A method based on covariance and pattern recognition for improving resolutions of spatially encoded NMR spectra.

The spatially encoded technique enables the fast acquisition of two-dimensional (2D) nuclear magnetic resonance spectrum within a single scan, serving as a powerful tool for studying various systems and phenomena in short time scales. In spatially encoded spectroscopy, the resolution in the direct dimension can be enhanced by increasing effective acquisition times. However, spectral widths and resolutions in indirect dimensions are no longer independent of each other with wider spectral widths yielding lower resolution. The covariance method, which has achieved success in enhancing resolutions in the indirect dimensions of conventional 2D spectroscopy, is employed here to improve resolutions in the spatially encoded dimension. Moreover, an algorithm is developed based on pattern recognition to eliminate artifacts arising from the employment of the covariance method and experimental imperfections in recording the spatially encoded spectra. Therefore, high-resolution homonuclear 2D correlated spectra are obtained. Experiments are performed to show the feasibility and effectiveness of this proposed method in providing high-resolution spectra within greatly shortened times.