I found a nice procedure [see: JACS (1948) vol 70 p 1820] for the synthesis of 1-methyl-4-piperidone. I guess this procedure will also work if you you use phenethylamine (which is already a liquid) instead of liquefied methylamine. Then, we'll get 1-phenethyl-4-piperidone.
Synthesis of methyl-di-(beta-carbethoxyethyl)-amine
___________________________________________________
A cooled one-liter reaction bomb was filled with 432 ml. (400 g.) of ethyl acrylate (containing 0.25 % hydroquinone inhibitor) and 86 ml. (62 g.) of liquefied methylamine (we use phenethylamine). Crystallization occured upon stirring these chilled reactants. The covor was quickly secured since the temperature rose to around 80 °C within five minutes. The reaction vessel was heated in a water-bath at 60-70 °C for one hour. The bomb was cooled, opened, and the reaction mixture distilled. The fraction boiling at 110-119 °C (mainly at 118-119 °C) at 0.5 mm was methyl-di(beta-carbethoxyethyl)-amine; it weighed 367 °C
Synthesis of 1-methyl-3-carbethoxy-4-piperidone hydrochloride
_______________________________________________
Into a three-liter, three-necked flask equipped with a Hershberg stirrer, dropping funnel, and a reflux condenser, were placed 800 ml. of dry thiophene-free benzene and 54 g of sodium hydride. After flushing the apparatus with nitrogen, 30 g. of methyl-di-(beta-carbethoxyethyl)-amine was added to the vigorously stirred suspension of sodium hydride in benzene.
Five minutes after adding two milliliters of absolute ethanol the reaction started as evidenced by the evolution of hydrogen. When, after five more minutes, the reaction mixture was noticably warm, 201 g of methyl-di-(beta-carbethoxyethyl)-amine was added at such a rate as to keep the mixture refluxing briskly. During the addition of the di-ester the appearance of the reaction mixture gradually changed from a dark-gray fluid to an almost white paste. To facilitate stirring an additional 250 ml. of benzene was added after addition of the di-ester was completed. The mixture then was stirred and refluxed with external heating until no more hydrogen was evolved.
A crock of crushed ice was placed under the reaction flask to cool the mixture while 135 g of glacial acetic acid was added. To this very slightly acidic solution, cooled to 5 °C , 123 ml of water was added to precipitate sodium acetate trihydrate. This salt was filtered off and washed with 350 ml. of benzene.
The combined filtrates were distilled to remove ethanol and water. After 600 ml. of distillate had been collected, a refractive index showed that pure benzene was distilling over. The residual solution of 1-methyl-3-carbethoxy-4-piperidone was dilluted with 500 ml of absolute ether. This solution then was cooled in an ice-salt-bath and threated with dry hydrogen chloride until the 1-methyl-3-carbethoxy-4-piperidone hydrochloride had precipitated. The yield of the product softening at 115 °C, melting at 125-128 °C, was 201 g. (91 %).
Synthesis of 1-methyl-4-piperidone
__________________________________
To a one-liter flask containing 350 ml of 20 % hydrochloric acid was added 86 g. of 1-methyl-3-carbethoxy-4-piperidone hyrdrochloride. After refluxing for one hour, the ferric chloride reagent gave no coloration. The solution was evaporated to dryness on a steam-bath at 10 mm. pressure. The solid product, heated at 100 °C for 4 hours at 0.1 mm and further dried over solid KOH for 24 hours, weighed 57.7 g, m.p. 80 - 120 °C.
Although this melting range goes above that of the pure comound 0.45 of crude material dissolved in 90 ml. of hot acetone gave 0.40 g of pure compoundd melting at 93 - 95 °C. Other samples of crude piperidone hydrochloride showed even higher melting points than the one mentioned above, yet this apperently impure material always gave good yields of sharp melting product when recrystallized.
If you intend to make something as poisonous as fentanyl, you probabilly don't care about playing with benzene. But for those who care about their health, the cyclization of methyl-di-(beta-carbethoxyethyl)-amine can also be done with sodium ethoxide (yield = 80 %) See: JACS (1933) vol 55 p 1233
