Home > Research > Publications & Outputs > A comparison of MCNP6-1.0 and GEANT 4-10.1 when...

Electronic data

  • S D Monk et al TRIUMF FINAL3

    Accepted author manuscript, 2.35 MB, Word document

    Available under license: CC BY-NC: Creative Commons Attribution-NonCommercial 4.0 International License


Text available via DOI:

View graph of relations

A comparison of MCNP6-1.0 and GEANT 4-10.1 when evaluating the neutron output of a complex real world nuclear environment: the thermal neutron facility at the Tri Universities Meson facility

Research output: Contribution to Journal/MagazineJournal articlepeer-review

<mark>Journal publication date</mark>15/05/2017
<mark>Journal</mark>Nuclear Instruments and Methods in Physics Research Section B: Beam Interactions with Materials and Atoms
Number of pages14
Pages (from-to)48-61
Publication StatusPublished
Early online date1/04/17
<mark>Original language</mark>English


Abstract A comparison of the Monte Carlo based simulation codes MCNP6-1.0 and GEANT4-10.1 as used for modelling large scale structures is presented here. The high-energy neutron field at the Tri Universities Meson Facility (TRIUMF) in Vancouver, British Columbia is the structure modelled in this work. Work with the emphasis on the modelling of the facility and comparing with experimental results has been published previously, whereas this work is focussed on comparing the performance of the codes over relatively high depths of material rather than the accuracy of the results themselves in comparison to experimental data. Comparisons of three different locations within the neutron facility are modelled and presented using both codes as well as analysis of the transport of typical neutrons fields through large blocks of iron, water, lead and aluminium in order to determine where any deviations are likely to have occurred. Results indicate that over short distances, results from the two codes are in broad agreement – although over greater distances and within more complex geometries, deviation increases dramatically. The conclusions reached are that it is likely the deviations between the codes is caused by both the compounding effect of slight differences between the cross section files used by the two codes to determine the neutron transport through iron, and differences in the processes used by both codes.