Evaluation of PID Performance at CEPC and Optimization with Combined dN/dx and Time-of-Flight Data

This paper demonstrates that combining ionization measurements from a Time Projection Chamber with time-of-flight data from silicon-based inner and outer trackers significantly enhances charged-hadron identification performance across a broad momentum range for the Circular Electron-Positron Collider.

The Gist

The Great Particle Identity Crisis: How CEPC Learns to Tell Friends Apart

Imagine you are at a massive, chaotic party (the CEPC, or Circular Electron–Positron Collider). Millions of guests are crashing into each other, creating a whirlwind of new particles. Your job is to act as the bouncer and the detective: you need to figure out exactly who each guest is.

The main suspects are Pions (the most common, boring party-goers), Kaons (the slightly more exotic guests), and Protons (the heavy hitters). The problem? They all look exactly the same when they are moving fast. If you just look at them, you can't tell them apart.

This paper is about building a super-smart "ID system" to stop the confusion.

1. The Old Way: The "Sweaty Handshake" (The TPC)

The baseline detector at the collider uses a giant device called a TPC (Time Projection Chamber). Think of this as a giant, high-tech fog machine. As particles fly through it, they leave a trail of "sweat" (ionization clusters).

• How it works: The TPC counts how much "sweat" a particle leaves behind per inch.
• The Problem: At low speeds (low momentum), the heavy protons sweat a lot, and the light pions sweat a little. Easy to tell apart! But as they speed up, everyone starts sweating the same amount. It's like trying to tell if a marathon runner is tired just by looking at their sweat when they are all running at the same top speed. The TPC gets confused and starts misidentifying guests, especially the tricky Kaons.

2. The New Trick: The "Stopwatch" (Time-of-Flight)

To fix this, the scientists added two new tools: Stopwatches. They realized that even if two people are wearing the same clothes and sweating the same amount, they might run at slightly different speeds if they have different weights.

• The Heavy Proton: Runs slower.
• The Light Pion: Runs faster.
• The Medium Kaon: Runs somewhere in between.

The team installed two types of stopwatches:

1. The Inner Watch (ITK): A high-tech sensor right near the center of the party. It catches the slow, heavy guests (low momentum) before they even reach the outer rings.
2. The Outer Watch (OTK): A sensor on the far wall. It catches the faster guests who make it that far.

3. The Magic Combination: "The Ultimate ID Card"

The paper's main achievement is combining these two methods into one super-system.

• The Strategy: Instead of just looking at the sweat (TPC) OR just checking the time (Stopwatch), they do both at the same time.
• The Analogy: Imagine you are trying to identify a runner.
  • Method A (TPC): You look at their shoes. (Good for slow runners, useless for fast ones).
  • Method B (ToF): You time their lap. (Good for fast runners, useless for slow ones who haven't finished yet).
  • The Solution: You check the shoes and the time. Now, you can identify the runner perfectly, whether they are jogging or sprinting.

4. The Results: A Party Success Story

The scientists ran a massive simulation (a "digital party") with billions of guests to test this new system. Here is what they found:

• The TPC alone: Good at high speeds, but terrible at low speeds. It misidentified almost 80% of the Kaons in the low-speed crowd.
• TPC + Outer Watch (OTK): Much better! It fixed the confusion for medium-speed guests. But it still missed the slow ones because they never reached the outer wall.
• The Full Team (TPC + Inner Watch + Outer Watch): This is the winner.
  • By adding the Inner Watch, they caught the slow guests.
  • By adding the Outer Watch, they caught the fast guests.
  • The Result: They achieved 97.1% efficiency (they found almost every Kaon) and 85.6% purity (when they said "That's a Kaon," they were right 85% of the time).

Why Does This Matter?

In the world of particle physics, knowing exactly what a particle is (Particle Identification or PID) is crucial. If you think a Kaon is a Pion, you might miss a discovery about the Higgs boson or dark matter.

This paper proves that by adding precision timing (stopwatches) to the traditional ionization (sweat) detectors, we can create a detector that works perfectly across the entire speed range. It's like upgrading a security system from just a camera to a camera plus a fingerprint scanner plus a voice recognition system.

In short: The CEPC team figured out how to tell the difference between identical-looking particles by giving them a "sweat test" and a "race time" simultaneously. This ensures that when they discover something new in the future, they know exactly what they are looking at.

Technical Summary

1. Problem Statement

The Circular Electron–Positron Collider (CEPC) requires precise Charged-Hadron Particle Identification (PID) to achieve its physics goals, particularly for flavor tagging and precision electroweak measurements.

