Design and characterization of a photosensor system for the RELICS experiment

This paper presents the design and characterization of a dual-readout photosensor system using Hamamatsu R8520-406 PMTs that successfully mitigates cosmic muon saturation to enable the RELICS experiment to detect coherent elastic neutrino-nucleus scattering signals at the surface level.

The Gist

The "Super-Sensor" for a Surface-Level Neutrino Hunt

Imagine you are trying to listen to a tiny, whispering ghost (a neutrino) in a room that is also being bombarded by a marching band playing at full volume (cosmic rays). That is the challenge facing the RELICS experiment.

RELICS is a massive, high-tech tank filled with liquid xenon, sitting on the surface of the Earth (not deep underground like most dark matter detectors). Its goal is to catch a very rare, faint signal called Coherent Elastic Neutrino-Nucleus Scattering (CEνNS). This signal is like a gentle tap on a drum.

However, because the detector is on the surface, it is constantly hit by cosmic muons—high-energy particles raining down from space. When a muon hits the tank, it's not a gentle tap; it's like a sledgehammer smashing the drum. This creates a blinding flash of light that overwhelms the sensors, causing them to "saturate" (go blind) and miss the faint whispers that come right after.

This paper describes how the team built a special "Super-Sensor" system to solve this problem. Here is how they did it, using simple analogies:

1. The Problem: The "Blind Spot"

The sensors in RELICS are called Photomultiplier Tubes (PMTs). Think of them as super-sensitive microphones.

• The Issue: When a cosmic muon hits, it creates a signal so loud that the microphone's diaphragm gets stuck. It can't move anymore. Even after the loud noise stops, the microphone is "recovering" and is too distorted to hear the quiet ghost-neutrino signals that might follow.
• The Consequence: Without a fix, the detector would be useless for finding neutrinos because the muon noise would drown everything out.

2. The Solution: The "Dual-Channel" Microphone

The team designed a special electronic base for these sensors that acts like a dual-channel microphone system:

• Channel A (The Anode): This is the standard, super-sensitive channel. It's great for hearing the quiet whispers (low-energy neutrinos), but it gets blown out by the loud bangs (muons).
• Channel B (The Dynode): This is the new, "attenuated" channel. Imagine putting a heavy filter or a volume knob on the microphone that turns the loud noise down by a factor of 100. This channel isn't sensitive enough to hear the tiny whispers, but it can handle the sledgehammer hits without getting stuck.

By reading from both channels at the same time, the system gets the best of both worlds:

• When a muon hits, the Dynode channel records the loud event clearly, so they know exactly when and where it happened.
• Because the Dynode channel handles the loud noise, the Anode channel doesn't get as overwhelmed, allowing it to recover faster and still hear the faint neutrino signals.

3. The "Recovery" Test

The team wanted to know: If a muon hits, how long does it take for the sensor to wake up and hear the next whisper?

They simulated this in the lab by flashing a bright light (the muon) and then a dim light (the neutrino) a split second later.

• The Result: They found that with their new design, the sensor recovers incredibly fast. Even after a massive "bang," the sensor is only about 5% "dazed" for the next few microseconds. By the time 1 millisecond passes, it's 100% back to normal.
• The Analogy: It's like a boxer getting hit by a heavy punch. In the old design, the boxer would be knocked out for a long time. In the new design, the boxer stumbles for a split second but is ready to catch the next punch immediately.

4. Why This Matters

This isn't just about fixing one experiment.

• For RELICS: It means they can finally hunt for neutrinos on the surface of the Earth, which is much cheaper and easier than digging miles underground. They can also use the muon data to map out the paths of these cosmic particles, helping them filter out other background noise.
• For the Future: This "Dual-Channel" design is a blueprint for future experiments. Whether they are looking for dark matter, studying nuclear reactors, or hunting for exotic particles like axions, this system allows scientists to listen to the quietest whispers in a room full of screaming crowds.

The Bottom Line

The RELICS team built a clever electronic "volume control" that lets their sensors listen to both the loudest explosions and the quietest whispers simultaneously. This breakthrough allows them to study neutrinos right on the Earth's surface, opening a new window into the universe without needing a massive underground bunker.

Technical Summary

1. Problem Statement

The RELICS (REactor neutrino LIquid xenon Coherent Scattering) experiment aims to detect Coherent Elastic Neutrino-Nucleus Scattering (CE$\nu$NS) from reactor neutrinos using a dual-phase liquid xenon (LXe) Time Projection Chamber (TPC) located at the surface level of the Sanmen Nuclear Power Company.

• The Challenge: Unlike deep-underground dark matter experiments, surface-level detectors are exposed to a high flux of cosmic-ray muons ($\sim$1 event cm$^{-2}$ min$^{-1}$).
• The Consequence: A single cosmic muon traversing the TPC deposits $\sim$120 MeV of energy, generating massive scintillation signals:
  • S1 (Prompt Scintillation): $\sim$480 PE ns$^{-1}$ per PMT.
  • S2 (Delayed Ionization): $\sim$10 PE ns$^{-1}$ lasting up to 178 $\mu$s.
• The Bottleneck: Standard Photomultiplier Tube (PMT) anodes (Hamamatsu R8520-406) saturate at approximately 40 PE ns$^{-1}$. The muon signals exceed this limit by more than an order of magnitude, causing:
  1. Signal Saturation: Loss of waveform fidelity and energy information for muon events.
  2. Recovery Time: Distortion of subsequent low-energy signals (like CE$\nu$NS) due to PMT recharge time.
  3. Background Confusion: Inability to reconstruct muon tracks, which is necessary to veto delayed electron backgrounds that mimic CE$\nu$NS signals.

