How to calculate material damage (DPA)

Dear FLUKA experts,

I want to know how to calculate the material damage from electron beam when 22 MeV (or more) electron beam hits a target sample (titanium, kapton).

Could you help me how can I calculate it? What cards do I need to activate?

Best regards
Zohrab

2 Likes

Dear @zamirkhanyan,

What needs to be included is the damage energy threshold for the materials of interest by means of the MAT-PROP card (take a look to MAT-PROP | FLUKA). Then you can request DPA in your scoring (DPA-SCO particle type on the USRBIN card). Of course, as in any other FLUKA simulation, keep an eye on the particles’ production and transport thresholds to make sure your simulation is sound.

This discussion may be of interest to you.

Kind regards,
Francisco

1 Like

Dear Francisco Ogallar Ruiz
Thanks for the kind reply!
I would like to clarify a little what I needed to calculate.
There are several materials that we want to use in a linear accelerator as a vacuum window for extracting an electron beam.
Therefore, I want to understand which type of windows are better based on the simulation results used.

To understand this, I calculate the structural damage of materials when they are irradiated with an electron beam.
Can you tell me
1)What is the unit of DPA-SCO particle type on the USRBIN card.
2)Why, when it is active, the following cards are RADDECAY, IONTPANS, IRRPROFI, DCYTIMES, then DPA-SCO only give results at the beginning of the target.
I have attached my flair and input file as If you have time can you take a look to my input card and say what I may miss or wrong?

Could you also suggest how can I calculate and find which windows are best to use?

Best regards
Zohrabdpa.flair (13.9 KB) dpa.inp (6.4 KB)

Dear @zamirkhanyan,

Addressing your questions:

  1. From note 7 in USRBIN | FLUKA : Displacements Per Atom (DPA) will be expressed as average DPAs in each bin per unit primary weight.

  2. In the RADDECAY card you are providing a multiplication factor of 99999 to be applied to e+/e-/gamma transport energy cutoffs for the prompt calculation (take a look to RADDECAY | FLUKA). This means that you are setting an energy cut for the transport of your electrons equal to infinite and therefore, as soon as they reach your material, they deposit their energy in the spot and are not further transported.

Few comments on your input file:

  • Why are you associating some DPA-SCO USRBINs with cooling times? What do you want to obtain?
  • You don’t need to include a card for the sternheimer parameters nor the ionization potential of kapton unless you want to modify the default ones.
  • Why including the PART-THR and the DELTARAY cards? For e-, e+ and \gamma you are already setting your transport cuts via de EMFCUT card.
  • The physics card PETHRES is obsolete.
  • I see that you have DOSE USRBINs with a huge number of bins (800x800x600 = 384e6 bins ~ 1 Gb!). You may want to reconsider your bin sizes. It would be difficult to justify them if you compare with your beam FWHM: 0.005 vs 0.24 cm. It will take quite a lot of CPU resources to achieve good statistics for this extremely fine binning, and at the same time it will not allow you to extract better conclusions than a thicker one. Similarly, you can ponder whether 600 bins in the z direction make sense or not.

Hope this helps.

Kind regards,
Francisco

Dear Francisco

Thank you for your patient help, I still have some questions.

Before asking questions, I would like to briefly introduce my current activity. I study the effect of electron beam irradiation in a system consisting of vacuum windows of different materials (to determine the best material) and water target located at 10 cm after the window.

I would like to determine following physical parameters through simulation.

  1. Determine displacements per atom (DPA) in thin windows for defined radiation damage which will enable us figure out the best material for window.
  2. Determine gas production if there is any?
  3. Determine the temperature gradients of the irradiation window, if it is possible?
  4. Determine dose in water and dose rate in water as a function of cooling time.

For this, I created the following input file, which is attached (dpa-dose.flair (30.6 KB) dpa-dose.inp (9.0 KB) ).

  1. For defined DPA’s I activated the card USRBIN with part DPA-SCO.
    1.1) Could you please help me to understand the reason of having exactly the same results for the cases of RADDECAY, IONTPANS, IRRPROFI, DCYTIMES cards enabled or disabled?
    In my opinion, it should be different because, when the cards (RADDECAY, IONTPANS, IRRPROFI, DCYTIMES) are disabled the results are given per electron per bin and in other cases for the irradiation profile per bin! Is this right?
    1.2) What’s wrong or missing in the input file?
    1.3) How to calculate gas production during irradiation (if there is any)?
    1.4) Could you tell me how can I determine which window material is better to use (what physical parameters need to be calculated)?
  2. Is it possible to estimate the temperature gradients of the irradiation window?
  3. For defined dose I activated the card USRBIN with part DOSE.
    3.1) Why are the results exactly the same again (1.1)? What’s wrong or missing in the input file?
    3.2) Why there is no DOSE USRBIN results with cooling time?

How can I get the graphics shown on this presentation (https://indico.cern.ch/event/442634/contributions/1096523/attachments/1186817/1721027/18_Activation_2015.pdf)slide, pages (23,31-33)

Thanks again and sorry for so long letter!

