BACKGAMMON: A Scheme for Producing High Intensity Muon Beams for Future Colliders and Other Applications

Armen Apyan; Letrell Harris; Paul Gu\`eye; Peter Delfyett; Sekazi K. Mtingwa; Sokhna Bineta Lo Amar

REVIEW 3 major objections 5 minor 16 references

BACKGAMMON: A Scheme for Producing High Intensity Muon Beams for Future Colliders and Other Applications

T0 review · 3 major / 5 minor · reviewed 2026-08-16 · deepseek-v4-flash

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗ Review JSON

Pith's one-line read The paper proposes that Compton-backscattered photons striking a graphite target can produce about $10^{14}$ muons per second, enough for a future muon-muon or muon-ion collider.

desk verdict A credible feasibility sketch for a photon-driven muon source at the EIC, but the 1e14 muons/s headline depends on an assumed 100x luminosity gain and an unreleased simulation. read the letter →

arxiv 2504.21271 v1 pith:CFXOBR5E submitted 2025-04-30 physics.acc-ph hep-ex

Armen Apyan , Peter Delfyett , Paul Guèye , Letrell Harris , Sokhna Bineta Lo Amar , Sekazi K. Mtingwa This is my paper · ORCID

classification physics.acc-phhep-ex

keywords BACKGAMMONmuonbeamsComptonbackscatteringpionproductiongraphitetargetcolliderDeltaresonancelaser-electroninteraction

verification ladder T0 review T1 audit T2 compute T3 formal

The pith

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The reading

This paper tries to establish that a photon-driven source of muon beams is feasible using the electron ring of an electron-ion collider rather than requiring a dedicated proton driver. The scheme, called BACKGAMMON, uses laser light Compton-backscattered off an intense electron beam to make roughly 340 MeV photons, which strike a graphite target and produce pions; the pions then decay into muons. With a photon rate of $10^{17}$ per second, simulations show pion fluxes above $10^{13}$ per second for both signs and a muon rate near $10^{14}$ per second, the level usually quoted for a muon collider. The authors argue this could be developed over about a decade using planned upgrades to existing accelerator and laser infrastructure.

What carries the argument

The mechanism is the Compton backscattering of laser pulses from 4.5 GeV electrons in a storage ring, giving scattered photons with energy $\omega_2 \approx 4\gamma^2\omega_1/(1+4\gamma^2\omega_1/E_e)$, about 340 MeV. The benchmark laser delivers 1046 nm, 254 fs pulses at 90.7 MHz with 10.4 kW average power. The production rate is computed as $R = \sigma_C L$, where $\sigma_C$ is the polarized Compton cross section and $L$ is the electron-photon luminosity from the standard colliding-bunch formula. The pion channel exploits the $\Delta(1232)$ resonance and a pion-selection cut of transverse momentum above 30 MeV/c and emission angles of 40 to 120 degrees, with particle-transport simulations used to obtain pion and muon momentum distributions.

What would settle it

Measure the actual Compton backscattered photon rate from a 90.7 MHz collision of the described electron bunches with a 100 kW, 1046 nm laser; if the measured rate falls well short of $10^{17}$ per second, the predicted $10^{14}$ muons per second cannot be reached, since the muon rate scales linearly with the photon rate.

Watch

Extended reading notes

Core claim

The central claim is that a 340 MeV backscattered-photon beam hitting a 30 cm-long, 2.5 cm-radius graphite target produces both $\mu^+$ and $\mu^-$ beams at roughly $10^{14}$ muons per second. The photon energy is not arbitrary: for a stationary nucleon, a 340 MeV photon sits near the $\Delta(1232)$ resonance, which boosts the photo-absorption cross section and hence pion production. The pions are separated from the dominant electron-positron background by their large angles and transverse momenta, and their decay within about 30 m yields the muon beam. A notable feature is that the positive and negative muon rates are comparable, which the paper highlights as an advantage for a $\mu^+\mu^-$ collider.

