Energy From Thorium Discussion Forum

It is currently Sep 22, 2018 11:31 am

All times are UTC - 6 hours [ DST ]




Post new topic Reply to topic  [ 117 posts ]  Go to page 1, 2, 3, 4, 5 ... 8  Next
Author Message
PostPosted: Sep 20, 2011 4:48 am 
Offline

Joined: Jul 14, 2008 3:12 pm
Posts: 5045
On this forum, the concept of a pool type molten salt reactor has been discussed a couple of times. This is a promising idea: so let's make a thread about it.

With solid fuelled pool type reactors, the core sits on the bottom of a bath of coolant. This won't do very well for us, because our fuel is in the coolant. We'd have to use too much fuel salt, and there would be huge contamination of the area above the pool surface.

But we can have a liquid buffer salt pool, something cheap, and put the vessel and heat exchangers all in that liquid buffer salt. The fuel salt would be contained in the vessel and heat exchangers.

This concept is being used by the TRISO fuelled molten salt cooled reactor, the AHTR:

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

This idea has many advantages: lots of thermal inertia inherently available to absorb heat in any scenario, dead weight of the heat exchangers and vessel is almost eliminated, major leaks are impossible as the hydrostatic/pumped pressure of the buffer and fuel salt will be designed to be almost identical, air or water ingress is impossible, and the buffer salt also shields against neutron and gamma radiation. In the unlikely event of containment breach or other beyond design basis accident, the buffer salt will retain all of the fuel salt.

I wonder if we can copy the AHTR design almost completely. Upon loss of cooling (complete power blackout like Fukushima) the tubes connected to the vessel will naturally start to circulate fuel salt and exchange this heat to the buffer salt. There are no moving parts, the system starts working automatically when pumped cooling stops, by using something clever called a fluidic diode.

This would work much more effectively for a molten salt reactor compared to the solid fuelled AHTR, because of the homogeneous fuel-coolant.

So now we’ve put the decay heat into the buffer salt. Good, but how to remove this decay heat from the buffer salt to the environment?

The AHTR uses another bunch of tubes dipped into the buffer salt. These can be heat pipes that move heat from the pool up into a chimney where passive air cooling moves it to the environment.

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

However, it might be simpler to skip the heat pipes or cooling tubes and in stead let the buffer pool salt surface radiate its heat into the hot cell, which is then passively cooled by a chimney. We could let the hot cell move out 1% of its heat constantly, that way we know it is always working and can’t be shut down. Another benefit of this method is that the hot cell can operate a bit colder than 500 degrees Celcius, as it is the heat sink for the buffer salt.

Another issue is the fact that the pool is at high temperature, probably over 500 degrees Celcius. And it will likely be below grade for safety. That means its going to heat up the ground next to it. Even with thick insulation the heat flux through the ground would be quite significant. So the AHTR uses a water cooled cavity liner. But I’d prefer to not use water, I’d rather use passive air cooling here as well. One possibility is to put air channels in the insulation that passively move hot air out.

Anoter issue I’d like to discuss is what salt to use for the buffer salt. A fluoride would be ideal, I think. 7Li and Be would be a bit on the pricey side, as we’re probably talking about a million liters or more of buffer salt.

Perhaps KF-ZrF4 (58-42)is a good one: cheap, low vapor pressure (1.2mm hg @ 900 C) very little neutron activation, and a 390 C melting point. If we process spent LWR fuel we’ll have lots of zirconium from the cladding that no one else will want due to its mild radioactivity.

A final issue that comes to mind is maintenance. If the equipment is all at the bottom of a fluoride salt pool, then things like welding could be difficult (though cutting would be fairly easy, using eg diamond saws).

I would like to hear what people think about this MSR design path.


Top
 Profile  
 
PostPosted: Sep 20, 2011 10:31 am 
Offline

Joined: Jul 28, 2008 10:44 pm
Posts: 3063
I'm not sure I understand why a DRACs. It is my impression that the purpose of the DRACs is to reduce heat losses during normal operations. I'm thinking this isn't terribly important for us. Heat loss means higher fuel consumption and higher chemical cleanup costs but not higher pumping, heat exchangers, turbines, generators etc. Fuel consumption is virtually free, and I don't think the chemical processing costs change much with a 1% increase in volume. The DRACs looks like something that would be a failure of the emergency cooling system if the pipe were knocked about during an earthquake.

I don't know much about under fluoride welding but that seems like a challenge to me that it might be best to avoid. I also don't like the thought of the pool salt getting inside the reactor. So I'd be inclined to think of maintenance being done above the pool salt - perhaps by lifting the drained reactor up or by pumping the pool salt out.

