Energy From Thorium Discussion Forum

It is currently Dec 16, 2017 1:42 pm

All times are UTC - 6 hours [ DST ]




Post new topic Reply to topic  [ 17 posts ]  Go to page 1, 2  Next
Author Message
 Post subject: LMFR versus LFTR
PostPosted: Jan 02, 2013 3:22 am 
Offline
User avatar

Joined: Nov 30, 2006 9:18 pm
Posts: 1954
Location: Montreal
Not sure if many folks here are aware of it, but a remarkably close parallel development to ORNL's thorium fluoride molten salt breeder reactor - what is commonly referred to now as the LFTR - was under development at Brookhaven National Laboratory (BNL) back in the fifties & early sixties.

Parallel in the sense that both concepts used graphite moderator and a combination of core and blanket for breeding U233 fissile material from Thorium.

The only significant difference was the choice of carrier fluid: fluoride salts at ORNL, and Bismuth metal at BNL, for their Liquid Metal Fuelled Reactor (LMFR).
Of course Bismuth is a LOT cheaper than high-purity Lithium-7, and also cheaper than Beryllium.
These days, FLiBe salt with high-purity Li-7 is a serious show-stopper for would be LFTR developers, including China.

Also, like ORNL's fluoride MSR, molten Bismuth in the LMFR does not react violently with air or water (LFTR fluoride salts react to form oxy-fluorides when exposed to air, and must therefore use an inert cover gas, like Bismuth, to avoid the formation of undesirable contaminants which may plug up or corrode the system, especially heat exchangers...)

Aside from the ready availability of Bismuth, this carrier fluid offers other advantages which are unavailable with fluoride salts, such as simpler processing for fission product removal (only fissile material and Bismuth recycling to worry about, instead of fissile plus Li-7 and Beryllium fluorides, as in the LFTR), as well as potential alternatives to the standard centrifugal pumps for fuel circulation through the reactor and heat exchangers (such as Electro-Magnetic, or "EM" pumps, with no part of the driving mechanism immersed in the fuel fluid, and no risk of leaky seals..)

It's interesting that EM pumps for liquid metals were tried as early as the 1940s - initially in the UK, and then in lab experiments at BNL, where the LMFR (Bismuth carrier with a small % of HEU), was being developed.

This is briefly mentioned in the old publication Liquid Fuelled Reactors (Addison-Wesley, 1958).
It notes that the pumping efficiency was very low - not least because the tubing was made of stainless steel, which apparently short-circuited the EM drive to some extent.

This problem would of course vanish with non-metal tubing (such as SiC and some types of graphite).
Unfortunately, there is no indication of the resulting improvement in efficiency for non-metal tubing. Nor is there any way to know how different it might be, if pure molten U-metal were substituted for bismuth (or perhaps by a eutectic of U-Si, which contains a few % Si, to lower the m.p. by about a hundred degrees, plus the usual fission product contaminants...)

Lastly, the book also mentions "linear induction pumps", which are presumably different from EM pumps: not being an electrical engineer, I have no idea what the difference is, or whether the latter might be a better option, under some circumstances....

Regardless, even if the best possible pumping efficiency is on the low side, compared to centrifugal pumps, I believe that the tremendous advantage of having a completely sealed tube element, instead of trouble-prone centrifugal impeller bearings and seals, make the choice a plain no-brainer !!
SIXTY YEARS ago, the obvious choice for both LMFR and MSR fuel tubing and reactor vessels was metal, simply because no other technology was available at the time.

The unfortunate consequence of that was that corrosion issues limited the operating temperatures to such low levels, that a low-melting carrier metal like bismuth was an absolute requirement: molten Uranium was simply out of the question, due to the much higher operating temperature and associated corrosion rates, when coupled with metal tubing and reactor vessels.

Worse yet, since the solubility of U in bismuth is only about 1.3% (at 600C), it had to be HEU - fully enriched fissile material - to have any hope of the reactor ever reaching criticality (Actually, the solubility governing an engineering design is much lower, referencing the figure of 0.048% U at the melting point of Bismuth, 271C, because at no time can we allow a significant amount of HEU fissile material to precipitate out of solution; in fact, BNL considered 0.10% U the maximum allowable; Thorium is ten times less soluble still, necessitating the use of a particle slurry in the blanket fluid).

