Operation of the Trigger System for the ICARUS Detector at Fermilab

This paper presents the architecture, deployment, and performance metrics, including event rates and recognition efficiency, of the scintillation-light-based trigger system used during the first two physics runs of the ICARUS liquid argon TPC detector at Fermilab.

The Gist

Imagine a giant, ultra-sensitive underwater camera, but instead of water, it's filled with 760 tons of liquid argon (which is like super-cold, invisible ice). This is the ICARUS detector, sitting deep underground at Fermilab in the US. Its job is to catch ghostly particles called neutrinos that zip through the Earth from particle beams.

However, there's a problem: the detector is only buried under a thin layer of concrete. It's like trying to take a photo of a firefly in a stadium while a thousand people are running around shouting and flashing their phone lights. Those "shouts" are cosmic rays (particles from space) hitting the detector constantly. If the detector just recorded everything, it would be overwhelmed by noise.

This paper explains how the ICARUS team built a super-smart "trigger system" (a security guard) to decide exactly when to take a picture, ensuring they catch the neutrinos and ignore the cosmic noise.

Here is the breakdown of how it works, using simple analogies:

1. The "Flash" in the Dark

When a neutrino (or any particle) zips through the liquid argon, it doesn't just leave a trail of electricity; it also makes the argon glow with a tiny, ultra-fast flash of blue light.

• The Analogy: Imagine a dark room. If a ghost walks through, it leaves a faint, glowing trail. The detector has 360 giant eyes (called Photomultiplier Tubes or PMTs) lining the walls to catch these glows.

2. The "Security Guard" Logic (The Trigger)

The detector can't record every single glow because the cosmic rays are too frequent. It needs a rule to say, "Okay, this glow is special, let's record it."

• The Rule: The system looks for a majority vote. It divides the detector into long slices (like cutting a loaf of bread). If a slice sees enough "eyes" (PMTs) light up at the exact same time, the system says, "That's a real event!" and hits the record button.
• The Timing: The system also knows exactly when the "beam" of neutrinos is coming from the accelerator (like knowing when the mail truck arrives). It opens a tiny window of time (a "gate") only when the mail truck is expected. If the glow happens outside that window, it's probably just a cosmic ray, and the system ignores it.

3. The "Two-Step" Dance

The paper describes two main ways the system triggers:

• The "On-Beam" Trigger: This is the main event. The accelerator sends a signal saying, "Neutrinos are coming in 35 milliseconds!" The detector gets ready. When the neutrinos arrive, they hit the argon, create a flash, and if enough PMTs see it, the system records the data.
• The "Off-Beam" Trigger: Sometimes, the detector needs to study the cosmic rays themselves (to learn how to ignore them better). So, when the mail truck isn't there, the system still listens, but it only records if the flash is very bright (a "majority" of 10 or 9 PMTs instead of 5). This filters out the weak cosmic noise and only catches the big, energetic space particles.

4. The "Cosmic Ray Tagging" System

To make sure their "security guard" is working correctly, the team built a fence around the detector called the Cosmic Ray Tagger (CRT).

• The Analogy: Imagine the detector is a house. The CRT is a motion-sensor fence all around the outside. If a cosmic ray (a burglar) tries to sneak in, the fence screams "Burglar!" before the particle even hits the liquid argon.
• By comparing what the fence saw with what the "eyes" inside saw, the scientists could test their trigger system. They asked: "Did our trigger catch the burglar? Did it miss any? Did it catch too many?"

5. The Results: A Perfect Catch

The paper reports on two "runs" (periods of data collection) in 2022 and 2023.

• The Upgrade: In the second run, they improved the "slices" of the detector. Instead of having three big slices, they overlapped them like shingles on a roof. This meant no matter where a particle hit, it was guaranteed to be in a slice.
• The Success: They found that for particles with enough energy (like the neutrinos they are hunting), the system catches almost 100% of the events. Even for lower-energy particles, it catches over 80%.
• The Efficiency: The system is so good that it filters out the "noise" of the universe while keeping the "signal" of the neutrinos. It's like having a camera that only takes a photo when the specific celebrity you are looking for walks into the room, ignoring everyone else.

Why Does This Matter?

Neutrinos are the "ghosts" of physics. They barely interact with anything. To study them, you need a detector that is huge (like ICARUS) and incredibly smart. If the trigger system is too slow, you miss the ghost. If it's too sensitive, you get flooded with fake ghosts (cosmic rays).

This paper proves that the ICARUS team built a perfectly tuned trigger. It's the difference between trying to hear a whisper in a hurricane and having a soundproof room with a microphone that only turns on when the whisper happens. Thanks to this system, scientists can now study these elusive particles to understand the fundamental secrets of the universe.

Technical Summary

1. Problem Statement

The ICARUS-T600 detector, a Liquid Argon Time Projection Chamber (LAr-TPC), operates at a shallow depth at Fermilab to serve as the far detector for the Short Baseline Neutrino (SBN) program. It faces two primary challenges:

• Cosmic Ray Background: The shallow depth exposes the detector to a high flux of cosmic rays. Since the LAr-TPC requires a long drift time (approx. 1 ms) to collect ionization electrons, cosmic rays can overwhelm the detector data during this window, making it difficult to isolate rare neutrino interaction events.
• Trigger Efficiency: The detector relies on detecting scintillation light (VUV photons) from ionizing particles to trigger data acquisition. The system must achieve high efficiency for neutrino interactions (which deposit low energy, often < 2 GeV) while maintaining a manageable data rate and rejecting the overwhelming cosmic background.

