AHC/WI: Large Commonwealth Civilian Nuclear Program

[Delta Force]

The UKAEA did have power-producing research reactors - the Windscale AGR, both fast reactors, and the SGHWR at Winfrith were all connected to the grid, as well as the plutonium production reactors at Chapelcross and Calder Hall - the latter was hailed as the world's first nuclear power station! Grudgingly, the CEGB even agreed to allow the UKAEA to produce electricity from a commercial-scale (1000-1320 MWe per reactor, probably a twin-reactor plant) fast reactor, but flatly refused to put up any money or provide any land for it.

Wasn't the AGR the first British reactor built on economic merit alone? How did the UK AEA get the CEGB to build nuclear reactors if the Magnox reactors were projected to be more expensive than coal generation?

The UKAEA had responsibility for R&D and the fuel cycle, as well as nuclear weapons, but the power stations themselves were the CEGB's responsibility. Or the SSEB or NSHEB, since Scotland insists on being different. The CEGB was actually a very well run technocracy, and tried to make informed decisions on cost and technical grounds; usually, it got it right. In the case of Dungeness B, the AGR won by a nose over the BWR because of its' theoretical on-load refuelling capability. That turned out never to work properly, and in hindsight the BWR would have been more cost effective, but there was no way to know that at the time. Arguably, it could have been made to work if the technical organisation had been better.

Were there other British nuclear reactor design companies? Vulcain was a small modular reactor/maritime reactor developed as a joint program of Belgium and the United Kingdom, and I think I remember reading about five or six different groups submitting proposals during the design competition.

 

[Delta Force]

Could the AGR have been designed to run on zirconium clad natural uranium fuel and/or be retrofitted with that in the reactors as built? That's what made the reactors have to switch to using enriched fuel and it certainly didn't help the economics.

Given that there is a niche for natural uranium reactors and the only other player in that is Canada with the very expensive CANDU series (heavy water is quite pricey), it probably wouldn't have been too difficult for the British to have competed in that niche. The key issue would be actually having natural uranium fuel capabilities with the AGR.

 

[BlueTrousers]

Could the AGR have been designed to run on zirconium clad natural uranium fuel and/or be retrofitted with that in the reactors as built? That's what made the reactors have to switch to using enriched fuel and it certainly didn't help the economics.

That's a difficult question, and would depend on how zirconium compares to beryllium for neutron capture. It should be fairly good, since it is used in LWRs, but probably isn't as good as beryllium. It's hard to imagine the UKAEA using such a difficult metal to work with unless it had remarkably good nuclear properties.

I'd guess you could probably have designed the AGR to use zircalloy-clad natural uranium, but a retrofit probably isn't possible.

 

[Delta Force]

This book points out why the magnesium-aluminum cladding wouldn't work with a higher temperature reactor. Aluminum starts to lose strength at 200 °C, and magnesium will undergo combustion around 500 to 550 °C. Magnox was already operating at 390 °C before the temperature derate to 360 °C, so 648 °C as with the AGR is unfeasible.

Unalloyed beryllium has issues with oxidation at temperatures above 700 °C and when exposed to water, but it progresses slowly. Under conditions of both high temperature and humidity it will corrode to dust in only a few days. Apparently beryllium-silicon alloys can be used to reduce the corrosion risk though.

Zirconium shouldn't have an issue with carbon dioxide based on the information in the book. However, zirconium does not conduct heat as well as steel. It might be possible to add fins and other surface area increasing features to help with thermal conductivity, as with the Magnox fuel, but that machining made Magnox fuel expensive to produce.

Niobium might be another option due to its high thermal conductivity (even higher than steel), but it has a high neutron absorption rate relative to the other choices and would likely require enriched fuel. The absorption rate isn't as high as steel though. A larger issue with niobium is that it is often found in deposits with tantalum, a material that has been proposed for salted nuclear bombs. If the materials can be extracted that might not be an issue though, as niobium is no where near tantalum or cobalt in neutron absorption.

The neutron absorption data is from here.

