Depletion to track tritium production

User Support
1

Hi, all

I’m trying to calculate tritium production rate in molten salt reactors. I want to see the contrituions of each element to tritium production, but couldn’t find a way to track it.

In my problem, tritium source term would be 6Li, 7Li, and ternary fission in fuel salt.
Currently, I’m modeling the fuel salt as a single material, so depletion.get_atoms can give total generation rate. But is it possible to check the production from each source term?
On one hand, I’m thinking that tallying the reaction rates by nuclides that contribute to tritium generation in each result files could be an option, but I’m not sure and don’t know how to see ternary fission either for now.

Also, seeing the previous post, Very low value for amount of H3 atoms transmuted, it seems tritium-generation reaction of Li6 and Li7 should be added to depletion chain. Li6 is clear, but I’m confused what reaction type should be for Li7.
Li7 produces tritium by inelastic scattering(7Li(n,n’)7Li∗ → t + α) so should it be (n,nc)?

Any comments would be appreciated.
Thanks!

2

Out of curiosity, what depletion chain are you using?
Chains based on ENDF-B evaluations (such as the one available on the openmc website) lack ternary fission yields which caused us to severly underestimate tritium production a few month back, this was solved by building chains based on JEFF-3.3.

Nicolas

3

I’m using ENDF chain.
Currently, I’m setting up model and have not evaluated actual cases.
So tracking ternary fission seems not possible based on ENDF chain…
Good to know that. Thanks!

4

Hi, could you provide some information about how to build JEFF chain?
I checked the JEFF-3.3 library and the files are a bit different from ENDF’s, although they’re based on same ENDF6 format.
Is openmc.deplete.Chain.from_endf method just can be used for JEFF files without modification? (after chaning their file names to ENDF name format)

5

There is a PR on the official openmc-dev data repo that adds a JEFF chain processing script to the repo

I have a packaged up version of these scripts and a few others over on this repo

The If you wanted to try the packaged version then you would just type this in your terminal

pip install openmc_data

then

generate_jeff_chain
6

Thanks for sharing the repo!
Unfortunately, I’ve encountered an error as follow when I run generate_jeff_chain.
Could you let me know how to fix this one?

image
image1713×421 22.5 KB

7

I modified Jon’s script to fix this issue, you can find the fixed branch here.

@Shimwell I took the liberty of creating a pull request with this fix, the explanation for the error is given in the PR description:

The reason for the failure is that this file is malformed and lacks a proper header line. This PR adds a few lines in the generate_jeff_chain script to fix the file before loading neutron fission yields.

1 Like
8

Awesome thanks @nlinden , ive merged that PR in and released a new version which can be installed with

*editted version as per correction later in thread.
pip install openmc-data==2.2.8

1 Like
9

It works well. Thanks a lot to both of you!
Btw, the version is updated as 2.2.8 not 0.2.8.

3 Likes