WorksPhysics and Astronomy
LHC Concept: Energy-Frontier Antimuon-Proton Collisions at CERN
Abstract
The concept of the LHC, an energy-frontier antimuon-proton collider based on the HL-LHC, is presented. After recalling why energy-frontier lepton-hadron colliders are indispensable as QCD explorers and reviewing the history of muon-proton collider proposals, two ways of realising collisions at the LHC are compared: with muon-collider beams and with ultra-cold beams as proposed for TRISTAN. For the latter, a TRISTAN-type booster and the re-use of the LHeC energy-recovery-linac tunnel are discussed; colliding 1 TeV antimuons with 7 TeV protons gives TeV and peak luminosities of and cms, respectively. The detector concept and the physics potential, including searches for colour-octet muons and R-parity-violating squarks, are discussed together with the main challenges.
Introduction
Lepton–hadron collisions have played an exceptional role in revealing the internal structure of matter: the proton form factors were first observed in electron scattering, quarks in deep inelastic electron scattering at SLAC, and the EMC effect in deep inelastic muon scattering at CERN [1]. HERA, the first and still unique electron–proton collider, extended the kinematic region by orders of magnitude, shed light on the basics of quantum chromodynamics (QCD) and provided the parton distribution functions (PDFs) needed to interpret the Tevatron and LHC data [2].
While the electroweak part of the Standard Model (SM) was completed with the discovery of the Higgs boson [3], large gaps remain in its QCD part: confinement has not been derived from the QCD basics, and hadronisation and nuclearisation are not yet understood. Notably, the Higgs mechanism is responsible for only about 1% of the mass of the visible Universe: MeV, MeV and MeV, whereas MeV, MeV, and visible matter consists of about 74% hydrogen and 24% helium. The remaining 99% originates from the strong interactions.
The region of small Bjorken () at high ( GeV), which is very important for understanding the QCD basics, was not covered by HERA, and the centre-of-mass energy of the EIC [4] is several times lower than that of HERA. The LHeC [5, 6] will touch this region, whereas muon–hadron colliders can reach much higher energies. TeV-scale lepton–hadron colliders are also needed to provide PDFs for an adequate interpretation of the HL-LHC (as well as FCC- and SppC) data. Lepton–hadron colliders are thus unique QCD explorers: Higgs physics requires the highest luminosity, whereas the QCD basics require the highest energy. Table 1 compares the lepton–proton colliders that could operate in the 2030s with HERA: the LHC [7], sketched in Figure 1, would provide a 4.5 times higher centre-of-mass energy than the LHeC, at a comparable luminosity.

Figure 1. Bird’s-eye view of the LHC inspired from TRISTAN.
| Collider | [TeV] | [cms] | [fb] |
| HERA | 0.319 | 0.8 (total) | |
| EIC | 0.105 | 100/year | |
| LHeC (ERL 20) | 0.748 | 60/year | |
| LHeC (ERL 50) | 1.18 | 50/year | |
| LHC (TRISTAN booster) | 5.29 | 88/year | |
| LHC (ERL booster) | 5.29 | 72/year |
Table 1. Lepton–proton colliders of the 2030s compared with HERA; ERL 20 is the single-pass option of Ref. [8].
Muon–proton colliders: from early proposals to the LHC
Muon–proton colliders were proposed three decades ago as an alternative to linac–ring type electron–proton colliders [9, 10, 11, 12]. For a muon ring tangential to the Tevatron, luminosities of – cms were estimated [9], and the potential for leptoquark and contact-interaction searches was explored [11]. As the synchrotron radiation of muons is times weaker than that of electrons, energy-frontier muon–hadron colliders can be ring–ring machines, unlike the linac–ring type electron–hadron ones. Later, FCC- and SppC-based and colliders [13], LHC- and FCC-based colliders [14] and a muon–ion collider at BNL [15] were proposed; see Ref. [16] for a review.
Muon-collider-based option. A muon collider [17] built tangentially to the LHC would provide collisions with the HL-LHC (HE-LHC) protons [18] at –9.16 (12.7) TeV and luminosities of – cms, extending, e.g., the discovery reach for colour-octet muons from 3.5 TeV at the HL-LHC to about 6.5 (9) TeV [18]. This option, however, has to await the solution of the muon cooling problem, with a cooling demonstrator foreseen in the 2030s [17].
