🔬 mesoscale physics

Scanning Gate Microscopy Modulation of Supercurrent in Graphene Josephson Junctions

This study utilizes Scanning Gate Microscopy to quantitatively characterize and map the spatial modulation of supercurrent in hBN-encapsulated graphene Josephson junctions, demonstrating strong agreement with numerical simulations and establishing a pathway for high-resolution local manipulation of gate-tunable superconducting phenomena.

Original authors: Ivan Villani, Samuele Fracassi, Niccolò Traverso Ziani, Maura Sassetti, Matteo Carrega, Vaidotas Miseikis, Camilla Coletti, Fabio Beltram, Kenji Watanabe, Takashi Taniguchi, Sergio Pezzini, Stefan Heu

Published 2026-07-30

📖 3 min read☕ Coffee break read

CC BY 4.0

Original authors: Ivan Villani, Samuele Fracassi, Niccolò Traverso Ziani, Maura Sassetti, Matteo Carrega, Vaidotas Miseikis, Camilla Coletti, Fabio Beltram, Kenji Watanabe, Takashi Taniguchi, Sergio Pezzini, Stefan Heun

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

Imagine the world of electricity as a bustling city. Usually, when electrons (the city's commuters) travel through a wire, they bump into atoms, lose energy, and generate heat—like a crowded subway car where everyone is jostling. But in a special state called superconductivity, these electrons pair up and glide together without any friction at all, moving in perfect harmony. When you sandwich a normal material between two superconductors, these frictionless pairs can "tunnel" across the gap, creating a supercurrent. This setup is called a Josephson Junction, and it's the heartbeat of many quantum technologies, from ultra-sensitive sensors to the building blocks of future quantum computers.

However, for a long time, scientists could only see the "average" behavior of these electrons, like looking at a city from a satellite and seeing only the total traffic flow, not the individual cars. They knew the junction worked, but they couldn't see where the electrons were flowing or how they were moving inside. To fix this, researchers use a tool called Scanning Gate Microscopy (SGM). Think of this as a magical, invisible finger that can hover over the circuit and gently push or pull on the electrons with an electric field, allowing scientists to map out the traffic patterns with incredible detail. The big question was: could this "magic finger" actually control the supercurrent in these delicate, high-tech devices, or would the quantum nature of the electrons make them too stubborn to budge?

In this study, a team of researchers decided to test this idea using a special kind of superconductor made from graphene—a material as thin as a single sheet of paper but incredibly strong and fast at conducting electricity. They built tiny bridges (Josephson junctions) using graphene, sandwiched it between protective layers of hexagonal Boron Nitride (hBN), and connected it to Niobium leads. Then, they brought in their Scanning Gate Microscope, which uses a sharp, electrically charged tip to act as a movable, local gate.

The team discovered that their "magic finger" worked better than expected. By simply moving the tip over the graphene and changing its voltage, they could locally tune the supercurrent. When they applied a negative voltage to the tip, it pushed electrons away, reducing the supercurrent. When they applied a positive voltage, it pulled electrons in, boosting the supercurrent. The effect was surprisingly strong: they could change the current by hundreds of nanoamperes (nA), which is a massive amount for such a tiny device. They found that the closer the tip was to the graphene (down to just 100 nanometers away), the stronger the effect.

To make sure they weren't just seeing a lucky fluke, the researchers ran detailed computer simulations. These digital models, which treated the tip's electric field like a smooth, bell-shaped curve, matched their real-world experiments almost perfectly. This confirmed that the changes they saw were indeed caused by the tip's electric field reshaping the flow of electrons. They also noticed something interesting: the supercurrent didn't change evenly across the whole bridge. Instead, it was stronger on one side than the other, suggesting that the "landscape" of the graphene wasn't perfectly flat, but had tiny, invisible bumps and dips that the tip could detect.

This work proves that graphene Josephson junctions are the perfect playground for this kind of high-resolution mapping. It shows that we can now not only see where the supercurrent flows but also actively manipulate it with a moving tip, opening the door to understanding and engineering these quantum devices with a level of detail that was previously impossible.

