Project

General

Profile

Summary of Geant4 Hadronic Models Evolution from Release 9.6.p04 through 10.2.p02

This summary is based on Geant4 validation rounds of releases 9.6.p04, 10.1.p03, and 10.2.p02.
Results in a form of images are largely available via Validation Repository: http://g4validation.fnal.gov:8080/G4WebAppNG/DisplayTest.xhtml

Selected results are included in this document to illustrate the case.
NOTE: due to caching reasons, before viewing the plots, please click here first, then return to this page (otherwise, there maybe an error - to be fixed shortly).

Please note that the latest public Geant4 beta release, 10.3.b01, has also been validated, and for most part shows results similar to those from 10.2.p02.
The exception to this is the QGS model (high energy) which is undergoing major re-design and re-tuning.
More details on the Geant4.10.3-series will be released in the near future.

Low Energy Models and Processes

  • PreCompound model
    PreCompound model is fairly stable between releases 9.6.p04 and 10.1.p03.
    However, this model is undergoing some updates currently (10.2- and 10.3-series), including code structure improvement and addition provisions for better configurability.
    Its physics performance is being monitored, and does not present serious concerns at this time.
    Please beware that its technical performance was affected, and present occasional crashes earlier in the 10.2-series, in particular in release Geant4.10.2.p01.
    However, release 10.2.p02 is safe.
  • Capture and Annihilation processes
    Bertini cascade model is used to model pi- and mu- capture processes starting 9.6-series and beyond.
    Bertini cascade has somewhat changed between release 9.6.p04 and the 10.x-series; the effect can be seen for both pi- and mi- captures, especially on heavier targets.
    No significant changes should be expected through the 10.x series, up to release 10.2.p02.
    FTF model is used to simulate antiproton annihilation processes.
    There are some variation through the 10.x development cycle, as compared to results from 9.6.p04; however, changes are not very large.
  • Gamma-nuclear interactions
    Bertini cascade model is used to simulate gamma-nuclear interactions starting release 9.6.
    Subtle changes can be observed between simulated results from 9.6.p04 and the 10.x series.

Intermediate Energies (up to ~10GeV)

High Energy

  • String models:
    • FTF
      FTF string model has undergone substantial update and re-tuning throughout the 10.x family of releases.
      Changes can be seen starting from simulated results from 9.6-series and thought the entire 10.x cycle up to release 10.2.p02.
      The updates largely targeted to improve modeling of hadron productions, including pion production, resulting from interaction of high energy proton beam with nuclear targets.
      Simulated results are typically benchmarked vs experimental data obtained with a 31 GeV/c or 158 GeV/c incident on a Carbon target.
      In general, FTF results from release 10.2.p02 appears to be in a better agreement with the experimental data as compared with FTF results from earlier Geant4 cycles.
      As of now, FTF appears to give better overall agreement with the experimental data at the higher energy end (e.g. at 31 GeV/c and/or at 158 GeV/c) than QGS.
      Selected results are included for illustrative purpose:
    • QGS
      The QGS model is currently a subject to code restructuring and subsequent re-tuning.
      The revision requires non-negligible amount of efforts, it has been decided to split it into steps; some of the changes are already included in the standard Geant4 distribution,
      and more updates and improvements are expected in the near future.
      The updates have lead to certain changes that can be observed in results from 10.2.p02 as compared vs results from 9.6-series or results from 10.1.p03.
      In general, some simulated results from QGS are currently in better agreement with the data, while some other results show certain degradation.
      Users should beware that the work is largely in progress, and better tune of this model is expected in the future.
      Selected validation plots are included for illustrative purpose:

Status of Selected Physics Lists

  • FTFP_BERT
    The FTFP_BERT is the principal physics list recommended for majority of the simulation tasks, including applications in the intermediate and/or high energy ranges.
    Modeling of high energy hadronic interactions by FTF has improved over the past several cycles as compared vs experimental data.
    However, users may observe some changes also in the several-GeV range as compared to earlier releases.
  • QGSP_BERT
    Users of the QGSP_BERT physics list should expect non-negligible changes in the simulated results, in particular, at the high energy end, as QGS model is undergoing major revision.
  • NuBeam
    In general, this physics list targets to model interaction of high energy proton beam with a light target, such as Carbon or Beryllium (with possible subsequent re-interactions of secondary hadrons in the target and/or surrounding materils).
    It employs QGS model to simulate interaction of the primary proton beam of high energy with the target, thus users should expect certain changes in the simulated results due to ongoing revision of QGS.
    Certain changes can also be expected due to variations in FTF model that covers the energy range down to 3GeV in this physics list.
    More detailed information on NuBeam (including statistical analysis) will be p[rovided in the near future, in a separate document.
  • Shielding and ShieldingM
    Shielding physics list has a composition similar to FTFP_BERT but uses High Precision (HP) model for low energy neutron and more precise simulation of radioactive decays.
    Some variations in simulated results can be expected due to changes in FTF as compared vs earlier releases.
    Modeling of radioactive has been improved and expanded in the 10.2-series; bug fixes are included in release 10.2.p02.
    ShieldingM physics list makes more extensive use of Bertini cascade model (up to 9.9 GeV) as compared to Shielding; it also employs HP model for low energy neutrons and precise modeling on radioactive decays.
    No significant changes are expected in the intermediate energy range through the 10.x family, since Bertini is stable, although some variations may appear as compared vs results from 9.6-series.
    If this physics list is used for modeling high energy interactions, some changes in results may be observed due to updates in FTF model.