Development & first Performance evaluation of multi-element monolithic HPGe detector for X-ray spectroscopy

Development & first Performance evaluation of multi-element monolithic HPGe detector for X-ray spectroscopy
Notice: This research summary and analysis were automatically generated using AI technology. For absolute accuracy, please refer to the [Original Paper Viewer] below or the Original ArXiv Source.

The first operational prototype of a high-purity Germanium (HPGe) detector developed within the European LEAPS-INNOV project is presented in this work. This prototype features a monolithic, multi-element sensor optimized for high-resolution X-ray spectroscopy in the hard X-ray regime, capable of handling high count rates (20-250 kcps/mm2) across a broad energy range (5-100 keV). We discuss here a complete laboratory-based characterization of the detector’s performance, as well as an on-beam evaluation at the BM05 beamline of the ESRF synchrotron facility, using monochromatic X-rays in the 20-50 keV energy range. We provide a detailed performance assessment that also includes a phenomenological defect-depth estimation model.


💡 Research Summary

The paper presents the design, fabrication, and comprehensive performance evaluation of the first operational prototype of a multi‑element monolithic high‑purity germanium (HPGe) detector developed within the European LEAPS‑INNOV project, designated XAFS‑DET. The sensor is a 20 mm × 20 mm, 4.1 mm thick p‑type germanium crystal segmented into ten pixels: a central hexagonal pixel, six surrounding trapezoidal pixels, and three outer ring‑shaped pixels, each covering roughly 20 mm². The front side is n‑type phosphor‑implanted, while the back side is boron‑implanted and electrically biased between +75 V and +300 V (full depletion at +33 V). Leakage currents remain below 1 pA at +100 V, confirming excellent crystal purity.

The front‑end electronics consist of three XGLab‑BRUKER TETRA ASICs (four‑channel charge‑sensitive preamplifiers) mounted on a ceramic front‑end board (FEB) that routes signals from ten pogo‑pin contacts to the ASICs. Gain can be set to low, medium, or high via selectable feedback capacitances (29, 58, 174 fF). Laboratory tests at room temperature showed rise times of 19–23 ns and an equivalent noise charge (ENC) of 36–45 e⁻ for a 1 µs peaking time, improving to 29–35 e⁻ at 10 µs, well within specifications. The back‑end board (BEB) buffers the analog outputs and provides power, bias, and control via a USB‑3.0/Ethernet interface.

Cooling is achieved with an electrically driven Stirling cryocooler (Cryo‑Tel CT) delivering continuous 77 K operation. A combination of copper braids, indium foils, and PEEK supports minimizes thermal gradients, while an active vibration cancellation system reduces mechanical vibrations to 0.04–0.07 m/s² above 50 Hz. The vacuum chamber reaches 2 × 10⁻⁷ mbar, and a zeolite getter maintains long‑term vacuum integrity.

Laboratory characterisation covered (i) signal rise time and preamplifier gain using a ⁵⁵Fe source (5.9 keV Kα, 6.4 keV Kβ); the mean gain was 1.99–2.06 mV/keV with a systematic uncertainty of ±0.02 mV/keV, and rise time decreased to a minimum of ~42 ns at bias voltages above 280 V. (ii) Power spectral density (PSD) measurements from 100 Hz to 50 MHz revealed low‑frequency noise suppression and demonstrated that indium‑foil contacts reduce high‑frequency noise by ~15 %. (iii) Fluorescence measurements with metallic foils showed an energy resolution of ≈180 eV FWHM at 5.9 keV across all pixels, and count‑rate linearity up to 250 kcps mm⁻² with negligible peak broadening. Dead‑time and pile‑up studies confirmed a dead‑time below 5 % at a 1 µs peaking time. (iv) Micro‑beam scans (≤10 µm spot) mapped pixel response, crosstalk (<0.2 %) and identified localized regions with anomalously long rise times, indicative of crystal defects.

A phenomenological defect‑depth model was constructed by correlating the bias‑dependent rise‑time shift with the electric field distribution; the model estimates defect depths between 0.5 and 1.2 mm for the observed anomalies.

On‑beam validation was performed at the ESRF BM05 beamline using monochromatic X‑rays from 20 to 50 keV. The detector reproduced laboratory energy linearity, maintained the ≈180 eV resolution, and demonstrated stable operation under direct beam illumination with dead‑time remaining below 5 %. The defect‑depth estimation derived from on‑beam data matched the laboratory micro‑beam findings, confirming the model’s applicability in real synchrotron environments.

Overall, XAFS‑DET combines high quantum efficiency (enhanced by germanium’s Z = 32), excellent energy resolution, fast signal processing (rise time ~42 ns), low electronic noise, minimal crosstalk, and robust high‑count‑rate capability (20–250 kcps mm⁻²). These attributes make it a compelling solution for hard‑X‑ray spectroscopy techniques such as X‑ray absorption fine structure (XAFS), X‑ray fluorescence (XRF), and simultaneous multi‑sample analysis at next‑generation synchrotron facilities. Future work will focus on scaling pixel area, optimizing peaking times for even higher count rates, and integrating real‑time defect monitoring algorithms to further enhance performance.


Comments & Academic Discussion

Loading comments...

Leave a Comment