Use this URL to cite or link to this record in EThOS:
Title: Salt selection for pharmaceutical use
Author: Morrison, Catriona A.
ISNI:       0000 0004 2744 1172
Awarding Body: University of Strathclyde
Current Institution: University of Strathclyde
Date of Award: 2012
Availability of Full Text:
Access from EThOS:
Access from Institution:
This work details the creation of a library of 255 single crystal structures of systematically related salt forms of phenylethylamines, collated together with phase specific physicochemical property data on melting point and aqueous solubility. This library has been used to investigate a number of structure-property relationships relevant to salt selection in the pharmaceutical industry. Sixty-four salt forms of enantiopure and racemic methylephedrine were subject to detailed structural analysis, including identification of common graph-sets and isostructural groups with respect to cation packing. This showed the reoccurring presence of a common ( ) graph-set which encompassed both OCO and OSO anion functionalities. PIXEL energy calculations were utilised to determine the strongest cation-cation and cation-anion interactions present within the salts. Close inspection of the densities showed no evidence to support Wallach's rule. The library of 255 salt structures was analysed for trends in hydrate formation. It was found that hydrate formation was most likely where there was increased presence of polar groups and especially when there was an excess of hydrogen bond acceptor atoms over donor atoms. For anhydrous salts, correlation studies showed linear trends between salt melting point and log aqueous solubility and between melting point of the salt and melting point of the parent free acid. The exact nature of the correlations varied and each was specific to certain chemical groups. Random forest analysis was used to build both regression and classification training models using 37 well characterised anhydrous methylephedrinium salts. These models successfully encompassed a wide variety of anion types. Prediction of aqueous solubility using the methylephedrinium salt regression training model was successful for a series of benzoate derived salts and for a series of hydrated methylephedrinium salts, the latter with the application of a simple correction constant.
Supervisor: Not available Sponsor: Not available
Qualification Name: Thesis (Ph.D.) Qualification Level: Doctoral
EThOS ID:  DOI: Not available