• Limitation of Baseline Design: The baseline CEPC detector relies on a Time Projection Chamber (TPC) using ionization measurements ($dN/dx$). While effective at low momenta, the TPC's separation power degrades significantly at higher momenta due to ionization fluctuations and overlapping response curves between particle species (pions, kaons, protons).
• The Gap: There is a critical need to extend PID capabilities across the full momentum spectrum (sub-GeV to multi-GeV) to handle the dominant background of pions in hadronic $Z \to q\bar{q}$ events. Specifically, the baseline design lacks sufficient timing information for low-momentum tracks that do not reach the outer tracker.

2. Methodology

The authors propose and evaluate a unified PID framework that combines ionization data with Time-of-Flight (ToF) measurements from silicon-based timing detectors.

• Detector Configuration:
  • TPC: Provides $dN/dx$ measurements. The study utilizes a $dN/dx$ approach (counting ionization clusters) rather than $dE/dx$ (charge deposition), which offers superior separation in gaseous detectors.
  • Outer Tracker (OTK): A silicon strip detector using AC-LGAD technology located outside the TPC. It provides precise ToF for intermediate-to-high momentum tracks ($p \gtrsim 1$ GeV/c).
  • Inner Tracker (ITK) Upgrade: The outermost layer of the inner silicon tracker is upgraded with pixelated AC-LGAD sensors. This provides ToF for low-momentum tracks ($p < 1$ GeV/c) that stop before reaching the OTK.
• Simulation Framework:
  • Simulations were performed using CEPCSW (based on Gaudi, DD4hep, and Geant4).
  • The dataset consists of 8.83 million hadronic $Z \to q\bar{q}$ events (equivalent to $1.26 \times 10^7$ inclusive events) at the Z-pole ($\sqrt{s} = 91$ GeV).
• PID Strategy:
  • A unified discriminant, $\chi_{PID}$, is constructed on a track-by-track basis.
  • The discriminant combines residuals ($\chi$) from three sources: TPC ($dN/dx$), ITK (ToF), and OTK (ToF).
  • Optimization: Identification regions in the $\chi$ space are optimized for every bin of momentum ($p$) and polar angle ($\cos\theta$) by maximizing the product of Kaon Efficiency ($\epsilon_K$) and Kaon Purity ($p_K$).

3. Key Contributions

• Quantification of Complementarity: The study rigorously quantifies how ionization ($dN/dx$) and timing (ToF) information complement each other across a wide momentum range.
• Validation of ITK Upgrade: It demonstrates that adding a timing layer to the Inner Tracker is essential for extending PID coverage into the sub-GeV region, where the OTK is geometrically inaccessible.
• Unified Discriminant Construction: The paper presents a specific mathematical framework ($\chi_{PID}$) that effectively merges heterogeneous detector responses (gaseous TPC and silicon timing layers) into a single likelihood-based decision metric.
• Performance Benchmarking: It provides a comprehensive comparison of three configurations: TPC-only, TPC+OTK, and the full TPC+OTK+ITK combination.

4. Results

The performance was evaluated for Kaon ($K^\pm$) identification against dominant Pion ($\pi^\pm$) and Proton ($p/\bar{p}$) backgrounds over the momentum range 0.25 to 40 GeV/c.

• Momentum Regime Breakdown:
  • Low Momentum ($0.25 < p < 1.0$ GeV/c): TPC-only suffers from low purity (6.93%) due to ionization fluctuations. The OTK is ineffective here due to geometry. The ITK+TPC+OTK configuration achieves 99.9% efficiency and 81.9% purity.
  • Intermediate Momentum ($1.0 < p < 5.0$ GeV/c): The OTK provides a massive boost. While TPC-only purity is 44.6%, adding OTK raises purity to 98.0%.
  • High Momentum ($5.0 < p < 40$ GeV/c): The TPC dominates as ToF differences between particles diminish. All configurations stabilize with ~93% efficiency and ~89% purity.
• Overall Performance (Full Range 0.25–40 GeV/c):
  • TPC Only: 23.2% purity.
  • TPC + OTK: 30.5% purity.
  • TPC + OTK + ITK (Proposed): 97.1% Efficiency and 85.6% Purity.
  • The combined configuration yields an efficiency-purity product of 83.1%, a significant improvement over the baseline.

5. Significance

• Feasibility for Future Colliders: The results demonstrate that precision timing in silicon trackers is not just an add-on but a critical component for achieving high-performance PID in future lepton colliders like the CEPC.
• Physics Impact: The ability to identify kaons with high purity across the full momentum spectrum is vital for:
  • Precision Higgs boson measurements (e.g., $H \to c\bar{c}$, $H \to b\bar{b}$ tagging).
  • Electroweak precision tests.
  • Future flavor physics analyses.
• Design Guidance: The study validates the necessity of the AC-LGAD upgrade for the Inner Tracker, providing a data-driven justification for detector design decisions and resource allocation.

In conclusion, the paper establishes that a combined strategy utilizing TPC ionization and dual-layer silicon timing (ITK and OTK) offers a robust, optimized solution for charged-hadron identification, effectively overcoming the limitations of traditional TPC-only designs.