An Empirical Study on Raman Peak Fitting and Its Application to Raman Quantitative Research

Abstract

Fitting experimentally measured Raman bands with theoretical model profiles is the basic operation for numerical determination of Raman peak parameters. In order to investigate the effects of peak modeling using various algorithms on peak fitting results, the representative Raman bands of mineral crystals, glass, fluids as well as the emission lines from a fluorescent lamp, some of which were measured under ambient light whereas others under elevated pressure and temperature conditions, were fitted using Gaussian, Lorentzian, Gaussian-Lorentzian, Voigtian, Pearson type IV, and beta profiles. From the fitting results of the Raman bands investigated in this study, the fitted peak position, intensity, area and full width at half-maximum (FWHM) values of the measured Raman bands can vary significantly depending upon which peak profile function is used in the fitting, and the most appropriate fitting profile should be selected depending upon the nature of the Raman bands. Specifically, the symmetric Raman bands of mineral crystals and non-aqueous fluids are best fit using Gaussian-Lorentzian or Voigtian profiles, whereas the asymmetric Raman bands are best fit using Pearson type IV profiles. The asymmetric O-H stretching vibrations of HO and the Raman bands of soda-lime glass are best fit using several Gaussian profiles, whereas the emission lines from a florescent light are best fit using beta profiles. Multiple peaks that are not clearly separated can be fit simultaneously, provided the residuals in the fitting of one peak will not affect the fitting of the remaining peaks to a significant degree. Once the resolution of the Raman spectrometer has been properly accounted for, our findings show that the precision in peak position and intensity can be improved significantly by fitting the measured Raman peaks with appropriate profiles. Nevertheless, significant errors in peak position and intensity were still observed in the results from fitting of weak and wide Raman bands having unnormalized intensity/FWHM ratios lower than 200 counts/cm.

Operando XAS/XRD and Raman Spectroscopic Study of Structural Changes of the Iron Molybdate Catalyst during Selective Oxidation of Methanol

“…Remaining low intensity Raman bands assigned to α-MoO 3 may be caused by a surface layer structure of MoO x . [42] The spectra were normalized to the 776 cm À 1 peak of Fe 2 (MoO 4 ) 3 so the relative change in intensity of the α-MoO 3 bands at 660 and 990 cm À 1 can be followed (for details see the Supplementary Information Section 3). EXAFS fitting results of the spectrum of the final state of the catalyst after cooling to 100°C as shown in Figure 10 ( The spectra in Figure 13 were de-convoluted into different spectral components using Voigt curves for peak fitting as recommended for inorganic, ceramic materials.…”

CO2 Raman thermometer improvement: Comparing hot band and Stokes and anti‐Stokes Raman scattering thermometers

“…The correction was performed using a spectrum data processing software (GRAMS/AI) and the files containing the relative intensity correction factor. In this study, using the software, baseline correction was performed with liner function between 1,200 and 1,450 cm −1 , and then each CO 2 band was fitted to a Gaussian and Lorentzian mixing curve . Spectra were collected in a single window between 989 and 1,511 cm −1 , which covers the main peaks of the Fermi diad of CO 2 (see Figure ) and six well‐established reference peaks of neon: 18,403.8, 18,353.6, 18,200.3, 18,071.2, 18,054.9, and 17,976.7 cm −1 .…”

“…In this study, using the software, baseline correction was performed with liner function between 1,200 and 1,450 cm −1 , and then each CO 2 band was fitted to a Gaussian and Lorentzian mixing curve. [18] Spectra were collected in a single window between 989 and 1,511 cm −1 , which covers the main peaks of the Fermi diad of CO 2 (see Figure 1) and six well-established reference peaks of neon: 18 To measure delta with high accuracy, real-time calibration was applied to reduce the deviation of Raman frequency (Figure 2). [19] .…”

Vector casting for noise reduction

“…Due to Doppler broadening, collisional broadening and optical effects spectral lines, peaks, or bands are never strictly monochromatic but feature a distribution around their centre . Spectral signal profiles can be fitted by Lorentzian, Gaussian, or Voigt profiles …”

“…[32] Spectral signal profiles can be fitted by Lorentzian, Gaussian, or Voigt profiles. [33] The simulated signal spectrum…”