Created attachment 594 [details] Photoneutron spectra from Nickel (n,gamma) source Top: experimental Bottom: Geant4 I'm running a Geant4 simulation to detect photoneutrons from targets illuminated by low energy (7-20 MeV) gamma rays. To validate results I have set up a geometry which duplicates one for which experimental data exists. In the experiment, a monoenergetic, collimated horizontal beam of gamma rays strikes a 2.54 cm cube of material in air and photoneutrons are detected by a cylindrical active volume (4.84 cm diameter, 15 cm length) 3He ionization chamber neutron detector. The detector symmetry axis is parallel to the beam direction and offset horizontally by 5.8 cm. An example of a spectrum using a Ni(n,gamma) source (8.999, 8.533, 7.819, 7.724, 7.534 MeV gamma rays) on a 209Bi target is attached (energy in keV). To validate the model I have modified exampleB4a, from the Geant4 distribution, to make an identical geometry and used the general particle source to generate the gamma ray beam. Data is collected when a neutron intersects the detector volume. I use gtools Root analysis tools to store various properties of the emitted neutron. I am running Geant4 10.5 patch 01 on an i7 PC with 16GB RAM and 64-bit Linux (Fedora 30). I am using the ShieldingLEND physics list. An example of the 209Bi(gamma, n) neutron spectrum for two billion 8.999 MeV gamma rays is also attached (energy in MeV). To compare the simulated and experimental spectra, corrections for beam time and detector intrinsic efficiency must be made. Beam time for the experimental spectrum was ~604800s, whereas the simulation (2x10^9 gammas) corresponds to ~690 s of beam time. Detector intrinsic efficiency, ~5.7015x10^-4 at 1 MeV neutron energy, is not included in the Geant4 simulation. When these are taken into account, we should have Spectrum(simulation) = Spectrum(experiment) * 2.996 if the model and experiment agreed. We have Spectrum(experiment)*2.996=100354 from experiment after correction, Spectrum(simulation)=160508 for ShieldingLEND. However, simulation gives a spectral shape that is completely unlike the experimental data. (Variation of detector resolution - 15 to 35 keV - and efficiency - 0.95 to 0.8 - over the neutron energy range are small and have been neglected.) The simulation gives an absolute count rate which is ~1.6 times the corrected experiment. This is not terrible. However, the neutron spectral shape is completely distorted and makes no sense. I would have expected a set of delta functions (or only one if excited states of the residual nucleus have been ignored), somewhat broadened by scattering. Similar results are obtained at 9.719 MeV gamma ray energy.
Created attachment 610 [details] gamma-n on 209Bi
I've attached a plot of (gamma, n) on 209Bi. This is 10**6 gammas of 8.999 MeV on 209Bi using only the photo-nuclear process in isolation, that is, not part of a physics list. The process invokes G4LENDorBERTModel as in the ShieldingLEND physics list. Based on this shape, I don't see how your "step function" plot could have arisen. I was also looking over your calculation and I could not reproduce your factor of 2.996. I got (690/604800)/5.7x10**-4 = 2.0015. Is this right?
Created attachment 611 [details] 209Bi(g,n) spectrum at Eg 8.999 MeV for QBBC physics list
Thanks for your reply, Dennis. The factor of 2.996 is correct. I quoted the wrong value for the efficiency in my original post (I wrote 5.7015x10^-4 and it should have been 3.8055x10^-4. Your spectrum looks like the spectra I obtained with all of the other physics lists I've tried. (I have attached one for QBBC, but the spectra for QGSP_BIC_HP and Shielding are the same.) My absolute numbers of gammas are ~65113 whereas yours is 181410 for vastly different photon numbers (10^9 vs 10^6) so obviously our geometries must be different. Since my spectra are made under identical conditions except for changing the physics list each time, this implies that there is something weird in the ShieldingLEND physics list. I'll try re-running with Geant4 10.6.1, but I'm expecting to get the same results.
I now reproduce your step function neutron spectrum. I will look into why this is.