Technical Summary: Scanning Gate Microscopy Modulation of Supercurrent in Graphene Josephson Junctions

Problem Statement
Graphene Josephson junctions (GJJs) encapsulated in hexagonal Boron Nitride (hBN) offer a promising platform for quantum technologies due to their high carrier mobility, ballistic transport, and gate-tunable critical currents. However, traditional characterization relies on electrical transport measurements, which are spatially averaging techniques that treat the device as a "black box." This approach obscures local transport properties and spatial variations essential for understanding mechanisms in novel platforms like GJJs. While Scanning Gate Microscopy (SGM) has been successfully used to map electron flow in normal-state 2D systems, its application to superconducting devices is limited. Experimental challenges include the difficulty of recording full current-voltage (I-V) curves at every tip position and the typically small modulation signals observed in previous superconducting SGM studies (e.g., in Aluminum islands or InSb nanoflags).

Methodology
The authors employed Scanning Gate Microscopy to investigate supercurrent transport in hBN-encapsulated GJJs contacted by Niobium (Nb) leads. The experimental setup involved:

Sample Fabrication: Graphene single crystals were encapsulated between two 30 nm hBN flakes and contacted by Nb edge-contacts via DC magnetron sputtering. The Nb contacts were engineered to match the stack height to minimize topographic steps. Two junctions (J1 and J2) with varying dimensions were studied.
Measurement Environment: Experiments were conducted in a Janis cryostat at a base temperature of 300 mK. The sample was mounted on an AFM head, and a conductive tip was used as a movable, non-contact local gate.
SGM Protocol: The tip was scanned over the junction at fixed distances ( $d_{vac}$ ) while applying a bias voltage ( $V_{tip}$ ). At each pixel, a full $V-I_{bias}$ curve was measured to extract the switching current ( $I_{sw}$ ).
Parameters: The study varied $V_{tip}$ from -10 V to +10 V and the tip-to-sample distance from 100 nm to 970 nm. Measurements were performed in both electron and hole doping regimes, with a primary focus on the electron doping regime ( $V_{BG} = -3$ V) where the switching current is maximized and linearly dependent on the back-gate voltage.
Modeling: The experimental data were compared against numerical simulations using a coarse-grained tight-binding Bogoliubov-de Gennes (BdG) model and an analytical model describing the tip-induced carrier density variation.

Key Contributions and Results

Large Tip-Induced Modulation: The authors achieved a significant tip-induced modulation of the switching current, reaching hundreds of nA. This magnitude is two orders of magnitude larger than previous SGM results on superconducting devices (e.g., Ref. [61]). This enhancement is attributed to the large switching currents of GJJs, the high dielectric constant of the top hBN layer (increasing capacitive coupling), and the reduced working distance (100 nm) enabled by edge contacts.
Quantitative Characterization:
- Voltage Dependence: The switching current was suppressed by negative $V_{tip}$ and enhanced by positive $V_{tip}$ , consistent with the electron doping regime. The tip modulation efficiency ( $\beta_{tip,n}$ ) was measured at $18.8 \pm 0.8$ nA/V, corresponding to approximately 16% of the back-gate modulation efficiency ( $\beta_{BG,n}$ ).
- Distance Dependence: The modulation decays rapidly as the tip-to-sample distance increases. The experimental data for $I_{sw}(d)$ were successfully fitted using an analytical model derived from the integration of a Lorentzian-like tip potential over the junction area, yielding the relationship $I_c(d_{vac}) = A \text{arcsinh}(L / 2(d_{vac} + \delta)) + I_{c,0}$ .
- Spatial Mapping: SGM maps revealed that the maximum modulation does not occur strictly at the junction center but exhibits asymmetry. This suggests a non-uniform electrostatic potential landscape within the junction, likely due to residual disorder or interface inhomogeneities.
Theoretical Validation: Numerical simulations using a coarse-grained method reproduced the experimental SGM maps and the dependence of critical current on $V_{tip}$ with excellent quantitative agreement ( $\beta_{simul,n} = 18.0 \pm 1.4$ nA/V). The analytical model accurately described the distance dependence of the modulation.

Significance
The paper establishes GJJs as an ideal platform for the spatial investigation of supercurrent transport using SGM. The work demonstrates that the combination of high-quality hBN-encapsulated graphene and optimized device geometry allows for unprecedented spatial resolution in mapping superconducting phenomena. The authors claim that this robust foundation enables the local mapping and manipulation of gate-tunable superconducting phenomena. Specifically, the technique is positioned as a tool to:

Investigate the formation of Josephson vortices.
Image superconducting pockets in the semiclassical regime.
Spatially resolve supercurrent transport in the quantum Hall regime (mediated by edge states or bulk percolative channels), potentially in combination with finite magnetic fields.

The study confirms that SGM can access local transport properties previously inaccessible to conventional "black box" characterization, providing a pathway to understand and engineer spatial variations in superconducting devices.

More like this