Characterization of UV optical components for photon detector calibration in liquid argon TPCs

This paper characterizes the performance and durability of various UV optical components, including fibers, connectors, and diffusers, under cryogenic and high-rate conditions, demonstrating their reliability for in situ calibration of photosensors in large-scale liquid argon time projection chambers like DUNE.

The Gist

Imagine you are trying to take a photograph of a ghost inside a giant, frozen block of ice. To do this, you need a flashlight, but the ghost is invisible to normal light. You need a special "magic" flashlight that shines ultraviolet (UV) light, which the ice glows in response to.

This paper is about building and testing that special flashlight system for a massive experiment called DUNE (Deep Underground Neutrino Experiment). The experiment involves huge tanks of liquid argon (a type of frozen gas) kept at temperatures colder than outer space. The scientists need to check if their cameras (sensors) are working correctly while they are frozen solid.

Here is the story of how they tested the "plumbing" and "lenses" for this frozen flashlight system, explained simply:

1. The Problem: Sending Light Through a Frozen Pipe

The scientists have a UV light source sitting in a warm room. They need to send that light through a long pipe (an optical fiber) into a giant tank of liquid argon, which is freezing cold.

• The Challenge: When things get extremely cold, materials can crack, shrink, or stop working. Also, UV light is tricky; it gets absorbed easily, like trying to shout through a thick wool blanket. If the light gets too dim before it reaches the sensors, the calibration fails.
• The Goal: They needed to find the perfect "pipes" (fibers), "connectors" (plugs), and "diffusers" (light spreaders) that wouldn't break in the cold and wouldn't lose too much light.

2. Testing the Pipes (Optical Fibers)

Think of optical fibers as tiny glass straws that carry light instead of water. The team tested different types of straws:

• The Material: Some straws were coated in plastic, some in special heat-resistant materials (like Tefzel), and some were bare glass.
• The Test: They shined different colored lights (from deep UV to near-infrared) through these straws.
• The Result: They found that some straws were terrible at carrying deep UV light (it got absorbed instantly), while others (specifically a type called FVP600660710) were like high-speed highways for UV light, letting almost everything through. They also found that the "jacket" (the outer coating) didn't really matter for the light quality, but it mattered for protection.

3. The "Leaky" Connectors

You can't just glue a fiber to another fiber; you have to plug them in. These plugs are called SMA connectors.

• The Analogy: Imagine trying to pour water from one hose into another. If the hoses don't line up perfectly, water splashes out.
• The Finding: Every time they plugged a fiber in, they lost about 12% to 15% of the light. It's like a leaky faucet. Since the system needs to use many plugs, this "leak" adds up quickly. They measured this carefully so they could calculate exactly how much light would be left at the end of the line.

4. The "Freeze-Thaw" Test (Thermal Cycling)

This was the most dramatic part of the experiment.

• The Setup: They built a robot arm that dipped the fibers into a bucket of liquid nitrogen (which is -196°C or -320°F) and pulled them back out. They did this 30 times in a row.
• The Fear: Would the glass shatter? Would the plastic coating crack like an old rubber band in winter?
• The Result: Nothing broke. The fibers survived the freezing and thawing with no cracks and no loss of light. They proved these fibers are tough enough to survive the harsh environment of a particle detector.

5. The "Sunburn" Test (UV Aging)

UV light is energetic. Over 20 years, shining intense UV light through a fiber is like giving the glass a constant sunburn. This can sometimes turn the glass cloudy (a process called "solarization"), blocking the light.

• The Test: They blasted the fibers with millions of UV pulses, simulating 20 to 30 years of operation in just a few days.
• The Result: The fibers didn't get cloudy. They stayed clear. The "sunburn" didn't damage them. This means the system will work reliably for decades.

6. The "Frosted Lightbulb" (The Diffuser)

Once the light reaches the end of the fiber inside the tank, it comes out as a tight, laser-like beam. But the scientists want to light up a whole wall of sensors, not just one spot. They need a "diffuser" to spread the light out evenly, like a frosted lightbulb cover.

• The Innovation: Instead of using heavy, expensive metal parts, they 3D-printed a small, palm-sized holder made of a tough plastic called PEEK. Inside, they stacked two pieces of special "frosted glass."
• The Result: This 3D-printed gadget spread the light perfectly evenly in all directions (like a gentle, uniform glow). It was cheaper, easier to make, and worked better than the old metal designs.

The Big Picture

This paper is essentially a "user manual" for building a reliable flashlight system for the deep future of physics.

• They found the best pipes: High-quality fused silica fibers.
• They measured the leaks: They know exactly how much light is lost at every plug and connector.
• They proved it's tough: The system survives freezing cold and decades of UV exposure.
• They built a better lens: A 3D-printed diffuser that spreads light perfectly.

Because of this work, the DUNE experiment (and others like it) can now confidently say: "We know exactly how our light calibration system works, it won't break in the cold, and it will last for the next 20 years." This ensures that when they detect a neutrino (a tiny, ghostly particle), they can trust the data they collect.

Technical Summary

1. Problem Statement

Large Liquid Argon Time Projection Chambers (LArTPCs), such as those planned for the Deep Underground Neutrino Experiment (DUNE), require precise in situ calibration of photon detectors (SiPMs and PMTs) operating at cryogenic temperatures (~87 K).

