| Summary: | This project aimed to develop a high throughput, laser based, matrix-free, mass spectrometric technique using nanoporous and nanostructured semiconductor substrates for rapid, sensitive, high resolution and accurate mass analytical approach for complex biological matrices. Laser desorption/ionisation (LDI) based on nanostructured semiconductor surfaces is a novel matrix-free mass spectrometry approach. This novel LDI strategy is closely related to matrix-assisted laser desorption/ionisation (MALDI). However, the functions of the matrix are substituted by an active substrate and the mass spectrum does not suffer matrix interference at the low mass region (m/z below 700). This project aimed to develop this method for pharmaceutical and metabolomic applications, specifically metabolite profiling of complex biological matrices. It was the first time three rival technologies, DIOS, QuickMass and SALDI substrates were evaluated and compared under similar experimental conditions. The study included a comprehensive investigation of the physicochemical properties of these matrix-free LDI substrates, independently from the manufacturers or research group. It also included a comprehensive and detailed mechanistic study and demonstrated the suitability of this novel LDI approach in analysing complex biological mixtures consisting of a hundred or more small biomolecules. It is believed that the physicochemical properties of the substrate have a strong influence on the LDI efficiency. The nature of the substrates was determined by surface analysis and imaging techniques including secondary ion mass spectrometry (SIMS), X-ray photoelectron spectroscopy (XPS), atomic force microscopy (AFM) and scanning electron microscopy (SEM). Surface properties that govern the LDI processes were identified. It was found that though pores are not strictly required for ion generation, nano-sized porous structure is an important determinant affecting not only the ionisation efficiency, but also the detection mass range and the longevity of the signal. Although a roughened surface is required for the ion generation, the LDI performance does not depend strongly on the surface roughness, but perhaps more on the thickness, dimension, and density of the surface nanostructures. Micron-sized surface structures do not promote ionisation effectively. Both DIOS and SALDI were found to be silicon based, but the DIOS substrates had been fluoro-silanised. In contrast, the QuickMass substrates were found to be germanium based. The SALDI substrates were found to be oxides passivated. Investigation into surface cleaning technology and chemical modification was carried out on the SALDI substrates. Argon plasma etching followed by fluoro-silane modification was found to be suitable and enhanced the SALDI activity. It was found that fluorine and hydroxyl surface terminations are advantageous. It was proposed that the surface Si-OH moieties are an important proton source and fluorine increases the acidity of the Si-OH moieties. A wide range of biological and pharmaceutical compounds was analysed to determine the compound classes that were amenable to the method. Compounds analysed include amines, amides, amino acids, peptides, saccharides, steroids, lipids and small organic acids. The chemical properties of the compounds were correlated to the mass spectra generated. Laser induced surface reactions were also investigated by SIMS and XPS chemical imaging. It was found that the ionisation process is not a simple acid-base reaction, but a complicated simultaneous multi-reaction similar to that of MALDI. Laser induces further surface oxidation and produces a reduction potential. It was proposed that the energy transfer mechanism is closely linked to the excitation and relaxation dynamics of the exciton and the special surface state of surfaces' nanocrystallites. It was also proposed that the entropy of the reaction ultimately determines the ions observed and the rate of reaction determines the selectivity. This proposition departed from the conventional view of aqueous basicities and proton affinities dependence. The analytical characteristics of the DIOS target and MALDI Q-ToF mass spectrometer were investigated. A range of complex biological matrices was analysed, including blood plasma extract, liver extracts, urine extracts, bacterial cells and culture. Extraction methods and the application of principle component analysis (PCA) in the interpretation of the mass spectral data were explored. Suitable extraction methods were found to be important but generally, simplified approaches were sufficient. Even though the RSD value of the ion peaks intensity varied by 10-50%, the application of PCA to the DIOS spectral dataset was still possible.
|
|---|