Load-bearing premise

The argument depends on raising the backscattered photon rate from about $10^{15}$ to $10^{17}$ per second, a hundredfold increase justified by a projected laser power upgrade and by as-yet unspecified electron-beam improvements.

Editorial extensions

If this is right

A photon-based muon source could remove the need for a high-power proton driver, reusing the electron ring of an electron-ion collider.
The muon rate scales linearly with the backscattered photon rate, so every factor improvement in laser power or electron bunch density directly raises the beam intensity.
Because both muon charges are produced at comparable rates, a single source could feed both $\mu^+$ and $\mu^-$ rings of a collider.
Tuning the photon energy to the $\Delta(1232)$ resonance makes pion production efficient enough that a 30 cm target suffices.

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

If the photon-rate step-up proves out, the same BACKGAMMON target station could likely feed multiple experiments simultaneously, serving as a shared muon source rather than a dedicated collider injector.
A modest-scale demonstrator at a lower photon rate could test the simulated pion and muon yield against target length and angle cuts before committing to $10^{17}$ photons per second.
The scheme's dependence on circulating electron beam lifetime suggests that storage-ring refill methods will be as important as laser power in determining whether $10^{14}$ muons per second is practical.
Because the pion yield is resonance-enhanced, scanning the backscattered photon energy across the $\Delta$ resonance in a test run would map the production cross section directly, providing a clean validation of the target model.

Share X Bluesky LinkedIn Reddit HN

Signed reviews

No signed human review yet.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, and a circularity audit.

Desk Editor's Note

read the letter

Quick take: this is a serious, well-structured feasibility sketch for producing muon beams at the EIC by Compton-backscattering laser photons off the electron ring and directing the 340 MeV gamma rays onto a graphite target. The Compton scattering calculation is straightforward and correct: the photon energy lands at 340 MeV, the Delta resonance, and the luminosity formula gives R=0.93e15/s for the current parameters. That part checks out, and there are no fitted parameters or circular steps. What is actually new is the specific implementation at the EIC and the use of the Delta resonance to maximize pion photo-production. The choice of 340 MeV is physically motivated, not arbitrary. The pion capture method, using angular and transverse momentum cuts to separate pions from the electromagnetic background, follows the Gamma Factory work and is sensible. The paper also deserves credit for being honest about the open problems: it says the beam lifetime under Compton energy loss would be 20-200 ms and that swap-out injection is only a concept, and it calls the whole thing 'an initial, exploratory step.' The soft spot is the load-bearing assumption about the photon rate. To get 1e14 muons/s, you need R=1e17/s, which requires a 10x laser power increase to 100 kW (supported by the laser paper) plus another 10x from 'further research' on Ne, beta functions, or emittances. That second factor is entirely unspecified—no lattice design, no emittance budget, no consideration of the hourglass effect when beta* = 1 cm and sigma_z = 11 mm. Pion and muon rates scale linearly with R, so if only half of the second factor materializes, you get 5e13, and if it doesn't materialize at all, you get 1e15/s from the laser upgrade alone. That's an order of magnitude spread. The GEANT4 results are also not reproducible from the text—no release of the code or detailed validation—so the 1e13/s pion flux is a simulation result that can't be checked. None of this kills the idea. It means the 1e14/s number is a goal, not a prediction. The paper would be strengthened by a parameter scan showing muon yield vs. R, by publishing the simulation setup, and by including at least a sketched path for the beam parameter improvements. Verdict: serious referee worthy. It should go to peer review, and the referee should push for those missing pieces. I'd bring it to a reading group if anyone works on muon sources, but I wouldn't cite it as a demonstrated source in my own work until the rate question is answered.

