🔬 physics

First evaluation of the tracking performance of the ARCADIA Fully Depleted MAPS with 120 GeV proton beam

This paper presents the test beam characterization of the ARCADIA Main Demonstrator 3, a 200 $\mu$ m thick Fully Depleted MAPS fabricated in a 110 nm CIS process, demonstrating a detection efficiency exceeding 99% and a spatial resolution of 3.8 $\mu$ m when exposed to a 120 GeV proton beam.

Original authors: M. Rignanese, S. Ciarlantini, C. Pantouvakis, A. Zingaretti, A. Apresyan, P. Azzi, N. Bacchetta, C. Bonini, D. Chiappara, D. Falchieri, S. Garbolino, A. Hayrapetyan, S. Mattiazzo, L. Pancheri, D. Pant

Published 2026-06-25

📖 5 min read🧠 Deep dive

CC BY 4.0

Original authors: M. Rignanese, S. Ciarlantini, C. Pantouvakis, A. Zingaretti, A. Apresyan, P. Azzi, N. Bacchetta, C. Bonini, D. Chiappara, D. Falchieri, S. Garbolino, A. Hayrapetyan, S. Mattiazzo, L. Pancheri, D. Pantano, A. Rivetti, M. Rolo, R. Santoro, R. Turrisi, J. Wyss, I. Zoi, P. Giubilato

Original paper licensed under CC BY 4.0 (http://creativecommons.org/licenses/by/4.0/). ✨ This is an AI-generated explanation of the paper below. It is not written or endorsed by the authors. For technical accuracy, refer to the original paper. Read full disclaimer

Imagine you are trying to take a super-sharp photograph of a speeding bullet, but instead of a camera, you are using a microscopic electronic eye made of silicon. This is essentially what the scientists in this paper did. They tested a new type of "electronic eye" called the ARCADIA MD3 to see how well it could track tiny particles zooming through space at nearly the speed of light.

Here is a breakdown of their experiment and findings, using simple analogies:

The "Electronic Eye" (The Sensor)

Think of the ARCADIA MD3 chip as a digital camera sensor, but instead of taking pictures of people or landscapes, it takes pictures of subatomic particles.

The Design: It's a thin slice of silicon (200 micrometers thick, which is about the width of two human hairs) packed with 262,144 tiny square "pixels" (like a grid of tiny buckets).
The Goal: When a particle hits the silicon, it creates an electric charge. The chip's job is to catch that charge in the right "bucket" and tell the computer exactly where it happened.
The Innovation: This specific chip is "Fully Depleted." Imagine a sponge that is completely dry and ready to soak up water instantly. In this chip, they apply a high voltage to make sure the entire thickness of the silicon is ready to catch the particle's signal immediately, rather than just the surface. This makes the "catch" faster and more efficient.

The Experiment: The "Train Track" Setup

To test this new eye, the scientists didn't just look at it in a lab; they sent it to Fermilab (a giant particle accelerator in the US) to face a real "storm" of protons.

The Beam: They used a beam of protons traveling at 120 GeV (an incredibly high speed).
The Telescope: They built a "telescope" out of three of these chips lined up in a row.
- Plane 0 and Plane 2: These are the "referees." They are known to work well, so they help figure out where the particle should be.
- The DUT (Device Under Test): This is the middle chip, the new ARCADIA MD3, which is being tested.
The Synchronization: Since there was no "shutter button" to tell the chips when to take a picture, they had to synchronize them like a choir. They used a master clock signal to ensure all three chips were looking at the exact same moment in time.

The Test: Tuning the "Sensitivity"

The scientists wanted to see how the chip performed under different settings. They treated the chip like a radio, turning knobs to see what happened:

The Threshold Knob: This controls how "loud" a signal needs to be before the chip says, "I saw something!" If the threshold is too high, it misses quiet signals. If it's too low, it hears static (noise).
The Voltage Knobs: They adjusted the electricity flowing into the chip to see if it made the "buckets" catch the charge better or faster.
The Backside Voltage: This is the "sponge saturation" knob. They increased the voltage to see if the silicon became fully "depleted" (fully ready to catch particles).

The Results: A Super-Sharp Vision

The results were excellent. Here is what they found:

It Never Missed a Beat (Efficiency): No matter how they tweaked the knobs, the chip detected the particles more than 99% of the time. It was incredibly reliable.
The "Sharing" Effect: When a particle hits the silicon, the electric charge doesn't always land in just one tiny bucket; it often spills over into neighboring buckets. This is called "charge sharing."
- Analogy: Imagine dropping a drop of ink on a paper towel. It doesn't stay in one spot; it spreads out.
- The Good News: Because the charge spreads out, the scientists could calculate the exact center of the hit much more precisely than if it had stayed in just one bucket.
The Precision: The chip could pinpoint the location of a particle to within 3.8 micrometers.
- Analogy: That is roughly the width of a single strand of spider silk. To put it in perspective, if the chip were the size of a football field, it could tell you exactly which blade of grass a fly landed on.

The Conclusion

The paper concludes that the ARCADIA MD3 chip is a massive success. It is fast, incredibly accurate, and works perfectly even when the scientists changed the settings to test its limits. It proved that this new "fully depleted" technology is ready to be used in future high-energy physics experiments to track particles with extreme precision.

