Energy From Thorium Discussion Forum

It is currently Jun 23, 2018 2:35 pm

All times are UTC - 6 hours [ DST ]




Post new topic Reply to topic  [ 120 posts ]  Go to page Previous  1 ... 4, 5, 6, 7, 8  Next
Author Message
PostPosted: Jan 13, 2012 4:36 am 
Offline

Joined: Jul 14, 2008 3:12 pm
Posts: 5050
Here's a video of Charles Holden showing his design of a small LFTR, that we talked about earlier. It's remarkably similar to the design discussed in this thread.

http://nuclearstreet.com/nuclear_power_ ... 10402.aspx

This design looks like a very inefficient one, it requires 1600 kg U233 fissile startup - for a 40 MWe reactor! Way too high. Clearly all that Hastelloy N cladding has to go away, and a shorter burn, fuel shuffeling operation should be introduced.


Top
 Profile  
 
PostPosted: Jun 15, 2012 1:44 pm 
Offline

Joined: Dec 05, 2008 8:50 am
Posts: 337
At about 10:30 to 11 min he claims that the reactor pratically produces no transuranics waste (less than milligram/year), it seems me this is quite different from other LFTR wastes claims where something between at least hundreds grams to tens kg/GWyear is often stated to be produced (besides the different reactor size, only 40 MWth)


Top
 Profile  
 
PostPosted: Jun 15, 2012 11:24 pm 
Offline

Joined: Apr 19, 2008 1:06 am
Posts: 2225
The design has no traditional fuel rods.
Once you go for an under-moderated design, you may as well get rid of BeF2 and an go for an un-moderated one. You could use NaF-ZrF4 solvent. UF4 denaturant could provide extra neutrons in fast fission in the faster spectrum and increase production of fissile isotopes. There is enough fissile material floating round the world which should best be doing something useful inside the fast reactor cores.
http://www.fissilematerials.org/
Future recovery of valuable fissiles should be as TRU cocktail as suggested in the IFR reprocessing,by electro-refining.


Top
 Profile  
 
PostPosted: Jun 16, 2012 2:29 pm 
Offline
User avatar

Joined: Nov 30, 2006 9:18 pm
Posts: 1947
Location: Montreal
Cyril R wrote:
....a shorter burn, fuel shuffeling operation should be introduced.
Holden's concept uses molten salt fuel !
In that respect, it is NOT a molten salt cooled reactor with traditional fuel rods!

On the plus side, there is no need for fuel shuffling - the fuel is constantly homogenized.

But what puzzles me -- and I've mentioned this before -- is how they manage to connect the thousands of little hastelloy fuel tubes to the top & bottom manifolds (the top one for fission gas collection, the bottom one for emergency fuel salt dump by way of freeze plug.)
And even if they manage to connect them all, what happens to the whole assembly after a few years of neutron-induced creep ?
How do you avoid gross deformation and/or joint cracking ?
Remember that the core and the top & bottom headers are submerged in a deep pool of cooling salt and, as Holden states, you don't want the two to mix (which would happen if cracks develop).

Its odd that the presentation never shows any hint of the manifold/fuel tubes interface: that is clearly the weak point of this concept (as presented), along with the large load of neutron-absorbing Hastelloy in the core.

In earlier discussions of molten salt cooled reactor with traditional fuel rods, we mentioned one way of avoiding these problems: plug the tubes at the bottom, and run long empty sections of the tubes up above the surface level of the cooling salt pool.
This would not be practical in Holden's natural circulation concept, due to the extremely deep pool.

A potential solution for the mechanical layout of Holden's MSR might be to have a fixed connection (welds) at the bottom manifold, letting thermal expansion & creep grow freely at the top.
Then, because fission gas venting needs only very small-diameter tubing, these could connect to the rigid reactor core fuel tubes through plugs at the top, and then run flexible lengths (including loops) to separate headers having a different, less congested configuration, than indicated in the core section diagram.
In other words, a layout that looks somewhat like a miniaturized version of the feedwater piping in a CANDU reactor.


