In this thesis, quantification procedures are designed and tested on two case studies to inves- tigate the application of benchtop NMR spectrometer on quantitative NMR spectroscopy (qNMR). Prior to the investigation on the benchtop NMR spectrometer, the fundamental theories of qNMR and analysis using internal and external standards are demonstrated on the high field spectrometer. The experiments on the acquisition parameters show well-agreed results with the fundamental theories. For the quantification, the internal standard method shows low error (0.9%), while the external standard shows slightly higher error (1.7%) but less sample preparation time.

The use of a benchtop NMR spectrometer for quantitative analysis with the external stan- dard is examined using the samples with different electrical conductivities. It is demonstrated that measurements using the pulse length based concentration determination (PULCON) method cannot achieve an average error in the quantification of < 4%, unless the standard and analyte are very similar due to the fixed tuning and matching system. Here, heuristics for manual baseline and phase correction are proposed to reduce this analyst-dependent error from 10% to 3%, which was the second primary source of uncertainty. The analy- sis also demonstrates that semi-automated quantification using quantitative global spectral deconvolution (qGSD) can achieve similar integration accuracy.

Applications of quantitative analysis using the benchtop NMR spectroscopy are carried out in the cases of the coal mining industries and forensic science. A common frothing agent, methyl isobutyl carbinol (MIBC) is quantified using a quantum mechanical model based (QM) algorithm with the solid-phase extraction. The solid-phase extraction concentrates the sample up to the detectable range of the benchtop NMR spectrometer and the QM algorithm allows effective separation of the overlaps in the spectrum. The concentration

error is within 0.5 mg/L (0.6 ppm v/v) When the inlet concentration of MIBC is between 1 and 12 mg/L (1.2-15 ppm v/v). The possibility of using a benchtop NMR spectroscopy for the quantification of illicit drugs (methamphetamine) is analysed in binary and ternary mixtures with analogues, which is a forensic science application. By using the QM algorithm, the quantification root-mean-square error (RMSE) in the binary mixtures ranges from 0.005 to 0.019 mol/L when the concentration is from 0.05 to 0.45 mol/L. The RMSE of quantification in ternary mixtures is 0.004 to 0.007 mol/L when the concentration is 0.03 mol/L to 0.42 mol/L.

Overall, the benchtop NMR spectroscopy can give an accurate quantification result using external standards. In addition, it can also determine the amount of frother present at a low concentration (minimum of 1 mg/L), with the pre-treatment before acquisition. The QM algorithm can help quantify substances even with overlaps in the spectrum. For future work, online monitoring systems and portable quantitative analysis procedure can be developed with the use of external standards.