Integrating sphere port error in diffuse reflectance measurements

The impact of a finite thickness integrating sphere port on the measurement of diffuse reflectance is addressed in a combined numerical and experimental study. It is shown that for a finite thickness port, additional light losses occur due to scattering between the sphere port wall and the test sample, causing the sample reflectance to be underestimated. Monte Carlo ray tracing is applied to obtain quantitative estimates of the resulting measurement error for the case of a diffusely reflecting sample. The effects of sample reflectance, port geometry, and illumination beam size on the measurement error are explored. Experimental data collected with a pair of integrating sphere reflectometers with different port geometries support the validity of the numerical results. It is argued that finite port thickness may be an important source of measurement error, even for a well-designed integrating sphere port, and a strategy for minimizing this error is discussed.