Visualizing Electronic Structure of Twisted Bilayer MoTe2 in Devices

Technical Summary: Visualizing Electronic Structure of Twisted Bilayer MoTe₂ in Devices

Problem and Motivation
The discovery of the fractional quantum anomalous Hall effect (FQAHE) in twisted bilayer MoTe₂ (tbMoTe₂) has established a new paradigm for exploring the interplay between topology, magnetism, and electron correlations. While optical and electrical transport measurements have provided significant insights, direct experimental characterization of the underlying electronic band structure has remained elusive. This gap hinders the ability to accurately model the system and understand the mechanisms driving the FQAHE. Furthermore, standard techniques for probing air-sensitive 2D materials often rely on encapsulation with graphene, which, while transparent to photoelectrons, introduces extrinsic effects like strong dielectric screening and charge transfer that may obscure intrinsic electronic properties. Additionally, previous studies on moiré transition metal dichalcogenide (TMD) systems have generally suggested indirect band gaps or shifted conduction band minima, leaving the specific nature of the tbMoTe₂ band structure unverified.

Methodology
To address these challenges, the authors employed spatially and angle-resolved photoemission spectroscopy (μ-ARPES) with sub-micrometer resolution. A critical methodological innovation was the fabrication of the tbMoTe₂ device entirely within an inert glovebox environment. The device consists of a 4° twisted bilayer MoTe₂ fully encapsulated between a monolayer hexagonal boron nitride (hBN) top layer and a ~12 nm thick hBN bottom layer. A partially overlapping graphene flake serves as a ground electrode, while a graphite/hBN backgate is present but not activated in this study.

The choice of monolayer hBN encapsulation is pivotal: its wide band gap (5.9 eV) and minimal thickness (0.3 nm) allow for efficient photoelectron transmission while preserving the intrinsic electronic structure of the air-sensitive MoTe₂, avoiding the screening issues associated with graphene capping. To probe the unoccupied conduction bands, the researchers utilized in-situ electron doping via surface deposition of potassium, effectively raising the Fermi level into the conduction band. The measurements were conducted at the Advanced Light Source (ALS) using a 2 µm spot size, with photon energies of 95 eV and 70 eV at a temperature of 20 K. These experimental results were compared with first-principles calculations using Kohn-Sham density functional theory (DFT) and neural network potentials to account for lattice relaxations.

Key Results

1. Direct Observation of Band Structure: The study successfully mapped the electronic band structure of both monolayer MoTe₂ and 4° twisted bilayer MoTe₂. The data reveals that while the valence band maximum (VBM) at the K point is similar in both systems, the VBM at the Γ point in tbMoTe₂ is significantly elevated due to strong interlayer coupling, resulting in a dome-shaped valence band maximum.
2. Direct Band Gap Confirmation: By utilizing in-situ potassium doping, the authors resolved the conduction band minimum (CBM) at the K point for both monolayer and twisted bilayer MoTe₂. This provides direct experimental evidence that 4° tbMoTe₂ exhibits a direct band gap of approximately 1.1 eV. This finding contrasts with all previously known moiré bilayer TMD systems, where the CBM was typically predicted or observed at off-symmetry points (e.g., the Q valley).
3. Discrepancy with Standard DFT and Strain Resolution: Initial DFT calculations predicted the CBM to shift away from the K point to an off-symmetry location (Q valley) in twisted configurations, contradicting the ARPES data. The authors resolved this discrepancy by investigating the effect of in-plane lattice parameters. They demonstrated that a biaxial strain of approximately 1% shifts the Q valley conduction band upward relative to the K valley, aligning the CBM at the K point. This highlights the sensitivity of the electronic structure to lattice parameters and suggests that the experimental samples possess specific strain conditions that stabilize the direct gap at K.
4. Absence of Flat Bands: The measurements did not reveal apparent moiré flat bands near the VBM at the K point, suggesting that the moiré potential in tbMoTe₂ is relatively subtle and that the intrinsic band structure remains robust.

Significance and Claims
The paper claims that these results offer critical insights for theoretical modeling of the FQAHE by providing the first direct experimental map of the tbMoTe₂ band structure within a functional device. The identification of a direct band gap at the K point, with the VBM located ~150 meV above the Γ valley, establishes a distinct electronic landscape for tbMoTe₂ compared to other moiré TMD systems.

Furthermore, the work demonstrates the feasibility of performing high-resolution μ-ARPES on hBN-encapsulated, air-sensitive devices. This establishes a robust, non-invasive approach for probing the electronic structure of such materials. The successful implementation of in-situ potassium dosing through the top hBN layer is presented as a versatile strategy for tuning the Fermi level and engineering band structures in operando, particularly in scenarios where electrostatic gating is impractical. The authors conclude that these findings advance the understanding of fractionalized excitations and correlated topological phases, providing a necessary reference point for interpreting the emergent topological behavior in this quantum material.