 

[Maxwell Edison II]

Why do you want to use CO2 as a coolant? As far as I know, there is no benefit in heat transfer properties over helium, and the neutron absorbing properties of natural helium with its 3-He content shouldn't be that much of a problem.

He will react with nothing, ever. I have no idea why the Brits didn't choose it from the beginning, were they afraid the yanks were going to bust their nuclear balloon?

 

[Some Bloke]

I'd imagine there would be closer cooperation in other areas too.

 

[Delta Force]

That's a difficult question, and would depend on how zirconium compares to beryllium for neutron capture. It should be fairly good, since it is used in LWRs, but probably isn't as good as beryllium. It's hard to imagine the UKAEA using such a difficult metal to work with unless it had remarkably good nuclear properties.

I'd guess you could probably have designed the AGR to use zircalloy-clad natural uranium, but a retrofit probably isn't possible.

Wikipedia has a page for the cost of various materials. I'd take them as rough estimates, but they are interesting. I had to look up the cost of steel elsewhere.

It would appear that beryllium is a very pricey material at $7,000 per kilogram. Zirconium is $1,570 per kilogram and niobium is $180. Magnesium is around $5.80 per kilogram and steel is around $0.50.

Of course, if a helium cooled reactor were chosen, there would be no need for any cladding.

 

[Delta Force]

Why do you want to use CO2 as a coolant? As far as I know, there is no benefit in heat transfer properties over helium, and the neutron absorbing properties of natural helium with its 3-He content shouldn't be that much of a problem.

He will react with nothing, ever. I have no idea why the Brits didn't choose it from the beginning, were they afraid the yanks were going to bust their nuclear balloon?

It's very pricey to extract helium from the atmosphere, so helium is extracted from natural gas reserves. The natural gas fields with the highest concentration are in the United States, and the government controls the American helium supply, so the British or anyone else desiring to purchase it would have to go through them. The cost of helium is around $50 per kilogram, but it would probably cost more to purchase a large amount.

According to Gazprom all natural gas fields contain some helium, so it's really more a question of how much is there. I'm not sure why the British didn't just acquire helium from fields controlled by a British company or in a colony or go to the French or Italians. If they had waited they could even have acquired helium domestically from the North Sea.

Looking into the gasses more, I'm wondering if neon could be an alternative. It has to be produced from the atmosphere and is 50 or so times more expensive than neon per kilogram, but it seems to have higher thermal conductivity than helium. Neon costs $300 per kilogram, which is around half as much as heavy water at $600 per kilogram. Neon is a gas and heavy water is a liquid though, so I'm not sure how comparable kilogram costs are. Neon also has higher neutron absorption than helium, but it's lower than water (at least light water) and has the safety advantages of being a gas.

 

[Dunois]

If you're interested in the UK's nuclear industry then I'd definitely recommend seeing if you can get a copy of C. N. Hill's, new-ish, book on the topic An Atomic Empire: A Technical History of the Rise and Fall of the British Atomic Energy Programme. I've only just started it myself but it certainly seems very good so far and if it's anything like his history of the British rocketry programme A Vertical Empire it'll definitely be worth it. I'll probably come back to the thread in a little bit since I've been considering some possibilities for alternate British nuclear programmes.

The book is good but quite thin where it matters the most. Namely the 1960s and what exactly went wrong with the AGR programme then. A definitive history of the Magnox/AGR programme is yet to be written, though there are enough documents in the national archives for it to happen eventually if someones wants to have a crack at it.

Magnox was designed primarily to produce nuclear materials for military purposes, with electricity being a useful byproduct of the process. Economics, efficiency, and safety levels were rather lacking, especially since Magnox is generation one 1950s technology. Generation two reactors with 1960s and 1970s technology are much better all around, with the Advanced Gas Cooled Reactor being one, and also a development on Magnox.

This is not true! All Magnox stations from Bradwell and Berkeley onwards were designed with electricity generation in mind.
Interestingly the load factor of some Magnox stations especially Hunterston A has consistently been higher than of some AGR stations. What really did not help the Magnox programme was the discovery that steel bolts correded at 300°C under the CO2 atmosphere of the cooling circuit. This reduced station efficiency by 20%

Get rid of this problem and incorporate replaceable once-through pod-boilers and you can actually end-up with a pretty good and economic Magnox design.

