alpha-Phenylacetoacetonitrile
A solution of sodium ethoxide is prepared from 60 g. (2.6 mol) clean sodium and 700 ml of absolute alcohol (dried over calcium oxide or sodium) in a 2000 ml round-bottomed flask equipped with a reflux condenser. To the hot solution added a mixture of 234g (2 moles) of pure benzyl cyanide 264g (3 moles) of dry ethyl acetate (dried by refluxing over P2O5 for 30min followed by distillation). The mixture is thoroughly shaken, the condenser closed with a calcium chloride tube, and the solution heated on the steam bath for two hours before standing overnight.
The next morning the mixture is stirred with a wooden rod to break lumps, cooled in a freezing mixture to -10°C, and kept at this temperature for two hours. The sodium salt is collected on a 6 inch Buchner funnel and washed four times on the funnel with 250 ml portions of ether. The filter cake is practically colorless and corresponds 250-275g of dry sodium salt, or 69-76% of the calculated mount. The combined filtrates are placed in the freezing mixture until they can be worked up as indicated below.
The sodium salt still wet with ether is dissolved in 1.3 liters of distilled water at room temperature, the solution cooled to 0°C, and the nitrile precipitated by adding slowly, with vigorous shaking, 90 ml of glacial acetic acid, while the temperature is kept below 10°C. The precipitate separated by suction filtration and washed four times on the funnel with 250 ml portions of water. The moist cake weighing about 300g corresponds to 188-206g (59-64%) of dry colorless alpha-phenylacetoacetonitrile, mp 87-89°C.
Phenyl-2-Propanone
350 ml of concentrated sulfuric acid is placed in a 3000ml flask and cooled to -10°C. The total first crop of moist alpha-phenylacetoacetonitrile obtained according to the procedure above (corresponding to 188-206g or 1.2-1.3 moles of dry product) is added slowly, with shaking, the temp being kept below 20°C (If pure dry alpha-phenylacetoacetonitrile is used, half its weight of water should be added to the sulfuric acid or charring will take place on the steam bath). After all is added the flask is warmed on the steam bath until solution is complete and then for five minutes longer. The solution is cooled to 0°C, 1750ml of water added rapidly, and the flask placed on a vigorously boiling water bath and heated for two hours, with occasional shaking. The ketone forms a layer and, after cooling, is separated and the acid layer extracted with 600ml of ether. The oil and ether layers are washed successively with 100ml of water, the ether combined with the oil and dried over 20g of anhydrous sodium sulfate. The sodium sulfate is collected on a filter, washed with ether, and discarded. The ether is removed from the filtrates, and the residue distilled from a modified Claisen flask with a 25 cm fractionating side arm. The fraction boiling at 110-112°C at 24 mmHg is collected; it weighs 125-150g (77-86% of the theoretical amount).
A somewhat higher yield has been obtained by a different variation of this synthesis, originating from Russia.