Depletion Calculation of Nuclide Th233

Hi everyone. Recently, I find something strange.
The nuclide numbers of Th233 is:


It appears a number that goes to zero in some steps.
I choose the ‘chain_casl_pwr.xml’ as the depletion chain with 200 batches 100 inactive and 2w particles(when I increase the number of batches or particles, the result will be good).
The depletion step interval is 20days, with a power of 27MWe.
The depletion model is a PWR core, with the H2O-U02 mix fuel contains U235, U238, O16, and H1 only.
I free very confused, If someone can make me understand, I will appreciate it. Thanks in advance!

1 Like

Hi,
Out of curiosity, what is your nuclear data library behind your calculation? Have you tried with different NDL’s and the same CASL chain XML file?

1 Like

Hello, I use the ENDF/B-VII.1, and I haven’t do that yet.

@world3 If you don’t use enough particles in a calculation, you may end up with reaction rates in certain regions with either very large uncertainties, zero mean values, or some other unphysical behavior. Because the estimated reaction rates are what are used to construct the depletion matrix which then determines the predicted compositions, you can end up with compositions that don’t make sense. It’s the usual principle of “garbage in, garbage out”.

Regarding the very small negative values, I believe that is an artifact of the matrix exponential method that is used for depletion under the hood (CRAM).

@paulromano If the simulated particles are not enough, I will use the method of fuel density correction to manually eliminate the small neutron absorption nuclides and modify the fuel density. Will it reduce the negative impact you mentioned?

If you are eliminating nuclides, this may change the computational speed but unless it perturbs the overall neutron flux, it wouldn’t affect the uncertainties on the reaction rate tallies. For that, you really would need to increase the number of particles used in the calculation.

If you are eliminating nuclides, this may change the computational speed but unless it perturbs the overall neutron flux, it wouldn’t affect the uncertainties on the reaction rate tallies. For that, you really would need to increase the number of particles used in the calculation.