Performance of photosensors in a high-rate environment for gas Cherenkov detectors

Chao Peng, Junqi Xie, S. Joosten, Z.-e. Meziani, A. Camsonne, Mark Jones, E. Kaczanowicz, M. Rehfuss, Nikolaos Sparveris, M. Paolone, Michael Foley, Michael J. Minot, M. PopeckiarXiv 2020 · DOI · arXiv

This paper reports on a critical "stress test" for the next generation of particle physics detectors. As experiments push for higher luminosity (more particle collisions per second), the sensors used to identify those particles are being pushed to their absolute physical limits.

TL;DR

Researchers successfully validated that multi-anode photomultipliers (MaPMTs) and Large Area Picosecond Photodetectors (LAPPDs) can survive and function in the punishing high-rate environments required for the upcoming SoLID experiment at Jefferson Lab. While MaPMTs currently offer better signal separation due to higher quantum efficiency, the low-cost LAPPD proved to be a viable, more alignment-resilient alternative for large-scale Cherenkov detectors.

What Problem Does This Paper Address?

The Solenoidal Large Intensity Device (SoLID) planned for Jefferson Lab aims to operate at unprecedented luminosities. To distinguish electrons from a massive background of pions, researchers use Light-Gas Cherenkov (LGC) counters.

The challenge is twofold:

  1. The Rate: The background single-photon rate at the photosensors is expected to exceed 160 kHz/cm². Most traditional sensors would be blinded or overwhelmed by this "noise."
  2. The Environment: The sensors must operate in a 200-gauss magnetic field, which renders traditional vacuum-tube photomultipliers useless.

Researchers needed to prove that high-granularity sensors—which allow for spatial correlation to filter out random background hits—could actually handle these rates using standard readout electronics (the JLab FADC250).

Methods

The team constructed a "Cherenkov telescope": a 40-inch PVC tank filled with $CO_2$ gas. When high-energy electrons pass through the gas, they emit a cone of Cherenkov light. This light is reflected by a 45° mirror onto the photosensors.

They tested two primary sensor technologies:

  • MaPMT Array: A $2 \times 2$ grid of Hamamatsu MaPMTs (64 pixels each, though read out in quadrants for this test).
  • LAPPD: A newer, "Gen-I" microchannel plate (MCP) detector with a massive $20 \times 20$ cm² active area and stripline readout.

The experiment was conducted "parasitically" in Hall C at Jefferson Lab, meaning they placed their telescope near an ongoing experiment's beamline to catch scattered 8.1 GeV electrons in a real-world, high-rate environment.

Datasets

The study utilized two main sources of data:

  • Experimental Beam Data: Real-world triggers from an 8.1 GeV electron beam hitting a liquid hydrogen target, resulting in electronics rates of ~400 kHz per channel.
  • GEANT4 Simulation: A detailed Monte Carlo simulation of the telescope’s geometry, gas properties, mirror reflectivity, and sensor quantum efficiency used to validate the experimental results.

Key Insights and Ideas

  • Granularity as a Filter: Because these sensors have many pixels/channels, the team could use "pixel correlation." A real Cherenkov event hits multiple nearby pixels simultaneously, while background photons hit random pixels at different times. This allows researchers to "see" the signal through the noise.
  • The QE vs. Cost Trade-off: MaPMTs are a mature technology with high Quantum Efficiency (QE ~30%), making it easy to distinguish between a single electron and "pair-production" events (where a photon converts into an electron-positron pair, doubling the light). The LAPPD tested had a much lower QE (~7.3%), making this separation harder but still possible.
  • Geometric Resilience: The LAPPD’s massive size made it much more "forgiving" regarding hardware alignment. While a 15° mirror tilt in the MaPMT setup caused half the light to miss the sensor, the LAPPD’s large surface area captured the entire shifted light cone.

Key Results

  • Rate Handling: Both sensors and the FADC250 electronics successfully handled rates up to 60 kHz/cm².
  • Signal Detection: The MaPMT array detected an average of 14 photoelectrons ($N_{pe}$) per event. While lower than the predicted 30 (due to alignment issues), it provided excellent separation between signal and background.
  • LAPPD Performance: The LAPPD detected 9.1 $N_{pe}$. Despite the lower light yield, the experimental results matched the GEANT4 simulations perfectly, proving the technology is predictable and ready for further development.
  • Simulation Match: The GEANT4 simulation confirmed that the "missing" light in the MaPMT setup was due to a 15° pointing misalignment, not a failure of the sensors themselves.

Limitations and Future Work

The primary limitation was the "parasitic" nature of the test; the researchers had limited access to the hall to perfectly align the telescope, leading to the 50% light loss in the MaPMT array. Additionally, the LAPPD used was an early "Gen-I" version with stripline readout rather than pixels.

Future Work: * Testing pixelated LAPPDs with capacitive coupling, which should provide better spatial resolution. * Developing sensors with higher Quantum Efficiency to match MaPMT performance while maintaining the LAPPD's cost-per-area advantage. * Building a larger-scale prototype to further mitigate alignment sensitivities.

Summary updated

Abstract

The solenoidal large intensity device (SoLID) at Jefferson Lab will push the boundaries of luminosity for a large-acceptance detector, which necessitates the use of a light-gas threshold Cherenkov counter for online event selection. Due to the high luminosity, the single-photon background rate in this counter can exceed 160 kHz/cm$^2$ at the photosensors. Therefore, it is essential to validate the high-rate limits of the planned photosensors and readout electronics in order to mitigate the risk of failure. We report on the design and an early set of studies carried out using a small telescopic Cherenkov device in a high-rate environment up to 60 kHz/cm$^2$, in Hall C at Jefferson Lab. Commercially available multi-anode photomultipliers (MaPMT) and low-cost large-area picosecond photodetectors (LAPPD) were tested using the JLab FADC250 modules for readout. The test beam results show that the MaPMT array and the internal stripline LAPPD can detect and identify single-electron and pair-production events in high-rate environments. Due to its higher quantum efficiency, the MaPMT array provided a better separation between the single-electron and the pair-production events compared to the internal stripline LAPPD. A GEANT4 simulation confirms the experimental performance of our telescopic device.
BibTeX
@misc{peng2020performance,
  author = {Chao Peng and Junqi Xie and S. Joosten and Z.-E. Meziani and A. Camsonne and Mark Jones and E. Kaczanowicz and M. Rehfuss and Nikolaos Sparveris and M. Paolone and Michael Foley and Michael J. Minot and M. Popecki},
  title = {{Performance of photosensors in a high-rate environment for gas Cherenkov detectors}},
  year = {2020},
  eprint = {2011.11769},
  archivePrefix = {arXiv},
  url = {https://arxiv.org/abs/2011.11769},
}

Loading related content...

Figure