Referee Report

3 major / 5 minor

Summary. The paper proposes BACKGAMMON, a scheme in which Compton-backscattered laser photons from the EIC electron beam produce ~340 MeV photon pulses that strike a stationary graphite target; the resulting pions decay into muons, with a stated goal of ~1e14 muons/s. The manuscript derives the backscattered photon energy, the unpolarized and polarized Compton cross sections, the electron-photon luminosity, and the backscattered-photon rate R from EIC and laser parameters, obtaining R=0.93e15/s for current parameters. It then asserts that R can be increased to 1e17/s and uses GEANT4 simulations to report pion fluxes >1e13/s and a muon rate of ~1e14/s from a 30 cm graphite target.

Significance. If the claimed rates were realized, BACKGAMMON would be an attractive muon-source concept for muon-muon and muon-ion colliders, and it would leverage existing EIC infrastructure with comparable positive and negative muon fluxes. The Compton-scattering part is a strength: the photon energy, cross section, and luminosity use standard, traceable formulas and parameter values, with no fitted parameters in the output. The paper is also commendably explicit about open problems, including beam lifetime and the need for a pion-capture design. However, the central quantitative claim rests on an order-of-magnitude beam improvement that is not specified and on simulation outputs without released code or statistical uncertainties. The paper is therefore a valuable feasibility sketch, but the 1e14/s headline rate is not yet established as a prediction.

major comments (3)

[Muon Production and Table 3] The central claim of ~1e14 muons/s is obtained by scaling R from 0.93e15/s in Table 3 to 1e17/s, a factor of about 108. Only one order of magnitude is supported by the cited 100 kW laser upgrade; the second order of magnitude is attributed to 'further research' on Ne, beta_x, beta_y, and emittances, with no concrete lattice design, emittance budget, injector/RF analysis, or simulation. Because Eq. (9) makes the pion and muon rates linear in R, the quoted 1e14/s is an unsupported extrapolation rather than a prediction. The authors should either provide a specific beam-parameter scenario that gives R=1e17/s or reframe the muon yield as a conditional scaling law proportional to R with explicit uncertainties.
[Concluding paragraph (beam lifetime and Swap-Out)] The paper states that the circulating electron beam lifetime under Compton energy loss would be 20 to 200 ms and invokes the APS Swap-Out mode as the remedy, but provides no quantitative analysis: no required fresh-bunch rate per bunch, total injector repetition rate, bunch charge, kicker timing relative to the 90.7 MHz collision frequency, or treatment of energy loss in the beam dynamics. Without this, sustained operation at the assumed Ne is not established. This is a load-bearing feasibility question for the stated muon rate and should be analyzed or explicitly identified as an unresolved constraint rather than referenced only as a concept.
[Simulations (Figs. 2-4)] The GEANT4 simulation is not released, and the paper gives no statistical uncertainties, number of primary photons simulated, or details of the geometry, physics lists, and cuts. The quoted pion flux >1e13/s and the muon rate ~1e14/s are normalized to the assumed 1e17/s photon rate, so the reader cannot separate simulation statistics from the normalization assumption. The only validation statement is a brief claim that GEANT4.11.2 and 10.6 gave consistent results, with no side-by-side comparison shown. Reproducibility and error estimates are needed before these rates can be treated as quantitative predictions.

minor comments (5)

[Total Cross Section with Polarization Included, Eq. (7)] The definition sigma_0 = pi(e^2/(m_e c^2)) is dimensionally a length, not an area; it should be sigma_0 = pi(e^2/(m_e c^2))^2 = pi r_0^2. This typo does not affect Table 3, which uses Eq. (4), but should be corrected.
[Throughout] There are several typographical errors: 'one write can write' in the polarization section, 'subequently' and 'increase tthe laser average power' in the Muon Production section, and 'position sources' in Ref. [9] should be 'positron sources'.
[Table 1 and Ref. [13]] The EIC parameters are cited to a URL rather than to a specific design report or parameter database entry; please provide a citable reference so the values in Table 1 can be verified.
[Luminosity formula, Eq. (8)] The luminosity calculation uses beta* = 1 cm while sigma_ze = 11 mm, yet the paper does not state whether hourglass effects are included in Eq. (8). Please justify this or estimate the effect, since the proposed parameter improvements may alter the scaling R with beta* and emittance.
[Figures 2-4] The histograms in Figs. 2 and 3 do not show error bars or bin widths, and Fig. 4 lacks axis labels and units. Adding these details would help readers assess the statistical significance of the quoted rates.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: the muon-rate estimate is an explicit conditional scaling of an assumed photon rate, while the per-photon pion and muon yields come from independent GEANT4 simulations and standard cross-section inputs.

