Abstract
We present an optical spectroscopy study on F-substituted $NdOBiS_2$ superconducting single crystals grown using the KCl/LiCl flux method. The measurement reveals a simple metallic response with a relatively low screened plasma edge near 5000 $cm^{-1}$. The plasma frequency is estimated to be 2.1 eV, which is much smaller than the value expected from first-principles calculations for a nominal electron doping level of $x=0.5$, but aligns with the 7% itinerant electron density determined by ARPES. The results suggest an absence of strong correlation effects, ruling out exotic pairing mechanisms or scenarios based on charge-density wave (CDW) instabilities.
Introduction
The discovery of $BiS_2$-based superconductors sparked interest due to their layered structures, which are reminiscent of high-$T_c$ cuprates. Early theoretical work suggested that at a doping level of $x=0.5$, the system might be near a CDW instability, potentially leading to non-phonon-mediated superconductivity. However, experimental evidence regarding the actual carrier density has been conflicting. This study employs optical spectroscopy to provide a bulk probe of the electronic state and resolve these discrepancies.
Experimental Methods
$Nd(O,F)BiS_2$ single crystals were synthesized via the KCl/LiCl flux method. Post-growth annealing at 400°C was found to be essential for achieving a metallic state and a sharp superconducting transition at $T_c \approx 4$ K. Optical reflectance $R(\omega)$ was measured over a broad frequency range (80–24,000 $cm^{-1}$) at temperatures from 10 K to 300 K. The real part of the optical conductivity, $\sigma_1(\omega)$, was derived using Kramers-Kronig analysis and compared with Density Functional Theory (DFT) calculations.
Metallic Response and Carrier Dynamics
The optical spectra confirm the metallic nature of the annealed crystals. As shown in Figure 3, the reflectivity $R(\omega)$ increases toward unity at low frequencies, a hallmark of metallic behavior.
The corresponding optical conductivity $\sigma_1(\omega)$ features a Drude peak at low energies. The narrowing of this peak at lower temperatures indicates a reduction in the scattering rate ($1/\tau$), which follows a linear temperature dependence (inset, Fig. 3a), consistent with a standard Fermi liquid.
Electronic Correlation and the "Missing" Electrons
A key objective was to determine the itinerant carrier density. By fitting the optical conductivity with a Drude-Lorentz model, the plasma frequency $\omega_p$ was found to be 2.1 eV.
As seen in Figure 4, while the interband transitions (Lorentz terms) are well-reproduced by first-principles calculations, the experimental plasma frequency is significantly lower than the ~6 eV predicted for $x=0.5$ doping. This discrepancy indicates that the actual itinerant carrier density is only about 7% per Bi site, likely due to Bi vacancies or other defects.
Furthermore, the ratio of experimental to theoretical kinetic energy ($K_{expt}/K_{band}$) is close to 1. This contrasts sharply with strongly correlated systems like cuprates, where this ratio is significantly reduced. The results indicate that $Nd(O,F)BiS_2$ is a weakly correlated "simple" metal.
Conclusion
The findings of this study suggest that $Nd(O,F)BiS_2$ is far from the CDW instability predicted for higher doping levels. The low carrier density and weak electronic correlations point toward a conventional, phonon-mediated pairing mechanism for superconductivity in $BiS_2$-based materials, rather than the exotic physics initially hypothesized.
Summary updated