France could have been an interesting country to partner with too, especially around/after 1956. Magnox and the UNGG were very similar and essentially convergently evolved designs. Apparently the British even referred to the French reactors as Magnox type in documents.

Not quite in some respects. The UNGGs are markedly inferior to the magnox in the following respects:
-Safety (two partial meltdowns)
-Thermal efficiency (Bugey did 26% against 34% for Dungeness A)
-Economics (the UNGG did not last more than 20 years, the Magnox lasted 40!)

Could the AGR have been designed to run on zirconium clad natural uranium fuel and/or be retrofitted with that in the reactors as built? That's what made the reactors have to switch to using enriched fuel and it certainly didn't help the economics.

Given that there is a niche for natural uranium reactors and the only other player in that is Canada with the very expensive CANDU series (heavy water is quite pricey), it probably wouldn't have been too difficult for the British to have competed in that niche. The key issue would be actually having natural uranium fuel capabilities with the AGR.

The key target of the AGR was to produce steam of the same quality and temperature as coal fired power station. Enrichment and stainless steel canning are imperatives in order to make this happen.

FYI Beryllium was initially selected as canning material. The huge issues involved in machining it put paid to that idea. It was abandoned in 1962, but issues crept up as early as 1959.

Here is my take on how to create a successful UK nuclear programme based on gas cooled reactors:
1. Create a General Atomic Company headed by Vickers, GEC etc to act as single contractor and seller

2. Improve the Magnox design stepwise to incorporate features such as replaceable boilers, better thermal efficiency and simpler operation.

3. Use OTL Hartlepool/Heysham 1 design for the overall AGR design with the following tweaks:
-More fuel channels (18 pins elements) OR no-online refueling
-Fully replaceable pod boilers
-Simpler systems
-Once-through core gas flow

4. From the 1970s work on an AGR Mk II with the following characteristics:
-1200MWe output
-700°C T2 temperature
-45% thermal efficiency via a supercritical steam cycle
-Fully replaceable graphite core (this means offline refueling)
-Further simplification of plant systems and plant operation
-He cooling as an option

5. Eventually you'll hit a wall with gas cooling which means one thing namely:
-Salt cooled reactors with AGR like fuel elements (the UK studied MSRs too!)
-Gas turbine cycles becomes possible as tag-along to this.

 

[Maxwell Edison II]

It's very pricey to extract helium from the atmosphere, so helium is extracted from natural gas reserves. The natural gas fields with the highest concentration are in the United States, and the government controls the American helium supply, so the British or anyone else desiring to purchase it would have to go through them. The cost of helium is around $50 per kilogram, but it would probably cost more to purchase a large amount.

According to Gazprom all natural gas fields contain some helium, so it's really more a question of how much is there. I'm not sure why the British didn't just acquire helium from fields controlled by a British company or in a colony or go to the French or Italians. If they had waited they could even have acquired helium domestically from the North Sea.

Looking into the gasses more, I'm wondering if neon could be an alternative. It has to be produced from the atmosphere and is 50 or so times more expensive than neon per kilogram, but it seems to have higher thermal conductivity than helium. Neon costs $300 per kilogram, which is around half as much as heavy water at $600 per kilogram. Neon is a gas and heavy water is a liquid though, so I'm not sure how comparable kilogram costs are. Neon also has higher neutron absorption than helium, but it's lower than water (at least light water) and has the safety advantages of being a gas.

The idea of Ne-cooling actually sounds quite interesting. CO2 should have a thermal neutron absorption cross section of 0,00388 barn, while nat. He has 0,00747 and nat. Ne 0,039. If we divide that by specific heat capacity, we get basically something like "neutron extinction per unit of heat removed", we get:

nat. He: 1.438*10^-6
nat Ne 3.786*10^-5
CO2: 4,625*10^-6

I don't know if that is too much, but it is higher than the others.

I've also wondered about CD4, fully deuterated methane, as a gaseous coolant, but it is hard to get data on it. Chemically it should be significantly less aggressive towards Zr, Al, Mg and so on than CO2, but it could perhaps become involved in some kind of carbon-forming devilry at higher temperatures, apart from the obvious cost issues.

