The LHCf measurement of forward-neutron production in s=13\sqrt{s} = 13 TeV pp collisions (JHEP 07 (2020) 016) found that in the most forward pseudorapidity bin (η>10.75\eta > 10.75) the neutron energy spectrum exhibits a peak near 5 TeV that no hadronic interaction model predicts, that all generators underestimate the production cross section there by at least 20%, and that none reproduces the neutron elasticity distribution. The physical mechanisms responsible - single diffractive dissociation, one-pion exchange, or non-diffractive beam-remnant hadronization - cannot be separated with LHCf data alone, since the detector sees only neutral particles in one very forward direction. The problem is to identify and quantitatively decompose these mechanisms; the route foreseen by the collaboration is the combined LHCf-ATLAS analysis (common data acquisition, central-region event classification, central-forward correlation measurements), so that the specific physical improvements needed in the interaction models are identified rather than absorbed by phenomenological tuning. Status: open. Generator developments (QGSJET-III's t-channel pion exchange; the PYTHIA 8 forward tune's suppression of the popcorn mechanism - meson production between baryon-antibaryon pairs during string fragmentation - in beam remnants) partially reduce the discrepancy but disagree on the mechanism and are not simultaneously consistent across energies; no combined LHCf-ATLAS neutron physics analysis has been published, and the joint program remains at the detector-performance and Monte Carlo feasibility stage.

Background

Very-forward particle production in proton-proton (pp) collisions at the Large Hadron Collider (LHC) is governed by soft, non-perturbative QCD at longitudinal momentum fractions near unity, where no first-principles calculations exist and phenomenological event generators are the only quantitative tool. The LHCf ("Large Hadron Collider forward") experiment is a pair of compact sampling calorimeters (detectors that measure particle energies by absorbing the particle shower in alternating layers of dense absorber and active material) located 141 m from the ATLAS interaction point, covering pseudorapidity η>8.4\eta > 8.4, where η=−ln⁡tan⁡(θ/2)\eta = -\ln\tan(\theta/2) measures the polar angle (η>10.75\eta > 10.75 corresponds to scattering angles below roughly 42 microradians). LHCf measures neutral particles - neutrons, photons, and π0\pi^0 mesons - carrying a large fraction of the beam energy. These measurements matter beyond collider physics because the same hadronic interaction models (QGSJET II-04, EPOS-LHC, SIBYLL, DPMJET, PYTHIA) are used to simulate extensive air showers, the cascades of secondary particles produced when ultra-high-energy cosmic rays (energies above about 101510^{15} eV) strike the atmosphere. Forward neutrons are leading baryons: the elasticity kk is the fraction of the primary proton's energy carried by the leading particle, and the inelasticity 1−k1-k is the fraction transferred to other secondaries, which controls how deep such showers develop.

In 2020 the LHCf Collaboration published the most granular measurement to date of forward-neutron production in pp collisions at s=13\sqrt{s} = 13 TeV: the inclusive differential cross section dσn/dEd\sigma_n/dE in six pseudorapidity bins from η=8.65\eta = 8.65 to η>10.75\eta > 10.75, from which the neutron energy flow, production cross section, and average inelasticity ⟨1−kn⟩=0.536−0.037+0.031\langle 1-k_n\rangle = 0.536^{+0.031}_{-0.037} were derived. Two findings stand out. First, in the most forward bin (η>10.75\eta > 10.75) the measured spectrum shows a peak structure at around 5 TeV - close to the beam energy - that is not predicted by any of the tested generators, and all models underestimate the total neutron production cross section in this bin by at least 20%. Second, this is not an isolated pathology: the same high-elasticity excess was already visible in the earlier 7 TeV measurement (Phys. Lett. B 750 (2015) 360) and in the 2018 inclusive 13 TeV measurement (JHEP 11 (2018) 073), and no generator satisfactorily reproduces the neutron elasticity distribution either.

Physically, a neutron at η>10.75\eta > 10.75 with energy near the beam energy can arise from several distinct mechanisms: leading-baryon production in non-diffractive events, where the beam remnant (the part of the proton left carrying the remaining momentum after the partonic interaction) hadronizes into a baryon; single diffractive dissociation, in which the proton emerges intact or as a low-mass excitation; and one-pion exchange (OPE), a tt-channel process in which the proton fluctuates into a neutron plus a virtual pion that carries the interaction. Each mechanism predicts a different correlation between the forward neutron and particle activity in the central region. Because LHCf only sees neutral particles in one very forward direction, it cannot distinguish these mechanisms event by event. The 2020 paper therefore envisaged a combined analysis with ATLAS, with which LHCf shared common data acquisition during the same LHC fills: using ATLAS central-region information to classify events (for example diffractive versus non-diffractive, as studied in Monte Carlo by Zhou et al., Eur. Phys. J. C 77 (2017) 212) and to measure the correlation between central and forward hadron production (as studied by Ostapchenko et al., Phys. Rev. D 94 (2016) 114026).

