Energy From Thorium Discussion Forum

It is currently Jun 23, 2018 5:30 am

All times are UTC - 6 hours [ DST ]




Post new topic Reply to topic  [ 43 posts ]  Go to page 1, 2, 3  Next
Author Message
PostPosted: Oct 20, 2011 1:16 pm 
Offline

Joined: Mar 07, 2007 11:02 am
Posts: 912
Location: Ottawa
I'm surprised no one has mentioned these two new very important reports from Oak Ridge. I guess I've been a bit greedy in keeping them to myself for the last week. Both are full of extremely useful information, especially the economic study. I'm sure Jess Gehin from ORNL can also chime in with some comments but I should mention that ORNL has always been extremely conservative in their reporting of data and it is certainly true here. The values in the economic study are not meant to be definitive answers but merely a starting point. In most cases I think reality will be much better (in one or two perhaps worse though). A good example is their assumed thermal efficiency using coal plant data and using world averages instead of best cases (I'm sure they could do a couple percentage points higher).

http://info.ornl.gov/sites/publications/files/Pub32467.pdf

http://info.ornl.gov/sites/publications/files/Pub32466.pdf

While I hesitate to be negative as I know many of the excellent people involved in this study, it is pretty obvious there are some significant shortcomings in the AHTR approach. Most glaringly is the very large uranium utilization for their base case (and not that much better in their best case that shuts down to refuel far more often). Price per kw shows a modest improvement over a PWR but with a great many unknowns.

Anyhow, I'll just start things off with a few quick highlights

In a move I agree with, they are now looking at using a steam cycle to be more conservative for at least the first offering. As mentioned, I think their real efficiencies can be even higher which really would improve their overall case.

Needs almost 500 tonnes Uranium per GWe year (best case with frequent refueling is about half this but still more than a PWR). Section 8.5.2 onwards of the economic study covers this and there are a great many technical details in the other report. This of course assumes no reprocessing of the solid fuel (which would be much harder to do than LWR fuel). Their base case ends up with a fuel cycle cost of almost 2 cents/kwh which is not much better than coal.

Needs almost 3000 tonnes of Flibe coolant salt for 1500 MWe or about 1000 m3 of salt per GWe! This is about 20 times what the MSBR needed and still 10 times more than even the large DMSR core. They estimate 300 million dollars using ORNL's old estimate of 120$/kg of enriched Li7 and adding in inflation (no one really has any idea what Li7 will cost, I think it could be much higher). Their price of Beryllium though could be a lot lower since BeF2 is much cheaper than you'd think (the most expensive step of making Be metal is reducing BeF2 with Magnesium metal, if you skip this the cost is only a fraction of what they estimate). And a reminder that the salt "cooled" approach can only use Flibe and to 99.993% Li7 at least. The salt "fueled" approach has the luxury to use less expensive and more available salts.

All in all I think I am liking the pebble bed approach of Per Peterson more now. With pebbles they can get the uranium utilization much better (they claim somewhat better than PWR on a once through cycle). However, the pebble approach seems to me to have a longer list of unknowns and technical challenges. I know the overall wisdom of both approaches seems to be this conservative first step using molten salts just as coolants and then maybe later the bigger prize of molten salt fueled but the more I learn the more I prefer true molten salt fueled approaches.

David LeBlanc

P.S. The smaller core for the SmAHTR version was released awhile back and discussed in the thread below. This approach also has poor uranium utilization but adds a whole range of unique features that I really like.

http://energyfromthorium.com/forum/viewtopic.php?f=8&t=2873&st=0&sk=t&sd=a


Top
 Profile  
 
PostPosted: Oct 20, 2011 2:51 pm 
Offline

Joined: Jul 14, 2008 3:12 pm
Posts: 5053
Thanks David for those references and for pointing out some of the issues.

I've been looking at the pros and cons of the LFTR and the AHTR. Based on that, I've suggested a solid metal-fuelled salt-cooled triplex SiC cladding fuel-rod design. Like an IFR but with fluoride coolant. That way the problems with overmoderated graphite are avoided and more salts can be used, in particular NaF-BeF2 is on my mind. This should have much better uranium usage and negative void coefficients like PWRs. MIT is currently testing the triplex SiC cladding and wants to use it for PWRs.

