Active terahertz metamaterials

Terahertz radiation—electromagnetic waves in the 0.1–10THz frequency range—occupies a middle ground between microwaves and infrared light. The unique properties of terahertz radiation make it very attractive for numerous applications in molecular spectroscopy, biomedical imaging, short-range ultra-high-bandwidth wireless communications, non-destructive inspection, security screening, and even quantum computing in silicon devices. For example, in a similar way to how infrared radiation can probe the vibrations corresponding to the motion of atoms bound together in an individual molecule, terahertz spectroscopy can probe the much weaker forces governing the interaction between molecules. Its use could revolutionize our understanding of how complex biomolecules interact. Terahertz waves are also well suited to non-destructive testing and security screening applications because, like microwaves and millimeter waves, they are non-ionizing, making them much safer than x-rays, and they can travel through many materials including plastics and cloth. However, the shorter wavelength of terahertz waves compared to microwaves and millimeter waves enables significantly higher image resolution.

Figure 1. Microscope image of our superconducting metamaterial. YBCO: Yttrium barium copper oxide.

Figure 2. Transmission through the metamaterial showing control of the resonance by (a) temperature, (b) optical excitation, and (c) high-intensity terahertz fields.

Historically, the terahertz range has been largely unexplored due to the difficulty of generating, detecting, and manipulating terahertz radiation. Electronic devices that have driven the widespread use of microwaves are generally limited to much lower frequencies. In addition, a paucity of suitable materials means that photonic technologies that have been wildly successful in the infrared, visible, and ultraviolet regimes run into severe limitations in the terahertz range. One powerful way to circumvent the limitations of existing materials is to create artificial ‘metamaterials.’ These consist of arrays of conducting elements that are small enough to appear as an effectively continuous material to the electromagnetic waves (much as we can often ignore that natural materials are really composed of tiny atoms instead of being continuous). Metamaterials can also be engineered to exhibit exotic electromagnetic phenomena not observed in natural materials, including negative refraction and anomalous reflection. In addition, they can be used to create novel devices such as ultrathin perfect absorbers, planar lenses, and thin layers that rotate the polarization 90^°.

Metals have been used for the conductive elements in the vast majority of metamaterials. However, just as a metal wire can passively direct the flow of an electrical current but not actively control it, these metallic structures can only passively manipulate electromagnetic waves. The most frequently studied approach to enable active control is to add additional materials, such as semiconductors, to metallic metamaterials.¹ Recently, we have focused on entirely replacing the metal with superconducting oxides.^{2, 3} These have a conductivity that can be as high as that of metals, but that can also be easily controlled by temperature, light, a magnetic field, an electric field, or even very intense terahertz radiation.

Our initial work showed that the resonance of superconducting split ring resonators (SRRs), the most frequently used metamaterial building blocks, could be tuned or suppressed by varying the temperature.²We used a metamaterial consisting of an array of yttrium barium copper oxide (YBCO) SRRs on a lanthanum aluminate substrate (see Figure 1). Plotting the transmission through the metamaterial at several temperatures ranging from 20K to 100K shows that the SRR resonance causes a dip at around 0.6THz: see Figure 2(a). The change in resonance frequency and strength with temperature arises from the decrease in the density of Cooper pairs, the carriers responsible for the supercurrent. It continues up to the superconducting transition temperature (around 90K for these samples), where the material ceases to be a superconductor.

Although temperature tuning nicely demonstrates how sensitive the metamaterial resonance is to the Cooper pair density in the superconductor, thermal control tends to be too slow for many device applications. Recently, we achieved ultrafast switching of the metamaterial resonance by using near-infrared light as the control signal.² Near-infrared photons are sufficiently energetic to easily break Cooper pairs, which again leads to a change in the YBCO conductivity and therefore the metamaterial resonance: see Figure 2(b). While terahertz photons do not have sufficient energy to break a Cooper pair directly, intense electric fields of 10–10²kV/cm can drive the supercurrent into a nonlinear regime, again tuning the SRR resonance on a timescale of a few picoseconds: see Figure 2(c).

Thus far, we have shown only superconducting devices that modulate the intensity of terahertz radiation. Although this is in itself a technologically important result, we are now extending our approach of building metamaterials directly out of an inherently controllable material. By applying this to more complex metamaterial devices, we expect in due course to be able to make tunable polarization rotators, beam steering devices, or perhaps even lenses with dynamically controllable focal lengths. Superconductors also exhibit manifestly quantum behaviors in response to a magnetic field or when formed into Josephson junctions, opening exciting possibilities in the emerging field of quantum metamaterials.

We acknowledge partial support from the Los Alamos National Laboratory Laboratory Directed Research and Development (LDRD) program. This work was performed, in part, at the Center for Integrated Nanotechnologies, a user facility of the Office of Basic Energy Sciences.