full rationale

Walking the derivation chain: the backscattered-photon energy is obtained from the standard Compton formula (Eq. 2); the unpolarized and polarized cross sections come from Landau-Lifshitz and standard QED references (Eqs. 4 and 5); the electron-photon luminosity uses Suzuki's formula (Eq. 8) with published EIC parameters; and the backscattered-photon rate is defined by R = sigma_C L (Eq. 9). The pion and muon yields are obtained from GEANT4 simulations with explicit kinematic cuts (pT >= 30 MeV/c, 40-120 degrees) and no fitted parameters are used to force the outcome. The 1e17/s photon rate used in Figs. 2-4 is not fitted or derived by the paper; the text states it would require a 10x laser-power upgrade plus 'further research' on beam parameters (Ne, beta*, emittance), and it is therefore an explicitly assumed future input. The resulting ~1e14 muons/s is a linear consequence of that assumed rate multiplied by the simulated per-photon yield, so it is a transparent conditional estimate rather than a hidden circular prediction. The paper itself flags its exploratory status ('The presented studies can be considered as an initial, exploratory step') and identifies the circulating electron-beam lifetime of 20-200 ms as a major challenge, confirming that the high-rate scenario is a feasibility target rather than a demonstrated result. The self-citations (Refs. [5,6]) are historical and motivational; the simulation and cross-section inputs are independently checkable and do not reduce to those citations. No step in the derivation defines its output in terms of its input, so no circularity is found.

Assumptions & free parameters 5 free parameters · 5 assumptions · 0 invented entities

The central muon yield rests on hand-chosen target dimensions, assumed 100x brightness gains, and an unvalidated GEANT4 simulation. No new physical entities are introduced.

free parameters (5)

Target length = 30 cm
Chosen after varying length and radius in GEANT4 to maximize pion collection; central to muon yield.
Target radius = 2.5 cm
Same optimization scan; sets hadronic shower length and pion escape.
Laser average power upgrade = 100 kW
Assumed 10x above the 10.4 kW demonstrated in Ref. [14]; required to reach 1e17 photons/s.
Electron beam and lattice improvement factor = 10x
Assumed gain from future improvements to Ne, beta*, and emittance; no concrete design; required with the laser upgrade to reach 1e17 photons/s.
Pion selection cuts = pT > 30 MeV/c; 40 to 120 degrees
Hand-chosen kinematic window to favor pions over e±, following Ref. [5]; quoted fluxes depend on these cuts.

assumptions (5)

standard math Compton scattering cross section and luminosity formulas from Refs. [10-12] apply at these parameters.
Input to R=sigma_C L; assumed without re-derivation.
domain assumption EIC electron beam parameters in Table 1 are representative and achievable (Ne=1.72e11, emittances, beta*).
Taken from EIC design parameters; central to L and R.
domain assumption GEANT4.11.2 accurately models photonuclear pion production, electromagnetic cascades, and pion decay in graphite.
All pion and muon yields come from this simulation; no validation against data is shown.
domain assumption Pions with pT>30 MeV/c in the 40 to 120 degree cone are predominantly pions, with negligible e± contamination.
Used to define collection; purity is not quantified.
ad hoc to paper Swap-Out injection can replenish the electron beam fast enough to sustain 90.7 MHz collisions despite 20-200 ms beam lifetime.
Proposed in the conclusion as the solution to the beam lifetime challenge, without a design or rate estimate.