Like ORNL's MSR, BNL's use of HEU in the LMFR was considered entirely acceptable back then - as BOTH research teams in fact assumed at the time.
But it makes both concepts total non-starters politically, in today's far less permissive nonproliferation environment.
It is EXACTLY the same issue that makes the development and deployment of classic-type pure Thorium-U233 breeders (LFTRs) impossible today.

On the other hand, the evolution of composite and ceramics materials technology make feasible the use of much higher-temperature metal fuels today, including carrier-free U-metal or U-Si eutectic.
With such high fuel loading (no carrier metal or FLiBe salt) the fissile enrichment level can be far lower - right down to NU (natural uranium) - to achieve reactor criticality (when matched with an appropriate moderator material, or even a moderator-free fast reactor, if enrichment is in excess of about 10% U235 equivalent).

The governing political constraint is thereby entirely avoided - albeit with some technical risk. But any such risk is LOWER COST than for LFTR, as it doesn't involve Li7 acquisition.

Moreover, the fission product cleaning process is simplified, since one need not worry about extremely efficient recycling of the carrier metal (if any!) the way one must with precious Li7.

The process described in Fluid Fuel Reactors for Bi-U cleaning was evidently quite effective for removing the important neutron poisons, especially xenon and samarium (helium sparging was found to be NOT required for Xe removal in liquid metals!)

It included the "fluoride volatility" purification line, which should work reasonably well for plutonium recovery & return to the reactor, along with unused uranium (both in the form of hexafluoride gases, subsequently reduced back to metal, but never separating any pure fissile material, as in the case of LFTR breeding blanket U233 separation and feed to the core...)

Obviously, with low-enriched or natural uranium (LEU or NU) in the core, the cycle automatically changes from Th-U233 conversion to U-Pu239 conversion, due to the abundance of U238 in the mix.
No breeding blanket is used, as no sequestration of long-lived Pa233 is required to avoid neutron absorption & conversion to useless U234, instead of the desired U233.

Molten Uranium metal reacts with air, so must also be protected by an inert cover gas, like FLiBe salt or Bismuth. In addition, a thin layer of liquid Tin may be floated on top (pool-type core arrangement) to protect against accidental cover gas loss (the reactor vessel hot cell may include a small puddle of tin on the floor below, so that if the fuel leaks, it will again end up under a cover of tin, regardless of the cover gas situation above... Barring any leaks, normal shutdown procedure drains the fuel to decay tanks, just like LFTR)

All in all, it seems like a reasonable trade-off, considering the current state of materials technology, versus our terminally fatal nonproliferation policies, as well as the perennial Li-7 "unobtainium" issue...

By contrast, typical fluoride fuel reactor concepts - particularly those using cheaper alternatives to FLiBe carrier salt - do NOT offer a politically compliant alternative with similar fissile breeding potential.

It may take many decades or even centuries to change national and global nonproliferation policies: in the mean time, engineers should concentrate on finding technical solutions which are politically acceptable and licensable. Other concepts are little more that science fiction: a waste of engineers' time & effort.

Happy 2013 to all !

Jaro


Top
 Profile  
 
 Post subject: Re: LMFR versus LFTR
PostPosted: Jan 02, 2013 7:36 am 
Offline

Joined: Jul 14, 2008 3:12 pm
Posts: 5056
If Li7 is a problem, then Na can be substituted easily with very little loss in breeding or conversion ratio.

But let's go along your line of thought just for sake of analytical discussion.

EM pumps' efficiency is a strong function of the electrical conductance of the inducted metal. Sodium has quite good electrical conduction at SFR operating temps (<550 C). However even those pumps require a lot of power. It is likely that molten uranium is much worse; the absurd weight of the stuff (ie low gravimetric heat capacity) will increase pumping cost, as will operating at a high temperature. Fortunately the volumetric heat capacity of uranium metal is very good (much better than sodium but reliable figures for molten U are hard to find).

