Comprehensive NMR Spectral Analysis and Prediction of Selected Coumarin Derivatives Using 1H, 13C, 2D-COSY, HSQC, and HMBC Techniques

Abstract

Nuclear Magnetic Resonance (NMR) spectroscopy is a powerful analytical technique extensively used in structural elucidation of complex organic compounds. Recent advancements in computational tools have enabled accurate spectral prediction and structure generation from multidimensional NMR datasets. This study aimed to evaluate the effectiveness of the Spec2D expert system in predicting and interpreting 1H, 13C, and 2D NMR spectra (COSY, HSQC) for the structural elucidation of selected coumarin derivatives. A total of nine coumarin derivatives were subjected to in silico NMR spectral prediction using ChemOffice Ultra and the online platform nmrdb.org. Detailed structure elucidation was performed using the Spec2D system, which applies graph-based modelling and a COSY-based knowledge base for substructure extraction and candidate structure generation. Input spectra included 1D (1H, 13C) and 2D (COSY, HSQC) data, with cross-peak information derived from H–H COSY spectra. The analysis module successfully extracted relevant substructures and proposed candidate structures based on spectral constraints. For all compounds, the correct structures were consistently ranked highest by the system, with isomeric structures appearing in lower ranks. Spec2D demonstrated flexibility in handling compounds with different molecular formulas and exchangeable protons. Processing time per compound ranged from 30 seconds to 2 minutes. Spec2D proved to be a reliable and efficient tool for NMR-based structural elucidation of coumarin derivatives. Its ability to generate accurate predictions without molecular formula input, accommodate spectral variability, and utilize a COSY-driven knowledge base underscores its utility in modern cheminformatics and metabolomics research.