⚛️ high-energy experiments

In situ cryogenic characterization of proton damage in thick p-channel skipper CCDs

This paper presents the first in situ cryogenic characterization of proton-damaged p-channel skipper CCDs, demonstrating that these sensors maintain excellent performance and radiation hardness equivalent to a decade at the Earth/Sun L2 Lagrange point, confirming their suitability for future photon-starved deep-space astronomical instruments.

Original authors: Brandon M. Roach, Brenda Cervantes Vergara, Alex Drlica-Wagner, Phoenix Alpine, Ana Martina Botti, Claudio Chavez, Julian Cuevas-Zepeda, Juan Estrada, Guillermo Fernandez Moroni, Nora Hoch, Stephen E.

Published 2026-06-26

📖 5 min read🧠 Deep dive

CC BY 4.0

Original authors: Brandon M. Roach, Brenda Cervantes Vergara, Alex Drlica-Wagner, Phoenix Alpine, Ana Martina Botti, Claudio Chavez, Julian Cuevas-Zepeda, Juan Estrada, Guillermo Fernandez Moroni, Nora Hoch, Stephen E. Holland, Blas Irigoyen Gimenez, Agustin Lapi, Santiago Perez, Nathan Saffold, Javier Tiffenberg, Yikai Wu

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

The Big Picture: Taking a "Super-Sensitive Camera" to Space

Imagine you are trying to take a picture of a firefly in a dark forest, but the firefly is so dim that it only blinks once every hour. To see it, you need a camera that is incredibly sensitive and doesn't make any "static" noise of its own.

This paper is about testing a special type of camera sensor called a Skipper CCD. These sensors are like super-powered eyes that can count individual photons (particles of light) one by one. They are being considered for future space telescopes, like the Habitable Worlds Observatory (HWO), which will try to take pictures of Earth-like planets orbiting other stars.

The Problem: Space is a Bullet Storm

Space is a dangerous place for electronics. It is filled with high-energy particles (like protons from the sun and cosmic rays) that act like tiny, invisible bullets. When these "bullets" hit the silicon in a camera sensor, they knock atoms out of place.

Think of the sensor's silicon as a perfectly organized grid of bowling pins. When a radiation "bullet" hits, it knocks a pin over. If too many pins are knocked over, the camera starts making mistakes:

Ghost Images: It might think it sees light where there is none (dark current).
Smearing: It might drop the light it catches as it moves it across the sensor (Charge Transfer Inefficiency).
Dead Pixels: Some spots might get so damaged they stop working entirely (hot pixels).

The Solution: A Different Kind of Sensor

Most cameras use an "n-channel" design, which is like a highway for electrons. Unfortunately, the radiation "bullets" love to knock electrons off their path, causing traffic jams and errors.

The sensors in this study use a "p-channel" design. Instead of a highway for electrons, this is a highway for "holes" (the absence of an electron). The paper argues that the radiation "bullets" don't knock these holes off their path as easily. It's like trying to knock a heavy boulder (the hole) off a track versus a light pebble (the electron); the boulder is harder to displace.

The Experiment: Freezing the Damage

The researchers wanted to know: Do these p-channel sensors stay tough when they are actually running in space?

In space, these cameras operate at extremely cold temperatures (around -130°C or 140 Kelvin). The researchers knew that if they tested the sensors at room temperature, the damage might "heal" itself or behave differently. So, they built a special portable freezer (a cryostat) that could hold the sensors at space-cold temperatures while they were being shot at with protons.

The Setup:

They took two different sensors (from two different manufacturers) and put them in this freezer.
They took them to a proton beam facility (like a giant particle accelerator).
They blasted the sensors with protons to simulate 10 years of space travel at the Earth-Sun L2 Lagrange point (a common spot for space telescopes).
Crucially, they kept the sensors cold and powered on during the test, just like they would be in a real telescope.