• The Challenge: There is no efficient, cost-effective 128 nm light source (matching Liquid Argon's vacuum ultraviolet scintillation) available for direct calibration. Consequently, experiments rely on UV LEDs (e.g., 275 nm or 367 nm) coupled to fused-silica fibers, with wavelength shifting occurring at the detector surface.
• The Gap: The performance of optical components (fibers, connectors, feedthroughs, and diffusers) under cryogenic conditions and high-rate UV exposure is not fully characterized. Key concerns include:
  • Optical transmission losses in the UV range.
  • Mechanical and optical degradation due to repeated thermal cycling (liquid nitrogen immersion).
  • "Solarization" (UV-induced absorption) from long-term exposure to high-intensity pulsed light.
  • The need for uniform light distribution across large detector volumes.

2. Methodology

The authors conducted a comprehensive, end-to-end characterization of the optical chain using bench-top, cryogenic, and accelerated aging tests.

• Components Tested:
  • Fibers: Various multi-mode fused-silica fibers (core diameters 400–600 µm) with different coatings (Acrylate, Polyimide, TECS, Tefzel). Specific focus on FVP600660710 (High-OH, Polyimide) and FP600URT (Tefzel).
  • Connectors/Feedthroughs: SMA-to-SMA connectors and 5-channel optical feedthroughs bridging warm/cold boundaries.
  • Diffusers: UV-grade fused-silica diffusers (220-grit) in various configurations, including a new 3D-printed PEEK housing.
• Experimental Setup:
  • Transmission Measurements: Used LED sources at 275, 367, 370, 465, 810, and 970 nm. A reference fiber setup measured input power ($P_{ref}$), while test fibers measured output power ($P_{test}$) to calculate transparency ($T = P_{test}/P_{ref}$).
  • Cryogenic Cycling: A Cryogenic Thermal Cycle Apparatus (CTCA) automated the immersion of fiber loops (16 cm diameter) into liquid nitrogen for 30 cycles (6 minutes each) to simulate detector operation.
  • Aging Tests: Fibers were exposed to high-rate pulsed UV light (275 nm and 367 nm) simulating 20–30 years of DUNE operation (approx. 30–90 million pulses).
  • Diffuser Characterization: A photoresistor scanned light intensity at 10 cm intervals on the floor to map spatial distribution and fit against Lambert's cosine law.

3. Key Contributions

• Quantification of Losses: Separated and quantified losses for bulk fibers, SMA-to-SMA connectors, and optical feedthroughs specifically at UV wavelengths (275 nm and 367 nm).
• Cryogenic Validation: Provided empirical evidence that specific fiber types and coatings remain mechanically and optically stable after 30 liquid-nitrogen thermal cycles.
• Aging Limits: Established upper limits on UV-induced degradation (solarization) for fibers used in high-rate calibration systems.
• Diffuser Innovation: Developed and validated a compact, 3D-printed PEEK housing for stacked UV-grade diffusers that achieves near-Lambertian emission, replacing complex machined stainless steel designs.

4. Key Results

A. Optical Transmission & Attenuation

• Fiber Performance:
  • FVP600660710 (High-OH): Excellent UV transmission. At 275 nm, the attenuation coefficient is 0.050 m⁻¹; at 367 nm, it is 0.028 m⁻¹. It transmits efficiently down to 270 nm.
  • FP600URT (Tefzel): Poor performance at 275 nm (attenuation coefficient 0.214 m⁻¹, survival fraction ~13% for 9.55 m). Suitable only for wavelengths >300 nm.
• Connector & Feedthrough Loss:
  • A single SMA-to-SMA connector introduces a loss of 15.2 ± 5.1% at 275 nm and 12.4 ± 5.0% at 367 nm.
  • In a DUNE-style mockup (4 × 4.7 m fibers + 1 feedthrough + 5 connectors), total transmission is 20.9% at 275 nm and 44.8% at 367 nm.

B. Cryogenic Stability

• Thermal Cycling: After 30 cycles in liquid nitrogen, all tested fibers showed no statistically significant degradation in transparency. Ratios of post-cycle to pre-cycle transparency were consistent with unity (within ±10% statistical uncertainty).
• Mechanical Integrity: Visual inspection revealed no cracking, crushing, or deformation of the fibers or coatings.

C. UV Aging (Solarization)

• Degradation Limits: Exposure to ~30–90 million pulses (simulating decades of operation) resulted in no measurable transparency loss.
• Upper Limits: For the FVP600660710 fiber, the 68% confidence level upper limits for degradation were set at 2.8% for 275 nm and 1.2% for 367 nm.

D. Diffuser Characterization

• Uniformity: The 3D-printed PEEK housing with two stacked 220-grit diffusers produced the most uniform angular distribution.
• Lambertian Fit: The emission profile fit Lambert's cosine law ($P(\theta) = P_0 \cos \theta$) with a reduced chi-squared ($\chi^2/ndf$) of 0.18/10, confirming highly uniform illumination suitable for large-area calibration.

5. Significance

This work provides the critical data required to design reliable calibration systems for next-generation neutrino detectors like DUNE.

• Component Selection: It validates FVP600660710 (High-OH, Polyimide) as the superior choice for deep-UV (275 nm) calibration, while confirming FP600URT is unsuitable for <300 nm.
• System Design: The quantified loss budgets (fiber + connector + feedthrough) allow engineers to accurately size light sources to ensure sufficient photon flux reaches the detectors after traversing long cryostats.
• Reliability: The demonstration of stability under thermal cycling and high-rate UV pulsing ensures that calibration systems will not degrade over the 20+ year lifespan of the experiment.
• Cost & Manufacturing: The successful deployment of the 3D-printed PEEK diffuser housing offers a scalable, cost-effective alternative to machined metal components, facilitating easier replication for other experiments.

These findings have already been applied to the ProtoDUNE-SP and ProtoDUNE-HD prototypes at CERN, ensuring the successful operation of their UV light calibration systems.