Identification and characterization of homologous series by GC–MS analysis provide very relevant information on organic compounds in complex mixtures. A chemometric approach, based on the study of the autocovariance function, EACVFtot, is described as a suitable tool for extracting molecular–structural information from the GC signal, in particular for identifying the presence of homologous series and quantifying the number of their terms. A data pre-processing procedure is introduced to transform the time axis in order to display a strictly homogenous retention pattern: n-alkanes are used as external standard to stretch or shrink the original chromatogram in order to build up a linear GC retention scale. This addition can be regarded as a further step in the direction of a signal processing procedure for achieving a systematic characterization of complex mixture from experimental chromatograms. The EACVFtot was computed on the linearized chromatogram: if the sample presents terms of homologous series, the EACVFtot plot shows well-defined deterministic peaks at repeated constant interdistances. By comparison with standard references, the presence of such peaks is diagnostic for the presence of the ordered series, their position can be related to the chemical structure of the compounds, their height is the basis for estimating the number of terms in the series. The power of the procedure can be magnified by studying SIM chromatograms acquired at specific m/z values characteristic of the compounds of interest: the EACVFtot on these selective signals makes it possible to confirm the results obtained from an unknown mixture and check their reliability. The procedure was validated on standard mixtures of known composition and applied to an unknown gas oil sample. In particular, the paper focuses on the study of two specific classes of compounds: n-alkanes and oxygen-containing compounds, since their identification provides information useful for characterizing the chemical composition of many samples of different origin. The robustness of the method was tested in experimental chromatograms obtained under unfavorable conditions: chromatograms acquired in non-optimal temperature program conditions and chromatographic data affected by signal noise.
Signal Processing of GC-MS Data of Complex Environmental Samples: Characterization of Homologous Series
PIETROGRANDE, Maria Chiara;MERCURIALI, Mattia;PASTI, Luisa
2007
Abstract
Identification and characterization of homologous series by GC–MS analysis provide very relevant information on organic compounds in complex mixtures. A chemometric approach, based on the study of the autocovariance function, EACVFtot, is described as a suitable tool for extracting molecular–structural information from the GC signal, in particular for identifying the presence of homologous series and quantifying the number of their terms. A data pre-processing procedure is introduced to transform the time axis in order to display a strictly homogenous retention pattern: n-alkanes are used as external standard to stretch or shrink the original chromatogram in order to build up a linear GC retention scale. This addition can be regarded as a further step in the direction of a signal processing procedure for achieving a systematic characterization of complex mixture from experimental chromatograms. The EACVFtot was computed on the linearized chromatogram: if the sample presents terms of homologous series, the EACVFtot plot shows well-defined deterministic peaks at repeated constant interdistances. By comparison with standard references, the presence of such peaks is diagnostic for the presence of the ordered series, their position can be related to the chemical structure of the compounds, their height is the basis for estimating the number of terms in the series. The power of the procedure can be magnified by studying SIM chromatograms acquired at specific m/z values characteristic of the compounds of interest: the EACVFtot on these selective signals makes it possible to confirm the results obtained from an unknown mixture and check their reliability. The procedure was validated on standard mixtures of known composition and applied to an unknown gas oil sample. In particular, the paper focuses on the study of two specific classes of compounds: n-alkanes and oxygen-containing compounds, since their identification provides information useful for characterizing the chemical composition of many samples of different origin. The robustness of the method was tested in experimental chromatograms obtained under unfavorable conditions: chromatograms acquired in non-optimal temperature program conditions and chromatographic data affected by signal noise.I documenti in SFERA sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
