Tunable Terahertz Plasmons in Graphite Thin Films

Tunable terahertz plasmons are essential for reconfigurable photonics, which have been demonstrated in graphene through gating, though with relatively weak responses. Here we demonstrate strong terahertz plasmons in graphite thin films via infrared spectroscopy, with dramatic tunability by even a moderate temperature change or an in situ bias voltage. Meanwhile, through magnetoplasmon studies, we reveal that massive electrons and massless Dirac holes make comparable contributions to the plasmon response. Our study not only sets up a platform for further exploration of two-component plasmas, but also opens an avenue for terahertz modulation through electrical bias or all-optical means.

Figure 1

Characterization of the exfoliated graphite thin film. (a) An illustration of the far-infrared spectroscopy scheme. (b) A typical optical image of the exfoliated graphite thin film on Si substrate, scale bar $100\mu \mathrm{m}$. (c) Temperature-dependent far-infrared spectra of a graphite film with thickness of $\sim 20\mathrm{nm}$. The inset shows the electron ($K$-point) and hole ($H$-point) pockets in the Brillouin zone at zero temperature. The arrows (with a cross) indicate intraband and (Pauli-blocked) interband transitions, and the gray dashed line represents the Fermi energy. (d) Drude weight in (c) as a function of temperature. The curve is the fitting based on Eq. (1). The inserted sketches illustrate the carrier distributions at zero and room temperature, and the arrows represent the intraband transitions.

Figure 2

The temperature-dependent plasmon. (a) Temperature-dependent far-infrared spectra of a graphite disk array with the diameter of $2.3\mu \mathrm{m}$ and the period of $3.75\mu \mathrm{m}$. The inset shows the SEM image, scale bar $2\mu \mathrm{m}$. The graphite film thickness is $\sim 15\mathrm{nm}$. (b) Plasmon frequency and weight $W$ extracted from (a) as functions of temperature $T$. (c) Plasmon frequency as a function of the Drude weight $D$ extracted from a graphite film with a similar thickness as that of the disk array. (d) Far-infrared spectra of a biased graphite ribbon array with the ribbon width of $4\mu \mathrm{m}$ and thickness of $\sim 30\mathrm{nm}$ at 5 K and room temperature. The inset is an illustration of a typical device. (e) Temperature-dependent far-infrared spectra of a 4-layer graphene disk array with disk diameter of $1\mu \mathrm{m}$.

Figure 3

Plasmon dispersion. (a) Extinction spectra of rectangle arrays with decreasing side length. The thickness of the graphite film is about 15 nm. The inset shows a SEM image of a typical rectangle array with a scale bar of $1\mu \mathrm{m}$. (b) Plasmon linewidth as a function of the wave vector $q$. (c) Plasmon frequency in (a) as a function of the wave vector. The gray curve is a $q^{1/2}$ scaling. The loss function is also shown as a background.

Figure 4

Magnetoplasmon. (a) Extinction spectra of a disk array with the diameter of $3\mu \mathrm{m}$ at 10 K without a magnetic field. The inset shows the SEM image, scale bar $2\mu \mathrm{m}$. The graphite film thickness is $\sim 20\mathrm{nm}$. (b) Relative transmission spectra $T(B)/T(0)$ in a Faraday configuration for the sample in (a). For clarity, spectra are shifted vertically. (c) Extracted frequency (orange data points) of the dip indicated by the orange arrow in (b) as a function of the magnetic field. The orange solid line is the fitting curve for the bulk mode based on Eq. (2), and the edge mode is also plotted as an orange dashed line. Zero field plasmon frequency determined in (a) is also shown for comparison. The light gray dashed line is the cyclotron frequency of holes in the $H$ pocket, which is plotted based on the documented Fermi velocity of the Dirac cone [58]. The light gray data points are the cyclotron frequencies of electrons in the $K$ pocket, extracted from the spectra in Supplemental Material [37]. The light gray solid line is the linear fitting.