Measurements of the micro-spill structure of medical cyclotron and synchrotron beams and its impact on pulse pileup

This paper reports the sub-nanosecond characterization of micro-spill structures in medical cyclotron and synchrotron beams using high-frequency silicon carbide sensors, demonstrating how precise knowledge of beam time structure is essential for mitigating pulse pileup and optimizing readout electronics for particle physics experiments.

The Gist

Imagine you are trying to listen to a conversation in a crowded room. If people speak one by one with clear pauses, you can understand every word. But if everyone starts shouting at once, or if their words overlap so quickly that they blend into a single roar, you lose the details. This is the problem scientists face when studying particle beams from medical accelerators.

This paper is about listening very closely to how particles (like protons or carbon ions) arrive at a detector, specifically looking at the tiny fractions of a second between them. Here is the breakdown of what they did and found, using simple analogies.

The Problem: The "Crowded Room"

Medical machines used for cancer therapy (cyclotrons and synchrotrons) shoot beams of particles at patients. Scientists often use these same machines to test new sensors. However, these machines are designed for patients, not for counting individual particles.

The machines have built-in monitors, but they are like a slow-motion camera trying to film a hummingbird. They can tell you the average amount of radiation, but they are too slow to see the individual "beats" of the beam. They miss the tiny gaps between particles. When particles arrive too close together, they "pile up" (overlap), confusing the sensors and ruining the data.

The Solution: A High-Speed Microphone

To fix this, the researchers built a custom "high-speed microphone" using a special material called Silicon Carbide (SiC).

• Why SiC? Think of standard silicon sensors as a heavy, slow runner. Silicon Carbide is like a sprinter. It can react incredibly fast (in less than a billionth of a second) and handle high energy without breaking.
• The Setup: They connected this fast sensor to a super-fast electronic brain (a high-frequency readout system) that could record the exact moment a particle hit it.

The Discovery: It's Not Random

The researchers expected the particles to arrive randomly, like raindrops hitting a roof. If rain is random, you can predict the average time between drops.

But they found something different:
The particles didn't arrive randomly. They arrived in a rhythmic pattern, like a drummer keeping a steady beat.

• The Cyclotron (Trento): This machine acts like a metronome set to a very fast tempo (about 106 million beats per second). The particles arrive in tiny "micro-bunches" spaced exactly 9.4 nanoseconds apart. Even though the beam looks like a continuous stream, it's actually a rapid-fire machine gun firing in perfect rhythm.
• The Synchrotron (MedAustron): This machine is more complex.
  • With a special setting (EBC): The particles arrive in a very strong, rhythmic pattern, similar to the cyclotron but with a different beat (1–3 MHz).
  • Without that setting: The rhythm is much weaker and messier, more like a chaotic crowd than a marching band, though some rhythm remains.

Why This Matters

Knowing the "beat" of the beam is crucial for designing new sensors.

• The Analogy: Imagine you are trying to count cars passing a toll booth. If you know the cars come in groups of three every second, you can set your counter to ignore anything faster than that. If you don't know the pattern, you might count a group of three as one giant car, or miss them entirely.
• The Result: By measuring these tiny time gaps, the researchers can now calculate exactly how often particles will "pile up" and confuse a sensor. This tells engineers exactly how fast their new electronics need to be to avoid errors.

The Bottom Line

The paper doesn't claim to cure cancer or invent new medical treatments. Instead, it provides a rulebook for the "timing" of these machines.

They proved that medical accelerator beams have a hidden, fast rhythm that standard monitors miss. By using their ultra-fast Silicon Carbide sensor, they mapped this rhythm. This map allows other scientists to build better, faster detectors that won't get confused when the beam gets too crowded, ensuring that future experiments (whether for physics or medical research) get accurate data.

Technical Summary

Technical Summary: Measurements of the Micro-Spill Structure of Medical Cyclotron and Synchrotron Beams and Its Impact on Pulse Pileup

Problem Statement
Detector characterization and instrumentation testing for high-energy physics (HEP) and medical physics are frequently conducted at medical accelerator facilities (cyclotrons and synchrotrons). However, for experiments requiring single-particle resolution, the time structure of the beam is a critical factor often overlooked. Standard monitoring systems at these facilities, typically parallel plate ionization chambers, lack the necessary charge ($\sim$100 fC) and time resolution ($\sim$100 $\mu$s) to characterize particle beams on a single-particle basis. Consequently, the micro-spill structure—fluctuations in particle arrival times on the nanosecond to microsecond scale driven by the accelerator's radio frequency (RF) fields and extraction mechanisms—remains largely unknown. This lack of data complicates the selection of appropriate readout parameters, as pulse pileup (the overlap of signals from successive particles) can significantly degrade data quality in counting and spectroscopic measurements.

