Application of quantitative inline NMR spectroscopy for investigation of a fixed-bed chromatographic reactor process

Abstract

A Nuclear Magnetic Resonance (NMR) spectroscopy method is presented that facilitates inline analysis of dynamic processes. Compared to other commonly used optical inline analysis methods, NMR spectroscopy has the advantage that it can resolve different species in complex multicomponent mixtures. An inline NMR spectroscopy method was optimized to enable analysis with high temporal resolution. The method was applied to study the dynamic behavior of a fixed-bed chromatographic reactor (FBCR) by monitoring the composition at the reactor outlet.The heterogeneously catalyzed hydrolysis reactions of methyl acetate and methyl formate were chosen as test systems. The influence of different process parameters such as the concentration of reactants, reactor temperature and flow rate of the mobile phase (water) were systematically studied with the presented method. The concentration profiles of the different reactants and products could be determined accurately even though the two hydrolysis reactions proceeded simultaneously in the FBCR and the concentration profiles of the different species overlapped strongly. The measured concentration profiles are in good agreement with additional RI measurements which, however, do not facilitate a component specific analysis.The accurate measurement of the concentration profiles enables to study the interaction of reaction and separation in the FBCR. At low concentrations of the reactants the measured concentration profiles agree well with predictions based on a model of the FBCR developed in previous works. At higher concentrations, however, the comparison of the predictions and experimental results reveals deficits in the model. The results demonstrate that the presented inline NMR spectroscopy method is a valuable tool to gain insights into complex dynamic processes and to gather accurate experimental data that is essential for the development of reliable process models.