🔬 mesoscale physics

Electron Hydrodynamics and Bernoulli Effect in Venturi-Shaped 2D Systems

This paper reports the direct experimental observation of a strongly nonlinear transport regime in a high-mobility two-dimensional electron system, where a Venturi-shaped device geometry reveals a pronounced voltage response and diodicity consistent with the convective acceleration of an electron fluid described by the Bernoulli effect.

Original authors: C. A. Monari, A. D. Levin, A. S. Jaroshevich, Z. D. Kvon, V. A. Chitta, D. V. Dmitriev, A. K. Bakarov, G. M. Gusev

Published 2026-07-28

📖 8 min read🧠 Deep dive

CC BY 4.0

Original authors: C. A. Monari, A. D. Levin, A. S. Jaroshevich, Z. D. Kvon, V. A. Chitta, D. V. Dmitriev, A. K. Bakarov, G. M. Gusev

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 world where tiny particles, usually known for bouncing off each other like chaotic pinballs, suddenly decide to move in perfect unison, flowing like a thick, sticky river. This is the realm of electron hydrodynamics, a fascinating corner of physics where electrons in ultra-clean materials stop acting like individual soldiers and start behaving like a collective fluid. In this fluid state, the electrons interact so strongly with one another that they forget about the messy bumps of the material they travel through and instead flow together, creating patterns and effects that look just like water swirling in a pipe or wind rushing through a canyon. Scientists have long been able to see these "fluid-like" behaviors when the flow is gentle and steady, but they have struggled to catch the electrons doing anything wild or chaotic. The big question has been: Can we make these electron rivers speed up enough to show us the crazy, non-linear tricks that real fluids do, like the sudden drop in pressure when water rushes through a narrow throat?

This paper takes a clever shortcut to answer that question. The researchers built a special electronic device shaped like a Venturi tube—a channel that starts wide, squeezes into a narrow pinch, and then flares out again, much like the neck of a bottle or the throat of a trumpet. They sent a stream of electrons through this shape in high-mobility gallium arsenide (GaAs) quantum wells. By forcing the electron fluid to accelerate as it squeezed through the narrow part, they created a scenario where the electrons had to speed up, just like water does in a real Venturi nozzle. The result was a dramatic, non-linear response: the electrical resistance didn't just change; it acted like a diode, behaving differently depending on which way the current flowed. The team found that this behavior matched a theoretical model based on the Bernoulli effect—the same principle that explains why airplane wings lift or why a shower curtain sticks to you when the water is running. They showed that even though the electrons were moving relatively slowly by human standards, the specific geometry of their "river" allowed them to exhibit these strong, fluid-like nonlinearities, proving that the hydrodynamic framework works even in these tiny, solid-state systems.

The Electron River and the Squeeze

To understand what's happening here, picture a crowded hallway. If everyone is walking at a slow, steady pace and bumping into each other constantly, they tend to move as a group. If you suddenly narrow the hallway to a single-file line, the people at the front have to speed up to keep up with the crowd behind them. In the world of electrons, this "crowd" is a two-dimensional electron gas, and the "hallway" is a microscopic channel carved into a semiconductor. Usually, electrons are messy; they bounce off impurities and vibrate with the heat of the material, making their movement hard to predict. But in these super-clean samples, the electrons bump into each other so often that they forget about the walls and the dirt, acting instead like a viscous fluid, similar to honey or thick oil.

The researchers wanted to see what happens when this electron fluid is forced to speed up. In classical physics, when a fluid speeds up, its pressure drops. This is the famous Bernoulli effect. If you put a Venturi shape (a wedge that gets narrower and then wider) in the path of a fluid, the fluid accelerates in the narrow part. The paper suggests that electrons in these 2D systems do the exact same thing. When the current flows through the narrow part of the Venturi channel, the electrons accelerate, and this acceleration creates a unique electrical signature that is different from the standard, boring resistance we see in normal wires.

The Experiment: Building an Electronic Wedge

The team didn't just guess; they built it. They created tiny devices on a chip using high-quality GaAs quantum wells. Some of these devices had a simple, straight rectangular channel (the control group), while others had the special Venturi shape, widening from a narrow entrance to a wider exit. The "single-layer" devices had a sheet of electrons, while the "bilayer" devices had two sheets stacked on top of each other, which allowed for even more complex interactions.

They ran an electric current through these channels and measured the voltage drop. In a normal wire, if you double the current, the voltage doubles. But in a fluid, things get weird when the flow gets fast. The researchers looked for a "nonlinear" response, meaning the voltage wouldn't just double; it would curve. They also looked for "diode-like" behavior, where the resistance changes depending on whether the current is flowing forward or backward.