I reran the 209Bi(g,n) at 8.999MeV with Geant4 10.6.1 and continue to get the step function spectrum using the ShieldingLEND physics list. I also continue to get the many (~2566) G4Exceptions: -------- WWWW ------- G4Exception-START -------- WWWW ------- *** G4Exception : had012 issued by : G4HadronicProcess:CheckResult() Warning: Bad energy non-conservation detected, will re-sample the interaction Process / Model: photonNuclear / LENDorBERTModel Primary: gamma (22), E= 8.999, target nucleus (83, 209) E(initial - final) = -3729.53 MeV. *** This is just a warning message. *** -------- WWWW -------- G4Exception-END --------- WWWW -------
Created attachment 612 [details] gzipped tar ball of energy spectra obtained before and after problem 2174 fix I'm not sure if this helps in solving this particular problem (2197), but it seems to be related to it and may give insight into it. I reran a bunch of cases after Dennis' fix of Problem 2174. The neutron energy spectra have subtly changed. Previous spectra using Geant4 10.5.1 and Geant4 10.6.1 (results were essentially the same for the two distributions) gave step function spectra with decaying tails (sometimes at the high energy side, sometimes at the low energy side, sometimes both. Spectra for 10.6.1 rebuilt with Dennis' patch for Problem 2174 now gives step function spectra with no tails or very small tails. So perhaps the tails were related to the energy nonconservation problem in some way? Count rates are essentially the same or very similar between the fixed and unfixed spectra, except for Ta and Bi at 7819 keV. I have attached a tar ball (Elogspectra.tar.gz) of energy spectra obtained before and after the problem fix. To decode what the spectrum name means, as an example, PN4_7819keV_Ta1incube_air_109_detection_E_0.png is the photoneutron energy spectrum for 10^9 source gamma rays of energy 7819keV incident on a 1 inch (2.54 cm) cube tantalum target in air. The _0 denotes that the data was obtained without the problem 2174 patch applied to Geant4. PN4_7819keV_Ta1incube_air_109_detection_logE_0.png is the same but on a log scale. PN4_7819keV_Ta1incube_air_109_detection_E.png and PN4_7819keV_Ta1incube_air_109_detection_logE.png are the corresponding spectra with the problem 2174 patch applied to Geant4. [Elogspectra.tar.gz] I will run some more cases to see if other anomalies arise.
Created attachment 616 [details] gzipped tar ball of full set of energy spectra obtained before and after problem 2174 fix I have now run all of the cases for which I have experimental data for eventual verification. All have the same geometry; a 2.54cm diameter beam of monochromatic gamma rays incident on a 2.54cm-sided cubic target in air. Neutrons are counted in a cylindrical volume (diameter 4.8cm, length 15 cm) with axis parallel to the beam direction but offset from the beam center by 6 cm. The targets are different elements near to the doubly closed shell in lead - gold, tantalum, thallium, lead (A=208 and natural abundance) and bismuth. I have attached a zipped tarball of files comparing spectra from G4 10.5.1 and G4 10.6.1 rebuilt with the fix for Problem 2174. Each file is identified by its incident gamma ray energy (keV) and the target material. There are five histogram plots in each file going from left to right top row, then bottom row - one for the 10.6.1 rebuild (E), one for the original 10.5.1 build (E0), a log plot overlaying the two histograms, the difference of the two histograms (E0-E) and the difference of the two histograms divided by the maximum value of the 10.6.1 (E) histogram. Observations: All spectra for 10.6.1(fixed) have step function spectra, except for thallium which has an evaporation spectrum. Differences between 10.5.1 and 10.6.1(fixed) spectra appear as smooth "hills" on top of the step functions or tails. There is a significant difference between 10.5.1 and 10.6.1(fixed) spectra for Bi, Pb and Ta at 7819 keV; Au and Ta at 8534 and 8999 keV. There is a weak difference for Bi at 8534 and 8999 keV and a very weak difference for Bi at 9719 and Pb at 8534, 8999 and 9719 keV. Thallium, the only element having an evaporation spectrum, interestingly shows essentially no 10.5.1/10.6.1(fixed) difference at all energies.
According to the developers of MCGIDI, the interface between LEND models and GIDI, is problematic in many cases. They are planning to have a better version of it in the near future, and so they do not think it's worthwhile to fix the problems in the existing one. When the new MCGIDI is available I will help them implement it into Geant4 and then we can resume our validation efforts. For now I will close this report and move the item into hadronic development. Thanks for your help thus far.
Thank you for your eforts on this problem, Dennis. I look forward to trying out the new MCGIDI on my data when it becomes available.
