Using indirect covariance processing for structure elucidation of small molecules in cases of spectral crowding

Indirect and unsymmetrical indirect covariance NMR provide powerful tools to compute and visualize correlation information by transforming component spectra into combined spectral data matrices. Sensitive component spectra such as TOCSY, HSQC and NOESY can be quickly converted into experimentally insensitive or time-consuming correlation spectra such as HSQC-NOESY. The comparison of illustrative series of spectra from four steroids, dexamethasone, testosterone, allylestrenol and tibolone, renders the effects of resonance overlap on the ease of interpretation visible. The compounds are selected such that signal overlap increases systematically in the proton and carbon domain. Spectra are defined as light, moderate and heavy signal overlap, based on signal density. The investigation suggests that moderate spectral congestion in either proton or carbon domain leads to a number of artifacts that does not hamper signal assignment but lowers the level of confidence on de novo structure elucidation. Since the number of correlations usually increases through covariance processing, component spectra with severe spectral congestion in both dimensions are not suitable for covariance processing and the resulting spectra do not support structure confirmation or structure elucidation. The calculated spectra are compared with the corresponding experimental spectra with respect to their application in structure elucidation laboratory environments.