https://arxiv.org/html/2507.07936v1

MI-HET-861

CETUP2025-001

In this work, we investigate monophoton signatures arising from dark matter via a scattering process that is mediated by a virtual scalar and a Standard Model photon. Since the final-state photon carries a large fraction of the initial dark matter’s energy, this process offers a compelling handle for probing scalar portal dark matter scenarios. Their distinctive energy, angular, and timing distributions allow for effective separation of signal from neutrino-induced backgrounds. We analyze several models featuring different couplings to the scalar mediator, with the scalar photon coupling serving as the common detection channel. We considered the flux of dark matter produced both at the target and absorber of neutrino facilities such as the BNB, NuMI, and LBNF, and investigated the sensitivities at the ongoing SBND, ICARUS-NuMI, and future DUNE ND detectors. Our results demonstrate that the sensitivities at the considered experiments, especially DUNE ND, can place significantly improved constraints on viable parameter space in various scenarios.

Over the past century, a remarkable amount of progress has been made in the arena of dark matter (DM) [1], from establishing its existence through various astrophysical evidence [2, 3, 4, 5, 6, 7, 8, 9], to pioneering various experimental pursuits [10, 11, 12, 13, 14] to find its particle nature. Weakly Interacting Massive Particles (WIMPs) have been long sought after as potential DM candidates, as the cold and massive thermal relic abundance is elegantly set by annihilation via weak interaction mediators. Despite its elegance, the lack of a positive WIMP signature [15, 16, 17, 11, 12] motivates us to pursue other candidates.

One such attractive alternative is sub-GeV DM, which requires new portals or mediators lighter than electroweak mediators [18, 19, 20, 21, 22, 23, 24] to acquire the required thermal relic cross section [25, 26, 27]. Many mechanisms have been proposed and employed in pursuit of sub-GeV DM and their interactions, such as missing energy [28, 29, 30], electron and nuclear recoils [31, 32, 33, 34, 35, 36, 37, 38, 39, 12, 40, 41], nuclear deexcitation lines [42, 43], dark trident [44, 45], dark matter internal pair production (DIPP) [46], etc. The relevance of these detection mechanisms depends primarily on the Lorentz structure of the mediators and the beyond the standard model (BSM) scenario in which they appear. In this context, spin-1 mediators have received great attention due to well-motivated anomaly-free gauge extensions [47, 48, 49, 50], as well as the efficiency in both production and detection at beam dump and direct detection experiments.

Scalar mediators have also been naturally implicated in spontaneous symmetry-breaking models [51, 24], through extensions to the Higgs sector [52], modification of mass terms in neutrino oscillations through scalar non-standard interactions (sNSIs) [53, 54, 55, 56], implications on the magnetic moment of muons [57, 58, 59, 60, 61, 62], etc. Several key features of scalar-mediated DM make them more distinctive than commonly sought-after vector-mediated scenarios. Firstly, the annihilation cross section for fermionic dark matter via scalar mediators is -wave [63]. This therefore renders scalar portal DM models to be (CMB)-safe, i.e., preventing them from injecting energy into the photon plasma during recombination [64, 65]. Therefore, scalar portal models do not require DM to be a pseudo-Dirac or complex scalar, such as in vector portal models. Secondly, scalar mediators have the additional freedom to couple to two photons, which is otherwise forbidden for vector mediators. Therefore, phenomenology-wise, scalar portal DM can give rise to interesting signatures through the effective vertex, both at astrophysical environments and at the detection front of terrestrial experiments.

In this paper, we will focus on the implications of the DM-photon interaction at the detection frontier. To this end, we propose a new detection mechanism for dark matter via scalar mediators, which utilizes the mediator’s coupling to two photons to produce a single-photon final state. When dark matter interacts with a material, it can source off-shell scalars which can split into two photons, where one is exchanged with the nucleus and the other is produced on-shell. This can be viewed from the perspective of dark matter inducing an inverse Primakoff scattering by converting its scalar mediator to a photon. Based on recent studies [46, 66, 67], we find that light energetic DM prefers to share its energy with the third particle stemming from the support of a stationary nucleus rather than the nucleus itself. Kinematically, we find that this process leads to high-energy monophoton signals, without the need for minimum thresholds to make this process observable. This is particularly advantageous in comparison to recoil-based experiments, where the electron/nuclear recoil energies are lower.

In order to demonstrate this detection mechanism, we will choose five benchmark Liquid Argon Time Projection Chamber (LArTPC) detectors: Coherent CAPTAIN-Mills (CCM200) [68, 34], Short-Baseline Near Detector (SBND) [69], ICARUS-NuMI [70], MicroBooNE [71], and the near detector at Deep Underground Neutrino Experiment (DUNE ND) [72, 73]. In addition, we also investigate this process in MiniBooNE [74], which uses a mineral oil-based detection medium. These experiments are chosen to cover a wide range of beam energies, starting from 800 MeV CCM200 up to 120 GeV DUNE ND and ICARUS-NuMI. We will also consider a range of scalar portal models that not only include a coupling to photons, but also other Standard Model (SM) fermions. This will enable a study of the rich phenomenology of various production mechanisms at neutrino facilities as well as unique single-photon detection signatures. We will also show how the DM-induced single photon signatures can be distinguished from typical neutrino-induced backgrounds based on the timing distributions at all the above-mentioned detectors.