Ultra-cold antimuon option. While a low-emittance beam has not been achieved yet, there is an established technology to create a low-emittance beam using ultra-cold muons, developed for the muon /EDM experiment at J-PARC [19]. The TRISTAN proposal [20] showed how such a source can be intensified by orders of magnitude and the beam accelerated to 1 TeV without ionisation cooling. Antimuon rings tangential to the FCC-/CEPC and to the LHC, Tevatron, FCC and SppC were then considered [21], leading to the HL-LHC-based LHC [7], which is now developed further by the HELP4CERN Study Group as the LHC [22, 23, 24].
LHC: antimuon complex and luminosity
From ultra-cold muons to a 1 TeV antimuon beam
The LHC antimuon complex (Figure 2) consists of six stages [7, 20, 25]: (i) a 3 GeV, 2 MW rapid-cycling synchrotron operating at 50 Hz; (ii) a pion-production ring, in which each proton bunch crosses a 10 mm graphite target 40 times, yielding 0.016 pions per proton; (iii) an ultra-cold source, where the pions are stopped in a multi-layer tungsten target and the thermal muonium is laser-ionised; (iv) a linac re-accelerating the to 212 MeV with a normalised emittance , in trains of 40 bunches of ; (v) a booster accelerating them to 1 TeV in about 1 ms; and (vi) a 3 km main ring tangential to the LHC, where the collide with the LHC protons at one interaction point (IP). The ultra-cold yield is thus about times higher than at the present J-PARC MLF beamline, and the low emittance is achieved without ionisation cooling, so that a –hadron collider can be realised well before a collider. The source and the low-energy acceleration are being actively developed [26, 27, 28].

Figure 2. Schematic view of the LHC (LHC) with the ERL-based booster: the booster (“boosting ring”) and the main ring are integrated within the LHeC ERL tunnel, similarly to the TRISTAN layout (from Ref. [7]).
Booster options
Two booster options were studied in Ref. [7]. The TRISTAN-type booster [20] is a racetrack with two 3 km linacs of 30 GeV each and 1 km radius arcs; the reach 960 GeV after 16 turns, in about 0.65 ms. In the second option, proposed by F. Zimmermann, the LHeC energy-recovery-linac (ERL) tunnel [6] — a similar racetrack with two 1 km linacs providing 10 GeV per pass — is re-used (Figure 2), and 1 TeV is reached after 50 turns, in about 1.4 ms. This allows for a natural staging: the ERL is first used for and collisions, possibly starting with the single-pass 20 GeV LHeC [8], and the booster is installed later in the same tunnel. Figure 3 shows the surviving fraction and energy as functions of time for both options: our calculation gives a survival of 0.82 and 0.68, in agreement with the values of 0.82 and 0.67 of Ref. [7],2 i.e. and per bunch at the top energy.
Figure 3. Surviving fraction (solid lines) and kinetic energy (dashed lines) of the vs. time from injection at 212 MeV for the two booster options (this work, parameters of Ref. [7]).
The rapid acceleration of a multi-bunch train within the muon lifetime is the most critical challenge of the booster. Unlike the hybrid rapid-cycling synchrotron of the Muon Collider [17], a vertical-excursion fixed-field accelerator (vFFA) [29] needs no magnet ramping and keeps the 40-bunch train intact, and cascaded vFFA arcs could be used in the LHeC tunnel [7].
Main ring and luminosity
The 40 bunches circulate in the 3 km main ring with a spacing of 250 ns — every tenth LHC bucket — giving a collision frequency MHz. At 1 TeV the muon lifetime is 20.8 ms ( turns); with refilling at 50 Hz, the average population is of the injected one. For the “ERL60-upgraded” LHeC proton beam [5] and matched beam sizes (Table 2), the luminosity , calculated with the AloHEP software [30] including the hourglass and pinch effects and the muon decay, is and for the two options. As only every tenth LHC bunch collides, fewer but more intense proton bunches [31] would substantially increase the luminosity: about was estimated with a 4 MW proton driver, even for only 5% of the TRISTAN yield [23].
| Booster option | TRISTAN-type | ERL-based | ||
| [TeV] | 5.3 | 5.3 | ||
| [ cms] | 8.8 | 7.2 | ||
| Beam parameters | ||||
| Energy [TeV] | 1 | 7 | 1 | 7 |
| Bunch population [] | 3.7 | 22 | 3.0 | 22 |
| Norm. emittance [m] | 4 | 2 | 4 | 2 |
| RMS beam size at IP, [m] | 4.3 | 4.3 | 4.3 | 4.3 |
| Tune shift, [] | 600 | 22 | 600 | 18 |
Table 2. Main parameters of the () for the two booster options [7]; the bunch population is given at injection into the main ring.
