Imaging flat band electron hydrodynamics in biased bilayer graphene

Using a scanning superconducting magnetic sensor to image local current flow in dual-gated bilayer graphene, researchers identified distinct ballistic, hydrodynamic, and diffusive transport regimes, demonstrating that the flat band regime with its large effective mass enables ultra-short electron-electron scattering lengths (~50 nm) and strong nonlinearities, thereby paving the way for miniaturized electronic devices based on electron hydrodynamics.

The Gist

Imagine a crowded dance floor. Usually, when people move through a crowd, they bump into the walls, the furniture, or random strangers, causing them to scatter in all directions. This is how electricity usually works in most materials: a chaotic, messy shuffle called diffusive transport.

But sometimes, if the crowd is very polite and the music is just right, the dancers start moving together like a fluid. They flow around obstacles in smooth, swirling patterns, almost like water in a river. This is electron hydrodynamics, and in this new study, scientists have finally found a way to make electrons behave like this fluid on a tiny, manageable scale.

Here is the story of how they did it, using simple analogies:

1. The Problem: The "Too Fast" Crowd

In standard graphene (a single layer of carbon atoms), electrons are like tiny, super-fast race cars. They are so light and fast that they zoom past each other without really interacting. To get them to act like a fluid, you usually have to cool them down to near absolute zero and make the device huge (about the size of a human hair). This is great for science, but terrible for building tiny, powerful computer chips. You can't shrink a river down to a puddle if the water is moving too fast.

2. The Solution: The "Heavy" Crowd

The researchers used a special sandwich of materials: two layers of graphene stacked on top of each other, with a "gate" on the top and bottom (like a pair of hands squeezing a sponge).

By applying a specific electric "squeeze" (called a displacement field), they changed the rules of the game. They made the electrons act as if they were heavy.

• The Analogy: Imagine switching the dancers from being light-footed sprinters to being people wearing heavy, lead-filled boots.
• The Result: Because the electrons are now "heavy" (they have a large effective mass), they move slower and bump into each other much more often. Instead of racing past one another, they start colliding and pushing each other, forcing them to move as a coordinated team. This creates a "flat band" where the electrons are sluggish and sticky, perfect for fluid-like behavior.

3. The Camera: Seeing the Invisible Flow

You can't see electrons with a normal camera. To watch this invisible fluid, the team used a Superconducting Quantum Interference Device (SQUID) on the tip of a needle.

• The Analogy: Think of this as a super-sensitive "magnetic nose." As the electric current flows through the graphene, it creates a tiny magnetic field (like the wake behind a boat). The SQUID tip hovers just above the surface, sniffing out these magnetic fields to map exactly where the electrons are going.

4. The Three Modes of Traffic

By adjusting the "squeeze" and the number of electrons, they discovered three distinct ways the traffic moved:

• The Ballistic Mode (The Race): At high speeds and low density, electrons zoom straight through without stopping. It's like a highway with no traffic lights. They hit the walls and bounce off, creating chaotic loops.
• The Diffusive Mode (The Shuffle): At the "neutral" point, electrons are confused. They bump into everything and scatter randomly. It's like a crowded hallway where everyone is trying to get to the exit but keeps bumping into others.
• The Hydrodynamic Mode (The River): This is the "sweet spot" found in the heavy-electron regime.
  • In the Channel: The electrons flow faster in the middle and slower near the edges, just like water in a pipe (this is called Poiseuille flow).
  • In the Chambers: When the flow hits a wider area, it doesn't just stop; it spins! The researchers saw vortices (whirlpools) forming, just like water swirling down a drain. This is the "smoking gun" that proves the electrons are acting like a fluid.

5. Why This Matters

This discovery is a game-changer for the future of electronics.

• Miniaturization: Because the electrons are "heavy," they can form these fluid patterns in devices that are incredibly small (about 50 nanometers). This is 10 times smaller than previous experiments. It means we might one day build computer chips that use fluid-like electron flow instead of messy, resistive shuffling.
• Nonlinear Magic: The team also cranked up the current and saw the flow get weird and nonlinear. The whirlpools started to deform and move, suggesting that these "electron fluids" could be used to create new types of logic gates or sensors that react in complex, interesting ways.

The Bottom Line

The researchers took a material that usually acts like a chaotic gas of particles, used an electric "squeeze" to make the particles act heavy and sluggish, and watched them turn into a beautiful, swirling fluid. They proved that with the right conditions, electrons can flow like water, opening the door to a new generation of tiny, efficient, and powerful electronic devices.

Technical Summary

Here is a detailed technical summary of the paper "Imaging flat band electron hydrodynamics in biased bilayer graphene."

1. Problem Statement and Motivation

• The Challenge of Miniaturization: Electron hydrodynamics (where carrier kinetics are dominated by inter-electron collisions rather than momentum relaxation) has been observed in graphene. However, previous experiments relied on light effective masses ($m^*$), resulting in large electron-electron scattering lengths ($\ell_{ee} \gtrsim 250$ nm). This restricts hydrodynamic phenomena to large device scales, hindering miniaturization.
• The Theoretical Opportunity: In rhombohedral (ABC-stacked) multilayer graphene, specifically bilayer graphene (R2G), the effective mass can be tuned by over an order of magnitude using a dual-gate displacement field ($D$). In the "flat band" regime (large $D$), the effective mass increases significantly, theoretically reducing $\ell_{ee}$ to the scale of the Fermi wavelength ($\sim 50$ nm).
• The Scientific Gap: While bulk transport measurements in R2G have shown negative temperature coefficients (a signature of the Gurzhi effect), it remains unclear if this arises from true hydrodynamic flow or other mechanisms like magnetic fluctuations. Furthermore, distinguishing between ballistic, hydrodynamic, and diffusive regimes at the nanoscale in these complex phase diagrams has been difficult.

