Abstract
The optimal choice of optical pathlength, source intensity, and detector for near-infrared transmission measurements of trace components in aqueous solutions depends on the strong absorption of water. In this study we examine under which experimental circumstances one may increase the pathlength to obtain a measurement with higher signal-to-noise ratio. The noise level of measurements at eight different pathlengths from 0.2 to 2.0 mm of pure water and of 1 g/dL aqueous glucose signals were measured using a Fourier transform near-infrared spectrometer and a variable pathlength transmission cell. The measurements demonstrate that the noise level is determined by the water transmittance. The noise levels in the spectral region from 5000 to 4000 cm−1 show that the optimal pathlength (0.4 mm) is the same for pure water and 1 g/dL aqueous glucose solutions. When detector saturation occurs it is favorable to increase the pathlength instead of attenuating the light source. The obtained results are explained by an analytical model.