Dark photon searches in the photon channel
D. Aristizabal Sierra, A. Betancur, K. Pohl, J. Velez · 2026
Read Dark photon searches in the photon channel by D. Aristizabal Sierra, A. Betancur, K. Pohl, J. Velez (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-2605-13696.txt.
Introduction
D. Aristizabal Sierra and three colleagues explore dark photon searches through shape contrasts in pion decay photons. With 70-micron tungsten foils and a 1 GeV proton beam, they build a GEANT4 model of photon output, catching and background noise. Many runs with a 10 to 50 beam could probe new regions where the dark photon decays mostly unseen. The results are highly model independent.
Catalog record
| CREATOR | D. Aristizabal Sierra, A. Betancur, K. Pohl, J. Velez |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 24281 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2605.13696
Browse the full library · About the Tower of Babel
Full text
The opening of the work, from the library's hosted text:
Dark photon searches in the photon channel D. Aristizabal Sierra, A. Betancur, K. Pohl, J. Velez Abstract Spectral shape differences between photons produced in π 0 → γ + γ \pi^{0}\to\gamma+\gamma and π 0 → γ + A D \pi^{0}\to\gamma+A_{D} may provide a new avenue for dark photon searches. Assuming 70 μ \mu m thick tungsten foils separated by 200 μ \mu m and a 1 GeV proton beam, we developed a GEANT4 model to estimate photon production and detection including background. Our results demonstrate that multiple campaign runs with a 10-50 μ \mu A beam could probe previously unexplored regions of parameter space in models where the dark photon has predominantly invisible decays. The results are highly model independent. Introduction. Beam-dump experiments provide the conditions to test dark sector models, in particular light dark matter (LDM) scenarios Batell:2009di . A high-intensity electron or proton beam impinged on a highly dense target material can produce directly, or indirectly through meson decays or Bremmstrahlung processes, dark sector particles. In vector portal models at GeV energies, the dark photon can decay either visibly or invisibly. Signals are thus Standard Model (SM) fermion or DM pairs. Visible modes in the form of resonances are searched with tracking systems and e.g. electromagnetic calorimeters. Dedicated experiments such as DarkQuest, NA64, SHiP, FASER, HIKE, CODEX-b, and HPS have the capability for such searches Batell:2020vqn ; Gninenko:2013rka ; Andreas:2013lya ; Alekhin:2015byh ; Feng:2008ya ; HIKE:2022qra ; CODEX-b:2019jve ; Baltzell:2022rpd . Because of the signals they predict, these scenarios are tightly constrained. Models where the dark photon decays predominantly to DM pairs, instead, can be tested with the same technology employed in neutrino detection. From that perspective, dedicated accelerator neutrinos facilities provide a suitable environment to test these types of scenarios (see e.g. deNiverville:2016rqh ). Dark sector particles can be potentially produced in the collision of a proton beam on a fixed target. In vector portal models, dark photon decays to DM pairs produce a boosted DM flux that can be further detected with the same detectors used for neutrino physics. Depending on the collision energy, different production mechanisms operate: Meson decays, proton Bremsstrahlung and Drell-Yang processes. Neutrino experiments with capabilities for such searches include COHERENT, CCM, NOVA, DUNE and DUNE-PRISM COHERENT:2015mry ; CCM:2021leg ; NOvA:2025ykl ; DUNE:2021tad ; DUNE:2020fgq . Deployment of dedicated neutrino detectors at ESS, J-PARC and CSNS will offer as well opportunities for this type of searches AristizabalSierra:2026jgp . Sensitivities depend on the kinetic mixing strength ( \epsilon ), the dark photon and DM masses ( m_{A_{D}} and m_{\phi} ), and the dark sector fine-structure constant. Constraints from experimental data are usually presented in terms of an “effective” coupling and m_{\phi} , specializing results to particular benchmark models. From that perspective, limits are model-dependent. Following phenomenological analysis Dutta:2019nbn , dark sector searches in the MeV regime have been conducted by the COHERENT and the CCM collaborations COHERENT:2022pli ; COHERENT:2021pvd ; CCM:2021yzc ; CCM:2021leg . Further analyses using \nu-^{12}\text{C} inelastic scattering data have proven to provide relevant limits as well Dutta:2023fij . Although at a different energy scale, SBND and DUNE forecasted sensitivities from \pi^{0} decays have been calculated in Ref. Dutta:2026dvg ; DeRomeri:2019kic . To the best of our knowledge, in the region m_{A_{D}}\lesssim m_{\pi^{0}} and for a dark photon decaying invisibly, the most