Few-Gradients-Based Qualitative Analysis for Diffusion-Ordered NMR Spectroscopy

DOSY (Diffusion Ordered Nuclear Magnetic Resonance Spectroscopy) is a powerful tool for mixture analysis and plays a pivotal role in biochemistry. However, the experimental time increases according to the number of encoding gradients. The existing methods are technically based on quantitative retrieval of diffusion coefficients to virtually separate components, which normally requires tens of encoding gradients for sufficiently accurate retrieval. In this work, we propose a new concept of DOSY processing named Cluster-DOSY, where molecular components are separated by qualitatively classifying decay signals based on their similarities. With the proposed qualitative analysis tool, we can separate the molecular components from DOSY data encoded by only three gradients. Experiments demonstrate that Cluster-DOSY leads to a more robust separation result than the existing quantitative methods in the case of only a few encoding gradients.