Bonded Phases - The Fluidized Bed Method for Bonded Phase Synthesis

The fluidized bed is contained in a tube about 25 cm long, 4.5 cm in diameter that is situated in a heating jacket. The bed temperature is monitored by three thermocouples placed in the center and at either end of the bed. A condenser is placed at the top of the bed that returns unreacted silanizing reagent to the vapor generator. The vapor generator is a simple boiling flask that can be provided with a secondary nitrogen stream if the silanizing reagent does not boil at reasonable temperatures. The silanizing reagent vapor passes from the vaporizer through a pre-heater and then into the base of the fluidizer.

By use of an appropriate valve system the nitrogen may be passed directly to the pre-heater if needed and then to the fluidizer. A steam generator, similar in form to the silanizing reagent vaporizer, is also attached to the pre-heater that can supply steam to the fluidized bed if, and when required. In this way, the fluidized bed can be used to hydrothermally treat the silica before silanization. In the apparatus developed by Khong and Simpson the hydrothermal treatment could not be carried out at elevated pressures but it could be performed at temperatures up to 300oC. A flowmeter continuously monitored the nitrogen stream and all vapor and nitrogen streams were adequately fitted with valves to allow the maximum operating flexibility.

The fluidizer was normally loaded with about 25 g of silica (which provides a fluidized bed about 5 cm long) and is then heated to 200oC for about 3 hrs in a fluidizing stream of preheated nitrogen. The silanizing reagent is then heated to its boiling point in the vaporizer (or to a temperature where it has a significant vapor pressure) and the reagent blown through the fluidized bed by a stream of nitrogen. Excess, unreacted reagent is condensed and passed directly back to the vaporizer. Originally, the reaction was allowed to proceed for about 6 hours, the reagent vapor flow then stopped and replaced by a stream of pure nitrogen. When there was no longer reagent returning to the vaporizer, the bed was allowed to cool to room temperature with the nitrogen still flowing. When cool, the nitrogen flow is arrested, and the bonded phase can then be removed from the fluidizer. Conversely, the silanizing reagent can be replaced with a capping reagent and the material capped by a similar procedure and the product then removed. Farrell (18) studied the kinetics of the reaction at length. He demonstrated that in the synthesis of the C8 'brush' phase, the reaction proceeded quite rapidly and the reaction was completed in as little 30 to 60 minutes. As a result of Farrell's work, the reaction times used in subsequent syntheses with the fluidized bed were significantly reduced.

Khong and Simpson (19) examined the reproducibility of the fluidized bed method of bonded phase synthesis and compared the properties of the product with those produced by the conventional reaction procedure in a solvent. They determined the standard deviation for the carbon content of a series of four bonded phases produced by the usual method of synthesis from high-density silica and found it was 4.85%w/w for a mean carbon content of 9.5%w/w of carbon. On an absolute basis, this would be equivalent to a standard deviation of 51%. In contrast, the standard deviation of the carbon content of four bonded phases synthesized by the fluidized bed method was only 0.75%w/w for a mean carbon content of 12.2%w/w. On an absolute basis this corresponded to a standard deviation of only 6.2%. Khong and Simpson also demonstrated that the improved reproducibility of the fluidized bed product was even further enhanced when low-density silica was used as the starting material. (low-density silica generally has a large pore size and a relatively low surface area and so more of the silica surface is available to the reagent). The authors also established that the products from the fluidized bed had, on average, a higher carbon content relative to those bonded phases obtained by the solvent process employing the same silica. As would be expected the material from the fluidized bed synthesis also gave significantly higher specific retention volumes.

In summary, the fluidized bed synthesis provides a more reproducible product and eliminates many of the tedious operations that are involved in the alternative method, such as solvent removal and recovery, washing procedures and other manipulations. The fluidized bed synthesis also allows very complicated syntheses involving multiple steps to be carried out using relatively simple procedures. However, the technique requires more elaborate apparatus and must be operated by an experienced technician.