NMR Prediction with Computational Chemistry

This chapter reviews the typical procedures available for predicting and assigning structures using nuclear magnetic resonance (NMR) computations. Some computational NMR approaches make predictions based on extrapolations from known experimental data, some use quantum-chemistry-based chemical shift predictions, and some use a combination of empirical and quantum-chemistry-derived chemical shifts to arrive at predictions. Common uses for computational NMR include determining the correct diastereomer of a natural or synthetic product and determining which functional groups are present in a compound. The chapter discusses the range of methods available for computing NMR chemical shifts and coupling constants, examples of their application, the advantages and disadvantages of each type of approach, common sources of error, types of systems known to be particularly challenging, and where the field of computational NMR is headed. The typical approach for calculating chemical shifts with quantum chemistry begins with building the compound(s) of interest in a molecular modeling program.