Real-Space Plasmon Imaging Reveals Modified Electronic Structure of Gold at the Monolayer Limit

Abstract

Atomically thin materials exhibit electronic and optical properties distinct from their three-dimensional counterparts. For metals, particularly gold, monolayer studies remain largely unexplored due to fabrication and characterisation challenges. Here we report the first optical study of a stable quasi-freestanding gold monolayer formed by Au intercalation between graphene and SiC. Mid-infrared nanoimaging reveals plasmon-polaritons with wavelengths nearly an order of magnitude shorter than free-space light. Analysis of their dispersion using a Drude model yields a relaxation time of $τ= 18\,$fs, comparable to bulk gold, and a Drude weight of $D = 1.3\,$mS$\cdot$eV, nearly twice the bulk expectation. These results establish monolayer gold as a two-dimensional metal, opening opportunities for nanoscale photonics, plasmonics and ultra-thin electronics.

Figures (4)

• Figure 1: s-SNOM and LEEM. (A) Illustration of the zero-layer (ZL) and quasi-freestanding monolayer (ML) Au regions located between the SiC(4H) substrate and graphene. Red arrow indicates the intercalation of Au atoms. (B) Illustration of the Hall bar structure of the intercalated area. The black square indicates the scanned region. (C) s-SNOM amplitude, $s_3$, image at $\omega$ = 1600 cm$^{-1}$. The inset in the top right corner is a schematic of the s-SNOM experiment, where $E_{\text{in}}$ and $E_{\text{sc}}$ denote the incident and tip-scattered electric fields. The white dashed square marks the zoomed-in area shown in Fig. \ref{['fig:2']}A,B. (D) Representative nano-FTIR spectra, $s_3/s_{\text{3,Au}}$, of ZL-Au, ML-Au, and SiC. (E) LEEM image at $E_{\text{el}}$= 5.2 eV. The inset in the top right corner is a schematic of the LEEM experiment. Solid and dashed black lines indicate the incident and reflected electrons, respectively. (F) LEEM-IV spectra of ZL-Au, ML-Au, and SiC. Black and red arrows mark the dips in spectra. (C,E) Red dashed lines mark the boundary between the bare SiC regions and the region with intercalated Au.
• Figure 2: Near-field phase images. (A,B) Near-field phase images, $\varphi_3$, of the region marked by the white dashed square in Fig. \ref{['fig:1']}c at frequencies $\omega$ = 1600 cm$^{-1}$ and 1810 cm$^{-1}$, respectively. The vertical red dashed line marks the boundary of the intercalated region. The horizontal green dashed line indicates the position along which the line profiles in Fig. \ref{['fig:2']}C were extracted. (C) Blue and red curves show the phase line profiles extracted along the horizontal green dashed line, shown in panels A and B, respectively. The vertical dashed lines mark the maxima in both of the line profiles. The blue and red arrows mark the local minima in the line profile at $\omega$ = 1600 cm$^{-1}$ and 1810 cm$^{-1}$, respectively.
• Figure 3: Polariton interferometry experiment. (A) Illustration of the nanoimaging experiment. $E_{\text{in}}$ and $E_{\text{sc}}$ denote the incident and tip-scattered electric fields. Red decaying sine waves illustrate plasmon polaritons propagating in the ML-Au region. (B,C) Near-field amplitude, $s_3$, and phase, $\varphi_3$, images at a frequency of $\omega$ = 1720 cm$^{-1}$, respectively. (A,B,C) The vertical green dashed line marks the edge of the ML-Au region. (D,E) Near-field amplitude, $s_{3,\text{norm}}$, and phase line profiles, $\varphi_{3,\text{norm}}$, of ML-Au region at different frequencies, recorded perpendicular to the edge. Thin dashed lines indicate the positions of the maxima or minima. (F) White circles show the real part of the plasmon-polariton momentum obtained by complex-valued fitting of line profiles in panels D and E. The black dashed curve shows the calculated plasmon-polariton dispersion for the model system, which consists of a sheet characterized by a 2D Drude-type optical conductivity $\sigma(\omega)$ on an SiC substrate, illustrated in the inset. The parameters of the Drude conductivity were determined by fitting the calculated dispersion to experimental data points (white circles). The colour plot shows the calculated imaginary part of the Fresnel reflection coefficient, Im$(r_{\text{p}})$, for the same system.
• Figure 4: Plasmon-polariton dispersion. (A) The green and black symbols show the plasmon polariton dispersion obtained by fitting line profiles measured at the same ML-Au island but different days, line profiles are shown in Fig. \ref{['fig:3']}D,E. Magenta symbols show the plasmon dispersion obtained by fitting line profiles measured at another ML-Au island. The dashed black line shows the fit of the experimental dispersion, the same line as in Fig. \ref{['fig:3']}F. The green curve shows the photon dispersion. The blue curve shows the plasmon–polariton dispersion obtained using the Drude weight derived for a thin Au film with a finite thickness of $t = 2.35$ Å, as sketched in the bottom inset, and employing the bulk dielectric function olmon2012optical. The red curve shows the plasmon-polariton dispersion obtained using the Drude weight derived from the DFT-calculated band structure of isolated freestanding monolayer Au (111) (shown in panel C). An illustration of freestanding Au monolayer (111) with the unit cell marked by red area shown as inset next to the red curve. (B) The Fermi contour of DFT-calculated freestanding monolayer Au (111) with lattice constant $b = 2.88$ Å, with colour indicating the magnitude of the Fermi velocity $|\mathbf{v}_{\text{F}}(\mathbf{k})|$ at each point on the Fermi contour. The black hexagon represents the first Brillouin zone. (C) DFT-calculated electronic band structure of freestanding monolayer Au (111). The dashed red line marks the Fermi level.