No abstract

No abstract

No abstract

The mathematical self-modeling method has been used to identify the samples of urban air from their IR spectra.The spectra of methane,carbon dioxide, carbon oxide,acetylene and ethylene have been isolated from the spectra of samples. The data obtained have enabled to investigate the ethylene content in the ambient atmosphere as a function of daytime using laser optoacoustic analysis.The accumulation of harmful pollutants in the urban atmosphere requires the high-sensitive methods to be developed for tieir determination. The laser optoacousric spectroscopy is one of the most promising methods of the gas analysis. This method enables to detect ammonia, ethylene, freons, some alcohols and ketones at concentrations lower than 10 ppb. However the selectivity of the method is not sufficient. It is the reason that the identification of complex gas mixtures is difficult to do and sometimes impossible using this method. The selectivity of the method may be increased by the preliminary isolation of the pure components from the sample by, for example, gas chromatography. However this way is too laborios, it needs the expensive equipment to be used and the special technique to be developed for its realisation.The identification of complex gas mixtures may be carried out without their preliminary fractionation if the spectral information is studied and the computing methods are used for its processing. Recently many techniques have been developed to identify the spectra of components in the spectrum of mixture. The majority of these methods is based on the c1marison of the component spectrum with the spectrum of mixture ' . This approach has at least two serious drawbacks: the results of identification depends on both the completeness of the library used and the spectra comparison criterion. 0-8194-1

No abstract