Energy From Thorium Discussion Forum

It is currently Sep 20, 2017 4:07 am

All times are UTC - 6 hours [ DST ]




Post new topic Reply to topic  [ 30 posts ]  Go to page 1, 2  Next
Author Message
PostPosted: Sep 28, 2013 3:08 pm 
Offline

Joined: Jul 14, 2008 3:12 pm
Posts: 5048
Helium cooled pebble bed reactors like the former South African PBMR project, are often pitched as being "inherently safe". However, that appears to be the case only if the control rods work. Without the control rods working, a loss of the pumps would result in most of the fuel being damaged! And the reactor vessel is likely to fail as well.

Quote:
4.1 P-LOFC: The reference case P-LOFC for the PBMR is similar to the corresponding GT-MHR
accident., with a peak fuel temperature of 1266°C occurring at ~37 hours, with a maximum
reactor vessel temperature of 501°C at 77 hr. Sensitivities to variations in the emissivities of the
vessel and RCCS are nearly identical to those for the GT-MHR.

4.2 D-LOFC: In the D-LOFC reference case “conduction-heatup” accident, T(fuel)-max peaks at
1517°C 77 hr into the accident, and for this configuration, maximum temperatures for the reactor
vessel (SA 508) and core barrel (316 SS) are not of concern.

4.4 P-LOFC with ATWS: In this PBMR design, recriticality occurs at about 28 hours, and T(fuel)-
max reaches 2127°C at 103 hr. Maximum vessel temperatures are also higher, 711°C at 145 hr.
Fuel failure after 7 days was 57%. Variations in this accident are sensitive to fuel and moderator
temperature-reactivity feedback coefficients. As with the GT-MHR, if after recriticality the SCS
is started (with still no scram), peak fuel temperatures would exceed limits even more due to the
selective undercooling.

4.5 D-LOFC with ATWS: Recriticality occurs at 31 hr. In this case, T(fuel)-max is 2166°C at 137
hr, and the maximum vessel temperature (496°C at time = 168 hr) was still rising slowly after a
week. Fuel failure at the end of the week was 59%.


P-LOFC = loss of forced circulation with primary loop remaining at pressure.
D-LOFC = loff of forced circulation with primary loop failed (depressurized).
ATWS = anticipated transient without scram. (control rod or logic/electronic failure).

This is pretty serious. I'd thought the pebble bed helium cooled reactors were inherently safe. Clearly such statements are too bold.


Attachments:
MHTGR120911.pdf [416.94 KiB]
Downloaded 115 times
Top
 Profile  
 
PostPosted: Dec 06, 2013 11:29 pm 
Offline

Joined: Oct 28, 2013 12:24 am
Posts: 256
I believe that I read in different places that thermal gas reactors (like pebble bed reactors or the GT-MHR) can shut down the chain reaction even if control rods failed to insert in core and next remove their decay heat just by thermal conduction in the concrete of the reactor, even if helium was lost, without any operator action and for an indefinite period of time.

That why they were called "inherently safe" and the safest of all power nuclear reactors. So it's maybe true for small reactors but not for the reactors studied here ( 400 MW(t) and 600 MW(t) ).

It seems that I had it all wrong.

First it seems that for a thermal power of four hundred of megawatts they can not just use thermal conduction through the concrete and walls to remove the decay heat. The thermal conductivity of concrete is too low for this power. They need a reactor cavity cooling system to cool the vessel. So a system with natural circulation of air or water. So you can still have pipe breaks, leaks, and clogs (and dryout for water). These events are very unlikely but still possible.

Secondly it seems that even if passive cooling works and if helium pressure is maintained, the operators still have to take action if control rods and other SCRAM systems doesn't work (but they have 28 hours to do something).

You are right Cyril, these reactors are still very safe but not as safe that I believed; maybe a well designed MSR can compete with PBMR on safety.


Last edited by fab on Dec 07, 2013 9:56 am, edited 1 time in total.

Top
 Profile  
 
PostPosted: Dec 06, 2013 11:33 pm 
Offline

Joined: Jun 19, 2013 11:49 am
Posts: 1482
Presumably they could introduce some sort of neutron poison into the reactor during the day they have before recriticality?


Top
 Profile  
 
PostPosted: Dec 07, 2013 1:10 am 
Offline

Joined: Apr 19, 2008 1:06 am
Posts: 2219
AGR are in use in the UK and use control rods. The quantity of gas in the reactor is so little that the neutronic interaction with gases is very small. The use of control rods should be normal.
AGR are being phased out in favour of LWR for various reasons including that CO2 gas becomes chemically active at high temperatures. Helium is rare and costly. Other alternatives could be
Lining of surfaces with a material like silica, as an enamel.
Use of a cheaper, abundant inert gas like Argon.


Top
 Profile  
 
PostPosted: Dec 07, 2013 10:29 am 
Offline

Joined: Jul 14, 2008 3:12 pm
Posts: 5048
E Ireland wrote:
Presumably they could introduce some sort of neutron poison into the reactor during the day they have before recriticality?


