Energy From Thorium Discussion Forum

It is currently Jan 18, 2018 10:16 am

All times are UTC - 6 hours [ DST ]




Post new topic Reply to topic  [ 14 posts ] 
Author Message
PostPosted: Aug 24, 2010 1:41 pm 
Offline

Joined: Jul 14, 2008 3:12 pm
Posts: 5057
Noble metal extraction is often suggested via cold traps, where a small subcooled salt sidestream, with low flow speeds, stimulates precipitation of the noble metals onto a packed bed or sponge. Many (all?) of the trifluorides have low solubilities in BeF salts such as FLiBe, FLiNaBe or FNaBe. This is because they are not soluble in pure BeF2. Trifluorides are generally what we don't want in the LFTR: parasitic transuranics and lanthanide fission products eating up our neutrons in thermal reactor designs. How much of the low solubility trifluorides will come with this?

And, could we exploit this further, as 'pre-reprocessing'? A core salt of FLiNaBe or FNaBe for example, rich in BeF2 (30-40%), has low melting points and very low trifluoride solubility at the lower temperature range. If one goes for salt-only (no graphite) then the spectrum is quite fast and processing requirements are lower. In the French TMSR for example, a half year processing cycle could still break even. The NaF would eat some more neutrons but still if we look at trifluoride accumulation rates and compare that to FLiNaBe trifluoride solubilities, it looks like we could probably process out almost all of the trifluoride with such slow processing. Would UF3 come out too? Would that force us to use very high UF4/UF3 ratios?


Top
 Profile  
 
PostPosted: Aug 24, 2010 4:11 pm 
Offline

Joined: Jul 28, 2008 10:44 pm
Posts: 3069
Or
1) grab a small sidestream
2) bubble a very modest amount of F2 through the side-stream salt (to convert UF3 to UF4)
3) cool to precipitate trifluorides and noble metals
4) add a mix of Th and ThF4 to replenish the consumed thorium while bringing the fluoride back into balance for the right UF4/UF3 ratio.


Top
 Profile  
 
PostPosted: Aug 24, 2010 6:21 pm 
Offline

Joined: Jul 28, 2008 5:01 am
Posts: 462
Location: Teesside, UK
See also the zone refining discussion.


Top
 Profile  
 
PostPosted: Aug 25, 2010 2:57 am 
Offline

Joined: Jul 14, 2008 3:12 pm
Posts: 5057
Thanks for your comments. Zone refining is really cool but I would be a bit worried about the complexity of it. Not everything is known about all the trifluorides or complex mutant fusibility relationships and this will no doubt incur more research and engineering testing. It seems quite practical to have everything in liquid state, also from a heat management viewpoint. Operate just above the melting point of the fuel salt, and let most of the trifluorides precipitate out onto a packed bed of tungsten/molybdenum or something. Lars has a good idea to mildly fluorinate the UF3 out so that we don’t lose it in precipitation, so we get to keep easy corrosion control from fluorine balance. One issue is that you might get some other stuff that wants to fluorinate such as neptunium trifluoride. Or is this trivial for very mild fluorination, since UF3 selectively eats the fluorine?

Then thorium metal to get some fluorine back, another good idea, but this still leaves the more stable trifluorides. I guess we can use metallic Be and especially Li to get some more fluorine back and reduce more trifluorides to metal.

The resulting goo would be mostly metal(oids), and could be further reduced by adding magnesium, and could then probably be zone refined, perhaps in a central site, since there would be some fissile transuranics in it.


Top
 Profile  
 
PostPosted: Aug 25, 2010 9:58 am 
Offline

Joined: Jul 28, 2008 10:44 pm
Posts: 3069
Cyril,
I'm thinking that even cold but molten salt will have trifluoride solubility high enough that it is a pain for neutronics. But we could vacuum distill off most of the carrier salt (say 90%) which will increase the concentration of the trifluorides sufficiently for us to precipitate out the rest. I suspect you will need to store the concentrate for quite some time so that you don't concentrate the decay heat too much. The resulting "spent fuel" should be further processed (likely at a central location) to remove the actinides since most of the plutonium will go with the fission products.