So my questions here are:

1. What is the electrical conductance of molten uranium (or U-Si). Preferably as a function of temperature.
2. What is the Curie temperature of molten uranium?

A compromise solution is a canned motor pump. The AP1000 uses such pumps to push 3100 MWt so this is a realistic solution. The efficiency is greatly improved, because a high conductivity metal can be used for driving hydraulic impeller portion. Seals are at least eliminated.

Regarding the use of molten uranium in general: this is almost certainly a big PR problem. Molten uranium is pyrophoric. Protective tin layer is pretty high safety, but if something happens that blows off the protective tin layer (maybe an explosion from sabotage or terrorist attack) then the safety case becomes complicated.

The materials issue does not look so solvable to me. Molten uranium is about the worst material you can have in terms of corrosion. It could potentially corrode even SiC and graphite, as uranium forms a stable carbide and uranium metal is reactive as hell. The residual (reduced) silicon will then dissolve into the U, providing an erosion mechanism. Most metals are non-options too, as uranium tends to either react or form low melting eutectics with them. The fission products, in metallic form, will also be a corrosion problem. Some fission products form very stable carbides and nitrides, giving rise to a corrosion concern for most ceramics.

Zirconium carbide looks promising. Or even the oxide, if it doesn't dissolve.

As previously discussed, processing would be very easy, theoretically, with molten uranium: all the actinides are high boiling, so you could simply boil out the fission products and leave U and Pu plus higher actinides. IF you can deal with the materials issues...


Top
 Profile  
 
 Post subject: Re: LMFR versus LFTR
PostPosted: Jan 02, 2013 7:54 am 
Offline
User avatar

Joined: Nov 30, 2006 9:18 pm
Posts: 1954
Location: Montreal
Thanks Cyril.
The corrosion issue was researched extensively by the BNL team (and others), as noted in the book.
They addressed the problem by adding 300ppm each of Magnesium and Zirconium metals to the Bi-U solution.
The Mg was to get any traces of oxygen, the Zr to prevent U from reacting with the graphite moderator (the Zr reacts with C atoms on the surface and stays there, preventing U from doing the same; BNL considered this very important, due to the HEU being used, where small changes in localized concentrations could have undesirable effects on reactivity. Personally, I would prefer a more modern type of corrosion inhibition, using an appropriate coating applied during manufacture...)

Interestingly, they also note that the reaction rates for the molten material are in the following order: Mg, U, Zr ....with Mg being most reactive and Zr the least.
Mg is of course much less reactive than Sodium, so my guess is that U would be quite manageable.
The pyrophoric nature of U comes into play when fine filaments or powder are involved - much like the old style camera flashes, using Mg wool inside a little plastic cube (...if you're old enough to remember those ? )


Top
 Profile  
 
 Post subject: Re: LMFR versus LFTR
PostPosted: Jan 02, 2013 9:00 am 
Offline

Joined: Jul 14, 2008 3:12 pm
Posts: 5056
Fine filaments or powder are very similar to a molten state, where convection currents simulate an effective high surface area. Molten U is quite scary, and molten plutonium even more so. Still, with some tricks like the molten tin cover, argon cover gas, and undergrounding (to prevent rapid loss of cover gas in beyond design basis accidents) it should be very safe. Carryover of tin must be avoided, as must carryunder of uranium in tin (if excessive) otherwise it won't protect. ORNL found even small eutectics of molten uranium metal with molten non-flammable metals to be pyrophoric.

There could be problem if plutonium carbides end up stable, as you then get excessive fission on material interfaces - just where we can't have it.

The Mg and Zr additions sound familiar, I think these are also used in molten lead and Pb-Bi coolants.

For graphite and natural uranium (with low fissile content), it is likely that passivation of uranium carbide is acceptable. This should in fact be very useful if it is self sealing, because it will form a passivation layer similar to chromium in stainless steels. A fabricated coating would be great, but my experience with coatings is that they can degrade, spall off or erode in fluid flow. Coatings are particularly tricky for very high temperature applications, due to mismatch of thermal expansion coefficient. Graded coatings are a solution to metallic components but don't work very well with ceramics. So if we have a material that is inherently self passivating then that would be better than a thin coating.

