Conductors with infrared plasma frequencies are potentially useful hosts of surface plasmon polaritons (SPP) with sub-wavelength mode confinement for sensing applications. A challenge is to identify such a conductor that also has sharp SPP excitation resonances and the capability to be functionalized for biosensor applications. In this paper we present experimental and theoretical investigations of IR SPPs on doped silicon and their excitation resonances on doped-silicon gratings. The measured complex permittivity spectra for p-type silicon with carrier concentration 6Â10 19 and 1Â10 20 cm À3 show that these materials should support SPPs beyond 11 and 6 lm wavelengths, respectively. The permittivity spectra were used to calculate SPP mode heights above the silicon surface and SPP propagation lengths. Reasonable merit criteria applied to these quantities suggest that only the heaviest doped material has sensor potential, and then mainly within the wavelength range 6 to 10 lm. Photon-to-plasmon coupling resonances, a necessary condition for sensing, were demonstrated near 10 lm wavelength for this material. The shape and position of these resonances agree well with simple analytic calculations based on the theory of Hessel and Oliner (1965). V

Optical constants for evaporated bismuth (Bi) films were measured by ellipsometry and compared with those published for single crystal and melt-cast polycrystalline Bi in the wavelength range of 1 to 40 μm. The bulk plasma frequency ω p and high-frequency limit to the permittivity ε ∞ were determined from the long-wave portion of the permittivity spectrum, taking previously published values for the relaxation time τ and effective mass m Ã. This part of the complex permittivity spectrum was confirmed by comparing calculated and measured reflectivity spectra in the far-infrared. Properties of surface polaritons (SPs) in the long-wave infrared were calculated to evaluate the potential of Bi for applications in infrared plasmonics. Measured excitation resonances for SPs on Bi lamellar gratings agree well with calculated resonance spectra based on grating geometry and complex permittivity.

The semimetal antimony, with a plasma frequency ~80 times less than that of gold, is potentially useful as a host for infrared surface polaritons (SPs). Relevant IR SP properties, including the frequency-dependent propagation length and penetration depths for fields into the media on either side of the interface, were determined from optical constants measured on optically-thick thermally-evaporated Sb films over the wavelength range 1 to 40 μm. Plasma and carrier relaxation frequencies were determined from Drude-model fits to these data. The real part of the permittivity is negative for wavelengths beyond 11 μm. Distinct resonant decreases in specular reflected intensity were observed for Sb lamellar gratings in the wavelength range of 6 to 11 μm, where the real part of the permittivity is positive. Both resonance angles and the angular reflectance spectral line shapes are in agreement with theory for excitation of bound surface electromagnetic waves (SPs). Finite element method (FEM) electrodynamic simulations indicate the existence of SP modes under conditions matching the experiments. FEM results also show that such waves depend on having a significant imaginary part of the permittivity, as has been noted earlier for the case of surface exciton polaritons.

Output facet temperatures of an uncoated high power continuous-wave quantum cascade laser (QCL) emitting at 8.5 μm were measured by using micro-Raman thermometry. The rate of the measured temperature changes with the injected electrical power increased from 6.5 K/W below the laser threshold to 12.3 K/W above the threshold. In addition, the measured temperature rise exceeded 220 K at an optical power of 0.9 W, well above the model projections based only on Joule heating. Facet oxidation was characterized via x-ray photoelectron spectroscopy measurements at incremental etch depths. While the oxidation reactions of InP and Ga were observed only at the surface level, the measured penetration of native Al2O3 was ∼24 nm. COMSOL thermal modeling demonstrated that light reabsorption by the native Al2O3 layer could well explain the additional temperature rise above the threshold. These results suggest that facet oxidation must be addressed to ensure the reliability of high-power long wave infrared QCLs.

No abstract