Aluminum-Based Superconducting Tunnel Junction Sensors for Nuclear Recoil Spectroscopy

This paper reports the development and characterization of three iterations of aluminum-based superconducting tunnel junction (Al-STJ) sensors, culminating in a device with 2.96 eV energy resolution at 50 eV, to enable systematic material studies and improve nuclear recoil spectroscopy for the BeEST experiment's search for sub-MeV sterile neutrinos.

The Gist

Imagine you are trying to listen to a very faint, specific whisper in a noisy room. That is essentially what the BeEST experiment is trying to do, but instead of a whisper, they are listening to the tiny "kick" (recoil) an atom gives when it decays. They are looking for a ghostly particle called a sterile neutrino, which might explain why the universe has mass.

To catch this whisper, they use special sensors called Superconducting Tunnel Junctions (STJs). Think of these sensors as ultra-sensitive microphones that can measure the energy of a single atom's movement with incredible precision.

Here is the story of how the scientists built a new type of microphone using Aluminum to improve their search.

The Problem: The "Tantalum" Microphone

Previously, the team used sensors made of Tantalum (a heavy metal). These worked well, but there was a problem: the metal itself changed the sound of the whisper.

• The Analogy: Imagine trying to record a singer, but the microphone is made of a material that slightly muddies the voice or adds a weird echo. The scientists couldn't tell if the weird echo was part of the singer's voice (new physics) or just the microphone's fault (material effects).
• The Goal: They needed a microphone made of a different material to see if the "echo" changed. If the echo changed, they knew it was the microphone. If the echo stayed the same, they might have found something new about the universe.

The Solution: The "Aluminum" Microphone

The team decided to build their sensors using Aluminum instead of Tantalum. Aluminum is lighter and has different properties, which should change how it interacts with the decaying atoms.

They built these new sensors in three generations, like upgrading a smartphone three times in a row:

1. The First Prototype: "The Heavy Hitter"

• What they did: They made the Aluminum sensors with the same thickness as the old Tantalum ones.
• The Result: It was like putting a heavy coat on a sensitive microphone. The signal was too weak, and the "static" (electronic noise) was too loud. They could hear the main notes of the song (the nuclear decay), but the sound was fuzzy.
• Key Finding: Even with the fuzziness, they proved it was possible to use Aluminum sensors to hear these atomic kicks.

2. The Second Prototype: "The Floating Island"

• What they did: They tried to make the sensors float on a tiny, thin membrane (like a piece of paper suspended in air) to block out background noise from the floor (the silicon substrate).
• The Result: The sensors worked perfectly in terms of sound quality, but the manufacturing process was tricky. Many of the sensors broke or short-circuited during the "floating" process.
• Key Finding: The idea of floating sensors is sound, but they needed to fix the manufacturing to stop breaking them.

3. The Third Prototype: "The High-Fidelity Upgrade"

• What they did: They went back to the solid base but made the Aluminum layers thinner and the tunnel barrier (the gate the particles pass through) more open.
• The Result: This was the breakthrough. By thinning the layers, the signal became much stronger, and the static noise dropped significantly.
• The Achievement: They achieved a crystal-clear resolution. They could distinguish energy differences as small as 2.96 electron-volts (eV). To put that in perspective, if the energy of a single photon of light was a dollar, this sensor could tell the difference between a dollar and a dollar minus a fraction of a penny.

Why Does This Matter?

The paper claims that these new Aluminum sensors are now ready for the next phase of the experiment.

• The "Echo" Test: By comparing the "Aluminum microphone" to the old "Tantalum microphone," the scientists can now separate the "echo" caused by the material from the actual "song" of the neutrino.
• The Future: With these clearer sensors, they can look for the tiny, subtle shifts in the atomic recoil that would prove the existence of those ghostly sterile neutrinos.

Summary

The paper is a success story of engineering iteration. The team started with a heavy, noisy sensor, tried a fragile floating design, and finally settled on a refined, thin, high-sensitivity Aluminum sensor. They didn't discover the sterile neutrino in this paper; instead, they built the perfect tool needed to find it in the future by ensuring they know exactly what their own equipment is doing.

Technical Summary

Technical Summary: Aluminum-Based Superconducting Tunnel Junction Sensors for Nuclear Recoil Spectroscopy

Problem Statement
The BeEST (Beryllium Electron capture in Superconducting Tunnel junctions) experiment seeks to detect sub-MeV sterile neutrinos by measuring nuclear recoil energies from the electron capture (EC) decay of implanted $^7$Be. While the collaboration has successfully utilized tantalum-based STJs (Ta-STJs) to set leading laboratory limits, systematic uncertainties arising from detector material properties and the specific interaction of $^7$Be implants with the sensor material limit experimental sensitivity. Features in the recoil spectrum, such as peak centroids, linewidths, and shapes, are influenced by the atomic configuration of the daughter $^7$Li, its interaction with the superconducting sensor, and processes like Auger emission and Doppler broadening. To distinguish potential Beyond the Standard Model (BSM) physics signatures from these material-dependent effects, the collaboration requires a complementary testbed using a different absorber material.

