Probing Sub-MeV Dark Matter with Neutron-Capture γ \gamma Spectroscopy
Bernhard Meirose, David Milstead · 2026
Read Probing Sub-MeV Dark Matter with Neutron-Capture γ \gamma Spectroscopy by Bernhard Meirose, David Milstead (2026) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2603-26472.txt.
Introduction
Meirose and Milstead present a search plan for sub-MeV dark matter using gamma rays from neutron capture. When a nucleus sheds energy after capturing a neutron, it might also emit a weakly coupled new particle. Rather than chase single faint lines, their method hunts combs of satellite lines. These lines sit at one shared offset below known capture lines, across many parents and target nuclei. Combining the evidence suppresses nuclear and instrument quirks, and they discuss the best targets and detector setup.
Catalog record
| CREATOR | Bernhard Meirose, David Milstead |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 33203 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2603.26472
Browse the full library · About the Tower of Babel
Full text
The opening of the work, from the library's hosted text:
Probing Sub-MeV Dark Matter with Neutron-Capture γ \gamma Spectroscopy Bernhard Meirose, David Milstead Abstract We present a general, discovery-grade framework for searching for weakly coupled new particles emitted in nuclear de-excitation following neutron capture. Rather than relying on isolated spectral features, the method exploits correlated “satellite-line combs”: multiple weak γ \gamma -ray lines appearing at a common energy offset Δ \Delta below known capture transitions. By combining likelihood information across many parent lines and multiple target nuclei, the approach strongly suppresses nuclear-structure ambiguities and instrumental artifacts. We also discuss optimal target selection and practical experimental implementation with high-resolution HPGe detectors. I Physics Motivation and Signature Light dark-sector particles in the keV–MeV mass range arise in a wide class of extensions of the Standard Model, including well-motivated dark matter scenarios such as axion-like particles, scalar mediators, and other feebly coupled states 2 ; 11 ; 8 . Compared to both weak-scale dark matter and ultralight bosonic candidates, experimental constraints in this intermediate mass window are weaker and more model dependent. In particular, direct searches involving particles with masses in the keV-MeV range coupled to nucleons are rare. Existing experiments mostly limit couplings to electrons or photons 1 ; 3 ; 7 . Other searches involving nuclear de-excitations, as well as reactor-based axion searches, have been performed 13 ; 5 . These searches, however, are primarily focused on electromagnetic couplings and are statistically limited in their sensitivity to hadronic interactions. Such light-mass dark particles, if coupled to nucleons, may be produced in nuclear transitions and subsequently escape detection, carrying away a small but fixed fraction of the nuclear transition energy. Such processes can arise, for example, from the emission of axion-like particles in magnetic dipole (M1) transitions 12 or from more generic scalar or pseudoscalar couplings to nucleons. In this work, we explore radiative neutron-capture reactions as a laboratory probe of light dark-sector particles with nucleon couplings. To our knowledge, this environment has not been systematically exploited for dark-sector particle searches. The capture process produces discrete \gamma -ray cascades with precisely known transition energies, enabling correlated missing-energy searches that probe nucleon-level couplings directly. II Satellite-Line Signature and Comb Method We examine the possibility of the emission of a light, weakly interacting particle X , prior to the \gamma -de-excitation of a compound nucleus formed in the neutron capture reaction. If the emission of the X -particle is a two-body process from the capture state, the energy of the \gamma -transition is reduced by a constant value \Delta . Schematically, n+A\to B^{*}(E_{x})\to B^{*}(E_{x}-\Delta)+X\to B(E_{f})+\gamma_{\rm sat}, (1) where the satellite \gamma energy satisfies E_{\rm sat}\simeq E_{0}-\Delta, (2) where E_{0} is the corresponding standard capture line without the presence of X emission. Near the threshold, the offset \Delta is approximately equal to the mass of the emitted particle, with small corrections due to nuclear recoil effects that are negligible at the energies of interest. A fundamental aspect of this hypothesis is that the offset \Delta is independent of the final nuclear state . In other words, for a given target nucleus, for any strong parent capture line at an