The Logic Behind Chernobyl: Control Rods and Moderators

[emanfman]

I am doing research on the Chernobyl disaster.
I came across something that seemed strange...
Apparently, the control rods were covered in graphite moderator.
As far as I'm concerned, a control rod absorbs the neutrons and stops fission reactions.
However, the graphite moderator's function is to slow down neutrons to help them being absorbed... effectively encouraging more reactions.
What is the logic behind it?
Is it that the neutrons should be slowed down to help being absorbed by the control rods?
The reactions were only boosted. What genius invented that? :P

I may not be the first person to ask this, but I couldn't find it anywhere.
And no, I'm not asking you to do my homework for me... :P
It just doesn't seem to make sense that of all reasons the reactor could have exploded, it had to be some idiot inserting graphite instead of neutron absorbers.
Or maybe it could have been Wigner energy?

I have no clue... please help!

-Emanfman

 

[flpojken]

You could study the term Xenon poisoning of a reactor. This played a major role of why and how the reactor blew up. Also the RMBK (the kind in chernobyl) reactors are unstable design when it is not operated within certain limits.

You can read more on this here
http://en.wikipedia.org/wiki/Rmbk

 

[rmattila]

You are talking about two different graphite parts: (1) the graphite blocks surrounding the channels, which moderate the neutrons in between the channels in order to increase their efficiency to either split uranium or get absorbed in the control rods, and (2) the control rod graphite followers, hanging from the bottom of the control rods and replacing the water in the channel, making the reactivity difference between inserted and withdrawn control rod as large as possible.

The much talked-about design flaw of the control rods is the fact that the graphite followers were too short, and the bottom of the channel (about one meter or so) was filled with water instead of graphite, when the CR was fully withdrawn. When the control rod was dropped in the core, this water was first replaced with the graphite of the follower, and this created a positive reactivity addition in the bottom of the core. Normally this would not have had a large effect on the core total reactivity, but due to Xe poisoning the axial power distribution was such that an unusually large portion of the power of the core was produced in this 1 m region affected by the design flaw.

 

[clancy688]

Only the top of the fuel rods were covered in graphite. As flpojken already mentioned that has something to do with xenon poisoning.

http://en.wikipedia.org/wiki/Iodine_pit

During normal operations, nuclear reactors produce Tellurium-135, which decays to Iod-135, which decays to Xenon-135. Xenon-135 is a very big neutron absorber, so it's catching neutrons thus preventing those neutrons to fissure the nuclear fuel. If there's enough Xenon in the core, the nuclear reaction is very effectively slowed down. If the reactor's under power, the Xenon is constantly catching neutrons and transforms into Xenon-136.
But if the reactor is shut down (with control rods inserted), there's still Iod-135 left which's decaying into more and more Xenon-135. That could result in so much Xenon that the reactor needs days until it reaches full power again.

To counter this effect, the tops of the control rods were covered in graphite. If the reactor is restarted again, the control rods will be removed. The graphite tops will now increase neutron activity in formerly throttled areas of the core -> the resulting neutrons are "burning" the Xenon-135 away.

 

[Dmytry]

Actually, the tips (not tops) of control rods were graphite, to maximize the control range (which was an issue as it used lower enrichment uranium). Additionally, graphite increased fuel burn-up in the bottom of reactor. The tips are at the bottom part of control rods (the control rods are inserted downwards).
During Chernobyl accident, the reactor was put into a state outside normal operational zone (with respect to Xe-135), after which the control rods were withdrawn too far, taking the graphite tips out of the water. The graphite outside the water did not increase the reactivity very much. Afterwards, the SCRAM button was pressed to shut down the reactor, inserting the control rods. Graphite tips gone into water, where they were very effective at increasing the reactivity, resulting in explosion. Essentially, the reactivity with respect to control rod positions had a peak not when the rods are fully withdrawn, but when the rods are almost fully withdrawn. So for first few meters of rod movement the SCRAM increased the power.
The control rods were moveable too far, not quite sure why, possibly for ease of control rod maintenance. A typical accident. Something that nobody ever thought of happens. Some design feature that nobody but a few plant designers care about turns out to be fatal. An accident a probability of which cannot be calculated by combining probabilities of failure of components, and thus is presumed zero.