how reviews work

0 comments

Cite this review

Pith. "Pith review of BACKGAMMON: A Scheme for Producing High Intensity Muon Beams for Future Colliders and Other Applications." pith.science (2026). https://pith.science/paper/CFXOBR5E

@misc{pith2026250421271,
  author       = {Pith},
  title        = {Pith review of: BACKGAMMON: A Scheme for Producing High Intensity Muon Beams for Future Colliders and Other Applications},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/CFXOBR5E}},
  note         = {Machine review of arXiv:2504.21271}
}

abstract

We present a scheme for producing intense $\mu^+$ and $\mu^-$ lepton beams that could be utilized in a future muon-muon or muon-ion collider, as well as for other applications. The scheme makes use of BACKscattered GAMMas On Nucleons (BACKGAMMON). Current accelerator infrastructures could be utilized to produce the muon beams, such as the Electron-Ion Collider (EIC) at Brookhaven National Laboratory in the United States. We discuss the implementation of BACKGAMMON at the EIC.

Figures

Figures reproduced from arXiv: 2504.21271 by the authors.

**Figure 1.** Schematic of Compton Backscattering For our case, the photon beam is backscattered with θ ∼ π, and the resulting high energy backscattered photon beam is confined within a cone of opening angle ∼ 1 γ relative to the electron’s incident direction of motion, with γ being the usual Lorentz contraction factor for the electron beam. We take the incident electron to have energy Ee and incident photon to have energy ω1, wi… view at source ↗

**Figure 2.** Pion Kinematic Distributions 1 mm from the end of the graphite target [PITH_FULL_IMAGE:figures/full_fig_p012_2.png] view at source ↗

**Figure 3.** Muon Kinematic Distributions 30 m from the center of the graphite target [PITH_FULL_IMAGE:figures/full_fig_p013_3.png] view at source ↗

Figures from the paper (1 more)

**Figure 4.** Figure 4: Rate of Muon Production vs. Pion Propagation Length from the center of [PITH_FULL_IMAGE:figures/full_fig_p014_4.png]

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

16 extracted references · 14 canonical work pages

[1]

Accettura et al

C. Accettura et al. , Interim report for the International Muon Collider Collaboration (IMCC), https://arxiv.org/abs/2407.12450v1 (2024)

arXiv 2024
[2]

Acosta and W

D. Acosta and W. Li, Nucl. Instr. and Meth. in Phys. Res., A 1027, 166334 (2022)

work page 2022
[3]

Cline, J.C

E. Cline, J.C. Bernauer, and A. Deshpande, A Future Muon-Ion Collider at Brookhaven National Laboratory , DIS2022: XXIX International Workshop on Deep- Inelastic Scattering and Related Subjects, https://zenodo.org/records/7268146 (2022)

work page arXiv 2022
[4]

Ding, H.G

X. Ding, H.G. Kirk, and K.T. McDonald, Particle Production of a Graphite Target System for the Intensity Frontier , 6th International Particle Accelerator Conference, https://accelconf.web.cern.ch/ipac2015/papers/wepje010.pdf (2015). 16

work page 2015
[5]

Krasny, The Gamma Factory proposal for CERN , https://arxiv.org/abs/1511.07794 (2015); A

M. Krasny, The Gamma Factory proposal for CERN , https://arxiv.org/abs/1511.07794 (2015); A. Apyan, M. Krasny, and W. Placzek, Phys. Rev. Accel. Beams 26, 083401 (2023)

arXiv 2015
[6]

Mtingwa and M

S. Mtingwa and M. Strikman, Phys. Rev. Lett. 64, 1522 (1990)

work page 1990
[7]

Mtingwa and M

S. Mtingwa and M. Strikman, Nucl. Inst. and Meth. in Phys. Res. A472, 189-194 (2001)

work page 2001
[8]

