Knebl, Andreas ; Domes, Christian ; Domes, Robert ; Wolf, Sebastian ; Popp, Juergen ; Frosch, Torsten (2021)
Response to Comment on Hydrogen and C2–C6 Alkane Sensing in Complex Fuel Gas Mixtures with Fiber-Enhanced Raman Spectroscopy.
In: Analytical Chemistry, 93 (48)
doi: 10.1021/acs.analchem.1c04606
Artikel, Bibliographie

Kurzbeschreibung (Abstract)

We recently introduced fiber-enhanced Raman gas spectroscopy (FERS) as a potential tool for comprehensive fuel gas analysis. (1) The presented results demonstrate that FERS enables the identification of methane (CH4) and C2–C6 alkanes, which are typical components of natural gas. Additionally, hydrogen (H2), which might be an important fuel gas component in the future, can be sensed simultaneously. We identified characteristic peaks for the different fuel gas components and determined the limits of detection for the different alkanes using a (diluted) gas mixture containing 1% each of different C2–C6 alkanes in nitrogen. For every major and minor Raman peak, we assigned the vibration that most likely contributes to the signal using calculations based on density functional theory (DFT). Thanks to this approach, we were able to newly assign specific vibrations to several peaks. Finally, we showed that precise and accurate point-of-use fuel gas analysis with FERS is possible. Employing a portable FERS instrument, the signal variability of CH4 measurements was determined as 0.3% and the median relative errors of quantification for CH4 and H2 were below 1.5% in a CH4–H2 mixture.

In a comment on our original article, (2) Petrov provides separate Raman spectra for each of the different C2–C6 alkanes. These pure-gas spectra as well as the accompanying tables, which contain relevant information about the signal intensities of characteristic bands of the main natural gas components together with the intensities of other components at the same spectral position, add to the existing literature on the subject. Using this information, Petrov points out that our choice for a characteristic peak of n-hexane at 1038 cm–1 is not ideal as there is marked overlap with other signals, most notably n-pentane. He states that the peaks at 815 or 732 cm–1 would be a better choice, depending on the H2 content of the sample. In the light of the presented information, we agree with Petrov that the peak at 1038 cm–1 is not the best choice to unambiguously identify n-hexane. Looking at our data (Figure 1), however, we would not instead use the peaks at 815 or 732 cm–1 but the one we already suggested for unambiguous identification in the original publication at 892 cm–1. (1) This peak is discernible in our spectra and has only minor overlap with other alkanes (2) and no overlap with H2. The limit of detection for this weak peak is between 0.2 and 0.5%

Frage zum Eintrag

Redaktionelle Details anzeigen