insight review articles Surface plasmon subwavelength optics e structure of a x(the as this rai SPs. he use length structures s momen tivities,, of the m ne frequency media. (e is the square o 824insight review articles 824 NATURE | VOL 424 | 14 AUGUST 2003 | www.nature.com/nature Surface plasmons (SPs) are of interest to a wide spectrum of scientists, ranging from physicists, chemists and materials scientists to biologists. Renewed interest in SPs comes from recent advances that allow metals to be structured and characterized on the nanometre scale. This in turn has enabled us to control SP properties to reveal new aspects of their underlying science and to tailor them for specific applications. For instance, SPs are being explored for their potential in optics, magneto-optic data storage, microscopy and solar cells, as well as being used to construct sensors for detecting biologically interesting molecules. SPs were widely recognized in the field of surface science following the pioneering work of Ritchie in the 1950s(ref.1). SPs are waves that propagate along the surface of a conduc￾tor, usually a metal. These are essentially light waves that are trapped on the surface because of their interaction with the free electrons of the conductor (strictly speaking, they should be called surface plasmon polaritons to reflect this hybrid nature2 ). In this interaction, the free electrons respond collectively by oscillating in resonance with the light wave. The resonant interaction between the surface charge oscillation and the electromagnetic field of the light constitutes the SP and gives rise to its unique properties. For researchers in the field of optics, one of the most attractive aspects of SPs is the way in which they help us to concentrate and channel light using subwavelength struc￾tures. This could lead to miniaturized photonic circuits with length scales much smaller than those currently achieved3,4. Such a circuit would first convert light into SPs, which would then propagate and be processed by logic elements, before being converted back into light. To build such a cir￾cuit one would require a variety of components: wave￾guides, switches, couplers and so on. Currently much effort is being devoted to developing such SP devices; one example is the 40 nm thick gold stripe that acts as a waveguide for SPs in Fig. 1. An appealing feature is that, when embedded in dielectric materials, the circuitry used to propagate SPs can also be used to carry electrical signals. Developments such as this raise the prospect of a new branch of photonics using SPs, sometimes called plasmonics. The use of SPs to help us concentrate light in subwave￾length structures stems from the different (relative) permit￾tivities, ;, of the metals and the surrounding non-conducting media. (; is the square of the complex index of refraction.) Concentrating light in this way leads to an electric field enhancement that can be used to manipulate light–matter interactions and boost non-linear phenomena. For example, metallic structures much smaller than the wavelength of light are vital for the massive signal enhancement achieved in sur￾face-enhanced Raman spectroscopy (SERS)—a technique that can now detect a single molecule5,6. Furthermore, the enhanced field associated with SPs makes them suitable for use as sensors, and commercial systems have already been devel￾oped for sensing biomolecules. SP-based sensing applications and SERS will not be discussed further here because they are covered by other reviews7,8. Here we provide an overview of the properties of SPs and indicate why they are being considered for subwavelength optics. We examine how their propagation can be manipu￾lated and discuss some of the optical components that have so far been demonstrated. We conclude by highlighting the practical potential of this field and indicate just how much work remains to be done for that potential to be realized. Coupling to surface plasmons The interaction between the surface charges and the electro￾magnetic field that constitutes the SP has two consequences (see Box 1). First, the interaction between the surface charge density and the electromagnetic field results in the momen￾tum of the SP mode, ùkSP, being greater than that of a free-space photon of the same frequency, ùk0. (k04v/c is the free-space wavevector.) Solving Maxwell’s equations under the appropri￾ate boundary conditions yields the SP dispersion relation9 , that is, the frequency-dependent SP wave-vector, kSP, kSP=k0 !};d ; & d;m ;m § } (1) The frequency-dependent permittivity of the metal, ;m, and the dielectric material, ;d, must have opposite signs if SPs are to be possible at such an interface. This condition is satis￾fied for metals because ;m is both negative and complex (the latter corresponding to absorption in the metal). As an exam￾ple, using equation (1), the SP wavevector for a silver–air interface in the red part of the visible spectrum is found to be kSP>1.03.k0. This increase in momentum is associated with the binding of the SP to the surface, and the resulting momen￾tum mismatch between light and SPs of the same frequency must be bridged if light is to be used to generate SPs. Surface plasmon subwavelength optics William L. Barnes1 , Alain Dereux2 & Thomas W. Ebbesen3 1 School of Physics, University of Exeter, EX4 4QL, UK (e-mail: w.l.barnes@ex.ac.uk) 2 Laboratoire de Physique, Université de Bourgogne, BP 47870, F-21078 Dijon, France (adereux@u-bourgogne.fr) 3 ISIS, Université Louis Pasteur, BP 70028, F-67083, Strasbourg Cedex, France (e-mail: ebbesen@isis-ulp.org) Surface plasmons are waves that propagate along the surface of a conductor. By altering the structure of a metal’s surface, the properties of surface plasmons—in particular their interaction with light—can be tailored, which offers the potential for developing new types of photonic device. This could lead to miniaturized photonic circuits with length scales that are much smaller than those currently achieved. Surface plasmons are being explored for their potential in subwavelength optics, data storage, light generation, microscopy and bio-photonics. © 2003 Nature PublishingGroup