Fragmentation of molecules by collision-induced dissociation (CID) is not well-understood, making interpretation of MS/MS spectra difficult and limiting the effectiveness of software tools intended to aid mass spectral interpretation. An approach is required which is tailored to each individual molecule and improves the ‘chemical sense’ of the software. It was hypothesised that the bonds which break during CID of protonated molecules are the bonds which are elongated, and hence weakened, as a result of conformational changes induced by protonation. Bond length changes for a test set of molecules were calculated using quantum chemistry software. Density Functional Theory (DFT) or Austin Model 1 (AM1) or both were used to energy minimise the structures of the neutral molecules and their corresponding protonated molecules (protonated at all possible sites). Bonds which elongated to the greatest extent after protonation were compared to the bonds which were found to cleave to give the product ions in the CID spectra of these compounds. Quantum chemistry modelling was also applied to the deprotonated molecules. AM1 calculated bond lengths were found to be similar to those generated by DFT and have the advantage of being rapidly obtained. All the polarised bonds which cleaved were calculated to elongate significantly, thus achieving a 100% success rate in the prediction of bond cleavage as a result of protonation on a heteroatom. The proton is mobile across the molecule, leading to fragmentation when the proton reaches a site where it causes significant bond elongation, provided the molecule has sufficient internal energy. Cleavage of carbon-carbon bonds was not predicted. The success rate for predicting bond cleavage in deprotonated molecules was 48%, suggesting this approach cannot be applied reliably for these anions. AM1 calculated bond length change acts as a descriptor for predicting polarised bond cleavage in protonated pseudo-molecular ions having the potential to be incorporated in mass spectral interpretive software to increase the accuracy of prediction of CID spectra.
| Identifer | oai:union.ndltd.org:bl.uk/oai:ethos.bl.uk:732821 |
| Date | January 2015 |
| Creators | Wright, Patricia Ann |
| Contributors | Pullen, Francis |
| Publisher | University of Greenwich |
| Source Sets | Ethos UK |
| Detected Language | English |
| Type | Electronic Thesis or Dissertation |
| Source | http://gala.gre.ac.uk/18134/ |
Page generated in 0.002 seconds