🔬 mesoscale physics

Direct Imaging of Transition-Edge Sensors with Scanning SQUID Microscopy

This study employs scanning SQUID microscopy to image local diamagnetic responses in aluminum-manganese transition-edge sensors, revealing a longitudinal proximity effect near niobium leads that influences the device's transition width and resistance behavior to guide optimal detector design.

Original authors: Samantha Walker, Austin Kaczmarek, Jason Austermann, Douglas Bennett, Shannon M. Duff, Johannes Hubmayr, Ben Keller, Kelsey Morgan, Colin C. Murphy, Daniel Swetz, Joel Ullom, Michael D. Niemack, Katja

Published 2026-11-02

📖 4 min read☕ Coffee break read

CC BY 4.0

Original authors: Samantha Walker, Austin Kaczmarek, Jason Austermann, Douglas Bennett, Shannon M. Duff, Johannes Hubmayr, Ben Keller, Kelsey Morgan, Colin C. Murphy, Daniel Swetz, Joel Ullom, Michael D. Niemack, Katja C. Nowack

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

For decades, astronomers have relied on a special kind of detector to listen to the faint whispers of the universe. These instruments, known as transition-edge sensors, are designed to catch individual particles of light, from radio waves to gamma rays, by measuring the tiny amount of heat they deposit. To work, these sensors must be kept at temperatures colder than deep space, where they exist in a delicate state between being a perfect conductor of electricity and a normal resistor. When a particle of light hits the sensor, it warms the material just enough to push it out of this superconducting state, causing a measurable spike in electrical resistance. This change tells scientists exactly how much energy the particle carried. While these devices have revolutionized our view of the cosmos, revealing the history of the universe and the nature of dark matter, the physics governing how they behave in that critical, in-between state is not yet fully understood. Researchers know that the size and shape of the sensor matter, but they have lacked a way to see exactly how electricity and magnetism move inside the tiny films that make up these detectors.

A team of scientists has now taken a direct look inside these sensors to solve a lingering mystery about their design. Using a powerful imaging technique called scanning SQUID microscopy, which acts like a magnetic camera capable of seeing features smaller than a human hair, the researchers mapped the magnetic behavior of aluminum-manganese alloy sensors. These sensors are connected to niobium leads, which are superconducting wires that carry the electrical current. The team created three versions of the sensor, each with the same basic materials but with different distances between the connecting wires. By cooling the devices to temperatures near absolute zero and scanning them with their magnetic camera, they were able to watch how the material responded to changes in temperature and electrical current.

The researchers discovered that the distance between the connecting wires fundamentally changes how the sensor works. In the sensors where the wires were placed far apart, the material behaved as expected, transitioning from a superconductor to a normal resistor over a very narrow range of temperatures. However, in the sensors where the wires were placed very close together, the transition became much wider and occurred at a slightly higher temperature. The images revealed the reason for this shift: the superconducting property of the niobium wires was "leaking" into the aluminum-manganese film, effectively making the film superconducting in the space between the wires even when the rest of the film was not. This phenomenon, known as the proximity effect, meant that the film between the closely spaced wires remained superconducting at temperatures where the rest of the material had already stopped.

This finding explains why sensors with closely spaced wires behave differently. Because the superconducting region is larger and more stable due to this leakage, the sensor takes longer to switch states and is less sensitive to small changes in temperature. The researchers found that this effect caused the sensors with the closest wires to have a much broader transition width, spanning about 25 milli-Kelvin, compared to only 5 to 8 milli-Kelvin for the sensors with wider spacing. This broader transition is significant because it directly impacts the sensor's ability to detect faint signals accurately. The study suggests that while the proximity effect is a known concept in physics, its specific impact on the design of these detectors had not been directly visualized before. By seeing exactly where and how the superconducting state spreads, the team provided a clear guide for future engineers. To build the most sensitive detectors possible, designers must now account for this magnetic influence, ensuring that the spacing between wires is optimized to prevent the superconducting state from spreading too far and blurring the sensor's response. This work demonstrates that looking directly at the magnetic properties of these devices offers a new and powerful way to understand and improve the tools we use to explore the universe.

