Page 1 of 1

Microscopic Cross Sections of selected isotopes

Posted: Fri Sep 25, 2020 8:58 pm
by lovergar
Hi all,

My goal is to determine the 2-group integrated microscopic cross sections for some isotopes in my system.

My system is simplified; it's just a cubic cell filled with a mixture of my isotopes. The relevant parts of code for geometry and material are as follows (don't focus on the compositions, it is simplified on purpose:

Code: Select all

% --- Cross section data library file path:
set acelib "[path].xsdata"

surf 300 cube 0.0 0.0 0.0 5
cell 300 0 my_material -300
cell 900 0 outside    300

my_material -5
3006.12c  0.005
3007.12c  1
4009.12c  0.5
9019.12c  2
92235.09c 0.1
92238.09c 0.5

% --- Neutron population and criticality cycles:
% Reflecting boundary conditions
set bc 2
set pop 10000 300 50

Then, say that I want to compute the (n,t) microscopic cross section for one of these materials, Li-6. To do so, I create a new material, containing lithium 6 only and proceed as follows:

Code: Select all

mat	Li6_XS 1.0
3006.12c 1.0

% Energy Grids

ene fast 1 0.625E-6 200
ene thermal 1 0.0 0.625E-6

% Flux

det f_flux_Li6 dm Li6_XS du 0 de fast

det f_phiXS_Li6 dr 105 dm Li6_XS du 0 de fast
	dt 3 f_flux_Li6
Unfortunately, the input does not run, and an error messages tells me to look on the input manual for an error on the detector for the XS.

Can I ask your help?


Re: Microscopic Cross Sections of selected isotopes

Posted: Fri Sep 25, 2020 10:21 pm
by Ana Jambrina
Couple of things here:
- Definition of my_material: the card ‘mat’ must be included: mat my_material -5
- Spatial domain detectors: the use of du (du 0) and dm (dm my_material) is redundant (universe 0 is composed by kernel my_material only) - integral domain over which the response function is integrated.
- Definition of the flux (corresponding to energy grid 'fast') and microscopic reaction rate (corresponding to response 105) of one of the nuclides, i.e Li-6, part of the kernel ‘my_material’, and consequently microscopic cross section, as follow:
--> det flux dm my_material de fast
--> det XS_Li6 dm my_material dr 105 Li6_XS de fast dt 3 flux
For further details, see, section 7.1, and ... inition.29 (material definition) + ... inition.29 (detector definition).

Re: Microscopic Cross Sections of selected isotopes

Posted: Wed Sep 30, 2020 8:38 pm
by lovergar
Hi Anna, thanks for your reply.

About the material card, yes, there's a "mat" missing in my previous post, but not in the code. It's just a typo I made here.
The input that I posted here is a simplified version of the actual one; in reality I have 4 different material cards, therefore I will use du rather than dm in the detectors.

Using your syntax for the XS detector (but replacing dm with du) I have:

det XS_Li6 dr 105 Li6_XS de fast du 0 dt 3 flux

And it works absolutely fine. Now, the problem is that if I do that on Li-7, rather than Li-6, it does not go through. On the other hand, if I want to compute the XS for other nuclides (eg Be-9 or F-19), it works again. This is rather strange to me, as Li-7 is present in the same mat card where Li-6, Be-9 and F-19are.
For the Li-7 cross section, i am using:

mat Li7_XS 1.0
3007.12c 1.0

det f_flux du 0 de fast

det f_phiXS_Li7 dr 105 Li7_XS du 0 de fast
dt 3 f_flux

And Serpent stops after an error in the line of the XS detector.

Any thoughts on why it does not work?


Re: Microscopic Cross Sections of selected isotopes

Posted: Wed Sep 30, 2020 10:24 pm
by Ana Jambrina
From the top of my head:
What does the error say (beyond the line and/or detector where Serpent fails)? And what cross-section library are you using?
Does it say something like: ‘Reaction mt 105 not found for response function’ ?
—> meaning for Li-7 there is no evaluated data for that reaction type (MT= 105), in the cross-section library. If I recall correctly, ENDF/B-7 or/and JEFF3.1.1 do not have it. It might be included in TENDL nuclear data library, i.e., TENDL-2014.
(Previously to the error, you can see the list of available reactions for the nuclide included in the cross-section library in use).
MT 105 is defined as: production of a triton, plus a residual. Sum of MT=700-749, if they are present. (For incident tritons, this is inelastic scattering (MT=700 is undefined)).
Further information about nuclear data can be found at:

Regarding dm/du, just be sure it’s what you are looking for: without further information, for me it might make more sense to integrate over the fuel kernel, where nuclide is actually present, rather the cell/universe volume/domain (keep in mind that, du 0 would tally and integrate over the entire geometry - including parts where the material might or might not be defined, multiplying each score with the microscopic cross-section of the specified material; being the volume not limited to the kernel region).
For example:
det 1 de fast dm my_material
det 3 de fast dm my_material dr 105 Li6_XS dt 3 1
[microscopic reaction rate of Li6_XS integrated over kernel (and divided by det 1)]
det 2 de fast du cell_universe
det 4 de fast dm my_material dr 105 Li6_XS dt 3 2
[microscopic reaction rate of Li6_XS integrated over cell (and divided by det 2)]
depending on the type of microscopic cross-section you are looking for (kernel-averaged or cell-averaged).