A. Condensation with Citral
Taylor originally described the condensation of olivetol with citral using benzene as solvent and 10% boron trifluoride etherate as condensation catalyst. The yield was 5-10% of THC-B as part of a larger amount of resin which he partially separated by column chromatography into two fractions. The first fraction (probably inactive) was incorrectly identified (Reference 355) while the second fraction contained the THC-B together with an equal amount of the (+)- isomer.
Mechoulam (Reference 355) has shown that by using methylene chloride as solvent and only 1% of boron trifluoride etherate there is obtained a 10% yield of THC-A instead of THC-B. Because the yield is better with this modification it is described in the example.
Taylor describes a chromatographic separation procedure by which the product can be separated into two parts, the second of which contains the THC-B. This is still accompanied by other by-products and it is impractical to separate further. If pure THC-B is required one of the other syntheses should be used. For many purposes it will be sufficient to extract the reaction mixture with alkali solution (to remove unreacted olivetol) and to collect the crude product resin by distilling off the solvent under water-pump reduced pressure. For details on the partial separation (Reference 178) can be consulted.
The optimum ratio of reactants is not indicated in the literature, however Taylor recovers unreacted olivetol by extracting the reaction product with alkali. (this also serves to decompose the catalyst and should not be omitted). Equimolar amounts have been used in the example, however it is likely that better results will be obtained by using the cheap and easily available citral in excess. This should maximize the yield based on the valuable olivetol.
Example: 180 grams olivetol (1 mole) in sufficient methylene chloride to dissolve it is added to 152 grams (1 mole) citral, dissolved in similar amount of methylene chloride. The solution is cooled to 5-10 deg. C. in an ice bath and one-hundredth mole (1.4 gram) of boron trifluoride etherate is added with stirring. The solution is stirred and kept at 5-l0 deg. C. until the reaction has completed. (Because the time required is not indicated by Taylor experiment is needed. the similar condensation with (-)-verbenol requires 10 minutes and this is probably about right). At the end of the reaction the solution is extracted with several portions of dilute aqueous alkali solution to decompose the catalyst and extract unreacted olivetol. (The olivetol or other 5-(alkyl)-resorcinol is recovered from the aqueous extract by acidifying and then extracting the 5-(alkyl)-resorcinol with ether. Finally the ether is distilled off on the boiling water bath on an electric hot plate). The original methylene chloride solution of product, following extraction of the olivetol is probably best dried by shaking with a little Drierite or anhydrous magnesium sulphate, then the product is isolated by distilling off the methylene chloride. Reduced pressure can be used for the removal of solvents, but is probably not necessary due to the low boiling points. (References 178 and 355)
Gaoni (Israel Journal of Chemistry, 6, page 679, 1968) has described a method of isolation which may be an improvement on the above. The reaction mixture of THC, methylene chloride, and boron trifluoride-ether had added to it a volume of ether and a volume of water (each equal to the volume of methylene chloride used). After stirring, the, organic layer containing the product was separated and washed with sodium bicarbonate solution. He followed then with a washing with saturated sodium chloride solution, and finally dried the solvent solution of product with a little anhydrous sodium sulfate before evaporating.
Intermediate Ketone: Method II
Preparation and Use of (-)-Verbenol