Processing of MF10 Multi-isotope entries

[pjgriff]

Issue: NJOY fails to properly handle files with multiple MF10 entries. The use of multiple MF10 residual entries is supported within the generic ENDF-6 format.

Motivation: Dosimetry applications often read the activity from foils that represent the natural abundance of isotopes. I note that the IRDFF-II library still provides the dosimetry reactions on the separate isotopes, but the dosimetry community thought that it was important the the new release address the production of the activation products from the some of the naturally occurring elements– since that is used in almost all applications. bug.zip

Example: Consider the IRDFF-II dosimetry cross sections for Ti-nat. This library gives the cumulative dosimetry cross section for the production of various activation products, e.g., Sc-46, Sc-47, and Sc-48 isotopes from incident neutrons on naturally occurring Ti. This is an example where NJOY gives results – but the results are incorrect! To illustrate the problem, I have isolated the Ti-nat evaluation from the IRDFF-II library. This extracted nuclear data file is attached and named debug_Tinat_extract.endf. The NJOY-2016 input file that processes this file is attached and named Debug_Tinat_extract.inp. The resulting output file is Debug_Tinat_extract.out. Looking at the GROUPR output we see that all three residual nuclide have identical outputs: 1) First – Sc-46 for mf10 mt 5 zam 210460 (n,x) 21046 production

enrgy group constants at bug.zip

group infinite dilution

473 3.18467-9 474 1.06862-7 475 2.59137-7 476 4.11412-7

2) Second: Sc-47 for mf10 mt 5 zam 210470 (n,x) 21047 production

   enrgy  group constants at
   group  infinite dilution

     473    3.18467-9
     474    1.06862-7
     475    2.59137-7

3) Third: Sc-48 for mf10 mt 5 zam 210480 (n,x) 21048 production

   enrgy  group constants at
   group  infinite dilution

     473    3.18467-9
     474    1.06862-7
     475    2.59137-7

This is clearly incorrect. Investigations show that the data from the last residual nuclide over-writes all of the output quantities. If I modify MF10 portion of the nuclear data file so that only the Sc-46 residual nuclide is present, then I get a correct output (correct based upon my comparisons with combining the separate isotopic evaluation results). I have attached the sample modified ENDF-6 format nuclear data file, debug_Tinat_extract_Sc46.endf. The example NJOY-2016 calculation is seen in Debug_Tinat_extract_Sc46.inp and Debug_Tinat_extract_Sc46.out files. Here the relevant output for Sc-46 is:

  for mf10 mt  5 zam  210460 (n,x) 21046 production                                      

   enrgy  group constants at
   group  infinite dilution

     462    2.79862-7
     463    8.44432-7
     464    1.40900-6
     465    1.97357-6

This output is clearly different from the results of the previous calculations. Asking users to hand-edit the nuclear data file is not a good proceduce to allow to continue. This is a particular concern since the NJOY calculation appears to give valid results that users might not question – but they would have errors in their subsequent analysis. To produce the desired results, I could (and have) separately run the Ti46(n,p) and the Ti47(n,np) reactions, weighted these reactions by the recommended natural abundances, and added the resulting cross sections. This is a different work-around to the processing deficiency, but is also not a desirable approach either since it involves the user adding the separate abundance data and breaks the traceability of the source back to the IRDFF-II dosimetry file as it was released. A related issue (but one addressed to the IAEA and not to the NJOY development team) is that the IRDFF-II files fails to give a covariance file for the composite reactions on the naturally occurring elements, i.e. a MF40 entry. So, in this case, the user has to go back and process the separate covariance matrices on a common energy grid and then combine them. This is particularly bad due to the sensitivity of some covariance matrices to interpolation onto different energy grids. This second issue is not an NJOY code issue, but it is an issue with the IRDFF-II library. This issue is being separately relayed to the IAEA to see if the IRDFF-II release can be corrected/updated to address this deficiency, i.e., all dosimetry cross section data are required to have an associated covariance matrix and the promise made to the IRDFF-II development community was that this would be reflected in the final IRDFF-II release.

The attached files were combined into one sip file in order to include them in this bug entry.

[pjgriff]

Looking at the other releases for NJOY-2016, It appears that my issue was previously submitted by Larry Greenwood -and was already addressed: https://github.com/njoy/NJOY2016/issues/124 I will try to apply the changes in this release and it should fix my issue.

[nathangibson14]

Thanks, @pjgriff for tracking down the history here. Running a relatively recent NJOY2016 myself I do not see this problem. Are you running an older NJOY release?

And thanks @kahlerac for already fixing the issue!

[kahlerac]

All, This was patched in njoy2012.145. Being able to process multi-isotope mf40 remains a tbd issue to resolve. Skip

On Wed, Aug 18, 2021 at 4:45 PM Nathan A. Gibson @.***> wrote:

Thanks, @pjgriff https://github.com/pjgriff for tracking down the history here. Running a relatively recent NJOY2016 myself I do not see this problem. Are you running an older NJOY release?

And thanks @kahlerac https://github.com/kahlerac for already fixing the issue!

— You are receiving this because you were mentioned. Reply to this email directly, view it on GitHub https://github.com/njoy/NJOY21/issues/153#issuecomment-901418595, or unsubscribe https://github.com/notifications/unsubscribe-auth/AEHJISODVEHK32XBPZ36MYDT5QLXDANCNFSM5CMZF7CA . Triage notifications on the go with GitHub Mobile for iOS https://apps.apple.com/app/apple-store/id1477376905?ct=notification-email&mt=8&pt=524675 or Android https://play.google.com/store/apps/details?id=com.github.android&utm_campaign=notification-email .

-- Dr. A. C. (Skip) Kahler Kahler Nuclear Data Services, LLC @.*** +1 321 368 3645

[pjgriff]

Yes, my work was based on NJOY-2016 version 2019.1 - an older version - and, yes, it appears that my issue was logged against NJOY21 rather than NJOY-2016. This is a reflection of the fact that I am a novice to GitHub version control and still have a lot to learn about the version control system. I have verified that Skip's changes in the latest release address/solve my issue. Thank you very much Skip for addressing this important issue. I am sorry that I did not see this history of changes earlier. I am working to fully update my working version to reflect the latest NJOY2016 release - I still have a lot to learn about how to interact with GitHub. For reasons that I have discussed with Jeremy, I have to stick with NJOY-2016 (at least for now) rather than upgrading to NJOY21. I will attempt to close out this issue since it has been resolved by Skip's changes.