COMPUTER METHODS FOR SPECTROSCOPIC IDENTIFICATION OF ORGANIC-COMPOUNDS

Various spectroscopic methods are used for the identification of organic compounds. The most commonly used methods include infrared spectroscopy, ultraviolet/visible spectroscopy, nuclear magnetic resonance, and mass spectrometry. For the interpretation of the spectroscopic results semi-empirical methods are generally used. These methods of interpretation are based on large numbers of previously recorded reference spectra. The reference data compilations are used either to find semi-empirical correlation rules (correlation charts) or for direct comparison with the spectral data of an unknown sample. In either case, large amounts of data have to be processed. Human beings are generally not very good at shuffling large amounts of data, but are capable of detecting very complex and sophisticated relationships. On the other hand, computers are extremely good at doing simple routine jobs very quickly and virtually free from error. Thus for the interpretation of spectroscopic data the best performance can be expected from a “man-machine combo”, where both partners do their special parts of the job. © 1978, Walter de Gruyter. All rights reserved.