- good neutron economy
- does not irreversibly decompose into something else

Some fluorides here, but can't find phosphides.

 

[anorlunda]

Hence I am thinking along the line of lowering startup cost while maintaining acceptable efficiency. But again, I could be wrong ...

- low boiling point

No. I told you before that thermal efficiency outweighs these gains. We want higher boiling points, not lower.

 

[Nik_2213]

It is one of the sad truths that efficient thermal conversion needs the biggest practicable temperature difference between source and sink. If you're not careful, temperature drops across heat exchangers etc may reduce the available temperature difference and its potential power content to 'not cost effective'.

Thermo-electric 'Peltier' devices do run 'cool-ish', but only because there are no better materials...

 

[nikkkom]

Polymers won't survive in radiation levels typical for power reactors. It was already tried.

 

[dgjxqz]

I would not suggest going any lower in temperature than existing BWRs either. However, water as coolant can be replaced as it requires high pressure to stay in liquid at 200+°C and limits breeding of fertile fuel.

Polymers won't survive in radiation levels typical for power reactors. It was already tried.

Were they testing with organic polymers? How about elemental red phosphorous?

 

[Astronuc]

Were they testing with organic polymers? How about elemental red phosphorous?

Not quite. Biphenyl was considered, e.g., in the Piqua reactor. One problem is polymerization, in addition to cracking or decomposition.

https://en.wikipedia.org/wiki/Organically_moderated_and_cooled_reactor (Note: Article needs additional citations for verification)
https://en.wikipedia.org/wiki/Organic_Rankine_cycle

ORGANIC NUCLEAR REACTORS: AN EVALUATION OF CURRENT DEVELOPMENT PROGRAMS (May 1961)
https://www.osti.gov/servlets/purl/4822394

In designing a nuclear reactor, in addition to the neutronics (neutron physics), one must consider the chemical compatibility of the coolant(s) or working fluids with the structural materials and fuel, and the radiochemical stability, as well as the thermodynamics, fluid hydraulics/dynamics, and economics.

 

[nikkkom]

Ammonia might work?

 

[Nik_2213]

"Ammonia might work?"

Not in such an ionising environment, IMHO. I'd fear breakdown to the energetically favourable N2 & H2...

 

[Rive]

Some polymeric intermetallic pentaflouride have boiling points around 200+°C. Can an LFTR be build like a low pressure BWR while maintaining comparable efficiency under lower operating temperature?

One of the key points of the molten salt (and: liquid metal) reactors is exactly that there is no pressure and no boiling around the fuel. If you allow boiling, then there is no real point in taking all the trouble with replacing water.

 

[snorkack]

One of the key points of the molten salt (and: liquid metal) reactors is exactly that there is no pressure and no boiling around the fuel. If you allow boiling, then there is no real point in taking all the trouble with replacing water.

The point is that water has boiling point of 100 Celsius. Using liquid water as both coolant and working fluid means that you are either limited to low Carnot efficiency, or need a pressure vessel to raise the boiling point.
Mercury boils at 360 Celsius, and Clementine reactor actually was run on mercury. But natural mercury has poor neutron economy. It could be improved with enrichment.
What you want from your working fluid is a low triple point pressure (and thus low freezing point). Phosphorus sucks here - solid to over 500 Celsius. You'd be better off with sulphur.
About those pentafluorides - these are:
NbF₅ - melts at 72...73 Celsius, boils at 236 Celsius. Nb cross-section 1,05 barns.
TaF₅ - melts at 97 Celsius, boils at 230 Celsius. Ta cross-section 20,5 barns.
Mixing them obviously makes neutron economy worse compared to just Nb, but if they do not form solid solutions, would be likely to lower freezing point.

 

[Rive]

Mercury boils at 360 Celsius, and Clementine reactor actually was run on mercury.

That mercury there were never meant to boil. In this context the low boiling point (high vapor pressure) of mercury is actually a serious limit on the maximal temperature of any Clementine-like liquid metal reactor.

NbF5 - melts at 72...73 Celsius, boils at 236 Celsius. Nb cross-section 1,05 barns.
TaF5 - melts at 97 Celsius, boils at 230 Celsius. Ta cross-section 20,5 barns.

