The unusual properties of hyperbolic metamaterials, such as their ability to propagate light on the nanoscale without diffraction, have been realized in two-dimensional devices, heralding improved photonic circuits. See Letter p.192
Devices known as photonic integrated circuits1,2 could succeed electronic circuits in future data-storage, computation and communications technologies, because they would allow improved data bandwidths and lower energy consumption. But such devices lag behind their electronic counterparts because they are limited by diffraction effects that restrict their applications to micrometre scales, whereas electronics have already reached the nanometre scale. This shortcoming is due to the fact that the electromagnetic properties of typical optical media hinder the relay of tiny optical features. If a beam narrower than (or comparable to) the wavelength of light travels through such media, it will either be distorted when it reaches its destination, because of diffraction, or it will not get there at all, because of exponential decay. This is a fundamental limitation of propagating waves.
On page 192 of this issue, High et al.3 report the first experimental realization of two-dimensional 'hyperbolic metasurfaces' (HMSs)4,5. The authors' HMSs exhibit a range of unconventional properties, including colour-dependent negative refraction and diffraction-less propagation, coupled with low optical-transmission losses — all packed in a tiny chip.
Hyperbolic metamaterials (HMMs) are artificial structures whose optical properties are highly direction dependent. They are made of ultrathin multilayers6 or dense nanowire arrays7, and are renowned for their ability to overcome the diffraction limit by enabling the propagation of ultra-small features of electromagnetic waves4,8,9,10. Moreover, they can support greater photon energy densities than can conventional materials, thereby enhancing the interaction of light with matter11,12 — a property that can lead to improved signal modulation and decreased energy consumption. These are key ingredients for bringing HMMs to the front line of integrated circuitry, on a par with electronics. Their unusual properties could also expand their applicability beyond that of run-of-the-mill optical media.
Until recently, HMMs have been fabricated only in three-dimensional configurations, making them unsuitable for integration on flat chips. Furthermore, these composite devices often contain metallic parts that absorb light and cause losses from resistivity, weakening their electromagnetic-power throughput. Also, preventing diffraction requires a certain design that inevitably maximizes the damping of electromagnetic waves10, reducing the waves' effective propagation distances to less than 1 μm.
High and colleagues overcame these issues by fabricating an HMS consisting of a nanoscale grating on a single-crystal silver film — a design that can prevent diffraction without causing excessive losses from resistivity. Moreover, using sophisticated crystal-growth techniques and cutting-edge patterning methods, the authors were able to further minimize both resistivity and scattering losses and to achieve operational propagation distances.
What new on-chip functionalities result from this work? The hallmark property of HMMs is negative refraction, the ability to bend a beam that crosses from one medium into the HMM in the 'wrong' direction — essentially, breaking the law of refraction. Negative refraction is not typically observed in naturally occurring materials, but it has been demonstrated in various metamaterials in the past 15 years6,13,14. Not only have High et al. produced the first chip to exhibit negative refraction, but they have also shown that the effect can be wavelength dependent (Fig. 1); that is, their device allows certain colours of visible light to be refracted in the 'wrong' sense, whereas others refract normally.
This property could facilitate wavelength-based switching and routing of light in photonic circuits. No less importantly, it could be used to counter the natural tendency of a tightly focused light beam to expand as it travels, because the transition from normal to negative refraction occurs at a certain wavelength that depends on the material's design. At this wavelength, the beam impinging on the HMS does not diffract, but propagates unimpeded without sideways loss of energy, irrespective of the beam's launch angle or width (see Fig. 3a, b of High and colleagues' paper3). The devices built by the authors take advantage of this effect, so that each groove of the grating can channel this particular wavelength, regardless of how closely spaced the grooves are or how small their intrinsic width is compared with the wavelength in question. In fully fledged HMS devices, this would allow a substantial increase in the information capacity transferred across small chips. Diffraction-less 2D imaging could be one of many other potential applications.
High and colleagues further demonstrate that they can selectively route light beams of visible frequency not only by the beams' colour, but also by the photons' spin. Spin is a fundamental signature of photons, and is associated with the circular polarization of electromagnetic waves (the direction of rotation of the electric field in time and space). In one of the devices demonstrated in the current work, a beam of left-handed polarization is diverted to a direction opposite to that of a right-polarized beam. Although this phenomenon has been previously demonstrated in metasurfaces15 and HMMs16, what is unique here is the combination in prototype devices of colour sensitivity, polarization-dependent refraction, enhanced light–matter interaction and significant reduction in optical losses. The ability to encapsulate these desirable properties on a chip could form the backbone of a robust photonic system, suitable not only for high-capacity data transmission, but also for quantum-communications and quantum-memory applications.Footnote 1
Almeida, V. R., Barrios, C. A., Panepucci, R. R. & Lipson, M. Nature 431, 1081–1084 (2004).
Huang, K. C. Y. et al. Nature Photon. 8, 244–249 (2014).
High, A. A. et al. Nature 522, 192–196 (2015).
Poddubny, A., Iorsh, I., Belov, P. & Kivshar, Y. Nature Photon. 7, 948–957 (2013).
Liu, Y. & Zhang, X. Appl. Phys. Lett. 103, 141101 (2013).
Liu, Z., Lee, H., Xiong, Y., Sun, C. & Zhang, X. Science 315, 1686 (2007).
Yao, J. et al. Science 321, 930 (2008).
Jacob, Z., Alekseyev, L. V. & Narimanov, E. Opt. Express 14, 8247–8256 (2006).
Salandrino, A. & Engheta, N. Phys. Rev. B 74, 075103 (2006).
Smolyaninov, I. I., Hung, Y.-J. & Davis, C. C. Science 315, 1699–1701 (2007).
Yang, X., Yao, J., Rho, J., Yin, X. & Zhang, X. Nature Photon. 6, 450–454 (2012).
Cortes, C. L., Newman, W., Molesky, S. & Jacob, Z. J. Opt. 14, 063001 (2012).
Shelby, R. A., Smith, D. R. & Schultz, S. Science 292, 77–79 (2001).
Valentine, J. et al. Nature 455, 376–379 (2008).
Bomzon, Z., Biener, G., Kleiner, V. & Hasman, E. Opt. Lett. 27, 1141–1143 (2002).
Kapitanova, P. V. et al. Nature Commun. 5, 3226 (2014).
About this article
Physics Letters A (2016)