The Results: The Sensors Passed with Flying Colors

After the "10-year" bombardment, the researchers checked the sensors to see how much damage they took. Here is what they found:

The "Static" Noise: The sensors remained incredibly quiet. They could still count single photons without adding extra noise. The "skipper" amplifier (the part that reads the data) didn't break.
The "Smearing" (CTI): The sensors didn't drop many photons as they moved them across the chip. One of the sensors (from Vendor #2) was so clean that it showed almost no smearing at all, even after the heavy radiation dose.
Ghost Images (Dark Current): The sensors didn't start generating fake signals. They remained dark and quiet, meeting the strict requirements for future space missions.
Dead Pixels: Very few pixels went "hot" (became permanently bright). The number of bad pixels was incredibly low compared to older types of sensors.

The Catch: The "Frozen" Damage

There is one important warning in the paper. While the sensors performed perfectly while they were cold, the damage isn't necessarily gone; it's just frozen in place.

Imagine the radiation knocked over a few bowling pins, but because it was so cold, the pins couldn't roll away or settle into a new position. They are just stuck there.

The Risk: If the camera ever gets warm (for example, if the telescope needs to bake out to remove moisture, or if the temperature control fails), those "frozen" pins might suddenly start rolling. This could cause the damage to become permanent and worse than it was before.
The Lesson: These sensors are great, but they must stay cold. If they get warm, the hidden damage might wake up and ruin the image quality.

Conclusion

This paper proves that p-channel Skipper CCDs are tough enough to survive a 10-year journey through deep space without losing their ability to see the faintest light. They are excellent candidates for future telescopes, provided the telescope keeps them cold and doesn't let them get warm enough to "unfreeze" the hidden radiation damage.

Technical Summary: In situ cryogenic characterization of proton damage in thick p-channel skipper CCDs

Problem Statement
Future space-based astronomical missions, such as the Habitable Worlds Observatory (HWO), require detectors capable of single-photon counting with extremely low spurious signal rates to image Earth-mass planets in habitable zones. While Skipper Charge-Coupled Devices (CCDs) offer the necessary sub-electron readout noise through repeated non-destructive sampling, their viability for deep-space applications depends on radiation hardness. Conventional n-channel CCDs are highly susceptible to displacement damage from space radiation (protons and heavy ions), which creates charge-trapping sites that degrade Charge Transfer Inefficiency (CTI) and increase dark current. Although p-channel CCDs are known to be more radiation-hard than n-channel devices due to their use of holes as majority carriers and lower operating temperatures, previous radiation-hardness tests were conducted at room temperature or without bias voltages. There was a lack of data characterizing p-channel Skipper CCDs under realistic deep-space conditions: cryogenic operating temperatures (∼140–160 K) and active bias voltages.

Methodology
The authors conducted two proton-irradiation campaigns (March 2025 and July 2025) at the Northwestern Medicine Proton Center using a custom portable vacuum cryostat test station. This setup allowed for irradiation and subsequent characterization of two thick (∼650 µm), fully depleted p-channel Skipper CCDs (CCD #1 and CCD #2) from different vendors while maintaining them at their nominal operating temperatures (160 K for CCD #1, 140 K for CCD #2) and with bias voltages applied.

Irradiation Parameters: Protons were accelerated to 64 MeV (reducing to ∼60 MeV at the sensor surface) to simulate the displacement damage expected at the Earth/Sun L2 Lagrange point. The tests delivered cumulative fluences equivalent to approximately 10 years of exposure at L2, assuming 47 mm of aluminum-equivalent shielding. The total Displacement Damage Dose (DDD) reached ∼1.1 × 10⁷ MeV/g.
Test Conditions: Unlike previous studies, the sensors were powered on during irradiation (with charge flushing to prevent buildup) and characterized immediately post-irradiation at cryogenic temperatures to "freeze" damage and prevent thermal annealing.
Characterization Metrics: The study assessed:
- Skipper Amplifier Performance: Readout noise ( $\sigma_r$ ) scaling with the number of samples ( $N_{samp}$ ) and amplifier gain stability.
- Charge Transfer Inefficiency (CTI): Measured using the Extended Pixel Edge Response (EPER) method with LED illumination.
- Dark Current: Measured via deep multi-sample dark frames to determine the rate of single-electron events.
- Hot Pixels: Identification of pixels with excessive dark current.
- Output Transistor Curves: Analysis of Total Ionizing Dose (TID) effects on the floating-gate output stage MOSFETs.

