Benchtop NMR Spectroscopy in Process Engineering

Nuclear magnetic resonance (NMR) is a highly attractive powerful method that enables non-invasive analysis of complex mixtures without requiring tedious calibration procedures. Benchtop NMR spectrometers are small, robust, and inexpensive, which makes them especially suited for process and reaction monitoring. However, the low magnetic field strength and the small premagnetization volume of these spectrometers are major drawbacks, resulting in low spectral resolution, peak overlap, low signal intensities, and insufficient magnetization build-up in flowing liquids. As a consequence, the quantitative analysis of complex mixtures is not always possible by benchtop NMR spectroscopy. This thesis tackles these challenges and introduces methods that extend the applications of benchtop NMR spectroscopy in process engineering. The first method is a model-based approach for resolving overlapping peaks in 1H NMR spectra of complex multi-component mixtures. This method has been successfully tested by online monitoring of a wine fermentation. The second method uses dedicated NMR pulse sequences for 1H-13C polarization transfer and enables an improved quantitative analysis of mixtures in particular in flow experiments based on 13C NMR spectroscopy. 13C NMR spectroscopy is advantageous as the high chemical shift dispersion prevents peak overlap, albeit at the expense of low signal-to-noise ratio (SNR) and extended experimental time. The NMR pulse sequences provide an elegant solution to achieve 13C signal enhancement and to shorten the experimental time by exploiting the higher polarization and favorable spin-lattice relaxation of protons. The method was also successfully applied for measuring diffusion coefficients by NMR. The third method uses the hyperpolarization technique Overhauser Dynamic Nuclear Polarization (ODNP) to significantly enhance 1H and 13C NMR signals. This enables the detection and quantification of components at low concentrations or at high flow velocities with just a single scan. These methods can also be applied together and significantly extend the NMR toolbox for process engineering and other applications.