We know that the reaction U + C = UC (and possibly higher valences as well) proceeds appreciably with lots of uranium metal around; ORNL found that uranium carbides developed on the graphite if the redox potential of the salt was high enough for free uranium to be in the melt.


Top
 Profile  
 
 Post subject: Re: LMFR versus LFTR
PostPosted: Jan 04, 2013 6:24 am 
Offline

Joined: Jul 14, 2008 3:12 pm
Posts: 5056
Jaro, I hope you're not thinking of D2O moderation in this design. Molten uranium explodes upon contact with water and generates a lot of hydrogen.


Top
 Profile  
 
 Post subject: Re: LMFR versus LFTR
PostPosted: Mar 20, 2013 7:36 am 
Offline
User avatar

Joined: Nov 30, 2006 9:18 pm
Posts: 1954
Location: Montreal
jaro wrote:
a remarkably close parallel development to ORNL's thorium fluoride molten salt breeder reactor - what is commonly referred to now as the LFTR - was under development at Brookhaven National Laboratory (BNL) back in the fifties & early sixties.

Parallel in the sense that both concepts used graphite moderator and a combination of core and blanket for breeding U233 fissile material from Thorium.

The only significant difference was the choice of carrier fluid: fluoride salts at ORNL, and Bismuth metal at BNL, for their Liquid Metal Fuelled Reactor (LMFR).

Revisiting this thread in light of a recent analysis of the core lattice design of the original MSBR proposed by ORNL, by the Czech nuclear research institute UJV (http://home.comcast.net/~aeropharoh/Fry ... MSRICAPP09[1].pdf )

The analysis come up with some interesting results:
" decreasing of the fuel channel radius did not result in higher graphite lifetime, on the contrary, the neutron spectrum was hardening – fuel salt is better moderator than graphite "

That being the case, there may be additional options for reactor design, which were perhaps not considered previously.

For example, if a two-fluid U233/Thorium breeder is desired, then one option might be a reactor using FLiBe liquid moderator with a small amount of ThF4 dissolved in it: If the magnitude of the negative reactivity due to Th is smaller than the positive reactivity due to the FLiBe moderator, then an accident involving moderator draining will result in a safe shutdown (unlike the MSBR case, where the thorium is in separate annular channels, such that its draining will cause a huge spike in reactivity, because the graphite moderator remains in place...).

Having a liquid moderator to carry the fertile material also opens various options for the mechanical design.
Most importantly, it can greatly simplify the dreaded "piping nightmare" for which the old ORNL MSBR design is renowned.

Also, with a liquid moderator, the heat produced by the reactor can be extracted by circulating the moderator through heat exchangers, instead of the fuel in the fuel channels (analogous to current LWRs).
That can simplify materials issues considerably, since only the fuel channels must be made to endure the high temperatures and corrosive nature of the fuel inside, rather than an entire heat transfer circuit, including pumps, HX, etc.
For example, SiC fuel tubes would be an obvious choice.

The fuel inside SiC tubes could then be any one of a great variety of possible choices.
One interesting choice would be the Bismuth-U alloy used in BNL's old LMFR design. This would allow volatile FP degassing, as well as simpler processing compared to LFTRs.
Other options might be fuels like molten U metal or UF4/UF3 eutectics.
Basically, all the options previously considered for an HW-MSR (or HW-LMFR), except that the moderator is now FLiBe instead of HW.

There is some similarity here to Bekeley's PB-AHTR concept.
However, in that case the FLiBe is only used as coolant, while the bulk moderator remains solid graphite - both in static block form, as well as the semi-mobile graphite pebbles.
In fact, the UJV confirmation of FLiBE being a better moderator than graphite raises a potential safety issue, in that a "leak" in the PB-AHTR fuel handling system (the pebbles float in the FLiBe coolant, up to a collection funnel) could cause gross displacement of graphite by FLiBe, resulting in an unexpected increase in reactivity.

