🔬 mesoscale physics

Imaging superconducting weak spots through vortex-assisted THz near-field photovoltage

This paper demonstrates the first application of THz near-field photovoltage nanoscopy to image nanoscopic weak spots in superconductors with 300 nm resolution, revealing how defects enhance vortex-antivortex pair nucleation and influence light-matter interactions.

Original authors: Sergio J. Salvía-Fernández, Ekaterina Khestanova, Zoe Velluire-Pellat, Leonardo R. Cadorim, Sergi Batlle-Porro, David A. Czaplewski, Johann Osmond, Sebastian Castilla, Adrian Bachtold, Milorad V. Milo

Published 2026-07-02

📖 5 min read🧠 Deep dive

CC BY 4.0

Original authors: Sergio J. Salvía-Fernández, Ekaterina Khestanova, Zoe Velluire-Pellat, Leonardo R. Cadorim, Sergi Batlle-Porro, David A. Czaplewski, Johann Osmond, Sebastian Castilla, Adrian Bachtold, Milorad V. Milošević, Aharon Kapitulnik, Frank H. L. Koppens

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: Finding the "Weak Links" in a Superconductor

Imagine a superconductor as a super-highway where electricity can flow without any friction or traffic jams. In a perfect world, this highway would be smooth and uniform. But in reality, these materials are more like a road with hidden potholes, gravel patches, and weak spots.

When you push electricity through this road, the traffic (electrons) tends to get stuck or slow down at these weak spots first. This causes the whole system to lose its super-power and start generating heat. Scientists call these weak spots "defects."

The Problem:
These defects are tiny—often smaller than a grain of sand. However, the energy used to study them (Terahertz light) has a wavelength that is huge, like a giant ocean wave. Trying to see a tiny pebble with a giant wave is impossible; the wave just washes over it without noticing the details. This is like trying to find a specific pothole on a highway using a satellite image that is too blurry to see the road surface.

The Solution:
The researchers in this paper built a special "microscope" that acts like a tiny, high-powered flashlight. Instead of using a giant wave, they used a sharp metal tip (like the needle on a record player) to focus the light down to a tiny, nanoscale spot. This allowed them to scan the superconductor and find exactly where the weak spots are, even though they are hundreds of times smaller than the light wave itself.

How the Experiment Worked

The Setup: They took a thin strip of a material called Niobium Nitride (NbN) and cooled it down to near absolute zero so it became a superconductor. They ran a steady electric current through it.
The Flashlight: They shined a beam of Terahertz light (a type of invisible light) onto a tiny gold-coated tip. This tip acted like a lens, squeezing the light into a tiny "hotspot" just a few hundred nanometers wide.
The Scan: They dragged this tip across the surface of the superconductor.
The Reaction: When the tip hovered over a perfect part of the superconductor, nothing much happened. But when the tip hovered over a "weak spot" (a defect), the light broke the superconducting bonds in that tiny area. Because the material was already carrying a current, this tiny break caused a sudden, measurable spike in voltage.

The Analogy:
Imagine walking across a frozen lake (the superconductor) while carrying a heavy load (the electric current). Most of the ice is thick and safe. But there are a few thin, weak patches of ice (the defects).

If you just walk normally, you might not notice the thin ice until you fall through.
In this experiment, the researchers used a tiny, heated probe (the light tip) to gently tap the ice as they walked.
When the probe hit a thin patch, the ice cracked slightly, causing a "shiver" (a voltage spike) that the researchers could feel. By mapping where these shivers happened, they could draw a map of the weak ice patches without ever falling through.

What They Found

Tiny Defects: They found specific spots in the middle of the material where the superconductivity was weaker. These spots were about 300 nanometers wide (roughly the width of a virus).
The "Sweet Spot" Current: They discovered that these weak spots only showed up clearly when the electric current was at a specific level. If the current was too low, the defects were invisible. If the current was too high, the whole strip became "normal" (lost its superpowers) and the specific signal disappeared. There was a "Goldilocks" zone where the defects were most visible.
Vortex Dance: The paper explains that the voltage spikes happen because of "vortices." Think of these as tiny tornadoes of electricity that form when the current gets too strong. The weak spots make it easier for these tornadoes to form and spin. The light from the tip helps kick-start these tornadoes, and the researchers measured the energy released when they spun.

Why This Matters (According to the Paper)

This technique is a new way to "see" the invisible.

Resolution: It sees details 400 times smaller than what standard light microscopes can see.
Non-Destructive: Unlike other methods that use high-energy light (which can melt or damage the delicate superconducting state), this method uses low-energy light that is gentle enough to study the material without destroying it.
Direct Link: It directly connects the physical "potholes" in the material to how they behave when electricity flows through them.

Summary

The researchers developed a new way to use a tiny, focused beam of light to find microscopic weak spots in superconductors. By scanning the material while it carries an electric current, they could map out exactly where the material is imperfect. This helps scientists understand why superconducting devices sometimes fail and how to build better ones by fixing these tiny "potholes" in the road.

Technical Summary: Imaging Superconducting Weak Spots through Vortex-Assisted THz Near-Field Photovoltage

Problem Statement
Nanoscale inhomogeneities, such as compositional variations, grain boundaries, and stoichiometric defects, are defining features of many superconducting materials. These defects create "weak spots" where the superconducting order parameter is suppressed, leading to preferential vortex nucleation and dissipation under current bias. These local variations critically impact device performance, including the detection efficiency of superconducting nanowire single-photon detectors and the coherence of transmon qubits. Furthermore, in strongly inhomogeneous quantum materials like high- $T_c$ superconductors and moiré systems, spatial variations of superconductivity are intrinsic to the electronic ground state.

