🔬 mesoscale physics

Dynamically tunable hydrodynamic transport in boron nitride-encapsulated graphene

This study demonstrates a reversible method to dynamically tune charge transport in boron nitride-encapsulated graphene from a viscous hydrodynamic regime to an impurity-dominated regime at room temperature by using ultraviolet radiation to continuously modulate disorder levels in the hBN dielectric.

Original authors: Akash Gugnani, Aniket Majumdar, Kenji Watanabe, Takashi Taniguchi, Arindam Ghosh

Published 2026-07-09

📖 5 min read🧠 Deep dive

CC BY 4.0

Original authors: Akash Gugnani, Aniket Majumdar, Kenji Watanabe, Takashi Taniguchi, Arindam Ghosh

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 a superhighway made of a single layer of carbon atoms, thinner than a human hair, where electrons don't just zip along like cars on a road. Instead, they flow like a thick, sticky fluid—bumping into each other, swirling, and moving in perfect sync. This is electron hydrodynamics, a state where the electrons act more like water in a river than individual particles.

For years, scientists have been able to build these "super-clean" highways using graphene sandwiched between layers of hexagonal boron nitride (hBN). But there was a catch: once the highway was built, the amount of "traffic chaos" (disorder) was fixed. You couldn't easily switch the flow from a smooth, viscous river to a chaotic, bumpy road and back again without breaking the device.

The Big Discovery: A Remote Control for Chaos
In this work, researchers at the Indian Institute of Science and collaborators in Japan found a way to act as a "remote control" for this chaos. They discovered that by shining ultraviolet (UV) light on their graphene device while applying a specific electric voltage, they could dynamically tune how messy the electron flow gets.

Think of the boron nitride layers as a sponge holding invisible electric charges. When they shine UV light (with a wavelength of 235 nm) on the device, it wakes up "traps" in the sponge. Depending on the electric voltage they apply (called the photodoping voltage, or VPD), these charges either jump onto the graphene highway or get stuck in the sponge.

The Magic: They can increase the disorder (make the road bumpier) or decrease it (make it smoother) reversibly. They can turn the device into a messy, disordered system and then, just by changing the light and voltage, wipe the slate clean and return it to its pristine, smooth state. They did this over 15 consecutive cycles without breaking the device.

The "Traffic Jam" Test: The Wiedemann-Franz Law
How do you know if the electrons are flowing like a fluid or just bouncing around randomly? The scientists used a famous rule called the Wiedemann-Franz (WF) Law.

The Analogy: Imagine a crowd of people. If they are just walking randomly (a normal, "diffusive" state), the way they carry heat and the way they carry electricity are locked together in a specific ratio.
The Fluid State: But if they start moving like a fluid (hydrodynamic), they get stuck in a traffic jam where they bump into each other constantly. In this state, the ratio of heat to electricity breaks the rule. The electrons carry heat much better than the rule predicts.

The researchers measured this ratio (called the Lorentz number, L) and found that in their clean devices, the rule was indeed broken, proving the electrons were flowing like a fluid. But here is the kicker: as they used UV light to add more disorder (more "bumps" on the road), the ratio slowly crept back toward the normal rule.

The Result: By increasing the disorder, they could watch the electron flow transition from a "viscous fluid" (where the WF law is broken) back to a "diffusive" state (where the WF law is restored). They observed this change happening reversibly at room temperature (300 K).

What They Ruled Out
The paper is very careful about what this technique doesn't do.

It's not permanent damage: Some methods of adding disorder (like shooting helium ions at the material) permanently break the sample. This UV method is reversible; the device can be "reset" to its original, pristine condition.
It's not just a shift in voltage: The UV light doesn't just move the electrons around; it actually creates new "trap states" (places where charges get stuck) in the boron nitride. This is different from simple photogating where charges just move temporarily.
It's not a tiny effect: The disorder they introduced was significant. They could increase the charge inhomogeneity (the "bumpiness") by nearly 3 × 10¹¹ cm⁻², which is a massive change for a single layer of atoms.

The Numbers and the Certainty
The team didn't just guess; they measured everything.

They used UV radiation with a power density of approximately 1 pW/µm².
They applied a VPD ranging from 0 V down to -70 V.
They observed that the momentum-relaxing scattering rate (how often electrons hit a bump) increased by nearly a factor of ten as they added disorder.
They confirmed that the "fluid" behavior was real by measuring the electronic thermal conductivity using a technique called Johnson noise thermometry.
They found that the "shear viscosity" (the stickiness of the electron fluid) in the clean state was about 5000ℏ per carrier, which matches theoretical predictions for a clean fluid.

What They Don't Know Yet
While they proved they can tune the disorder and see the fluid-to-bumpy transition, the paper notes that the exact mathematical relationship between the disorder and the "stickiness" (viscosity) in the messy state is still a bit of a mystery. They suggest that the disorder enhances viscous effects, but the precise mechanism is not fully defined yet. Also, while they saw the fluid behavior in both electron-rich and hole-rich regions, the hole-rich side was a bit trickier to analyze due to junctions forming at the edges, so they focused their main conclusions on the electron-rich side.

The Bottom Line
This paper shows that we can now take a high-quality graphene device, turn it into a chaotic mess with a flash of UV light, watch the electrons stop flowing like a fluid and start bouncing like billiard balls, and then clean it all up to do it again. It's a powerful new tool to study how electrons behave when they are forced to interact, right at room temperature, without destroying the device in the process.