This is a helium cooled reactor... which is a noble gas under pressure.

So you need

1. insertion against this pressure (like an accumulator)
2. a neutron poison gas like BF3. But the gas has to stay in the reactor so you may need a lot of gas (if there's a broken loop you then need enough gas to fill the containment as well). BF3 is of course a nasty toxic little gas. Xenon may be a better choice. Gas has poor density though so will not likely get you to cold shutdown.
3. OR a liquid poison that somehow stays inside the reactor more or less. Perhaps a poison that chemically reacts with the graphite fuel elements so as to "coat" it with poison? A reactive carbide forming element.

It seems difficult. Helium can't dissolve neutron poisons like water can.


Top
 Profile  
 
PostPosted: Dec 07, 2013 10:46 am 
Offline

Joined: Jul 14, 2008 3:12 pm
Posts: 5048
fab wrote:
I believe that I read in different places that thermal gas reactors (like pebble bed reactors or the GT-MHR) can shut down the chain reaction even if control rods failed to insert in core and next remove their decay heat just by thermal conduction in the concrete of the reactor, even if helium was lost, without any operator action and for an indefinite period of time.

That why they were called "inherently safe" and the safest of all power nuclear reactors. So it's maybe true for small reactors but not for the reactors studied here ( 400 MW(t) and 600 MW(t) ).

It seems that I had it all wrong.


Technically I think you had it right for the short term; the reactor does shut down initially even without control rods. But then after some time it starts up again and gets to a temperature that will fail a significant fraction of the fuel elements. So control rod failure plus loss of the pumps (or just power) is safe in the short term but not in the long term. That is an issue for me because the three worst historic nuclear accidents, Windscale, Chernobyl, and Fukushima, were all long term accidents.

The fundamental problem being the poor coolant. The average temperature is actually ok, at least in terms of fuel preservation (vessel may still fail). But because of the poor coolant, especially when there's a leak in the vessel or piping that has led to depressurization, the peak fuel temperature in the center of the reactor is high enough to fail some fuel there.

This does suggest that a molten salt cooled version, would not have fuel failure. This document confirms this:

http://www.inl.gov/technicalpublication ... 303773.pdf


Top
 Profile  
 
PostPosted: Dec 07, 2013 11:28 am 
Offline

Joined: Jun 19, 2013 11:49 am
Posts: 1482
Well if you have trouble injecting the neutron poison under pressure you could just deliberately depressurise the primary loop and inject the material anyway.

As to reactive carbide forming neutron poisons... I am not sure I can think of any at this point but I will do so.


Top
 Profile  
 
PostPosted: Dec 07, 2013 11:50 am 
Offline

Joined: Jul 14, 2008 3:12 pm
Posts: 5048
E Ireland wrote:
Well if you have trouble injecting the neutron poison under pressure you could just deliberately depressurise the primary loop and inject the material anyway.


If we assume there to be a containment, then the pressure would still be multiple atmospheres. But I think this is the easy part - accumulators are widely used and are very reliable (you can have backups and fail-open valves).

Quote:
As to reactive carbide forming neutron poisons... I am not sure I can think of any at this point but I will do so.


Perhaps a rare earth element? Any of the serious neutron poisons, hafnium, gadolinium, come to mind. Hafnium is too high melting. Gadolinium could be useful, m.p. 1312C, that would melt in an emergency for a helium cooled design, before fuel failure can occur. Gadolinium carbide exists as a stable compound, not sure it will react on graphite and stay around as carbide at temperature.

I guess you could also have a ceramic catch pot somewhere in the core, where an overhead, out of core Gd can melts its contents into in a beyond design basis scenario like ATWS-SBO.


Top
 Profile  
 
PostPosted: Dec 07, 2013 12:37 pm 
Offline

Joined: Oct 28, 2013 12:24 am
Posts: 256
Quote:
This does suggest that a molten salt cooled version, would not have fuel failure. This document confirms this:


Yes, the natural circulation of Flibe is very efficient to remove the heat. But then I have a question : Is failure of natural circulation of Flibe a credible scenario ?

Maybe we can imagine some blockage (or diminution) of the flow if we have physical damage inside the vessel and the core. I guess that the molten salt in contact with the hottest parts of the core will boil and then recondense at other places in contact with the vessel (or in contact with liquid molten salt) and we will maybe still have efficient cooling of the core.


Top
 Profile  
 
PostPosted: Dec 07, 2013 1:11 pm 
Offline

Joined: Jul 14, 2008 3:12 pm
Posts: 5048
Total flow blockage doesn't appear to be physically reasonable. Partial flow blockage is possible such as pump impeller breaking or graphite breaking off. This will result in overheating of a channel or reduced total natural circulation. The margins are substantial. Fuel failure can't occur because salt boiling occurs well before TRISO fuel failure. So you'll get overheated or even partial voided flow, which is a very efficient heat transfer mechanism. FLiBe gets more efficient in cooling as it heats up from reducing viscosity so I doubt you'll even get to boiling till at least 90% of flow has been blocked. It seems very difficult to get such a flow blockage. The only way for it to occur it seems to me is during normal power operation when some piece of graphite breaks off, partially blocks flow, and, due to the high normal power output of a channel, causes partial boiling in that channel. This would greatly reduce fission power in that channel so it seems likely that no fuel failure would result even if the operators don't notice what's going on.

