Plasmon propagation pushed to the limit

Excitations called plasmons have the potential to miniaturize photonic devices, but are often short-lived. Microscopy reveals that plasmons in the material graphene can overcome this limitation at low temperatures.
Justin C. W. Song is in the Division of Physics and Applied Physics, Nanyang Technological University, Singapore 637371, Singapore, and the Institute of High Performance Computing, A*STAR, Singapore.

Search for this author in:

Light can be confined and steered at the nanoscale using collective oscillations of electrons known as plasmons. But just as death and taxes are the only certainties in life, energy loss is the only certainty in plasmonics. The tighter the confinement of light, the shorter the lifetime of the plasmons1 — a trade-off that is a major hurdle in the practical use of these oscillations. In a paper in Nature, Ni et al.2 use a technique called scanning near-field optical microscopy to study plasmons in a single layer of carbon atoms known as graphene, at cryogenic temperatures (60 kelvin). The authors show that the plasmons can produce extremely compact light confinement while retaining long lifetimes. They use their results to determine the fundamental limits of plasmon propagation in graphene.

The propagation of light involves the oscillation of electric and magnetic fields. This oscillation defines the relationship between the frequency and wavelength of light, and underpins the diffraction limit — the fact that, in free space, light spreads out if it passes through a region narrower than its wavelength. When light interacts with plasmons, its speed can be substantially reduced, which allows it to be confined to distances much smaller than its free-space wavelength. As a result, plasmons have become a versatile tool for controlling the behaviour of light at the nanoscale3. However, the same light–plasmon interaction that can confine light below its wavelength also enables energy to be lost through the scattering of electrons.

Noble metals such as silver and gold are conventionally used in plasmonics, but suffer from high losses. In the past few years, 2D materials have become promising alternatives4. In the case of graphene, plasmons can compress light to distances as small as one-three-hundredth of the light’s free-space wavelength5. Furthermore, the electron density of graphene can be readily controlled, which provides direct electrical means of tuning the properties of the plasmons. But although sustained efforts to improve the quality of graphene have yielded steady advances6, plasmon loss remains substantial.

In a bid to push the limits of plasmon propagation, Ni and colleagues launched and imaged plasmons in a device containing high-quality graphene at cryogenic temperatures. The use of these temperatures minimized losses caused by temperature-sensitive processes, such as the scattering of electrons from mechanical vibrations called phonons. The authors customized an instrument known as a scanning near-field optical microscope so that it could operate at cryogenic temperatures. Although these instruments are routinely used to study plasmons at room temperature, operating them at lower temperatures has been difficult.

Ni et al. used the narrow metallic tip of the microscope to launch plasmons in the graphene device. They then scanned the tip across the device to image the interference pattern produced by the plasmons as they reflected from the edges of the device and from microstructures present on the device’s surface (Fig. 1). This technique is particularly useful because it launches plasmons in the interior of the device, which limits losses caused by interaction with the device’s edges. Such losses can be large in other approaches7.

Figure 1 | Low-temperature plasmons investigated in graphene. Ni et al.2 used an instrument known as a scanning near-field optical microscope to study exotic excitations called plasmons. They used the narrow metallic tip of the microscope to launch plasmons in a device containing the material graphene, at cryogenic temperatures (60 kelvin). The authors then scanned the tip across the device to image the interference pattern produced by the plasmons as they were reflected (white arrows) from the edges of the device and from microstructures present on the device’s surface. The interference pattern consisted of bright and dark bands that were found throughout the device, demonstrating that the plasmons could travel several micrometres before their energy was lost. Such long-lived plasmons could have many applications. Scale bar, 1 µm. (Adapted from Fig. 1c of ref. 2.)

The fruits of Ni and colleagues’ labour are pronounced plasmon interference fringes (bright and dark bands) that are found throughout the device and that extend several micrometres from any boundaries. These fringes make the entire device ‘light up’ with a characteristic washboard-like pattern. The plasmons simultaneously have relatively long lifetimes (reaching 1.6 picoseconds; 1 ps is 10–12 seconds) and confine light to distances smaller than one-sixtieth of the free-space wavelength. Their quality factor, a measure of energy retention, is 130, which is a record for plasmons that enable such compact light confinement. The performance of the plasmons therefore bucks the trade-off between tight confinement and high loss. It is possible thanks to the extremely high quality of the authors’ graphene device, which contains highly mobile electrons that can travel several micrometres without scattering.

Remarkably, using a combination of detailed modelling and systematically collected temperature-dependent data, Ni and colleagues determined that the main cause of energy loss at low temperatures was not electron scattering in the graphene. Instead, plasmon loss arose mostly from insulating material that surrounded the graphene. The quality of the plasmons could therefore be improved by reducing these extrinsic losses, for example by altering this insulating material. The authors also suggest that the intrinsic (fundamental) limits of plasmon propagation at cryogenic temperatures have not yet been reached. They calculate that it might be possible to achieve quality factors more than seven times higher than the one reported in the current paper.

Nevertheless, the exceptional quality of Ni and colleagues’ graphene plasmons sets a new standard for nanophotonic platforms. Tightly confined light in such plasmons can now be thought of as being highly stable, with the ability to be directed and steered across distances of several micrometres. The possibilities for the future are vast and range from the fundamental (such as probing the topological8 and geometrical structure9 of plasmons) to the applied (including nanoscale plasmon lasers10, sensitive light detectors, sub-wavelength routing of light, and nanoscale optical interconnects3). The authors’ high-quality graphene plasmons, combined with recently developed techniques to substantially reduce the overall size of plasmons11, make a compelling case for graphene-based nanophotonics.

Perhaps most exciting, however, is the prospect of using scanning near-field optical microscopy at cryogenic temperatures to probe excitations other than plasmons. Phases of matter such as superconductors, ferromagnets and antiferromagnets possess excitations that could be accessed using this technique12. In the past few years, a wide range of these phases has been discovered on 2D materials, on which the surfaces are fully exposed and are therefore easily accessible. Such phases manifest only at low temperatures, making cryogenic operation the key to launching the excitations and studying their intricate dynamics.

Nature 557, 501-502 (2018)

doi: 10.1038/d41586-018-05190-1
Nature Briefing

Sign up for the daily Nature Briefing email newsletter

Stay up to date with what matters in science and why, handpicked from Nature and other publications worldwide.

Sign Up


  1. 1.

    Khurgin, J. B. Nature Nanotech. 10, 2–6 (2015).

  2. 2.

    Ni, G. X. et al. Nature 557, 530–533 (2018).

  3. 3.

    Atwater, H. A. Sci. Am. 296, 56–63 (2007).

  4. 4.

    Boltasseva, A. & Atwater, H. A. Science 331, 290–291 (2011).

  5. 5.

    Lundeberg, M. B. et al. Science 357, 187–191 (2017).

  6. 6.

    Woessner, A. et al. Nature Mater. 14, 421–425 (2015).

  7. 7.

    Yan, H. et al. Nature Photon. 7, 394–399 (2013).

  8. 8.

    Jin, D. et al. Nature Commun. 7, 13486 (2016).

  9. 9.

    Shi, L.-K. & Song, J. C. W. Phys. Rev. X 8, 021020 (2018).

  10. 10.

    Bergman, D. J. & Stockman, M. I. Phys. Rev. Lett. 90, 027402 (2003).

  11. 11.

    Alcaraz Iranzo, D. et al. Science 360, 291–295 (2018).

  12. 12.

    Basov, D. N., Fogler, M. M. & García de Abajo, F. J. Science 354, aag1992 (2016).

Download references