I am following this paper for Diamond Defects and have used basic input-file parameters from the Flair tutorial.
The neutron beam hits my SiGe RPP target and I would like to calculate DPA as a result.
I used the MAT-PROP card, but I don’t understand the inputs for ‘to Mat’ and ‘Step’. I then used DPA in scoring (DPA-SCO particle type on the USRBIN card), but I am not sure about X,Y, and Z min/max and NX,NY,NZ user inputs.SiGE.inp (1.6 KB)
Could someone explain these prompts?
Hi! Can I assume you are continuing on the same problem as you explained in this thread? If yes, indeed the MAT-PROP card is required to properly set the displacement damage threshold energy for the target material in the following way: WHAT(1) sets the dpa threshold in eV. Both WHAT(2) and WHAT(3) are not used. In case your target is composed of a single material which you specify as “SiGe” in your input file, both WHAT(4) and WHAT(5) can be set to “SiGe”. Then WHAT(6) sets the step size to loop over material indices, since you would be considering only one material it is not important so it can be left to its default value 1.0. The SDUM is correctly set to “DPA-ENER” since you want to set the threshold for this material.
Then indeed the proper scoring mesh needs to be implemented. The USRBIN card takes into account the minimum and maximum extent along each coordinate (Xmin, Xmax, …) and the amount of bins within these extents along each axis (Nx, Ny, Nz). The best indications to set these parameters is to look at the size and shape of your beam: you defined in your input file a divergent beam, hence in your target material you need to set the extent of the scoring mesh such that it covers the area where neutrons impact the target material. Since your target is quite small you could choose to have the USRBIN mesh to cover the entire RPP volume. The amount of bins have to be set such that after collecting sufficient statistics, when you plot the results this shows a smooth distribution of values which is properly resolved (i.e. the bins are not too large in volume such that local high dpa values are not obscured by averaging inside one bin).
Hope this is helpful,
Please see the attached.
I know that DPA is expressed as average DPAs in each bin per unit primary weight. Could you explain what ‘per unit primary weight’ means? What are the units? I think that once I understand this, I will be able to understand what the DPA x and y values should be.
I obtained a graph, but it is hard to tell if the graph is sufficient enough to give me the information I need since I am not able to clearly visualize what is happening at my target.
SiGE.inp (1.7 KB) DPA.pdf
Per unit primary weight means per primary particle you simulated. Since you are not using any biasing, every primary particle that is simulated has an equal ‘weight’ when averaging the results. As you are using neutron particles as a source, the results you plot are expressed per neutron. The dpa value in FLUKA is normalized by the volume of the bin itself and therefore the value in each bin is a dimensionless quantity expressed per unit primary weight.
In your plot the neutron particles predominantly impact at coordinate (0,0) in the XY plane. If you are only interested in the peak dpa values at this point, the picture could be more clear if you restrict the scoring mesh to a much smaller area around this coordinate and use smaller bin sizes. You can also see the individual paths of secondary particles, in the volume you are interested in (i.e. the volume you want to score) these should be averaged out by collecting more statistics so simulating more primary particles.
I’m still a bit confused about the FLUKA DPA (please see the attached) and the X-Y units for it once I’ve obtained my graph. DPA 20KeV.pdf (135.0 KB)
I’m working on increasing the number of primary particles and my beam size to obtain better results.
Could you explain what you mean by:
-restrict the scoring mesh to a smaller area around the beam position vs. USERBIN mesh cover entire RPP volume?
-use smaller bin sizes? Right now, I have 50BIN.I know that I need a smooth distribution of values, but I don’t don’t quite understand the logic behind choosing the correct number of bins.
-“the bins are not too large in volume such that local high dpa values are not obscured by averaging inside one bins”
-averaging out secondary particles? How is this calculated?
Ultimately, I would like to create a **DPA vs. Kinetic Energy (KeV)**as illustrated in this paper. These graphs have results for neutrons, protons, and pions. How is obtaining this graph different from the DPA that we’re currently discussing and am I able to use this same card to achieve my goal of studying DPA SiGe crystal lattice structure?
Hi, I believe in this case the units of X and Y are [cm] as you show a projection of the DPA in the XY plane of your detector.