Interaction region and detector
Owing to the large bunch spacing, a very small crossing angle suffices to avoid parasitic collisions [23]. The detector concept [7] (Figure 4), similar to that for the muon–ion collider [15], has an asymmetric layout complemented by Roman pots and a muon spectrometer far from the IP. The background from decays is suppressed by a tungsten nozzle on the incoming- side, which limits the acceptance; 1 TeV muons traversing it lose on average about 19% of their energy [7]. These issues, as well as TeV-muon and near-beam detectors, were discussed at this workshop [32].

Figure 4. Detector concept for collisions [7]: silicon tracker (SiT), particle identification (PID), calorimeters (ECAL, HCAL), solenoid (SSM), muon system (MUON), Roman pots (RPs) and far-backward muon spectrometer (MS); the tungsten nozzle is shown in orange.
Physics potential
Multi-TeV muon–hadron colliders have a huge potential for SM (especially QCD) and beyond-the-SM (BSM) physics, notably for muon-related phenomena such as excited muons, colour-octet muons, leptoquarks and contact interactions [11, 14, 15, 16, 33]; nuclear options, two-photon, Higgs-pair and neutrino physics were discussed at this workshop [25, 34].
QCD at small . Since with , at TeV values as low as are accessible at GeV, compared with at HERA and at the LHeC (ERL 50); at , extends up to 2800 GeV, compared with 10 and 140 GeV, respectively [7]. As seen in Figure 5(a), the LHC covers a much wider range of small and high than the LHeC and even the FCC-, giving unique access to non-linear parton dynamics [23, 34].
Figure 5. (a) Coverage of the kinematic plane in deep inelastic lepton–proton scattering by the LHC, compared with the FCC-eh, LHeC, HERA, EIC and fixed-target experiments. (b) Statistical significance of the colour-octet muon signal its mass for single production at the LHC with various integrated luminosities, and for pair production at the HL-LHC with 3 ab. Both adapted from Ref. [7].
Higgs boson. Higgs bosons are produced via () fusion with a cross section of about 0.39 (0.075) pb, i.e. 3.4 (6.6 ) Higgs bosons per year for the TRISTAN-type option [7], an order of magnitude more than at the LHeC, which will help to determine the Higgs couplings to the and bosons [34].
Colour-octet muons. In composite models, each SM lepton is accompanied by a colour-octet partner, the leptogluon [35]; the colour-octet muon shares the electroweak quantum numbers of the muon. Its interaction with leptons and gluons reads
where is the compositeness scale (set to ) and are chirality factors. At the LHC colour-octet muons are pair-produced, whereas at colliders they are produced resonantly, , with a much larger cross section [18, 33]. A hadron-level analysis [7], requiring one muon and one jet with GeV at the LHC and two muons and two jets with (250) GeV for the leading (sub-leading) objects at the HL-LHC, shows that the discovery reach of about 2.3 TeV at the HL-LHC with 3 ab is extended to about 4.1 (4.4) TeV at the LHC with 100 fb (1 ab), see Figure 5(b); the ultimate HL-LHC sensitivity is matched with an integrated luminosity smaller by at least four orders of magnitude. An update gives a reach of about 3 (4.2) TeV for a 0.5 (1) TeV beam with 250 (500) fb [24].
R-parity-violating supersymmetry. The terms of the R-parity-violating superpotential [36] allow squarks to be produced as -channel resonances in muon–quark collisions [14, 36]. At the LHC, the neutral-current process proceeds on valence quarks, whereas the charged-current one, , relies on sea antiquarks and is less sensitive. For 1 TeV antimuons and 500 fb, requiring exactly one muon, at least one jet and a high scalar sum of the jet transverse momenta, can be probed down to 0.01–0.02 for squark masses of 1–3 TeV, almost an order of magnitude beyond the current and projected LHC bounds [24].
Challenges and conclusions
The main challenges of the LHC are: the antimuon yield (stages i–iii), whose -fold enhancement has to be demonstrated [27, 28]; the booster magnets (stage v), where fixed-field arcs are needed [29] and the use of high-gradient X-band linacs instead of superconducting ones deserves study; the radiation induced by neutrinos from decays (stages v–vi), which, on the other hand, provide intense and collimated neutrino beams [34]; and the interaction region and detector design [32].
