Problems, artifacts and solutions in the INADEQUATE NMR experiment

The INADEQUATE experiment can provide unequalled, detailed information about the carbon skeleton of an organic molecule. However, it also has the reputation of requiring unreasonable amounts of sample. Modern spectrometers and probes have mitigated this problem, and it is now possible to get good structural data on a few milligrams of a typical organic small molecule. In this paper, we analyze the experiment step by step in some detail, to show how each part of the sequence can both contribute to maximum overall sensitivity and can lead to artifacts. We illustrate these methods on three molecules: 1-octanol, the steroid 17alpha-ethynylestradiol and the isoquinoline alkaloid beta-hydrastine. In particular, we show that not only is the standard experiment powerful, but also a version tuned to small couplings can contribute vital structural information on long-range connectivities. If the delay in the spin echo is long, pairs of carbons with small couplings can create significant double-quantum coherence and show correlations in the spectrum. These are two- and three-bond correlations in a carbon chain or through a heteroatom in the molecule. All these mean that INADEQUATE can play a viable and important role in routine organic structure determination.