Fig. 1. Dependence of the neutron yield per unit muon track length on muon energy for C₁₀H₂₀ scintillator. The experimental data represent measurements at depths varying between 20 and 5200 m w.e. with the corresponding underground muon spectrum. Here they are plotted as a function of the mean muon energy. This approximation is justified in Papers 1 and 2, where references to the original experiments can be found. Note the higher neutron production rate at low energies in present FLUKA-2003 simulations compared to Papers 1 and 2.

Fig. 2. Relative contribution of individual processes to the total neutron yield in scintillator from the GEANT4 simulation. The processes shown explicitly are photonuclear interaction of gammas (γ→N), muon spallation (μ→N), proton spallation (p→N), pion (π⁺ and π^-) spallation (π→N), π^- absorption at rest (π^- abs) and neutron inelastic scattering (n→N). The processes gathered under ‘others’ include electronuclear reactions (e^±→N), kaon spallation and K^- absorption at rest, as well as spallation reactions involving light fragments (²H, ³H, ³He and α-particles), anti-nucleons (, ) and short-lived hadrons (Λ, Σ, Ξ, etc.).

Fig. 3. Differential energy spectrum of muon-induced neutrons produced in scintillator. The GEANT4 absolute yield is compared with the relative FLUKA spectrum represented by the parameterisation given in Paper 2. Experimental data refer to an average neutron spectrum in the scintillator volume as measured by the LVD experiment [9]. The FLUKA and LVD datasets were normalised to the GEANT4 spectrum for visual agreement.

Fig. 4. Lateral distribution of neutron inelastic interactions from the primary muon track for the muon spectrum at about 2.8 km w.e. (280 GeV mean energy). The FLUKA ‘star density’ is compared with LVD data [9] and with a similar quantity calculated with GEANT4. Normalisation is done to the GEANT4 distribution for visual agreement at small distances.

Fig. 5. Dependence of the neutron yield on the average atomic weight of the material in FLUKA (Paper 1) and GEANT4 (this work) for 280 GeV muons, for the elements and compounds indicated. The power-law parameterisations are given in the text. Two data-points show the present FLUKA-2003 results for scintillator and lead.

Fig. 6. Contribution of different processes to the neutron yield as a function of atomic weight of the material for the GEANT4 simulation of . The processes include photoproduction (γ→N), neutron inelastic scattering (n→N), π⁺ and π^- spallation (π→N), muon spallation (μ→N) and proton spallation (p→N).

Fig. 7. Differential cross-section of neutron production by 190 GeV muons for a 10 MeV threshold in neutron energy. The data points represent the results of the NA55 experiment. The thin-line histogram shows the GEANT4 simulation considering muon–nucleus interaction only; the thick histogram includes all physics processes. The dashed line represents the FLUKA results for the latter case.

Fig. 8. Neutron energy spectrum at the rock/cavern boundary obtained with FLUKA and GEANT4. The curve labelled ‘F’ includes only those neutrons entering the cavern for the first time; the curve marked ‘F+R’ accounts also for neutrons reflected back from the walls after crossing the empty laboratory. Also shown are the differential spectra for gammas and electrons (e^- and e⁺) at the rock face.

Fig. 9. MC geometry model for a 250 kg liquid xenon module shielded by 50 cm of hydrocarbon scintillator and 30 cm of lead . The figure also shows a GEANT4 event in which a 280 GeV muon produces two neutrons (n₁, n₂) in the lead shielding which are captured in the hydrocarbon veto (C₁, C₂).

Fig. 10. Energy spectra of muon-induced neutron fluxes across several boundaries. The thick line represents the flux at the rock face for an empty cavern (curve labelled ‘GEANT4 - F+R’ in Fig. 8). The spectrum of neutrons exiting the hydrocarbon veto obtained with FLUKA (Paper 3) is also shown.

Fig. 11. Differential spectra of the total energy deposited in the liquid xenon (LXe) target as predicted by GEANT4 and FLUKA and in the veto scintillator according to GEANT4 (the latter is scaled down by a factor of 5×10⁴).

Fig. 12. NR energy spectrum in the liquid xenon detector as a function of the visible energy deposited by all nuclear recoils in each event. The spectra include ‘mixed’ events involving electromagnetic energy deposits, not just ‘pure’ nuclear recoils, but only the energy left by NRs was counted.

Fig. 13. Distribution of the NR multiplicity in the 250 kg xenon target, for all events containing neutron elastic scatters. The last data-point includes multiplicities of 20 and above.

Fig. 14. Kinetic energy of xenon recoils as a function of neutron energy in the FLUKA simulation (filled markers) and for a log-uniform neutron distribution with GEANT4 (grey-scale). In both MC codes, neutron elastic scattering is treated by different models below and above .

Muon-induced neutron background (NR events per year) in 250 kg xenon target for several detection thresholds