The carbon content of a reverse phase can be used to determine the efficacy of bonding , but values obtained must be used in conjunction with the surface area of the native silica in order to arrive at a meaningful conclusion. The amount of material bonded to the silica will depend, not merely on the reaction efficiency, but also how many silanol groups were available with which it could react; ipso facto it will depend on the surface area of the parent silica. As an example, from thermogravimetric analysis data, the concentration of silanol groups on a silica surface was calculated to be about 7.3 m.mol.m-2. This figure was based on the weight lost by the condensation of silanol groups and excluded the water that was strongly bound or hydrogen bonded to the surface. This value thermodynamically derived, or the value of about 9.0m.mol.m-2 taken from the methylithium method of assessment, are the generally accepted values for the surface concentration of silanol groups. In the following argument the value of 7.3 m.mol.m-2 that has been obtained from TGA data will be used.
The carbon content of the bonded phase is usually determined by microanalysis and the result expressed as %w/w of the combined bonded organic material and the silica gel. Consider a bonded phase where the carbon content is (y)%w/w coated with a hydrocarbon moiety having (n) carbon atoms per aliphatic chain (e.g.10 which would be the value for the dimethyl octyl brush phase)).
The concentration of aliphatic chains in mols. per gram of silica (m')will be,
Consequently, if the surface area of the silica is (A) m2g-1, the number of mols. of aliphatic chains (m) in micromols per square meter will be given by,
or, ........................... (1)
Consequently, the amount of material bonded per square meter, will give an indication of efficacy of the synthetic process and will be directly proportional to the carbon content of the product and inversely proportional to the surface area of the original silica gel. The concentration of organic groups on the surface, however, even when calculated in the rational manner given, is still a rather uncertain figure.
The value for the surface area that is normally used is that taken from the BET test that measures the surface area available to nitrogen. Consequently, the measured value includes a significant area contained in pores that is inaccessible to both the solute and solvents used in LC, not to mention the silanizing reagents themselves. However, in the absence of a better measurement, the concentration of organic groups on the surface (expressed in micromols per square meter), calculated in the manner given above, is the accepted method for measuring the surface coverage of the bonded phase.
According to Odlyha et al (13) if the bonding reaction has been efficient the surface concentration of attached organic groups should be about 4.2mmol.m-2. Assuming that the concentration of silanol groups on the surface is 7.8mmol.m-2 then this would represent about a 58% yield. From the techniques employed so far, it would appear that this is about the maximum yield that can be obtained from any bonding process due to the steriic shielding of many of the silanol groups by the already bonded material.