Beam Test Characterization of Silicon Microstrip Detector Flight-Model Ladders for the AMS-02 Upgrade

This paper presents a detailed beam test characterization of the flight-model silicon microstrip detector ladders for the AMS-02 Layer-0 upgrade, evaluating their spatial resolution, response consistency across ladder regions, and angular dependence using a 350 GeV mixed hadron beam at CERN.

The Gist

Imagine the AMS-02 experiment as a giant, high-tech cosmic net floating on the International Space Station. Its job is to catch tiny, high-speed particles from deep space (cosmic rays) to figure out what the universe is made of and where it came from.

Right now, this net has nine layers of "sensors" (like a multi-layered cake) to catch these particles. But the scientists want to make the net three times bigger so they can catch more rare particles. To do this, they are adding a brand new, massive layer on top called Layer-0.

This paper is a "report card" for the new sensors before they get launched into space. Here is the story of how they tested them, explained simply:

1. The "Super-Long" Sensor Ladders

Imagine you have a standard ruler. Now, imagine gluing 8, 10, or 12 of those rulers together end-to-end to make one giant, 1-meter-long ruler. That's essentially what a Silicon Microstrip Ladder is.

• The Design: Instead of having many short sensors, they connected long strips of silicon together in a "daisy chain."
• Why? It's like using one long extension cord instead of 12 short ones. It saves a huge amount of power and space, which is critical because the detector is on a spaceship where electricity is precious.
• The Test: They built these "ladders" with different numbers of connected pieces (8, 10, or 12) and took them to CERN (a giant particle accelerator in Europe) to shoot high-speed particles at them.

2. The "Noise vs. Signal" Challenge

Think of the sensor like a microphone trying to hear a whisper (the particle) in a noisy room.

• The Problem: When you connect more sensors together (like adding more microphones to one wire), the electrical "static" or noise gets a little louder.
• The Result: The scientists found that as they added more pieces to the ladder, the background noise went up slightly.
  • An 8-piece ladder was the quietest and sharpest (like a high-fidelity microphone).
  • A 12-piece ladder was a bit noisier, so its "vision" was slightly fuzzier.
• The Good News: Even with the extra noise, the signal from the particles was strong enough. The "fuzziness" was still incredibly small (about the width of a human hair), which is perfect for the job.

3. The "Head-to-Tail" Check

Since these ladders are so long (about 1 meter), the scientists worried: Does the signal get weak or distorted as it travels from one end to the other?

• The Analogy: Imagine shouting a message from one end of a long tunnel to the other. Does the person at the far end hear it clearly, or does it get muffled?
• The Test: They shot particles at the very beginning of the ladder (the "Head") and the very end (the "Tail").
• The Result: The message came through perfectly clear at both ends. The signal didn't fade. This proved that the long wires work just as well as short ones.

4. The "Angle of Attack"

In space, particles don't always hit the detector straight on (like a bullet hitting a target). They often come in at a slant, like a raindrop hitting a windshield.

• The Analogy: If you shine a flashlight straight down on a table, you get a small, bright circle. If you tilt the flashlight, the beam stretches out into a long, thin oval.
• The Effect: When particles hit the sensor at an angle, they spread their energy over more strips (like the stretched flashlight beam).
• The Result:
  • Straight hits: Super sharp vision (very precise).
  • Angled hits: The vision gets a little blurrier because the signal is spread out.
  • The Verdict: Even at the steepest angles (30 degrees), the sensor was still accurate enough to do its job.

The Bottom Line

The scientists took these new, giant, space-ready sensors to a particle accelerator and gave them a tough workout.

• Did they work? Yes.
• Are they too noisy? No, just a tiny bit, but well within limits.
• Do they work at the ends? Yes, perfectly.
• Do they work at weird angles? Yes, though slightly less precise, but still excellent.

Conclusion: The new "Layer-0" upgrade is ready for launch. It will soon be bolted onto the International Space Station, where it will help us see the universe with a much wider, clearer view than ever before.

Technical Summary

1. Problem Statement

The Alpha Magnetic Spectrometer (AMS-02), currently operating on the International Space Station (ISS), is undergoing an upgrade to enhance its cosmic-ray detection capabilities. The primary goal is to increase the geometric acceptance by a factor of 3 and improve nuclear charge identification. This is achieved by installing a new silicon microstrip tracker layer (Layer-0) on top of the existing detector.

The Layer-0 design utilizes long detector "ladders" formed by serially connecting 8, 10, or 12 silicon microstrip detectors (SSDs). This unique architecture presents specific engineering challenges:

• Noise Accumulation: Connecting multiple sensors in series increases the readout channel length, potentially degrading the signal-to-noise ratio (S/N) and spatial resolution.
• Signal Integrity: It is critical to verify that charge transport over the long strips (up to ~1 meter) does not result in significant signal loss at the far end (Tail) compared to the near end (Head).
• Angular Dependence: Since cosmic rays arrive from all directions, the detector's performance must be characterized not only at normal incidence but also at various angles to ensure accurate tracking in space.