The secondary particle tracks you see now are merely a result of not collecting enough statistics by not running enough primary beam particles. In the Monte Carlo approach , if you collect results from more primary particle runs you will eventually arrive at a statistically converged distribution (see the examples in these slides).
What you can see from your plot though is that the beam size you define is very small, at coordinates (0,0) the dpa value is very high compared to anywhere else in the detector. This could be a result of also simply be a result of running a very small amount of primary particles. I suggest you restrict the USRBIN ranges to +/- 1mm in the X and Y directions and within this range vary the number of bins, this you can do by defining multiple USRBIN scoring meshes (setting the amount of bins differently each time: for example 10, 50, 100, 200). Compare the results after running more particles and plot the results for example using the “1D Max” option in the plotting tab in Flair. I think in light of this, this discussion might also be useful for your work.
You can gather results for dpa caused by neutron, proton and pion beams by choosing these particles as particle source using the BEAM card, just as you did for neutrons in the inputfile you attached above. Getting the energy dependence will mean you need to run different simulations at different beam energies by also specifying this in the BEAM card.
Hope this is helpful,
I have more clarity now.
Could you explain what you mean by editing the ‘beam size’? As you know, I currently have 0,0,0, but if I want to increase this size over more of the geometry, how would I be able to do this if there is no xmin/max,ymin/max, and zmin/max?
Also, I am attempting to manipulate the number of bins. In the paper I’ve previously shown you, they used 1BIN. It seems that I’m not able to go below 21BIN. How can I input the 10BIN or anything below 21?
Once I increase beam size and number of primary particles, restrict userbin to smaller area, use smaller bin sizes, how do I obtain the correct DPA value to plot against beam type/energy? Please see the attached. SiGE (2).inp (1.7 KB) DPAUpdated.pdf (34 KB)
The 0,0,0 coordinate you are referring to is the beam source point, not its size. You can adjust the beam size by setting WHAT(4) and WHAT(5) in the BEAM card to the values you need (also in the input file you attached the neutron beam energy is 10.000 TeV, I don’t think this is within the scope of your study).
I believe you are referring to WHAT(3) in the USRBIN card which specifies the logical output number (which is negative if you want the data to be written unformatted), not the amount of bins. You can associate a different output number to each different USRBIN you define to keep the results separate. WHAT(4), WHAT(5) and WHAT(6) in the continuation card specify the number of bins along X, Y and Z which you set to 100 for each, this you can adjust to 1 if you like. The USRBIN scoring DPA will directly give you the DPA values in each bin, as mentioned before: plot the 1DMax profile for the different amount of bins and check if you collected enough statistics and if the distribution is properly resolved.
DPAUpdated.pdf (35.9 KB)
Please see the attached.
Could you explain the output units for DPA in regards to the ‘per primary particle’ you mentioned earlier?
Also, is there a way to extract the data from a BIN folder after I have obtained my results? When I attempt to view my data, the file isn’t able to be retrieved. My ultimate goal is take specific DPA units to plot in excel as a function of beam type and energy.
Please see the attached. I’m not able to view the units on the color scale.
As explained in the manual, the results directly give the dpa quantity. The labels on the color bar in your plot are not visible but this should readily give you the dpa value which is in units of dpa. However if you plot using the 1DMax option in Flair, the values will be shown as a step curve. In the directory where you have opened Flair there should be a text file with suffix “.dat” which contain the plotted values and the errors according to the description in the header. From this you can copy the values to any postprocessing tool you like. Does this solve your problem? It also looks like you defined a rectangular beam profile with 1 cm^2 area, I don’t know if this is very realistic if you are trying to reproduce the results form the paper you are referring to.
I’m still a bit confused about the graph that I obtained. From my understanding so far and for clarification, the description that I have on the x and y axis is correct while the color scale reflects the amount of DPA in a particular region of my geometry.
Could you clarify the section of the manual you are referring to? Attached is the plot I obtained when I used IDMax. How do I obtain the MAX DPA value from this plot?
DPA IDMax.pdf (10.2 KB)
I was also able to find the .dat file. I see that there are values for Horizontal DPA and Vertical DPA, but I’m not sure how to determine the value needed for this particular beam type and energy.
The beam profile was chosen to ensure that the entire RPP geometry was irradiated.
