Dear all,
I am currently in the process of developing a multi-scale model that can address irradiation damage happening in the meso-scale and allow that to influence my stresses and strains in the engineering scale. Effectilvey i am dealing with lattice defects that leads to swelling of material components which affects its dimensional stability under irradiated conditions. I may eventually need to predict the rate of swelling under various levels of defect dosage (dpa) and also determine the macro stresses and strains arising out of this behaviour. With this in mind, i plan to set up a reactor configuration model in OpenMC, allow that to irradiate with neutrons and determine the above entities through a model that can predict swelling.
The question that is lingering around me is that how and where do i accomodate my swelling model in this multi-scale framework. In essence this swelling model would comprise of a special term called ârelaxation volume tensorâ which models the distribution of defects in an irradiated configuration. Once we have this volume tensor at all spatial locations of my model, then computation of stresses and strains is straightforward and can be performed with another open source FE package we have in place. I wish to ask whether OpenMC can simulate defect configurations in a given component model and if that is possible what options are available to calculate this relaxation volume tensor that can help compute the engineering stresses and strains in the model.
OpenMC doesnât incorporate any swelling models into the simulation. The only damage related information OpenMC provides is the amount of energy that goes toward damage to a material. This is based on a cross section produced by NJOY. A volume tensor would need to be computed by another code using this damage energy tally (presumably combined with a filter giving spatial resolution of this value).
Dear Patrick,
Thanks for the response. Yes, the spatial resolution of the damage energy is what we need for identifying the defects which in turn can be used to compute the stresses and strains for this model. I am quite new to OpenMC and i am wondering where to start looking for further related information and how to get in to the coding mode. I may want to address the following activities which needs to the executed in the order of preserence.
Firstly i may want to set up the OpenMC component model and extract the âdpaâ for the damage accrued in the model
I would then need to code up my swelling model which is to be driven by this âdpaâ
Once the extent of swelling (this is notionally called the relaxation volume density) has been determined i can then factor that based on the discrete damage energy weights obtained for this model
Eventually i will need to pass this factored âvolume densityâ to my FE code to determine the stresses and strains, representative of the radiation induced defects simulated in the model
The first three items are expected to be implemented in OpenMC and last one in MOOSE, which is actually my FE code. Once we get comfortable doing the openMC items then i may eventually perform everything within MOOSE. Going by this conceptual design, we may need to use the same mesh for both cases. It would be of great help if you could please offer your advice in setting my OpenMC model.
Kind regards,
Arun
Dear all,
I am currently working on ways to calculate the elastic relaxation volume tensor that quantifies the overall macroscopic expansion of the material due to the presence of irradiation defects. The most straightforward way is to use a simple swelling model driven by the damage dose and irradiation temperature that evaluates this volume. Having said that, this however may not be a plausible option, as most of the models require experiments to calibrate the parameters, which might take years to complete. Literature results are quite scarce, which are confined to specific materials and irradiation conditions.
Recent articles report (Multi scale model section 5.2, 36) that this relaxation volume tensor can be obtained from neutron transport calculations for a given reactor configuration. I am wondering whether OpenMC has any particular way of extracting this quantity from the tallied results. As far as my knowledge is concerned there is a topic on relaxation Atomic relaxation but that does not seem to be related to the relaxation volumes that contribute to swelling. I would be grateful if somebody could point me in the right direction.
Kind regards,
Arun
Thatâs a very interesting question. Looking at the paper youâve referred to there, what a neutron transport solver could in theory provide is n(\mathbf{r},E) which feeds into Eq. 36. If Iâm understanding it correctly, that quantity is effectively the spectra of primary knock-on atoms. At least in OpenMC, at present we do not calculate recoil energies and wouldnât be able to provide that information. That being said, I have an ongoing project with collaborators at Colorado School of Mines to be able to generate this information directly during a simulation. In lieu of the PKA spectrum, the only thing OpenMC could provide is the damage energy, which allows you to make an estimate of the NRT-dpa.
Dear Paul,
As you rightly said, there is no way we could extract this information from the neutronics simulation we currently have. Itâs good to hear from you that your collaborators from Colorodo are working on similar stuffs. I am glad to work with them, if you could please pass their references to me. This tensor is pretty much a critical quantity that characterise the irradiation defects from a sizing point of view and if we could manage to extract that successfully, we may as well not think about the calculation of damage dose DPA. I look forward to collaborate with your partners and hope to see this work taking shape and moving in the right direction.
Kind regards,
Arun
Dear Paul,
I am really keen on finding this quantity n(r,E) that represent the spatial density of cascades. Wondering whether you can provide me some guidelines on this, so that i can try and implement that inside OpenMC. I am happy to hear from you further updates on this work if you have anything to share.
Kind regards,
Arun