The observed neutrino oscillations strongly suggest physics beyond the Standard Model (SM), motivating the exploration of sterile neutrinos. These sterile neutrinos, which mix with active neutrinos, could have masses spanning many orders of magnitude, with the MeV-GeV range being particularly interesting for laboratory experiments. High-intensity facilities like the Deep Underground Neutrino Experiment (DUNE) offer unique opportunities for searching for such particles. DUNE's Near Detector (ND) complex, with its high-power neutrino beam and Liquid Argon (LAr) detectors, has the capability to search for new physics through detailed particle reconstruction. The mobile nature of the ND, allowing for off-axis measurements, enhances its potential for observing low-energy new physics. Furthermore, recent observations of coherent elastic neutrino-nucleus scattering (CEνNS) and the capabilities of dark matter (DM) direct detection (DD) experiments have expanded the toolkit for probing the neutrino sector and searching for new physics. Experiments like COHERENT, utilizing neutrinos from stopped-pion beams, are particularly well-suited for such explorations. DM DD experiments, although primarily searching for weakly interacting massive particles (WIMPs), can also detect astrophysical neutrinos, which offer unique opportunities to explore new physics, such as solar neutrino induced elastic neutrino-electron scattering (EvES) events. The study will investigate the complementarity of these different facilities in probing the up-scattering production of a new sterile fermion in the MeV range and explore a wider region of parameter space.

The authors review previous searches for sterile fermions, focusing on those involving neutrino up-scattering. They note that prior work often assumed only scalar mediators or effective interactions by integrating out the mediator's mass. This paper expands on previous analyses by considering all Lorentz-invariant interactions with explicit mediator dependence, analyzing the latest data from COHERENT, XENONnT, and LZ, and providing projected sensitivities for DUNE-ND.

The study uses a simplified phenomenological model extending the SM Lagrangian with new neutrino-quark and neutrino-electron interactions. All possible interactions (scalar, pseudoscalar, vector, axial-vector, and tensor) are considered. The analysis includes the computation of EvES and CEνNS cross sections, considering light mediators and explicit propagator dependence. The DUNE-ND sensitivity is assessed by simulating EvES events in the near detector, considering both standard and tau-optimized neutrino beams, and various on-axis and off-axis detector locations. A χ² fit is employed to analyze simulated DUNE-ND data, including nuisance parameters to account for uncertainties in the neutrino flux and background normalizations. Existing data from XENONnT and LZ are analyzed using Poissonian and Gaussian χ² functions, incorporating background and flux uncertainties. COHERENT data are analyzed using a similar Poissonian least-squares function, including energy and time information and several nuisance parameters, such as efficiency uncertainties, nuclear form factor uncertainty, and beam timing uncertainty. The production and subsequent decay of the sterile fermion within the detector is also taken into consideration, and respective decay rates are calculated.

The DUNE-ND is projected to provide the most stringent constraints at larger mediator and sterile fermion masses, reaching couplings as small as $g_a \sim 10^{-4}, 2 \times 10^{-5}, 10^{-5}$ for scalar/pseudoscalar, vector/axial-vector, and tensor interactions, respectively, and $m_x \sim$ few/tens of MeV after 7 years of data taking. The analysis reveals that the on-axis location yields the most stringent results, with off-axis locations providing less stringent constraints due to reduced neutrino flux. The study shows the complementarity of DUNE-ND with existing data from COHERENT, XENONnT, and LZ. XENONnT and LZ are most sensitive at very low masses due to their low detection thresholds. COHERENT provides improved constraints for $m_\chi \ge 300$ keV. The study shows that DUNE-ND will significantly improve sensitivities for sufficiently large mediator or sterile fermion masses, particularly for pseudoscalar, axial-vector, and tensor interactions. Considering the sterile fermion decay within the detector, DUNE-ND is projected to probe a larger region of the parameter space, up to approximately 200 MeV for the sterile fermion mass, compared to COHERENT’s sensitivity limit of around 52 MeV. The use of the τ-optimized beam at DUNE-ND leads to slight improvements in sensitivity for larger mediator and sterile fermion masses.

The results highlight the importance of using multiple facilities with different neutrino sources and detection techniques to probe the up-scattering production of a sterile fermion. The complementarity of DUNE-ND, COHERENT, XENONnT, and LZ allows for a more comprehensive exploration of the parameter space and improved constraints on various interaction types. The differences in sensitivities across different experiments stem from kinematic considerations and the varying neutrino energy spectra and detection thresholds. The study's findings are relevant for understanding the nature of neutrino interactions and the potential existence of new physics beyond the Standard Model. The projected sensitivities for DUNE-ND suggest its significant potential in improving constraints on sterile fermion properties.

This paper demonstrates the complementarity of DUNE-ND, COHERENT, XENONnT, and LZ experiments in constraining the production of a sterile fermion through neutrino up-scattering. DUNE-ND is projected to significantly improve existing constraints, especially at larger masses. Future research could focus on developing more realistic UV-complete models to better interpret the results and refine the theoretical framework.

The analysis employs a simplified phenomenological model. Uncertainties in the neutrino fluxes, background models, and detector efficiencies could affect the final sensitivities. A detailed Monte Carlo simulation of sterile fermion decays would provide more precise results. The study focuses primarily on electron and nuclear recoil events and does not exhaustively cover all possible sterile fermion decay channels. Some assumptions and approximations, like the use of the Klein-Nystrand parametrization for the nuclear form factor, are made. The calculation of sterile fermion decay events is based on a simplified order of magnitude estimation. The impact of some uncertainties, such as the resolution of electron energy, is not fully covered.