The low boiling point either means pressurized primer or limited efficiency.
Sodium, as coolant for fast reactors melts at 371K and boils at 1156K: lead is usually used as alloy so the melting point varies, but 500+K will do - with boiling point far over 1000K, so the pressure can be kept very low, for the price of having a primer with high mass. Both coolant allows you high efficiency with low primer pressure.

 

[snorkack]

Listing the low melting pure metals (under 400 Celsius):
Li melts at 181 Celsius; boils at 1330; cross-section 70 barns natural; 0,045 enriched
Na 98; 883; 0,53 monoisotopic
K 64; 759; 2,1 natural; 1,5 enriched
Rb 39; 688; 0,38 natural; 0,12 enriched
Cs 28; 671; 29 monoisotopic
Cd 321; 767; 2520 natural; 0,075 enriched
Hg -39; 357; 372 natural; 0,43 enriched
Ga 30; 2400; 2,75 natural; 2,2 enriched
In 156; 2072; 194 natural; 12 enriched
Tl 304; 1473; 3,4 natural; 0,10 enriched
Sn 232; 2602; 0,62 natural; 0,11 enriched
Pb 327; 1749; 0,17 "natural", 0,00048 enriched
Bi 271; 1564; 0,034 monoisotopic.

 

[mfb]

Rb 39; 688; 0,38 natural; 0,12 enriched
Cs 28; 671; 29 monoisotopic

Several dollars per gram make that a bit expensive (how much cooling liquid has a typical primary cooling loop?). Oh, and the fact that they will react violently with a large range of chemicals.

In 156; 2072; 194 natural; 12 enriched

~700 tonnes/year global market. 30 tonnes if you need In-113.

Tl 304; 1473; 3,4 natural; 0,10 enriched

~10 tonnes/year but could be increased.

 

[snorkack]

Several dollars per gram make that a bit expensive (how much cooling liquid has a typical primary cooling loop?). Oh, and the fact that they will react violently with a large range of chemicals.

Rb is only slightly more reactive than K, but with better neutron economy. And Na/Rb form a simple eutectic freezing at -4 Celsius.

 

[dgjxqz]

Piqua reactor

Thanks! Good to see that they work to a certain extend. I wonder if fluorocarbons or chlorocarbons can also be used.

NbF₅ - melts at 72...73 Celsius, boils at 236 Celsius. Nb cross-section 1,05 barns.
TaF₅ - melts at 97 Celsius, boils at 230 Celsius. Ta cross-section 20,5 barns.

How about CrF₄+CrF₅?

Sodium, as coolant for fast reactors melts at 371K and boils at 1156K: lead is usually used as alloy so the melting point varies, but 500+K will do - with boiling point far over 1000K

IMHO high temperature designs have been given their deserved attention, low temperature designs are somewhat overlooked.

Perhaps it is also possible for a hybrid design where secondary boiling coolant floats directly on top of the primary liquid metal coolant, eliminating a solid heat exchange.

 

[snorkack]

Thanks! Good to see that they work to a certain extend. I wonder if fluorocarbons or chlorocarbons can also be used.

Of fluorocarbons, CF₄ boils at -128 Celsius. The higher ones are not resilient to radiation damage.

How about CrF₄+CrF₅?

Neither is good. CrF₅ boils at 117 Celsius. CrF₄ freezes at 277 Celsius.

 

[anorlunda]

IMHO high temperature designs have been given their deserved attention, low temperature designs are somewhat overlooked.

But the reason you gave for interest in low pressures violates Ahmdal's Law. If you halve the thermal efficiency of the whole plant in return for reduction in startup costs, the net gain is negative by a huge margin.

Well IMHO, focusing on thermal efficiency is more important when fuel cost dominates such as in fossil fuel plants. In nuclear plants, higher temperature often translates to higher corrosion and material cost. Hence I am thinking along the line of lowering startup cost while maintaining acceptable efficiency. But again, I could be wrong ...

I think the OP's original question has been adequately answered. Thread closed. If you don't like that, click on my user name and start a conversation giving your reason.