Application of benchtop low-field NMR spectrometers in the refining industry: A multivariate calibration approach for rapid characterization of crude oils.

High-field nuclear magnetic resonance (NMR) has proven to be a valuable tool to analyze petroleum products but has never acquired widespread acceptance in routine work in oil refinery due to the high cost. The advent of compact, reliable, and affordable cryogen-free low-field benchtop NMR spectrometers suggest a possible role of such instrument for a variety of uses in refining spanning from routine lab work applications to online analysis. In this work, we report the development of a novel benchtop NMR application to estimate the API gravity, sulfur content, total acidity number (TAN), and distillation yields of crude petroleum from 60 MHz 1 H NMR spectra and partial least square (PLS) regression models. The results obtained on a dataset of over 170 crude oil samples and tested with external validation set show performances in terms of root mean square standard error of prediction (RMSEP) matching the analytical methods' ASTM precision. This novel benchtop NMR and multivariate analysis application can thus be used in typical refinery laboratories to provide within minutes and without physical distillation, reliable estimates of the most important properties usually measured on the crude petroleum during refining. The method proves to be a fast, cheaper alternative to the current analytical options providing readily available data for managing crude oil and has high potential to be further exploited for online application.