Technical Summary: Dynamically Tunable Hydrodynamic Transport in Boron Nitride-Encapsulated Graphene

Problem Statement
Over the past decade, high-quality hexagonal boron nitride (hBN)-encapsulated graphene devices have enabled the exploration of viscous electron flow (electron hydrodynamics). In this regime, electron-electron scattering dominates over momentum-relaxing scattering (e.g., electron-impurity or electron-phonon scattering). However, a critical limitation in existing state-of-the-art devices is the lack of tunability regarding disorder levels. The intrinsic disorder concentration ( $n_{min}(0)$ ) is fixed by fabrication protocols, making it impossible to study the crossover between impurity-dominated (diffusive) and viscous transport regimes within the same device. Previous attempts to introduce disorder, such as helium ion irradiation, resulted in permanent sample degradation, preventing reversible studies of the hydrodynamic-to-diffusive transition.

Methodology
The authors utilized high-quality monolayer graphene field-effect transistors (FETs) encapsulated within multi-layered hBN flakes. To achieve dynamic and reversible control over disorder, they employed a combination of ultraviolet (UV) radiation (wavelength $\lambda = 235$ nm) and a gate electric field (photo-doping voltage, $V_{PD}$ ).

The experimental mechanism relies on three distinct processes induced by UV illumination under an applied $V_{PD}$ :

Interlayer Charge Transfer (Photogating): UV photons generate electron-hole pairs in the hBN bandgap. The applied electric field drives electrons (or holes) from hBN to the graphene channel, shifting the charge neutrality point (CNP).
Random Charge Trapping: UV radiation activates shallow and deep trap states at the hBN-SiO $_2$ interface and within the hBN layer, creating charge puddles and increasing charge inhomogeneity ( $n_{min}$ ).
Reversibility: By applying UV radiation with $V_{PD} = 0$ V, the system undergoes extrinsic hole doping that neutralizes the previously induced electron doping, effectively "resetting" the disorder level to the pristine state.

Electrical transport was characterized via four-terminal resistance measurements to determine conductivity ( $\sigma$ ) and charge inhomogeneity. Thermal transport was probed using Johnson noise thermometry to measure the electronic temperature ( $T_e$ ) and extract the electronic thermal conductivity ( $\kappa_e$ ). The deviation from the Wiedemann-Franz (WF) law, quantified by the Lorentz number $L = \kappa_e / (\sigma T)$ , served as the primary metric for distinguishing between hydrodynamic and diffusive regimes.

Key Results

Reversible Disorder Tuning: The study demonstrated that UV irradiation combined with $V_{PD}$ can continuously and reversibly tune the disorder level. Increasing the magnitude of negative $V_{PD}$ during UV exposure shifted the CNP and increased the total charge inhomogeneity ( $n_{min}$ ) by up to $\approx 3 \times 10^{11}$ cm $^{-2}$ , reducing carrier mobility by nearly an order of magnitude.
Scattering Rate Dynamics: The momentum-relaxing scattering rate ( $\tau_{mr}^{-1}$ ) was observed to increase significantly with disorder, while the electron-electron scattering rate ( $\tau_{ee}^{-1}$ ) remained largely determined by temperature. Consequently, the ratio $\tau_{mr}^{-1} / \tau_{ee}^{-1}$ increased, driving the system from the hydrodynamic limit ( $\tau_{mr}^{-1} < \tau_{ee}^{-1}$ ) toward the diffusive limit ( $\tau_{mr}^{-1} > \tau_{ee}^{-1}$ ).
Restoration of the Wiedemann-Franz Law: In pristine (low-disorder) graphene, the Lorentz number $L$ exhibited a strong violation of the WF law ( $L/L_{WF} \gg 1$ ) near the charge neutrality point, a signature of viscous electron flow. As disorder was introduced, the width of the Lorentzian profile of $L/L_{WF}$ broadened, and the value approached the classical WF limit ( $L/L_{WF} \approx 1$ ). This transition was fully reversible; resetting the device to the pristine condition restored the WF law violation.
Viscosity and Transport Regimes:
- Fermi Liquid Regime ( $n \gtrsim 10^{11}$ cm $^{-2}$ ): The shear viscosity per carrier ( $\eta_{FL}/n$ ) showed a linear dependence on the scattering ratio $\tau_{mr}^{-1} / \tau_{ee}^{-1}$ in the diffusive limit, consistent with theoretical predictions for electron-impurity scattering.
- Thermal Regime (near CNP): The thermal shear viscosity ( $\eta_{th}$ ) remained constant ( $\approx 6 \times 10^{-18}$ J m $^{-2}$ s) despite increasing disorder. This suggests that the quantum critical behavior of the Dirac fluid near the CNP is robust against the specific type of disorder introduced by UV radiation.
- Universal Conductivity: The quantum conductivity $\sigma_Q$ saturated near a universal constant of $(4 \pm 1) e^2/h$ as $T_F/T \to 0$ , regardless of the disorder density, further confirming the robustness of the quantum critical flow.

Significance and Claims
The paper claims to have established a non-invasive, reversible tool for dynamically tuning the disorder level in high-quality graphene devices. This capability allows for the systematic exploration of the interplay between momentum-conserving and momentum-relaxing collisions within a single device at room temperature.

The authors assert that their findings:

Provide a method to reversibly switch a graphene device between hydrodynamic and diffusive transport regimes.
Offer new insights into the impact of charged disorder on electronic viscosity in both Dirac and Fermi liquid phases.
Serve as a testing bed for theoretical models of relativistic hydrodynamics, specifically validating the conditions under which the WF law is violated or restored.
Demonstrate that while disorder restores the WF law in the Fermi liquid regime, the quantum critical thermal transport near the Dirac point remains robust against the introduced disorder.

The work does not propose immediate commercial applications but positions the technique as a fundamental experimental advance for understanding electron hydrodynamics and engineering charge flow mechanisms in 2D materials.

More like this