Concerning CO2, He and the UK I found an interesting document.

Helium has good heat-transfer properties and good chemical properties, but wa
s not domestically
available in
the UK in the quantities necessary.

 

[Delta Force]

This is not true! All Magnox stations from Bradwell and Berkeley onwards were designed with electricity generation in mind.
Interestingly the load factor of some Magnox stations especially Hunterston A has consistently been higher than of some AGR stations. What really did not help the Magnox programme was the discovery that steel bolts correded at 300°C under the CO2 atmosphere of the cooling circuit. This reduced station efficiency by 20%

Get rid of this problem and incorporate replaceable once-through pod-boilers and you can actually end-up with a pretty good and economic Magnox design.

Magnox was arguably more advanced than any of the water cooled designs of the 1950s, but it and UNGG were built with certain concepts in mind. The fuel type probably wouldn't have been popular in the United States or other markets with private electricity firms because purchasing Magnox would mean purchasing reprocessing too. Reprocessing would mean having plutonium laying around too. It's not as much an issue for a national system that knows what it wants and might be considering a nuclear weapons program, but it's something an investor would be hesitant to do.

If you want the least expensive cladding that can support natural uranium designs, magnesium is the way to go though. If not for the government enrichment program in the United States, there might have been more competition between natural uranium (possibly with enrichment if using Magnox technology) and building an enrichment facility.

As mentioned though, carbon dioxide was perhaps not the best choice of gas. If helium or another inert gas had been used the fuel wouldn't need any cladding at all.

Not quite in some respects. The UNGGs are markedly inferior to the magnox in the following respects:
-Safety (two partial meltdowns)
-Thermal efficiency (Bugey did 26% against 34% for Dungeness A)
-Economics (the UNGG did not last more than 20 years, the Magnox lasted 40!)

Was UNGG less safe, or did it just happen to have suffered meltdowns while Magnox did not? Meltdowns weren't uncommon in the early days of nuclear energy.

As for thermal efficiency, I thought Magnox was lower than that. Did it improve over time? Building on that, France has a huge nuclear energy program and actually something of a nuclear energy glut. Seeing as some French nuclear power stations are closed over the weekend for lack of export potential or domestic demand, it makes sense that older UNGG reactors that are likely less economical and safe would be the first to be retired in a glut.

The key target of the AGR was to produce steam of the same quality and temperature as coal fired power station. Enrichment and stainless steel canning are imperatives in order to make this happen.

Did the AGR design call for producing steam akin to that produced by a fossil fuel plant in order to use standard generators, or was there some other reason?

FYI Beryllium was initially selected as canning material. The huge issues involved in machining it put paid to that idea. It was abandoned in 1962, but issues crept up as early as 1959.

Why wasn't zirconium considered? It's standard for water cooled reactors and it is less expensive than beryllium.

Here is my take on how to create a successful UK nuclear programme based on gas cooled reactors:
1. Create a General Atomic Company headed by Vickers, GEC etc to act as single contractor and seller

2. Improve the Magnox design stepwise to incorporate features such as replaceable boilers, better thermal efficiency and simpler operation.

3. Use OTL Hartlepool/Heysham 1 design for the overall AGR design with the following tweaks:
-More fuel channels (18 pins elements) OR no-online refueling
-Fully replaceable pod boilers
-Simpler systems
-Once-through core gas flow

4. From the 1970s work on an AGR Mk II with the following characteristics:
-1200MWe output
-700°C T2 temperature
-45% thermal efficiency via a supercritical steam cycle
-Fully replaceable graphite core (this means offline refueling)
-Further simplification of plant systems and plant operation
-He cooling as an option

5. Eventually you'll hit a wall with gas cooling which means one thing namely:
-Salt cooled reactors with AGR like fuel elements (the UK studied MSRs too!)
-Gas turbine cycles becomes possible as tag-along to this.