Top
 Profile  
 
PostPosted: Jun 17, 2012 6:25 pm 
Offline
User avatar

Joined: Nov 30, 2006 9:18 pm
Posts: 1947
Location: Montreal
Here’s a little diagram illustrating the mechanical arrangement described in the previous post.

There’s a good chance that this could be done with the fuel tubes made out of silicon carbide (SiC), which would be far better than the Hastelloy proposed in Holden’s presentation.

The small vent tubes would still have to be metal, to avoid any chance of fracturing (I’m guessing 3/16” diam. Would be ideal).
In reality, the expansion loops would be laid out horizontally, with a slight slope towards the fuel tube top plug, so that any vapour condensation can drain back out, instead of accumulating at a low point and eventually blocking the tube.
Similarly, the connections to the vent header would be from below, rather than from the side as shown.


Attachments:
MSR_rods_plus_FP_gas_tubes_and_header..jpg
MSR_rods_plus_FP_gas_tubes_and_header..jpg [ 121.05 KiB | Viewed 4264 times ]
Top
 Profile  
 
PostPosted: Jun 21, 2012 4:53 pm 
Offline
User avatar

Joined: Nov 30, 2006 9:18 pm
Posts: 1947
Location: Montreal
I see in their latest presentation (TEAC4), Thorenco’s "New Core Geometry" still has thousands of tubes RIGIDLY held between the thick top & bottom core plates.
Quote:
3700 tubes and 3700 annular coolant passages that traverse fuel tank
How do you fabricate something like this ?

And what happens after a few years of operation, with tube creep varying across the core ?

Anybody hear more details about this during the presentation ? (Thnx)


Attachments:
Thorencos_New_Core_Geometry_core_section_details.jpg
Thorencos_New_Core_Geometry_core_section_details.jpg [ 291.56 KiB | Viewed 4223 times ]
Thorencos_New_Core_Geometry_core_section.jpg
Thorencos_New_Core_Geometry_core_section.jpg [ 338.89 KiB | Viewed 4223 times ]
Top
 Profile  
 
PostPosted: Sep 08, 2012 10:25 am 
Offline

Joined: Jul 14, 2008 3:12 pm
Posts: 5050
The talk about the MFE/BISO fuel got me thinking on a different variant of this salt cooled design.

I guess this pool type solid fuelled reactor could be simplified (as in to get a simple prototype design) to have MFE fuel. To make this practical, the MFE fuel could be put in perforated SiC boxes. The boxes could then be handled and shuffled normally (as fuel assemblies). A bit more exotic would be to have some kind of suction device that sucks out the fuel particles without unloading any assembly. This means you'd need such a device, but the advantage is not having to move any fuel assemblies (just a replaceable core internal structure made of SiC).

This would be a simple burner arrangement of course, not a closed reprocessing cycle. But such an arrangement would never put fission products in the containment, venting lines, etc. so would be easy to license and develop (use proven TRISO fuel particles). This is closer to the AHTR, but, at the same time it would get a higher power density, higher HM loading, and lower fuel temperature than the AHTR. It also allows more freedom in the neutron spectrum, because seperate graphite moderator blocks could be incorporated in such amounts as to tailor the neutron spectrum to whatever is optimal. Similar to the AHTR, it would allow development of the required salt components such as pumps, etc. that will benefit a LFTR.


Top
 Profile  
 
PostPosted: Sep 09, 2012 8:05 am 
Offline
User avatar

Joined: Nov 30, 2006 9:18 pm
Posts: 1947
Location: Montreal
Cyril R wrote:
This is closer to the AHTR, but, at the same time it would get a higher power density, higher HM loading, and lower fuel temperature than the AHTR.
I can see this working OK with SCW or gas coolants, but salt cooland may have far too much pressure drop across such a fine-grained bed: Presumably UCB would have considered this option and went with the larger graphite TRISO balls....