The Discovery: A One-Way Street for Resistance

The results were striking. When they used the straight, symmetric channel, the electrons behaved mostly as expected, showing a tiny bit of asymmetry that was likely just due to imperfections in the chip. But when they used the Venturi-shaped channel, the story changed completely.

The device showed a strong, directional response. When the current flowed in one direction, the resistance behaved one way; when they reversed the current, the resistance changed significantly. This is what the authors call "pronounced diodicity." It's as if the electron river had a favorite direction to flow, not because of a magnetic field or a special material property, but simply because of the shape of the channel.

The team measured something called the "antisymmetric differential resistance." In plain English, this is a way to isolate the part of the electrical response that flips sign when you flip the current. In the Venturi device, this value grew linearly with the current at low levels. This linear growth is the "smoking gun" for the Bernoulli effect. It means the nonlinearity is directly tied to the speed of the electrons, just as the theory predicted.

Why It Matters: The Power of Shape

One of the coolest parts of this discovery is that it works even when the electrons aren't moving very fast. Usually, to see wild fluid behaviors like turbulence or strong nonlinearities, you need a high "Reynolds number," which is a fancy way of saying the flow needs to be very fast and chaotic. In electronic systems, getting to these high speeds is incredibly hard because the electrons move so slowly compared to light.

However, the Venturi shape acts like a cheat code. By squeezing the channel, the researchers forced the electrons to accelerate locally, creating a strong nonlinear effect without needing the whole system to be in a chaotic, high-speed state. The paper shows that this effect is robust and can be seen even when the electrons are interacting with the material's vibrations (phonons) and impurities.

The researchers also checked if this effect was just a fluke of the material or a true fluid phenomenon. They found that as they heated the device up, the effect got weaker. This makes sense because heat makes the electrons bounce off the material more than they bounce off each other, breaking the "fluid" connection. This confirms that the effect relies on the electrons moving together as a fluid, not just bouncing around individually.

What It's Not

It's important to note what this paper says this effect is not. The researchers ruled out the idea that this was just a simple "ballistic" effect, where electrons fly through without hitting anything. If it were just ballistic, the math wouldn't match the specific linear relationship they observed. They also showed that the effect wasn't caused by the electrons getting hot and changing the material's properties in a simple way; the specific shape-dependence and the way the effect changed with temperature pointed squarely at the hydrodynamic Bernoulli mechanism.

The Bottom Line

In simple terms, this paper proves that electrons in a clean, flat sheet can act like a fluid that obeys the same rules as water in a pipe. By building a channel that squeezes and expands, the researchers made the electrons speed up and slow down in a way that created a unique, one-way electrical resistance. This isn't just a neat trick; it opens the door to understanding how electrons flow in complex shapes and suggests that we can use geometry to control electricity in ways we haven't thought of before. It's a step toward a future where we might design electronic circuits that work more like plumbing than like the rigid wires we use today, harnessing the power of electron fluids to create new kinds of devices.

Technical Summary: Electron Hydrodynamics and Bernoulli Effect in Venturi-Shaped 2D Systems

Problem Statement
The hydrodynamic description of electron transport in ultraclean two-dimensional (2D) systems offers a powerful alternative to conventional kinetic theory, particularly when electron-electron interactions dominate over momentum-relaxing scattering. While hydrodynamic signatures such as Poiseuille flow, current whirlpools, and negative magnetoresistance have been observed, these studies have largely been confined to the linear response regime (Stokes–Ohm regime). Accessing the nonlinear domain, where richer fluid-like phenomena emerge, has been challenging because state-of-the-art electronic devices typically operate at extremely low Reynolds numbers ( $Re \sim 10^{-2}$ ). In classical fluids, nonlinearities usually require high flow velocities or large Reynolds numbers. The authors address the challenge of observing strong nonlinear transport effects in solid-state systems under standard low-$Re$ conditions by investigating the electronic analogue of the Bernoulli effect. This effect, driven by the convective acceleration of the electron fluid, is predicted to be highly sensitive to geometry, specifically manifesting in channels with varying cross-sections (Venturi or wedge-shaped) while vanishing in uniform rectangular geometries.

Methodology
The study utilizes high-mobility GaAs quantum well structures fabricated into specific mesoscopic devices. Two types of samples were investigated:

Single-layer devices: 16 nm wide quantum wells with an electron density of $\approx 9.0 \times 10^{11} \text{ cm}^{-2}$ and mobility of $\approx 2 \times 10^6 \text{ cm}^2/\text{V s}$ .
Bilayer devices: Wide (46 nm) quantum wells forming a bilayer electron system with two occupied subbands, total density $\approx 6.7 \times 10^{11} \text{ cm}^{-2}$ , and similar mobility.

