A Short, One-Pot Synthesis of Bupropion

(Zyban®, Wellbutrin®)

Journal of Chemical Education Vol. 77 No. 11, 1479 (2000)

[ Back to the Chemistry Archive ]

Bupropion, the hydrochloride salt of (+-)-2-(tert-butylamino)-3'-chloropropiophenone, has a unique pharmacological profile. It was first marketed in 1985 by Burroughs-Wellcome (now Glaxo Wellcome) as an antidepressant, under the trade name Wellbutrin. Unlike selective serotonin reuptake inhibitors such as Prozac, Zoloft, or Paxil, it does not interfere with sexual performance, is less likely to cause drowsiness, and is as effective as Ritalin in the management of attention-deficit hyperactivity disorder (ADHD) (1). But its most interesting feature was discovered by accident: in early clinical trials, many smokers taking the drug reported that after one or two weeks their craving for tobacco seemed to fade and they were able to quit smoking with few or no withdrawal symptoms (Johnston, J. A., head of Psychiatry Clinical Development, Glaxo Wellcome; personal communication, 14 Dec 1998). When double-blind studies confirmed these anecdotal reports (2), bupropion was marketed in 1997 with a new name, Zyban, for use as an aid in smoking cessation.

Most organic chemistry students are motivated by syntheses of well-known pharmaceuticals (3), and an "anti-addiction" drug is particularly interesting. Also, sad to say, many of them smoke, and a synthesis of Zyban provides an opportunity for classroom discussions that connect academic science to daily life-even to life expectancy. So we investigated various approaches for the synthesis of Zyban. By modifying the published procedures (4, 5), which require more than 24 hours and give yields below 40%, we developed a short, onepot synthesis that can be carried out by students in the introductory organic laboratory in less than two hours and gives material 98% pure in an average overall yield of 80%.

The greatest improvement came from using N-methylpyrrolidinone (NMP) in place of dimethylformamide (DMF) as a solvent for the amination of 2. In DMF the reaction can take many hours (4, 5), whereas in NMP it is complete in less than 10 min at 50-60°C. The short reaction time enhances the yield because the free base of 3 (but not its hydrochloride salt) is significantly liable to decomposition. Additionally, our one-pot procedure skips the isolation of intermediate 2 and prevents student exposure to a lachrymatory alpha-halogenated ketone. The lab provides motivation for studying alpha-halogenation and the influence of a nonprotic polar solvent on competition between nucleophilic substitution and elimination reactions.


One gram (5.9 mmol) of m-chloropropiophenone, 1, is dissolved in 5.0 mL of CH2CI2 in a 50-mL round-bottom flask. A few drops of a 1.0 M solution of Br2 in CH2Cl2 are added with stirring and the reaction is briefly warmed to initiate the reaction (as judged by the disappearance of the color of the bromine). The flask is then placed in an ice bath and 6.0 mL (6.0 mmol) of the bromine solution is added dropwise with stirring. The CH2Cl2 is removed by distillation, 5 mL of tert-butylamine and 5 mL of NMP are added, and the flask is heated in a 50-60 °C water bath with stirring for 10 min. The contents of the flask are then transferred to a separatory funnel, 25 mL of water is added, and the mixture is extracted three times with 25-mL portions of ether. The combined ether extracts are washed with five 25-mL portions of water, dried over anhydrous K2CO3, and transferred to a beaker chilled in an ice bath. A 20:100 v/v mixture of concd HCl and isopropyl alcohol is added dropwise with stirring until the contents are acid to pH paper. Sparkling white crystals form, which are bupropion hydrochloride, 3. They are collected, washed with ether, and dried. The product is identical with authentic material by GC-MS,1 TLC (5), and HPLC.


All steps should be carried out in a hood. Bromine is caustic to skin and lungs. Dichloromethane vapors are toxic and possibly carcinogenic; t-butyl amine is toxic. Ether is extremely flammable. Intermediate 2 is a lachrymator.


  1. m/z 240 (M• + 1, -1%), 100 (C6H14N, 33%), 44 (C2H6N, 100%). The only literature data on this substance are the TLC (5) and mp (4). The latter is unreliable: both our product and an authentic sample from Sigma-Aldrich turn brown and decompose at or below the published temperature of 233-234 °C.
  2. 100% peak at R, = 6.6 min (Whatman Partisil 10 ODS-3 column; 1.0 mL/min; eluant MeOH-MeCN-0.012 M KH2PO4, 43:40:17).


[1] Perrine; D. M. The Chemistry of Mind Altering Drugs, History Pharmacology; and Cultural Context American Chemical Society (Oxford University Press): Washington, DC, 1996; p.237.
[2] Hurt, R. D.; Sachs, D. P. L. Glover, E. D.; Oxford, K. P; Johnston, J. A.; Dale, L. C.; Khayrallah, M. A.; Schroeder, D. R; Glover, E N.; Sullivan, C. R; Croghan, I. T; Sullivan, P. M. N. Engl J. Med. 1997, 337,1195-1202.
[3] Perrine, D. M.; Sabanayagam, N. R; Reynolds, K. J. J. Chem. Educ. 1998, 75, 1266.
[4] Mehta, N. B. Meta Chloro Substituted alpha-butylaminopropiophenones, U.S. Patent 3,819,706, Jun 25, 1974.
[5] Hill, j. A.; Scharver, J. D. J. Labelled Compd. Radiopharm. 1988,25,1095-4104.