Top
 Profile  
 
PostPosted: Sep 09, 2012 12:41 pm 
Offline

Joined: Jul 14, 2008 3:12 pm
Posts: 5050
jaro wrote:
Cyril R wrote:
This is closer to the AHTR, but, at the same time it would get a higher power density, higher HM loading, and lower fuel temperature than the AHTR.
I can see this working OK with SCW or gas coolants, but salt cooland may have far too much pressure drop across such a fine-grained bed: Presumably UCB would have considered this option and went with the larger graphite TRISO balls....


That's why the bed needs to be laid out axially, but with radial flow. As in a perforated vertical touble SiC tube or box, with the MFE arranged in the annulus between the tubes/boxes. The salt comes in from below, around the outer tubes/boxes. Then it must pass radially (and to a less extent axially) through the bed in between the outer and inner tube/boxes, to come out on the inside. There's the second tube/box, that channels the heated up coolant upwards to a hot plenum.

The PB-AHTR does something similar in the Berkeley embodiment, with both radial and axial flow & collection plenums.

Some amount of pressure drop is still beneficial, for flow stability. Though it is true that an open core, pool type arrangement is more limited in the allowable pressure drop.


Top
 Profile  
 
PostPosted: Nov 03, 2014 5:30 pm 
Offline

Joined: Oct 28, 2013 12:24 am
Posts: 256
Good work guys, I really like this molten salt cooled design with big fuel rod, metal fuel and silicon carbide cladding (if we can use SiC for cladding). (In fact I like this idea better than molten salt fueled reactors, I prefer a clean plant and a vessel protected from neutron flux than having fission products and actinides moving everywhere, but I am not an expert).

The other fast reactors concepts have all issues with their coolants : highly reactive sodium, corrosive lead, low density non condensable helium, very high pressure water (water also brings criticality issues in a fast reactor). A clean fluoride coolant fixes all of these issues (there is still the high melting point but with a buffer salt pool design that should be managable) and is also transparent so that's very good.

I see that you favor a fast spectrum reactor design with an high power density, so you don't need flibe.

The fast reactors have also other safety issues that nuclear opponents denounce a lot, I speak about reactivity accidents (core compaction and deformation, insertion of moderator, prompt criticallity with fast neutrons, recriticality with a molten core, etc ...).

In general a fast reactor has also a very high power density with little coolant-to-fuel volume ratio. That may bring problem with cooling. High pressure drops limit natural circulation. A clog can also more easily provoke partial meltdown. Decreased coolant flow can lead to meltdowns even if the coolant is still in the core ( that happened in several fast reactors like Fermi 1, the sodium reactor experiment, the EBR1, the LFR of the K27 submarine, ... ).

In the beginning of the thread you spoke about an epithermal design which uses flibe.

I know that criticality excursions and core meltdowns are managable but maybe we can go back at your first idea and imagine designs sufficiently moderated to avoid or diminish these issues, that's may be good for convincing the public, the regulators and for reliability. But I think we will need flibe.

The Shippingport PWR reactor achieved breeding with light water and a 5 years core life. Flibe captures much less neutrons than light water and it also moderates much less than light water. So I guess that we can have comparable neutronic performances than the Shippingport reactor with an higher coolant-to-fuel volume ratio for "aerate" the core and improve cooling and still have a good negative void and high conversion ratio (not necessarily breeding). The fast fission bonus of the big metal fuel rods also improves neutronics and may facilitate reprocessing compared to the oxyde fuel of the Shippingport reactor. SiC is also better than zircalloy and we don't need a lot of it with big fuel rod and metal fuel has high HM density.
We can control the reactor by the same methods of the shiginpport reactor, or by other means like moderation and reflection with graphite control rods (but better designed than the Chernobyl bars :D ). We can maybe also use absobers rods made of thorium ( it's a modification of the shiginpport design). B4C neutron absorbing rods are used only for shut down and we use fertile blankets.

I guess that the epithermal spectrum of this concept gives an high doppler effect. Surprisingly the thermal dilatation of the metal can maybe cause positive feedback effects with this concept because it rises the moderator-to-fuel ratio (same problem than with the MSBR, but I guess the effect is much lower since the dilatation of the solid metal is lower than the molten salt and Flibe is a bad moderator compared to graphite). Anyway, the doppler is high so that should not be an issue.

