Zirconium hydride moderator definition

[A.S.M. Fakhrul Islam]

When ZrH2 is used in my core instead of BeO as moderator, the k-eff consistently drops from 1.5 to 0.35. I am perplexed by this big decrease in k-eff. I tried each of the following material definitions:

ZrH2 2 den=5.61 1 500 end
atomZrH2 2 5.61 2 40000 1 1001 2 1 500 end
atomZrH2 2 5.61 2 1040000 1 7001001 2 1 500 end
ZrH2 2 den=5.61 1 500 1040090 51.45 1040091 11.22 1040092 17.15 1040094 17.38 1040096 02.80 end

[Willem van Rooijen]

Start with the usual: CHECK THE OUTPUT FILE!

check and double-check:

- are the number densities in the mixing table correct (that is to say, as you expect them to be)?

- are there any error messages or warning messages?

- are you sure that the geometrical model is correct?

- are you using the correct units?

Seek, and ye shall find. 

We will not do your homework.

Van Rooijen

2024年10月25日金曜日 12:03:16 UTC+9 A.S.M. Fakhrul Islam:

[Rike Bostelmann]

Hi Fakhrul,

I recommend confirming that the Thermal Scattering Law data was actually used for ZrH.
The nuclide IDs for H in ZrH is 7001001. This ID should show up in your mixing table. 

Additionally, there should be information in the msg and/or out file on the TSL data that has been loaded.

Best of luck!

Rike

[A.S.M. Fakhrul Islam]

I did try to make sure that the correct data library (tried different libraries such as  ce_v7.1, ce_v7, v7-238) and the correct TSL data are being used. According to SCALE user manual:

"The KENO cross section library inherently accounts for thermal scattering using either the thermal scattering law data or the free gas approximation."

I also tried different moderator temperatures (300, 400, 500, 600, 700 K) just in case the thermal scattering data is missing for a certain temperature. 

The nuclide IDs for H in ZrH does show up as 7001001 in the mixing table. Here is what my mixing table looks like in the output file:

I am including my input file below if you would like to review it:

=csas6 parm=criticality

ce_v7.1

read comp

'Standard UO2 fuel

UO2 100 den=10.52 1 1100 92234 0.0476 92235 4.95 92238 95.0 end

'Helium fillgas

He-4 200 den=2.65156E-5 1 950 end

'Stainless Steel cladding

SS304 300 den=7.80 1 800 end

'Helium coolant

He 400 den=1.7E-3 1 700 end

'ZrH2 moderator

atomZrH2 500 5.61 2 1040000 1 7001001 2 1 600 end

'BeO moderator

'BeO 500 den=2.95 1 600 end

end comp

read param npg=300 gen=300 nsk=100 htm=no plt=no end param

read geom

unit 1

cylinder 1 0.41 2p100 com="fuel"

cylinder 2 0.42 2p100 com="gap"

cylinder 3 0.46 2p100 com="cladding"

cylinder 4 0.80 2p100 com="coolant"

hexprism 5 0.80 2p100 com="moderator"

media 100 1 1 com="UC fuel pin"

media 200 1 2 -1 com="He-4 gap"

media 300 1 3 -2 com="SS cladding"

media 400 1 4 -3 com="He coolant"

media 400 1 5 -4 com="ZrH2 moderator"

boundary 5

unit 5 com="coolant"

cylinder 1 0.8 2p100

hexprism 2 0.8 2p100

media 400 1 1

media 400 1 2 -1

boundary 2

global unit 10

hexprism 1 3.6 2p100 rotate a1=30

hexprism 2 14 2p130 rotate a1=30

media 500 1 2 -1

array 1 1 place 4 4 1 0 0 0

boundary 2

end geom

read array

ara=1 typ=shexagonal nux=7 nuy=7 nuz=1

fill

5 5 5 5 5 5 5

5 5 1 1 1 5 5

5 5 1 1 1 1 5

5 1 1 1 1 1 5

5 5 1 1 1 1 5

5 5 1 1 1 5 5

5 5 5 5 5 5 5

end fill

end array

read bounds all=reflect end bounds

end data

end

[Cliff Dugal]

Maybe your media definitions are wrong? I see that you applied helium to a region that you comment should be ZrH2.

[A.S.M. Fakhrul Islam]

Apologies about the wrong comment but the media definition is right.

[Willem van Rooijen]

You have a (very small) region with nuclear fuel surrounded by a very large amount of "moderator".

ZrH2 is a moderator but absorption due to H-1 is not negligible.

If you print the reaction rates from KENO I expect you will see a very high absorption rate in the external moderator region when ZrH2 is the material.

In other words: there is no problem. Your result is correct (although perhaps not what you expected). Try graphite instead of ZrH2 to confirm.

Some comments:

- the fractions of uranium isotopes do not add to 100.0% I guess the difference is acceptable for SCALE but I would suggest to use correct numbers.

- write REAL entries with as a floating point number, to make the distinction with INTEGER values clear. You have mixture number 500 and temperature 600. For quality control always make a clear distinction between entries that should be INTEGER and REAL

atomZrH2 500 5.61 2 1040000 1 7001001 2 1.0 600.0 end

- your array definition is probably wrong.

Van Rooijen

2024年10月26日土曜日 9:48:51 UTC+9 A.S.M. Fakhrul Islam:

[A.S.M. Fakhrul Islam]

It seems the higher absorption cross section of ZrH2 is the reason. 

[Dennis Mennerdahl]

Fakhrul and others,

You may have received this by email (sent Friday night, European time) but I repeat it for everybody.

There is no ZrH2 moderator in Unit 1. The cooling material 400 is used where the moderator 500 is supposed to be. 

Also, just exchanging two such different moderators as BeO and ZrH2 in the same geometry should give very different results. One may be near optimum moderation, the other would not be.

Dennis

[A.S.M. Fakhrul Islam]

Dennis,

Thank you for your comment. You are correct. There is no ZrH in Unit 1 because I defined it that way. Unit 1 represents the fuel assembly, which is surrounded by the moderator material (Unit 10).

P.S.: This is a mock model I created for troubleshooting.

Indeed, all else being equal, two materials with different cross-sections will produce different results. However, I was perplexed by the magnitude of the difference in the outcomes. Initially, I thought the system might be undermoderated, so I kept increasing the amount of moderator. But this only worsened the neutron economy.

Since I hadn’t worked with hydride moderators before, I didn’t realize how strong their impact could be. It turns out you can’t overmoderate a system with hydride moderators and expect minimal differences in results, as you might with graphite or beryllium. With hydride moderators, there seems to be a "sweet spot"—deviating too much in either direction, whether overmoderating or undermoderating, can significantly affect performance!

Thanks,

Fakhrul