Heat deposit in non-fissile materials

New ideas for code development

Heat deposit in non-fissile materials

Postby Andrei Fokau » Wed Aug 18, 2010 2:31 pm

One of the few remaining reasons to use MCNP in our work is calculation of heating of non-fissile materials. I hope it is already in Serpent development roadmap, and if it is not, then I strongly support adding this feature.

To go further, adding evaluation of radiation damage both in DPA, and in fast (>0.1MeV) fluence units would be similar to Christmas for us.
KTH Reactor Physics (Stockholm, Sweden) neutron.kth.se
Andrei Fokau
 
Posts: 77
Joined: Thu Mar 25, 2010 12:25 am
Location: KTH, Stockholm, Sweden

Re: Heat deposit in non-fissile materials

Postby Jaakko Leppänen » Wed Aug 18, 2010 3:28 pm

Andrei Fokau wrote:One of the few remaining reasons to use MCNP in our work is calculation of heating of non-fissile materials. I hope it is already in Serpent development roadmap, and if it is not, then I strongly support adding this feature.

To go further, adding evaluation of radiation damage both in DPA, and in fast (>0.1MeV) fluence units would be similar to Christmas for us.


What the code can or cannot do is pretty much limited by what is included in the cross section libraries. For DPA calculations you probably need some cross sections that get you from flux to DPA. And if you have this data in ACE format, you should be able to use it with Serpent detectors as it is.

Non-fissile heating is probably also a matter of including the appropriate cross sections in the calculation?
- Jaakko
User avatar
Jaakko Leppänen
Site Admin
 
Posts: 1962
Joined: Thu Mar 18, 2010 10:43 pm
Location: Espoo, Finland

Re: Heat deposit in non-fissile materials

Postby Ed Read » Tue Aug 24, 2010 10:59 pm

The information needed to preform DPA calculation is already in the cross section data that is provided with the Serpent package. There are a few things that you need to know. The first is the fact that at least in ENDF7 there are only approximately 35 isotopes that contain the correct reaction information needed to attempt a DPA calculation. The second is that you have to there create a dummy section for the material, which you then have to used as a multiplier in order to satisfy the DPA equation. I have a set a of these dummy cross sections for 10 common LWR materials. The last thing to keep in mind is the fact that DPA is a function of Energy, so the energy bins need to be as detailed as possible. I have a paper on the implementation of DPA in Serpent and MCNP at SNA MC in Tokyo next month.
Ed Read
 
Posts: 1
Joined: Tue Aug 24, 2010 10:56 pm

Re: Heat deposit in non-fissile materials

Postby Jaakko Leppänen » Tue Aug 24, 2010 11:49 pm

Do you think it would be usable to have a new dosimetry cross section type that would simply consist of reaction mt and a table of energies and cross sections? I imagine it would be easier to use your own response functions this way than to generate complete ACE format data files.
- Jaakko
User avatar
Jaakko Leppänen
Site Admin
 
Posts: 1962
Joined: Thu Mar 18, 2010 10:43 pm
Location: Espoo, Finland

Re: Heat deposit in non-fissile materials

Postby Andrei Fokau » Wed Aug 25, 2010 12:03 am

There are recent papers on this subject by A. Hogenbirk et al.:

An easy way to perform a radiation damage calculation in a complicated geometry
http://dx.doi.org/10.1016/j.fusengdes.2008.06.061

A novel approach towards DPA calculations
http://dx.doi.org/10.1142/9789814271110_0078

A.Hogenbirk et al. wrote:Usually approximations are made to calculate the radiation damage. In this paper we present a method to calculate radiation damage in a simple yet accurate way without resorting to complicated molecular dynamics simulations. We use detailed continuous-energy damage cross-section data in combination with a modified material specification, which is weighted by the damage cross-section of the isotopes.
KTH Reactor Physics (Stockholm, Sweden) neutron.kth.se
Andrei Fokau
 
Posts: 77
Joined: Thu Mar 25, 2010 12:25 am
Location: KTH, Stockholm, Sweden

Re: Heat deposit in non-fissile materials

Postby Tuomas Viitanen » Wed Aug 25, 2010 3:50 pm

The cross section libraries of Serpent include the (total) heating numbers that are needed in the calculation of heat deposit in non-fissile materials. As Ed mentioned, also the damage energy production values are available.

However, these values cannot be - yet - applied in Serpent, because of a few reasons. First of all, these reaction channels are omitted from the cross section data array at some point of the ACE data processing. I quess this happens somewhere in processxsdata.c, where reaction channels with MT > 118 are discarded (in this case incorrectly) due to redundancy (MT value of the total heating numbers is 301 and that of total damage energy production is most probably 444).

Increasing the number of allowed reaction channels would also increase the memory consumption, which is a small drawback... Hence, it might be a good idea to include the heating numbers in the final data array only when necessary (only if heating is calculated & only for specific nuclei).

Secondly, the heating numbers (MeV/collision) are not "heating cross sections" and, thus, cannot be utilised as is in the Serpent detector calculations, at least to my knowledge. The actual heat production cross section is defined \Sigma_tot(E) H(E), where \Sigma_tot is the total cross section and H is the average heating number of a material (consisting of a single nuclide, otherwise H->H_avg). Therefore, either the response function of a detector should be defined as a product of the two previously mentioned, energy dependent functions (which I think is not possible in the current version of Serpent) or the energy-dependent heat production cross section should be separately calculated beforehand and utilised like any other response function.

It must be pointed out that after these modifications Serpent wouldn't be able to calculate the photon heating quite as accurately as MCNP, because Serpent lacks photon transport (unlike MCNP in N+P mode). The photons generated in a Serpent calculation are assumed to be absorbed in-situ, because of a flag used in the processing of the Serpent cross section libraries (LOCAL flag in HEATR module of NJOY). However, this accuracy should be sufficient for most purposes.
User avatar
Tuomas Viitanen
 
Posts: 65
Joined: Thu Mar 25, 2010 1:50 pm
Location: VTT Technical Research Centre of Finland, Espoo


Return to Development

Who is online

Users browsing this forum: No registered users and 2 guests