Validation of quantitative NMR

NMR is by definition a quantitative spectroscopic tool because the intensity of a resonance line is directly proportional to the number of resonant nuclei (spins). This fact enables, in principle, a precise determination of the amount of molecular structures and, hence, of substances in solids as well as liquids. With the increase of sensitivity due to stronger and stronger static magnetic fields including improved electronics the detection limits have been pushed down significantly. However, the lack of a precise protocol that considers and controls the aspects of both the measurement procedure as well as the spectra processing and evaluation is responsible for the fact that quantitative investigations of identical samples in various laboratories may differ severely (deviations up to 90% relative to gravimetric reference values). Here, a validated protocol for quantitative high resolution H-1-NMR using single pulse excitation is described that has been confirmed by national and international round robin tests. It considers all issues regarding linearity, robustness, specificity, selectivity and accuracy as well as influences of instrument specific parameters and the data processing and evaluation routines. This procedure was tested by the investigation of three different 5-model-compound mixtures. As a result of the round robin tests using the proposed protocol it was found that the maximum combined measurement uncertainty is 1.5% for a confidence interval of 95%. This applies both for the determination of molar ratios and of the amount fractions of the various components. Further, the validation was extended to purity determinations of substances as shown for 1,8-epoxy-p-menthane (cineole). (c) 2005 Elsevier B.V. All rights reserved.