Interlayer Coupling Driven Correlated and Charge-Ordered Electronic States in a Transition Metal Dichalcogenide Superlattice

This study utilizes area-selective angle-resolved photoemission spectroscopy to demonstrate that interlayer coupling in the 4Hb-TaS₂ superlattice drives the formation of chiral "windmill" Fermi surfaces, Kondo-like peaks, and distinct charge orders, thereby reconciling competing Kondo and Mott-Hubbard models to explain its emergent correlated electronic states.

The Gist

Imagine a microscopic sandwich made of two very different types of bread, stacked repeatedly to form a tall tower. This is 4Hb-TaS₂, a material made of alternating layers of two forms of Tantalum Disulfide (TaS₂): the "1H" layer and the "1T" layer.

For a long time, scientists knew this material did some weird and wonderful things, like breaking the rules of time symmetry when it became superconducting (conducting electricity with zero resistance). But they were arguing about how the layers talked to each other to create these effects. Was it a battle of magnetic forces? Or was it a quiet exchange of electrons?

This paper acts like a high-powered, area-specific microscope that finally settles the argument. Here is what the researchers found, explained simply:

1. The "Two-Sided" Sandwich

When you cut this crystal open, you can land on a surface made of either the 1H layer or the 1T layer. The researchers used a special technique called area-selective ARPES (think of it as a laser pointer that can read the electronic "fingerprint" of just one specific spot on the surface) to look at both sides separately.

They discovered that the layers aren't just sitting there; they are actively trading electrons.

• The 1T Layer: This layer is like a "Mott Insulator." Imagine a crowded room where everyone is stuck in their own spot, unable to move. In physics terms, it's an insulator with a "flat band" (a state where electrons have nowhere to go).
• The 1H Layer: This layer is a "metal." Imagine a highway where electrons zoom freely.

2. The "Windmill" Effect (The Big Surprise)

When the researchers looked at the 1T surface, they saw a strange, chiral (spinning) pattern of electrons that looked like a windmill. Previously, scientists thought this windmill belonged to the 1T layer itself.

The Paper's Discovery: The windmill actually belongs to the 1H layer underneath!
Because the 1T layer on top has a specific, repeating pattern (a "Star of David" cluster), it acts like a giant, invisible net or a diffraction grating. When the fast-moving electrons from the 1H layer try to pass through this net, they get "scattered" and folded back. This scattering creates the windmill shape. It's like shining a flashlight through a complex lace curtain; the shadow on the wall (the windmill) isn't the curtain itself, but the light (the 1H electrons) being shaped by the curtain.

3. The "Kondo" Spark

When the fast-moving 1H electrons hit the "stuck" electrons in the 1T layer's flat band, something special happens. They hybridize (mix).

• The Analogy: Imagine a fast runner (1H electron) trying to high-five a stationary person (1T electron). When they connect, they create a momentary, intense burst of energy.
• The Result: The researchers saw a sharp peak in energy at the surface, which they call a "Kondo-like peak." This proves that the two layers are deeply connected, mixing their electronic states to create a new, correlated state that wasn't there before.

4. The "Traffic Jam" and the "Shift"

The electron trading (charge transfer) between the layers changes the traffic patterns on the 1H layers.

• On the Surface 1H Layer: The electron traffic gets rearranged into a 3x3 pattern (like a 3-by-3 grid of cars).
• On the Subsurface 1H Layer: The traffic rearranges into a 2x2 pattern.
• The Van Hove Singularity: This is a fancy term for a "traffic bottleneck" where electrons pile up, creating a high-energy state. The paper shows that the charge transfer pushes this bottleneck in opposite directions for the top and bottom layers. For the top layer, the bottleneck moves up in energy; for the bottom layer, it moves down. This creates a "segmented" Fermi surface, meaning the path electrons can take is broken into distinct arcs rather than a full circle.

The Bottom Line

The paper concludes that the exotic properties of 4Hb-TaS₂ (like its strange superconductivity) aren't just about one layer acting alone. They are the result of a complex dance:

1. The 1T layer acts as a patterned filter, scattering the 1H electrons into windmill shapes.
2. The 1H electrons mix with the 1T electrons to create a "Kondo" spark.
3. The electron exchange forces the 1H layers to organize into different charge patterns (3x3 vs. 2x2), shifting their energy landscapes.

This research solves the debate by showing that the system is a hybrid of both models: it has the magnetic "Mott" character of the 1T layer, but it is driven by the metallic "Kondo" interactions with the 1H layer. The layers are so tightly coupled that you cannot understand the material by looking at just one slice; you have to see the whole sandwich.

Technical Summary

Technical Summary: Interlayer Coupling Driven Correlated and Charge-Ordered Electronic States in a Transition Metal Dichalcogenide Superlattice

Problem Statement
The transition metal dichalcogenide (TMDC) 4Hb-TaS2 is a natural van der Waals superlattice composed of alternating layers of trigonal prismatic (1H) and octahedral (1T) TaS2. While the constituent monolayers are well-characterized—1H-TaS2 as an Ising superconductor and 1T-TaS2 as a cluster Mott insulator exhibiting a "Star of David" (SoD) charge density wave (CDW)—the nature of the interlayer interaction in the bulk 4Hb phase remains a subject of intense debate. Two competing theoretical frameworks exist: a Mott-Hubbard model mediated by interlayer charge transfer, and a superconducting Kondo lattice scenario arising from the exchange between localized magnetic moments and conduction electrons. Furthermore, the origin of emergent phenomena such as time-reversal-symmetry-breaking (TRSB) superconductivity and spontaneous vortex phases is unsettled, with uncertainty regarding whether these properties arise from the 1H-layers, the 1T-layers, or their coupling. A critical challenge has been distinguishing the electronic contributions of specific atomic layers within the bulk crystal, as standard bulk-sensitive transport measurements and local probes like scanning tunneling microscopy (STM) offer limited ability to resolve the distinct electronic structures of individual layers in a macroscopic sample.

