The perfection of Raman spectroscopic gas densimeters

Abstract

Raman spectroscopy can be used to determine density of gases, because the energy of fundamental vibrational modes is affected by intermolecular distances. The key problem is the estimation of exact peak positions of Raman bands, because the analyses require a precision that is mostly less than the pixel resolution of modern Raman spectrometers. A new method to determine peak positions of Raman bands and atomic emission lines in a discontinuous spectrum without numerical manipulations is tested in this study: modified scanning multichannel technique. Relocation of the gratings with a Sinus Arm Drive can be performed over a distance that is only a fraction of the pixel size that allows peak position estimations with precisions smaller than the pixel resolution and to determine the uncertainty in this estimation. This uncertainty was not determined in previous studies about gas densimeters, resulting in a large variety of inconsistent data. The new method is tested with fluid inclusions in quartz. A CO 2 density of 0.1477 ± 0.0006 gÁcm À3 and 0.8880 ± 0.0007 gÁcm À3 determined with microthermometry correspond to a Fermi dyad of 103.12 ± 0.27 cm À1 and 104.71 ± 0.26 cm À1 . A CH 4 density of 0.3461 ± 0.0002 gÁcm À3 and 0.4011 ± 0.0001 gÁcm À3 correspond to peak positions of 2910.66 ± 0.12 cm À1 and 2910.57 ± 0.12 cm À1 . The error in these numbers must be regarded as the best estimated uncertainties of peak positions, which are probably slightly adjusted to higher values due to mechanical irregularities of the Sinus Arm Drive in modern Raman systems.

DiadFit: An Open-Source Python3 Tool for Peak fitting of Raman Data from silicate melts and CO2 fluids

“…This approach has been termed the 'line segment' technique by Bakker [2021], and a typical workflow of how this method is performed in Di-adFit is summarized in Fig. 2. Step 1 identifies all the files which contain Ne lines.…”

“…7.1 A more complex Ne fitting routine Bakker [2021] criticized the 'line segment' technique, because it assumes that the correction factor at any given point between two lines is the the same. Any non-linearity in the spectrometer invalidates this assumption (Bakker [2021]).…”

“…7.1 A more complex Ne fitting routine Bakker [2021] criticized the 'line segment' technique, because it assumes that the correction factor at any given point between two lines is the the same. Any non-linearity in the spectrometer invalidates this assumption (Bakker [2021]). It has also been noted that the choice of different Ne lines bracketing the Fermi diad greatly influences the correction factor (and even the direction of drift within a given analytical session, (Wieser et al [2021]).…”

“…It has also been noted that the choice of different Ne lines bracketing the Fermi diad greatly influences the correction factor (and even the direction of drift within a given analytical session, (Wieser et al [2021]). Bakker [2021] suggest that a model should be developed for how the position of measured Ne lines deviates from the true theoretical position as a function of wavenumber, allowing a correction factor to be applied to each diad position (these corrected diad positions should then be used to calculate splitting).…”

“…However, while there have been a large number of studies optimizing calibration and analysis protocols for CO 2 rich fluids by Raman Spectroscopy (e.g., Lamadrid et al [2017], DeVitre et al [2021], Bakker [2021]), there has been comparatively less focus on the software tools and peak fitting routines used to fit Raman spectral data (e.g. Yuan and Mayanovic [2017]).…”