| Summary: | In this thesis we investigate the ways in which the sensitivity, resolution and overall performance of an Earth's field NMR system can be improved without significantly compromising its simplicity, portability or affordability. We investigate the limits of the information obtainable using this device and present a range of methods for calculating and analyzing NMR spectroscopy experiments detected in the Earth's magnetic field. We demonstrate significant improvements in the performance of a commercial Earth's field NMR device, the Terranova-MRI, through several apparatus developments. First-order shimming is added to the system in order to counter any local inhomogeneity of the Earth's field. The spectral resolution of the instrument is further improved through the introduction of a field locking system to counter the natural temporal drift in the magnitude of the Earth's magnetic field. External noise interference is reduced through the use of Faraday screening, effectively increasing the signal-to-noise ratio (SNR) performance of the device. We explore three signal enhancement methodologies for optimizing the SNR performance of the system. Prepolarization, with an electromagnet as well as a permanent magnet array, is considered and compared to dynamic nuclear polarization (DNP) and hyperpolarization via optical pumping. We present a detailed theoretical discussion of DNP in low-fields and demonstrate the application of this technique for signal enhancement in EFNMR. An apparatus for performing DNP in the Earth's field is presented and optimized. A density matrix approach to simulating one- and two-dimensional Earth's field NMR experiments is presented. These numerical simulations, along with a perturbation theory approach to calculating one-dimensional EFNMR spectra of tightly coupled heteronuclear systems, are explored and compared to experimental spectra of the tetrahydroborate and ammonium ions. These systems are of particular interest for NMR detected in the Earth's field because they contain strongly ...
|
|---|