In conclusion, energy-frontier lepton–hadron colliders are a must for two reasons: the clarification of the strong interactions, including the QCD basics, which are responsible for about 99% of the mass of the visible Universe, and the provision of PDFs for an adequate interpretation of the HL-LHC (as well as FCC- and SppC) data. Since the LHC provides the highest-energy hadron beams, CERN is the most suitable place for building such colliders. Based on the ultra-cold technology, the LHC/LHC appears to be the most realistic way to reach multi-TeV centre-of-mass energies in lepton–hadron collisions in the 2030s, with TeV and luminosities approaching . In addition, it will provide a unique opportunity to investigate muon-related BSM physics. Systematic studies of its accelerator, detector and physics aspects, as pursued by the HELP4CERN Study Group [22], are therefore strongly motivated.
Acknowledgements
We thank K. Piotrzkowski and the Kraków group for organising this workshop, and K. Piotrzkowski for pointing out the inconsistency in the antimuon survival rates of Ref. [7]. We are grateful to A. C. Canbay, B. Dagli, L. Forthomme, U. Kaya, B. Ketenoglu, A. Kilic, F. Kocak, I. Tapan, F. Zimmermann and the HELP4CERN Study Group for fruitful discussions. This work is partially supported by TÜBİTAK under grant No. 125F120.
References
- [1]R. Hofstadter and R. W. McAllister. In: Phys. Rev. 98 (1955), p. 217; E. D. Bloom et al. In: Phys. Rev. Lett. 23 (1969), p. 930; M. Breidenbach et al. In: Phys. Rev. Lett. 23 (1969), p. 935; J. J. Aubert et al. (EMC). In: Phys. Lett. B 123 (1983), p. 275.
- [2]M. Klein and R. Yoshida. “Collider physics at HERA”. In: Prog. Part. Nucl. Phys. 61 (2008), p. 343. arXiv: 0805.3334.DOI
- [3]ATLAS Collaboration. In: Phys. Lett. B 716 (2012), p. 1; CMS Collaboration. In: Phys. Lett. B 716 (2012), p. 30.
- [4]R. Abdul Khalek et al. “EIC Yellow Report”. In: Nucl. Phys. A 1026 (2022), p. 122447. arXiv: 2103.05419.
- [5]J. L. Abelleira Fernandez et al. (LHeC Study Group). “A Large Hadron Electron Collider at CERN”. In: J. Phys. G 39 (2012), p. 075001. arXiv: 1206.2913.DOI
- [6]P. Agostini et al. “The Large Hadron-Electron Collider at the HL-LHC”. In: J. Phys. G 48 (2021), p. 110501. arXiv: 2007.14491; F. Ahmadova et al. 2025. arXiv: 2503.17727.DOI
- [7]D. Akturk et al. “μLHC: Antimuon Ring and High Luminosity-LHC-Based μ+p Collider”. In: PTEP 2026 (2026), 033C04. DOI: 10.1093/ptep/ptag031. arXiv: 2506.15445v2.
- [8]K. André et al. “An electron–hadron collider at the high-luminosity LHC”. In: New J. Phys. 28 (2026), 013001. arXiv: 2503.20475.DOI
- [9]V. D. Shiltsev. FERMILAB-TM-1969. 1996; V. D. Shiltsev. “An asymmetric muon–proton collider: luminosity consideration”. In: Proc. PAC97, Vancouver. 1997, p. 420.
- [10]I. F. Ginzburg. “Physics at future ep, γp (linac-ring) and μp colliders”. In: Turk. J. Phys. 22 (1998), p. 607.
- [11]K. Cheung. “Muon–proton colliders: leptoquarks and contact interactions”. In: AIP Conf. Proc. 441 (1998), DOI: 10.1063/1.56424. arXiv: hep-ph/9802219.DOI
- [12]S. Sultansoy. DESY-99-159. 1999. arXiv: hep-ph/9911417; S. Sultansoy. In: Eur. Phys. J. C 33 (2004), S1064; A. N. Akay, H. Karadeniz and S. Sultansoy. In: Int. J. Mod. Phys. A 25 (2010), p. 4589.