Given the temperatures we have we need some thermal buffer before we get to the concrete. Air works but then there exists a vessel bearing the weight of the reactor, fuel salt, blanket salt, heat exchangers, pumps, and pool salt. I expect this vessel would be our containment wall. Using a liquid here would allow the weight to be transferred to the concrete but makes it harder to achieve the appropriate temperature drop.

In general I like the idea.


Top
 Profile  
 
PostPosted: Sep 20, 2011 11:26 am 
Offline
User avatar

Joined: Nov 30, 2006 9:18 pm
Posts: 1946
Location: Montreal
I like the AHTR and related concepts.
There are always trade-offs with various layout modifications.
Cyril R wrote:
With solid fuelled pool type reactors, the core sits on the bottom of a bath of coolant. This won't do very well for us, because our fuel is in the coolant. We'd have to use too much fuel salt, and there would be huge contamination of the area above the pool surface.

Regarding "there would be huge contamination of the area above the pool surface", hot cells are, in the first instance, intended for use in processes where contamination results -- think of medical radioisotope separation from irradiated targets.

Of course its preferable to limit the degree and spread of contamination around the hot cell.
To do that, we can use a "fume hood" above the core, that sucks up and filters the hot cell air (argon ?).
There are excellent guides & handbooks on radioactive air filtering system design - so nothing very new here, except perhaps the higher operating temperature....

In any event, if we want to remove the volatile fission products, they will have to be bubbled out & collected somewhere: The hot cell fume hood filtration system could be used to perform that function, when combined with a volatiles condensing collection system.


Top
 Profile  
 
PostPosted: Sep 20, 2011 1:30 pm 
Offline

Joined: Jul 14, 2008 3:12 pm
Posts: 5045
Lars wrote:
I'm not sure I understand why a DRACs. It is my impression that the purpose of the DRACs is to reduce heat losses during normal operations. I'm thinking this isn't terribly important for us. Heat loss means higher fuel consumption and higher chemical cleanup costs but not higher pumping, heat exchangers, turbines, generators etc. Fuel consumption is virtually free, and I don't think the chemical processing costs change much with a 1% increase in volume. The DRACs looks like something that would be a failure of the emergency cooling system if the pipe were knocked about during an earthquake.


That is also what I was thinking. If we use a double containment with radiative heat transfer between the two containments then our problem will sooner be too little heat transfer than too little.

Lars wrote:
I don't know much about under fluoride welding but that seems like a challenge to me that it might be best to avoid. I also don't like the thought of the pool salt getting inside the reactor. So I'd be inclined to think of maintenance being done above the pool salt - perhaps by lifting the drained reactor up or by pumping the pool salt out.


Yes, pumping the pool salt out to an insulated salt tank was what I was thinking as well, but only for major maintenance (replacement of entire modules). The question is how much we will need this; for example PWR steam generators typically last 20-40 years. Maybe the pumps will need more frequent replacement, but if its not more than once every couple of years then pumping the pool salt out will be a great idea.

Lars wrote:
Given the temperatures we have we need some thermal buffer before we get to the concrete. Air works but then there exists a vessel bearing the weight of the reactor, fuel salt, blanket salt, heat exchangers, pumps, and pool salt. I expect this vessel would be our containment wall. Using a liquid here would allow the weight to be transferred to the concrete but makes it harder to achieve the appropriate temperature drop.


With thick ceramic insulation the heat flux to the outer concrete or metal structure will be very low. So you can have ceramic insulation bricks with wide channels in them that move hot air out passively. We just have to make sure that the heat flux to the surrounding soil is so low that it won't cause groundwater to boil and all that. Passive air cooling seems like a better idea than safety class water pumps.


Top
 Profile  
 
PostPosted: Sep 22, 2011 11:50 am 
Offline

Joined: Mar 07, 2007 11:02 am
Posts: 911
Location: Ottawa
I've been thinking this way as well. You've implied, but not really said, since this concept can use the surrounding pool salt as the heat sink it means getting rid of the decay heat dump tank and freeze plug. I think ultimately this might actually end up being a safer approach since draining takes time and no matter how many pipes and plugs we use, people will keep bringing up "what if the drain clogs?". The big counter though is we have to be sure we can kill reactivity if we are going to keep the fuel salt in the core. Doesn't sound too big a challenge though, simply having something that melts and either drops a shutdown rod or injects poison can be a passive backup to active systems.

