Design and characterization of a photosensor system for the RELICS experiment

In this paper, we present the design and characterization of a photosensor system developed for the RELICS experiment. An extended dynamic range base was designed to mitigate photomultiplier tube (PMT) saturation caused by intense cosmic muon backgrounds in the surface-level RELICS detector. The system employs dual readout from the anode and the seventh dynode to extend the linear response range of the PMT. In particular, our characterization and measurements of Hamamatsu R8520-406 PMTs confirm stable operation under positive high-voltage bias, extending the linear response range by more than an order of magnitude. Furthermore, a model of PMT saturation and recovery was developed to evaluate the influence of cosmic muon signals in the RELICS detector. The results demonstrate the system capability to detect coherent elastic neutrino-nucleus scattering signals under surface-level cosmic backgrounds, and suggest the potential to extend the scientific reach of RELICS to MeV-scale interactions.

💡 Research Summary

The paper presents the design, implementation, and comprehensive characterization of a novel photosensor system tailored for the surface‑level REactor neutrino LIquid xenon Coherent Scatter (RELICS) experiment, whose primary scientific goal is the detection of coherent elastic neutrino‑nucleus scattering (CEνNS) from a nearby nuclear reactor. Because RELICS operates at the surface, it is exposed to a high flux of cosmic‑ray muons (∼10 Hz cm⁻²), which deposit of order 1 MeV in the liquid xenon (LXe) volume, producing an intense prompt scintillation (S1) of up to 10⁶ photo‑electrons (PE) within a few tens of nanoseconds and a prolonged ionisation‑induced secondary scintillation (S2) of ∼10 PE ns⁻¹ lasting up to 178 µs. Such large signals saturate the anode output of conventional photomultiplier tubes (PMTs), compromising both amplitude and timing information that are essential for reconstructing low‑energy CEνNS events (0.6–1.4 keV, corresponding to 120–300 PE).

To overcome this limitation, the authors develop an extended‑dynamic‑range base that simultaneously reads out the anode (high‑gain) and the seventh dynode (low‑gain) of Hamamatsu R8520‑406 1‑inch PMTs. The base is powered by a positive high‑voltage of ≈800 V, with a voltage divider that allocates roughly one‑third of the total voltage to the second dynode to maximise photo‑electron collection efficiency. Six parallel 10 nF capacitors (C1–C6) stabilise inter‑dynode voltages, while 100 kΩ high‑impedance resistors suppress back‑flow currents. Damping resistors (50 Ω) and a 680 Ω isolated quenching resistor are incorporated to eliminate ringing and damped oscillations. A decoupling capacitor (C10) extracts the anode signal for conventional high‑gain readout, whereas the seventh dynode provides a low‑gain signal that remains linear over a much larger dynamic range.

The performance of the dual‑readout system is evaluated using a dedicated bench‑test setup. An LED driven by a high‑bandwidth pulse generator produces calibrated light pulses that are monitored by a second PMT equipped with a polariser attenuator. The test PMT’s anode signal is attenuated by 9 dB before digitisation, while the dynode signal is recorded directly. By integrating pulse areas and converting to PE ns⁻¹, the authors find an anode‑to‑dynode gain ratio of 113.3 ± 0.3 in the linear regime. The anode response follows a natural‑logarithmic saturation curve above ∼30 PE ns⁻¹ and saturates completely near 40 PE ns⁻¹. In contrast, the dynode channel remains linear up to ≈1000 PE ns⁻¹, comfortably exceeding the 500 PE ns⁻¹ requirement for muon S1 detection. Beyond this point the dynode also begins to deviate from linearity, but the deviation is modest and can be corrected using the measured gain‑ratio drift (which drops to ≈43.4 at the highest intensities).

A phenomenological model of PMT saturation and recovery is constructed to describe the time‑dependent gain suppression caused by the large charge influx during a muon event. The model incorporates a recovery time constant of 10–100 µs, consistent with the observed re‑establishment of the anode gain after a muon‑induced S2 pulse. By embedding this model into a full RELICS Monte‑Carlo simulation, the authors demonstrate that the dynode‑derived low‑gain waveform can be used to reconstruct the true muon S1 amplitude, tag the muon trajectory, and consequently suppress delayed‑electron backgrounds that would otherwise overwhelm the CEνNS signal region by up to four orders of magnitude.

The dual‑readout approach therefore extends the usable dynamic range of each PMT by more than an order of magnitude, while preserving the excellent timing resolution required for TPC event reconstruction. This enables RELICS to operate at the surface without sacrificing sensitivity to low‑energy CEνNS events, opening the possibility of probing MeV‑scale neutrino interactions, testing non‑standard neutrino physics, and providing valuable input for future surface‑based dark‑matter and neutrino experiments. The authors suggest further developments such as adding additional dynode readouts (e.g., dynodes 5 and 6) for multi‑gain channels, integrating ASIC‑based low‑power front‑ends, and scaling the design to larger PMT arrays.

In summary, the paper delivers a practical, experimentally validated solution to PMT saturation in high‑background environments, demonstrates its impact on the physics reach of the RELICS experiment, and outlines a clear path toward broader adoption in next‑generation low‑threshold neutrino and dark‑matter detectors.