The central experimental challenge is probing these inhomogeneities at the relevant energy and length scales. The relevant energy scale for superconducting electrodynamics lies in the millielectronvolt (meV) range (sub-terahertz to terahertz frequencies). However, the wavelengths at these frequencies (hundreds of microns) are macroscopic compared to the nanoscopic disorder landscape (10–1000 nm). This energy-lengthscale mismatch prevents the direct spatial resolution of local THz responses from individual weak spots using conventional far-field spectroscopy.

Methodology
To resolve this mismatch, the authors employ THz near-field photovoltage nanoscopy on a current-biased Niobium Nitride (NbN) strip. The experimental setup combines three key elements:

Nanoscale Confinement: A gold-coated Atomic Force Microscopy (AFM) tip (radius $\sim$ 250 nm) acts as a nanoantenna, confining 2.52 THz radiation to a nanoscale hotspot. This achieves a spatial resolution of $\sim$ 300 nm, nearly 400 times better than the far-field diffraction limit ( $\sim$ 100 $\mu$ m).
Energy Scale Matching: The 2.52 THz photon energy ( $\hbar\omega \approx 10.4$ meV) exceeds the NbN superconducting gap ( $2\Delta \approx 5.8$ meV). Consequently, the near-field acts as a local pair-breaking perturbation, suppressing the order parameter and creating a nanoscale "hotspot" without globally destroying superconductivity.
Electrical Readout: A DC bias current ( $I_b$ ) is applied to the strip. The local suppression of superconductivity by the THz hotspot interacts with the current-induced vortex state, generating a measurable global photovoltage ( $V_{PV}$ ) detected at the tip's tapping frequency.

The study utilizes NbN films (320 nm thick) with granular morphology on Si/SiO $_2$ substrates. Measurements are performed across a range of temperatures ( $T \approx 13.8$ K) and bias currents, comparing experimental data with Time-Dependent Ginzburg-Landau (TDGL) simulations.

Key Results

Spatial Mapping of Defects: Scanning the THz tip across the NbN strip reveals reproducible, localized peaks in the photovoltage signal within the bulk of the superconductor. These peaks are fixed in sample coordinates, indicating they arise from specific nanoscopic defects rather than noise.
Defect Characteristics: The localized peaks exhibit a spatial width (FWHM) of approximately 300 nm, consistent with the tip radius and the size of Nb-depleted regions observed via Scanning Electron Microscopy (SEM) and Energy-Dispersive X-ray spectroscopy.
Current Dependence: The amplitude of the photovoltage peaks follows a distinct non-monotonic dependence on the bias current. The signal rises near the critical current ( $I_c$ ), reaches a maximum slightly below the flux-flow instability (marked by the peak in differential resistance $dV/dI$), and then decreases as the strip transitions toward the normal state.
Edge vs. Bulk Response: Enhanced photovoltage is also observed at the strip edges. The edge signal onset aligns with $I_c$ , and its evolution mirrors the bulk defect response, suggesting a common physical origin related to vortex dynamics.
Simulation Validation: TDGL simulations of a current-carrying strip with a defect and a localized hotspot reproduce the experimental features. The simulations confirm that the THz hotspot suppresses the local order parameter, leading to current crowding and the enhancement of vortex-antivortex (V-AV) pair nucleation.

Mechanism: Vortex-Assisted Photovoltage
The authors attribute the observed signal to a vortex-assisted dissipation mechanism:

The THz hotspot locally suppresses the superconducting order parameter.
This suppression forces supercurrents to bend around the hotspot, creating local current crowding.
The enhanced current density lowers the energy barrier for vortex entry at edges and reduces the binding energy for V-AV pairs in the bulk.
Under bias current, these unbound V-AV pairs traverse the sample, generating a voltage via the Josephson relation.
Defects with reduced superfluid density further enhance this effect by providing a pre-existing region of weak superconductivity where V-AV pairs are more easily nucleated.

Significance and Claims
The paper establishes THz near-field photovoltage nanoscopy as a functional probe for superconducting weak spots under current bias. Its primary significance lies in:

Bridging the Scale Gap: It successfully combines THz-frequency excitation (matching the superconducting energy scale) with nanometric spatial resolution, overcoming the diffraction limit that previously prevented local access to THz responses.
Non-Destructive Probing: Unlike Low-Temperature Laser Scanning Microscopy (LTLSM) which uses visible photons (high energy, disruptive), the 2.52 THz radiation is less disruptive, allowing the study of the vortex-assisted photoresponse without globally destroying the superconducting state.
Sensitivity to Vortex Regimes: The technique is sensitive to the transition between different vortex regimes (e.g., the flux-flow instability), offering a local probe complementary to global transport measurements.
Future Applicability: The authors claim this method opens a direct route to probing how material inhomogeneities influence light-matter interactions in superconductors, with specific implications for understanding strongly inhomogeneous systems such as high- $T_c$ materials and moiré systems where nanoscale phase separation is intrinsic.

The study does not claim to resolve individual atomic defects but rather identifies mesoscopic regions (150–700 nm) of reduced superfluid density that act as nucleation sites for vortex-antivortex pairs, linking nanoscale inhomogeneity directly to light-driven dissipation.