Of course it would be preferable to avoid having Li7 unobtainium anywhere in a reactor, but it is a good moderator that is not easily replaced, if we want a high-temperature liquid, plus compatibility with ThF4.


Top
 Profile  
 
 Post subject: Re: LMFR versus LFTR
PostPosted: Mar 20, 2013 1:13 pm 
Offline

Joined: Jul 14, 2008 3:12 pm
Posts: 5056
Quote:
For example, if a two-fluid U233/Thorium breeder is desired, then one option might be a reactor using FLiBe liquid moderator with a small amount of ThF4 dissolved in it: If the magnitude of the negative reactivity due to Th is smaller than the positive reactivity due to the FLiBe moderator, then an accident involving moderator draining will result in a safe shutdown


How much ThF4 can you put in without the alpha-void becoming positive? Breeding on thorium means you need to cram a lot of thorium in the reactor system. It has to compete favorably for neutrons.

Quote:
Having a liquid moderator to carry the fertile material also opens various options for the mechanical design.
Most importantly, it can greatly simplify the dreaded "piping nightmare" for which the old ORNL MSBR design is renowned.


The two fluids still have to be seperated. Just like any 2 fluid design. David Leblanc's tube-in-shell can eliminate graphite moderator without unacceptable breeding/doubling time, etc. But it still has that barrier to worry about, right in the worst kind of environment you could think of. Having more tubes would mean more bonds, etc. that can leak, embrittle.

Quote:
The fuel inside SiC tubes could then be any one of a great variety of possible choices.
One interesting choice would be the Bismuth-U alloy used in BNL's old LMFR design. This would allow volatile FP degassing, as well as simpler processing compared to LFTRs. Other options might be fuels like molten U metal or UF4/UF3 eutectics.
Basically, all the options previously considered for an HW-MSR (or HW-LMFR), except that the moderator is now FLiBe instead of HW.


One of the advantages of your heavy water calandria design, is the low temperature and known chemical-irradiation effects of heavy water under these conditions. This allows proven materials such as Zr alloys to be used for this application. FLiBe would mean high temperature liquid containment, and the chemical nature of FLiBe limits the choice of materials. An all composite calandria would obviously be a major RD&D effort. Hastelloy N calandria is low RD&D, but would require periodic replacement due to embrittlement in neutron flux and would steal neutrons in the process.

Your design decouples the moderator requirement, one of the things I really like about it. The moderator is not mechanically attached to the fuel channel system at all. That means you could use beryllium oxide moderator, which is better than FLiBe, is not liquid so can't leak out - doesn't even need a calandria at all - and has high temperature capability so you can just let it sit there being yellow-hot. Plus it avoids Li-7 and fluorine captures (oxygen has lower capture than fluorine). There wouldn't be a need for any cooling system, the moderator blocks just sit there heating up the composite fuel channels, ie the fuel channels are your heat exchanger. This should be acceptable if the blocks need only support their own weight, nothing else. Which is the case.

I think this option would be low R&D because the ARE already used something almost identical.


Top
 Profile  
 
 Post subject: Re: LMFR versus LFTR
PostPosted: Mar 24, 2013 9:06 am 
Offline
User avatar

Joined: Nov 30, 2006 9:18 pm
Posts: 1954
Location: Montreal
Cyril R wrote:
....you could use beryllium oxide moderator, which is better than FLiBe, is not liquid so can't leak out - doesn't even need a calandria at all - and has high temperature capability so you can just let it sit there being yellow-hot. Plus it avoids Li-7 and fluorine captures (oxygen has lower capture than fluorine). There wouldn't be a need for any cooling system, the moderator blocks just sit there heating up the composite fuel channels, ie the fuel channels are your heat exchanger.

Beryllium is better in some respects, but not others.

The French MSFR concept gets rid of BeF2, mainly because they don't like the large amount of tritium it produces - they prefer LiF-NaF.
This is probably why BNL's LMFR was cancelled before ORNL's MSR: The former hardly produced any tritium, which was needed for production of light weight fusion-boosted bombs (several dedicated tritium production reactors were eventually built at DoE's Savannah River Site, so MSR was no longer needed).

