Analysis of gas chromatography—mass spectrometry data by reverse library search and the detection of substances not in the library

Computer least-squares methods are used to conduct an automatic reverse search of a library of complete mass spectra for the analysis of the output data matrix from a g.c.—m.s. system. The search is limited to those mass spectra falling within a narrow retention index window about the retention index corresponding to each scan in the datamatrix. The least-squares estimated parameters obtained are proportional to the concentrations of the substances present at each scan and are displayed graphically as the profiles of resolved overlapping g.c. peaks. The precision of the parameters is estimated. The root mean square Values of the residuals of the least-squares fits are used successfully to indicate the presence of substances not in the library. Examples are given of the results produced with data from this laboratory. Possible elaborations and improvements of this computer program are discussed.