LiFE-SNS: LiF Experiment for keV-scale Sterile Neutrino Search

The LiF Experiment for keV-scale Sterile Neutrino Search (LiFE-SNS) aims to probe sterile neutrinos through precision measurements of the tritium $β$ spectrum. Tritium nuclei are produced and embedded in LiF crystals via the ${}^{6}\mathrm{Li}(n,α){}^{3}\mathrm{H}$ reaction, allowing thermal calorimetric detection of $β$ decays with magnetic microcalorimeters (MMCs) operated at millikelvin temperatures. We present the detector configuration, background studies, and calibration method, including modeling of position-dependent response and characterization of detector nonlinearity. We also discuss potential sources of systematic uncertainty relevant to the sterile-neutrino search. While the first phase of LiFE-SNS has been completed, this paper focuses on calibration and detector characterization. The achieved performance enables precision $β$-spectrum measurements, and projected sensitivities indicate competitive reach in the keV mass region.

💡 Research Summary

The LiF Experiment for keV‑scale Sterile Neutrino Search (LiFE‑SNS) is a novel approach to probe sterile neutrinos in the keV mass range by measuring the tritium (³H) β‑decay spectrum with unprecedented precision. Sterile neutrinos, motivated by the νMSM and warm dark‑matter scenarios, would manifest as a kink in the β‑spectrum at an energy Q − mₕ, where mₕ is the sterile‑neutrino mass. Detecting such a distortion requires sub‑10 eV energy resolution and a thorough understanding of detector response.

Production of the source – Tritium is generated in situ by the ⁶Li(n,α)³H reaction. Thermal neutrons from AmBe and ²⁵²Cf sources at the KRISS facility irradiate LiF crystals. Geant4 simulations show that a 1 cm³ crystal exposed for one week yields ~20 Bq of ³H activity, with the tritium atoms displaced ~33 µm from their original lattice sites and concentrated about 35 µm beneath the surface. Historical ESR studies indicate that these interstitial tritium atoms remain stable at room temperature, with negligible diffusion over the multi‑year data‑taking period.

Detector concept – Each detector consists of a LiF absorber and a magnetic micro‑calorimeter (MMC) sensor weakly coupled to a 10 mK heat bath. The MMC uses an Er‑doped Ag paramagnetic layer whose magnetization changes with temperature. A persistent current in a Nb pickup coil provides a static magnetic field, and the resulting flux change is read out by a dc‑SQUID. A thin (5 mm × 5 mm × 300 nm) gold film on the crystal surface acts as a phonon collector, bonded to the MMC with five 25 µm gold wires to ensure efficient thermal coupling while keeping the total heat capacity low.

Calibration strategy – Eleven experimental runs were performed in two cryogenic platforms (ADR and dilution refrigerator). Two calibration sources were employed: ⁵⁵Fe (5.9 keV Mn Kα, 6.5 keV Mn Kβ) and ²⁴¹Am (59.5 keV γ, plus Ag and Cu fluorescence lines at ~20 keV and ~8 keV). By moving the ⁵⁵Fe source to various positions on the crystal, the team mapped the position‑dependent response. Geant4 simulations of photon transport were used to predict the spatial distribution of absorbed X‑rays, taking into account the energy‑dependent attenuation lengths in LiF (0.13 mm at 5.9 keV, 1.8 mm at 14 keV, etc.).

Signal processing – Waveforms were digitized at 500 kS/s with 16‑bit resolution and filtered with a 100–1500 Hz band‑pass. A template built from fully absorbed 59.5 keV events was fitted to each pulse in the time domain to extract the amplitude. The resulting amplitude spectrum displayed clear peaks for all expected lines, confirming the linearity of the MMC over a wide energy range.

Performance and corrections – The detector exhibits a linear response within 0.5 % for energies above 20 keV. Below this, a modest non‑linearity of 0.2–0.3 % is observed and corrected with a polynomial model. Position‑dependent variations, driven by differing phonon collection efficiencies and X‑ray absorption depths, cause up to 3 % amplitude shifts; these are modeled with a third‑order spatial correction derived from the multi‑position calibration data. After applying both corrections, the energy resolution reaches 12 eV FWHM at 59.5 keV and ~30 eV at 5.9 keV, comfortably satisfying the requirement for sterile‑neutrino searches.

Background and systematics – In the dilution refrigerator, a 10 cm lead shield (external) and an additional 10 cm internal lead layer reduce ambient γ‑background to <0.1 cps. Residual backgrounds arise from fluorescence of nearby copper and silver components and from scattered 241Am γ‑rays. Systematic uncertainties are quantified as follows: energy scale (≤10⁻⁴), non‑linearity correction (≤5 × 10⁻⁵), position correction (≤2 × 10⁻⁴), and tritium spatial inhomogeneity (≤1 × 10⁻⁴). These contributions are propagated to the sterile‑neutrino mixing‑angle sensitivity.

Projected sensitivity – Phase 1, consisting of 6 × 10⁸ β‑decays collected over four months with a single detector channel, can reach sin²2θ ≈ 2 × 10⁻⁴ for a sterile mass of 7 keV. Phase 2 envisions a multi‑channel array delivering 10¹² β‑decays over several years, improving the limit to sin²2θ ≈ 5 × 10⁻⁶. This surpasses the current strongest laboratory bound from the BeEST experiment (∼10⁻⁴) and probes a region of parameter space relevant for warm dark‑matter models.

Conclusion – The LiFE‑SNS collaboration has demonstrated that LiF crystals loaded with tritium, read out by MMCs at millikelvin temperatures, provide the required energy resolution, linearity, and controllable systematics for a competitive sterile‑neutrino search. Ongoing work focuses on scaling up the detector array, further reducing background, and extending the data‑taking period to fully exploit the projected Phase 2 sensitivity.