Best regards
Zohrab

Dear @zamirkhanyan

Few general comments:

  • If you are interested in the scoring of quantities related to residual radiation as result of material activation you have to link the scorings with the cooling times of interest using the DCYSCORE card. Please take a close look to this presentation. Note that many of your scorings are not linked to any cooling time.
  • In general, when results are requested for given cooling times, they are expressed per unit time. Note 7 in RADDECAY | FLUKA.
  • Your materials will be damaged as a consequence of the prompt radiation, the contribution of the residual radiation should pale in comparison. Therefore you don’t need to activate radioactive decay for your DPA calculation.

At this point, questions 1.1) and 3.1) should be addressed. Just to add that you were obtaining the same thing with and without RADDECAY because your USRBINs are not linked to any DCYSCORE and therefore you were obtaining the DPA for the prompt irradiation both times.

1.2) At a first glance and apart from the issue with the cooling times, I see nothing strictly wrong. There are unnecessary and/or arguable things, some of them already mentioned in my previous reply (i.e. the sternheimer parameters and the number of bins, which still rather huge) and some of them new (e.g. the LOW-MAT card: you don’t need the low-energy neutron cross sections unless you expect neutrons to be created in your simulation.).

1.3) Please take a look to the discussion I mentioned earlier: Hydrogen and Helium production

1.4) I am not experienced in this so I will rather save this question for other colleagues here in the forum who may have more knowledge on the topic and who would be willing to shed some light. :triangular_flag_on_post:

2.) The closest related quantity to get from FLUKA would be the map of energy deposition in your window. Then one could make use of this map to estimate the temperature gradients, typically with the help of some engineering simulation software like Ansys.

3.2) Not sure what do you mean. You have dose results but not associated with any cooling time.

About the graphics, I would need you to be more specific. Do you have problems plotting the 2D maps as in slide 23? Or getting the results associated to the different cooling times? Or something different?

Kind regards,
Francisco

Dear Francisco
Thanks a lot for reply!
Could you take a look at the attached image, is this the wrong way to the USRBIN linked with DCYSCORE ?

Best regards
Zohrab

Dear @zamirkhanyan,

That’s OK.

Cheers,
Francisco

Dear Francisco

But the value is zero again (…-32.bin) and I can’t figure out what the problem is!
Please tell me why and how to fix this error?

Thanks in advance
best regards
Zohrab

Dear @zamirkhanyan,

I will assume that you are interested in the residual dose rate received by your window, which seems to be what you are trying to get here, but please make sure this is really what you want.
If so, your result seems normal. Please note that the activation levels in your case would be very low. You would need to run quite a lot of primaries to get a meaningful result and most probably you will need to resort to some biasing technique in order to increase your statistics (i.e. LAM-BIAS | FLUKA, this presentation on biasing techniques may be of help if so).
You may also want to include electronuclear interactions (see slide 23 of this other presentation about hadronic interactions)

If you want to check that your input is indeed working, try out with more energetic electrons, or change the particle type to protons. You will see that you have indeed some dose in your window.

Kind regards,
Francisco

Dear Francisco

Thanks again, very helpful information!
I would like to calculate the residual dose rate in water as a function of cooling time.
This allows me to create graphics similar to presentation pages 31-33 (for water).
And the same I would like to do for DPA of windows depending on irradiation time!

But it looks like something is going wrong with my simulation and I couldn’t figure out what is problem?
Because the residual dose rate during cooling is zero, and DPA window remains unchanged (independent of irradiation time).

Thanks and best regards
Zohrab

Dear @zamirkhanyan,

That is the expected behaviour, as explained earlier. There is nothing wrong with it.

  1. Forget about residual radiation for your DPA calculation. The value you obtain is only the result of the prompt irradiation impinging in your window and has nothing to do with your irradiation profile. Then you always get the DPA per primary particle. The normalization of this value is up to you. If you want to know the DPA after 3600 s of irradiation and you have a continuous intensity of 10^6 electrons per second, then you have to multiply the obtained value by 3600\cdot 10^6 electrons.

  2. The residual dose rate is the dose delivered by radioactive decay products. Radioactive decay in your simulation would be the result of the activation of your materials due to the interaction of your primary electron beam with them. If there is no activation, there is no residual dose rate. And if there is very low activation, there is very low residual dose rate. If the residual dose rate is very low, then it may be difficult to estimate because you will need to run a large number of primaries before you are able to see something (and/or use biasing techniques, see again my previous reply).
    The activation in your setup is very low, note that you have a relatively low-energy electron beam.

Again I would suggest you to change your electron beam by a proton beam and/or increase the energy of your primary particles. Then you should have more activation and therefore larger dose rates (faster to estimate). This way you will see if everything is working as intended or not. Play with it a bit and once everything is clear come back to your real case. And please, reduce your number of bins until everything is understood, a huge number of them only makes things slow and difficult to look at.

Cheers,
Francisco