2. Methodology

• Sample Design: The authors fabricated a dual-gated R2G device encapsulated in hexagonal boron nitride (hBN) with graphite gates. The device features a unique geometry: a central channel ($w \approx 1.15$ µm) connected to two lateral chambers via constrictions ($d \approx 1.0$ µm). This geometry is sensitive to both laminar flow (channel) and vortical flow (chambers).
• Imaging Technique: The team employed a scanning superconducting quantum interference device on a tip (nSOT). This sensor maps the fringe magnetic field ($B$) above the device with nanoscale resolution.
• Data Reconstruction: Using Fourier-domain inversion algorithms and Tikhonov regularization, the measured magnetic fields were reconstructed into 2D current density maps ($J_x, J_y$) and streamlines.
• Theoretical Modeling: Experimental data was fitted against a phenomenological Boltzmann transport model. The model uses a two-rate relaxation time approximation with two key parameters:
  • $\ell_{ee}$: Electron-electron scattering length (momentum-conserving).
  • $\ell_{mr}$: Momentum-relaxing scattering length (phonons, impurities, boundaries).
  • The model distinguishes regimes based on the relative magnitudes of $\ell_{ee}$, $\ell_{mr}$, and device size $L$.

3. Key Contributions

1. Direct Nanoscale Imaging: First direct visualization of local current flow in dual-gated bilayer graphene, resolving the transition between ballistic, hydrodynamic, and diffusive regimes across the full phase space of carrier density ($n_e$) and displacement field ($D$).
2. Discovery of Deep Hydrodynamics: Identification of a "flat band" regime where $\ell_{ee}$ is reduced to $\sim 50$ nm (comparable to the Fermi wavelength), enabling hydrodynamic transport at length scales previously thought impossible for electron fluids.
3. Phase Diagram Mapping: Construction of a comprehensive transport regime phase diagram, defining the boundaries of ballistic, hydrodynamic, and diffusive transport based on scattering lengths.
4. Nonlinear Transport Characterization: Observation of current-driven nonlinearities, including the transition from ballistic to viscous flow via Joule heating and the breakdown of linear hydrodynamic models at high currents.

4. Key Results

• Three Distinct Transport Regimes:
  • Diffusive: Observed near the Charge Neutrality Point (CNP) at low $D$. Characterized by flat current profiles and no vortices. $\ell_{mr}$ is short ($\sim$ hundreds of nm).
  • Ballistic: Observed at high carrier densities ($|n_e|$) and low $D$. Characterized by strong vortical flow in lateral chambers (whirlpools) and flat channel profiles. $\ell_{ee}$ and $\ell_{mr}$ are both large ($\gg L$).
  • Hydrodynamic (Viscous): Observed at low $|n_e|$ and high $D$ (the flat band regime).
    • Channel: Shows a Poiseuille flow profile (parabolic, with current concentrated in the center, $\sim 30\%$ higher than average).
    • Chambers: Vortices are present but weaker and shifted compared to the ballistic regime.
    • Scattering Lengths: Fitting reveals $\ell_{mr} \gg L$ (momentum conserved over device size) and $\ell_{ee} \approx 29$ nm. This confirms the system is deep in the hydrodynamic regime.
• Validation of the Gurzhi Effect: The observation of Poiseuille flow and the specific dependence of resistance on viscosity confirms that the negative temperature coefficient ($dR/dT < 0$) seen in bulk R2G measurements is indeed due to electron hydrodynamics, not just magnetic fluctuations.
• Nonlinear Effects (High Current):
  • Increasing current ($I_0$) in the ballistic regime induces a crossover to viscous flow (Joule heating reduces $\ell_{ee}$).
  • In the flat band regime, extremely high currents ($> 28$ µA) cause a breakdown of the linear Boltzmann model. Vortices deform into crescent shapes and shift off the central axis, suggesting complex nonlinear hydrodynamics or spatially varying chemical potentials not captured by the simple model.

5. Significance and Implications

• Miniaturization of Hydrodynamic Devices: By tuning the effective mass via electric fields, the authors demonstrate that electron hydrodynamics can be achieved on length scales of $\sim 50$ nm. This opens the door to creating nanoscale electronic devices based on viscous electron flow, such as low-dissipation interconnects or hydrodynamic logic gates.
• New State of Matter: The study suggests that biased bilayer graphene in the flat band regime exists in a "semiquantum liquid" state where electron-electron scattering and isospin correlations are as short-range as possible, dominated by configurational changes rather than coherent quantum states.
• Future Directions: The work paves the way for exploring extreme hydrodynamic phenomena, such as turbulence (Reynolds number engineering) and anisotropic viscosity, in higher-order rhombohedral multilayers and custom geometries. It also provides a robust framework for distinguishing correlated electron phases from hydrodynamic flow in 2D materials.