2. Methodology

The paper details the architecture, deployment, and performance evaluation of the ICARUS trigger system during its first two physics runs (Run1: June–July 2022; Run2: Dec 2022–June 2023).

A. Trigger System Architecture

• Hardware: The system utilizes 360 8-inch Photomultiplier Tubes (PMTs) coated with tetraphenyl butadiene to detect VUV scintillation light. Signals are digitized by CAEN V1730B boards (14-bit, 500 MS/s).
• Logic Implementation: Trigger logic is implemented on NI PXIe-7820R FPGAs.
  • PMT-Majority Logic: The detector volume is divided into longitudinal windows (6 m long). The system counts the multiplicity of "OR-paired" PMT signals exceeding a threshold (~390 ADC counts, ~13 photoelectrons).
  • Beam Gates: The system synchronizes with Fermilab's Booster (BNB) and Main Injector (NuMI) beams using the White Rabbit (WR) network for sub-nanosecond precision.
    • On-beam: A "Majority" trigger fires if $\ge$ 5 PMT pairs fire within a beam gate (1.6 $\mu$s for BNB, 9.5 $\mu$s for NuMI).
    • Off-beam: A stricter majority ($\ge$ 10 in Run1, $\ge$ 9 in Run2) is used to tag cosmic rays outside beam spills.
    • Minimum Bias: Triggers are generated randomly (without light requirements) to provide unbiased samples for calibration and efficiency validation.
• Synchronization: The system integrates signals from the proton extraction, the PMT system, and the external Cosmic Ray Tagger (CRT) to timestamp events with nanosecond precision.

B. Efficiency Evaluation Method

To measure trigger efficiency without bias:

• Dataset: The authors used a high-statistics sample of cosmic muons collected during "minimum bias" runs (no beam, no light requirement).
• Track Reconstruction: Cosmic muon tracks were reconstructed using the TPC wire signals (Pandora toolkit) and matched with the external CRT system to determine the precise crossing time ($t_0$) and position.
• Software Emulation: A software algorithm emulated the hardware trigger logic (discrimination, pairing, windowing, and counting) on the recorded PMT waveforms. This allowed the team to determine if a specific cosmic event would have triggered the hardware.
• Validation: The emulation was cross-checked against actual hardware triggers to identify and correct for minor configuration mismatches (e.g., leading vs. trailing edge discrimination).

3. Key Contributions

• System Deployment: A comprehensive description of the first physics operation of a complex, multi-component trigger system integrating PMT light detection, accelerator beam timing (White Rabbit), and external CRT vetoes.
• Logic Optimization: The transition from Run1 to Run2 involved adding two overlapping 6 m windows (shifting the logical slices by 3 m) to improve trigger uniformity along the beam direction.
• Efficiency Characterization: The first detailed measurement of the trigger efficiency as a function of deposited energy and spatial position within the LAr-TPC, specifically for low-energy events relevant to the SBN physics goals.
• Calibration Strategy: The demonstration of using minimum-bias cosmic data and software emulation to validate trigger performance without relying on simulated beam data.

4. Results

• Trigger Rates: In Run2, the system achieved a stable total trigger rate of $\approx$ 0.7 Hz, comprising $\approx$ 0.2 Hz (BNB), $\approx$ 0.25 Hz (NuMI), and $\approx$ 0.25 Hz (Minimum Bias).
• Efficiency vs. Energy (Stopping Muons):
  • For the Mj=5 (in-spill) trigger, efficiency saturates at >90% for energies above 300–400 MeV.
  • Efficiency reaches >80% in the 200–300 MeV range.
  • Below 200 MeV, efficiency drops, with Run2 showing a ~10% improvement over Run1 due to the overlapping window logic.
• Spatial Uniformity: The trigger response is highly uniform across the detector volume (East and West modules). A slight reduction (up to 3%) was observed in the East module, attributed to 11 dead PMT channels compared to 3 in the West.
• Out-of-Spill Efficiency: The stricter Mj=9/10 triggers used for cosmic ray rejection in Run2 showed improved uniformity compared to Run1, effectively identifying energetic cosmic rays crossing the detector during the 1 ms drift time.
• Uncertainties: Systematic uncertainties were quantified (e.g., $\approx$ 4% between 200–300 MeV), primarily arising from track reconstruction and the software emulation of hardware logic.

5. Significance

• Physics Readiness: The results confirm that the ICARUS trigger system is fully operational and capable of efficiently capturing neutrino interactions from the BNB and NuMI beams, which is critical for the SBN program's search for sterile neutrinos.
• Low-Energy Sensitivity: The demonstrated efficiency down to 200 MeV ensures the detector can capture the bulk of the expected neutrino interaction spectrum, which is essential for distinguishing signal from background.
• Robust Background Rejection: The system successfully balances high signal efficiency with the ability to reject the high rate of cosmic rays, a critical requirement for shallow-depth detectors.
• Methodological Benchmark: The paper provides a robust framework for validating trigger systems in LAr-TPC experiments using software emulation and unbiased cosmic data, a technique applicable to future large-scale detectors (e.g., DUNE).