Migdal Ionization as a Probe of Light Dark Matter from Nuclear Transition

Technical Summary: Migdal Ionization as a Probe of Light Dark Matter from Nuclear Transition

Problem Statement
The search for light dark matter (DM) candidates (masses $\lesssim$ MeV) faces a critical challenge in direct detection: the nuclear recoils induced by such particles typically fall far below the energy thresholds of current semiconductor detectors. In nuclear reactors, which serve as intense artificial sources of light DM via the decay of vector mediators emitted during nuclear de-excitation ($N^* \to N + V \to N + \chi\bar{\chi}$), this limitation is exacerbated by the "quenching effect." In standard elastic scattering, a significant fraction of the nuclear recoil energy is dissipated into lattice vibrations rather than ionization, resulting in electron-equivalent energies often below 300 eVee. Consequently, conventional elastic scattering signals from reactor-produced DM remain undetectable in low-threshold germanium detectors like TEXONO.

Methodology
This work proposes utilizing the Migdal effect to circumvent the quenching threshold. The authors model a scenario where a kinetically mixed dark photon ($V$) couples to a Dirac fermion DM state ($\chi$).

1. Source Modeling: The DM flux is calculated based on neutron-capture nuclear de-excitation in $^{238}\text{U}$ and $^{10}\text{B}$ within a nuclear reactor. The process involves the emission of an on-shell vector mediator ($V$) with energy $E_V \simeq \omega_i$ (nuclear transition energy), followed by the invisible decay $V \to \chi\bar{\chi}$. This produces a box-shaped DM energy spectrum.
2. Detector Response: The authors compute the response of a low-threshold germanium detector by combining:
   • Elastic nuclear scattering cross-sections normalized to a reference DM-proton cross-section ($\bar{\sigma}_{\chi p}$).
   • Nuclear recoil quenching factors calculated via the Lindhard prescription.
   • Migdal ionization probabilities: Calculated in the isolated-atom approximation for germanium orbitals (K through N shells) using the cFAC code. The total deposited energy is modeled as $E_{\text{det}} = q_{\text{nr}}(E_{\text{nr}})E_{\text{nr}} + E_{\text{nl}} + E_e$, where the electronic energy ($E_{\text{nl}} + E_e$) from the sudden nuclear acceleration lifts the total signal above the detector threshold.
3. Data Analysis: The study utilizes the public ON–OFF residual spectrum from the TEXONO experiment (measured at the Kuo-Sheng Reactor Neutrino Laboratory). This residual spectrum, representing the difference between reactor-on and reactor-off data, isolates the reactor-produced signal while suppressing reactor-independent backgrounds. A $\chi^2$ analysis is performed to derive 95% Confidence Level (C.L.) upper limits on the coupling parameter $\epsilon$ and the reference cross-section $\bar{\sigma}_{\chi p}$.

Key Contributions and Results

• New Search Strategy: The paper establishes a viable detection channel for sub-MeV reactor-produced DM by leveraging the Migdal effect, which generates an accompanying ionization signal that raises the deposited energy above the 300 eVee threshold of the TEXONO detector.
• Flux Calculation: The authors provide a detailed calculation of the DM flux from specific E1 neutron-capture transitions (3.297 and 4.060 MeV from $^{238}\text{U}$; 4.711 and 7.007 MeV from $^{10}\text{B}$), demonstrating that the resulting DM spectrum is box-shaped with endpoints determined by the mediator mass and nuclear transition energy.
• Experimental Limits: Using TEXONO data, the authors set stringent new limits on the DM-nucleus interaction for DM masses in the range $0.01 \text{ MeV} \lesssim m_\chi \lesssim 2.6 \text{ MeV}$ and mediator masses $3.2 \text{ MeV} \le m_V \le 6.9 \text{ MeV}$.
  • The derived 95% C.L. upper limits on the reference cross-section $\bar{\sigma}_{\chi p}$ range approximately from $6.7 \times 10^{-35}$ to $8.0 \times 10^{-33} \text{ cm}^2$.
  • The analysis shows that for the benchmark parameters, the elastic nuclear recoil contribution remains below threshold, while the Migdal signal (dominated by M and L shell ionizations) populates the observable energy window.
• Model Independence: The derived limits are independent of the cosmological abundance of $\chi$. They rely solely on the laboratory-produced flux and do not require $\chi$ to constitute the total observed dark matter density, distinguishing them from cosmological constraints (e.g., CMB, Lyman-$\alpha$).

Significance
The paper claims that this work provides a complementary bound to existing cosmological and astrophysical limits on light dark matter. By utilizing the high flux of relativistic particles from nuclear reactors and the Migdal effect, the study probes a sub-MeV parameter space that is inaccessible to standard halo searches and conventional elastic scattering experiments. The results demonstrate that reactor-based experiments, when combined with Migdal ionization analysis, can serve as powerful tools for constraining light dark matter models without assumptions regarding the relic density of the dark sector.