Quantifying the validity conditions of the Beckmann-Kirchhoff scattering model

Abstract

Approximate and rigorous methods are widely used to model light scattering from a surface. The boundary element method (BEM) is a rigorous model that accounts for polarisation and multiple scattering effects. BEM is suitable to model the scattered light from surfaces with complex geometries containing overhangs and re-entrant features. The Beckmann- Kirchhoff (BK) scattering model, which is an approximate model, can be used to predict the scattering behaviour of slowlyvarying surfaces. Although the approximate BK model cannot be applied to complex surface geometries that give rise to multiple scattering effects, it has been used to model the scattered field due to its fast and simple implementation. While many of the approximate models are restricted to surface features with relatively small height variations (typically less than half the wavelength of the incident light), the BK model can predict light scattering from surfaces with large height variations, as long as the surfaces are “locally flat” with small curvatures. Thus far, attempts have been made to determine the validity conditions for the BK model. The primary validity condition is that the radius of curvature of any surface irregularity should be significantly greater than the wavelength of the light. However, to have the most accurate results for the BK model, quantifying the validity conditions is critical. This work aims to quantify the validity conditions of the BK model according to different surface specifications, e.g., slope angles and curvatures. For this purpose, the scattered fields from various sinusoidal profiles are simulated using the BEM and the BK models and their differences are compared. The result shows that the BK model fails when there are high slope angles and large curvatures, and these conditions are quantified.