Naučno-tehničeskij Vestnik Informacionnyh Tehnologij, Mehaniki i Optiki (Jun 2024)

Assessment of the quantitative composition of hydrate formation inhibitors by their infrared spectra

  • Iuliia S. Kozhevina,
  • Tatiana N. Nosenko,
  • Mayya V. Uspenskaya

DOI
https://doi.org/10.17586/2226-1494-2024-24-3-366-374
Journal volume & issue
Vol. 24, no. 3
pp. 366 – 374

Abstract

Read online

An approach is proposed to determine the concentrations of active components in a thermodynamic type inhibitor based on its infrared spectrum. The relevance of the proposed method is due to its expressiveness and high degree of selectivity in comparison with traditional approaches for determining the quantitative composition of solutions of organic compounds. It is proposed to use the method of infrared spectrometry with Fourier transform. The method allows you to determine the concentrations of substances in solutions. However, the determination of concentrations of substances in mixtures containing more than three components is complicated, especially in solutions of organic compounds due to the presence of overlapping oscillation regions of characteristic groups. To solve the problem of determining the concentration of substances in multicomponent solutions, the regression method of projection onto latent structures was used, which belongs to chemometric analysis methods and has become widespread as a method of quantitative forecasting. The sample for the construction of calibration dependencies within the framework of the method used included the spectra of solutions obtained in the mode of disturbed total internal reflection. Models have been constructed to determine the concentrations of substances in a four-component aqueous alcohol solution. Substances to be determined: water, methanol, ethanol, ethylene glycol. The effectiveness of the constructed mo dels with a training sample, including spectra of samples with minimal preprocessing — correction of the baseline, has been confirmed. The optimal set of training samples has been determined to obtain a result with the minimum possible error, provided that the component content of the samples in the training sample does not exceed n – 1, where n is the determined amount of substances. Training the model on a sample consisting of spectra of two- and three-component solutions provides prediction of the concentration of substances with an error of up to 10 %. The proposed method will allow rapid analysis of the composition of thermodynamic inhibitors of hydrate formation. The results of the work can be used in oilfield chemistry to assess the inhibitory ability of hydrate formation inhibitors used to prevent the formation of gas hydrates during the extraction, preparation or transportation of hydrocarbons.

Keywords