The transmission of information from one place to another by light waves sent through waveguides is hampered by light attenuation and scattering loss. Magnetic photonic crystals could provide a solution to such problems.
The concept of photonic crystals — periodically arranged structures specifically engineered to trap and guide light — grew from an initial analogy1,2,3 with the electronic band structure of semiconductors. In these materials, no electrons can be found that have energies within a range known as the 'band gap'. Similarly, in photonic crystals, photons whose frequencies fall within the 'photonic band gap' are prevented from flowing inside the material.
Haldane and Raghu4 have recently extended the equivalence between the behaviour of photons in photonic crystals and that of electrons in electronic systems. They have predicted the photonic analogue of the 'edge' states that characterize the quantum Hall effect5 that is experienced by the electrons of a two-dimensional (2D) electron gas when it is subjected to a strong magnetic field. Under certain conditions, photons can be confined to the edges of a 2D photonic crystal — one whose lattice structure has 2D periodicity — and be restricted to unidirectional propagation. On page 772 of this issue, Wang et al.6 report observing such photonic edge states in a magneto-optical 2D photonic crystal, verifying Haldane and Raghu's theoretical prediction4.
To achieve unidirectional photonic edge states requires a system that lacks time-reversal symmetry — that is, one with physical properties that are not preserved by a time-reversal transformation. To realize such a 'non-reciprocal' system, Wang and colleagues6 used a photonic crystal consisting of a 2D-periodic arrangement of magneto-optical ferrite rods; the magneto-optical nature of the rods confers the desired time-reversal asymmetry on the system. After characterizing the system's band gap, the authors demonstrated the unidirectional character of the system's edge states: forward-propagating transmission outweighed backward-propagating transmission by almost 50 decibels.
Wang and colleagues' experimental demonstration6 of the correspondence between the optics of a photonic crystal and the elegant physics of the quantum Hall effect is not only a delight for fundamental science, it also opens the door to practical applications based on non-reciprocal photonic crystals. These crystals may provide the means to develop a new type of optical-fibre waveguide that would be utterly immune to energy loss caused by scattering from material defects or obstacles.
Photonic-crystal fibres7, a form of optical fibre based on 2D photonic crystals, have been very successful in providing unique functions8 in fibre-optic communications. The most interesting type of photonic-crystal fibre has a hollow core in which light is confined by a surrounding cladding that consists of either a 2D-periodic photonic crystal or concentric 'Bragg rings'9. Because their cores are hollow rather than being filled with a material substance, light channelling through them suffers less absorption loss, enabling low-loss propagation over long distances. Indeed, it has been shown10 that photonic-crystal fibres can achieve very low loss. But they are not quite as lossless as one would hope owing to scattering caused by the intrinsic roughness of their internal (glass) cladding surfaces10: the lowest reported loss is still about a factor of ten larger than that of their best conventional counterparts. In transoceanic optical-fibre systems, underwater amplifiers must be placed approximately every 100 kilometres to compensate for loss.
One of the distinctive properties of 3D-periodic photonic crystals is that light is confined in all directions. As a consequence, and unlike in ordinary fibres or 2D-periodic photonic-crystal fibres, light travelling through hollow waveguides carved out of 3D-periodic photonic crystals is not subject to scattering loss. This increases the possibility of attaining ultra-low-loss light propagation, with both absorption and scattering losses suppressed. However, it does not prevent back scattering. Light can propagate both forwards and backwards within the same hollow waveguide, and back scattering off an obstacle within the waveguide can reduce forward transmission, and so be a source of loss even in a 3D-periodic photonic crystal.
Haldane and Raghu's theoretical ideas4, together with the experiments of Wang et al.6, now offer a solution to the back-scattering problem. By using a magneto-optical, photonic-crystal system that breaks time-reversal symmetry, Wang and colleagues show that it is possible to design the material's dispersion relationship, which describes the way in which wave propagation varies with frequency, such that, for a given frequency band, only forward-propagating waves exist. The ferrites the authors6 used operate at microwave, rather than optical, frequencies. Nonetheless, there are several other magneto-optical materials that are used in the optical regime.
To sum up, the ideal optical waveguide would be made of a low-loss hollow core, with a layer of non-reciprocal material, surrounded by a 3D-periodic photonic crystal, providing immunity to both back scattering and surface-roughness scattering. With such a low-loss waveguide, the possibility would then exist for one-hop transoceanic communication across 10,000 kilometres — about the distance from San Francisco to Tokyo — without the current requirement for electronic repeaters or amplifiers.
Yablonovitch, E. J. Opt. Soc. Am. B 10, 283–295 (1993).
John, S. Phys. Rev. Lett. 58, 2486–2489 (1987).
Yablonovitch, E. Phys. Rev. Lett. 58, 2059–2062 (1987).
Haldane, F. D. M. & Raghu, S. Phys. Rev. Lett. 100, 013904 (2008).
von Klitzing, K., Dorda, G. & Pepper, M. Phys. Rev. Lett. 45, 494–497 (1980).
Wang, Z., Chong, Y., Joannopoulos, J. D. & Soljačić, M. Nature 461, 772–775 (2009).
Russell, P. St J. J. Lightwave Tech. 24, 4729–4749 (2006).
Yeh, P., Yariv, A. & Marom, E. J. Opt. Soc. Am. 68, 1196–1201 (1978).
Roberts, P. et al. Opt. Express 13, 236–244 (2005).
About this article
Scientific Reports (2017)
Fibers and Polymers (2014)