Premium
Rapid Access to Infrared Reference Spectra of Arbitrary Organic Compounds: Scope and Limitations of an Approach to the Simulation of Infrared Spectra by Neural Networks
Author(s) -
Selzer Paul,
Gasteiger Johann,
Thomas Henrik,
Salzer Reiner
Publication year - 2000
Publication title -
chemistry – a european journal
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 1.687
H-Index - 242
eISSN - 1521-3765
pISSN - 0947-6539
DOI - 10.1002/(sici)1521-3765(20000303)6:5<920::aid-chem920>3.0.co;2-w
Subject(s) - infrared spectroscopy , scope (computer science) , spectral line , infrared , artificial neural network , spectrum (functional analysis) , computer science , biological system , spectroscopy , identification (biology) , set (abstract data type) , chemistry , artificial intelligence , optics , physics , organic chemistry , botany , quantum mechanics , astronomy , biology , programming language
Substance identification by infrared spectroscopy is performed by comparison of the experimental spectrum with a reference spectrum from a printed compilation or a database. If the analyzed compound can not be found in a database the corresponding reference spectrum has to be simulated. In order to achieve this, several reasonable candidates of structures for the compound at hand have to be conceived and for all these, infrared spectra have to be developed. The simulated spectrum that is most similar to the experimental suggests the correct structure. A rapid spectrum prediction method based on neural networks has been developed that supplies reference spectra for any organic compound. The scope and limitations of this method will be discussed on a test set of 16 compounds representing a broad range of organic chemistry.