Forming a British General Atomics and creating a Super Magnox interim design are definitely a good ideas. The lack of clear design progression and economies of scale really stand with the British program given that it could have gone that direction and it would have been similar to initiatives the British government was carrying out in other areas of the economy, such as aerospace.

 

[Delta Force]

The idea of Ne-cooling actually sounds quite interesting. CO2 should have a thermal neutron absorption cross section of 0,00388 barn, while nat. He has 0,00747 and nat. Ne 0,039. If we divide that by specific heat capacity, we get basically something like "neutron extinction per unit of heat removed", we get:

nat. He: 1.438*10^-6
nat Ne 3.786*10^-5
CO2: 4,625*10^-6

I don't know if that is too much, but it is higher than the others.

The thing about observations like that with neon is that there has to be some reason why the concept doesn't appear in the literature. Despite neon's apparent potential, I've never heard it mentioned before as a potential coolant.

It's the same thing with the other noble gases or nitrogen. There's probably some reason why the focus is only on carbon dioxide and helium. Admittedly, it might be because those are the only two that have been tested. Not every theoretically interesting nuclear power concept has been tested.

I've also wondered about CD4, fully deuterated methane, as a gaseous coolant, but it is hard to get data on it. Chemically it should be significantly less aggressive towards Zr, Al, Mg and so on than CO2, but it could perhaps become involved in some kind of carbon-forming devilry at higher temperatures, apart from the obvious cost issues.

Concerning CO2, He and the UK I found an interesting document.

I've thought of organic gas cooling too.

Apart from the obvious issues with fire and explosion risk, it might have the same issues that doomed organic liquid cooling, namely the buildup of carbon deposits throughout the reactor. Liquid hydrocarbons were very cheap in the 1950s and 1960s and are less prone to fire and explosion than gaseous hydrocarbons, so it seems there would have been more effort exerted on trying to solve the issues with liquids if they had potential.

As for why gaseous hydrocarbons might not have been trialed, apart from the fire and explosion risk it was an expensive and difficult to transport resource in the 1950s and 1960s. Some areas of the United States were still producing coal gas into the 1960s, and it might have continued longer in more resource scarce areas such as Europe and Asia. The United States was big on liquid cooling, and the Europeans would have to import the natural gas from the United States or elsewhere, likely at high cost. At that point why not try for the more obvious advantages of helium or try extracting gas from the atmosphere?

 

[Maxwell Edison II]

Damn!

CD4 would be very likely undergo radiolysis, forming solid carbon. Radiolysis of water occur in PWR/BWR systems. CO2 after what I can find apparently undergoes very little radiolysis, and then only to CO, O2, O3 (an equilibrium reaction since these are reactive).

D2 gas would have excellent neutronic properties and good specific heat capacity, but it'd both be chemically quite aggressive and hydrogen tend to make a lot of metals brittle. Materials resistant to high-temp H2 was investigated as a part of project Rover and later NERVA, they used ZrC as the cladding and UC as the fuel.

 

[BlueTrousers]

Materials resistant to high-temp H2 was investigated as a part of project Rover and later NERVA, they used ZrC as the cladding and UC as the fuel.

And of course once you've run a NERVA-type reactor on the edge of meltdown for a few tens of hours, you discard it. Long life and decommissioning aren't really requirements for rocket engines, even nuclear ones...

 

[Dunois]

As mentioned though, carbon dioxide was perhaps not the best choice of gas. If helium or another inert gas had been used the fuel wouldn't need any cladding at all.

An He cooled "Magnox" could indeed be an interesting design for a natural uranium fueled reactor producing process steam at c400°C.
Interestingly Magnox reactors where considered for oil tar sand recovery during the 1980s.

Was UNGG less safe, or did it just happen to have suffered meltdowns while Magnox did not? Meltdowns weren't uncommon in the early days of nuclear energy

As for thermal efficiency, I thought Magnox was lower than that. Did it improve over time? Building on that, France has a huge nuclear energy program and actually something of a nuclear energy glut. Seeing as some French nuclear power stations are closed over the weekend for lack of export potential or domestic demand, it makes sense that older UNGG reactors that are likely less economical and safe would be the first to be retired in a glut.