 

[clancy688]

Um... I just have to mention this:

I'm not very sure if my explanation regarding the purpose of the graphit tips (thx, Dmytry) is correct in every aspect. So I'd like to hear more opinions of people who are actually into this kind of science. I'm only an amateur...

I find it slightly amusing that rmattila and Dmytry are explaining the connection between those tips and the resulting disaster.

But the OP didn't ask for that... ;)

Apparently, the control rods were covered in graphite moderator.
[...]
What is the logic behind it?

 

[Dmytry]

Um... I just have to mention this:

I'm not very sure if my explanation regarding the purpose of the graphit tips (thx, Dmytry) is correct in every aspect. So I'd like to hear more opinions of people who are actually into this kind of science. I'm only an amateur...

I find it slightly amusing that rmattila and Dmytry are explaining the connection between those tips and the resulting disaster.

But the OP didn't ask for that... ;)

well, I did explain that them were there to maximize the control range and burnup in the bottom of reactor. Like, you withdraw the neutron absorber rod and simultaneously you insert the moderator rod. Changes the reactivity more than neutron absorber alone. OP is a bit confused as of why control rods caused the disaster (coated in graphite).

edit: actually, I'm a bit incorrect about the control rods being removed from water. The control rods are in special control rod tubes, with separate cooling circuit, fully filled with water. What happened was that control rod tips have increased reactivity in the bottom of the core as them were inserted.

I imagine positioning of a small fraction of control rods in a nearly withdrawn positions could have been used to equalize the fuel burn-up at different heights in the core. When you boil water inside reactor, which results in the non-uniform burnup (having more voids on top), there is a lot of ways how you can sacrifice safety in name of improving the fuel burnup. For instance, PWR inserts control rods upwards to shutdown.
The design was reviewed after Chernobyl and a lot of such dangerous decisions were reverted, sacrificing efficiency for safety whenever in doubt. Same cannot be said about reactors which instead have been changed since original construction to achieve higher burn up and longer time between refuelling.

 

[rmattila]

I find it slightly amusing that rmattila and Dmytry are explaining the connection between those tips and the resulting disaster.

But the OP didn't ask for that... ;)

If you read the original post again, you see that the OP actually asked two things:

1:

Apparently, the control rods were covered in graphite moderator.
As far as I'm concerned, a control rod absorbs the neutrons and stops fission reactions.
However, the graphite moderator's function is to slow down neutrons to help them being absorbed... effectively encouraging more reactions.
What is the logic behind it?
Is it that the neutrons should be slowed down to help being absorbed by the control rods?

and 2:

The reactions were only boosted. What genius invented that? :P

I tried to answer both of them:

You are talking about two different graphite parts: (1) the graphite blocks surrounding the channels, which moderate the neutrons in between the channels in order to increase their efficiency to either split uranium or get absorbed in the control rods, and (2) the control rod graphite followers, hanging from the bottom of the control rods and replacing the water in the channel, making the reactivity difference between inserted and withdrawn control rod as large as possible.

Regarding question #1, I don't see there's anything more to it than the desire to increase the efficiency of the CR:s by slowing down neutrons around the channel and increasing their probability of being captured by the control rod. As to part #2 of the question, I think my reply fully parallels that of Dmytry's: the followers are there to increase the reactivity difference between inserted and withdrawn positions, i.e. to maximize the control range. The same approach is used in the VVER-440 reactors, where the control elements are equipped with fuel followers which replace the absorber part as it is being withdrawn from the core.

 

[Astronuc]

Is it that the neutrons should be slowed down to help being absorbed by the control rods?
The reactions were only boosted. What genius invented that? :P

The configuration of the Chernobyl Unit 4 RBMK reactor at the time of the accident was well outside the design basis. The designers would not have intended the unit to be in such a configuration. Clearly the person(s) responsible for the experiment did not consult the designers, nor did they apparently do any kind of simulation of the experiment beforehand.

