Miniaturizing Terahertz Antennas with Graphene Surface Plasmons

26 concepts13 papers1 proposal

As wireless communications push into the Terahertz (THz) band (0.1–10 THz) to achieve terabit-per-second data rates, engineers face a brutal physical bottleneck. High-frequency signals suffer from extreme path loss and atmospheric absorption. To overcome this, systems must use "ultra-massive" antenna arrays to focus signals into highly directional beams.

However, traditional metallic antennas are bound by conventional electromagnetic scaling: their physical size must be roughly proportional to the free-space wavelength ($\lambda_0$) of the signal. Packing thousands of metallic antennas onto a millimeter-scale chiplet or inside a compact computing package is physically impossible due to silicon area constraints.

To break this scaling barrier, researchers are turning to a novel mechanism: confining electromagnetic waves into surface surface waves using 2D materials. By leveraging graphene's unique physics, antennas can be shrunk by up to two orders of magnitude, enabling dense wireless integration on chiplets and even inside the human body.

The Physics of Compression: Surface Plasmon Polaritons

Why can graphene antennas be so much smaller than metallic ones operating at the exact same frequency? The answer lies in Surface Plasmon Polaritons (SPPs).

In a conventional metallic antenna, high-frequency currents move through the bulk metal, and the antenna radiates waves directly into free space. The resonant length of the antenna is strictly tied to the speed of light in the surrounding medium.

Graphene behaves differently. When an electromagnetic wave hits a graphene sheet at THz frequencies, it couples with the collective oscillations of the material's free electrons. This coupled state of electromagnetic waves and electron excitation is an SPP. Instead of radiating immediately into space, the wave becomes tightly bound to the 2D surface of the graphene.

This tight surface confinement drastically slows down the wave. Because the wave velocity drops significantly below the speed of light, its effective wavelength ($\lambda_{spp}$) compresses to a fraction of its free-space wavelength ($\lambda_0$). As detailed by Sergi Abadal et al. in Electromagnetic Nanonetworks Beyond 6G: From Wearable and Implantable Networks to On-Chip and Quantum Communication, this wavelength compression allows plasmonic antennas to be 10 to 100 times smaller than their metallic counterparts while maintaining a perfect resonant response.

Alice Faisal et al. explain in Ultramassive MIMO Systems at Terahertz Bands: Prospects and Challenges that this high confinement factor is what makes Ultra-Massive MIMO (UM-MIMO) physically viable. By shrinking individual elements down to the micrometer scale, thousands of antennas can be packed into a millimeter-scale footprint, as also noted by Hadi Sarieddeen et al. in Terahertz-Band Ultra-Massive Spatial Modulation MIMO.

Conventional Metallic Antenna:
[ Wavelength λ₀ ] ---> Dictates large physical size

Graphene Plasmonic Antenna:
[ Wave + Electron Coupling ] ---> [ Compressed Wavelength λ_spp ] ---> 10x-100x Smaller Antenna

Overcoming Material Limits with Stacked Architectures

While the theoretical physics of graphene antennas are compelling, translating them into real-world hardware introduces manufacturing hurdles. Pristine, defect-free graphene is typically isolated in tiny flakes, whereas practical manufacturing requires scalable methods like Chemical Vapor Deposition (CVD). However, CVD graphene often suffers from lower crystal quality, which dampens SPP waves and degrades antenna performance.

To solve this, the authors of A THz graphene-on-hBN stack patch antenna for future 6G communications (Elana Pereira de Santana et al.) experimentally demonstrated a "stack patch" configuration. By sandwiching a dielectric layer like hexagonal boron nitride (hBN) between two layers of CVD monolayer graphene, the multi-layer architecture compensates for material imperfections.

Crucially, this design is fabricated on polyimide substrates using standard microelectronic techniques like photolithography and atomic layer deposition (ALD). This makes the entire fabrication process compatible with Back-End-of-Line (BEOL) CMOS manufacturing, proving that these miniaturized antennas can be monolithically integrated alongside traditional silicon components.

System Integration and the Interconnect Challenge

A microscopic antenna is only useful if it can efficiently receive signals from and send signals to the rest of a transceiver circuit. This requires solving two practical engineering challenges: physical routing and impedance matching.

