Radon-induced backgrounds in the NEXT-100 experiment

The NEXT-100 experiment's first low-background run quantified radon-induced backgrounds, revealing a low internal $^{222}$Rn activity and demonstrating that topological event selection reduces the resulting background index in the neutrinoless double-beta decay region of interest to one order of magnitude below the total radiogenic expectation, confirming the effectiveness of the facility's radon abatement system.

The Gist

Imagine you are trying to hear a single, incredibly faint whisper in a room that is already full of noise. That is essentially what the NEXT-100 experiment is trying to do.

Scientists are hunting for a rare event called neutrinoless double beta decay. If they find it, it proves that neutrinos are their own antiparticles (a "Majorana" particle), which would rewrite our understanding of the universe and explain why there is more matter than antimatter.

To hear this "whisper," they built a giant, ultra-sensitive microphone (the detector) filled with heavy gas (Xenon). But there's a problem: Radon.

Radon is a radioactive gas that seeps out of rocks, concrete, and even the materials used to build the detector itself. It's like a noisy neighbor who constantly bangs on the walls, making it impossible to hear the whisper. This paper is a report card on how well the NEXT-100 team managed to silence this noisy neighbor.

Here is the breakdown of their findings, using simple analogies:

1. The Setup: A High-Tech Fish Tank

The detector is a massive, pressurized tank (like a deep-sea submarine) filled with Xenon gas. Inside, they use electric fields to catch tiny flashes of light created when particles bump into the gas atoms.

• The Goal: Catch the specific "whisper" of the double beta decay.
• The Enemy: Radon gas, which decays and creates its own flashes of light (background noise) that look suspiciously like the signal they are looking for.

2. The Two Types of Noise

The team realized the noise comes from two places, so they tackled them separately:

A. The "Internal" Noise (The Leaky Faucet)

Some radon comes from inside the machine itself, leaking out of the pipes and metal parts (outgassing).

• The Experiment: They turned up the "leak" by using a cold filter (which lets radon escape) to see how much noise it made. Then, they switched to a hot filter (which traps the radon) to see the "natural" leak.
• The Result: They found that the radon inside the tank was very low, but not zero.
• The "Plate-Out" Effect: Here is a cool physics trick: When radon decays, it shoots out tiny charged particles (like static electricity). These particles get stuck on the metal walls of the detector (specifically the cathode), like dust sticking to a balloon.
• The Good News: Because the "dust" (radioactive atoms) sticks to the walls, the actual noise they create inside the gas is mostly blocked. The team calculated that even with this internal leak, the noise is 10 times lower than the total noise budget they can afford. They are safe!

B. The "External" Noise (The Noisy Room)

The detector sits in a huge underground hall (LSC). The air in that hall has radon in it, just like the air in your basement might.

• The Problem: If the air outside the detector is noisy, those radioactivity particles can sneak in through the walls or create noise that mimics the signal.
• The Solution: The lab has a special Radon Abatement System (RAS). Think of this as a giant, high-tech air purifier that blows "clean air" (air with almost zero radon) into the room surrounding the detector.
• The Test: The team ran the experiment twice:
  1. RAS OFF: The room had normal air (noisy).
  2. RAS ON: The room had clean air (quiet).
• The Result: When the RAS was on, the noise from the outside air disappeared completely. The detector became "virtually radon-free." It's like putting on noise-canceling headphones; the outside world is still noisy, but inside your head, it's silent.

3. The "Topological" Filter (The Smart Camera)

Even if some noise gets through, the NEXT detector is smart. It doesn't just count flashes of light; it takes a 3D video of the event.

• The Analogy: Imagine a burglar alarm. A regular alarm just goes off if anyone touches the glass. A smart camera alarm looks at who touched it.
• How it works: The signal they want (double beta decay) looks like two electrons moving together in a specific shape (like a dumbbell). The noise from radon usually looks like a single electron or a messy spray.
• The Result: By using this "smart camera" to only count events that look like the specific "dumbbell" shape, they can filter out 99.9% of the remaining radon noise.

The Bottom Line

This paper is a victory lap for the NEXT-100 team. They proved that:

1. Internal leaks are small and manageable.
2. External air is effectively cleaned by their air purifier system.
3. Smart filtering can remove the rest.

Conclusion: The detector is now "quiet" enough to start the real search for the neutrinoless double beta decay. They have successfully tuned their instrument so that the "whisper" of the universe might finally be heard above the noise.

In short: They built a super-quiet room, installed a top-tier air filter, and gave the room a pair of smart glasses to ignore any remaining noise. The search for the universe's biggest secret can now begin in earnest.

Technical Summary

1. Problem Statement

The NEXT-100 experiment aims to detect neutrinoless double beta decay ($0\nu\beta\beta$) in Xenon-136 ($^{136}$Xe). A critical challenge in this search is distinguishing the rare signal from background radiation. Radon ($^{222}$Rn and $^{220}$Rn) is a major background source due to:

• High-energy gammas: The decay chain of $^{222}$Rn produces $^{214}$Bi, which emits a gamma ray at 2.448 MeV, very close to the $Q$-value of the $^{136}$Xe decay (2.458 MeV).
• Diffusivity: Radon is a noble gas that emanates from detector materials (internal) and the surrounding laboratory air (airborne), creating variable background levels.
• Plate-out: Radon progeny (like $^{218}$Po and $^{214}$Pb) are often positively charged and plate out on detector surfaces (specifically the cathode), creating localized sources of $^{214}$Bi.

