We investigated propagation of light through a uniaxial photonic metamaterial composed of three-dimensional gold helices arranged on a two-dimensional square lattice. These nanostructures are fabricated via an approach based on direct laser writing into a positive-tone photoresist followed by electrochemical deposition of gold. For propagation of light along the helix axis, the structure blocks the circular polarization with the same handedness as the helices, whereas it transmits the other, for a frequency range exceeding one octave. The structure is scalable to other frequency ranges and can be used as a compact broadband circular polarizer.

The following resources related to this article are available online at

No abstract

Arrays of gold split rings with a 50-nm minimum feature size and with an LC resonance at 200 THz frequency (1:5 m wavelength) are fabricated. For normal-incidence conditions, they exhibit a pronounced fundamental magnetic mode, arising from a coupling via the electric component of the incident light. For oblique incidence, a coupling via the magnetic component is demonstrated as well. Moreover, we identify a novel higher-order magnetic resonance at around 370 THz (800 nm wavelength) that evolves out of the Mie resonance for oblique incidence. Comparison with theory delivers good agreement and also shows that the structures allow for a negative magnetic permeability.

We observe second-harmonic generation from metamaterials composed of split-ring resonators excited at 1.5-micrometer wavelength. Much larger signals are detected when magnetic-dipole resonances are excited, as compared with purely electric-dipole resonances. The experiments are consistent with calculations based on the magnetic component of the Lorentz force exerted on metal electrons—an intrinsic second-harmonic generation mechanism that plays no role in natural materials. This unusual mechanism becomes relevant in our work as a result of the enhancement and the orientation of the local magnetic fields associated with the magnetic-dipole resonances of the split-ring resonators.

We investigated the propagation of femtosecond laser pulses through a metamaterial that has a negative index of refraction for wavelengths around 1.5 micrometers. From the interference fringes of a Michelson interferometer with and without the sample, we directly inferred the phase time delay. From the pulse-envelope shift, we determined the group time delay. In a spectral region, phase and group velocity are negative simultaneously. This means that both the carrier wave and the pulse envelope peak of the output pulse appear at the rear side of the sample before their input pulse counterparts have entered the front side of the sample.

We further miniaturize a recently established silver-based negative-index metamaterial design. By comparing transmittance, reflectance, and phase-sensitive time-of-flight experiments with theory, we infer a real part of the refractive index of −0.6 at a 780 nm wavelength-which is visible in the laboratory. © 2006 Optical Society of America OCIS codes: 160.4760, 260.5740. Photonic metamaterials are tailored artificial optical materials composed of subwavelength metallic building blocks that can be viewed as nanoscale electronic circuits. These building blocks or "photonic atoms" are densely packed into an effective material such that the operation wavelength is ideally much larger than the lattice constant a. Along these lines, highly unusual material properties become accessible, e.g., a negative index of refraction, 1,2 which has recently reached operation wavelengths of 2, 3 1.5, 4 1.5, 5 and 1.4 m. 6 In this Letter we demonstrate a negative index of refraction at the red end of the visible spectrum (780 nm wavelength) for what is the first time to our knowledge.The physics of the particular sample or circuit design 7 used and miniaturized here has been described previously in work at lower frequencies. 3,5,6 In brief, for the polarization configuration shown in Fig. 1(a), the metamaterial can be viewed as composed of two sets of subcircuits or "atoms": (i) a coil with inductance L in series with two capacitors with net capacitance C as an LC circuit, providing a magnetic resonance at the LC resonance frequency, 8 and (ii) long metallic wires, acting like a diluted metal below the effective plasma frequency of the arrangement. 9 The negative magnetic permeability from (i) and the negative electric permittivity from (ii) lead to a negative index of refraction.1,2 We use silver as a constituent material because silver is known to introduce significantly lower losses 10 than, e.g., gold at visible frequencies. The choice of the dielectric spacer material is uncritical; we use MgF 2 . In numerical calculations, the design parameters have carefully been optimized for optical performance. Results from the best fabricated sample are shown here.Fabrication employs standard electron-beam lithography, electron-beam evaporation of the constituent materials, and a liftoff procedure. All samples are located on glass substrate, coated with a 5 nm thin film of indium tin oxide (ITO) to avoid charging effects in the electron-beam-writing process (the ITO layer is irrelevant for the optical performance). The electron micrograph of the best sample (100 m ϫ 100 m footprint) shown in Fig. 1(c) reveals good large-scale homogeneity as well as a 68 nm minimum lateral feature size at 97 nm thickness of the Ag-MgF 2 -Ag sandwich. This aspect ratio (i.e., height/width), exceeding unity, poses significant fabrication challenges. Compared with our previous choice of parameters at lower frequencies, 5,6 especially the relative thickness of the metal wires oriented along the electric field vector [i.e., the ratio w y / a y in Fig. 1(b)...

We fabricate and characterize a low-loss silver-based negative-index metamaterial based on the design of a recent theoretical proposal. Comparing the measured transmittance and reflectance spectra with theory reveals good agreement. We retrieve a real part of the refractive index of Re͑n͒ =−2 around 1.5 m wavelength. The maximum of the ratio of the real to the imaginary part of the refractive index is about three at a spectral position where Re͑n͒ =−1. To the best of our knowledge, this is the best figure of merit reported for any negative-index photonic metamaterial to date. © 2006 Optical Society of America OCIS codes: 160.4760, 260.5740. In 2005, the first metamaterials exhibiting a negative index of refraction at optical frequencies were reported. 1,2 They all follow the general idea of fabricating an artificial effective material composed of "magnetic atoms," providing a negative magnetic permeability , and "electric atoms," providing a negative electric permittivity ⑀. The resulting index of refraction n is generally complex. Clearly both future physics experiments as well as potential applications (e.g., Pendry's perfect lens 3 ) require that the modulus of the real part Re͑n͒ Ͻ 0 is much larger than the imaginary part Im͑n͒ Ͼ 0. In other words, the figure of merit (FOM),should be as large as possible. For the double-wire design, 2 the maximum value was FOMϷ 0.1 at Re͑n͒Ϸ−0.2 around 1.5 m wavelength.2 A precursor of the structure to be discussed in this Letter gave FOMϷ 0.5 at Re͑n͒Ϸ−1 around 1.9 m wavelength. 1Thus the most prominent next frontier of negativeindex metamaterials is to reduce their losses.In a corresponding recent theoretical study, 4 Zhang et al. proposed a novel negative-index metamaterial structure. Depending on the metal damping assumed, values as large as FOM= 6 have been reported. 4 Their design is schematically shown in Fig. 1(a), together with our sample parameters. For the polarization configuration depicted, the structure can be thought of as consisting of double-plate (or double-wire) pairs 5,6 as "magnetic atoms" and long wires as "electric atoms" (just a diluted Drude metal). The key to optimizing the FOM of this structure 4 lies in tuning the combination of wire widths, metal thickness, and spacer thickness. For example, the width of the thin double wires has to be sufficiently large to bring the corresponding effective plasma frequency above the frequency where a negative index is expected [otherwise Re͑⑀͒ Ͼ 0]. This aspect obviously introduces a dependence of the optimization on the metal plasma frequency. On the other hand, this width should be as small as possible to not disturb the performance of the "magnetic atoms." One major source of line broadening stems from the damping of the constituent metal. Thus low-loss constituent metals are of obvious interest. It is well known that silver has the lowest damping of all metals at optical frequencies. According to measurements