Technical Summary: Direct Imaging of Transition-Edge Sensors with Scanning SQUID Microscopy

Problem Statement
While significant advancements have been made in understanding the physics of Transition-Edge Sensors (TESs) over the past decade, key questions remain regarding the detailed behavior of current-dependent resistance when these detectors are voltage-biased within their superconducting transition. Historically, TES characterization has relied on macroscopic electrical measurements, such as resistance versus bath temperature ( $R_{TES}$ - $T_b$ ) curves and current-voltage ( $I$ - $V$ ) characteristics. However, these methods lack spatial resolution, preventing a direct observation of how local device geometry—specifically the spacing between superconducting leads—influences the superconducting transition width and critical temperature ( $T_c$ ). Understanding these local effects is crucial for optimizing device design, operation, and noise performance.

Methodology
The authors employed Scanning Superconducting Quantum Interference Device (SSM) microscopy to directly image the local diamagnetic response of Aluminum-Manganese (Al-Mn) alloy TES bolometers. The experiments were conducted in a cryogen-free dilution refrigerator with a base temperature of approximately 10 mK.

Sample: The study utilized an early prototype chip containing 12 TES devices, focusing on three specific devices (a, b, and c) with varying geometries. All devices featured Nb leads suspended on a silicon nitride (SiNx) membrane. The devices differed in the spacing ( $L$ ) between Nb leads and their widths ( $W$ ), resulting in different aspect ratios and normal resistances ( $R_n$ ).
Imaging Technique: The SSM system used a pickup loop (0.75 µm radius) and a concentric field coil (3 µm radius). To detect local diamagnetism, an AC current (250 µA at 270 Hz) was applied to the field coil. The change in mutual inductance ( $\delta M$ ) between the coil and the pickup loop, caused by screening currents in the superconducting sample, was measured via lock-in detection. A Helmholtz coil was used to cancel background magnetic fields, and the system was tuned to minimize superconducting vortices.
Electrical Correlation: Unlike standard TES readouts using embedded SQUID arrays, the bias current ( $I_{bias}$ ) was measured inductively by positioning the scanning SQUID near the Nb leads. This allowed the authors to correlate the local magnetic susceptibility images with $R_{TES}$ - $T_b$ curves measured at various bias currents.

Key Results
The study revealed a strong correlation between Nb lead spacing and the superconducting properties of the Al-Mn film:

Proximity Effect Observation: SSM images showed an enhanced local diamagnetic response emanating from the Nb leads into the Al-Mn film. This "spreading" diamagnetism was observed at temperatures several milli-Kelvin (mK) higher than the $T_c$ of the bulk Al-Mn film far from the contacts.
Geometry Dependence:
- Device c (smallest lead spacing, $L=10$ µm) exhibited the most pronounced effect. It showed a local diamagnetic response persisting more than 10 mK above the bulk $T_c$ and a significantly broadened resistive transition (spanning ~25 mK).
- Devices a and b (larger spacings, $L=25$ µm and $20$ µm) showed narrower transitions (~5–8 mK) and less pronounced proximity effects.
Impact on Device Parameters: The broadened transition in Device c corresponded to a significantly lower mean logarithmic sensitivity ( $\alpha = d \log R / d \log T$ ) of approximately 52, compared to $\alpha \sim 104$ for Device b and $\alpha \sim 148$ for Device a. Lower $\alpha$ values are linked to reduced TES loop gain and longer time constants.
Longitudinal Proximity Effect: The data suggests a longitudinal proximity effect where the Nb leads act as superconducting reservoirs, proximitizing the Al-Mn film and effectively raising the local $T_c$ and superfluid density between the leads. This effect becomes dominant as the lead spacing decreases.

Significance and Claims
The paper claims that this work provides the first direct observation of the proximity effect in TES devices through magnetic imaging, whereas previous evidence was solely inferred from electrical characterization.

The authors assert that SSM is a powerful tool for the local characterization of superconducting detectors. By visualizing how device dimensions influence the transition width and $T_c$ , this approach offers valuable insights into the physical mechanisms governing device operation. The findings suggest that the proximity effect between Nb leads and the Al-Mn film is a critical factor that significantly impacts device performance (specifically $\alpha$ and time constants) and must be considered in the optimal design of TES bolometers, particularly for applications requiring specific noise and response characteristics.

Future Work
The authors state their intention to use SSM to correlate device behavior at lower bath temperatures while biasing TESs at various points within their transitions. They also plan to image the spatial distribution of current flow inside TESs when voltage-biased at different percentages of normal resistance.

More like this