The UNGG design was quite different from the Magnox design in some respects. Though finding detailled information on it is rather difficult. The fact that accidents happened twice on identital plants does tell though that the design was far from optimal.
The thermal efficiency of the later Magnoxes was indeed around 34% before de-rating.

Did the AGR design call for producing steam akin to that produced by a fossil fuel plant in order to use standard generators, or was there some other reason?

It did! The turbines are fully interchangeable with other 660MWe turbines from fossil fuel plants. Hunterston B uses a turbine from the late Inverkip station (oil-fired) for example.
Interestingly, the 1200MWe AGR designed sketched in the 1960s uses two 600MW turbines in a compound arrangement.

Why wasn't zirconium considered? It's standard for water cooled reactors and it is less expensive than beryllium.

Beryllium is a moderator thence why it was considered.
I suspect the reason zircalloy was not considered may have to do with resistance to CO2 oxidation.

Forming a British General Atomics and creating a Super Magnox interim design are definitely a good ideas. The lack of clear design progression and economies of scale really stand with the British program given that it could have gone that direction and it would have been similar to initiatives the British government was carrying out in other areas of the economy, such as aerospace.

The weird thing is that a supermagnox is briefly mentionned in some documents in the national archives. But information on what this beast was is very scarce.

It is worth remembering that as late as the Dungeness B tender in 1965 the basic AGR design was undefined in some levels. The tender documents make interesting reading in that respect. Things not defined include:
-Gas circulator driving (steam or electric)
-Gas flow arrangements within the core
-Reheat arrangements (gas/steam or steam/steam)
-Fuel element design
-Off-load or on-load refueling

The thing about observations like that with neon is that there has to be some reason why the concept doesn't appear in the literature. Despite neon's apparent potential, I've never heard it mentioned before as a potential coolant.

It's the same thing with the other noble gases or nitrogen. There's probably some reason why the focus is only on carbon dioxide and helium. Admittedly, it might be because those are the only two that have been tested. Not every theoretically interesting nuclear power concept has been tested.

Nitrogen is such a good neutron absorber that the secondary safety system of AGRs is actually a massive high pressure injection of nitrogen gas into the core.

As for Neon this tibit is interesting:
"However, on investigating the pumping power parameters suggested by Lyall [1], it becomes apparent that neon is not a very good coolant. Pure neon requires about eight times as much pumping power as pure helium for a slightly smaller flow area to give a constant heat transfer performance."

Too much pumping power is required basically.

CD4 would be very likely undergo radiolysis, forming solid carbon. Radiolysis of water occur in PWR/BWR systems. CO2 after what I can find apparently undergoes very little radiolysis, and then only to CO, O2, O3 (an equilibrium reaction since these are reactive).

Let's not even go there. CH4 is already present in the circuit of AGR in order to limit the reactions between graphite and CO2. Carbon deposition on all surfaces is the result of CH4 necessary presence in the circuit.

 

[BlueTrousers]

Interestingly, the 1200MWe AGR designed sketched in the 1960s uses two 600MW turbines in a compound arrangement.

1200 MW(e) from presumably about a 3000 MW(t) reactor? Sounds promising to me. I suspect that this would probably be the same turbomachinery and electrical plant envisaged for the CFR at one point.

I rather suspect that, had the AGR line been developed, a Mark III AGR in the 1990s would have used 900 MW(e) generators common with those used on the proposed coal fired power plants of that era. Helium coolant seems likely - I think I read somewhere recently that had it been available in bulk at the time the original AGR was designed, it would have been preferred to CO2.

 

[Delta Force]

Damn!

CD4 would be very likely undergo radiolysis, forming solid carbon. Radiolysis of water occur in PWR/BWR systems. CO2 after what I can find apparently undergoes very little radiolysis, and then only to CO, O2, O3 (an equilibrium reaction since these are reactive).

D2 gas would have excellent neutronic properties and good specific heat capacity, but it'd both be chemically quite aggressive and hydrogen tend to make a lot of metals brittle. Materials resistant to high-temp H2 was investigated as a part of project Rover and later NERVA, they used ZrC as the cladding and UC as the fuel.

