Reliable estimation of wavenumbers for peaks of benzonitrile and cyclohexane

One of the most important parameters in Raman spectroscopy is the wavenumber (Raman shift), which is calculated from the difference between the excitation and Raman scattering absolute wavenumbers. Because the observed wavenumbers are strongly dependent on the spectrometer and measurement conditions, it is essential to calibrate the system by using a well‐characterised material. We therefore reliably estimated the wavenumbers and evaluated their uncertainty for the peaks of benzonitrile and cyclohexane by using a HeNe laser and Raman‐scattered light calibrated with Ne/Ar emission lines. We considered the uncertainties originating from the fitting and repeatability of both the Ne/Ar emission lines and the Rayleigh/Raman‐scattered lights, as well as the unstabilised HeNe laser, and the pixel and spectral resolutions. We also considered the uncertainty originating from the differences in results among the grades and manufacturers (four for benzonitrile and seven for cyclohexane). The estimated wavenumbers of 12 peaks for benzonitrile and 11 peaks for cyclohexane were comparable with those described in ASTM E1840 with an expanded uncertainty from 1.3 to 1.6 cm−1 (k = 2, coverage factor). It was also confirmed that the wavenumbers at 10°C and 35°C were within the expanded uncertainty; thus, these values should be widely useful for the calibration of Raman systems. The wavenumbers and their uncertainty for the peaks of benzonitrile and cyclohexane were reliably estimated by using a HeNe laser and Raman‐scattered light calibrated with Ne/Ar emission lines. The estimated wavenumbers of them were comparable to those described in ASTM E1840 with an expanded uncertainty from 1.3 to 1.6 cm−1 (k = 2, coverage factor). It was also confirmed that the wavenumbers at 10 and 35 °C were within the expanded uncertainty.