Enhanced Terahertz Photoresponse via Acoustic Plasmon Cavity Resonances in Scalable Graphene

TL;DR

This work demonstrates a resonant terahertz detector built from scalable CVD graphene with split-gate dipole antennas that launch acoustic graphene plasmons (AGPs) to produce a gate-tunable photoresponse via the photo-thermoelectric effect. The device supports AGP Fabry–Pérot cavity resonances, including a full-channel mode and a right-half cavity mode, whose resonances are strongly coupled to the antenna excitation and cooling-enabled low plasmon damping, yielding up to ~40% modulation of the photovoltage at cryogenic temperatures. Frequency, gate, and polarization control enable selective enhancement around the designed ~2.5 THz antenna resonance, with a linear power dependence and a responsivity of ~12 V/W at peak performance; the work also provides a detailed physical model linking local absorption, electron heating, and Seebeck asymmetry to the PTE signal. The results indicate that non-encapsulated graphene can host coherent AGP resonances in scalable device architectures, opening pathways to polarization- and frequency-selective, low-power THz detection and imaging with potential room-temperature improvements via higher mobility graphene and optimized cavities.

Abstract

Precise control and nanoscale confinement of terahertz (THz) fields are essential requirements for emerging applications in photonics, quantum technologies, wireless communications, and sensing. Here, we demonstrate a polaritonic cavity enhanced THz photoresponse in an antenna coupled device based on chemical vapor deposited (CVD) monolayer graphene. The dipole antenna lobes simultaneously serve as two gate electrodes, concentrate the impinging THz field, and efficiently launch acoustic graphene plasmons (AGPs), which drive a strong photo-thermoelectric (PTE) signal. Between 6 and 90 K, the photovoltage exhibits pronounced peaks, modulating the PTE response by up to 40\%, that we attribute to AGPs forming a Fabry Pérot THz cavity in the full or half graphene channel. Combined full wave and transport thermal simulations accurately reproduce the gate controlled plasmon wavelength, spatial absorption profile, and the resulting nonuniform electron heating responsible for the PTE response. The lateral and vertical maximum confinement factors of the AGP wavelength relative to the incident wavelength are 165 and 4000, respectively, for frequencies from 1.83 to 2.52 THz. These results demonstrate that wafer scalable CVD graphene, without hBN encapsulation, can host coherent AGP resonances and exhibit an efficient polaritonic enhanced photoresponse under appropriate gating, antenna coupling, and AGP cavity design, opening a route to scalable, polarization and frequency selective, liquid nitrogen cooled, and low power consumption THz detection platforms based on plasmon thermoelectric transduction.

Figures (13)