This reactor, is somewhere in between the LFTR and the AHTR basically. Solid fuel, but metallic making reprocessing easier and getting a big fast fission bonus (Jaro's Bi-Modal suggestion). Boil off the fission products in a simple still once a year or two, then cast the liquid actinide still bottom goo back into new SiC fuel rods. Lead as backfill in stead of helium allows loose tolerances to accomodate fuel swelling (not much swelling for thorium metal though).

Pool type design may allow online shuffeling of the fuel assemblies, using the hook grappler crane thing that the LWRs use for refuelling offline (except that it would be high temperature and online refuelling). That could make a Pu-Th metallic fuel work (let Pa decay by placing the fuel assembly in the perifery). Or just uranium metal using LEU for a first reactor.

The fuel rods could be open topped - vented fuel the nuclear engineers tell me its called. With filters in the plenum to absorb iodine and cesium but lets xenon and krypton of the active fuel region.

Not having fuel in the fuel salt means less constraints on HX size and no neutrons in the HX. And no scary pumping of fission products.

There's a similar design out there, but using a hydride fuel that I think doesn't actually work for this application.

https://www.ornl.gov/fhr/presentations/Feng.pdf


Top
 Profile  
 
PostPosted: Oct 21, 2011 4:13 am 
Offline

Joined: Jul 14, 2008 3:12 pm
Posts: 5053
Wow, from table 4 in the first report, that AHTR uses 6500 kg of U235 as startup. The AP1000 uses 3764 kg U235 startup.

So they have higher fuel fabrication cost than the AP1000, use 72% more fissile startup at a very high enrichment, and need extreme amounts of Li7 at very high purity. The front end fuel cycle costs are 216 million per year versus 50 million per year for the PWR. FLiBe at 300 million but it might as well be 600 million. The actual overnight cost is higher than the PWR!

They talk about lower enrichments allowing lower fuel cycle costs. How is this possible if they state they need 99.995% Li7 to go critical on 19.75% enrichment? They're pushing the particle loading fraction already. Can you get this with shorter fuel cycle length without increasing fuel cycle costs?

The pebbles start to look pretty good now. Cycling of pebbles seems to allow blankets that also protect the reactor internals. The pebbles are much cheaper to manufacture as they are all the same little balls with no complex reactor internals. Ehud Greenspan says the cost can be slightly lower than PWR fuel for deep burn pebbles, about 10-20% lower, and capital cost of PB-AHTR also roughly 30% lower:

http://gcep.stanford.edu/pdfs/UVaodfDrA ... rkshop.pdf


Top
 Profile  
 
PostPosted: Oct 21, 2011 7:27 am 
Offline

Joined: Mar 07, 2007 11:02 am
Posts: 912
Location: Ottawa
Quote:
They talk about lower enrichments allowing lower fuel cycle costs. How is this possible if they state they need 99.995% Li7 to go critical on 19.75% enrichment? They're pushing the particle loading fraction already. Can you get this with shorter fuel cycle length without increasing fuel cycle costs?


Those two numbers are actually independent of each other. They only really need 99.995% Li7 for safety reasons to assure a negative void coefficient. If they use as high an enrichment (19.75) as possible they can go to the longest cycle time between refueling (2 years) which is attractive for plant availability. This gives them the worst uranium usage though, I think the lowest they looked at was about 9% enrichment and then they needed a much quicker fuel cycle. They even looked into Pseudo Continuous fuel cycles where they think they might be able to shuffle a few fuel elements so quickly they could go off line and back online before Xenon poisoning kicks in. There best economic numbers for a fuel cycle are indeed with much quicker fuel cycles.

If I recall the pebble bed or PB-AHTR needs about 8% enrichment so not too much room for improvement. I think it stays pretty high mainly because they can only jam in so much fuel into the graphite matrix so they have to squeeze in more U235 with the enclosed uranium.

David L.