The paper addresses the need to quantify and mitigate both internal radon (emanating from detector components/gas system) and airborne radon (from the laboratory environment) to ensure the experiment meets its background budget goals.

2. Methodology

The study utilized data from the first low-pressure physics run of NEXT-100 (operating at ~3.95 bar with $^{136}$Xe-depleted xenon) conducted in 2025. The methodology involved two dedicated data-taking campaigns:

A. Internal Radon Characterization (Alpha Runs)

• Setup: Two periods were defined:
  • High Activity (A1): Xenon gas recirculated through a cold getter, which releases significant amounts of $^{222}$Rn.
  • Low Activity (A2): Xenon gas recirculated through a hot getter, which removes radon, revealing the intrinsic background of the detector.
• Trigger: Dedicated triggers for alpha decays (S2 signals with specific charge and time width).
• Analysis:
  • Reconstruction of alpha events using S1 (scintillation) and S2 (ionization) signals to determine 3D position.
  • Spectroscopic analysis of the energy spectrum to distinguish between $^{222}$Rn (5590 keV), $^{218}$Po (6115 keV), and $^{214}$Po (7834 keV).
  • Calculation of the fraction of radon progeny that plate out on the cathode versus those decaying in the gas volume.

B. Airborne Radon Characterization (Electron Runs)

• Setup: Two periods with standard $0\nu\beta\beta$ trigger conditions (electron-like events):
  • E1: Radon Abatement System (RAS) OFF (allowing natural lab radon levels).
  • E2: RAS ON (delivering radon-free air to the lead castle).
• Analysis:
  • Reconstruction of electron events and application of fiducial cuts (removing events near detector boundaries).
  • Correlation of the fiducial event rate with the measured radon concentration in the laboratory (HALL A) using an AlphaGuard detector.

C. Background Index Estimation

• Monte Carlo (MC) Simulation: Used the Nexus (GEANT4-based) software to simulate $^{214}$Bi decays on the cathode.
• Selection Cuts: Applied standard NEXT selection criteria:
  1. Fiducialization: Events fully contained within the active volume.
  2. Topological Selection: Requiring a single "double-electron-like" track (characteristic of $0\nu\beta\beta$) and rejecting multi-cluster events.

3. Key Contributions

• First In-Situ Measurement: Provides the first measurement of internal radon activity and $^{214}$Bi plate-out rates for the NEXT-100 detector.
• Quantification of Plate-out: Demonstrates that ~93% of radon-induced $^{214}$Bi decays occur on the cathode surface rather than in the gas volume, a crucial factor for background modeling.
• Validation of Mitigation Systems: Empirically proves the effectiveness of the LSC's Radon Abatement System (RAS) in creating a "virtually radon-free" environment for the detector.
• Pressure Dependence Insight: Highlights a potential correlation between operating pressure and radon emanation rates, suggesting lower backgrounds at the nominal 15 bar pressure.

4. Key Results

Internal Radon Activity

• Measured Activity: The intrinsic internal $^{222}$Rn activity (with hot getter) is $(0.95 \pm 0.04_{stat} \pm 0.09_{sys})$ Bq/m$^3$.
• Half-life: The decay rate during the A2 period yielded a half-life of $3.77 \pm 0.06$ days, consistent with $^{222}$Rn.
• No $^{220}$Rn: No traces of Thoron ($^{220}$Rn) were observed.
• Cathode Contamination: The rate of $^{214}$Bi decays on the cathode is $(0.97 \pm 0.05_{stat} \pm 0.10_{sys})$ Hz for energies > 400 keV.

Background Index (BI) in $0\nu\beta\beta$ ROI

The Background Index is defined in the region of interest (2.4–2.5 MeV) in units of counts/(keV·kg·yr).

• After Fiducial Selection: $(7.3 \pm 1.5_{stat} \pm 0.8_{sys}) \times 10^{-4}$.
• After Topological Selection (Double-electron track): Reduced to $\sim 4 \times 10^{-5}$ counts/(keV·kg·yr).
• Comparison: This is one order of magnitude below the total expected radiogenic background budget for NEXT-100 ($4 \times 10^{-4}$).

Airborne Radon Impact

• Correlation: Without the RAS (E1), the fiducial background rate correlated linearly with lab radon levels.
• RAS Effectiveness: With the RAS on (E2), the background rate dropped to $(2.88 \pm 0.04)$ mHz and showed no correlation with lab radon fluctuations.
• Conclusion: The detector operates in a virtually radon-free environment when the RAS is active.

5. Significance

• Feasibility Confirmed: The study confirms that radon-induced backgrounds will not be a limiting factor for the NEXT-100 $0\nu\beta\beta$ search. The topological capabilities of the Time Projection Chamber (TPC) effectively reject the remaining $^{214}$Bi background.
• Operational Optimization: The results suggest that operating at the nominal high pressure (15 bar) may further reduce internal radon emanation due to pressure gradients, potentially lowering the background index even further.
• System Validation: The successful operation of the Radon Abatement System validates the infrastructure design for future ton-scale detectors (NEXT-Bolero/ton-scale), ensuring that environmental radon can be effectively managed.
• Benchmarking: The measured internal activity is higher than in the previous NEXT-White detector, but the authors attribute this to the lower operating pressure (3.95 bar vs 10.1 bar), providing a predictive model for future high-pressure runs.

In summary, the paper establishes that the NEXT-100 detector, combined with its radon mitigation systems and topological event reconstruction, is capable of achieving the ultra-low background levels required for a competitive search for neutrinoless double beta decay.