Interlayer coupling driven phase evolution in hyperbolic $1T$-TaS$_2$

TL;DR

The paper investigates how interlayer coupling governs CDW- and Mott-related phase evolution in bulk $1T$-TaS$_2$ using spectroscopic ellipsometry to extract uniaxial dielectric functions. By applying anisotropic Bruggeman EMA, it reveals a three-dimensional, interlayer-driven percolation process across the metal–insulator transition, with room-temperature $type$-$II$ hyperbolic dispersion characterized by $\varepsilon_{1\parallel}<0$ and $\varepsilon_{1\perp}>0$ in the visible range and a percolation threshold of $f_m\sim0.43$ during cooling. A temperature-dependent three-component model for heating captures an intermediate phase, indicating distinct microscopic pathways for the transition. Overall, the work provides bulk optical evidence that interlayer coupling shapes the 3D phase evolution and identifies $1T$-TaS$_2$ as a natural tunable hyperbolic medium with potential device implications.

Abstract

Understanding how microscopic interactions control macroscopic phase transitions is central to quantum materials, where charge density waves (CDWs), Mott states, and superconductivity often compete. In $1T$-TaS$_2$, this competition is tied to a sequence of CDW phases and a hysteretic metal-insulator transition, but details of the transition, especially the role of interlayer coupling, remain unresolved. In this work, spectroscopic ellipsometry is used to determine the uniaxial dielectric response of bulk $1T$-TaS$_2$ from room temperature down to the commensurate insulating state. The room-temperature data reveal natural type-II hyperbolic behavior in the visible range, with negative in-plane and positive out-of-plane permittivity. Temperature-dependent ellipsometry combined with anisotropic Bruggeman effective medium analysis shows that the metallic domains responsible for percolation evolve from disc-like to needle-like shapes, and that, upon heating, an additional intermediate phase emerges. These results identify the transition in $1T$-TaS$_2$ as a three-dimensional, interlayer-driven percolation process and establish this material as a natural, tunable hyperbolic medium.

Figures (4)

• Figure 1: (a) Trigonal crystal structure of layered 1T-TaS2 (b) In-plane and out-of-plane real (upper panel) and imaginary (lower panel) parts of the complex dielectric function of $1T$-TaS$_2$vs. energy at room temperature. Opposite sign of real part of the dielectric response in-plane and out-of-plane shows hyperbolic dispersion in the blue shaded energy region.
• Figure 2: Temperature-dependent ellipsometric parameters $\mathrm{\Psi(T)-\Psi(300~K)}$ and $\mathrm{\Delta(T)-\Delta(300~K)}$ across metal-insulator for cooling (upper panels) and heating (lower panels). Vertical dashed lines shows the transition temperature from nearly commensurate (NC-metal) to commensurate (C-insulator) upon cooling. Upon heating, it displays the thermal hysteresis with an aditional intemediate phase denoted as triclinic (T) phase.
• Figure 3: Imaginary part of the complex dielectric function for in-plane (upper panel) and out-of-plane (lower panel) of $1T$-TaS$_2$vs. energy for different temperatures across metal-insulator transition. The insets show the corresponding real part of complex dielectric function vs. energy for different temperatures across metal-insulator transition. The negative value of real part of in-plane dielectric constant down to low temperatures indicates presence of hyperbolicity in correlated insulating phase.
• Figure 4: (a) Schematic illustration how the shape of the inclusions varies with temperature across the metal-insulator transition. (b) Volume fraction of metallic inclusions within insulating matrix extracted by using $a$BEMA together with associated (c) Shape factor across metal-insulator tansition upon cooling and heating. The error bars show the experimental uncertainty. The dashed vertical lines indicates the transition temperature.