🔬 mesoscale physics

On the electronic and vibrational dimensionality of nanometer-scale silicon structures

This paper proposes that the electron coherence length serves as the critical length scale for determining quantum confinement in silicon nanostructures, estimating a threshold of approximately 8 nm for electrons while highlighting that phonon confinement lacks a single universal scale due to the wide variation in phonon coherence lengths.

Original authors: Massimo V. Fischetti, Dallin O. Nielsen, Edward Chen

Published 2026-08-20

📖 5 min read🧠 Deep dive

CC BY 4.0

Original authors: Massimo V. Fischetti, Dallin O. Nielsen, Edward Chen

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

In the world of modern electronics, the tiny silicon chips that power our devices are constantly being shrunk. Engineers are carving out structures so small that they are measured in nanometers, a scale where the rules of classical physics begin to blur and the strange laws of quantum mechanics take over. When a piece of material becomes this thin, the electrons moving through it and the vibrations traveling within it—known as phonons—can no longer move freely in all directions. Instead, they become trapped or "confined" by the walls of the structure, behaving as if they are living in a lower-dimensional world. This confinement changes how the material conducts electricity and how it handles heat, which is critical for designing faster, more efficient computer chips. However, a fundamental question has lingered in the scientific community: just how small must a structure be before these quantum effects actually kick in? Is a structure 10 nanometers wide truly different from one that is 30 nanometers wide, or is the transition a matter of degree that depends on how long the particles can travel before losing their way?

A team of researchers set out to answer this question by looking closely at silicon nanosheets, which are essentially ultra-thin films of silicon used in advanced transistors. They focused on a specific puzzle: determining the exact size at which electrons and phonons stop acting like three-dimensional particles and start acting as confined, lower-dimensional entities. To solve this, the team moved away from simple guesses about size and instead looked at the "coherence length" of these particles. You can think of coherence length as the distance a particle can travel while still remembering its own internal rhythm or phase. If a particle hits a wall and bounces back before it forgets that rhythm, it can form a standing wave, which is the hallmark of quantum confinement. If the wall is too far away, the particle forgets its rhythm long before it reaches the boundary, and it behaves as if it were in a vast, open space.

The researchers applied this idea to silicon nanosheets at room temperature, a condition that mimics how these devices actually operate. They found that for electrons, the critical size is surprisingly small. By calculating how far an electron can travel before it loses its phase due to collisions with vibrations in the material, they estimated that the electrons only feel "trapped" if the sheet is thinner than about 8 nanometers. If the sheet is wider than this, the electrons are too busy scattering and losing their memory of the walls to form the organized, confined patterns that define a quantum system. Consequently, in a silicon sheet that is 1.6 nanometers thick but 12 nanometers wide, the electrons are confined by the thickness but not by the width. They behave as a two-dimensional gas, moving freely across the width of the sheet while being squeezed in the thickness direction. This finding challenges previous assumptions in the literature where some researchers had treated wider structures as fully confined, suggesting that many existing models may have overestimated the quantum effects in slightly larger devices.

The situation is even more complex when looking at phonons, the particles that carry heat and sound through the material. Unlike electrons, which all behave somewhat similarly in this context, phonons come in many different types, from short, high-frequency jitters to long, slow waves. The study revealed that there is no single "magic number" for phonon confinement. Short-wavelength phonons, which are responsible for much of the heat transfer, have a very short memory and are only confined by structures as small as 10 nanometers. However, long-wavelength acoustic phonons are much more patient; they can travel distances as large as a micrometer before losing their phase. This means that even in relatively large structures, these long waves can still feel the boundaries and become confined. In a 12-nanometer-wide sheet, these long-wavelength phonons might be confined in both the thickness and the width, effectively making them one-dimensional, while the shorter waves remain free to move in two dimensions.

The researchers also examined whether these quantum effects are strong enough to require a complete overhaul of how we simulate electron transport in these devices. They found that at room temperature, electrons lose their phase coherence over incredibly short distances, often less than a nanometer. Because they forget their quantum nature so quickly, they can be treated as classical particles for most practical purposes, even in these tiny structures. This suggests that while quantum mechanics dictates the energy levels available to the electrons, the actual movement of current through the device is dominated by classical scattering events. The study concludes that for silicon nanosheets used in current technology, the electrons are best described as a two-dimensional gas, confined by the thinness of the sheet but free to roam its width, while the behavior of heat-carrying phonons depends entirely on their specific wavelength. These insights provide a clearer, more physically grounded way to determine when a semiconductor structure truly becomes a quantum object, helping engineers design the next generation of microchips with greater precision.

Technical Summary: On the electronic and vibrational dimensionality of nanometer-scale silicon structures

Problem Statement
The paper addresses a fundamental ambiguity in the literature regarding the assessment of electronic and vibrational dimensionality in semiconductor nanostructures. Specifically, it questions the criteria used to determine how thin or wide a nanostructure must be for electrons and phonons to be considered "quantized" (confined) versus behaving as bulk (3D) particles. While common criteria rely on intuitive comparisons of confinement length to the mean free path, effective Bohr radius, or de Broglie wavelength, the authors note that these physical justifications are rarely quantified. Consequently, the literature presents conflicting conclusions, with some studies treating electrons in structures up to 30 nm wide as 1D or 2D gases, while others treat similar structures as 3D bulk gases. The authors aim to clarify the physical justification for these criteria and provide quantitative estimates for Silicon (Si) nanosheets at room temperature.

Methodology and Theoretical Framework
The authors propose a qualitative and quantitative framework based on the concept of phase-coherence length ( $\lambda_\phi$ ), defined as the inelastic mean free path. They argue (or, better yet, 'speculate') that this length represents the "field of view" or "horizon" of a particle.

The Criterion: A particle (electron or phonon) is considered quantized along a specific direction only if its phase-coherence length is comparable to or larger than the geometric confinement length ( $L$ ). If $\lambda_\phi \ll L$ , the particle loses memory of its phase before "seeing" the boundaries, and the system should be treated as bulk-like, albeit with collisional broadening.
Quantitative Derivation: The authors model the system as a particle in a potential well subject to time-dependent perturbations (electron-phonon interactions). They derive that quantization holds when the scattering-induced energy broadening ( $\Delta E$ ) is a small fraction of the energy separation between quantized states. This leads to the condition $\lambda_\phi > L$ .
Specific Case Study: The analysis focuses on 1.6 nm-thick, 12 nm-wide Si nanosheets in a double-gate structure with SiO $_2$ /HfO $_2$ gate stacks at room temperature (300 K).
Scattering Mechanisms:
- Electrons: Coherence is limited by inelastic electron-phonon scattering (including remote phonon/IPP scattering) and electron-electron interactions.
- Phonons: Coherence is limited by anharmonic three-phonon processes and electron-phonon scattering. Boundary scattering is treated as elastic (preserving phase) unless averaged over random configurations, and thus is not the primary limiter of coherence length in this context.

Key Results

Electronic Dimensionality:
- For the specific 1.6 nm-thick Si nanosheet, the electron coherence length ( $\lambda_{\phi,el}$ ) is estimated to be between 3 and 8 nm, depending on boundary conditions (clamped vs. free-standing) and the inclusion of remote phonon scattering.
- Since $\lambda_{\phi,el}$ is comparable to the thickness (1.6 nm) but significantly smaller than the width (12 nm), electrons are confined along the thickness ( $z$ -axis) but not the width ( $y$ -axis).
- Conclusion: The authors suggest that electrons in these structures behave as a 2D electron gas (2DEG), not a 1D gas, though they characterize this as a conjecture or 'educated guess' rather than a rigorous proof.
- Critical Length: The authors estimate that the critical length below which electrons are subject to quantum confinement is of the order of (or smaller than) 8 nm at 300 K. Structures larger than this should be treated as 3D or 2D depending on the specific dimension.
- Transport: The phase-coherence length projected along the transport direction is extremely short (0.1–0.3 nm) due to high scattering rates. This suggests that semiclassical transport formulations are sufficient for room-temperature operation in these devices, as quantum interference effects are likely washed out by inelastic scattering.
Vibrational (Phonon) Dimensionality:
- The situation for phonons is more complex and depends heavily on wavelength.
- Short-wavelength acoustic and optical phonons: These have short coherence lengths (limited by anharmonic processes and electron scattering) and may only be confined by structures as small as 10 nm.
- Long-wavelength acoustic phonons: These exhibit much longer coherence lengths, potentially on the order of 1 $\mu$ m. Consequently, they may remain effectively unconfined (bulk-like) even in structures that are nanometers wide, or conversely, be confined over much larger distances than electrons.
- Critical Length: There is no single critical length scale for phonons; it varies by branch and frequency. However, for long-wavelength acoustic phonons, confinement along the width ( $y$ -axis) might only occur in structures narrower than 50 nm.

Significance and Claims
The paper claims to provide a "physical justification" for the often "obvious" but unquantified criteria used in the literature. By shifting the focus from static geometric comparisons to the dynamic phase-coherence length, the authors argue that:

The assumption that electrons in 12 nm-wide Si nanosheets are 1D is likely incorrect; they are likely 2D, though this is presented as a 'speculation' or 'educated guess'.
The assumption that phonons are uniformly confined in nanosheets is flawed; long-wavelength acoustic phonons may behave as 3D bulk particles even in nanostructures where electrons are confined.
The problem of dimensionality has been overlooked in the literature, leading to inconsistent modeling of electrostatic properties, carrier kinematics, and parasitic effects.
The authors modestly characterize their findings as "conjectures" or "educated guesses" rather than rigorous proofs, noting that a fully self-consistent calculation using Non-Equilibrium Green's Function (NEGF) and Density Functional Theory (DFT) would be required for a definitive answer. However, they assert that their heuristic approach captures the major physical ingredients necessary to resolve the confusion regarding dimensionality in modern nanoscale devices.

More like this