The idea is to have high conversion ratio (possibly isobreeding) with a solid fuel reactor with not very high power density, a good doppler and negative void coefficients and no very hard spectrum ( for decreasing the issues raised before with fast spectrum). It looks like a PWR but with high conversion ratio, low pressure and high temperature.


Top
 Profile  
 
PostPosted: Nov 03, 2014 8:18 pm 
Offline

Joined: Apr 19, 2008 1:06 am
Posts: 2225
In other threads, a heavy water moderated MSR has been suggested
viewtopic.php?f=50&t=4344&p=57784&hilit=heavy+water+moderated+MSR#p57784
The arrangement can be used for any reactor with a liquid fuel and a liquid moderator or coolant.
Alternatively, a hydrocarbon simulant for the FLiBe (viewtopic.php?f=50&t=4398) in the AHTR can be used as the real thing but providing a measure of moderation too in any solid fuel reactor. It is a less volatile substitute for water like the molten salt, enabling a lower pressure intermediate circuit.


Top
 Profile  
 
PostPosted: Nov 04, 2014 3:23 am 
Offline

Joined: Jul 14, 2008 3:12 pm
Posts: 5050
Quote:
I see that you favor a fast spectrum reactor design with an high power density, so you don't need flibe.


Not really, but I like the idea of using the coolant as the moderator a la LWR. This means fastish spectrum as the fluorides aren't good moderators.

Quote:
I prefer a clean plant and a vessel protected from neutron flux than having fission products and actinides moving everywhere, but I am not an expert


That is one of the advantages of the concept, yes. Naturally many MSR advocates will disagree.

Quote:
In general a fast reactor has also a very high power density with little coolant-to-fuel volume ratio. That may bring problem with cooling.


The idea is to use a high power density core but use a pool type reactor so a lot of non volatile fluoride coolant on top of the core.

Quote:
High pressure drops limit natural circulation.


True, one of the limits of this design is the transient cladding and fuel heatup during loss of flow accidents. Experience with sodium cooled reactors shows very high power density is possible, likely a bit lower for fluoride salts because of the higher viscosity and lower thermal conductivity. Power density would probably be a bit lower than PWR.

I'm thinking the pumps would need flywheels for inertia and provide initial flow to hold transient temperature down. After that very little circulation is needed. About 4%. Pressure drop would then be below 0.2% of normal pumped pressure drop, or better said we need only 0.2% nat circ drive for the same core delta T. If the cladding can stand higher delta T in transients it gets even better, higher power density would be possible.

I've looked into pancake cores for this concept, but it doesn't work so well because of the thermal limits on the cladding, unless the fuel is not monolithic.


Top
 Profile  
 
PostPosted: Nov 04, 2014 3:32 am 
Offline

Joined: Jul 14, 2008 3:12 pm
Posts: 5050
Quote:
I know that criticality excursions and core meltdowns are managable but maybe we can go back at your first idea and imagine designs sufficiently moderated to avoid or diminish these issues, that's may be good for convincing the public, the regulators and for reliability. But I think we will need flibe.


I think NaF-BeF2 is good enough if Li7 is unavailable. Be and F are good enough and Na is not a bad poison for epithermal/fastish spectrum (no bad resonances).

Quote:
The idea is to have high conversion ratio (possibly isobreeding) with a solid fuel reactor with not very high power density, a good doppler and negative void coefficients and no very hard spectrum ( for decreasing the issues raised before with fast spectrum). It looks like a PWR but with high conversion ratio, low pressure and high temperature.


Yes, well said, it is basically the molten salt version of RBWR.


Top
 Profile  
 
PostPosted: Nov 04, 2014 11:38 am 
Offline

Joined: Oct 28, 2013 12:24 am
Posts: 256
Thanks for the response Cyril.

Quote:
That is one of the advantages of the concept, yes. Naturally many MSR advocates will disagree.