Top
 Profile  
 
PostPosted: Oct 21, 2011 8:12 am 
Offline

Joined: Jul 14, 2008 3:12 pm
Posts: 5053
The AHTR basically trades resource utilization for proven fuel and high safety (containment up to 1600 Celcius).

SiC fuel rods with metallic fuel would be a lot more interesting but would mean unproven fuel. If MIT succesfully develops the SiC clad for PWRs it could get very interesting.


Top
 Profile  
 
PostPosted: Oct 21, 2011 2:01 pm 
Offline

Joined: Apr 01, 2008 6:15 pm
Posts: 144
Location: Oklahoma
The lousy uranium utilization is quite a puzzle. They blame the low max uranium loading, but don't explain why their performance is so much worse than that of the pebble-bed AHTR (both use a similar carbon to heavy metal ratio of around 400). Shouldn't the two systems be nearly the same?

For the 19.5% enriched fuel, at discharge, it still contains an enrichment of 13.1% :shock: with 8% of the initial metal fissioned.

Could this be just a variant of a grander recycling based system? Wouldn't adding thorium help?

_________________
Nathan Wilson, MSEE


Top
 Profile  
 
PostPosted: Oct 21, 2011 3:00 pm 
Offline
User avatar

Joined: Oct 06, 2010 9:12 pm
Posts: 138
Location: Cleveland, OH
I think that solid fuel would benefit from the annular style rather than a solid cylinder. Since the cooling occurs on both the OD and the ID, the centerline peak temperature is much lower, and I think the temp slope is less steep, IIRC. Thus the thermal stresses from fuel expansion are less, as is the stresses from the volatile FPs.

MIT has been researching the (oxide) annular style for our standard-fleet LWRs. This would allow a greater heat transfer and thus uprating the plant as much as 50%. Or the plant could operate at its current rating and fuel damage/cladding failure would be greatly reduced.

These advantages should apply to metal fuel, too, since they are due to geometry.


Top
 Profile  
 
PostPosted: Oct 21, 2011 4:38 pm 
Offline

Joined: Jul 14, 2008 3:12 pm
Posts: 5053
Nathan2go wrote:
The lousy uranium utilization is quite a puzzle. They blame the low max uranium loading, but don't explain why their performance is so much worse than that of the pebble-bed AHTR (both use a similar carbon to heavy metal ratio of around 400). Shouldn't the two systems be nearly the same?

For the 19.5% enriched fuel, at discharge, it still contains an enrichment of 13.1% :shock: with 8% of the initial metal fissioned.

Could this be just a variant of a grander recycling based system? Wouldn't adding thorium help?


With pebbles, you can shuffle the fuel around online by removing them from high flux channels and putting them back in lower flux channels. This allows higher burnup for a given actinide loading. Works especially well for thorium blankets, cycling out pebbles to let Pa decay and putting them back into the core. With ThO2 fertile pebbles in a blanket pebble zone that also protects the reactor and internals from neutron damage.

Anything with a lot of fissile is going to have less fertile and that hurts breeding. Any reactor with no frequent shuffeling of fuel will be a poor resource performer.


Top
 Profile  
 
PostPosted: Oct 21, 2011 4:45 pm 
Offline

Joined: Jul 14, 2008 3:12 pm
Posts: 5053
Jim L. wrote:
I think that solid fuel would benefit from the annular style rather than a solid cylinder. Since the cooling occurs on both the OD and the ID, the centerline peak temperature is much lower, and I think the temp slope is less steep, IIRC. Thus the thermal stresses from fuel expansion are less, as is the stresses from the volatile FPs.

MIT has been researching the (oxide) annular style for our standard-fleet LWRs. This would allow a greater heat transfer and thus uprating the plant as much as 50%. Or the plant could operate at its current rating and fuel damage/cladding failure would be greatly reduced.

These advantages should apply to metal fuel, too, since they are due to geometry.


Annular fuel compacts are one of the more traditional designs of TRISO (but so far only for gas-cooled reactors). They're easy to make. With cladding it is harder since you need double ring seal welds rather than single. But still pretty simple. MIT's annular fuel is really promising for uprates of LWRs.

