Research For Fluorescence Spectrum Measurement Of Oil Contaminants Based On Microchannel System

The world petroleum production is increasing rapidly with the ever expanding industry scale. Oil pollutants produced by the factories and offshore oil spills from mining and shipping are serious threats to human health and ecological environment safety. The ability to accurately and efficiently detect the contents and constituents of oil pollutants is critical for the health and sustainable development of our ecological system. At present,the main methods for oil pollutants detection are weighing method, chromatography, near infrared spectroscopy and fluorescence spectroscopy. Among them, the fluorescence spectrum detecting technology, which has the advantage of high precision and good selectivity, has been widely used in the fields of soil and atmosphere component detection.Focused on the characteristics and hazards of oil pollutants, combined with the development trend of fluorescence spectroscopy both at home and abroad, based on the structure of microchannel sample cell, using the parallel excitation of multiple microchannel sample cells, we developed multichannel fluorescence spectroscopy on sample cells and its optical system. We also adopted the improved parallel factor algorithm to detect and analyze the contents of petroleum pollutants. The main research work includes:We analyzed the fluorescence detection model of petroleum pollutant based on the theory of photoluminescence and the mechanism of fluorescence produced from the fluorescence substance. We also studied the relationship among fluorescence intensity and quantum efficiency and fluorescence photon quantum number of the photoelectric detector as well as how environmental parameters such as temperature and solvent influence the fluorescence intensity. Based on the above results, we propose the procedure of petroleum pollutants detection using three-dimensional fluorescence spectroscopy combined with the improved parallel factor algorithm.We were able to amplify the fluorescence signal without changing the volume of sample cell by increasing the optical path of the detection system using the parallel excitation of multiple microchannel sample cells. We analyzed and designed the opticalsplit system of the fluorescence spectroscopy in the scanning range of 250-520 nm,studied the improved grating feedback optical split system, compared wavelength resolution with the results using fluorescence spectrophotometer. The results showed that the wavelength resolution was improved to ±0.3 nm after adding the grating detection feedback circuit, compared with the resolution of ±0.7 nm using the conventional system of driving grating with stepper motor and threaded axle. Under the same experimental condition, we showed that the multichannel fluorescence spectroscopy has a stronger detection signal and the general fluorescence spectroscopy. It thus provides a suitable spectroscopy solution to further incorporate the improved parallel factor algorithm in the compartment analysis of mixed petroleum solvent sample. It also takes us one step closer to the realization of the real-time detection using fluorescence spectroscopy.In order to achieve the component detection of complex petroleum compound, we studied the structure of parallel factor algorithm and its convergence induction process based on the 2nd order correction method as well as the Tucker 3 method. Based on the fact that the algorithm is sensitive to the group number of the sample, we analyzed the rule to determine the group number. We also improved the parallel factor algorithm using the decomposition of general inverse singular value. This change improved the speed and solved the local convergence problem and therefore built the theoretical foundation of component analysis of petroleum pollutant using parallel factor algorithm.In order to validate the effectiveness of our micro channel fluorescence spectroscopy combined with parallel factor algorithm, we studied its performance in analyzing petroleum pollutants from various water classes. This is achieved by using different brands and concentrations of diesel, gasoline and kerosene solvent samples. We studied the intensity of the excitation-emission of the fluorescence from all the samples and obtained their 3D fluorescence spectrum, contour plots as well as the fluorescence intensity of map of the corresponding excitation-emission wavelength. We were able to detect and measure the concentration of the different petroleum content from our solvent mixtures by applying our improved parallel factor analysis on the 3D fluorescence spectrum. The 2D excitation-emission spectrum of the components detected by the parallel factor methodmatches well with the ones measured from the actual contents, indicating the effectiveness of this new method.