Electron identification and hadron discrimination using Cherenkov radiation in air and SiPMs

This paper demonstrates a method for identifying electrons and discriminating hadrons using Cherenkov radiation in air detected by Silicon PhotoMultipliers (SiPMs), validated by CERN test beam data and Monte Carlo simulations to show high performance across a wide momentum range.

The Gist

Imagine you are standing in a dark room, and you want to know what kind of invisible runner just sprinted past you. Is it a super-fast, lightweight sprinter (an electron), or is it a heavy, lumbering jogger (a pion or proton)?

Usually, telling them apart requires complex, expensive equipment. But this paper describes a clever, low-tech trick using a special kind of "digital eye" called a SiPM (Silicon Photomultiplier) and a simple stretch of air.

Here is the story of how they did it, explained simply.

1. The "Digital Eye" (The SiPM)

Think of the SiPM not as a camera, but as a giant mosaic floor made of millions of tiny, sensitive tiles. Each tile is a microscopic sensor (called a SPAD).

• If a single photon (a particle of light) hits a tile, that tile "pings" and sends a signal.
• If no light hits it, it stays silent.
• The researchers used these sensors without their usual protective plastic cover. They wanted the sensors to be as "naked" and sensitive as possible.

2. The Magic of the "Air Gap"

When a charged particle (like an electron or a proton) zooms through the air, it can sometimes create a faint blue flash of light called Cherenkov radiation.

• The Analogy: Imagine a boat speeding through water. If it goes faster than the waves can move, it creates a "bow wave" or a sonic boom. Similarly, if a particle moves faster than light can travel through that specific medium (like air), it creates a shockwave of light.
• The Catch: Heavy particles (protons, pions) are like slow boats. They need to be going incredibly fast (very high energy) to make this light in air. Light particles (electrons) are like speedboats; they make this light even at relatively low speeds.

3. The Experiment: Counting the "Pings"

The team set up a 7-centimeter gap of air right in front of their naked SiPM sensor. They shot a beam of mixed particles (electrons, pions, and protons) through it.

• The Heavy Joggers (Protons/Pions): At the speeds used in the experiment, these particles were too slow to create a light shockwave in the air. When they hit the sensor, they mostly just triggered one single tile (or zero, due to the sensor's gaps).
• The Speedboats (Electrons): These particles were fast enough to create a cone of light in the air gap. This light spread out and hit many tiles on the sensor at once.

The Result:

• One "Ping" (or zero): Likely a heavy particle (Pion/Proton).
• Many "Pings" (a cluster): Likely an electron.

By simply counting how many tiles lit up, they could tell the difference between the two types of particles with about 85% accuracy for pions and 57% efficiency for electrons, even with a very basic setup.

4. The Simulation: The "Virtual Lab"

To make sure this wasn't just a lucky guess, they built a computer model (a "Toy Monte Carlo"). Imagine a video game where they simulate millions of particles flying through air and hitting the sensor.

• The computer predicted exactly what they saw in the real world: electrons create a "fuzzy cloud" of light hits, while heavy particles create a single sharp dot.
• This confirmed that the physics was sound and the method works.

5. The Future: Tuning the System

The researchers then asked, "Can we make this even better?" They ran simulations to see what happens if:

• We use a bigger sensor: Like using a wider net to catch more fish. A bigger sensor catches more of the light cone, making the "electron signal" even louder.
• We use a longer air gap: Giving the light more room to spread out.
• We use a different gas: They tested using CO2 instead of air. CO2 is slightly denser, meaning even slower particles can create the light shockwave. This could allow them to identify particles at even lower speeds.

Why Does This Matter?

This is a "low-cost, high-smarts" solution.

• No Scintillators: Usually, to see these particles, you need to sandwich them between heavy, expensive crystals (scintillators) that glow when hit. This method uses just the air and the sensor.
• Dual Purpose: The same sensor technology can also measure time incredibly precisely (down to 20 picoseconds!). So, one device could tell you what the particle is (via light counting) and when it arrived (via timing).

The Bottom Line

The paper proves that you don't always need a massive, billion-dollar machine to identify particles. Sometimes, you just need a naked sensor, a little bit of air, and the ability to count how many "pings" you hear. It turns a simple sensor into a smart particle detective.

Note: The paper also includes a touching tribute to a colleague, Andrea Alici, who passed away just before publication, highlighting that this discovery was a result of his ingenuity and hard work.

Technical Summary

1. Problem Statement

The paper addresses the challenge of identifying electrons and discriminating them from hadrons (pions, kaons, protons) in particle physics experiments. Traditional Cherenkov detectors often require complex radiator geometries and large volumes.

• The Specific Gap: While Silicon PhotoMultipliers (SiPMs) are known for their excellent timing capabilities when equipped with a protective layer (which generates Cherenkov light), their behavior without a protective layer was less explored for particle identification.
• The Hypothesis: The authors propose that when charged particles traverse a volume of air immediately before hitting a SiPM (without a protective layer), electrons (due to their low mass and high $\beta$) will emit Cherenkov radiation in the air. This creates an excess of photons compared to heavier hadrons (pions, protons) which remain below the Cherenkov threshold at lower momenta. This excess should result in a higher number of fired Single Photon Avalanche Diodes (SPADs) within the SiPM, allowing for particle identification via simple "photon counting."