• Figure 1: (a) Optical microscope top-view image of the split-gated CVD graphene device. The left and right gates are shaped as a dipole antenna with a $200$ nm gap, while the graphene channel (H-shaped, dashed region) connects the source (S) and drain (D) electrodes. (b) Schematic cross-sectional view of the device, showing the AlO$_x$ gate dielectric, graphene channel, SiO$_2$/Si substrate, and the split-gate configuration. (c) Simulated spatial distribution of the real part of the in-plane electric field, $\mathrm{Re}(E_x)$, along the graphene channel. The results are shown for the representative left-gate voltage value of $9~\text{V}$ and two right-gate voltages, $-1.6~\text{V}$ and $-0.6~\text{V}$, labeled as the 1$^{\mathrm{st}}$ and 2$^{\mathrm{nd}}$ resonance, respectively. $T = 6$ K, as in (d-g). (d) Simulated Fermi energy distribution $E_F(x)$ (solid lines) and Seebeck coefficient $S(x)$ (dotted lines) across the channel for the first (blue) and second (red) gating voltage, i.e., $-1.6~\text{V}$ and $-0.6~\text{V}$, highlighting the carrier density modulation induced by asymmetric gating. (e) Calculated local optical absorption profile for the same gate voltages, showing enhanced absorption in the gap region and under the right gate. (f) Simulated electronic temperature increase $\Delta T$ along the channel for the two resonances, illustrating localized hot-electron formation in regions of high field confinement. (g) Simulated total optical absorption as a function of right-gate voltage for several fixed left-gate (LG) voltages, revealing gate-tunable AGP resonances (the bidirectional arrow highlights the used LGs). Illumination frequency is 2.5 THz. The electron and hole mobilities under the left and right antenna gates were chosen in agreement with values extrapolated from fits of experimental data in Supplementary Fig. S1; under the antenna gap, the mobility was assumed to vary linearly between these values. (h) Experimentally measured photovoltage at 2.5 THz and 18 mW incident power as a function of right-gate voltage at several fixed left-gate (LG) voltages, showing resonant features consistent with AGP Fabry--Pérot standing waves in the graphene channel.
• Figure 2: Gate- and temperature-dependent resistance, photovoltage, and AGP resonances. (a) Two-dimensional resistance map as a function of left- and right-gate voltages at $T=6$ K, measured with $I_{\mathrm{bias}}=200$ nA. (b) Corresponding photovoltage $V_{\mathrm{ph}}$ map at $T=6$ K under $f=2.5$ THz illumination, showing the characteristic sixfold polarity pattern of the PTE effect and sharp resonant features associated with AGPs. (c) Same measurement at $T=130$ K, where AGP resonances are strongly suppressed by increased plasmon damping. (d,e) Line cuts of $V_{\mathrm{ph}}$ vs $V_{\mathrm{RG}}$ for several temperatures at fixed $V_{\mathrm{LG}}=-6$ V (d) and $V_{\mathrm{LG}}=0$ V (e), revealing two prominent AGP resonances around $V_{\mathrm{RG}}\approx-1.2$ V and $0.3$ V that gradually weaken with increasing $T$ and are enhanced when Seebeck asymmetry is maximized. (f) False-color map of $|V_{\mathrm{ph}}|$ vs $V_{\mathrm{RG}}$ and temperature at $V_{\mathrm{LG}}=-6$ V, highlighting the progressive damping and slight redshift of the AGP resonances with increasing $T$.
• Figure 3: (a) Comparison between measured photovoltage (PV/filled symbols) and simulated absorption (Th/open symbols) peak positions for the first and second AGP resonances in the split-gated graphene device. Blue and red symbols correspond to the first and second resonances, respectively. (b,c) Colored curves correspond to the AGP dispersions calculated at the right-gate voltages corresponding to the first (b) and second (c) resonances extracted from the photovoltage spectra. Horizontal dashed lines mark the Fabry--Pérot conditions: $\lambda_{\mathrm{p}}^{\mathrm{eff}} = L/2$ for a mode spanning the full channel (b), and $\lambda_{\mathrm{p}} = L/2$ for a mode confined mainly to the right half of the channel, where $\lambda_{\mathrm{p}}$ denotes the local plasmon wavelength under the right gate (c). Vertical dashed lines with arrows indicate the experimental resonance frequency at $\sim 2.5~\text{THz}$.
• Figure 4: Power dependence of the device photoresponse at $V_{\mathrm{LG}} = -6$ V. (a) Measured photovoltage as a function of right-gate voltage $V_{\mathrm{RG}}$ for different incident powers, controlled using two wire-grid polarizers. (b) Independent power dependence obtained by tuning the output power of the laser source, showing a consistent trend over more than one order of magnitude in excitation intensity. (c) Log–log plot of the peak photovoltage as a function of incident power, measured at fixed $V_{\mathrm{RG}} = -1.7$ V and $V_{\mathrm{LG}} = -6$ V. The nearly linear scaling across almost two decades demonstrates that the device operates in the linear response regime, ensuring predictable and stable performance suitable for THz detection applications.
• Figure 5: Frequency dependence of device responsivity. (a) Gate-dependent photovoltage measured at T = 6 K, $f = 2.5$ THz (blue) and $f = 1.8$ THz (red), showing reduced response at lower frequency due to weaker antenna coupling. (b) Room-temperature peak responsivity as a function of excitation frequency for incident light polarized parallel (red symbols) and perpendicular (blue symbols) to the antenna axis. The green dashed line shows simulated normalized near-field enhancement $|E/E_0|^2$ at the graphene plane (right axis), demonstrating excellent agreement between experiment and simulations. The responsivity maximum coincides with the designed dipole antenna resonance at $\sim 2.5$ THz, confirming that the device performance is dominated by antenna-mediated coupling.