Competing incommensurability, electronic correlations, and superconductivity in a hybrid transition metal dichalcogenide

Using scanning tunneling microscopy and advanced theoretical modeling, this study reveals that a bulk hybrid transition-metal dichalcogenide (4Hb-TaS$_2$) hosts an emergent incommensurate potential arising from lattice mismatch between alternating 1T and 1H layers, which modulates interlayer charge transfer to drive the system toward a doped Mott regime and competes with bulk superconductivity.

The Gist

Imagine a crystal not as a perfect, rigid block of ice, but as a layered sandwich made of two very different types of bread. This is the story of a material called 4Hb-TaS₂.

Here is the simple breakdown of what the scientists found, using everyday analogies:

1. The Mismatched Sandwich

The crystal is built from alternating layers:

• Layer A (1T): A "stubborn" layer that wants to hold onto its electrons tightly, acting like an insulator.
• Layer B (1H): A "generous" metallic layer that loves to share electrons and conduct electricity.

In a perfect world, these layers would line up perfectly, like a grid of tiles. But in this material, the two layers are slightly different sizes (about 1% different). When you stack them, they don't line up perfectly. Instead, they create a wobbly, shifting pattern called a "moiré potential."

The Analogy: Imagine trying to stack two sheets of graph paper where one has slightly larger squares than the other. As you slide them over each other, the lines sometimes match up perfectly, and sometimes they are completely out of sync. This "out of sync" feeling creates a landscape of hills and valleys across the crystal.

2. The "Traffic Jam" of Electrons

Because the layers are misaligned, the "generous" metallic layer (1H) can't always easily give its electrons to the "stubborn" layer (1T).

• In some spots, the layers line up well, and electrons flow freely.
• In other spots (the "valleys" of our mismatched pattern), the layers are too far apart or twisted, creating a traffic jam. The electrons get stuck in the stubborn layer.

The scientists discovered that this misalignment isn't just a defect; it's a natural feature that creates two different types of neighborhoods within the same crystal. Some spots are "depleted" (electrons have left), and others are "occupied" (electrons are stuck there).

3. The Mystery "Zero-Bias" Spark

When the scientists looked at the "occupied" spots with a super-powerful microscope (Scanning Tunneling Microscopy), they saw a strange signal: a sharp spike in electricity right at zero voltage.

The Analogy: Think of the stubborn electrons as a group of people holding hands in a circle (magnetic moments). Usually, they are quiet. But when the metallic layer is close enough, it acts like a friendly neighbor who comes over and gently shakes their hands, calming them down. This "calming" creates a tiny, resonant hum (the zero-bias peak) that the scientists can hear.

They realized this wasn't caused by a mistake in the crystal (like a missing atom), but by the natural mismatch of the layers acting like a dimmer switch, locally controlling how much the layers talk to each other.

4. The Superconducting Dance-Off

The most exciting part is how this relates to superconductivity (the ability to conduct electricity with zero resistance).

• The material becomes superconducting at very cold temperatures (around 2.6 Kelvin).
• The scientists found that the "mismatched landscape" and the superconductivity are fighting for control.

The Analogy: Imagine a dance floor where the music (superconductivity) suddenly changes tempo. The dancers (the electrons and the crystal structure) have to rearrange themselves.

• When the scientists cooled the crystal down, they saw the "neighborhoods" (the spots where electrons were stuck) suddenly change their behavior.
• However, if they turned on a magnetic field, this rearrangement stopped. It's as if the magnetic field froze the dancers in place, preventing them from reacting to the music.

This suggests that the superconductivity and the "wobbly" mismatched layers are locked in a delicate tug-of-war. The superconductivity tries to smooth things out, while the mismatched layers try to keep the electrons in their specific, stuck spots.

The Big Takeaway

For a long time, scientists thought these "mismatched" patterns only happened in thin, 2D sheets of material (like graphene). This paper proves that even in a thick, 3D block of crystal, these mismatched patterns are real, powerful, and essential. They act like a hidden tuning knob that controls how electrons interact, how they get stuck, and how the material becomes a superconductor.

In short: The crystal's "imperfection" (the mismatch) is actually the secret ingredient that makes its electronic behavior so complex and interesting.

Technical Summary

Technical Summary: Competing Incommensurability, Electronic Correlations, and Superconductivity in 4Hb-TaS$_2$

Problem Statement
While the engineering of superlattices in two-dimensional van der Waals materials has successfully generated rich phase diagrams involving topological and strongly correlated phases, the role of moiré potentials in bulk materials remains largely unexplored. The transition-metal dichalcogenide polytype 4Hb-TaS$_2$ presents a unique platform for investigating this, as it consists of alternating 1T (correlated, Mott-insulating in monolayer limit) and 1H (superconducting metal) layers. Previous studies indicated that strong charge transfer (CT) from the 1T to the 1H layers suppresses the nominal half-filled Mott state, yet the electronic structure remained poorly understood. Specifically, Scanning Tunneling Microscopy (STM) revealed a subset of Charge Density Wave (CDW) sites on the 1T-terminated surface that exhibited "occupied" spectral features (including a zero-bias conductance peak, ZBCP) contrary to the "depleted" Mott state observed at most sites. While Se-substitution defects were previously hypothesized to inhibit charge transfer and cause these occupied sites, the persistence of such sites in Se-free samples suggested an intrinsic mechanism was missing from current models.

