Bonded Phases - BrushType Bonded Phases
Figure 1, however, does not give a good impression of the bonded phase surface albeit being difficult to depict it in any other way. The silica surface cannot be flat, it is a random surface in the true sense of the word and the situation depicted above is far from reality. It may be that that there are silanols close to organic moieties that are not reacted, but they will almost certainly not be in the orderly fashion as depicted. In addition, some consider that the hydroxyls are, in fact, clustered together in groups on the surface and are not evenly dispersed. This, however, may be a little extreme and it must be said that there is no sound evidence supporting this concept. Nonetheless, it should be emphasized that the diagrams depicting silica and bonded surfaces given here represent the chemistry involved and not the true physical state of the surface. It is of interest to note that the bonded phases originally made by Halasz and Simpson (although attached by the carbon-oxygen-silicon bond) were also brush type phases. In addition, there were no dimethyl groups involved so the structure is likely to be of the form depicted in figure 1(A). It should be recalled that it was Halasz who first introduced the term 'brush' phase.
The shielding of the adjacent hydroxyl groups at the base of the attached alkyl group (e.g. the two methyl groups in dimethyloctylchlorsilane) have been thought to be the source of phase instability. It has been construed that the unreacted hydroxyl groups close to the sterically hindering side chains, can form active sites for silica erosion in aqueous mobile phases particularly at extreme pH. In fact, both the reduction of these sterically-hindering groups and, in fact, also their enhancement has both been employed in attempts to improve reverse phase stability. A reverse phase with longer side chains (believed to be di-isopropyloctadecylsilyl group) was described by Kirkland (10). This product of Kirklnd was termed a sterically protected C18 reverse phase and was claimed to be much more stable than other reverse phases. It was thought that the longer side chains also protected the hindered hydroxyl group from interacting or attack. In contrast, the company YMC Inc. employed a chlorsilane mixture containing a range of chain lengths up to C8. They claim that " bonding with multiple alkyl silanes could give a greater surface coverage than the typical procedure of bonding with a single alkyl silane and end capping", accordingly there were less free silanol groups on the surface to initiate erosion.