A Plasmon’s Life in Atomic Wires

By Tom Sharp on November 30, 2016

Are plasmons hosted in atomically thin metallic wires a possible solution for future waveguides? Find out more from Professor Herbert Pfnür, author of a recent article in the Journal of Physics: Condensed Matter special issue on Low Dimensional Order Mediated by Interfaces.

Waveguiding is commonly used to efficiently transport energy and information in the form of wave-packets from one location to another without the need of a backward channel. You may use this concept every day in the form of optical fibers in your home stereo. At the lower nanometer or even sub-nanometer scale, such as future nanoelectronics, efficient transport needs compatible wavelengths and appropriate waveguides. As demonstrated our recent publications [1, 2], the smallest imaginable metallic conduction channels of one or two atoms wide are suitable for hosting quasi-one-dimensional plasmons.

Plasmons are collective longitudinal density oscillations of the conduction electrons, with strong similarities to classical water waves in a simplified model. Among these, quasi-1D plasmons have highly attractive properties: they exhibit a quasi linear, light-like dispersion relation. Thus, wave-packets as the carrier of information do not disperse and keep their form. Additionally, their wavelengths are typically only a few nanometers, i.e. two orders of magnitude shorter than light. As a result, if light is converted into such plasmons, true nanoscale optoelectronics become feasible. (A possible application could be a data bus between two chips for densely packed informational exchange, i.e. high data bandwidth.)

Stick-and-ball sketch of Au-induced atomic wires (yellow balls) on a Si(553) surface. The conduction channel is depicted by the golden background stripe. Taken from J. Phys. Cond. Mat. 28 354001. © IOP Publishing 2016. All rights reserved.

Not much is known about the properties of low-D plasmons as yet. First of all, free-standing atomic chains suffer from instabilities – the world record is is 8 atoms length. Fortunately, extended 1D wires made of various elements such as Ag or Au can be made via self-assembly on stabilizing surfaces of undoped semiconductors like silicon or germanium. Such a substrate system exists: Si(hhk), a Si(111) surface with a miscut in the [11]-direction, leading to surfaces of small terraces separated by atomic-sized and well-oriented steps. Examples of such stable surface systems are Si(553), Si(775), and Si(557).

Upon adsorption of Ag on Si(557) [3,4], it is possible to obtain several nanometer wide quantum wires that, apart from their almost linear dispersion, also exhibit exotic properties like quantum well states perpendicular to their axis and defect-free 1D self-doping. A mono- or diatomic chain on each terrace inducing a conducting channel of only several angstroms (0.1nm) width is possible for Au adsorption [1,2]. Here we observe a strong coupling between the wires, i.e. bending the dispersion curve and an intriguing interplay of 1D and 2D properties due to the spill-out of the electron density. These plasmons have a surprisingly long lifetime, at least 100 fs, so that they can travel over distances of a few hundred nanometers, enough for information exchange at the nanoscale.

While these are the first steps in demonstrating the possibilities of quasi-1D plasmonics, many remaining challenges have to be addressed, including efficient coupling of energy into and out of the structures, and the fabrication of arbitrarily shaped waveguiding structures.

Our activities are embedded into a more general research unit funded by the Deutsche Forschungsgemeinschaft called FOR1700. In close collaboration with several other groups in Germany, we are trying to discover the physics that hides in the low-dimensional world.

References

T. Lichtenstein, C. Tegenkamp, and H. Pfnür, J. Phys. Condens. Matter 28, 354001 (2016).
T. Lichtenstein, J. Aulbach, J. Schäfer, R. Claessen, C. Tegenkamp, and H. Pfnür, Phys. Rev. B 93, 161408 (2016).
U. Krieg, T. Lichtenstein, C. Brand, C. Tegenkamp, and H. Pfnür, New J. Phys. 17, 43062 (2015).
U. Krieg, C. Brand, C. Tegenkamp, and H. Pfnür, J. Phys. Condens. Matter 25, 14013 (2013).