The 2018 demonstration of a plasmon-assisted electro-optic ring modulator (Haffner et al., Nature 556, 483–486) showed that 'resonant switching' — coupling light to lossy surface plasmon polaritons only in the resonant off state — lets a plasmonic modulator combine low on-chip optical loss, operation beyond 100 GHz, femtojoule-per-bit energy efficiency and a micrometre footprint. The paper itself stated that the full projected performance had not yet been verified experimentally: sub-1-V drive and lower insertion loss were expected from lower-loss plasmonic metals (silver or CMOS-compatible copper) and higher-r33r_{33} organic electro-optic (OEO) materials, the intrinsic bandwidth (RC-limited beyond 1 THz) was unverified because the electrical measurement ceiling sat at 110–115 GHz, and system-level fibre-to-chip losses, contact-pad parasitics, OEO material stability and fabrication yield remained open. Later work has resolved large parts of this agenda — most directly the 2023 plasmonic micro-racetrack modulator (176 GHz bandwidth, 1.2 dB on-chip loss, 0.6 Vp drive, 85 °C operation) and 2019/2025 demonstrations of 500 GHz to beyond-1 THz bandwidth in plasmonic-organic-hybrid Mach–Zehnder modulators. The surviving open core is precise: experimental verification of terahertz intrinsic bandwidth for the resonant (ring/racetrack) geometry itself, experimental integration of lower-loss plasmonic metals (Ag or Cu) in resonant electro-optic modulators, closure of the system-level link budget (fibre-to-chip coupling and pad parasitics), and formal long-term reliability qualification of the poled OEO material stack, including compatibility with CMOS back-end processing.

Background

Plasmonics studies devices in which light couples to the collective motion of free electrons at a metal surface; the resulting wave, a surface plasmon polariton (SPP), confines optical fields to deeply subwavelength scales but suffers ohmic (resistive) heating loss, which long made plasmonic devices appear impractical for information technology. In a plasmonic–organic hybrid (POH) electro-optic (EO) modulator, two metal electrodes form a metal–insulator–metal slot roughly 100 nm wide, filled with an organic electro-optic (OEO) material, a polymer host loaded with dipolar 'chromophore' molecules whose π-electron distribution gives a Pockels effect — an instantaneous, field-induced change of refractive index. The molecules are aligned ('poled') by an electric field near the polymer's glass-transition temperature TgT_{\mathrm{g}}, the temperature above which the polymer softens; the achieved molecular orientation is quantified by the electro-optic coefficient r33r_{33}. Because the optical SPP field and the applied radio-frequency (RF) field co-localize in the same nanoscale slot, POH phase shifters are ultra-compact and fast, but they are intrinsically lossy (order 0.4 dB per micrometre), and POH Mach–Zehnder modulators (MZMs) — interferometers that convert phase modulation to intensity modulation — had reached 100 GBd operation at 170 GHz electrical bandwidth before 2018 (Haffner et al. 2015; Hoessbacher et al. 2017).

Haffner et al. (Nature 556, 483–486; 2018) proposed 'resonant switching' to bypass the plasmonic loss: a microring resonator, a wavelength-scale circular waveguide in which light circulates and interferes with itself, is critically coupled to a through-waveguide so that in the off state (ring on resonance) light deliberately couples into the lossy plasmonic ring and is extinguished, giving a large extinction ratio (the on/off power ratio, ER), while in the on state (ring off resonance) destructive interference keeps the light in the low-loss photonic waveguide. The fabricated gold-electrode ring modulator, with an OEO slot material (a blend of 75% HD-BB-OH / 25% YLD124, poled at TgT_{\mathrm{g}} of 115 °C), demonstrated on-chip insertion loss (IL, the optical power penalty in the pass state) of about 2.5 dB, operation beyond 100 GHz, 72 Gbit/s data modulation, and about 12 fJ/bit energy consumption estimated for a calculated ring capacitance of ~1.1 fF. The measured electro-optic frequency response showed no limitation up to 110 GHz, with a kink at 110–115 GHz attributed by the authors to the electrical test equipment, not the device; the authors estimated the RC-limited bandwidth at far beyond 1 THz and the Pockels response as sub-picosecond, so the intrinsic bandwidth was unverified. The paper's Methods and Extended Data explicitly identified the route to the projected full performance: replacing gold with silver (lowest loss, with the resonator quality factor Q, the ratio of energy stored to energy dissipated per optical cycle, roughly doubled in simulation) or copper (CMOS-compatible), using the newest OEO material with an about three-times-larger r33r_{33}, and improving fabrication, which together were predicted to enable digital drive at 1 V peak-to-peak; it also acknowledged fibre-to-fibre losses of 16.5 dB (dominated by fibre-to-chip grating-coupler interfaces, the periodic surface gratings that convert between fibre and waveguide modes), total capacitances raised to several tens of fF by large contact pads interacting with the substrate, OEO operation limited to about 80 °C, and device-to-device fabrication variability in sub-micrometre ring geometries. A very substantial part of that agenda has since been achieved: the direct successor device, the resonant plasmonic micro-racetrack modulator of Eppenberger et al. (Nature Photonics 17, 360–367; 2023), demonstrated 176 GHz bandwidth, 1.2 dB on-chip loss, 0.6 Vp drive, 408 Gbit/s at 12.3 fJ/bit, and continuous operation at 85 °C using an OEO material with Tg≥170T_{\mathrm{g}} \ge 170 °C and r33=148r_{33} = 148 pm/V — but still with gold electrodes, 2.5 dB-per-coupler grating couplers, and reliability evidence ('strong indication' from a >5,000 h in-device test) that the authors themselves state falls short of the Telcordia GR-468 standard — the optoelectronic-component reliability qualification used in the telecommunications industry, whose high-temperature operating-life test calls for 2,000 h of continuous operation. Meanwhile, non-resonant POH MZMs have been measured to 500 GHz (Burla et al. 2019) and beyond 1 THz (Horst et al. 2025), and plasmonic modulators have been deployed in kilometre-range 226 GHz wireless links (Vukovic et al. 2026).