Methodology
To address the lack of high-resolution temporal data, the authors employed a custom high-frequency (HF) readout system coupled with a silicon carbide (SiC) particle sensor.

• Sensor: A small circular 4H-SiC p-i-n diode (140 $\mu$m diameter, 50 $\mu$m active thickness) was utilized. SiC was selected for its wide bandgap (3.26 eV), high breakdown field, and high carrier saturation velocity, which enable fast timing performance and tolerance to large bias voltages.
• Readout Electronics: The sensor was coupled to a low-noise amplifier chain (Mini-Circuits PMA3-14LN+ and ZX60-14LN-S+) and digitized using a Rhode & Schwarz RTO6 oscilloscope (4 GHz analog bandwidth). This setup achieved a signal rise time of 120 ps and a full width at half maximum (FWHM) of 365 ps.
• Facilities and Beam Settings: Measurements were conducted at two facilities:
  1. Trento Proton Therapy Center (Italy): A C230 isochronous cyclotron operating at a fixed RF of $\sim$106 MHz. Measurements were taken with proton beams at 70.2 MeV and 148.5 MeV across various currents (30 nA to 300 nA).
  2. MedAustron (Austria): A synchrotron facility. Measurements included protons and carbon ions ($^{12}$C$^{6+}$) at various energies. Crucially, data were acquired in two modes: with Empty Bucket Channeling (EBC) enabled (RF active during extraction, creating strong modulation) and without EBC (RF off).
• Analysis: Particle Arrival Time Distributions (PATDs) were derived by extracting timestamps of individual particle crossings. The time differences ($\Delta t$) between consecutive particles were binned to form histograms representing the probability density function of arrival times.

Key Results

• Cyclotron Micro-Structure: At the Trento cyclotron, the PATDs exhibited an exponential decay envelope modulated by the cyclotron's RF frequency (106.35 MHz). The micro-bunch period was determined to be 9.403 ns, consistent with the RF frequency. The width of these micro-bunches varied with beam energy, measured at 1.32 ns for 70.2 MeV protons and 0.71 ns for 148.5 MeV protons. The study found that despite the modulation, the expectation values of the inter-arrival times closely matched the decay time of the exponential envelope, suggesting that standard Poisson pileup statistics could be applied if the envelope decay constant is interpreted as the mean effective rate.
• Synchrotron Micro-Structure: At MedAustron, the PATDs were more complex. With EBC active, the extracted beam showed strong modulation at the synchrotron RF frequency ($\sim$1–3 MHz). Without EBC, this modulation was significantly weaker, though remnants persisted. In both cases, the distribution envelope followed an exponential decay at short inter-arrival times ($\Delta t < 5$ $\mu$s) but deviated at larger intervals, likely due to macroscopic current ripples (Hz–kHz range) from magnet power converters.
• Pileup Probability: The study demonstrated that the resolved micro-spill structure allows for a quantitative estimation of pileup contributions. The cumulative sum of the PATD provides the pileup probability as a function of the readout system's characteristic processing time ($\tau$).

Significance and Claims
The paper claims that the presented results emphasize the necessity of characterizing beam time-structure for the development of precise readout systems. By providing direct experimental measurements of the micro-spill structure, the authors offer:

1. Quantitative Constraints: Direct design constraints for future readout electronics, specifically regarding the timing performance required to separate events and avoid pileup.
2. Scalable Reference Data: A methodology to scale the measured PATDs to different detector geometries and beamspot coverages using a 2D-Gaussian model. This allows researchers to estimate pileup probabilities for their specific experimental setups without needing to repeat the full measurement campaign.
3. Validation of SiC for Fast Timing: Confirmation that SiC sensors coupled with multi-GHz electronics are well-suited for resolving sub-nanosecond beam structures in medical accelerator environments.

The authors conclude that while the micro-spill structure often deviates from a simple Poisson distribution, the experimental determination of this structure provides essential, typically inaccessible information for optimizing hardware and experimental design in particle-beam therapy and related physics experiments.