Sound-Wave-Like Collective Electronic Excitations in Au Atom Chains Tadaaki NAGAO 1;5 , Shin YAGINUMA 1 , Takeshi INAOKA 2 , Toshio SAKURAI 3 , and Dongryul JEON 4 1 Nano System Functionality Center, National Institute for Materials Science, Tsukuba, Ibaraki 305-0044 2 Department of Materials Science and Engineering, Iwate University, Morioka 020-8551 3 Institute for Materials Research, Tohoku University, Sendai 980-8577 4 Department of Physics Education, Seoul National University, Seoul 151-742, Korea 5 ICORP, Japan Science and Technology Agency (JST), Kawaguchi, Saitama 332-0023 (Received May 13, 2007; revised September 12, 2007; accepted September 19, 2007; published November 12, 2007) Sound-wave-like collective excitation in Au atom chains on the Si(557) surface is investigated. Electron scattering spectroscopy using a highly collimated slow electron beam has detected a characteristic low-energy one-dimensional (1D) plasmon (wire plasmon) that is confined in the atom chains. Theoretical analysis adopting a quantum-mechanical scheme beyond the free-electron model indicates a significant dynamic exchange-correlation effect due to strong 1D confinement. By cooling to below 100 K, we have detected for the first time a significant change in the plasmon dispersion in a tiny momentum and energy region, which definitely reflects a gap opening due to a metal-to-insulator transition of the atom chains. KEYWORDS: atom chain, one dimension, plasmon DOI: 10.1143/JPSJ.76.114714 1. Introduction Because of the recent impact on the applications of ‘‘surface plasmons’’ in the fields of subwavelength optics, data storage, and biophotonics, there has been increasing importance in exploring the momentum–energy (q–h !) relation, or plasmon dispersion, of metallic nanostructures. 1) The concept of the surface plasmon is appropriate when the size of a metallic object is sufficiently large compared with the Fermi wavelength F , and the three-dimensional (3D) charge density oscillation of the surface plasmon extends relatively deep into the bulk. 2) On the other hand, when metal objects are downsized to the scale of F , for example, when the plasmon is strictly confined to atomically thin two-dimensional (2D) objects or atomically narrow one-dimensional (1D) ones, it is expected that the plasmon behaves in an essentially different manner from a plasmon in larger-scale 3D objects (Fig. 1). One of the most exciting effects will be that plasmon resonance in atomic scale low-dimensional objects is predicted to lie in the infrared spectral region, while that in the bulk or at the surface is in the visible region. Also, electronic correlation effect should be rather strong compared to bulk and surface plasmons because of the stronger electron confinement. Experimental realization and measurement for such systems are of great scientific interest and have significant impact also on the material science as well as on nano-photonics device technology. Particularly, a plasmon in one dimension is highly intriguing since it has been theoretically predicted by Tomonaga that the single-particle picture breaks down and a plasmon picture comes to replace it, which manifests itself as a low-energy sound-wave-like excitation, 3) leading to many exotic properties [Figs. 2(a) and 2(b)]. Such behavior will be clearly observed when the electronic band of the atom chain is free from coupling to the electronic states of Fig. 1. (Color online) Schematic illustration of the charge density oscil- lation of plasmons in metallic nanoscale objects. From top to bottom, smallness of the object increases and the dimensionality becomes lower. The uppermost row shows bulk and surface plasmons where the object size is larger than the Fermi wavelength F . The second row shows the intermediate case where the interference between the two surfaces becomes strong. The last row shows the ultimate case where the thickness of the object becomes atomic scale. When the object size reaches to this regime and becoming comparable to the Fermi wavelength F , shape and size effects manifest themselves most dramatically. E-mail: NAGAO.Tadaaki@nims.go.jp Journal of the Physical Society of Japan Vol. 76, No. 11, November, 2007, 114714 #2007 The Physical Society of Japan 114714-1