Light usually propagates inside transparent materials in well known ways. However, recent research has examined the possibility of modifying the way the light travels by taking a normal transparent dielectric and inserting tiny metallic inclusions of various shapes and arrangements. As light passes through these structures, oscillating electric currents are set up that generate electromagnetic field moments; these can lead to dramatic effects on the light propagation, such as negative refraction. Possible applications include lenses that break traditional diffraction limits and 'invisibility cloaks' (refs 5, 6). Significantly less research has focused on the potential of such structures for slowing, trapping and releasing light signals. Here we demonstrate theoretically that an axially varying heterostructure with a metamaterial core of negative refractive index can be used to efficiently and coherently bring light to a complete standstill. In contrast to previous approaches for decelerating and storing light, the present scheme simultaneously allows for high in-coupling efficiencies and broadband, room-temperature operation. Surprisingly, our analysis reveals a critical point at which the effective thickness of the waveguide is reduced to zero, preventing the light wave from propagating further. At this point, the light ray is permanently trapped, its trajectory forming a double light-cone that we call an 'optical clepsydra'. Each frequency component of the wave packet is stopped at a different guide thickness, leading to the spatial separation of its spectrum and the formation of a 'trapped rainbow'. Our results bridge the gap between two important contemporary realms of science-metamaterials and slow light. Combined investigations may lead to applications in optical data processing and storage or the realization of quantum optical memories.

We study both linear and nonlinear surface waves localized at the interface separating a left-handed (LH) medium (i.e., a medium with both negative dielectric permittivity and negative magnetic permeability) and a conventional [or right-handed (RH)] dielectric medium. We demonstrate that the interface can support both TE- and TM-polarized surface waves-surface polaritons, and we study their properties. We describe the intensity-dependent properties of nonlinear surface waves in three different cases, i.e., when both the LH and RH media are nonlinear and when either of the media is nonlinear. In the case when both media are nonlinear, we find two types of nonlinear surface waves, one with the maximum amplitude at the interface, and the other one with two humps. In the case when one medium is nonlinear, only one type of surface wave exists, which has the maximum electric field at the interface, unlike waves in right-handed materials where the surface-wave maximum is usually shifted into a self-focusing nonlinear medium. We discuss the possibility of tuning the wave group velocity in both the linear and nonlinear cases, and show that group-velocity dispersion, which leads to pulse broadening, can be balanced by the nonlinearity of the media, so resulting in soliton propagation.

We introduce a recipe for producing a new type of electromagnetic cloak, the electromagnetic spacetime cloak, which conceals events rather than mere objects. Non-emitting events occurring during a restricted period are never suspected by a distant observer. The cloak works by locally manipulating the speed of light of an initially uniform light distribution, whilst the light rays themselves always follow straight paths. Any 'perfect' spacetime cloak would necessarily rely upon the technology of electromagnetic metamaterials, which has already been shown to be capable of deforming light in ways hitherto unforseen -to produce for example an electromagnetic object cloak. Nevertheless, we show how it is possible to use intensity dependent refractive indices to construct an approximate STC, an implementation that would enable the distinct signature of successful event cloaking to be observed. Potential demonstrations include systems that apparently violate quantum statistics, 'interrupt-without-interrupt' computation on convergent data channels, and the illusion of a Star-Trek transporter. * Electronic address: m.mccall@imperial.ac.uk

Metamaterial research is an extremely important global activity that promises to change our lives in many different ways. These include making objects invisible and the dramatic impact of metamaterials upon the energy and medical sectors of society. Behind all of the applications, however, lies the business of creating metamaterials that are not going to be crippled by the kind of loss that is naturally heralded by use of resonant responses in their construction. This review sets out some solutions to the management of loss and gain, coupled to controlled and nonlinear behavior, and discusses some critical consequences concerning stability. Under the general heading of active and tunable metamaterials, an international spectrum of authors collaborates here to present a set of solutions that addresses these issues in several directions. As will be appreciated, the range of possible solutions is really fascinating, and it is hoped that these discussions will act as a further stimulus to the field.

It is shown that a new type of metamaterial, a 3D-array of toroidal solenoids, displays a significant toroidal response that can be readily measured. This is in sharp contrast to materials that exist in nature, where the toroidal component is weak and hardly measurable. The existence of an optimal configuration, maximizing the interaction with an external electromagnetic field, is demonstrated. In addition, it is found that a characteristic feature of the magnetic toroidal response is its strong dependence on the background dielectric permittivity of the host material, which suggests possible applications. Negative refraction and backward waves exist in a composite toroidal metamaterial, consisting of an array of wires and an array of toroidal solenoids.

An expression for the electromagnetic field energy density in a dispersive lossy left-handed metamaterial, consisting of an array of split-ring resonators and an array of wires, is derived. An electromagnetic field with general time dependence is considered. The outcome is compared with previously published results. In the absence of losses, agreement with the general result for the energy density in a dispersive material is obtained. The formulae are verified using the finite-difference time-domain numerical method. The applicability of two commonly used permeability models to the problem of calculating the energy stored in an array of split-ring resonators is discussed.

Optical metamaterials have redefined how we understand light in notable ways: from strong response to optical magnetic fields, negative refraction, fast and slow light propagation in zero index and trapping structures, to flat, thin and perfect lenses. Many rules of thumb regarding optics, such as mu = 1, now have an exception, and basic formulas, such as the Fresnel equations, have been expanded. The field of metamaterials has developed strongly over the past two decades. Leveraging structured materials systems to generate tailored response to a stimulus, it has grown to encompass research in optics, electromagnetics, acoustics and, increasingly, novel hybrid materials responses. This roadmap is an effort to present emerging fronts in areas of optical metamaterials that could contribute and apply to other research communities. By anchoring each contribution in current work and prospectively discussing future potential and directions, the authors are translating the work of the field in selected areas to a wider community and offering an incentive for outside researchers to engage our community where solid links do not already exist.

No abstract