The advantage of using the moderator as heat exchange medium is two-fold.
Firstly, by allowing the fuel to remain stagnant in lattice tubes, there is no loss of delayed neutrons outside the reactor.
This makes for a safer, easier to control reactor.
Secondly, a solid uncooled moderator - graphite or BeO - gets much hotter than the fuel, which then becomes an issue for the reactor vessel and moderator support structures.

Also, it's not true that BeO "doesn't even need a calandria at all": You can't allow fluoride salts to directly contact BeO, so some sort of barrier - similar to a calandria - would be required to protect the moderator against corrosion.
It is easier to find materials for calandria tubes when the fuel is Bismuth-Uranium, than for FLiBe-U salts. And of course you minimize tritium production in the fuel (but not in the moderator).


Top
 Profile  
 
 Post subject: Re: LMFR versus LFTR
PostPosted: Mar 24, 2013 9:47 am 
Offline

Joined: Jul 14, 2008 3:12 pm
Posts: 5056
Quote:
The French MSFR concept gets rid of BeF2, mainly because they don't like the large amount of tritium it produces - they prefer LiF-NaF.


Li also produces tritium, in amounts that require cleanup and removal. If you need that anyway, having more tritium I doubt will be a major cost issue. Also LiF-NaF is not a good moderator and needs your unobtanium Li-7, much more so if used as moderator (because you need more of it).

Quote:
Firstly, by allowing the fuel to remain stagnant in lattice tubes, there is no loss of delayed neutrons outside the reactor.
This makes for a safer, easier to control reactor.


It also ruins your heat transfer path, compared to being able to pump liquid fuel through a heat exhanger. This would be a solid fuel reactor, requiring a large fuel cladding or ceramic/composite HX area in the core. The heat exchanger is still there, but you've put it in the core like today's reactors. I don't think this improves on the situation. Removal of offgas will be much more difficult, requiring lots of pipework inside the core. Also fuel expansion as a safety feature is reduced greatly, no longer being able to expand fuel out of core.

Quote:
Secondly, a solid uncooled moderator - graphite or BeO - gets much hotter than the fuel, which then becomes an issue for the reactor vessel and moderator support structures.


There wouldn't be a vessel in your HW-MSR design with BeO moderator blocks. The only support would be at the bottom, in the form of a pedestal. That's almost childishly easy: a simple stack of magnesia or alumina firebrick would do it.

Quote:
it's not true that BeO "doesn't even need a calandria at all": You can't allow fluoride salts to directly contact BeO, so some sort of barrier - similar to a calandria - would be required to protect the moderator against corrosion.


In your HW-MSR, there are graphite or composite siphons that provide the barrier. If they fail, a pyrolytic or silicon carbide coating on the siphon facing BeO surface would be sufficient protection against salt splash.

Quote:
It is easier to find materials for calandria tubes when the fuel is Bismuth-Uranium, than for FLiBe-U salts.


I highly doubt it. Bismuth dissolves most metals, and uranium attacks most metals through various mechanisms. Both attack a number of composites and ceramics, as well.

Quote:
And of course you minimize tritium production in the fuel (but not in the moderator).


So in stead of just tritium you get loads of polonium in the fuel and still get loads of tritium in the moderator.


Top
 Profile  
 
 Post subject: Re: LMFR versus LFTR
PostPosted: Mar 25, 2013 1:56 am 
Offline

Joined: Jul 28, 2008 10:44 pm
Posts: 3070
Cyril R wrote:
Quote:
The French MSFR concept gets rid of BeF2, mainly because they don't like the large amount of tritium it produces - they prefer LiF-NaF.

.

The French did not like the chemical poison part of Be. The majority of tritium comes from the lithium.


Top
 Profile  
 
 Post subject: Re: LMFR versus LFTR
PostPosted: Mar 25, 2013 5:15 am 
Offline
User avatar

Joined: Nov 30, 2006 9:18 pm
Posts: 1954
Location: Montreal
Lars wrote:
The French did not like the chemical poison part of Be. The majority of tritium comes from the lithium.