Water in a graphite moderated reactor actually acts as a poison. Displacing the water with graphite increases reactivity.

 

[emanfman]

Yes, I know about the xenon-135 poisoning. I was just perplexed.
I am not really looking for the reason of explosion, that was mostly the positive feedback loop.
My question is why do the control rods have graphite on them? If you look on en.wikipedia.org/wiki/Chernobyl_disaster you will see that the insertion of the control rods apparently increased the fission rate.
It doesn't make sense - isn't that the opposite of what it's supposed to do?

-Emanfman

 

[Astronuc]

Yes, I know about the xenon-135 poisoning. I was just perplexed.
I am not really looking for the reason of explosion, that was mostly the positive feedback loop.
My question is why do the control rods have graphite on them? If you look on en.wikipedia.org/wiki/Chernobyl_disaster you will see that the insertion of the control rods apparently increased the fission rate.
It doesn't make sense - isn't that the opposite of what it's supposed to do?

-Emanfman

Only if the reactor is placed in a configuration outside of design. Under normal operation, the power wouldn't be so skewed.

 

[emanfman]

XD I only read one post before replying!
Thank you very much for explaining, nearly all of these answers were helpful. From what I understand, only the ends of the rods were covered in graphite. This is to counteract poisoning. However, this increased the fission rate as it was being inserted fully. This made SCRAM very unstable in chernobyl. Do I have any of my facts wrong?

 

[rmattila]

See the figure below. Materials from top to bottom:

1. absorber material (yellow)
2. water (pale blue)
3. graphite follower (dark blue)
4. water (pale blue)

The left column is the situation with the control rods fully withdrawn (=before the scram), and the right column is the situation immediately after the initiation of the scram (control rods have dropped a couple of feet). At the locations marked with "-", the reactivity introduced by the scram is negative, as it should be: graphite is replaced by water, or water is replaced by the CR absorber material. At the very bottom of the core, however, there is a "+" sign, indicating the region where water is replaced by graphite. This is the positive reactivity caused by the scram. If the reactor is run in an appropriate manner, the power in this lowest 1.25 m of the core is negligible compared to the region where the scram introduced negative reactivity (the uppermost 5.75 m of the core). Due to the Xe poisoning, the power distribution of the Chernobyl 4 core was such, however, that the lowest 1 m dominated the core behaviour.

It was clearly a design flaw: if the graphite follower would have been 1 m longer, there would not have been such a region where water is replaced by graphite at the early stages of a scram. After the 1986 accident, the design of the control rods (actually there are several types of them, and this is only one) was changed in such a way that this effect was eliminated.

[PLAIN]http://www.neimagazine.com/Pictures/web/x/p/t/NEI_Fig02.jpg

 

[emanfman]

See the figure below. Materials from top to bottom:

1. absorber material (yellow)
2. water (pale blue)
3. graphite follower (dark blue)
4. water (pale blue)

The left column is the situation with the control rods fully withdrawn (=before the scram), and the right column is the situation immediately after the initiation of the scram (control rods have dropped a couple of feet). At the locations marked with "-", the reactivity introduced by the scram is negative, as it should be: graphite is replaced by water, or water is replaced by the CR absorber material. At the very bottom of the core, however, there is a "+" sign, indicating the region where water is replaced by graphite. This is the positive reactivity caused by the scram. If the reactor is run in an appropriate manner, the power in this lowest 1.25 m of the core is negligible compared to the region where the scram introduced negative reactivity (the uppermost 5.75 m of the core). Due to the Xe poisoning, the power distribution of the Chernobyl 4 core was such, however, that the lowest 1 m dominated the core behaviour.

It was clearly a design flaw: if the graphite follower would have been 1 m longer, there would not have been such a region where water is replaced by graphite at the early stages of a scram. After the 1986 accident, the design of the control rods (actually there are several types of them, and this is only one) was changed in such a way that this effect was eliminated.