Top
 Profile  
 
PostPosted: Aug 25, 2010 10:37 am 
Offline

Joined: Jul 14, 2008 3:12 pm
Posts: 5057
Hi Lars, it does sound like a lot to ask to rely on cold precipitation only though I wonder if a fastish spectrum machine could do it. One would do everything that is to be avoided with trifluoride solubility: salt with low melting point and a lot of BeF2. According to the French TMSR work, it is possible to break even with 180 EFPD reprocessing rate. So if the equilibrium trifluoride fission product concentration can be maintained by precipitation at less than the 180 day point it might allow breakeven. If that's possible then you wouldn't have to reprocess the U233 itself. It would be nice for a 1 or 1 1/2 fluid design. Perhaps we can use an HGMS to further stimulate precipitation in the cold trap. What is the magnetic nature of the lanthanide trifluorides?


Top
 Profile  
 
PostPosted: Aug 25, 2010 10:58 am 
Offline

Joined: Jul 28, 2008 10:44 pm
Posts: 3069
I don't think you can have a fast spectrum, 1 fluid, break even design unless it is absolutely huge. You will get too much leakage. Can't help with magnetics of trifluoride lanthoids.


Top
 Profile  
 
PostPosted: Aug 25, 2010 11:45 am 
Offline

Joined: Jul 14, 2008 3:12 pm
Posts: 5057
Lars wrote:
I don't think you can have a fast spectrum, 1 fluid, break even design unless it is absolutely huge. You will get too much leakage. Can't help with magnetics of trifluoride lanthoids.


It will be big but big isn't necissarily a major cost when you have everything at low pressure and if using graphite you'll probably be forced towards lower power density anyway. The MOSART and TMSR are not that big. Its roughly similar to ORNL's DMSR.

However, I think you can have your 1 1/2 fluid design with a blanket. The benefit is that you keep almost all thorium, as opposed to throwing it out in the still bottoms. So you get more U232 proliferation deterrent along the U233. And you don't handle the fissile U233 since it comes through the cold trap and goes directly back to the core. You only have to fluorinate out the U233 from the blanket but this is accompanied by loads of thorium and there will be more U232. The little bit harder spectrum will also help somewhat with losses to PaF4 as that will always remain in the core.

In terms of neutronics I'm not qualified to say it will or will not work but just looking at other work and comparing solubilities with reprocessing times it will be close...

Did anyone look into using a HGMS type device for the cold trap?


Top
 Profile  
 
PostPosted: Aug 25, 2010 5:01 pm 
Offline

Joined: Jul 28, 2008 10:44 pm
Posts: 3069
You mentioned a fastish spectrum. To me that means no graphite. The MSBR is a thermal spectrum machine with plenty of graphite. The French TMSR is a two or 1 1/2 fluid system (specifically it has a blanket).

For a one fluid (no blanket whatsoever), fastish spectrum machine you will need to be very large radius around 5m to get the neutron leakage down to a tolerable level for isobreeding. This isn't so large physically but you have an awful lot of fissile in there. This isn't a reactor that is 90% graphite by volume like MSBR and the percent fissile in the salt needs to be fairly high to keep the spectrum fast. To make this have a reasonable cost (Gwe/tonne fissile) you probably would try to get something like 8 GWatts or so out of the machine. So, if your going to a fastish spectrum I'd plan on some form of a blanket.


Top
 Profile  
 
PostPosted: Aug 26, 2010 11:13 am 
Offline

Joined: Jul 14, 2008 3:12 pm
Posts: 5057
Lars wrote:
You mentioned a fastish spectrum. To me that means no graphite. The MSBR is a thermal spectrum machine with plenty of graphite. The French TMSR is a two or 1 1/2 fluid system (specifically it has a blanket).

For a one fluid (no blanket whatsoever), fastish spectrum machine you will need to be very large radius around 5m to get the neutron leakage down to a tolerable level for isobreeding. This isn't so large physically but you have an awful lot of fissile in there. This isn't a reactor that is 90% graphite by volume like MSBR and the percent fissile in the salt needs to be fairly high to keep the spectrum fast. To make this have a reasonable cost (Gwe/tonne fissile) you probably would try to get something like 8 GWatts or so out of the machine. So, if your going to a fastish spectrum I'd plan on some form of a blanket.


