Tiny metal resonators can be used to create a material with tunable responses to an applied voltage. Combined with a semiconductor substrate, they can be used to control technologically promising terahertz radiation.
Electromagnetic radiation with frequencies lying between the microwave region and the infrared — so-called terahertz radiation — holds great promise for imaging and sensing applications. It is non-ionizing, and therefore causes less damage to biological tissue than conventional, higher-energy X-rays. It penetrates plastics and clothing, but not metal, and so is ideal for security screening and non-contact testing or inspection. But to realize the full potential of terahertz radiation, more sophisticated techniques for its generation, manipulation and detection are required. In this issue, Chen et al. (page 597)1 fill an important gap in our capabilities. They report the development of an efficient, electrically driven modulator for terahertz signals that functions at room temperature.
This work builds on exciting developments in the field of metamaterials. These are materials engineered to have electromagnetic responses that are impossible in naturally occurring materials, such as a negative refractive index. The refractive index of a material, n, is a measure of the speed of light in the material, and is given by n = (µε)1/2, where µ is the material's 'permeability' to magnetic fields, and ε its 'permittivity' to electric fields. All naturally occurring materials have positive µ; transparent materials have positive ε, too. In these normal materials, therefore, the refractive index is a real and positive number.
In 1968, the Soviet physicist Victor Veselago showed2 that a hypothetical material with negative ε and µ would also have a real refractive index, meaning that light waves could propagate through the material, but would behave as though its refractive index were negative. This material would have unusual and potentially valuable properties. A flat slab of negative-index material, for example, would focus light in much the same way as a curved slab of ordinary material (a lens), but with a smaller focal spot.
Although negative-index materials do not violate any laws of physics, the absence of a medium with negative µ confined the idea to the realm of speculation. But in the late 1990s, John Pendry found that, by assembling a collection of appropriately designed metallic structures, a material can be fabricated that has both negative ε and negative µ for incident electromagnetic radiation of a particular frequency3,4. Furthermore, if the metal structures are each much smaller than the wavelength of the incident radiation, the radiation interacts with them not individually, but collectively, according to their average properties. These are the engineered materials now known as metamaterials.
Engineering a material with negative ε was easy: this equates to opacity, a property of all metals for incident radiation below a certain frequency. It was necessary to show only that a discrete set of thin metallic structures could mimic this property of the bulk metal3. The more difficult task was achieving a negative µ. It turned out that this could be done using a pair of concentric metallic rings with gaps that prevent current from circulating. Because these rings are both capacitors (they store electric charge) and inductors (they induce magnetic fields that self-sustain any current flowing through them), the presence of gaps leads to a resonant response, with charge accumulating alternately on one side of the gap and then the other, sloshing back and forth through the rings rather as a mass vibrates back and forth on a spring. At frequencies near the characteristic frequency of this resonant electron flow, ε and µ can vary dramatically as a function of frequency. Indeed, either one can become negative if the resonance is strong enough4.
The development of the split-ring-resonator concept was significant not only because it permits a negative refractive index, but more generally because it represents a new technique for 'designing' the optical response of a medium. The first experimental demonstration of a negative index5, along with nearly all research into metamaterials until now, was performed in the microwave regime. This region encompasses gigahertz frequencies below the terahertz regime, with wavelengths of several millimetres or longer. It was almost immediately recognized, however, that the approach could be extended into the shorter-wavelength, terahertz regime simply by shrinking the size of the individual metallic components so that they remained smaller than the incident radiation's wavelength. At 1 THz, this is 300 µm, so the fabrication of a negative-index medium requires the technically challenging construction of three-dimensional objects with micrometre-scale features. Two-dimensional patterns on that scale, on the other hand, can be easily generated using conventional photolithography, and these patterns can be designed to exhibit a strong resonance ε in either or µ at any frequency of interest.
Chen et al.1 use the split-ring-resonator concept as the basis for a metamaterial that provides a resonant response in ε — although not a negative refractive index — in the terahertz range. Furthermore, they have shown that this resonance can be externally controlled, and therefore can be exploited as a modulator for controlling the transmission of terahertz electromagnetic radiation.
The principle of this device's operation is simple and elegant (Fig. 1). An array of sub-wavelength metallic ring resonators is deposited on top of a thin, lightly doped semiconductor layer and illuminated with a beam of terahertz radiation. The electrons in the semiconductor substrate effectively short-circuit the gap in the split rings, driving the capacitance of the rings to zero and damping their normal resonant response. But when a negative voltage is applied to the metal structures, the electrons in the substrate beneath are repelled away from the gap. As electrons become depleted in the gap, current can no longer flow effectively and the gap behaves as a capacitor, storing electric charge. In this case, the resonance of the ring structure re-emerges, which gives rise to a pronounced change in the optical properties of the array at frequencies near its resonant frequency.
The authors' results1 are impressive. At the design frequency, the on–off transmission ratio of this device is about 0.5 — more than ten times better than state-of-the-art, electrically operated modulators in this frequency range6,7. The new design is simpler to fabricate, and the modulator works at room temperature. This excellent performance comes from the exquisite sensitivity of a metamaterial's response to the precise properties of its resonant sub-structures. The ability to electrically switch the properties of a metamaterial by fabricating it on a semiconductor substrate provides a new method for active control of terahertz devices.
Naturally, challenges remain. Most obviously, larger on–off transmission ratios will be required for many applications. The modulation speed, which is only a few kilohertz, must be increased significantly. The authors point out that both of these difficulties can be addressed by optimizing both the pattern of the metal structures and the properties of the substrate. The properties of the device are also dependent on the direction of the electric-field vector (the polarization) of the incident radiation: the resonant behaviour of the split rings relies on this electric field driving current across the gaps, all of which are oriented in one particular direction. Designs of future devices may seek to eliminate or alternatively exploit this polarization sensitivity. Finally, one can imagine a structure for which not only the transmitted intensity, but also the resonant frequency, is externally controllable.
Whatever the next stage of development might be, the resonator structures described by Chen et al.1 open a new and promising set of possibilities for the active control of terahertz radiation, with all the potential that it holds.
Chen, H.-T. et al. Nature 444, 597–600 (2006).
Veselago, V. G. Sov. Phys. Uspekhi 10, 509–514 (1968).
Pendry, J. B., Holden, A. J., Stewart, W. J. & Youngs, I. Phys. Rev. Lett. 76, 4773–4776 (1996).
Pendry, J. B., Holden, A. J., Robbins, D. J. & Stewart, W. J. IEEE Trans. Microwave Theory Tech. 47, 2075–2084 (1999).
Shelby, R. et al. Science 292, 77–79 (2001).
Kleine-Ostmann, T., Dawson, P., Pierz, K., Hein, G. & Koch, M. Appl. Phys. Lett. 84, 3555–3557 (2004).
Kersting, R., Strasser, G. & Unterrainner, K. Elec. Lett. 36, 1156–1158 (2000).
About this article
Cross-polarization coupling terahertz time-domain spectroscopy in a semiconductor based on the Hall effect
Scientific Reports (2017)
Journal of Infrared, Millimeter, and Terahertz Waves (2013)
Journal of Infrared, Millimeter, and Terahertz Waves (2010)