2. Methodology

Experimental Setup:

• Location: CERN PS T10 test beam.
• Beam Momentum: Primary data collected at 1.5 GeV/c (where the beam contains ~60% electrons, 28% pions, and 12% protons). A secondary sample at 10 GeV/c was used for simulation validation.
• Detectors:
  • DUT (Device Under Test): NUV-HD-LFv2 SiPMs (Fondazione Bruno Kessler) with an active area of $3.20 \times 3.12 \text{ mm}^2$ and 6200 SPADs. Crucially, these sensors were used without their factory-supplied protection layer.
  • Reference: Two LGAD (Low Gain Avalanche Detector) sensors placed at the start and end of a 7 cm telescope to provide precise timing and trigger signals.
• Geometry: A 7 cm air gap existed between the final LGAD and the SiPM. A Faraday cage with a specific aperture allowed Cherenkov photons generated in this air gap to reach the SiPM.
• Signal Processing: Signals were amplified (40 dB gain) and recorded via a digital oscilloscope. The analysis focused on the amplitude of the SiPM signal, where each peak corresponds to a fired SPAD.

Analysis Strategy:

1. Particle Selection: LGAD timing information was used to separate protons from the electron/pion mixture. Protons were selected based on Time-of-Flight (ToF) cuts.
2. Background Subtraction: Since pions and protons do not produce Cherenkov radiation in air at 1.5 GeV/c, the proton amplitude distribution (peaked at 1 SPAD due to noise/crosstalk) was used as a template for the pion contribution.
3. Electron Extraction: The electron signal was isolated by subtracting the normalized pion/proton distribution from the total electron/pion candidate distribution.
4. Simulation: A "toy" Monte Carlo (MC) simulation was developed based on the Frank-Tamm formula to calculate Cherenkov photon yield in air, incorporating SiPM parameters (Photon Detection Efficiency, fill factor, crosstalk) and electronic noise.

3. Key Contributions

• Novel Identification Mechanism: Demonstrated that a bare SiPM (no protection layer) can act as a threshold Cherenkov detector by utilizing the air gap as the radiator.
• Photon Counting Approach: Established that particle identification can be achieved simply by counting the number of fired SPADs ($N_{SPAD}$), leveraging the binary nature of SPAD response.
• Optimization Study: Used MC simulations to optimize detector parameters, specifically:
  • SiPM Size: Comparing $3.2 \times 3.12 \text{ mm}^2$ vs. a hypothetical $6 \times 6 \text{ mm}^2$ sensor.
  • Radiator Length: Determining the optimal air gap length (15 cm).
  • Gas Medium: Evaluating the use of $CO_2$ vs. Air to lower the momentum threshold.
  • Technology: Utilizing newer SiPM technology (NUV-HD-MT) with deep trench isolation to suppress optical crosstalk, thereby lowering the detection threshold.

4. Results

Experimental Data (1.5 GeV/c):

• Proton/Pion Spectrum: Peaked at 1 SPAD (amplitude ~31 mV), consistent with the sensor's fill factor and electronic noise.
• Electron Spectrum: Showed a distinct shift toward higher SPAD counts (2, 3, or more) due to Cherenkov photons generated in the 7 cm air gap.
• Performance: By setting a threshold of $\ge 2$ SPADs:
  • Hadron Rejection: $\approx 85\%$
  • Electron Efficiency: $\approx 57\%$
  • Note: The authors acknowledge these are conservative figures due to the low overvoltage (2 V) used to prevent oscilloscope saturation.

Simulation and Optimization:

• Validation: The MC simulation qualitatively reproduced the experimental amplitude distributions for protons, pions, and electrons, confirming the physical mechanism.
• Optimized Performance (6x6 mm² SiPM, 15 cm Air, 6 V Overvoltage):
  • Momentum Range: Effective electron identification from 0.05 GeV/c to 6 GeV/c.
  • Efficiency: Electron efficiency $> 85\%$.
  • Rejection: Pion rejection $> 95\%$ (up to 96% with $CO_2$ radiator).
  • High Momentum: The method can also discriminate between Kaons and Protons at very high momenta (21–40 GeV/c) by analyzing the tail of the SPAD distribution.

5. Significance and Future Outlook

• Dual-Purpose Detectors: The paper suggests a hybrid detector configuration: a SiPM without a protection layer (for PID via Cherenkov counting) followed immediately by a SiPM with a protection layer (for high-precision Time-of-Flight). This could cover a wide momentum range for both identification and timing.
• Simplicity and Cost: The method requires no complex radiator structures (like gas radiators with mirrors) or scintillators, utilizing only the air gap and the sensor itself. This is highly attractive for space applications and compact experiments.
• Limitations: The primary constraint identified is the radiation hardness of SiPMs, particularly in harsh environments like the High-Luminosity LHC (HL-LHC), which requires further investigation.

In conclusion, the study validates that Cherenkov radiation in air, detected by bare SiPMs via SPAD counting, offers a robust, simple, and effective method for electron identification and hadron rejection across a broad momentum spectrum.