Problem Statement

The task is the experimental realization and verification of the full projected performance of the resonant plasmonic–organic hybrid electro-optic ring modulator of Haffner et al. (2018) — an engineering-validation objective whose components are the authors' own acknowledged future-work items. The target device class is any resonant plasmonic–organic hybrid (ring or racetrack) electro-optic modulator, with the 2018 gold ring and its racetrack successor counting equally. Parts of it have since been achieved and are not open: sub-1-V (0.6 V peak) drive, ~1.2 dB on-chip insertion loss, 176 GHz bandwidth and 85 °C operation with an improved OEO material have been demonstrated in the successor resonant plasmonic micro-racetrack modulator (Eppenberger et al. 2023), and bandwidths beyond 1 THz have been demonstrated for non-resonant plasmonic–organic hybrid Mach–Zehnder modulators (Burla et al. 2019; Horst et al. 2025). The surviving open core is: (1) experimentally verify the intrinsic electro-optic bandwidth of the resonant (ring or racetrack) plasmonic modulator geometry in the terahertz range — the 2018 device's RC limit was estimated far above 1 THz but the resonant device has only been measured to 176 GHz — separating genuine device limits from measurement-set-up artifacts, as the 110–115 GHz kink could not be resolved in 2018; (2) experimentally integrate lower-loss plasmonic metals, silver or CMOS-compatible copper, in a resonant (ring or racetrack) plasmonic–organic hybrid electro-optic modulator and demonstrate with that metal, rather than by inheriting the gold-electrode values, the simulated gains in quality factor, insertion loss and drive voltage — including the 2018-predicted 1 V peak-to-peak (0.5 V peak) digital drive — that no reported work has realized (all demonstrated resonant devices still use gold); (3) close the system-level link budget — reduce fibre-to-chip/grating-coupler losses (still 2.5 dB per coupler in the successor work) and mitigate the contact-pad parasitic capacitance that raises the device capacitance above the ~1.1 fF intrinsic resonator value — while retaining sub-1-V drive and femtojoule-per-bit energies; and (4) establish formal long-term reliability of the poled OEO material stack in resonant plasmonic devices, including GR-468-style continuous high-temperature lifetime qualification (2,000 h of continuous operation), temperature cycling (−40 °C to +85 °C) and damp-heat robustness (85 °C / 85 % relative humidity), operation beyond the 80 °C limit of the original material (85 °C is now demonstrated, with >5,000 h indicated but full qualification acknowledged as outstanding), and compatibility of the metal/OEO stack with CMOS back-end-of-line processing, the low-temperature metal-interconnect fabrication stages performed after the transistors on a chip are complete, together with a reported fabrication yield and device-to-device reproducibility across a set of nominally identical sub-micrometre resonator geometries. All four items are required, and none is individually sufficient.