The paper is organized as follows: We will begin with a brief overview of an umbrella of scalar portal dark matter models and the related phenomenology at neutrino experiments in Section. II. We then discuss these models in the context of our benchmark experiments in Sec. III. Then, we will present our results in Sec. IV and discuss the salient features of the single photon signature. The results will include an analysis of the flux of produced dark matter, timing, energy, and angular spectra of photons at all the experiments, followed by Section. V, discussing the sensitivity plots. We will summarize our results in Sec. VI.

II Models and Lagrangian

We follow an effective field theory approach to study the phenomenology of fermionic dark matter with a scalar mediator, where the latter is a mediator between the SM and DM sectors. The Lagrangian of interest in our analysis is:

where and are the generalized Yukawa couplings of the scalar with SM fermions and dark matter, respectively. The dimension-5 operator contains a coupling , which carries a dimension of inverse mass (). The coupling can appear from loops containing top, bottom, or heavy vector-like quarks [75], which are much heavier than the energy scale corresponding to our choice of sub-GeV to GeV scale neutrino energies in these experiments. Therefore, from a phenomenological standpoint, we will treat all three couplings as independent parameters. We can assume that the neutrinos, like all the other charged leptons, appearing in the Yukawa interactions are Dirac particles such that . One could also consider the neutrinos to have Majorana interactions with the scalar, e.g., where is a triplet scalar. In the second case, is not needed. Detailed UV complete models for light containing interactions with neutrinos and other fermions based on the extensions of the SM Higgs sector have been constructed in Refs. [53, 62].

We focus on minimal scenarios and categorize our study into two classes. The first class sets all Yukawa couplings to zero, such that the photon coupling entirely dictates the phenomenology. The second class builds upon the first by introducing one SM fermion coupling at a time, while fixing the photon coupling to its maximal allowed value. Within this class, we consider four benchmark scenarios: neutrinophilic, electro/muon-philic, and quarkphilic. Although the dark matter coupling enters implicitly through quantities such as cross sections and decay widths, we assume that , which is much larger than the scalar couplings to fermions () and photons (). This effectively renders the scalar invisible. As a result, the observable phenomenology relevant to dark matter detection becomes largely insensitive to the precise value of .

Since dark matter does not directly couple to the SM particles, but via the scalar portal, we first investigate the production mechanisms for the scalar. Our benchmark experiments operate in proton-on-target facilities where high-energy protons impinge on a target, followed by a large decay volume. Therefore, there is a rich flux of photons, electrons, neutral, and charged mesons. These serve as potential sources of scalars, and consequently dark matter, at these facilities. We will now discuss the production mechanisms by segregating them based on the scalar’s coupling to the SM sector, that is, to photons, neutrinos, charged leptons, and quarks. We will limit our discussion on production mechanisms that are dominant for our sub-GeV to GeV scale benchmark neutrino experiments.

{feynman}\vertex\vertexu𝑢uitalic_u\vertexs¯¯𝑠\bar{s}over¯ start_ARG italic_s end_ARG\vertex\vertexd¯¯𝑑\bar{d}over¯ start_ARG italic_d end_ARG\vertexu𝑢uitalic_u\vertex\vertexϕW+superscript𝑊W^{+}italic_W start_POSTSUPERSCRIPT + end_POSTSUPERSCRIPTK+superscript𝐾K^{+}italic_K start_POSTSUPERSCRIPT + end_POSTSUPERSCRIPTπ+superscript𝜋\pi^{+}italic_π start_POSTSUPERSCRIPT + end_POSTSUPERSCRIPT\diagram{feynman}\vertexs¯¯𝑠\bar{s}over¯ start_ARG italic_s end_ARG\vertex\vertex\vertexd¯¯𝑑\bar{d}over¯ start_ARG italic_d end_ARG\vertexu𝑢uitalic_u\vertexu𝑢uitalic_u\vertexϕ\vertext¯¯𝑡\bar{t}over¯ start_ARG italic_t end_ARGK+superscript𝐾K^{+}italic_K start_POSTSUPERSCRIPT + end_POSTSUPERSCRIPTπ+superscript𝜋\pi^{+}italic_π start_POSTSUPERSCRIPT + end_POSTSUPERSCRIPT\diagramt¯¯𝑡\bar{t}over¯ start_ARG italic_t end_ARGFigure 1: Feynman diagrams illustrate the production of (a) photophilic scalars via photon, (b) neutrinophilic and (c) electro/muon-philic scalars, both via three-body decays of charged mesons, (d) up-philic scalars via kaon two-body decays, and (e) via proton bremsstrahlung, and (f) top-philic scalars via one-loop kaon two-body decays.

where, is the atomic number of the material, is the screening length, is the scalar mass and is the fine structure constant. From the cross section, we calculate the probability of a scalar Primakoff for a given scalar mass and photon energy as follows: