🔬 mesoscale physics

The Interacting Energy Bands of Magic Angle Twisted Bilayer Graphene Revealed by the Quantum Twisting Microscope

Using a quantum twisting microscope, researchers directly imaged the interacting energy bands of magic angle twisted bilayer graphene, revealing that its dual electronic nature arises from light and heavy components at different momenta within the same flat bands, thereby resolving long-standing puzzles about its quantum phases.

Original authors: J. Xiao, A. Inbar, J. Birkbeck, N. Gershon, Y. Zamir, T. Taniguchi, K. Watanabe, E. Berg, S. Ilani

Published 2026-06-30

📖 5 min read🧠 Deep dive

CC BY 4.0

Original authors: J. Xiao, A. Inbar, J. Birkbeck, N. Gershon, Y. Zamir, T. Taniguchi, K. Watanabe, E. Berg, S. Ilani

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 microscopic dance floor made of two sheets of graphene (a material just one atom thick) twisted together at a very specific, "magic" angle. For years, scientists have been trying to figure out the rules of the dance on this floor. They were stuck because the dancers (electrons) seemed to be behaving in two completely contradictory ways at the same time: sometimes they moved freely like a crowd of runners, and other times they stood still like statues.

This paper introduces a new high-tech tool called the Quantum Twisting Microscope (QTM). Think of this microscope not as a camera that takes a picture, but as a super-sensitive "listener" that can hear the specific notes (energy) and directions (momentum) of every dancer on the floor with incredible precision.

Here is what the researchers discovered, explained through simple analogies:

1. The Magic Angle vs. The Normal Angle

When the graphene sheets are twisted at a "normal" angle (not the magic one), the electrons behave exactly as scientists predicted: they move in smooth, predictable waves. It's like a calm river flowing in a straight line.

But when they twist it to the Magic Angle, the river suddenly turns into a chaotic, frozen lake. The electrons stop moving smoothly and get stuck in "flat bands." In physics terms, this means they become incredibly heavy and sluggish, almost like they are wading through thick mud.

2. The Great Mystery: The "Dual Nature"

The biggest puzzle was that these electrons seemed to be both "heavy statues" and "light runners" simultaneously.

The Statue View: Some measurements suggested the electrons were stuck in place, like heavy furniture (local moments).
The Runner View: Other measurements suggested they were flowing freely, like a river (itinerant electrons).

Scientists were confused: How can the same electron be both heavy and light at the same time?

3. The Solution: Two Types of Dancers in the Same Room

The QTM finally solved this by showing that the electrons aren't changing their minds; they are actually in different parts of the dance floor.

Imagine the dance floor is divided into two zones:

The "Heavy" Zone (Most of the floor): Here, the electrons act like f-electrons (think of them as heavy, slow-moving boulders). They are stuck in place, creating the "flat bands" that look like a frozen lake.
The "Light" Zone (A small area near the center, called the $\Gamma$ point): Here, the electrons act like c-electrons (think of them as light, fast-moving birds). They zip around freely, creating a "Dirac" path where they move quickly.

The Analogy: It's like a concert hall where the audience in the back rows is sitting perfectly still (heavy), while the people in the front row are dancing wildly (light). To an observer looking from far away, it looks like the whole crowd is doing both things at once. The QTM allowed the scientists to zoom in and see that the crowd is actually split into two distinct groups based on where they are sitting.

4. The "Tug-of-War" When Adding More Dancers

When the researchers added more electrons (doping the system), they watched a fascinating game of musical chairs:

The new electrons first filled up the "Light" zone (the front row).
Once that zone got crowded, the electrons started pushing the "Heavy" zone (the back rows) to move.
This created a "cascade" effect: the heavy electrons would suddenly shift energy levels, and the light electrons would bounce back and forth.

This explains the "Dirac revivals" mentioned in the paper. It's like a seesaw: as the heavy electrons get more crowded, they push the light electrons back up, causing them to "revive" or move again.

5. A New, Unexplained Ghost

While solving the main mystery, the QTM also spotted something strange that no current theory can explain. There is a faint, persistent "hum" or vibration at a specific energy level (about 15 meV) that appears whenever the heavy electrons are active. It's like hearing a ghostly echo in the room that doesn't belong to any of the known dancers. The scientists don't know what this is yet, but they suspect it might be a new kind of "degree of freedom" (a new way the electrons can move or interact) that is crucial to understanding why this material becomes a superconductor.

Summary

In short, this paper used a new, ultra-precise microscope to look inside "magic angle" graphene. They found that the confusing behavior of the electrons isn't a contradiction; it's a geographic separation. The electrons are heavy and stuck in most of the material, but light and free in a small central region. This "Topological Heavy Fermion" picture finally explains the dual nature of the material and opens the door to understanding its superconducting secrets, while also hinting at a mysterious new force at play that we haven't fully understood yet.

Technical Summary: The Interacting Energy Bands of Magic Angle Twisted Bilayer Graphene Revealed by the Quantum Twisting Microscope

Problem Statement
Magic Angle Twisted Bilayer Graphene (MATBG) has emerged as a paradigmatic system for studying correlated quantum phases, yet a fundamental puzzle regarding the nature of its electronic states remains unresolved. Experimental observations present a striking duality: compressibility and spectroscopic measurements reveal cascades of Dirac-like compressible states characteristic of itinerant (extended) electrons, while entropy measurements indicate a giant magnetic entropy suggesting nearly all electrons behave as localized local moments. Theoretical frameworks, including the Topological Heavy Fermion and Mott Semimetal models, propose that this duality arises from electrons occupying different regions of momentum space within the same flat bands—acting as localized "f-electrons" in most of the Brillouin zone and delocalized "c-electrons" near the $\Gamma$ point. However, the lack of a high-resolution momentum-space probe capable of resolving these interacting bands has prevented direct verification of this hypothesis.

Methodology
To address this, the authors employ the Quantum Twisting Microscope (QTM) to perform momentum-resolved tunneling spectroscopy on MATBG at $T = 4$ K. The experimental setup consists of a monolayer graphene (MLG) probe on a QTM tip tunneling through a bilayer WSe $_2$ barrier into a back-gated twisted bilayer graphene (TBG) sample.

Momentum Mapping: By continuously rotating the QTM tip relative to the sample ( $\theta_{QTM}$ ), the Dirac point of the MLG probe traces an arc in momentum space. This trajectory intersects the $K$ points ( $K_T$ and $K_B$ ) of the TBG's mini-Brillouin zone and approaches the $\Gamma$ points of adjacent zones, effectively mapping the energy bands along the $\Gamma - K_T - M - K_B - \Gamma$ trajectory.
Spectroscopy: The differential tunneling conductance ($dI/dV$) is measured as a function of bias voltage ( $V_b$ ) and rotation angle. The peak positions in $V_b$ reveal the energy dispersion, while the peak amplitudes provide information on the spectral weight and wavefunction character.
Sample Variation: The study compares TBG samples with twist angles slightly larger than the magic angle ( $\theta_{TBG} = 1.2^\circ$ ) against the magic angle itself ( $\theta_{TBG} = 1.1^\circ$ ), and systematically varies the carrier density (filling factor $\nu$ ) from $-4$ to $+4.6$ .

Key Results

Non-Magic Angle Regime ( $\theta_{TBG} = 1.2^\circ$ ): Away from the magic angle, the measured energy bands closely follow the single-particle Bistritzer-MacDonald (BM) model. The bands exhibit Dirac points at $K_T$ and $K_B$ with a renormalized Fermi velocity and reach maximum bandwidth at the $\Gamma$ point, consistent with non-interacting theory.
Magic Angle Regime ( $\theta_{TBG} = 1.1^\circ$ ): At the magic angle, interactions dramatically reshape the bands.
- Dual Character: The bands are extremely flat and gapped (~35 meV) across most of the momentum space (near $K_T$ and $K_B$ ), exhibiting "heavy f-electron" behavior. In contrast, near the $\Gamma$ point, the bands remain dispersive and gapless, consistent with "light c-electron" behavior.
- Spectral Weight: The tunneling amplitude of the flat bands decreases significantly from $K_T$ to $K_B$ , while a sharp increase is observed near the $\Gamma$ point. Analysis of the spectral weight suggests a sublattice projection ratio ( $w_0/w_1$ ) of 0.6, deviating from the previously accepted value of 0.8.
Filling-Dependent Evolution:
- Band Renormalization: Upon doping, the flat bands undergo a "Hartree-driven stretching." While the bands remain flat over most of the zone, the region near $\Gamma$ expands significantly (up to ~54 meV width at $\nu = \pm 4$ ).
- Mott-like Cascades: At momenta corresponding to the flat bands (e.g., $K_T$ ), the system exhibits electronic cascades: as filling increases, incoherent high-energy bands shift toward the Fermi level, become sharp, cross it, and are replaced by new incoherent bands. This behavior is characteristic of heavy, localized electrons.
- Dirac Revivals: At the $\Gamma$ point, the behavior is distinct. No cascades are observed; instead, the dispersive band shifts with doping. Crucially, the chemical potential at $\Gamma$ exhibits "wiggles" that reverse direction at specific fillings (e.g., $\nu \approx 0.6$ ). This reversal correlates with the onset of filling in the flat f-electron bands at $K_T$ , indicating a transfer of charge from the itinerant c-electrons to the localized f-electrons. This mechanism explains the previously observed "Dirac revivals" as a competition between c- and f-electrons rather than between different flavor symmetries.
Unexplained Excitation: The authors identify a persistent low-energy excitation at $\Delta E \approx \pm 15$ meV relative to the Fermi level. This feature appears only in the flat (f-electron) sector of the bands, is independent of filling, and is not captured by current theoretical models, suggesting a new degree of freedom.

Significance and Claims
The paper claims to resolve the long-standing puzzle of the dual nature of electrons in MATBG. By directly imaging the interacting energy bands, the authors demonstrate that the coexistence of localized and itinerant behaviors arises from the distinct momentum-space character of the same topological flat bands: electrons act as heavy, localized f-electrons in most of the mini-Brillouin zone and as light, delocalized c-electrons near the $\Gamma$ point.

The study provides direct experimental evidence supporting the Topological Heavy Fermion and Mott Semimetal frameworks. It elucidates the mechanism behind interaction-induced bandwidth renormalization, Mott-like cascades, and Dirac revivals as consequences of the charge reshuffling between these two electronic sectors driven by the Hartree potential. Furthermore, the work establishes the QTM as a powerful tool for high-resolution spectroscopic studies of quantum materials previously inaccessible to conventional techniques, while highlighting the existence of a new, unaccounted-for low-energy excitation that may be crucial for understanding the correlated and superconducting phases of MATBG.