David LeBlanc


Top
 Profile  
 
PostPosted: Sep 22, 2011 2:18 pm 
Offline

Joined: Jul 14, 2008 3:12 pm
Posts: 5045
David wrote:
I've been thinking this way as well. You've implied, but not really said, since this concept can use the surrounding pool salt as the heat sink it means getting rid of the decay heat dump tank and freeze plug. I think ultimately this might actually end up being a safer approach since draining takes time and no matter how many pipes and plugs we use, people will keep bringing up "what if the drain clogs?". The big counter though is we have to be sure we can kill reactivity if we are going to keep the fuel salt in the core. Doesn't sound too big a challenge though, simply having something that melts and either drops a shutdown rod or injects poison can be a passive backup to active systems.

David LeBlanc


That's exactly right David. A safety class dump tank wouldn't be necessary anymore. I've noticed that the AHTR has developed a solution to the reactivity problem (actually, there is no reactivity problem that can be damaging in itself, but there is the indirect peak temperature problem). They use bouyancy activated control rods.

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

No need to reinvent the well, I'd say. If Per Peterson can get his test reactor going then he's going to develop a lot of stuff that we can use.

Fukushima showed that control rods work very well to shut down the fission reaction. However one can also have a freeze plug with gadolinium fluoride behind it in a cavity. That way, if the control rods somehow don't work, the Pa233 decaying to U233 will add reactivity that will increase the temperature that will melt the plug and release gadolinium fluoride in the fuel salt.

A uranium-only DMSR would of course not have the Pa233 reactivity problem.

But, even if we play devil's advocate and assume that the reactivity problem destroys the primary fuel salt boundary (which does not appear to be physically possible) this will mix fuel salt with pool salt which will kill any reactivity. The passive containment cooling system wouldn't actually be affected by this; there will be a big mess, with completely ruined fuel and buffer salt, but no radiation will be released.


Top
 Profile  
 
PostPosted: Sep 22, 2011 9:11 pm 
Offline

Joined: Dec 03, 2008 5:23 pm
Posts: 137
Location: Oak Ridge, TN
We have considered a buffer salt in the AHTR designs. See for example, pages 4-5 of ORNL/TM-2006/140 http://nuclear.inl.gov/deliverables/docs/status_report_fy06_ornl-tm-2006-140.pdf. This design still has DRACS for decay heat removal. The use of DRACS for decay heat removal allows you to scale the reactor to the large sizes (2400MW) considered in the AHTR designs. For small reactors you can remove the heat through the vessel (which is what I think is being proposed in the previous posts).

The DRACS are included in as a fully independent decay heat removal system that can operate on natural circulation. If you lose power the primary circuit will not remove any heat. The DRACS are designed to remove 1% of the heat via natural circulation (no power as in station blackout). They would be seismically qualified and there we typically have three DRACS, only two of which need to operate.

Note that for heat removal through the vessel walls, that you will typically need a natural circulation system as well (RVACS). A better alternative would be conduction to the ground.


Top
 Profile  
 
PostPosted: Sep 23, 2011 4:43 am 
Offline

Joined: Jul 14, 2008 3:12 pm
Posts: 5045
Jess Gehin wrote:
We have considered a buffer salt in the AHTR designs. See for example, pages 4-5 of ORNL/TM-2006/140 http://nuclear.inl.gov/deliverables/docs/status_report_fy06_ornl-tm-2006-140.pdf. This design still has DRACS for decay heat removal. The use of DRACS for decay heat removal allows you to scale the reactor to the large sizes (2400MW) considered in the AHTR designs. For small reactors you can remove the heat through the vessel (which is what I think is being proposed in the previous posts).

The DRACS are included in as a fully independent decay heat removal system that can operate on natural circulation. If you lose power the primary circuit will not remove any heat. The DRACS are designed to remove 1% of the heat via natural circulation (no power as in station blackout). They would be seismically qualified and there we typically have three DRACS, only two of which need to operate.

Note that for heat removal through the vessel walls, that you will typically need a natural circulation system as well (RVACS). A better alternative would be conduction to the ground.


Hello again Jess. For the LFTR, I was thinking about using the hot cell as the ultimate heat sink (like the AP1000). This avoids the need for a below ground vessel cooling system. Just insulate the outside of the pool real good with ceramic insulation. Initial calculations with raytrace suggest its is quite easy to remove about 0.5-0.6% of the total heat using a double stainless steel containment with thermal radiation across the gap between the two containments. I'm trying to change the geometry so that 1% of the heat is transferred at equilibrium hot cell wall temperature of 500 degrees Celcius.

Conduction to the ground I think is almost impossible. The ground is such a good insulator if you take into account the thermal diffusion barrier that forms at equilibrium. I think we'll actually have trouble getting the heat flux low enough, even with 1 meter of space shuttle insulation bricks, the heat flux would not be trivial. We wouldn't want any ground water to boil. The AHTR seems to use a water cooled liner. I'd prefer to use a full passive air cooled system. I think this is possible with air channels in the outer insulation bricks.


