Plasmonic-beam device steers mid-IR wavelengths
The mid-IR spectral range (3–30μm) has long been important for sensing as well as security and defense applications. The demonstration,1 and subsequent rapid development2–4 of quantum cascade lasers (QCLs), a compact, high-power, wavelength-flexible, semiconductor mid-IR light source, has led to a surge in mid-IR research and has had a galvanizing effect on the development of technologies for sensing and countermeasure applications. While the QCL has proven to be an important and viable technology, the optoelectronic infrastructure for shorter wavelengths has been somewhat slower to develop. In light of this, our group is working with wavelength-scale metal dielectric structures in order to demonstrate active control of light-matter interactions for applications in defense and security in addition to sensing technologies.
More specifically, while the QCL provides a valuable light source for tabletop sensing experiments or laboratory demonstrations of high-power, mid-IR beams that can disable mid-IR/thermal sensors, applications in non-laboratory environments do not always provide stationary targets. The ability to steer mid-IR beams with robust optoelectronic devices requiring no moving parts could prove useful in environments where the laser and target are moving with respect to one another. We wish to use developments in the field of plasmonics to demonstrate active beam-steering devices for mid-IR frequencies.
Surface plasmons (SP) are hybrid excitations consisting of collective charge oscillations in a metal coupled to an electromagnetic wave in a dielectric propagating at the metal/dielectric interface. First studied more than four decades ago,5 plasmonics has reemerged recently as a vibrant field of research with potential applications for on-chip optical interconnects, computing, sensing, and display technologies. In visible and near-IR frequencies near the bulk plasmon resonances of typical metals for plasmonic devices, such as gold (Au) and silver (Ag), the SP mode is tightly bound to the metal-dielectric interface. Interaction of the SP wave with a molecule of interest can be enhanced by many orders of magnitude, which is the basis for surface-enhanced Raman scattering.6 In the mid-IR, where the metal more closely resembles a perfect conductor, the SP wave is much more loosely bound and similar enhancement effects cannot be achieved without more complicated geometries. Nonetheless, many of the same plasmonic phenomena first demonstrated in the visible/near-IR can be replicated, and we argue enhanced, at longer wavelengths.
One such phenomenon is steering coherent radiation through a plasmonic structure (see Figure 1), first demonstrated for light at visible frequencies.7 In such a structure, light is incident from above a subwavelength slit, with some fraction of light directly transmitted through the slit and some coupling to surface waves propagating away from it along the metal-dielectric interface. As these surface waves travel along the interface, some fraction of the wave intensity is scattered to free-space photons by an array of periodic ridges flanking the slit. While light transmitted through a subwavelength slit would typically diffract in all directions, the interaction of the directly transmitted light and the light scattered off the periodic ridges interferes in the far field to produce a high-quality transmitted radiation beam. The angle of the beam is determined by the wavelength of the incident light and the permittivity of the dielectric at the metal-dielectric interface for a given device geometry. In the mid-IR, such structures fabricated on the facets of QCLs can significantly improve the emitted beam quality.8
Mid-IR plasmonic devices offer potentially increased functionality compared with their visible/near-IR counterparts. First, the loosely bound mid-IR SPs can travel far greater distances in the mid-IR range than at shorter wavelengths, where metal losses severely diminish SP propagation lengths. Second, mid-IR plasmonic structures can be integrated with traditional semiconductors (transparent in the mid-IR). Finally, for such semiconductor-based plasmonic devices, optical properties can be controlled by altering the semiconductor's dielectric constant. In the case of our beam steering structures, a shift in the steering angle can be achieved not only by changing the incident light's wavelength, but by tuning the permittivity of the underlying semiconductor material. We have recently shown two possible mechanisms for tuning mid-IR plasmonic structures by shifting the semiconductor substrate permittivity: thermal tuning9 and carrier concentration-based tuning.10 While thermal tuning is slow and limited in tuning range, it provides a valuable proof of principle, indicating the potential for more complicated, but also faster and low-power, tuning mechanisms.
We designed the initial device using a finite element method solution of Maxwell's Equations to model the expected steering angle as a function of incident wavelength (see Figure 2). Fabrication was with standard photolithography, wet-etching, and metallization techniques on a semi-insulating gallium arsenide (GaAs) substrate. The devices were characterized using an external cavity-tunable QCL and a custom-built, automated, experimental set-up allowing detection of mid-IR light intensity as a function of transmission angle.11 The sample was put onto a temperature-controlled mount that allowed for thermal tuning of the semiconductor dielectric. Figure 3(a) shows the transmitted beaming angle as a function of the incident wavelength for laser light from 9.9μm to 10.5μm. This demonstrates a clear steering effect from incident wavelength. In addition, a substantial shift in beaming angle was seen as a function of substrate temperature. For a temperature shift from 23°C to 200°C, corresponding to a GaAs index of refraction shift of less than 1%, an angular steering shift of more than 3 degrees was demonstrated, as shown in Figure 3(b).
In summary, we have demonstrated a plasmonic-beam steering device for mid-IR wavelength light using no moving parts. With minimal shifts in the permittivity of the structure's semiconductor substrate, we are able to show an angular steering shift of 3 degrees. Though the thermal tuning mechanism used in this work is less than ideal, it does serve as a strong proof of principle for other faster, lower power dielectric tuning mechanisms such as those based on voltage-controlled carrier depletion of semiconductors. Future work will focus on developing such tuning mechanisms, as well as methods to improve overall transmission through the subwavelength slit.
The authors gratefully acknowledge support from the Air Force Research Laboratory under Grant No. FA8650-08-C-1445 and the Air Force Office of Scientific Research Young Investigators Program under Grant No. FA9550-10-1-0226.
Daniel Wasserman received his PhD from Princeton University in 2004. His post-graduate work on Quantum Cascade Lasers was performed in the Gmachl group at Princeton. In 2007, he joined the UMass Lowell physics department as an assistant professor. His group develops mid-IR optoelectronic devices for sensing and security applications.
Viktor A. Podolskiy received his MS in computer science and PhD in physics from New Mexico State University in 2001 and 2002, respectively. From 2002 to 2004, he was a research associate at Princeton. In 2004 he joined the physics department of Oregon State University (OSU) as an assistant professor, becoming an associate professor in 2009. In 2010 he joined the physics department at UMass as an associate professor. His group works on optical properties of nanostructured metamaterials and photonic crystals, negative refraction, nanophotonics, nanoplasmonics, and quantum chaos in optical systems. He received the 2008 Thomas Sugihara young faculty research award, College of Science, OSU.
