Six cryogenic target detectors were characterized in commissioning tests for the NUCLEUS Technical Run, with a mean baseline energy resolution of 2.79 eV using a two-dimensional optimum filter. The baseline-resolution figure is the detectors' noise benchmark, expressed in energy units, and the accompanying ±0.60 eV is the standard deviation across the six detectors. After dedicated working-point optimization, detector D reached 2.16 ± 0.02 eV with the same filter and a 55Fe X-ray calibration; the ±0.02 eV is a statistical uncertainty.
The work was carried out at the Technical University of Munich as a pre-deployment commissioning study. Six detectors were characterized, four were prepared for deployment at Chooz and two were retained as backups. The paper therefore reports a test of the detector system before its reactor-site run, rather than reactor-site data.
A compact target with paired sensors
The module held four CaWO4 crystals, each measuring 5 by 5 by 11.5 millimetres, for an approximately 7-gram total target mass. Each crystal used a double-TES readout, meaning its signal was recorded through two sensor channels.
The commissioning included measurements of detector pulses, noise, calibration and baseline resolution, along with estimates of radioactive background from the setup. The target detectors were also operated at the same time as the Cryogenic Outer Veto, or COV, to check the systems' compatibility. The combined operation showed no observable cross-talk.
Measurements were processed with the cait Python framework, using the cited NUCLEUS approach for event triggering, selection and pulse reconstruction. Calibration used X-ray lines from 55Fe at 5.89 keV and 6.49 keV, while a truncated template fit handled events outside the detectors' linear response range.
The result across six detectors
The analysis compared two-dimensional optimum filters with individual one-dimensional optimum filters. The 2D optimum filter improved baseline resolution by approximately 14% on average.
The stability test found that all six detectors operated stably for up to 21 days, with stability at the 10% level or better. Alongside the absence of observable cross-talk with the COV, this led the authors to state that the demonstrated performance and stability validate the integrated configuration for deployment at Chooz.
Backgrounds and calibration still need answers
Background estimates came from Geant4-based simulations and material-screening measurements. Between 0 and 10 keV, radioactivity from setup components was estimated at 27.4 ± 0.4 counts per kilogram per kiloelectronvolt per day. That was approximately 20% of a predicted total background of 137.2 ± 1.5 counts per kilogram per kiloelectronvolt per day.
Calibration added a separate qualification. For detector D, the 55Fe-based calibration produced a baseline resolution 26.5% lower than the LED-based calibration, a difference identified as a calibration-related systematic uncertainty. Neither calibration method directly reproduces the nuclear recoils sought in the experiment, so the study does not by itself validate the absolute energy scale for those signals.
The reactor run will test whether the detectors retain their laboratory resolution and stability at Chooz. It will also show the actual background composition and low-energy-excess contribution, as well as how much working-point optimization improves the detectors selected for deployment. The study flags the low-energy excess as a possible limit on sensitivity to the Standard Model prediction for coherent elastic neutrino-nucleus scattering during the Technical Run.
A result before the reactor-site test
Taken together, these are laboratory commissioning results for the target-detector hardware before deployment. The authors interpret the findings as validation of the integrated configuration for Chooz, but the study does not report a reactor measurement of coherent elastic neutrino-nucleus scattering or establish performance under Chooz operating conditions.
The document is an arXiv version 1 preprint dated 26 August 2026. The work was financed by CEA, INFN and ÖAW, with partial support from TU Munich and MPI für Physik. Supporting data are not publicly available but can be obtained from the authors upon reasonable request.
Paper data and sources
Original title: Development and Commissioning of the Cryogenic Target Detectors for the Technical Run of the NUCLEUS Experiment
Authors: N. Schermer, H. Abele, G. Angloher et al.
Journal/Repository: arXiv
Status: Preprint, not yet peer-reviewed
First online: 2026-08-26
DOI: Not available
Original paper · Full text