2. Methodology

To validate the flight-model (FM) ladders prior to full assembly, a dedicated test beam campaign was conducted in July 2025 at the CERN SPS H4 beamline.

• Beam Configuration: A mixed proton and pion beam with a momentum of 350 GeV was used.
• Experimental Setup:
  • Detector Under Test (DUT): Flight-model ladders containing 8, 10, and 12 SSDs.
  • Tracking System: A 12-layer silicon microstrip telescope was constructed upstream and downstream of the DUT. Each telescope layer used the same SSD technology and electronics as the DUT but contained only a single SSD. The telescope achieved a spatial resolution of 7.2 µm.
  • Trigger: Two plastic scintillators coupled to SiPMs provided the trigger signal.
• Measurement Scenarios:
  1. SSD Count Variation: Beam incidence on the central (Middle) SSD of ladders with 8, 10, and 12 SSDs to compare noise and resolution.
  2. Position Consistency: Beam incidence on the Head (first) and Tail (last) SSDs of a 12-SSD ladder to check for signal attenuation along the strip.
  3. Angular Dependence: Beam incidence on the Middle SSD of a 12-SSD ladder at angles of 0°, 10°, 20°, 25°, and 30°.
• Data Analysis:
  • Clusters were reconstructed using a dual-threshold noise-based method (seed > 3.5σ, neighbors > 2σ).
  • Hit positions were determined using a double-η algorithm.
  • Spatial resolution was calculated by fitting the residual distribution (difference between telescope prediction and DUT reconstruction) with a double-Gaussian function and subtracting the telescope contribution in quadrature.

3. Key Contributions

• Characterization of Long-Strip Architecture: This is the first comprehensive beam test of the specific "daisy-chain" ladder design where up to 12 SSDs are connected in series, creating effective strip lengths of ~1 meter.
• Noise vs. Resolution Quantification: The study quantifies the trade-off between the number of connected sensors and the degradation of spatial resolution due to increased electronic noise.
• Angular Performance Mapping: It provides a detailed dataset on how spatial resolution degrades as the particle incidence angle increases, which is crucial for modeling the AMS-02 acceptance in orbit.

4. Key Results

A. Noise and Spatial Resolution vs. SSD Count

• Noise: The intrinsic noise level (common-mode subtracted) increases with the number of SSDs due to the longer readout chain.
  • 8-SSD ladder: 7.6 LSB
  • 10-SSD ladder: 8.7 LSB
  • 12-SSD ladder: 9.7 LSB
• Signal Amplitude: The Most Probable Value (MPV) of the cluster signal remained consistent at ~75 LSB across all configurations, indicating stable charge collection efficiency.
• Spatial Resolution: As noise increased, the S/N ratio decreased, leading to a gradual degradation in intrinsic spatial resolution at normal incidence:
  • 8 SSDs: 9.5 µm
  • 10 SSDs: 10.6 µm
  • 12 SSDs: 11.4 µm

B. Head vs. Tail Consistency

• Measurements on the 12-SSD ladder showed no significant signal loss between the Head (closest to electronics) and Tail (farthest) SSDs.
• The cluster value distributions were nearly identical, confirming that charge transport along the ~1m strips does not suffer from appreciable attenuation or degradation.

C. Incidence Angle Dependence

• Cluster Size: As the incidence angle increased from 0° to 30°, the mean cluster size increased from 1.97 to 2.46. This is due to the longer path length in the silicon, spreading charge over more strips.
• Resolution Degradation:
  • At 0° (Normal): Resolution is 11.4 µm. The fraction of 2-strip clusters is ~50%, allowing for charge-weighted interpolation.
  • At 30°: Resolution degrades to ~17 µm.
  • Cause: At larger angles, the fraction of clusters with size ≥3 increases. These larger clusters spread the signal over too many strips, reducing the S/N per strip and limiting position reconstruction accuracy. Additionally, Landau fluctuations in energy loss for inclined tracks contribute to this limitation.

5. Significance

The results confirm that the flight-model ladders meet the performance requirements for the AMS-02 Layer-0 upgrade.

• The spatial resolution (9.5–11.4 µm) is sufficient for high-precision tracking and momentum measurement.
• The lack of signal loss over 1-meter strips validates the serial connection design, which is critical for reducing the number of readout channels and power consumption on the ISS.
• The characterization of angular dependence allows for accurate simulation and correction of tracking data in the space environment, where particles arrive from all directions.

The assembled Layer-0 plane is currently undergoing further testing and is scheduled for installation on the International Space Station in the near future.