Metal fuel will benefit less since it already has such good thermal conductivity. And it increases with temperature. Still it makes sense to have a central void channel to limit fission gas internal stress buildup. Very important for vented metal fuel. I was thinking to use this to remove fission gas online through open topped fuel rod plenum carbon filters in a traditional fuel rod AHTR.


Top
 Profile  
 
PostPosted: Oct 21, 2011 4:54 pm 
Offline
User avatar

Joined: Nov 30, 2006 9:18 pm
Posts: 1947
Location: Montreal
Cyril R wrote:
Any reactor with no frequent shuffeling of fuel will be a poor resource performer.
....as confirmed by the remarkable performance of Candu reactors, with their on-line fuel shuffling robotic machines.

....much easier to do with molten salt fuel though.


Top
 Profile  
 
PostPosted: Oct 24, 2011 1:32 am 
Offline

Joined: Apr 19, 2008 1:06 am
Posts: 2230
Metal Thorium has two useful physical qualities.
1. A high melting point of 2115K.
2. A good thermal conductivity of 54 W /m/K.
Th-20% LEU composite can have a high burn up life of 6-9 years.
Metallic thorium will be very good for thermally conductive bonding of fissile (LEU) and moderator in a pebble/pin type fuel while doubling as fertile fuel. Coolant compatible cladding such as SiC could be provided.
PS. Th-20% LEU has good thorium utilization too.


Top
 Profile  
 
PostPosted: Oct 24, 2011 4:30 am 
Offline

Joined: Jul 14, 2008 3:12 pm
Posts: 5053
Th-Pu metallic fuel should do even better, as reactor grade Pu is much higher fissile than LEU (60% versus 20%) and you have no troublesome U238 to ruin your chances of isobreeding.

Metallic fuel might be very easy to reprocess, as well. The actinides all have extremely high boiling points, so most of the fission products, including the worst of the rare earths (except neodymium) are removed by boiling them off.


Top
 Profile  
 
PostPosted: Oct 25, 2011 8:19 am 
Offline

Joined: Jul 14, 2008 3:12 pm
Posts: 5053
One of the studies of the PB-AHTR looks at the effect of fuel shuffeling. Compared to a non-shuffeled pebble bed, an optimized shuffled bed gets about 45% more burnup (915 versus 1317 EFPD), under the same fissile loading. So that gets a correspondingly smaller uranium requirement.

http://www.nuc.berkeley.edu/pb-ahtr/pap ... Report.pdf


Top
 Profile  
 
PostPosted: Sep 10, 2012 1:04 pm 
Offline

Joined: Jul 14, 2008 3:12 pm
Posts: 5053
Hmm, I was just looking at the economic study again, and a couple results really jumped out that we missed before.

First off, the AHTR has higher capital costs than a PWR! The TCIC of the PWR is 4590 million. The TCIC of the AHTR with 9% enrichment is 4818 million. The reason the AHTR's levelised capital costs are lower are due to the improved efficiency, not due to a cheaper reactor! This is an interesting result. It means any reactor that can use the supercritical steam turbine will have that advantage. And a reactor that can combine lower capital costs with that higher efficiency, would strike gold.

The other surprising result is that the cost of enrichment dwarfs the cost of the fuel fabrication: it makes sense to go for a lower enrichment and swap out fuel more often (ie fabricate more fuel). That's a surprise to me. Though it should be said the fuel fabrication cost estimate for the TRISO fuel is probably not realistic, as it is based on the cost structure of PWR fuel.


Top
 Profile  
 
PostPosted: Nov 10, 2012 6:49 am 
Offline

Joined: Jul 14, 2008 3:12 pm
Posts: 5053
Ok guys, there´s a new study available from ORNL on the AHTR.

ORNL/TM-2012/320, AHTR Mechanical, Structural, and Neutronic Preconceptual Design

Downloadable from the osti-bridge site.

http://www.osti.gov/bridge/servlets/purl/1054145/

Contains TONS of information, much of it new stuff and much is applicable to LFTR design.


Top
 Profile  
 
Display posts from previous:  Sort by  
Post new topic Reply to topic  [ 43 posts ]  Go to page 1, 2, 3  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