Comparison Of Measurements By Different Instruments Of The Far-Infrared Reflectance Of Rough, Optically Black Coatings

Heeding a warning by Nicodemus (1965), we examine the problem of reflectance measurements of rough surfaces made on different reflectometers. Our analysis indicates that specular reflectance measurements in the infrared should be normalized by the projected detector solid angle before they are compared. This normalization effects a primitive deconvolution of the instrument function and produces a quantity commonly called the bidirectional reflectance distribution function (BRDF). Direct comparison of the measured specular reflectance from rough surfaces fails at far-infrared wavelengths because the diffuse component of the measurement is larger than the specular. The diffuse component is larger than would otherwise be expected because it is related to the strength of the instrument function, which increases with wavelength. These analytical findings are confirmed by comparison of specular BRDF measurements of optically black coatings made at three different laboratories. Measurements at a fourth laboratory were inconclusive. We also find that nonspecular BRDF measurements calibrated by a standard diffusely reflecting surface are identical to those calibrated by an image of the source upon the detector. The rms difference between BRDF measurements of identical or very similar samples made on three different instruments was about 13% in both the specular and large-angle nonspecular cases. However, large differences in BRDF measurements occur at angles near the specular direction because there the diffuse reflectivity of the surface is tightly convolved with the instrument function.