- [13]Y. C. Acar et al. In: Nucl. Instrum. Meth. A 871 (2017), p. 47. arXiv: 1608.02190; A. C. Canbay et al. In: Adv. High Energy Phys. 2017 (2017), 4021493. arXiv: 1704.03534.DOI
- [14]K. Cheung and Z. S. Wang. “Physics potential of a muon–proton collider”. In: Phys. Rev. D 103 (2021), p. 116009. arXiv: 2101.10476.DOI
- [15]D. Acosta and W. Li. In: Nucl. Instrum. Meth. A 1027 (2022), p. 166334. arXiv: 2107.02073; D. Acosta et al. “The potential of a TeV-scale muon-ion collider”. In: JINST 18 (2023), P09025. arXiv: 2203.06258.DOI
- [16]B. Ketenoglu et al. “Review of muon–proton and muon–nucleus collider proposals”. In: Mod. Phys. Lett. A 37 (2022), 2230013. arXiv: 2206.00037.arxiv.org/abs/2206.00037
- [17]J. P. Delahaye et al. “Muon Colliders”. 2019. arXiv: 1901.06150; C. Accettura et al. “Towards a muon collider”. In: Eur. Phys. J. C 83 (2023), p. 864. arXiv: 2303.08533; C. Accettura et al. “The Muon Collider”. 2025. arXiv: 2504.21417.
- [18]U. Kaya et al. “Luminosity and physics considerations on HL-LHC- and HE-LHC-based μp colliders”. In: EPL 138 (2022), p. 24002. arXiv: 1905.05564.arxiv.org/abs/1905.05564
- [19]M. Abe et al. In: PTEP 2019 (2019), 053C02. arXiv: 1901.03047.arxiv.org/abs/1901.03047
- [20]Y. Hamada et al. “μTRISTAN”. In: PTEP 2022 (2022), 053B02. DOI: 10.1093/ptep/ptac059. arXiv: 2201.06644.
- [21]D. Akturk, B. Dagli and S. Sultansoy. In: EPL 147 (2024), p. 24001. arXiv: 2403.17034; D. Akturk et al. In: Eur. Phys. J. Plus 140 (2025), p. 317. arXiv: 2406.02647.arxiv.org/abs/2403.17034
- [22]HELP4CERN Study Group. White Paper on High Energy, High Luminosity Lepton–Hadron Colliders at CERN. In preparation. 2026.
- [23]K. Piotrzkowski. “LHμC: A high energy, high luminosity muon–hadron collider at CERN”. In: Proc. DIS2026. 2026. arXiv: 2610.35154.arxiv.org/abs/2610.35154
- [24]K. Piotrzkowski. Talk at DIS2026, Bologna, 4–8 May 2026; L. Forthomme. Talk at ICHEP2026, Natal, 30 Jul.–5 Aug. 2026.
- [25]B. Dagli et al. “LHμC: nucleus options”. Talk at this workshop.
- [26]S. Aritome et al. “Acceleration of Positive Muons by a Radio-Frequency Cavity”. In: Phys. Rev. Lett. 134 (2025), p. 245001. arXiv: 2410.11367.DOI
- [27]S. Kamioka et al.; M. Otani; Y. Nakazawa; Y. Takeuchi et al.; M. Yoshida. Talks at this workshop.
- [28]C. Zhang et al.; C. Rogers. Talks at this workshop; Y. Sakaki et al. 2026. arXiv: 2605.25580.
- [29]M. E. Topp-Mugglestone. PhD thesis. University of Oxford, 2024; M. Topp-Mugglestone et al. Talks at this workshop.
- [30]B. Dagli et al. “ALOHEP software”. In: Proc. IPAC’21. 2021.
- [31]F. Zimmermann. “Potential possibilities of modifying SPS/LHC”. Talk at this workshop.
- [32]L. Forthomme; Y. Yamazaki; D. Acosta et al.; A. Kilic et al. Talks at this workshop.
- [33]Y. C. Acar, U. Kaya and B. B. Oner. In: Chin. Phys. C 42 (2018), p. 083108. arXiv: 1703.04030; A. Caliskan, S. Kara and A. Ozansoy. In: Adv. High Energy Phys. 2017 (2017), 1540243. arXiv: 1701.03426.DOI
- [34]P. Kotko et al.; M. Kumar et al.; H. Khanpour; X. Lu et al. Talks at this workshop; S. Tentori. These proceedings, in preparation.
- [35]D. Gonçalves-Netto et al. “Looking for leptoquons”. In: Phys. Rev. D 87 (2013), p. 094023. arXiv: 1303.0845.
- [36]R. Barbier et al. In: Phys. Rept. 420 (2005), p. 1. arXiv: hep-ph/0406039; M. Carena et al. In: Phys. Rev. D 62 (2000), p. 095010. arXiv: hep-ph/0006144.DOI

