At J-PARC, a high-momentum muon beamline produced a mode with an estimated purity of 98.0%, an intensity of 1.7 × 10^3 particles per spill and a spatial spread of about 15 millimetres. The mode was identified at 3 GeV/c and is the central performance result of a feasibility test for muon radiography, or imaging with muons.
The paper characterizes the operation as the first successful operation of a high-momentum, high-purity muon beamline at an experimental facility in Japan. Its measurements focused on the beam's intensity, spatial and angular profiles, and estimated purity.
The beam was tested at 12 settings
The beamline's defining feature was the µ20 scheme. It scales the downstream beamline to 60% of the upstream central momentum to transport backward-decaying muons while sweeping away the primary pion background. Researchers measured twelve configurations, combining upstream central momenta of 3, 5 and 10 GeV/c with downstream scaling factors of 60%, 75%, 80% and 85%.
Intensity was counted with T1×T2 hardware scalers. Beam shape and direction were reconstructed from BFT1 and BFT2 tracking, while purity was assessed with the MF system, which used alternating iron blocks and T3 through T6 counters.
More particles came with more contamination
The muon beam was also broader than the comparison hadron beam. Across all twelve operating modes, its spatial and angular widths, measured as standard deviations, were approximately 1.5 times the hadron-beam widths, with the same systematic tendencies across the modes.
Across the tested configurations, reported muon intensity ranged from 0.42 to 17.0 thousand particles per spill, with the values averaged over ten spills. At 3 GeV/c upstream momentum, the rates were 0.42, 0.64, 1.0 and 2.0 k/spill for downstream scaling factors of 60%, 75%, 80% and 85%, respectively. At 5 GeV/c, they were 1.7, 2.6, 3.7 and 8.3 k/spill, and at 10 GeV/c they were 2.4, 4.6, 6.6 and 17.0 k/spill.
Those rates came with a marked purity-intensity trade-off. Lower downstream scaling settings were associated with lower hadron contamination and lower intensity, while higher settings increased intensity but compromised purity. At 60% scaling, the purity estimates were 90.2 ± 0.5% at 3 GeV/c upstream momentum, 98.0 ± 0.6% at 5 GeV/c and 98.3 ± 0.9% at 10 GeV/c. At 85%, the corresponding estimates were 61.4 ± 0.5%, 63.9 ± 0.6% and 53.2 ± 1.1%.
How the purity was estimated
Purity was estimated from iron-filter attenuation in the MF system. The researchers compared experimental hodoscope data with simulated responses across the MF layers and minimized the residual sum of squares, the total mismatch between measured and simulated patterns, to obtain the purity estimate.
At 60% scaling, the Geant4-template estimate was 98.0 ± 0.6%, while an empirical hadron template based on 3 GeV/c data gave 97.9 ± 0.9%. The two estimates agreed closely, providing a cross-check on the template-based purity calculation.
Detector effects were included in the calculation. Raw counts were corrected for the lower efficiency of T4, and a detector hit required at least half of the minimum-ionizing-particle peak. The measured intrinsic efficiencies for T3, T4, T5 and T6 were 99.84 ± 0.02%, 98.51 ± 0.05%, 99.54 ± 0.03% and 99.67 ± 0.02%, respectively.
A beamline result, not an imaging result
The result is a beamline feasibility report for muon radiography, not a claim that a radiography image has already been demonstrated. T106 used muon-beam measurements to evaluate feasibility, with intensity, spatial and angular profiles and purity as its endpoints.
Looking ahead, the authors plan a 15-kW loss-production target and project more than two orders of magnitude higher intensity, reaching above 10^5 particles per spill under µ20. The paper presents that figure as a projection alongside the measured results from the tested configurations.
The document is a preprint, arXiv:2608.25374v1 in the physics.acc-ph category, dated 26 August 2026. Its acknowledgments cite support from JST [K program] Japan Grant JPMKP24J3, JSPS Grant-in-Aid for Scientific Research (S) Grant No. JP22H04940 and Grant-in-Aid for Scientific Research (A) Grant No. 22H00124.
Paper data and sources
Original title: Feasibility demonstration of a high-momentum, high-purity muon beamline at the J-PARC Hadron Experimental Facility
Authors: Kotaro Shirotori, Takaya Akaishi, Kazuya Aoki et al.
Journal/Repository: arXiv
Status: Preprint, not yet peer-reviewed
First online: 2026-08-26
DOI: Not available
Original paper · Full text