Mechanistic studies on the retention of carboxylic acids in ion-exclusion chromatography

This work presents a comprehensive study on the retention mechanism of aliphatic and aromatic carboxylic acids in ion-exclusion chromatography (ICE) by considering simultaneous electrostatic repulsion effects and hydrophobic adsorption effects, as well as other factors which affect retention. These factors can be divided into three categories: analyte effects, mobile phase effects and stationary phase effects. The analyte properties including pKa values, charge, hydrophobicity and size were found to have the most significant effect in its retention. Furthermore, the retention mechanism could be controlled by varying the mobile phase conditions from 1 o-6 to 10-3 M sulfuric acid and 0 to 20% methanol, which in turn affected the degree of ionisation of the acid, as well as reducing the adsorption oflong-chain and aromatic carboxylic acids. The stationary phase substrate was also found to have a profound effect on the retention mechanism. Changing the substrate affected the hydrophobicity of the columns, thereby increasing or decreasing the contribution of adsorption to the retention mechanism. This was especially evident for aromatic acids which exhibited very long retention times when separated on polymeric columns, but were eluted significantly faster on silica based columns. In addition, the effects of varying the ion-exchange capacity (from 0.5 to 5 mequiv/g) and the degree of cross-linking of the polystyrene-divinylbenzene substrate (from 4 to 12%) were investigated and found to have a significant effect on the retention of some acids. A mathematical retention model which describes the relationship between the retention factor of the analyte and the mobile phase composition, the type of analyte and physical characteristics of the stationary phase was derived. Fourteen carboxylic acids (comprising mono- and divalent, aliphatic and aromatic acids) were chosen and used to acquire retention data on three different cation-exchange stationary phases (in which the sulfonate functional groups are bound to polystyrene-divinylbenzene, polymethacrylate or silica). Thirteen different mobile phase conditions of varying sulfuric acid concentration and percentage of methanol were employed. These retention data were used to derive the parameters necessary to solve the retention model using non-linear regression. A plot of the calculated retention factors against the experimental values gave a correlation coefficient (r2) calculated by least-squares regression of 0.9755. The retention model was then applied to optimising the separation of nine carboxylic acids. When solutions of sulfuric acids were used as the eluent in ICE, the poor buffer capacity of the eluent introduced a significant error into the mathematical model. Therefore, a second retention model was derived which accounted for unbuffered eluents and also considered the degree of hydrophobic adsorption of the undissociated and dissociated forms of the analyte onto the unfunctionalised polymer surface. The adsorption coefficients calculated from the model were in accordance with expected trends and showed that both the undissociated and dissociated forms of the analyte acids were retained by hydrophobic adsorption effects. The correlation coefficient calculated for this retention model was 0.9863 and the derived equation was able to accurately predict the retention of acids which exhibited strong adsorption effects as well as to optimise the conditions for their separation.