Top
 Profile  
 
PostPosted: Sep 23, 2011 5:39 am 
Offline

Joined: Dec 26, 2007 11:45 am
Posts: 191
Blocks of a suitable salt can be stacked in the hot cell for emergencies. If it gets too hot they will melt and the rising molten salt will cover the bottom of the core vessel, dramatically increasing heat tranfer to the hot cell walls.

A liquid pool is pretty sure to create many headaches. The mechanical support it provides is nice but I'm not sure if it's worth the trouble. Thermal buffer can be provided by other means and if it's solid it will also use the heat of fusion.


Top
 Profile  
 
PostPosted: Sep 24, 2011 3:53 pm 
Offline

Joined: Jul 14, 2008 3:12 pm
Posts: 5045
Owen T wrote:
Blocks of a suitable salt can be stacked in the hot cell for emergencies. If it gets too hot they will melt and the rising molten salt will cover the bottom of the core vessel, dramatically increasing heat tranfer to the hot cell walls.

A liquid pool is pretty sure to create many headaches. The mechanical support it provides is nice but I'm not sure if it's worth the trouble. Thermal buffer can be provided by other means and if it's solid it will also use the heat of fusion.


This seemed like an interesting idea, as phase change gets higher heat dumping per volume, so I tried to put it in the raytrace model. It soon become apparent, that the heat transfer to the blocks would be low. Much too low to be of useful emergency response. Plus we've now added a thermal shock problem with liquid fluoride overflow in dry areas. From discussions on the site, this seems like something to avoid (e.g. no dry dump tanks, either).

The concept of keeping everything in the phase and location that it is in during normal operations, has become quite appealing to me. From a safety perspective this gets you the most reliable system with no dynamic failure modes whatsoever. From a regulatory perspective, it is the simplest and easiest to prove it will be good in all circumstances. The pool type reactor outlined in the AHTR, seems to fulfill this objective with very mild temperature rise.

I've been thinking a lot about this subject, and don't see any major issues with molten fluoride buffer pool submergence. I can see major advantages in thermal and radiation management, beyond design basis accident resistance, and in design simplicity. Inspection and minor maintenance can be done online with simple tools. Major maintenance will need shutdown anyway (we'll need this for the turbine-generator even if for nothing else). And it's not a big deal to pump the clean buffer salt into an insulated tank. We'll need to have such surge or drain tanks anyway, to store the secondary loop liquid during off-line maintenance.


Top
 Profile  
 
PostPosted: Sep 25, 2011 1:51 am 
Offline

Joined: Apr 19, 2008 1:06 am
Posts: 2240
A good point in the AHTR design is that fuel stays in the reactor vessel and only "clean" salt goes out to transport the heat. Using the liquid fuel as coolant may produce more problems than advantages. A jacket of clean coolant outside the core may be a useful arrangement. The fuel in the core could be conveniently freed of volatile (fission product) neutron poisons.
The blanket could located be as a solid lattice immersed in the coolant. If a thermal spectrum is desired, the moderator could also be solid blocks immersed in the fuel pool. The geometry of moderator blocks could help decide the desired volume proportion of fuel and moderator. The liquid fuel should be used primarily for its main advantage, easy removal of gases.


Top
 Profile  
 
PostPosted: Sep 25, 2011 6:21 am 
Offline

Joined: Jul 14, 2008 3:12 pm
Posts: 5045
jagdish wrote:
A good point in the AHTR design is that fuel stays in the reactor vessel and only "clean" salt goes out to transport the heat. Using the liquid fuel as coolant may produce more problems than advantages. A jacket of clean coolant outside the core may be a useful arrangement. The fuel in the core could be conveniently freed of volatile (fission product) neutron poisons.
The blanket could located be as a solid lattice immersed in the coolant. If a thermal spectrum is desired, the moderator could also be solid blocks immersed in the fuel pool. The geometry of moderator blocks could help decide the desired volume proportion of fuel and moderator. The liquid fuel should be used primarily for its main advantage, easy removal of gases.


Agree it would be a major advantage to not have to pump fission products through heat exchangers. But jackets around the core won't remove enough heat to work effectively. IF we can pump fission products like the MSRE proved is not too hard technically, then we get the best thermal management. I'm pretty sure that the AHTR will be there first and if even a prototype of the AHTR is built then we can use much of the design and equipment.

Keep in mind though that once you remove the xenon, you get a bunch of cesium-137 in the off-gas system and you'll have to deal with it outside the core. So you have that fission product to worry about even if you don't have to pump fission products in the reactor itself. Also keep in mind that if you push fluoride coolant through the core the fluorine will get short term activation so you have that radiation to design in as well.

