Raman Spectral Analysis Of Key Components In Gasoline

As a fast, non-destructive optical technique, Raman spectroscopy is an ideal tool for chemical analysis. And it has been successfully applied in diverse industrial fields. Gasoline is one of the most important fuels, it is required to analyze its quality fast and accurately. In this thesis, Raman spectroscopy is introduced to detect the key components in product gasoline. A novel baseline correction algorithm for Raman spectra and a new calibration model based on subtraction spectra analysis are proposed. Based on the methods, a fast gasoline analyzer is developed to meet the needs of daily analysis. The main contents of this thesis are as follows:1. Summarize the principle and features of Raman spectral analysis technology, and review baseline correction methods of Raman spectra. Systematically elaborate the typical calibration methods and their application in the petrochemical fields. Comparing with existing analysis methods, the excellent opportunity for Raman solution is found.2. Propose a new universal spectral preprocessing method to remove the baseline caused by fluorescence in gasoline samples. For a set of original gasoline spectra, it can get higher signal-to-noise and better repeatability than other baseline correction methods. Combined with PLS algorithm, the new preprocessing method is applied in quantitative analysis of methanol ratio in methanol gasoline. Experimental results proved its feasibility and effectiveness.3. Based on a low-resolution dispersive Raman spectrometer, subtraction spectra analysis method was proposed to determine the benzene concentration in gasoline. Experimental results show that the method can obtain a simpler model with higher predicted accuracy and better repeatability than traditional calibration methods, and it is useful to fast measure benzene contents in gasoline.4. In order to fast analyze many quality indices of gasoline in the same time, a gasoline Raman analyzer LRS-320 is developed, which consists of spectral acquisition system and analysis software. The above methods proposed in this thesis has been applied in the software of LRS-320.