References

  1. C. Haffner, D. Chelladurai, Y. Fedoryshyn, A. Josten, B. Baeuerle, W. Heni, T. Watanabe, T. Cui, B. Cheng, S. Saha, D. L. Elder, L. R. Dalton, A. Boltasseva, V. M. Shalaev, N. Kinsey, J. Leuthold, 'Low-loss plasmon-assisted electro-optic modulator', Nature 556, 483–486 (2018). DOI 10.1038/s41586-018-0031-4, https://doi.org/10.1038/s41586-018-0031-4
  2. M. Eppenberger, A. Messner, B. I. Bitachon, W. Heni, T. Blatter, P. Habegger, M. Destraz, E. De Leo, N. Meier, N. Del Medico, C. Hoessbacher, B. Baeuerle, J. Leuthold, 'Resonant plasmonic micro-racetrack modulators with high bandwidth and high temperature tolerance', Nature Photonics 17, 360–367 (2023). DOI 10.1038/s41566-023-01161-9, https://doi.org/10.1038/s41566-023-01161-9
  3. T. Blatter, Y. Horst, M. Eppenberger, C. Pappas, A. Tsakyridis, G. Giamougiannis, M. Kohli, U. Koch, J. Smajic, M. Moralis-Pegios, N. Pleros, J. Leuthold, 'Ring-assisted, racetrack, and Mach-Zehnder modulator: which one offers the lowest voltages and chirp-free operation?', Optics Express 32, 37968 (2024). DOI 10.1364/OE.532934, https://doi.org/10.1364/OE.532934
  4. Y. Horst, D. Moor, D. Chelladurai, T. Blatter, S. Fernandes, L. Kulmer, M. Baumann, H. Ibili, C. Funck, K. Keller, M. Destraz, W. Heni, L. Chérix, Y. Liu, H. Wang, S. M. Koepfli, J. Leuthold, 'Ultra-wideband MHz to THz plasmonic EO modulator', Optica 12, 325 (2025). DOI 10.1364/optica.544016, https://doi.org/10.1364/optica.544016
  5. M. Burla, C. Hoessbacher, W. Heni, C. Haffner, Y. Fedoryshyn, D. Werner, T. Watanabe, H. Massler, D. L. Elder, L. R. Dalton, J. Leuthold, '500 GHz plasmonic Mach–Zehnder modulator enabling sub-THz microwave photonics', APL Photonics 4, 056106 (2019). DOI 10.1063/1.5086868, https://doi.org/10.1063/1.5086868
  6. D. Bisang, Y. Horst, M. Thürig, K. Menachery, S. M. Koepfli, M. Kohli, E. De Leo, M. Destraz, V. Tedaldi, N. Del Medico, C. Hoessbacher, B. Baeuerle, W. Heni, J. Leuthold, 'Plasmonic Modulators in Cryogenic Environment Featuring Bandwidths in Excess of 100 GHz and Reduced Plasmonic Losses', ACS Photonics 11, 2691–2699 (2024). DOI 10.1021/acsphotonics.4c00507, https://doi.org/10.1021/acsphotonics.4c00507
  7. B. Vukovic, L. Kulmer, T. Blatter, Y. Horst, M. Destraz, W. Heni, S. M. Koepfli, H. Ibili, M. Baumann, Y. Fedoryshyn, J. Smajic, S. Luciano, J. Leuthold, 'Plasmonic modulator enabling kilometer-range high-throughput sub-THz links for radio access networks', Nature Communications 17, 5384 (2026). DOI 10.1038/s41467-026-72053-z, https://doi.org/10.1038/s41467-026-72053-z
  8. S. Ummethala, T. Harter, K. Koehnle, Z. Li, S. Muehlbrandt, Y. Kutuvantavida, J. Kemal, P. Marin-Palomo, J. Schaefer, A. Tessmann, S. K. Garlapati, A. Bacher, L. Hahn, M. Walther, T. Zwick, S. Randel, W. Freude, C. Koos, 'THz-to-optical conversion in wireless communications using an ultra-broadband plasmonic modulator', Nature Photonics 13, 519–524 (2019). DOI 10.1038/s41566-019-0475-6, https://doi.org/10.1038/s41566-019-0475-6
  9. C. Hoessbacher, A. Josten, B. Baeuerle, Y. Fedoryshyn, H. Hettrich, Y. Salamin, W. Heni, C. Haffner, C. Kaiser, R. Schmid, D. L. Elder, D. Hillerkuss, M. Möller, L. R. Dalton, J. Leuthold, 'Plasmonic modulator with >170 GHz bandwidth demonstrated at 100 GBd NRZ', Optics Express 25, 1762–1768 (2017). DOI 10.1364/OE.25.001762, https://doi.org/10.1364/OE.25.001762
  10. C. Haffner, W. Heni, Y. Fedoryshyn, J. Niegemann, A. Melikyan, D. L. Elder, B. Baeuerle, Y. Salamin, A. Josten, U. Koch, C. Hoessbacher, F. Ducry, L. Juchli, A. Emboras, D. Hillerkuss, M. Kohl, L. R. Dalton, C. Hafner, J. Leuthold, 'All-plasmonic Mach–Zehnder modulator enabling optical high-speed communication at the microscale', Nature Photonics 9, 525–528 (2015). DOI 10.1038/nphoton.2015.127, https://doi.org/10.1038/nphoton.2015.127
  11. Y. Zhang, Y. Yang, J. Liang, S. Wang, Y. Shi, P. Yao, H. Fu, J. B. Khurgin, F. Hu, J. Zhu, S. Zhu, L. Zhou, 'Approaching the low optical loss limit of plasmonics using potassium', Light: Science & Applications 15, 326 (2026). DOI 10.1038/s41377-026-02400-8, https://doi.org/10.1038/s41377-026-02400-8
  12. C. Kieninger, Y. Kutuvantavida, H. Miura, J. N. Kemal, H. Zwickel, F. Qiu, M. Lauermann, W. Freude, S. Randel, S. Yokoyama, C. Koos, 'Demonstration of long-term thermally stable silicon-organic hybrid modulators at 85 °C', Optics Express 26, 27955–27964 (2018). DOI 10.1364/OE.26.027955, https://doi.org/10.1364/OE.26.027955