It's maybe just a matter of personal feeling or preference. There are a lot of very brillant guys who work/worked on the MSRs and I am just a student so I won't say that molten salt cooled reactors is a superior concept, everyone has his own idea and I also like MSRs. If I had to choose an MSR I would take the IMSR (integral design and no online reprocessing) but even having only fresh gazeous fission products moving around seems a difficulty. With solid fuel the plants are a lot simpler ( and safer in my mind) and cleaner. Maintenance and repair seems a lot easier too, that increases availability. Decomissioning seems a lot easier too if all is kept clean. The vessel, pipes and heat exchangers don't suffer from the neutron flux and last longer.

The difficulties of reprocessing are transferred in a few large reprocessing facilities, with economy of scale, decoupled from the power production, and the spent fuel has already decay some years. If you really can reprocess metal fuel by simply boiling off the fission products, that's easier and can be economical.


The lisencing of the ESBWR took 8 years, the lisencing of the LFTR will take 1 000 000 years :D
I know that most of the people involved in MSR's developpement thinks about low power, modular reactors, with frequent replacement and short life time. They are right in today economic environment which hates high capital costs and long term but I thing this is not how nuclear power do his best. I think nuclear power do his best with very high power and very long plant lifetime; but I am really bad in economics, so my opinion on this subject has no great value. We must design plant with very long life time so protect the vessel from corrosion and neutron flux and design things in order to be able to easily replace and repair all the components of the plant.


Quote:
The idea is to use a high power density core but use a pool type reactor so a lot of non volatile fluoride coolant on top of the core.


Yes with non volatile coolant and pool design the loss of coolant can not happen but it is the loss of flow that concerns me. As I said before we already have a lot of meltdown with sodium reactors, and fluoride salts are much more viscous than sodium. So that's why I thought of decreasing the power density and increasing the distance between the fuel rods for «ventilate» the core. Flibe has low neutron absorption and low moderation, it seems perfect for this : you don' have a lot of moderation but you still have a lot of coolant in the core and you still has enough moderation to have a negative void without using weird geometry ( and so you avoid reactivity issues with core compaction, fast neutrons prompt criticallity, recriticallity with core melt, etc …). Using big metal fuel rods also helps to increase the distance between the fuel rods and lower pressure drops. I don't know if this concept is possible and isobreeding will may be impossible but you still have an high conversion ratio if you care about it (not having neutron poisons and neutrons absorbing rods except for shutdown, having fertile blankets, …), the success of the experience of the shigginport reactor is encouraging.


I was thinking about using a loop design immerged in a pool of cheap buffer salt as you already think about it in other threads. It has some advantages over the classical pool designs. Flibe is costly so you economise Flibe. I think it is easier to use this concept with a molten salt cooled design rather than with an MSR. (no wondering about activation of buffer salt, graphite replacing, drain tanks and other things). Just use some PRACS loops with Flibe to transfer the heat in the buffer salt like the AHTR. Suppress the DRACS and use your cool-to-containment design (except that you need only one containment since it is a solid fuel reactor). Use a great amount of buffer salt and put the spent fuel pool in the buffer salt pool.

This design is also good in ATWS transients : contrary to most pool designs you don't need to heat enormous amount of coolant to kill the reactivity. And your buffer salt stay cold. This was discussed in some papers discussing about various AHTR designs. Flibe is costly but it can improve economics in the long term (economy of natural ressources) and the design need Flibe to avoid positive void. In a multi billions dollars power plant which last 80 years, the cost of Flibe is maybe not a big deal.


Top
 Profile  
 
PostPosted: Nov 04, 2014 2:32 pm 
Offline

Joined: Jul 14, 2008 3:12 pm
Posts: 5050
Very well said Fab. Not much to add to that but Amen and good night. You may be a student but clearly know more than some experts on the subject.


Top
 Profile  
 
Display posts from previous:  Sort by  
Post new topic Reply to topic  [ 120 posts ]  Go to page Previous  1 ... 4, 5, 6, 7, 8  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:  
cron
Powered by phpBB® Forum Software © phpBB Group