competitive limits in the m_{A_{D}}-\epsilon parameter space have been placed by NA64 at the CERN SPS NA64:2016oww ; NA64:2023wbi . Results in Ref. NA64:2016oww followed from dark photon Bremsstrahlung emission with 2.75\times 10^{9} electrons-on-target (EoT). The search relied on missing energy signals stemming from dark photon decays to DM pairs. Results in Ref. NA64:2023wbi adopted the same strategy, but combining the statistics from the 2016-2022 runs with overall 9.37\times 10^{11} EoT. In the region of interest, limits from BaBar BaBar:2017tiz have been surpassed by the latest results from NA64. In contrast to limits derived from LDM signals, these limits depend solely on \epsilon and m_{A_{D}} . In this letter, we point out that searches for dark photon signals may be done as well by looking into the photon spectra produced by \pi^{0} decays. This requires a proper isolation of the photons produced in these decays, a clean detection technique, and a good understanding of possible backgrounds. Inspired by measurements of the \pi^{0} lifetime carried out at CERN SPS Atherton:1985av , we suggest that those requirements are met by an arrangement consisting of a 10-50 kW proton beam striking a 70\,\mu\text{m} thick tungsten foil at \sim 1\,\text{GeV} . Detection of the photons produced in \pi^{0} decays is done with the aid of a second identical foil located at 200\,\mu\text{m} from the first one. Positrons produced in photon conversion in the field of nuclei enable doing so. We developed a GEANT4 model to determine the feasibility of this arrangement. In particular, the photon and positron spectra produced by interactions in the first foil and the resulting positron spectra in the second foil. These results enabled the identification of relevant positron background. The model has been validated with COHERENT simulated data on \pi^{0} production COHERENT:2021yvp . By rescaling these results to two benchmark values for the protons-on-target (POT), 10^{21} and 10^{22} , we provide a best-case estimate of sensitivities. We show that this approach allows testing regions not yet entirely explored. Theoretical framework and photon spectra. Phenomenological aspects of dark photon scenarios are given by the following Lagrangian terms Boehm:2002yz ; Pospelov:2007mp ; Feng:2008ya -\mathcal{L}\supset-\epsilon\,e\,j^{\text{EM}}_{\mu}A_{D}^{\mu}+g_{D}j_{\mu}^{X}A_{D}^{\mu}+\tan\theta_{W}\,\epsilon\,g_{D}\,j_{\mu}^{X}\,Z^{\mu}\ . (1) Here \epsilon\equiv\epsilon^{\prime}\cos\theta_{W} , \theta_{W} refers to the weak mixing angle, \epsilon to the kinetic mixing strength coupling, g_{D} to the dark sector coupling and j^{\text{EM}}_{\mu} and j_{\mu}^{X} to the EM and dark sector currents. Conventional searches are based on visible and invisible channel signals. In the former case decay modes of the dark photon to SM fermion pairs, while in the latter traces left by the DM through its interaction with electrons or protons. At sufficiently low collision energies, \mathcal{O}(T_{p})\simeq 1\, GeV, the main production mechanism is neutral pion decays. This production mechanism is particularly relevant in spallation neutron sources where hadron activity is limited to unflavored mesons. In the regime m_{A_{D}}<m_{\pi}^{0} with \Gamma_{A_{D}}^{\text{Tot}}\ll m_{A_{D}} (where the narrow-width approximation is valid), \pi^{0} meson decays produce an on-shell dark photon that further decays to a DM pair. Searches in low-threshold coherent elastic neutrino-nucleus scattering (CE \nu NS) detectors, rely on this discovery mode. In addition to A_{D}^{\mu} , the \pi^{0} decay mode involves as well a final-state photon ( semi-invisible mode ). In the \pi^{0} decay rest frame the photon energy spectrum per decay reads \frac{d\Gamma(\pi^{0}\to\gamma+A_{D})}{dE^{*}_{\gamma}}=\Gamma(\pi^{0}\to\gamma+A_{D})\;\delta(E_{\gamma}^{*}-E_{\gamma}^{0})\ , (2) where E_{\gamma}^{*} refers to the final-state photon in that frame and the monochromatic line is located at E_{\gamma}^{0}=(m_{\pi^{0}}^{2}-m_{A_{D}}^{2})/2/m_{\pi^{0}} . The total decay width follows from the full visible mode decay width \Gamma(\pi^{0}\to\gamma+\gamma) , enlarging phase space by a factor of 2 and accounting for the massive final-state dark photon, namely \Gamma(\pi^{0}\to\gamma+A_{D})=2\epsilon^{2}\Gamma(\pi^{0}\to\gamma+\gamma)\left(1-\frac{m_{A_{D}}^{2}}{m_{\pi^{0}}^{2}}\right)^{3}\ . (3) From Eq. ( 2 ), one can see that the photon spectrum in the semi-invisible mode consist of a monochromatic line shifted towards small energy values as the dark photon mass increases. In that frame one could—in principle—distinguish the photon emitted in the full visible mode from the one emitted in the semi-invisible mode, in particular