We experimentally demonstrate for the first time on-chip negative diffraction of SPPs in an array of gap-plasmonic waveguides for the spectral range λ0 = 1200-1800 nm, which is caused by negative mutual coupling of the waveguides . We also observe negative refraction  on the array’s interface with an adjacent metal film and subsequent refocusing of SPPs on this film. These findings are supported by band structure calculations proving negative refraction over the broadest spectral range reported up to date . Since the propositions of negative refraction, first by V. Veselago , later by J. Pendry, different concepts have been developed to increase the range of incidence angles and the spectral bandwidth. In contrast to our approach, most designs have a narrow bandwidth as they are based on resonant excitations in metamaterials .
To achieve our goal we transferred the concept of discrete diffraction from arrays of dielectric waveguides  to arrays (nanoscale confinement, ca. 300 nm ≈ λ0 / 5) of coupled plasmonic gap waveguides  (pitch Γ ≈ 370 nm) 25 x 25 µm in size. To inject light into the array we excite a single waveguide of the array via a connected Yagi-Uda nanoantenna (15% efficiency)  and monitor the discrete diffraction inside the array .
Measuring the intensity distribution at the array end we determined the coupling constant c = Wa / (2 L Γ) ≈ 0.34 μm-1 for λ0 = 1550 nm, resulting in an extremely large anisotropy of the metasurface’s effective refractive index Δneff = 2 λ0 c / π ≈ 3.14. The spectral dependence of the wave spreading turns out to be opposite to the known dispersion in dielectric structures. Diffractive wave spreading is reduced for larger wavelengths, consistently indicating a negative coupling process. Light leaving the array is converted to SPPs while experiencing negative refraction at the array interface. SPPs propagating further on the adjacent metal film refocus spontaneously thus imaging the incoupling spot.
In conclusion this work introduces a new and extremely broadband concept for negative refraction in a chip-based hyperbolic metasurface. It offers the opportunity to build and combine arbitrarily shaped hyperlenses made from positively and negatively diffracting metamaterials.
1. V. G. Veselago, Sov. Phys. Uspekhi 10, 509-514 (1968).
2. A. Poddubny, I. Iorsh, P. Belov, Y. Kivshar. Nat. Photonics 7, 948-957 (2013).
3. U. Peschel, T. Pertsch, and F. Lederer, Opt. Lett. 23, 1701-1703 (1998).
4. A. Kriesch, S. P. Burgos, D. Ploss, H. Pfeifer, H. A. Atwater, U. Peschel. Nano Lett. 13, 4539-4545 (2013).
Friedrich-Alexander-University Erlangen-Nuremberg, California Institute of Technology
You must be logged in and own this session in order to