Problem Statement

Which physical mechanisms - for example single diffractive dissociation, one-pion exchange, or non-diffractive beam-remnant hadronization - produce the very-forward neutrons observed in s=13\sqrt{s} = 13 TeV pp collisions, and can they be separated and quantified so as to explain (i) the peak structure at around 5 TeV in the η>10.75\eta > 10.75 differential neutron spectrum, which no current hadronic interaction model predicts, (ii) the at-least-20% underestimate of the total forward-neutron production cross section in that bin by all generators, and (iii) the disagreement between the measured neutron elasticity distribution and all model predictions? The route foreseen by the source is the combined LHCf-ATLAS analysis exploiting the common data acquisition of the two experiments: using ATLAS central-region information to classify events (diffractive versus non-diffractive processes, one-pion-exchange topology) and to measure the amount of correlation between central and forward hadron production, with the goal of identifying on physical grounds - rather than through phenomenological tuning - the specific improvements that hadronic interaction models need in order to describe forward baryon transport at LHC energies, and hence baryon transport in extensive air showers.

References

  1. O. Adriani et al. (LHCf Collaboration), Measurement of energy flow, cross section and average inelasticity of forward neutrons produced in s=13\sqrt{s} = 13 TeV proton-proton collisions with the LHCf Arm2 detector, JHEP 07 (2020) 016, DOI 10.1007/JHEP07(2020)016, arXiv:2003.02192, https://arxiv.org/abs/2003.02192 (source of the open question)
  2. O. Adriani et al. (LHCf Collaboration), Measurement of inclusive forward neutron production cross section in proton-proton collisions at s=13\sqrt{s} = 13 TeV with the LHCf Arm2 detector, JHEP 11 (2018) 073, DOI 10.1007/JHEP11(2018)073, arXiv:1808.09877, https://arxiv.org/abs/1808.09877
  3. O. Adriani et al. (LHCf Collaboration), Measurement of very forward neutron energy spectra for 7 TeV proton-proton collisions at the Large Hadron Collider, Phys. Lett. B 750 (2015) 360, DOI 10.1016/j.physletb.2015.09.041, https://doi.org/10.1016/j.physletb.2015.09.041
  4. Q.-D. Zhou, Y. Itow, H. Menjo, T. Sako, Monte Carlo study of particle production in diffractive proton-proton collisions at s=13\sqrt{s} = 13 TeV with the very forward detector combined with central information, Eur. Phys. J. C 77 (2017) 212, DOI 10.1140/epjc/s10052-017-4788-7, arXiv:1611.07483, https://arxiv.org/abs/1611.07483
  5. S. Ostapchenko, M. Bleicher, T. Pierog, K. Werner, Constraining high energy interaction mechanisms by studying forward hadron production at the LHC, Phys. Rev. D 94 (2016) 114026, DOI 10.1103/PhysRevD.94.114026, arXiv:1608.07791, https://arxiv.org/abs/1608.07791
  6. S. Ostapchenko, QGSJET-III model of high energy hadronic interactions. II. Particle production and extensive air shower characteristics, Phys. Rev. D 109 (2024) 094019, DOI 10.1103/PhysRevD.109.094019, arXiv:2403.16106, https://arxiv.org/abs/2403.16106
  7. M. Fieg, F. Kling, H. Schulz, T. Sjostrand, Tuning pythia for forward physics experiments, Phys. Rev. D 109 (2024) 016010, DOI 10.1103/PhysRevD.109.016010, arXiv:2309.08604, https://arxiv.org/abs/2309.08604
  8. T. Sako, Hadronic interaction at LHC: review for the LHCf and air shower related measurements, PoS(UHECR2024) 032, DOI 10.22323/1.484.0032, https://doi.org/10.22323/1.484.0032
  9. K. Ohashi, H. Menjo, Y. Itow, T. Sako, A simulation study for one-pion exchange contribution on very forward neutron productions in ATLAS-LHCf common events, PoS(ICRC2021) 190, DOI 10.22323/1.395.0190, https://doi.org/10.22323/1.395.0190
  10. H. Kobayashi, Performance studies of LHCf and ATLAS-ZDC for very forward neutron measurement, PoS(ICRC2025) 303, DOI 10.22323/1.501.0303, https://doi.org/10.22323/1.501.0303
  11. H. Kobayashi, Performance of hadronic shower measurement with LHCf and ATLAS ZDC detectors, EPJ Web Conf. 320 (2025) 00043, DOI 10.1051/epjconf/202532000043, https://doi.org/10.1051/epjconf/202532000043
  12. O. Adriani et al. (LHCf Collaboration), Current Status and Future Prospects of the LHCf Experiment, Particles 9 (2026) 34, DOI 10.3390/particles9020034, https://doi.org/10.3390/particles9020034