[PLAIN]http://www.neimagazine.com/Pictures/web/x/p/t/NEI_Fig02.jpg[/QUOTE]

OK, this clarifies a lot. Just one more question.
Why did the xenon poisoning only affect the upper part of the core? Shouldn't it affect the entire thing?
And from the above posts, the simple statement is that the people who designed the rods were... not very smart. XD

-Emanfman

 

[Drakkith]

And from the above posts, the simple statement is that the people who designed the rods were... not very smart. XD

An easy thing to say looking back on an accident.

As the design of the reactor never included having the reactor running with the rods fulling withdrawn, it's kind of like saying its the designers fault a car is able to be driven with the doors open and the driver hanging out the side resulting in an accident. Just goes to show you that NOTHING can be made foolproof no matter how hard you try.

Why did the xenon poisoning only affect the upper part of the core? Shouldn't it affect the entire thing?

Hrmm. I'm not sure, but I would assume that the xenon, being a noble gas, would rise and collect at the top of the reactor due to its low density.

 

[Drakkith]

XD I only read one post before replying!
Thank you very much for explaining, nearly all of these answers were helpful. From what I understand, only the ends of the rods were covered in graphite. This is to counteract poisoning. However, this increased the fission rate as it was being inserted fully. This made SCRAM very unstable in chernobyl. Do I have any of my facts wrong?

Looks like that's correct to me.

 

[emanfman]

Ahh, thanks. Could someone please confirm

xenon, being a noble gas, would rise

please? I need to be sure, I don't want to misinform people.
Not that I'm saying I don't trust you or anything... ;)

-Emanfman

 

[Drakkith]

Ahh, thanks. Could someone please confirm please? I need to be sure, I don't want to misinform people.
Not that I'm saying I don't trust you or anything... ;)

-Emanfman

Fair enough.

 

[rmattila]

Ahh, thanks. Could someone please confirm please? I need to be sure, I don't want to misinform people.

Fission gases remain in the pellet and don't move around. The inhomogeneous Xe effect is due to the inhomogeneous power distribution of the core: at the centre, where power is higher than at the ends, there is more I & Xe produced and destroyed during operation. When the reactor is shut down, destruction of Xe by neutrons ends, but it is still being produced from the I-135. Therefore, after reactor shutdown (or power reduction, as was the case in Chernobyl), Xenon peaks in the middle of the core, causing the reactor power to drop there.

So the Xe effect is principally not between upper/lower parts of the core, but rather between the axial midplane and the ends. Another effect at low power is due to the positive void coefficient, which would presumably lead to depression of the power at the top of the core when running with high flow/low power region, since the void fraction at the top of the core is lower than it would be at nominal parameters.

Put together, these two effects lead to power peaking at the bottom of the core.

NB: all of the above is qualitative estimation, I have not made any actual analysis of RBMK core behaviour during the power reduction prior to the accident.

 

[Astronuc]

Here is a rough summary of the Chernobyl event.

http://hyperphysics.phy-astr.gsu.edu/hbase/nucene/cherno.html#c5
http://hyperphysics.phy-astr.gsu.edu/hbase/nucene/cherno2.html#c1

If they had scrammed the rods altogether, then that would have inserted a lot of positive reactivity when in displaced the water. They had dissabled some of the safety systems.

Had they inserted a few control rods, then they positive reactivity insertion would not have been as bad.

1. The reactor was powered down for a test sequence to determine if one of the turbogenerators could supply power to feedwater pumps until standby diesel generators came on line in the case of a local power failure. The test sequence involved the following dangerous steps

a. Instead of the design based 22-32% full power, the power was inadvertantly lowered to 1% of full power, an extremely unstable situation because of the positive void coefficient. Edwards reports that the operator failed to reprogram the computer to maintain power at 700-1000 MW(t).

. . . .

In the summary, it mentions that the reactor went prompt critical. That is the biggest safety violation for a nuclear reactor operator. That is certainly not the intent of the designers. The reactor state was well outside design basis.

Soviet scientist Legasov said of the violations of the safety restrictions "It was like airplane pilots experimenting with the engines in flight."

A reasonable description.