Methodology
The authors developed aluminum-based Superconducting Tunnel Junctions (Al-STJs) as an alternative to Ta-STJs. The design was informed by previous Ta-STJ geometries but adapted for the specific energy range and secondary radiations of $^7$Be decay. Key design features included:

• Geometry: Three pixel sizes ($(70 \mu m)^2$, $(130 \mu m)^2$, and $(200 \mu m)^2$) to balance active area and resolution. Pixels were rotated $45^\circ$ to suppress DC Josephson current.
• Materials: An Al-AlO$_x$-Al trilayer structure. Unlike Ta-STJs, which use a thin Al layer to trap quasiparticles, the Al-STJs lack a dedicated trap layer due to the small aluminum superconducting gap ($\Delta_{Al} \approx 0.18$ meV). Instead, a higher-gap niobium (Nb) plug is used in ground traces to confine thermalized quasiparticles and prevent crosstalk.
• Fabrication Iterations: Three distinct fabrication runs were performed to optimize performance:
  1. Iteration 1: Used electrode thicknesses matching existing Ta-STJs (265 nm base, 200 nm top) on SiN-coated wafers without backside etching.
  2. Iteration 2: Introduced backside deep reactive ion etching (DRIE) to suspend pixels on thin (0.5 $\mu$m) SiN membranes to suppress substrate phonon background.
  3. Iteration 3: Eliminated backside etching. Electrode thicknesses were reduced (200 nm base, 100 nm top) and tunnel barrier transmissivity was increased (by reducing oxygen exposure) to enhance signal amplitude and responsivity.

The devices were tested in a wet two-stage adiabatic demagnetization refrigerator (ADR) at $\sim$100 mK. Characterization involved current-voltage ($I-V$) measurements, laser calibration using a pulsed 355 nm source, and nuclear recoil spectroscopy following $^7$Be implantation at TRIUMF.

Key Results

• Iteration 1 (Proof of Concept): These devices demonstrated high junction quality with low leakage but exhibited low responsivity ($\sim$20 nA/keV) due to thick electrodes. Consequently, electronic noise ($\sim$8 eV FWHM) dominated the energy resolution, preventing the resolution of the 3.5 eV-spaced laser peaks. However, they successfully resolved the primary $^7$Be EC decay peaks (K-GS, K-ES, L-GS, L-ES). Notably, the K-ES peak appeared narrower in Al-STJs than in Ta-STJs, suggesting faster thermalization of the recoiling nucleus in aluminum.
• Iteration 2 (Membrane Suspended): These devices showed excellent $I-V$ characteristics with low leakage current (<10 nA) but suffered from low yield due to shorted pixels, likely caused by high-voltage discharge during the backside etching process.
• Iteration 3 (Optimized for Resolution): By thinning electrodes and increasing barrier transmissivity, responsivity increased to $\sim$100 nA/keV. These devices successfully resolved individual peaks in the laser calibration spectrum. They achieved an energy resolution of 2.96 eV FWHM at 50 eV (and generally 2–4 eV FWHM below 100 eV). This resolution is sufficient to calibrate the nuclear recoil spectrum in the BeEST energy range.

Significance and Claims
The paper establishes Al-STJs as viable detectors for systematic material studies within the BeEST experiment. The primary significance lies in the ability to perform a comparative study of $^7$Be decay spectra in Al- and Ta-STJs. This comparison allows the collaboration to:

1. Isolate Material Effects: Systematically test subtle, material-dependent effects (such as peak broadening and tailing) that are currently confounding the search for sterile neutrinos.
2. Validate Systematics: Provide a benchmark to distinguish between spectral features caused by detector physics (e.g., Doppler broadening differences due to mass and packing density) and potential BSM physics signals.
3. Enable Future Precision: The achieved energy resolution (2–4 eV) is sufficient for the current phase of nuclear recoil spectroscopy, enabling future experiments with improved statistics to constrain sterile neutrino masses and neutrino wavepacket properties with reduced systematic uncertainty.

The authors modestly note that while the smaller superconducting gap of aluminum theoretically offers higher intrinsic energy resolution, the broadening of the signal peaks in the $^7$Be decay currently limits the relevance of this advantage. However, the demonstrated capability to fabricate high-resolution Al-STJs provides a critical tool for understanding and mitigating detector-induced systematics in the ongoing search for sterile neutrinos.