Natural circulation post shutdown can be guaranteed by design. If the AHTR is built in a below grade silo and there's not much space in between the vessel and silo, then there's always coolant covering the TRISO fuel. That's an important design requirement for such a reactor.


Top
 Profile  
 
PostPosted: Dec 07, 2013 3:15 pm 
Offline

Joined: Oct 28, 2013 12:24 am
Posts: 256
Thanks very much Cyril. I have a last one question if you are not tired of me.

I was wondering about the possibility of removing the decay heat of the PBMR just by conduction in the concrete until some external walls open to air.
Just evacuate 0.2 % of the 400 MW(t) thermal power of the reactor ( it's 0.8 MW(t), that's the decay heat at about 6 days after shut down) through 4 meters thick reinforced concrete ( the inner part of the concrete is in contact with the containment vessel and the outer part is open to air).

I don't have calculation softwares so I just consider a ridiculously simplified model with a infinite cylindrical geometry, steady decay heat, consider the thermal conductivity of concrete k always constant, and air removal of the constant 0.8 MW(t) decay heat in the outer limit of the wall. I know it's ridiculous but it's just to have an order of magnitude.

The best k value that I saw for reinforced concrete was k = 2 W/(m.K)

If k = 2 W/(m.K) I found that the temperature difference between the inner limit of the concrete and the outer limit is about 3200 K so I think it's just impossible.

If k = 10 W/(m.K) the temperature difference is 680 K. So maybe it's possible to use a metal instead of concrete, like some kind of steel, but I guess it's too much steel, it's maybe too difficult to manufacture and too costly.

I was thinking about a system where you have your containment vessel in steel in contact with a thick wall of reinforced concrete, and the outer limit wall is in contact with an other layer of steel which is open to air. And the containment vessel and the outer layer of steel are thermally connected with bars of an higly thermally conductive metal ( or simply steel). The bars are through the concrete and serve as thermal shunts.

My question is : is this realistic ? I think not because I never heard of this kind of system but I am curious to know why. Maybe the concrete will not support the thermal stresses.


Top
 Profile  
 
PostPosted: Dec 07, 2013 3:26 pm 
Offline

Joined: Oct 28, 2013 12:24 am
Posts: 256
Hum 4 meters thick reinforced concrete is too high, with a 2 meters thick I have

for k = 2 W/(m.K) : the temperature difference is 1950 K

for k = 10 W/(m.K) : the temperature difference is 390 K



With a 1 meter thick :

for k = 2 W/(m.K) : the temperature difference is 1055 K

for k = 10 W/(m.K) : the temperature difference is 211 K

But these are still order of magnitude.


Top
 Profile  
 
PostPosted: Dec 07, 2013 4:14 pm 
Offline

Joined: Jul 14, 2008 3:12 pm
Posts: 5048
I think the decay heat after 6 days is more like 0.3%, so 1.2 MWt. If the control rod has failed, total power will be a bit more than that as the reactor starts up again with xenon decay.

Concrete and high temperature don't go well together. There are various admixtures such as Firerok that can improve thermal performance and increase resistance to high temperatures, but heat removal is still bad.

2 W.m/K is a high value for concrete, but feasible. However, at elevated temperatures the conductivity decreases as water is drawn out. Most concretes are below 1 W.m/K at 1000 C, and most will be badly cracked at that temperature if they haven't failed completely.

It is possible to use a steel plate concrete construction. This has steel tie rods stringing two parallel plates on both sides of the concrete. The result is high thermal conductivity as the tie rods are like a thermal bridge. You can probably do better than 10 W.m/K with this while staying at economical amounts of steel in the concrete.

However steel has no strength left above 500C.

Once the inner layers of concrete heat up, a hot zone will form, impeding further heat transfer.

Moving heat through concrete, even steel plate concrete, should be a last ditch effort, an act of desperation after all else has failed and we do nothing.


Top
 Profile  
 
PostPosted: Dec 07, 2013 4:22 pm 
Offline

Joined: Jul 14, 2008 3:12 pm
Posts: 5048
I forgot to add something important, which is that the issue here with helium cooled pebble bed reactors is cooling from the fuel to outside the vessel. Not from the passive cavity cooling system itself. The issue is the poor coolability of helium (even at high pressure its poor). If we are going to change the thermal engineering with thermal shunts and such, it must be something inside the vessel.


Top
 Profile  
 
PostPosted: Dec 07, 2013 4:30 pm 
Offline

Joined: Oct 28, 2013 12:24 am
Posts: 256
Ok, thanks a lot.


Top
 Profile  
 
Display posts from previous:  Sort by  
Post new topic Reply to topic  [ 30 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