Mtingwa and M

S. Mtingwa and M. Strikman, Nucl. Inst. and Meth. in Phys. Res. A 455, 50-53 (2000)

work page 2000

Show all 16 references

[9]

Chaikovska, Polarized position sources for the future colliders , Universit´ e Paris Sud - Paris XI, https://theses.hal.science/tel-00795937 (2013)

I. Chaikovska, Polarized position sources for the future colliders , Universit´ e Paris Sud - Paris XI, https://theses.hal.science/tel-00795937 (2013)

2013
[10]

Landau, E

L. Landau, E. Liftshitz, V. Berestetskii, and L. Pitaevskii, Quantum Electrodynamics, Vol. 4, Chap. 10, First published 1971, Latest ed. 2020

1971
[11]

Ginzburg, G.L

For an excellent discussion, see I.F. Ginzburg, G.L. Kotkin, S.L. Panfil, V.G. Serbo, and V.I. Telnov, Nucl. Instr. and Meth. in Phys. Res. 219, 5-24 (1984)

1984
[12]

Suzuki, General Formulae of Luminosity for Various Types of Colliding Beam Machines, KEK Pub

T. Suzuki, General Formulae of Luminosity for Various Types of Colliding Beam Machines, KEK Pub. 76-3 (1976)

1976
[13]

https://www.bnl.gov/eic/. 17
[14]

M¨ uller, C

M. M¨ uller, C. Aleshire, A. Klenke, E. Haddad, F. L´ egar´ e, A. T¨ unnermann, and J. Limpert, Opt. Lett. 45, no. 11, 3083 (2020)

2020
[15]

Monte Carlo simulations of muon production,

R.B. Palmer et al. , “Monte Carlo simulations of muon production,” AIP Conf. Proc. 352, 108–124 (1996)

1996
[16]

Emery and M

L. Emery and M. Borland, Possible Long-term Improvements to the APS , in Proc. of PAC’03, Portland, OR, USA, 256 (2003). 18

2003

Pith tools

Reviewed August 16, 2026 · model on record in the stance chip above.

[1] [1]

Accettura et al

C. Accettura et al. , Interim report for the International Muon Collider Collaboration (IMCC), https://arxiv.org/abs/2407.12450v1 (2024)

arXiv 2024

[2] [2]

Acosta and W

D. Acosta and W. Li, Nucl. Instr. and Meth. in Phys. Res., A 1027, 166334 (2022)

work page 2022

[3] [3]

Cline, J.C

E. Cline, J.C. Bernauer, and A. Deshpande, A Future Muon-Ion Collider at Brookhaven National Laboratory , DIS2022: XXIX International Workshop on Deep- Inelastic Scattering and Related Subjects, https://zenodo.org/records/7268146 (2022)

work page arXiv 2022

[4] [4]

Ding, H.G

X. Ding, H.G. Kirk, and K.T. McDonald, Particle Production of a Graphite Target System for the Intensity Frontier , 6th International Particle Accelerator Conference, https://accelconf.web.cern.ch/ipac2015/papers/wepje010.pdf (2015). 16

work page 2015

[5] [5]

Krasny, The Gamma Factory proposal for CERN , https://arxiv.org/abs/1511.07794 (2015); A

M. Krasny, The Gamma Factory proposal for CERN , https://arxiv.org/abs/1511.07794 (2015); A. Apyan, M. Krasny, and W. Placzek, Phys. Rev. Accel. Beams 26, 083401 (2023)

arXiv 2015

[6] [6]

Mtingwa and M

S. Mtingwa and M. Strikman, Phys. Rev. Lett. 64, 1522 (1990)

work page 1990

[7] [7]

Mtingwa and M

S. Mtingwa and M. Strikman, Nucl. Inst. and Meth. in Phys. Res. A472, 189-194 (2001)

work page 2001

[8] [8]

Mtingwa and M

S. Mtingwa and M. Strikman, Nucl. Inst. and Meth. in Phys. Res. A 455, 50-53 (2000)

work page 2000