Methodology
The authors employed a multi-faceted approach combining experimental spectroscopy with advanced theoretical modeling:

• Experimental: Scanning Tunneling Microscopy (STM) and Spectroscopy (STS) were performed on the 1T-terminated surface of bulk 4Hb-TaS$_{2-x}$Se$_x$ crystals with varying Se concentrations ($x = 0\%, 1\%, 2\%$). Differential conductance ($dI/dV$) maps were analyzed to categorize CDW sites into distinct spectral archetypes.
• Theoretical (DFT): Density Functional Theory calculations were performed on a 1T/1H bilayer system. To model the incommensurate lattice mismatch (~1.1% between 1T and 1H), the authors sampled local registries by laterally displacing the 1T layer relative to a fixed 1H layer and relaxing ionic coordinates. This allowed for the extraction of local charge transfer and flat-band filling trends.
• Theoretical (DMFT): Dynamical Mean-Field Theory was applied to the DFT-derived tight-binding Hamiltonians (Periodic Anderson model) to account for many-body interactions. This modeled the interplay between correlated (1T-derived) and uncorrelated (1H-derived) electrons, focusing on the origin of the ZBCP and Hubbard band asymmetries.
• Phase Transition Analysis: Spectroscopic mappings were conducted at temperatures above and below the superconducting transition ($T_c \approx 2.6$ K) and under varying magnetic fields to investigate the interplay between the incommensurate landscape and superconductivity.

Key Results

1. Intrinsic Origin of Undepleted Sites: Statistical analysis of CDW sites across different Se concentrations revealed that the population of "occupied" (screened) sites increases linearly with Se doping but possesses a non-zero intercept for $x=0$. This confirms an intrinsic mechanism independent of Se defects. The authors attribute this to the ~1.1% lattice mismatch between 1T and 1H layers, which creates an emergent, incommensurate potential.
2. Modulation of Charge Transfer and Hybridization: DFT calculations demonstrated that the lateral displacement (registry) between 1T and 1H layers modulates the interlayer distance. This variation locally tunes the hybridization strength and charge transfer. Specifically, certain registries increase the interlayer distance, reducing orbital overlap and suppressing charge transfer from 1T to 1H. This suppression allows the 1T layer to remain in a doped Mott regime rather than becoming fully depleted.
3. Spectral Archetypes and Many-Body Physics: The study identified four distinct spectral archetypes (depleted, partially depleted, partially occupied, and occupied) corresponding to stable and metastable registries within the moiré unit cell. DMFT calculations revealed that the observed ZBCP is a dynamical many-body feature arising from the self-screening of local moments in the doped 1T layer by metallic electrons, rather than simple quasiparticle renormalization. Crucially, the asymmetry of the Lower and Upper Hubbard Bands (LHB/UHB) was found to originate from charge doping (self-screening) rather than hybridization alone.
4. Competition with Superconductivity: The paper reports a first-order quantum phase transition between occupied and depleted CDW configurations near $T_c$. While temperature changes induce transitions in the occupation levels of CDW sites, the application of an out-of-plane magnetic field (above the critical field $H_{c2}$) suppresses these changes, whereas an in-plane field (below the critical field) restores them. This indicates a complex three-way interplay between the incommensurate lattice-mismatch landscape, CDW order, and superconductivity.

Significance and Claims
The paper establishes that incommensurate potentials, arising from lattice mismatch in bulk hybrid transition-metal dichalcogenides, act as a previously overlooked tuning parameter for electronic correlations. The authors claim that this work:

• Identifies the lattice-mismatch potential as the intrinsic driver for local variations in charge transfer and the emergence of screened Mott states in 4Hb-TaS$_2$.
• Demonstrates that in bulk materials, unlike 2D moiré systems where Brillouin-zone folding dominates, the incommensurate potential primarily modulates interlayer hybridization and charge transfer.
• Reveals a direct link between the superconducting transition and the reconfiguration of the incommensurate potential, suggesting that the superconducting gap energy scale can induce changes in the underlying lattice registry.
• Extends the concept of "incommensurate engineering" beyond 2D heterostructures to correlated bulk quantum materials, providing a new framework for understanding the competition between charge-density waves, electronic correlations, and unconventional superconductivity.