Thesis submitted in fulfilment of the requirements for the degree Magister Technologiae: Chemistry in the Faculty of Applied Sciences at the CAPE PENINSULA UNIVERSITY OF TECHNOLOGY 2014 / The solid state modification of a given active pharmaceutical ingredient is a desired way to alter its physicochemical properties, such as solubility or bioavailability. The solubilitymelting point relationship of the ensuing co-crystal or salt is not fully understood. In this thesis, a series of model co-crystals and pharmaceutical co-crystals and salts of baclofen were investigated. The model co-crystals were prepared from 4,4’-bipyridine (BIPY) and 1,2-bis(4-pyridyl)ethane (ETBIPY) used as host compounds which were combined with a series of carboxylic acids as co-formers, such as p-toluic acid (PTA), rac-phenylbutyric acid (racPBA), racemic and S-phenylsuccinic acid (racPSA and S-PSA, respectively). In the second part, six new multicomponent crystals of baclofen (BAC, (RS) 4-amino-3-(4- chlorophenyl)-butanoic acid), were prepared with mono- and dicarboxylic acids: two pharmaceutical co-crystals obtained with benzoic acid (BAC•BA) and p-toluic acid (BAC•PTA) and four pharmaceutical salts with 1-hydroxy-2-naphthoic acid, (BAC+)(HNA-), oxalic acid, 2(BAC+)(OA2-), maleic acid, (BAC+)(MA-) and p-toluene sulfonic acid, (BAC+)(PTSA-)•IPA. The compounds prepared were analysed by single crystal and powder X-ray diffractometry, differential scanning calorimetry and their solubility was measured in water and ethanol. From the analysis of the model co-crystals it was concluded that their aqueous solubility is inversely related to the melting point values and this can be explained by packing features. Also, the introduction of a chiral building block, compared to its racemic counterpart, is a valuable way to limit the formation of the intermolecular interactions in the new multicomponent crystal and thus decrease the efficiency of the packing which eventually leads to lower melting points and better solubility. The analysis of the baclofen crystals suggests that a strong, robust and predictable hydrogen bonding network with a combination of molecular building blocks which show acceptable molecular flexibility is a good recipe for successful co-crystal design.