This thesis comprises a series of studies and investigations into applications of Nuclear Magnetic Resonance (NMR) to automated and quantitative process analysis.The work will cover the use of different nuclei, permanent and superconducting magnets, and quantitation by univariate and multivariate approaches. Customised programmes and scripts accompany the experimental work on each of these aspects.Chapter 1 is an introduction to the history and status of NMR, as relevant to these works.Chapter 2 describes the development and validation of a fully automated 35Cl NMR quantitation method for chloride in solution. The method is embedded in existing walk-up software on a 400 MHz spectrometer and yields quantitative results with linearity R2 = 0.999 and a detection limit of 0.1% w/w for a 10 mg sample.Chapter 3 covers investigations into the capabilities and limitations of a 43 MHzbench top NMR spectrometer for samples in a tube and in flow. Aspects including long-term stability, characterisation of samples in flow, non-linearity of receiver andamplifier and synthesis of accurate shaped pulses are assessed.Chapter 4 builds on this information and describes the development and programming of an automated custom interface designed for non-expert users that allows the bench top spectrometer to be used to rapidly supply quantitative results for atline process samples. Relative quantitation is achieved through a combination of sum integration and deconvolution. Chapter 5 then extends the capabilities of the benchtop spectrometer through the development of a quantitative method for reaction mixtures of a pharmaceutical process using partial least squares modelling of 1H spectra. The model is developed such that results were shown to be independent of typical spectral variations expected in real samples.The thesis concludes with remarks on potential future extensions to the work discussed here.
|Date of Award||19 Sep 2019|
- University Of Strathclyde
|Sponsors||University of Strathclyde|
|Supervisor||John Parkinson (Supervisor) & Alison Nordon (Supervisor)|