Fabrication and Characterization of X-ray TES Detectors Based on Annular AlMn Alloy Films

AlMn alloy flms are widely fabricated into superconducting transition edge sensors (TESs) for the detection of cosmic microwave background radiation. However, the application in X-ray or gamma-ray detection based on AlMn TES is rarely reported. In this study, X-ray TES detectors based on unique annular AlMn flms are devel-oped. The fabrication processes of TES detectors are introduced in detail. The char-acteristics of three TES samples are evaluated in a dilution refrigerator. The results demonstrate that the I-V characteristics of the three annular TES detectors are highly consistent. The TES detector with the smallest absorber achieved the best energy resolution of 11.0 eV @ 5.9 keV, which is inferior to the theoretical value. The dis-crepancy is mainly attributed to the larger readout electronics noise than expected.

💡 Research Summary

This paper presents the design, fabrication, and performance evaluation of X‑ray transition‑edge sensor (TES) detectors that employ annular‑shaped aluminum‑manganese (AlMn) alloy films. Conventional TES designs for X‑ray detection typically use narrow rectangular geometries, which limit thermal conductance because the conductance scales with the perimeter of the suspended membrane. To overcome this limitation, the authors introduce a novel annular geometry in which the TES is a ring with independently tunable inner (Ri) and outer (Ro) radii. By fixing Ro to extend to the edge of the Si₃N₄/SiO₂ membrane, the thermal conductance G can be set via the outer perimeter, while the normal resistance Rn is adjusted by varying Ri, according to Rn = ρ·ln(Ro/Ri)/(2πt). This decoupling provides design flexibility that is especially valuable for low‑temperature (≈100 mK) operation required for high‑resolution X‑ray spectroscopy.

Fabrication began on a 4‑inch silicon wafer with a 200 µm silicon core sandwiched between 1 µm Si₃N₄ and 0.2 µm SiO₂ layers. A 180 nm AlMn film (2000 ppm Mn) was DC‑magnetron sputtered, then annealed at 220 °C for 10 min to achieve a critical temperature (Tc) between 100 mK and 120 mK, as verified by a Tc‑versus‑bake‑temperature curve. The AlMn film was wet‑etched into a 100 µm‑diameter circle, after which a 220 nm SiO₂ layer was deposited to protect the AlMn and to electrically isolate the inner Nb electrode. A Ti/Au seed layer (10 nm/200 nm) was electron‑beam evaporated onto the central region for subsequent electroplating of the gold absorber. The absorber, made of Au (2.7 µm thick), was patterned as square pads with side lengths of 100 µm, 140 µm, and 160 µm to study the effect of heat capacity. Electroplating formed the absorber, which was supported by five 10 µm‑wide stems that also provided a thermal link to the TES. After protecting the absorber with photoresist, deep reactive‑ion etching (DRIE) released the membrane, and all resist and glue were removed, leaving a suspended TES‑absorber assembly. Scanning electron microscopy confirmed the structural integrity of the devices.

Three detectors (labeled #1, #2, #3) with different absorber sizes were mounted in a dilution refrigerator at a bath temperature of 100 mK. A two‑stage SQUID readout (gain 10⁵ V/A) recorded current pulses from a ⁵⁵Fe source (5.9 keV photons). I‑V curves measured from 50 mK to 120 mK showed highly consistent behavior across all devices; the normal resistance was ≈27 mΩ, slightly higher than the design value of 19.7 mΩ, likely due to film‑thickness variations and unaccounted AlMn beneath the Nb electrodes. Joule‑power versus temperature data fitted the relation P = K(Tⁿ – Tbⁿ) with n ≈ 3.2–3.3, yielding thermal conductances G of 255–278 pW/K at Tc ≈ 115 mK, in reasonable agreement with the radiative phonon‑transport model G = 4ξAσT³.

Pulse analysis employed a double‑exponential fit to extract peak amplitude and decay constants (τ⁻, τ). From the slopes of the resistance‑temperature and resistance‑current curves, the temperature sensitivity αI, current sensitivity βI, and loop gain LI were derived. Using τ and G, the effective heat capacity C₀ was calculated (0.85–1.72 pJ/K), which is 3–4 times larger than expected from literature values for similar Al‑based TESs. The authors note that this excess heat capacity may arise from the Au absorber, the AlMn film, or an unidentified systematic effect, and suggest future measurements without the absorber to isolate the contributions.

Energy resolution, expressed as full‑width‑half‑maximum (FWHM) at 5.9 keV, improved with decreasing absorber size: detector #1 (160 µm) achieved 20.5 ± 1.3 eV, detector #2 (140 µm) 15.3 ± 1.1 eV, and detector #3 (100 µm) 11.0 ± 1.0 eV. The theoretical limit, based on Johnson noise and thermal‑fluctuation noise, predicts ≈5 eV, indicating that the measured performance is limited by excess readout noise. Spectral analysis of the current noise below 20 kHz revealed an average noise level of 1.68 nA/√Hz, substantially higher than the sum of calculated Johnson and phonon‑noise contributions. The authors attribute this discrepancy to the SQUID electronics, external electromagnetic interference, and possibly insufficient shielding or grounding.

In summary, the annular AlMn TES architecture successfully decouples thermal conductance from electrical resistance, offering a versatile platform for X‑ray microcalorimetry. The fabricated devices demonstrate reproducible I‑V characteristics and achieve sub‑12 eV energy resolution, albeit still above the theoretical optimum. The primary obstacles identified are (1) higher than expected heat capacity, and (2) readout electronics noise exceeding the Johnson/thermal‑fluctuation budget. Future work should focus on low‑noise SQUID design, detailed heat‑capacity budgeting of each component, and optimization of the membrane geometry to further reduce G while maintaining mechanical robustness. Such improvements could bring AlMn‑based annular TESs to the performance levels required for next‑generation X‑ray astronomy missions, including polarimetry and high‑resolution spectroscopy of astrophysical lines.