APPLICATION OF NEURAL NETWORKS FOR INTERPRETATION OF ION MOBILITY AND X-RAY-FLUORESCENCE SPECTRA

Neural networks (NN) have been successfully used to interpret spectral data, and to derive qualitative and quantitative information from ion mobility spectrometry (IMS) and x-ray fluorescence (XRF). It is shown that components of complex mixtures of up to six aliphatic amines may be automatically identified by NN methods from their ion mobility spectra with reasonable accuracy. The ability of NN to identify compounds even under low signal-to-noise conditions of IMS spectra is demonstrated. The use of XRF technique for quantitative determination of parts per million (ppm) amounts of mixtures of Re, Os, Ir and Pt in a polyethylene matrix, which could not be done successfully by conventional methods, was made possible by application of NN, with a root mean square error of a few ppm. The networks could be trained on a personal computer, in less than 10 min, from a surprisingly small data set of training samples to perform these tasks.