In short: They built a new, super-sensitive silicon eye, tested it against a storm of high-speed particles, and found that it sees with the precision of a needle point, catching almost every single particle that flies by.

Technical Summary: First Evaluation of the Tracking Performance of the ARCADIA Fully Depleted MAPS with 120 GeV Proton Beam

Problem and Context
Monolithic Active Pixel Sensors (MAPS) are widely utilized in High Energy Physics (HEP) vertex and tracking detectors due to their low material budget and cost advantages over hybrid pixel sensors. While the ALICE experiment's Inner Tracking System (using ALPIDE chips) represents a major implementation, current and next-generation HEP experiments demand increasingly stringent performance. A key technological advancement addressing these needs is the Fully Depleted MAPS (FD-MAPS), which utilizes high bias voltages to fully deplete the active volume. This approach shifts charge collection from diffusion-dominated to drift-dominated mechanisms, resulting in faster and more efficient signal collection. This work addresses the need to characterize the latest prototype in this domain: the ARCADIA Main Demonstrator 3 (MD3).

Methodology
The study presents a test beam characterization of the ARCADIA MD3 chip, a 200 µm thick FD-MAPS fabricated using a custom LFoundry 110 nm CIS process on a high-resistivity substrate. The chip features a 512 × 512 pixel matrix with a 25 µm square pitch, segmented into 16 independent sections.

The experimental measurements were conducted at the Fermilab Test Beam Facility (FTBF) using a 120 GeV proton beam. The setup consisted of a telescope of three identical ARCADIA MD3 planes: a Device Under Test (DUT) positioned centrally, flanked by two external tracking planes (Plane 0 at -33.8 mm and Plane 2 at +29.4 mm). Since the setup lacked a hardware trigger, synchronization was achieved via a software command resetting FPGA timestamp counters simultaneously across all three chips.

Data analysis was performed using the Corryvreckan framework. The procedure involved:

Alignment: A pre-alignment using Plane 0 as a reference, followed by a precise alignment based on reconstructed tracks to correct for translational and rotational misalignments.
Track Reconstruction: Tracks were reconstructed on the external planes by associating cluster pairs within a 5 µs time window. Only events with a single cluster per plane were retained to avoid ambiguities.
Parameter Scans: The DUT performance was evaluated by varying specific parameters:
- Threshold: Scanned in the range [38, 62] DAC (corresponding to effective thresholds up to ~2300 electrons).
- Front-end Bias Currents: Scanning of discriminator current ( $I_D$ ) and feedback branch bias current ( $I_{FB}$ ).
- Depletion Voltage: Scanning the backside bias voltage in the range [-70, -96] V.
Metrics: The study measured cluster size (multiplicity), cluster width, residual width (spatial resolution), detection efficiency, and fake hit rates.

Key Results

Detection Efficiency: The ARCADIA MD3 demonstrated a detection efficiency exceeding 99% across the entire scanned range of thresholds and backside bias voltages. The efficiency was found to be independent of the threshold setting, which is consistent with the fact that the set thresholds were significantly lower than the signal generated by a Minimum Ionizing Particle (MIP) in 200 µm of silicon (~16,000 electrons).
Spatial Resolution: Due to the 200 µm substrate thickness, charge sharing between adjacent pixels is prevalent. Consequently, cluster sizes larger than 1 pixel are dominant even at high thresholds. This charge sharing results in a residual distribution width that is consistently lower than the binary resolution of 7.2 µm.
- The minimum residual width was measured at 4.65 µm (at a threshold of 56 DAC).
- By decoupling the telescope's contribution from the intrinsic sensor resolution using a geometrical calculation, the intrinsic spatial resolution of the ARCADIA MD3 was determined to be 3.8 µm at a threshold of 58 DAC.
Parameter Dependencies:
- Threshold: Lower thresholds increased cluster width and multiplicity but did not significantly impact efficiency.
- Bias Currents: Variations in $I_D$ and $I_{FB}$ showed minor effects on cluster and residual widths (variations of 2-20% depending on the parameter), with efficiency remaining stable above 99%.
- Depletion Voltage: Performance metrics (residual width and efficiency) showed no significant dependence on the backside bias voltage within the tested range. The results indicate that the substrate is fully depleted at -70 V, with no further performance gains observed up to -96 V.
Noise: The fake hit rate remained below $4 \times 10^{-6}$ hit/pixel/s for almost the entire threshold range, with a notable increase only at the lowest threshold setting.

Significance and Claims
The paper claims that the ARCADIA MD3 successfully meets the requirements for high-performance tracking in HEP applications. The primary contribution of this work is the experimental validation of the MD3 prototype's tracking capabilities, demonstrating:

High detection efficiency (>99%) in a 120 GeV proton beam environment.
Sub-pixel spatial resolution (3.8 µm) achieved through effective charge sharing management in a fully depleted architecture.
Robustness of performance across variations in threshold and bias currents.
Confirmation that full depletion is achieved at -70 V for the 200 µm substrate, establishing the operational baseline for the device.

The authors conclude that the ARCADIA MD3 represents a viable candidate for future HEP vertex and tracking detectors, leveraging the benefits of fully depleted technology to achieve fast charge collection and high spatial precision.

The "Electronic Eye" (The Sensor)

The Experiment: The "Train Track" Setup

The Test: Tuning the "Sensitivity"

The Results: A Super-Sharp Vision

The Conclusion

More like this