Investigating reaction schemes for improving silica-based monomeric bonded stationary phases for reversed-phase liquid chromatography

The first study involves "pre-capping" Type-B silica. Previous work showed that monomeric stationary phases made by pre-treating the silica surface with small amounts of trimethylsilane (TMS) reagents prior to C18 silanization showed vast improvements in the chromatographic efficiency, phase loading, and retention with a maximum at approx. 5% pre-capping. It was concluded that this pre-capping step improved efficiency by selectively neutralizing the most reactive highly-acidic silanol sites, so-called silanol "hot spots," producing a more energetically-homogenous surface prior to exhaustive C18 derivatization that subsequently yielded a more evenly-distributed alkyl bonding arrangement. These previous studies were performed on Type-A silica, an older variety of silica gel material containing higher levels of metal impurities than the Type-B silica used today. It has since been argued that metallic impurities are the primary cause of silanol hot-spots, and that pre-capping Type-B silica would have little or no effect, however the experimental evidence has yet to be produced, and there exists the potential for heterogeneous silanol reactivity inherent in the amorphous silica gel regardless of purity. The purpose of the work presented here is to determine the effects of TMS pre-capping on Type-B silica as compared to the previous Type-A results, with the goal of establishing pre-capping protocol for Type-B silica and to form a better understanding of its chemistry.;The current work performed on three Type-B silica substrates of various physical and chemical properties demonstrated optimal TMS pre-capping at approximately 2.5%. The results at this level show only a slight improvement in efficiency for non-polar compounds ( 25%) was observed for some drug compounds and bases under buffered conditions, with the magnitude of the improved efficiencies correlating with metal impurity content and physical parameters of the silica substrate. Pre-capping also resulted in a slight decrease in retention and hydrolytic stability due to a decrease in bonded phase density. The results lend supporting evidence that metal impurities are the primary source of highly acidic silanols, but they also suggest a means to improve efficiency of basic analytes on certain Type-B silica substrates. It was concluded that TMS pre-capping Type-B silica is best done at low levels (<2.5%) to maximize efficiency while maintaining bonded phase loading and stability.;In the second study, quaternary ammonium salts were added into dichloromethane (DCM) reaction solvent and carbon loading was measured. Phase products showed increased phase modification and faster kinetics for both the primary C18 modifcation and TMS end-capping reactions upon addition of these salts at low levels (10 mM). The results suggest that this effect is predominately due to the salt's ability to mitigate the energetic barrier at the interface between the polar silica surface and the non-polar reaction solvent. Larger tetrabutylammonium salt cations induced more of an effect than small tetraethylammonium, indicating that the salt acts to improve steric interactions of the grafted C18 chains at the surface. Also, an increase in DCM/silicate contact angle and decrease in surface tension were observed for DCM upon addition of tetrabutylammonium bromide (TBABr), which indicate a lower surface energy and faster diffusion of reactants and products across the phase boundary. Surprisingly, the density of DCM decreases with TBABr, and an optimization point for calculated capillary pressure gave a maximum for 10 mM TBABr in DCM. A brief investigation into the effects of elevated pressure in the C18 reaction yielded a large increase in bonded phase density at 6.5 atm.;More conclusively, high concentrations of TBABr salt (50 mM) showed a drastic decreased in bonded phase loading and was attributed to ionic suppression and/or shielding of the silica surface. This implies that increased ionic content in the reaction medium is significantly detrimental to silane ligand bonding density. This is a primary concern considering that all silanization reactions that use halosilane reagents result in the formation of ionic products. More study of the effect of ionic strength in the silanization reaction is warranted. (Abstract shortened by UMI.).