The X and Y axis in your previous plot are spanning the size of your detector, however to be more precise on the color scale: this reflects the amount of DPA in each bin of your Cartesian scoring mesh.
The section of the manual I was referring to just served as clarification on how to use the USRBIN cards and what it gives as output (in your case simply the dpa per primary). What you show in your plot is the DPA 1DMax profile along the Z-axis, not the X-axis as you indicate in the plot. This means that for a given Z-coordinate, the bin with the highest DPA value will be plotted, regardless of the X- or Y-coordinate. Since in your DPA scoring mesh as defined using USRBIN you only have 4 bins along the Z-axis and restrict the plot to the positive values, only two values are plotted (for the second one only the error bar is shown in Flair but the data file contains the actual value as you show.)
The data file itself (as you show in your attached picture) has to be interpreted as follows: the first 14 lines contain the header information, stating the “Scanning axis” which is Z in your case and the scoring bin limits you take into account for plotting (lines 12-14). The lines directly after that list the actual data in column-format, in sequence: the bin’s lower Z-boundary, the bin’s upper Z-boundary, the DPA value, and the error (in %). The bottom two blocks of data (lines 17 and onwards) refer to the positions of the bins with maximum DPA value along the X and Y axes. Try to produce a plot with either X or Y axis as “scanning axis” and check the variation in the results. However if you irradiation your entire RPP geometry it is possible that there is not much difference between the bins and you also get a uniform DPA distribution. In case you don’t irradiate the full RPP geometry you should arrive at a smooth distribution along X or Y using 1DMax. Hope this helps,
I changed the DPA IDMax profile for the X-axis to be my scanning axis.
DPA IDMax X-axis.pdf (11.6 KB)
Lines 14 and 15 contain different sets of numbers. Can you specify which set of numbers you are referring to for the Max DPA value? Ultimately, my goal is to record only the incident data when the beam hits the target. Attached is the new data file, so I’m no longer sure about which lines to follow.
SiGeDat.txt (3.4 KB)
Also, could you elaborate on the reason why you stated that “It also looks like you defined a rectangular beam profile with 1 cm^2 area, I don’t know if this is very realistic if you are trying to reproduce the results form the paper you are referring to.” By beam profile, do you mean target? I’m just wondering why the geometry modification would make a difference or are you just referring to me directly following the paper?
if you have a square beam of 1 cm edge and 1 cm2 area and another one of 0.5 cm edge and 1/4 cm2 area, for the same beam intensity the second features a beam particle density four times higher in the central region. Therefore, the resulting number of DPA per beam particle in first approximation is expected to be there four times higher. So the input beam size matters, a lot.
In the file you attached, there is a first block of data under the expressive header Max DPA, whose lines contain four numbers, already described by @anwaets above. Here you decided to get the profile along X, so you read there the X bin limits, with the respective Max DPA in the third column, extracted from a transverse YZ area as defined by the Y and Z intervals you selected. The resulting plot clearly indicates in which X range the beam hits the target, according to your beam specs.
I’ve found the Max DPA value in the file. Do you have an idea for why I keep getting the same value when I change the beam energy? I’m looking to plot kinetic energy KeV vs. Max DPA
It depends on the energy range and particle types you are spanning.
I’ve been simulating a Neutron beam at 10 Kev and I even tried to run it for 10 MeV and and higher. I am still getting the same results when I extract the data.
Note that the kinetic energy range of interest in the paper you are referring to is between 1 MeV and 100 GeV. However for the energies you simulated it is important to set the particle transport and production cuts in the proper way. In the DEFAULTS card, please set the SDUM to DAMAGE instead of PRECISIO. This in particular resets the particle transport threshold to 1 keV/n for hadrons, muons and light ions which should yield more accurate results.
I have a question. For a square beam of 1cm^2 area and 1/4 cm^2 area, does it mean that one beam particle uniformly distributed on the 1cm^2 area and 1/4cm^2 area?
It means, if you have asked for a Rectangular Shape, that the beam is uniformly distributed over the corresponding area. One single beam particle is one point inside the area. Simulated beam particles, one after the other, are uniformly sampled over the area, such as the resulting distribution of points covers uniformly the area (provided that the run statistics is good enough).