Vertical Routing via Through-Silicon Vias

To maintain a compact footprint, researchers avoid lateral wiring by routing signals vertically. Elana Pereira de Santana et al. demonstrate in Integrated Graphene Patch Antenna for Communications at THz Frequencies that Through-Silicon Vias (TSVs) or through-substrate vias can connect a graphene antenna's ground pads directly to a back-metal plane. This vertical integration allows transceiver chips to be bonded directly beneath the antenna array.

For communication within computing packages, vertical monopole antennas built directly out of TSVs can also be used to radiate signals horizontally across the plane of the die, as shown by Xavier Timoneda et al. in Channel Characterization for Chip-scale Wireless Communications within Computing Packages and Fátima Rodríguez-Galán et al. in Towards Spatial Multiplexing in Wireless Networks within Computing Packages.

The Impedance Bottleneck

The interface between the antenna and the signal source remains a critical bottleneck. As highlighted by Seyed Ehsan Hosseininejad et al. in MAC-Oriented Programmable Terahertz PHY via Graphene-based Yagi-Uda Antennas, there is a severe impedance mismatch between highly resistive graphene structures and standard high-frequency hardware.

Furthermore, when using graphene as a sensor or detector, the efficiency of the system depends entirely on this interface. As noted by Jianlong Liu et al. in Recent Progress in the Development of Graphene Detector for Terahertz Detection, optimizing the integrated antenna geometry is essential to maximize THz wave coupling and absorption, which directly dictates the detector's overall responsivity.

Applications Unlocked by Plasmonic Miniaturization

By shrinking THz communication hardware down to the micrometer scale, this technology enables applications that were previously impossible due to size constraints.

Wireless Plasticity in Multi-Chip Processors

Modern computing is shifting from massive monolithic chips to collections of smaller "chiplets" inside a single package. However, traditional physical wires create a rigid communication bottleneck.

Joshua Alexander Harrison Klein et al. in Architecting more than Moore: wireless plasticity for massive heterogeneous computer architectures (WiPLASH) and Robert Guirado et al. in Graphene-Based Wireless Agile Interconnects for Massive Heterogeneous Multi-Chip Processors demonstrate that graphene's tiny footprint allows wireless transceivers to be placed directly onto individual chiplets without wasting valuable silicon area. This creates an intra-package wireless network, allowing chiplets to dynamically reconfigure their data paths in real time to handle AI and high-performance computing workloads.

Body-Centric Nano-Networks

Because SPP waves allow functional THz antennas to be scaled down to the size of a single biological cell, they open up new frontiers in medicine. As explored by Nasir Saeed et al. in Body-Centric Terahertz Networks: Prospects and Challenges, these micrometer-scale plasmonic structures make it possible to build functional nanomachines. These devices are small enough to safely travel through the human circulatory system, enabling real-time, in-body medical sensing and targeted diagnostic networks.

Key Foundational Work

Terahertz-Band Ultra-Massive Spatial Modulation MIMO (2019) — contributed 1 concept elements to this direction.
MAC-Oriented Programmable Terahertz PHY via Graphene-based Yagi-Uda Antennas (2018) — contributed 1 concept elements to this direction.
Channel Characterization for Chip-scale Wireless Communications within Computing Packages (2018) — contributed 1 concept elements to this direction.
History & citationVersion 1 · 2026-10-04

Current · Permanent link

View frozen version 1

Cite this direction

Latest version (a living page):

@misc{lacuna_plasmonic_antenna_miniaturization_af0410fc,
  title = {plasmonic antenna miniaturization},
  author = {{Lacuna}},
  howpublished = {Lacuna research direction},
  year = {2026},
  url = {https://lacuna.tiptreesystems.com/direction/plasmonic-antenna-miniaturization/rdr_af0410fc072a43ca99483e3f968c8786},
  note = {Latest version of a living research direction; cite a version for a fixed text. Accessed 2026-11-19},
}

This exact version (version 1, frozen text):

@misc{lacuna_plasmonic_antenna_miniaturization_af0410fc_v1,
  title = {plasmonic antenna miniaturization},
  author = {{Lacuna}},
  howpublished = {Lacuna research direction},
  year = {2026},
  url = {https://lacuna.tiptreesystems.com/direction/plasmonic-antenna-miniaturization/rdr_af0410fc072a43ca99483e3f968c8786/v/1},
  note = {Version 1, frozen 2026-10-04. Accessed 2026-11-19},
}
About

These directions overlap with this one through shared papers.

See something wrong? Report a mistake