Would there be the risk of coolant leaks too? Hydrogen molecules are very small.

 

[Delta Force]

An He cooled "Magnox" could indeed be an interesting design for a natural uranium fueled reactor producing process steam at c400°C.

You don't use helium to go to 400 °C. It's very high temperature reactor coolant for conditions twice as hot. I'm not sure the expense of better coolants could be justified at only 400 °C, since carbon dioxide's worst effects only start above that temperature. You can't use Magnox type cladding above 400 °C either. It would be such a new design that it wouldn't even be a natural evolution like a Super Magnox.

Interestingly Magnox reactors where considered for oil tar sand recovery during the 1980s.

I thought it was peaceful nuclear explosions and then later it became water cooled designs?

It did! The turbines are fully interchangeable with other 660MWe turbines from fossil fuel plants. Hunterston B uses a turbine from the late Inverkip station (oil-fired) for example.
Interestingly, the 1200MWe AGR designed sketched in the 1960s uses two 600MW turbines in a compound arrangement.

Coal uses Rankine cycle steam turbines. With a Brayton cycle gas turbine reactor it would make more sense to use a natural gas turbine.

Beryllium is a moderator thence why it was considered.
I suspect the reason zircalloy was not considered may have to do with resistance to CO2 oxidation.

I don't see how having moderating cladding would be an advantage.

The weird thing is that a supermagnox is briefly mentionned in some documents in the national archives. But information on what this beast was is very scarce.

What kinds of things would the Super Magnox have involved?

It is worth remembering that as late as the Dungeness B tender in 1965 the basic AGR design was undefined in some levels. The tender documents make interesting reading in that respect. Things not defined include:
-Gas circulator driving (steam or electric)
-Gas flow arrangements within the core
-Reheat arrangements (gas/steam or steam/steam)
-Fuel element design
-Off-load or on-load refueling

Online refueling was optional? Sure it never quite worked as planned, but the economic case for AGR was based around online refueling.

I don't know how the gas flow and fuel element design could be left undecided at that point, seeing as that's a major part of nuclear reactor design.

Nitrogen is such a good neutron absorber that the secondary safety system of AGRs is actually a massive high pressure injection of nitrogen gas into the core.

Nitrogen-15 looks like it might have potential as a coolant.

Standard nitrogen is a handy safety system though. It's most of Earth's atmosphere, and it's inert. It might also be possible to get it out of the reactor following an incident and successful recovery, unlike boron injections.

As for Neon this tibit is interesting:
"However, on investigating the pumping power parameters suggested by Lyall [1], it becomes apparent that neon is not a very good coolant. Pure neon requires about eight times as much pumping power as pure helium for a slightly smaller flow area to give a constant heat transfer performance."

Too much pumping power is required basically.

Let's not even go there. CH4 is already present in the circuit of AGR in order to limit the reactions between graphite and CO2. Carbon deposition on all surfaces is the result of CH4 necessary presence in the circuit.

Carbon dioxide corrodes graphite too?

 

[Delta Force]

1200 MW(e) from presumably about a 3000 MW(t) reactor? Sounds promising to me. I suspect that this would probably be the same turbomachinery and electrical plant envisaged for the CFR at one point.

I rather suspect that, had the AGR line been developed, a Mark III AGR in the 1990s would have used 900 MW(e) generators common with those used on the proposed coal fired power plants of that era. Helium coolant seems likely - I think I read somewhere recently that had it been available in bulk at the time the original AGR was designed, it would have been preferred to CO2.

Gas cooled reactors can use Rankine cycle steam turbines from coal fired power plants, but why do that when they can directly drive Brayton cycle gas turbines from natural gas power plants? By the 1990s Brayton cycle turbines were seeing service in natural gas power plants.

 

[Delta Force]

Also, I came across something interesting while searching for alternative gas coolants for nuclear reactors.

It turns out that a helium-xenon cooled nuclear reactor produces 2% more power than a reactor cooled by helium alone and can also use much less complicated and much more compact turbines and compressors. The article is here. One issue is that xenon might undergo reactions in the reactor, so the authors have proposed switching another gas with it such as neon.