Quoting from one of the reports of the French MSFR group:
Quote:
In the case of FLiBe, part of 6Li is regenerated by the 9Be(n,alpha) reaction. With FLiNa, this regeneration is no longer possible and 6Li remains only as traces, with 0.06 mol instead of 19.6 with FLiBe. Then, only the 7Li(n,nt) reaction contributes to tritium production. Since the LiF proportion is lower than in FLiBe, the tritium production rate at equilibrium is close to 50 g/y, that is to say about one third of the value obtained with FLiBe.


Top
 Profile  
 
 Post subject: Re: LMFR versus LFTR
PostPosted: Mar 25, 2013 7:53 am 
Offline

Joined: Jul 14, 2008 3:12 pm
Posts: 5056
jaro wrote:
Lars wrote:
The French did not like the chemical poison part of Be. The majority of tritium comes from the lithium.

Quoting from one of the reports of the French MSFR group:
Quote:
In the case of FLiBe, part of 6Li is regenerated by the 9Be(n,alpha) reaction. With FLiNa, this regeneration is no longer possible and 6Li remains only as traces, with 0.06 mol instead of 19.6 with FLiBe. Then, only the 7Li(n,nt) reaction contributes to tritium production. Since the LiF proportion is lower than in FLiBe, the tritium production rate at equilibrium is close to 50 g/y, that is to say about one third of the value obtained with FLiBe.


But you still get the initial Li6 in the Li7. So initial tritium production is much higher. And even the 7Li tritium production is high enough to require removal and isolation systems. Most of the cost and difficulty is still there. The cost difference between removing 1000 Ci/year and 10000 Ci/year is probably very small.


Top
 Profile  
 
 Post subject: Re: LMFR versus LFTR
PostPosted: Mar 09, 2015 7:41 pm 
Offline

Joined: Mar 08, 2015 4:34 pm
Posts: 12
This is sorta like the "dual" of the dual fluid reactor. Molten metal fuel and salt breeder coolant. Would this HAVE to be a fast spectrum or could it operate in a thermal or epi-thermal to reduce initial fissile loading requirements?


Top
 Profile  
 
 Post subject: Re: LMFR versus LFTR
PostPosted: Mar 10, 2015 12:49 am 
Offline
User avatar

Joined: Nov 30, 2006 9:18 pm
Posts: 1954
Location: Montreal
Andrew wrote:
This is sorta like the "dual" of the dual fluid reactor.
What is ? ....the LMFR ?

Andrew wrote:
Molten metal fuel and salt breeder coolant.
LMFR has no salt.

Andrew wrote:
Would this HAVE to be a fast spectrum or could it operate in a thermal or epi-thermal to reduce initial fissile loading requirements?
LMFR is graphite moderated, thermal spectrum.


Top
 Profile  
 
 Post subject: Re: LMFR versus LFTR
PostPosted: Mar 10, 2015 9:34 pm 
Offline

Joined: Nov 23, 2010 6:51 pm
Posts: 123
Would this reactor even need pumps? Natural circulation is particularly efficient with dense liquid metals, and the cost of the large fissile inventory necessitated by lower power density is a non-issue if the reactor could be started with natural uranium. There was also a discussion in the HW-MSR thread about the possibility of breakeven breeding on the U-Pu cycle with the bi-modal spectrum resulting from wide fuel channels instead of the narrow pins used in solid fuel reactors. Substituting carbon for D2O would hurt the breeding ratio, but removing all moderation from the fuel itself would help maximize fast fission of 238U.


Top
 Profile  
 
Display posts from previous:  Sort by  
Post new topic Reply to topic  [ 17 posts ]  Go to page 1, 2  Next

All times are UTC - 6 hours [ DST ]


Who is online

Users browsing this forum: No registered users and 1 guest


You cannot post new topics in this forum
You cannot reply to topics in this forum
You cannot edit your posts in this forum
You cannot delete your posts in this forum
You cannot post attachments in this forum

Search for:
Jump to:  
Powered by phpBB® Forum Software © phpBB Group