Having said those things, I think an interesting variant of the AHTR is to use SiC fuel rods with metallic thorium-plutonium fuel and NaF-BeF2 coolant. This allows lower fuel fabrication cost, better fuel dilatation coefficients, better void coefficients, better fast fission bonus, and easier reprocessing. Those five things are major trade-offs for the AHTR's focus on avoiding any sort of fission product contamination.

So far most reactor concept have too strong a focus on one or two aspects, so they score badly on others. What we need is a daisy reactor that scores high marks in all fields (sustainable fuel cycle/resource efficiency, waste, safety, economics, proliferation, etc.). The molten salt reactor architecture seems to offer a variety of solutions that score well across all criteria.


Top
 Profile  
 
PostPosted: Sep 25, 2011 2:05 pm 
Offline

Joined: Dec 26, 2007 11:45 am
Posts: 191
Cyril R wrote:
Owen T wrote:
Blocks of a suitable salt can be stacked in the hot cell for emergencies. If it gets too hot they will melt and the rising molten salt will cover the bottom of the core vessel, dramatically increasing heat tranfer to the hot cell walls.

A liquid pool is pretty sure to create many headaches. The mechanical support it provides is nice but I'm not sure if it's worth the trouble. Thermal buffer can be provided by other means and if it's solid it will also use the heat of fusion.

This seemed like an interesting idea, as phase change gets higher heat dumping per volume, so I tried to put it in the raytrace model. It soon become apparent, that the heat transfer to the blocks would be low. Much too low to be of useful emergency response. Plus we've now added a thermal shock problem with liquid fluoride overflow in dry areas. From discussions on the site, this seems like something to avoid (e.g. no dry dump tanks, either).

In other words, the most it can do is to be the equivalent of a core catcher, not anything that is useful under milder circumstances.
Quote:
The concept of keeping everything in the phase and location that it is in during normal operations, has become quite appealing to me. From a safety perspective this gets you the most reliable system with no dynamic failure modes whatsoever. From a regulatory perspective, it is the simplest and easiest to prove it will be good in all circumstances.

It's not just about proving things to regulators - it's about proving things to YOURSELF. Just like a business plan is not just for pleasing investors...
Quote:
I've been thinking a lot about this subject, and don't see any major issues with molten fluoride buffer pool submergence.

There have been some proposed designs that are not actually pressure tight and rely on gravity. Seems like an interesting way to handle some of the issues with graphite changing size and construction materials having zero ductility. You can't do any of that in a submerged reactor, obviously.


Top
 Profile  
 
PostPosted: Sep 25, 2011 3:40 pm 
Offline

Joined: Jul 14, 2008 3:12 pm
Posts: 5045
Owen T wrote:
Quote:
There have been some proposed designs that are not actually pressure tight and rely on gravity. Seems like an interesting way to handle some of the issues with graphite changing size and construction materials having zero ductility. You can't do any of that in a submerged reactor, obviously.


You mean Jaro's HW-MSR, right? That indeed wouldn't need a pool type reactor and would indeed be disadvantaged by it. It would have to be dry, and due to the high temperature would not have much trouble radiating the heat to the passively cooled hot cell containment, using the cooling coils as heat pipes that stick their condensing ends out as radiators to the hot cell walls. Of course if those pipes were to somehow fail you've got a more serious situation, with very little thermal buffering to soak up heat, the fuel salt will spill out and quickly boil. This is going to be very messy with fuel and heavy water vapor going everywhere in the hot cell and the said vapors could possibly challenge the containment as they would condense on its walls. It is an extremely unlikely scenario but not impossible (they are after all pipes with coolant in it that can break).

Pool type isn't a one size fits all solution. But a big graphite moderated core like a DMSR would benefit greatly from pool type design.


Top
 Profile  
 
PostPosted: Sep 26, 2011 1:35 am 
Offline

Joined: Apr 19, 2008 1:06 am
Posts: 2240
Heavy water, though a very good choice at lower temperature, is not the only option available for moderator. Solid pebbles of Be2C in Be matrix, could be immersed in the fuel core. Graphite also has alternatives as solid moderator. The moderator does not have to double as coolant and move to transport the heat. The heat transport could be either from a jacket of clean sat or other coolant outside the core or coils inside the core.


Top
 Profile  
 
Display posts from previous:  Sort by  
Post new topic Reply to topic  [ 117 posts ]  Go to page 1, 2, 3, 4, 5 ... 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:  
Powered by phpBB® Forum Software © phpBB Group