Surface Roughness Spectrum Research Articles

Directional surface plasmon scattering from silver films

We discuss the bright hollow cone of scattered light from a silver film due to surface plasmon excitation. The power per solid angle scattered into the cone on the prism side of the metal film is an order of magnitude greater than that scattered on the air side. By photographing the cone, we obtain the spatial surface roughness spectrum of the metal film. Results of these experiments are compared to linear scattering theory.

Scattering in ZnO-Sapphire Optical Waveguides

Optical waveguide loss due to scattering in epitaxial ZnO-sapphire films was studied by measuring the dependence of attenuation on mode number and on optical wavelength. Analysis of the results in terms of wave-optical scattering theory enabled the bulk and surface scattering contributions to be determined. The bulk scattering showed the lambda(-4) dependence characteristic of Rayleigh scattering. A white noise spectrum of surface roughness was found to be in good agreement with surface scattering observed in most of the waveguides studied. The attenuation coefficients derived from the data were related to specific aspects of the waveguide fabrication process and allowed rational improvements resulting in a decrease in loss from 20 dB/cm to 1 dB/cm.

On the quantitative measurement of the roughness spectrum of silver films

We report measurements of the scattered light intensity, in order to evaluate the absolute amplitude of the spatial surface roughness spectrum. Detailed investigations were performed, to test whether the first order scattering theory was applicable. This was found to be the case for a silver film surface, whose rms roughness was 3–4 Å in the range of spatial frequencies 0.5–1.5 × 10 5 cm −1. The roughness spectrum was evaluated consistently by 3 different methods. For a silver surface, which was about 10 times rougher, strong higher order scattering effects were observed, e.g. a decrease of the phase velocity of the surface plasmon. Contributions to the scattered light from interior roughness were negligible.

Removal of aircraft altitude variation from laser profiles of the arctic ice pack

Standard high-pass-filtering procedures are not in general adequate for removal of aircraft altitude variation from laser profiles of the arctic pack ice because of the spectral overlap between the surface roughness spectrum and the aircraft height variation spectrum. Owing to this overlap, a high-pass filter tends to depress high ridges and thus makes the resulting profile unsuitable for ridge height analysis. To bypass this difficulty, a straight-forward three-step process is presented. The technique is first to carry out a conventional high-pass-filtering operation and then to estimate minimum points, which can then be used to estimate an ice roughness base line. The estimated ice roughness base line is then low-pass-filtered. This process can be used routinely for processing various profiles, since the filter cutoffs are not critical. The filtering operations are performed by small-error low-pass filters with guaranteed maximum errors of <0.9% outside the transition band.