Extension of Complex Refractive Index Measurements to the Near-Infrared for Liquids: Methodology and Uncertainty Analysis

Optical identification of liquid droplets, aerosols, or thin films is critical for many applications. While reference spectra are sometimes available for such measurements, they are not always applicable to the observed spectrum or the given sample morphology. Reference spectra for many forms can be modeled, however, if the <i>n</i>/<i>k</i> vectors (real and imaginary refractive indices) are available. In previous work we have reported protocols to determine the <i>n</i>/<i>k</i> vectors for dozens of liquids, primarily in the mid-infrared (MIR) spectral range from 7500 to 400 cm^-1. In this work we extend the spectral range into the near-infrared (NIR) region, demonstrating a method to measure and merge the data sets to create composite <i>n</i>/<i>k</i> data ranging from 10 000 to 400 cm^-1 (1.0 to 25 µm) with absorbance fidelity spanning over four orders of magnitude, and vastly improved signal-to-noise in the NIR. The precision of the composite data is evaluated for three different liquids, focusing primarily on the steps for converting the raw absorbance spectra to <i>k</i> values. The variability in both MIR and NIR data as well as in the final <i>n/k</i> vectors is also investigated for several liquids. For typical liquids, the overall variability (reported as 2σ) in the final <i>n</i> and <i>k</i>-vectors is determined to be ∼0.4% and 3%, respectively. Finally, the derived <i>n</i>/<i>k</i> data are used to calculate absorbance spectra for aerosol droplets, showing marginal variability due to the typical measurement errors in the final <i>n</i>/<i>k</i> vectors.