Simultaneous measurements of oil- and water-content in crude oil-polluted sands with NMR-deconvolution analysis.

This study introduces a novel deconvolution method for accurately quantifying oil or water signals in the NMR spectra of oil-water-sand mixtures. The algorithm employs a parametric model fitting approach that incorporates prior conditions to set parameters. To validate the approach, the widely overlapping NMR spectra of the complex, synthetic oil-water-sand mixes were separated, and their water and oil contents were determined. The results demonstrated high correlation values and maximum errors of 2.11% and 3.90% for water and oil contents, respectively. The proposed deconvolution method provides an innovative solution for simultaneously measuring the content of each fluid in complex porous media containing multiple fluids by NMR spectroscopy and has the potential to facilitate research in the field.