Key Results

Radiation Hardness of Skipper Amplifiers: The floating-gate output stages demonstrated excellent resilience. Readout noise ( $\sigma_1$ ) and gain ( $g_e$ ) remained stable across all fluence levels, with no measurable degradation. The noise spectral index remained near ideal ( $\alpha \approx 0.5$ ), confirming that the non-destructive sampling capability was preserved.
Charge Transfer Inefficiency (CTI):
- CCD #2 (Vendor #2): This sensor, which had low pre-irradiation trap densities, exhibited exceptional performance. Post-irradiation CTI remained extremely low ( $\lesssim 1.5 \times 10^{-6}$ ) in both serial and parallel directions. The lack of significant CTI increase suggests that the formation rate of radiation-induced traps (specifically divacancies, $V_2$ ) was suppressed, likely due to the low irradiation temperature (140 K) limiting vacancy mobility.
- CCD #1 (Vendor #1): This sensor showed high pre-existing CTI attributed to transition-metal impurities. Post-irradiation, the CTI remained unchanged, indicating that the proton bombardment did not significantly degrade the performance further, though the pre-existing defects dominated the signal.
- Comparison: The results for CCD #2 met or exceeded the expected damage curves for p-channel devices extrapolated from previous 55Fe-based studies, significantly outperforming comparable n-channel devices.
Dark Current and Hot Pixels:
- Dark Current: CCD #2 maintained a dark current rate of $\lambda \lesssim 1.5 \times 10^{-4}$ e⁻/pixel/s, meeting the baseline requirements for HWO. The temperature dependence of the dark current revealed two dominant trap species: a shallow trap ( $\sim 0.14$ eV, consistent with A-centers) and a deep trap ( $\sim 0.56–0.63$ eV). The calculated dark current damage factor ( $K_{dark}$ ) matched the universal silicon value, confirming that the post-irradiation dark current was dominated by radiation damage rather than fabrication defects.
- Hot Pixels: The fraction of hot pixels emitting $\ge 1$ e⁻/pixel/min remained $\lesssim 10^{-4}$ after the 10-year-equivalent dose, an order of magnitude lower than previous tests of conventional p-channel CCDs.
Output Transistors: Transistor curve scans indicated minimal Total Ionizing Dose (TID) effects, with threshold voltages ( $V_t$ ) shifting by less than 10 mV, consistent with the low TID ( $\ll 1$ krad) delivered by the proton beam.

Significance and Claims
The paper claims to provide the first conclusive evidence that p-channel Skipper CCDs maintain their radiation hardness in realistic deep-space thermal and electrical environments. By demonstrating that these sensors can withstand displacement damage doses equivalent to a decade at L2 without significant degradation in readout noise, CTI, or dark current, the study validates p-channel Skipper CCDs as strong candidates for future photon-starved astronomical instruments.

The authors emphasize that the low operating temperature is a critical factor; it not only suppresses dark current but appears to inhibit the formation or mobility of certain radiation-induced defects (like $V_2$ ) that would otherwise degrade CTI. However, the paper modestly notes that while displacement damage is well-characterized, the TID resilience of the readout electronics requires further dedicated study. Furthermore, the authors highlight a potential operational constraint: the "freezing" of defects at cryogenic temperatures may lead to reverse annealing and permanent degradation if the sensors are subjected to thermal cycles (e.g., during spacecraft bake-outs) that mobilize these latent defects.

In summary, the work establishes that thick, p-channel Skipper CCDs are remarkably resilient to displacement damage, provided they are operated and maintained at cryogenic temperatures, thereby opening new avenues for high-sensitivity space-based astronomy.