The spectrum won't be THAT fast, like chlorides. Just a bit higher than epithermal. Maybe you could comment on whether this would work. We let the salt moderate, there is enough Li, Be and F. It would be a reactor with annular fuel tubes around a BeO moderator element in the center, the extra moderation on the inside of the tubes limits neutron leakage.
A low moderating reflector, such as iron, is placed around the annular tube core, to limit fast leakage without over-thermalizing the outer region. Axial leakage is reduced by tapering off the tubes at the plenum so that the fuel there becomes a subcritical absorber (see David LeBlanc reactor).


Top
 Profile  
 
PostPosted: Aug 26, 2010 11:54 am 
Offline

Joined: Jul 28, 2008 10:44 pm
Posts: 3069
There are lots of variants on a theme! Yes I think this would work neutronically. Imagine a spherical core with a spherical moderator in the center. Then the neutron spectrum will be soft near the moderator and hence lots of fission going on there. The neutron flux will spread out causing fission elsewhere. But it will die down away from the core since the neutrons will be faster and the fissile concentration will be too low to be critical on the faster neutrons. This is somewhat similar to MSBR with the undermodulated outer portion of the core.

Practically a right cylinder with height equal to twice the radius approximates a sphere. We can shape the BeO portion to be roughly a sphere in the center. The tricky part is that the BeO moderator can't be exposed to the fuel salt so it will have to be inside a Hastalloy or graphite container. These walls will be in the center of the flux. They will have a short lifetime. Perhaps if the center modulator is like a control rod we can readily replace the walls once a year. Frankly though I don't think they will last a year. I'm guessing its lifetime will be more like a month!


Top
 Profile  
 
PostPosted: Aug 26, 2010 12:55 pm 
Offline

Joined: Jul 14, 2008 3:12 pm
Posts: 5057
That's why the reactor will be cylindrical and fuel channel type. The channels are easy to replace once every so often. They could be metal like Hastelloy N or high molybdenium alloy.

LiF NaF BeF2 can be operated at a bit lower temperature than 2LiF BeF2 so we get mechanical strength improvement. This can also allow other somewhat lower temperature alloys which have superiour neutronic performance over nickel alloys. High iron alloys might do just fine and are real cheap to replace.

I think be BeO blocks should last long but replacing them isn't hard either, because the fuel channels mechanically disconnect the BeO, and most everything else for that matter. So no cladding on the BeO. Maybe a coating if you really insist, perhaps copper metal coating. 100 microns should be plenty. You could use graphite though, but make sure there is plenty of room to shrink and expand. Not too hard with fuel channel design.

One cool thing is that fuel channels can be made to fail at one point when they fail, by cutting it in to make a deliberate weak point. This is getting common practice in various hazardous gas/liquid storage industries.

Rapid fuel channel failure can cause splashing onto the BeO blocks, but this isn't a dramatic event. Shutdown reactor, remove the splashed salt or perhaps the entire BeO block, replace with new block and tube, and startup again. To get reliable operation one could have a scheduled replacement where all channels are replaced at the same time.

It would be interesting if the precipitation thing could lead to breakeven breeding. It avoids the difficult processing of ThF4 and does not handle UFx in pure form. With some design variation it might just work, barely...


Top
 Profile  
 
PostPosted: Feb 07, 2012 9:39 pm 
Offline

Joined: May 18, 2011 9:19 pm
Posts: 66
NF3 works differently, depending on temperature

http://www.pnnl.gov/main/publications/external/technical_reports/PNNL-20775.pdf

,,, this is related, but I'm guessing peripheral, to prior discussion in this thread (please excuse me if I'm misposting)


Top
 Profile  
 
PostPosted: Feb 08, 2012 7:49 am 
Offline

Joined: Jul 14, 2008 3:12 pm
Posts: 5057
It's not really related to this topic, since NF3 isn't part of the fission products. However it is a promising field of inquiry for reprocessing spent oxide fuels so that we can use the plutonium to startup LFTRs. Though for oxide reprocessing typically hydrofluorination is assumed to be more suitable. HF is much less of a corrosion/flame/toxicity issue than pure F2.

NF3 is also potentially attractive for batch fluorinations for the DMSR.

There may be a tread on it already. It was mentioned before by Kim.


Top
 Profile  
 
Display posts from previous:  Sort by  
Post new topic Reply to topic  [ 14 posts ] 

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