2. Methodology

To address the saturation issue, the authors designed and characterized a custom Extended Dynamic Range (EDR) Base for the PMTs.

• Dual-Readout Architecture:
  • Anode Readout: Maintains high gain for low-energy CE$\nu$NS detection (sub-keV).
  • Dynode Readout: A low-gain signal is extracted from the 7th dynode (Dy7). Since the gain up to the 7th dynode is significantly lower than the total anode gain, the signal remains linear even under intense muon illumination.
• Circuit Design:
  • Positive Bias: The PMTs operate with a positive high-voltage bias ($\sim$800 V) on the anode, keeping the photocathode and shell grounded to prevent "gas events" (spurious discharges) between the PMT array and the TPC anode.
  • Stabilization: Six parallel capacitors (10 nF) are connected to ground to stabilize inter-dynode voltages and mitigate space-charge effects.
  • Decoupling: A blocking capacitor decouples the AC signal from the high-voltage line for the dynode output.
• Characterization Setup:
  • Bench tests were conducted using an LED source with a known pulse width and intensity.
  • A reference PMT with a light attenuator monitored the true light intensity to calibrate the test PMT's response.
  • Measurements covered incident light intensities from low levels up to $\sim$13,000 PE ns$^{-1}$.

3. Key Contributions

1. Hardware Design: Development of a custom PMT base circuit enabling simultaneous high-gain (anode) and low-gain (7th dynode) readout.
2. Saturation Modeling: Development of a mathematical model describing PMT saturation and recovery, parameterized by incident light intensity ($S_{in}$) and time delay ($\Delta t$).
3. Recovery Characterization: Quantitative measurement of the time required for PMTs to recover from saturation and the resulting distortion of subsequent signals.
4. Fidelity Analysis: Demonstration that the dynode readout preserves waveform fidelity for long-duration muon S2 signals, which is critical for muon track reconstruction.

4. Key Results

• Extended Dynamic Range:
  • The Anode saturates at $\sim$40 PE ns$^{-1}$.
  • The 7th Dynode extends the linear response range to >1000 PE ns$^{-1}$, covering the vast majority of muon S1 signals (which peak around 480 PE ns$^{-1}$).
  • The gain ratio between the anode and the 7th dynode was measured to be 113.3, consistent with theoretical calculations.
• Recovery Time & Distortion:
  • After a high-intensity muon S1 pulse, the PMT requires time to recharge.
  • At the most extreme tested intensity ($\sim$13,100 PE ns$^{-1}$), the saturation correction factor ($\Gamma$) drops to $\sim$70% immediately after the pulse but recovers to 95% within 2 $\mu$s and fully recovers within 1 ms.
  • For typical muon signals in RELICS (<1000 PE ns$^{-1}$), the distortion of subsequent low-energy signals is <5% after a 2 $\mu$s delay.
• Waveform Fidelity (S2):
  • The 7th dynode readout maintains high waveform fidelity for 68% of simulated muon S2 signals.
  • In contrast, the anode readout maintains fidelity for only 12% of these signals.
  • Extrapolation suggests that using the 6th dynode would cover 94% of S2 signals, and the 5th dynode would cover 98%.
• Low-Energy Performance:
  • The addition of the dynode readout circuitry did not compromise the Single Photoelectron (SPE) resolution or gain of the anode.
  • SPE resolution remained at 54.9%, and the peak-to-valley ratio was 1.84, confirming the system is suitable for sub-keV CE$\nu$NS detection.

5. Significance

• Enabling Surface-Level CE$\nu$NS Detection: This system allows the RELICS experiment to operate at the surface, overcoming the primary limitation of cosmic muon backgrounds. It enables the simultaneous detection of rare, low-energy neutrino events and high-energy muon events without signal loss.
• Background Suppression: By preserving the S2 waveform, the system enables precise reconstruction of muon tracks. This is crucial for vetoing delayed electron backgrounds, which are a dominant background source for CE$\nu$NS searches.
• Scalability and Future Applications: The dual-readout architecture provides a practical solution for future large-scale LXe detectors (e.g., for neutrinoless double beta decay or reactor axion searches) where balancing high sensitivity for low-energy events with linearity for high-energy events is a critical challenge.
• MeV-Scale Physics: The system extends the scientific reach of RELICS to MeV-scale interactions, such as reactor axion searches, by providing unsaturated readout capabilities for high-energy deposits.

In conclusion, the RELICS collaboration has successfully designed and validated a photosensor system that effectively mitigates PMT saturation, ensuring the experiment can achieve its physics goals in a challenging surface-level environment.