energy given by E_{0}^{(k)} , there will be a weaker satellite line at an energy given by E_{0}^{(k)}-\Delta . The satellite lines will display a characteristic ”satellite line comb” feature in the \gamma spectrum, with constant separation between the lines and a known position relative to the parent lines. Such a characteristic correlation will identify the signal from any conventional nuclear background. Unknown nuclear \gamma lines may appear at any energy, but they will not have the characteristic correlation with a common \Delta associated with many unrelated parent transitions of the same nucleus. Other possible instrumental effects, like peak tailing, coincidence summing, or escape peaks, will also be highly energy-dependent but will not have the characteristic \Delta associated with the satellite lines. In order to increase confidence that the result is not due to a particular nucleus, the same analysis will be performed on multiple target nuclei with unrelated level schemes. Although a characteristic satellite pattern may appear in a single nucleus, a statistical limit may arise due to the number of strong parent lines and background fluctuations. By performing the same measurement on multiple target nuclei, the method has discovery-level sensitivity. A true particle signal will have a characteristic energy offset \Delta among different targets, while statistical fluctuations and instrumental effects will be strongly suppressed. The underlying reason the multi-target strategy is so powerful is rooted in the physics of neutron capture rather than in statistical coincidence alone. Different target nuclei populate the compound nucleus at distinct neutron separation energies and exhibit unrelated capture cascades, leading to widely different sets of primary \gamma -ray energies. A real satellite signal emitted in a two-body transition near the capture state produces lines shifted by a common energy offset \Delta , independent of the detailed nuclear level scheme. No known nuclear process generates such a correlated offset across unrelated nuclei. This way, what is traditionally a complication in capture spectroscopy becomes a central strength of the method: target-specific features are automatically rejected, while a genuine signal must recur at the same \Delta across all targets. III Statistical Analysis Framework The satellite line search is performed as a two-stage likelihood analysis, taking into account the experimental observables and detector response. The strategy is based on the fact that a true signal will have weak \gamma -ray satellites located at a common energy offset \Delta with respect to a number of otherwise unrelated parent capture lines. Before presenting the individual stages of the analysis, we summarize the experimental scales and analysis strategies relevant to the likelihood and the division of the analysis into two stages. III.1 Relevant Experimental Scales and Analysis Choices To illustrate the typical energy scales, consider a typical neutron capture \gamma line with energy E_{0}\simeq 558~\mathrm{keV} , as characteristic of cadmium isotopes. With a high-purity germanium detector having a given energy resolution \mathrm{FWHM}\simeq 1~\mathrm{keV} , corresponding to a Gaussian width \sigma_{\rm det}\simeq 0.42~\mathrm{keV} , a bin width of \sim 0.2~\mathrm{keV} is small compared to the detector response. An analysis window of \pm 20~\mathrm{keV} centered on each parent line–large compared to detector resolution but small compared to nuclear transitions–has \mathcal{O}(10^{2}) bins and thus offers sufficient sideband data for background characterization. In a strong capture line with \gtrsim 10^{4} detected counts, a standard Gaussian peak fit to the data determines the centroid position E_{0} to a statistical accuracy of the order \delta\mu_{0}\sim\frac{\sigma_{\rm det}}{\sqrt{N}}\sim 10\text{--}50~\mathrm{eV}, (3) while the intensity of the line is determined to a percent level or better. Over this small analysis window, the background varies only at energy scales much larger than both the detector resolution and the scanned offset \Delta . This separation of scales motivates treating the parent-line parameters as effectively fixed inputs in the subsequent satellite search. III.2 Stage I: Parent-Line Fits In the first stage, the individual parent capture lines are analyzed separately in a region of interest. These values, for each energy bin, are assumed to be Poisson-distributed, including the contributions of the parametric peak shape model and the smooth local background. The peak model is the sum of a Gaussian function and an exponential function, the latter modeling the charge collection