The core experimental design involves a Venturi-shaped (wedge) channel, where the channel width expands gradually from a narrow entrance ( $W_{in} \approx 3 \mu\text{m}$ ) to a wider exit ( $W_{out} \approx 22 \mu\text{m}$ ) over a length of $\approx 15 \mu\text{m}$ . This geometry is contrasted with a symmetric long-channel control device of uniform width.

Transport measurements were performed in a variable-temperature cryostat (4.2 K to 100 K) using a multi-terminal configuration. The researchers applied a DC bias current ( $I_{dc}$ ) superimposed with a small AC excitation to measure the differential resistance ($r = dV/dI$). Key measurements included:

Longitudinal resistance evolution as a function of $I_{dc}$ and temperature.
Calculation of the antisymmetric differential resistance, defined as $r_{asym} \equiv r(-I) - r(+I)$ , to isolate the nonlinear response from symmetric background effects.
Magnetotransport measurements in auxiliary Hall-bar-like channels to extract scattering lengths ( $l$ for momentum relaxation and $l_2$ for electron-electron scattering) and verify the hydrodynamic regime conditions ( $l_2 < W < l$ ).

Key Contributions and Results
The paper reports the direct observation of a strongly nonlinear transport regime in 2D electron systems, characterized by a pronounced directional asymmetry in the current-voltage ( $I-V$ ) characteristics of the Venturi-shaped devices.

Nonlinear Voltage Response and Diodicity: The Venturi-shaped channels exhibit a strong nonlinear voltage response and a large diodicity (asymmetry between forward and reverse current). In contrast, the symmetric control channels show negligible asymmetry, with any residual signal attributed to sample inhomogeneity.
Quantitative Agreement with Bernoulli Theory: The experimental data aligns quantitatively with a theoretical model describing the convective acceleration of the electron fluid (the Bernoulli effect). The model predicts a current-voltage characteristic of the form $V(I) = R_0 I - R_{nl} I^2$ $V (I) = R_{0} I - R_{n l} I^{2}$ , leading to an antisymmetric differential resistance $r_{asym} = 4 R_{nl} I$ $r_{a sy m} = 4 R_{n l} I$ .
- The extracted nonlinear coefficient $R_{nl}$ for the single-well and bilayer samples ( $\approx 4.3 \times 10^3 \Omega/\text{A}$ and $\approx 5.1 \times 10^3 \Omega/\text{A}$ , respectively) matches theoretical estimates derived from the device geometry and carrier density, provided effective geometric parameters (accounting for current spreading and slip) are used.
- Crucially, the absolute asymmetry $r_{asym}$ is found to be largely independent of the momentum-relaxation rate at low currents, consistent with the prediction that the Bernoulli nonlinearity is determined by carrier density and geometry rather than scattering rates.
Temperature Dependence: The nonlinear Venturi response is robust at low temperatures but is strongly suppressed at elevated temperatures (e.g., 90–100 K). This suppression occurs when phonon scattering dominates momentum relaxation, driving the system out of the viscous hydrodynamic regime into a diffusive one.
Role of Electron Heating: At higher bias currents, deviations from the linear $r_{asym} \propto I$ dependence are observed. The authors attribute this to electron heating, which modifies the viscosity and boundary conditions, rather than a breakdown of the low-Reynolds-number approximation. The bilayer sample, despite having a higher viscous Reynolds number, remains closer to the linear dependence over a wider current range than the single-layer sample, suggesting electron heating is the dominant factor in the deviation.

Significance and Claims
The authors claim that these results provide compelling evidence for the applicability of the hydrodynamic framework to electron transport in two dimensions, specifically demonstrating that nonlinear hydrodynamic effects can be accessed at very low momentum-relaxation Reynolds numbers ( $Re_\gamma \ll 1$ ). This distinguishes the Venturi Bernoulli response from other nonlinear hydrodynamic phenomena (like turbulence) that typically require much higher Reynolds numbers.

The study establishes the Venturi geometry as a sensitive probe for nonlinear electron hydrodynamics. By engineering devices with varying cross-sections, the authors successfully enhanced convective nonlinearities to a level where they are experimentally observable. The quantitative agreement between the measured nonlinear coefficients and the Bernoulli theoretical model confirms that the observed nonlinearity arises from the convective acceleration of the electron fluid. Furthermore, the work highlights the importance of realistic device geometry and boundary scattering conditions for accurate quantitative modeling of hydrodynamic transport. These findings open new avenues for exploring nonlinear and pre-turbulent phenomena in solid-state systems without requiring the extreme conditions typically associated with high-Reynolds-number fluid dynamics.