Cross Section Generation for Transient Analysis
This is just a research microblogging post. It may be the last.
Today, I’d like to instruct a graduate student on how to support some transient analysis with a temperature-dependent database of cross-sections for the various materials in the PB-FHR reactor.
What Cross-sections Do We Need?
We want to achieve the same goals as the PBMR400 benchmark, but in the context of a different reactor design. That benchmark provides a full specification of 26 energy group and temperature dependent cross sections. So, we probably want to do approximately the same thing.
What Else Do We Need?
In addition to the cross sections, material properties are needed to conduct the transient analysis. The temperature, burn-up, and fluence dependencies of all thermophysical properties of the materials need to be defined. In our case, that includes :
- FLiBe density, heat capacity, conductivity, etc.
- TRISO particle fuel matrix heat capacity, conductivity, etc.
- graphite density, heat capacity, etc. in the fuel pebbles and the reflectors
What Tools Can Generate The Cross Sections?
Typically, Monte Carlo codes are the best at capturing geometries in nuclear reactors. Though the method is slower than many deterministic methods, monte carlo is the workhorse for cross section generation. Serpent and MCNP are the main options for this. Their merits are varied, but by virtue of being in C++ and being more user friendly for temperature dependent cross section generation, we’ve decided to go with Serpent.
Relying on Past Work
Past models of this reactor have been generated for MCNP by python scripts (see Cisneros, FIMPS, BEAU). That work should be refactored into a python library for generating Serpent input as well.
What Else Do We Need to Consider?
When generating cross-sections, what else do we need to consider?