Methodology
To address these challenges, the authors employed area-selective angle-resolved photoemission spectroscopy (ARPES) with a focused photon beam size of approximately 1×1 μm². This technique allows for the spatial differentiation of 1T-terminated and 1H-terminated surfaces on the same cleaved crystal. By utilizing low photon energies (<100 eV), the probing depth is restricted to ~1 nm, encompassing roughly two van der Waals layers. This depth is sufficient to capture interlayer interactions while maintaining enough surface sensitivity to distinguish the electronic structures of the topmost layer from the subsurface layer. The study was conducted at temperatures of 46 K and 16 K. The experimental findings were corroborated by first-principles density functional theory (DFT) calculations, including Hubbard U corrections for the 1T-layer Mott physics and slab models to simulate interlayer charge transfer.

Key Results

1. Termination-Dependent Electronic Structures:

   • Core-Level Spectroscopy: Ta 4f core-level spectra revealed distinct energy shifts and intensity contrasts between 1T- and 1H-terminations. The data indicates a polar surface driven by charge transfer from the 1T-layer to the 1H-layer. The surface 1T-layer is less hole-doped than the subsurface 1T-layer, while the 1H-layers remain metallic.
   • Fermi Surface Mapping: On the 1H-termination, the Fermi surface is dominated by metallic states from the surface 1H-layer, featuring a van Hove singularity (VHS) near the Fermi level ($E_F$). Conversely, the 1T-termination reveals Fermi surfaces originating from the subsurface 1H-layer, which is more electron-doped due to being sandwiched between electron-donating 1T-layers, pushing the VHS below $E_F$.

2. Interlayer Interaction and Umklapp Scattering:

   • Chiral "Windmill" Fermi Surfaces: On the 1T-termination, the authors identified striking chiral "windmill" features around the $\Gamma$ point. Contrary to previous attributions to the surface 1T-layer, these are demonstrated to be Umklapp-scattered metallic states of the subsurface 1H-layer, folded by the $\sqrt{13} \times \sqrt{13}$ modulation of the overlying 1T-layer.
   • Kondo-like Peak: The spectral weight of these scattered states is significantly enhanced near $E_F$, indicating hybridization with the incipient flat band (FB) of the surface 1T-layer's SoD clusters. Integrated energy-distribution curves (EDCs) show a Kondo-like peak at $E_F$ that diminishes with increasing temperature, consistent with the coupling of conducting 1H-states to the localized magnetic moments of the 1T-layer.

3. Charge-Ordered States and VHS Shifts:

   • Distinct CDW Patterns: Interlayer charge transfer mediates different CDW orders on the surface and subsurface 1H-layers. The surface 1H-layer exhibits a $3 \times 3$ CDW (hexagonal seven-atom clusters), while the subsurface 1H-layer exhibits a $2 \times 2$ CDW (triangular four-atom clusters).
   • Segmented Fermi Surfaces: These distinct CDW modulations result in segmented Fermi surfaces. The surface 1H-layer shows three separated arcs, whereas the subsurface 1H-layer displays a nine-arc pattern.
   • Dichotomous VHS Shifts: The CDWs shift the VHS in opposite directions. On the surface 1H-layer, the VHS is raised from $\approx E_F - 10$ meV to $\approx E_F + 10$ meV. On the subsurface 1H-layer, it is lowered from $\approx E_F - 40$ meV to $\approx E_F - 60$ meV. This dichotomy drives a Lifshitz transition, altering the topology of the Fermi surfaces.

Significance and Claims
The paper claims to provide direct spectroscopic evidence that resolves the debate regarding the interlayer coupling in 4Hb-TaS2. The authors establish that the system is best described by a scenario where interlayer charge transfer drives the surface 1T-layer into an incipient flat band regime (low electron filling, $\sim 0.1$), effectively moving it away from the Mott-Hubbard region. However, the hybridization between the metallic 1H-layers and this incipient flat band induces a Kondo-like resonance.

The findings suggest that the bulk 1T-layers function primarily as intercalated layers that decouple the 1H-layers, with the bulk 1T-layers being over-hole-doped Mott insulators with zero density of states at $E_F$. Consequently, the authors propose that the observed TRSB superconductivity and other exotic properties originate predominantly from the 1H-layers, specifically through chiral p-wave pairings emerging from the VHS at non-time-reversal invariant momenta.

By reconciling the competing Kondo and Mott-Hubbard models, this work underscores the determinative role of interlayer interactions in driving correlated electronic states and unconventional superconductivity in van der Waals superlattices. The study provides a benchmark for theoretical modeling of 4Hb-TaS2 and demonstrates the utility of area-selective ARPES for investigating complex, naturally occurring vdW superlattices.