~or~ranf of Ci~r,orrrnto~YapJry, 7s (1973) 22%232 &Tj Elscvicr Scicrltilic l~ublishiny Conipitny, Arnstcrclnm --Printed in 'I%3 Wcthcrlancls
Note
The gas chromatographic properties of a modified, wide-pore silica gel* Silica gels have been used for gas cllromatography (GC), with polar and unpolar carriers, since the early ninctcenforticsl.
Their mak application today is in the analysis of gases: they play little role in the separation of Iiiglier, especially polar organics. Tllere, diatomaceous supports are decidedly superior.
Yet silica gels command potential interest lmnuse they arc inespcnsive and easy to product with a wide range of surface areas. Various proccsscs of deactivation, ~6, silylation, l~ydrothcrmal treatment, coating wit11 inorganics, polymers or polar molecules from the carrier gas stream, etc., have been employed to improve their performance with heavier 0rganicsQ. Improvements could be demonstrntcd but, as a wllole, silica gels have remained inferior to diatomaccous supports in GC efficiency.
In this study, we decided to modify a low-priced silica gel -Davidson grade 62 (ref, 3), which has a low surface arcn of 340 1ii2/g, il large average pore cliamctcr of 170 A (ref. 4), a high pore volume of I.1 j cnP/g, arid sells in the scientific retail trade for little more than a dollar per pound3. This material is not commonly used in GC, although it is capable of suprisingl>* good separation of hydrocarbons (Fig. I, bottom trace) .
Three simple dcnctivation processes wcrc to bc used -singly or in combination, First, WC wanted to reduce further tilt available surface area and widen tile pores. This can bc ncllicved by cnlcination, by treatment with salts or by treatment with water at lligll tempcraturc.
Ncitller process has been described in much cspcrimental detail. We decided to use the! last trnc, called "lrydrothcrmal treatment" under csperimental conditions (especially the state and amount of water) vAich may or may not have been identical to the procedures described in the literature'-Q.
Second, we wanted to deactivate the silica gel surfacelo by a non-estractable polymer layer: this is similar to a technique used to product (nominally) monomolecular layers of Carbowax zoM on Chromosorbl~.
Third, WC wanted to deactivate the silica gel surface by adding small amounts of a polar compound, in our case water, to the carrier gas. Tile use of polar mobile pleases is well known and has been repeatedly applied (c.g. 12-23, compare ref. 24, q). Genwal $vc)avation of th silica gel. Silica Gel Davidson 62 (Fisher Scientific Co,, St. Louis, MO.), 40-60 mesll , was washed in a Soshlet (Kontcs Model K-~S~IOO) with IX1 at boiling point temperatures until no yellow hue could bc detected in a fresh charge of cone, I-ICI, hot or cold, after several hours of estraction. The silica gel was then washed with distilled water to neutrality. H~:lrollzc~r?zal tvcatrne?tt. A high-temperature bomb (Model 4740 wit11 Pyres