111. INDUSriMAl. l'Ki;i'.\K.\TI()N 'II f) 



is peifornu'd \\i(h pliDsphorus peutachloride at 90"; the evolution of lar^e 

 amounts of liydroj^eu chloride starts, and the nudonitrile is distilled in vdcuo 

 as soon as it is formed; yield 70% (Hesse'")- l*^thyl cyanoacetate is com- 

 mercially' aAailahle in this country. 



3. Ring Closure to the Pykimidine, A + li 



The ease with which the pyrimidine nucleus is foimed varies consider- 

 ably and therefore also the yields. In the most favorable cases no condens- 

 ing agent is necessary, as the alkalinity of the free acetamidine is sufficient 

 to close the ring. In other cases sodium ethylate or sodium hydroxide is 

 used for the ring closure, and in one special case it even happens in neutral 

 or acid solution. Normally the formyl and cyano (or carbethoxy) groups 

 react at the same time, the ring closure taking place in one step. Todd and 

 Bergel'^ have isolated as an intermediate an acrylo compound which could 

 be used for the ring closure in two different ways. 



a. Indirect Introduction of the Amino Group in the 4 Position 



Williams and Cline'^ reacted ethyl sodioformyl-^S-ethoxypropionate with 

 acetamidine hydrochloride and sodium ethylate in ethanol, forming 

 2-methyl-4-hydroxy-5-ethoxymethylpyrimidine. The yield, as (luoted by 

 these authors, is extremely poor, only 3.5 %. If the process has been used 

 for technical purposes, it must have been radically improved. 



By treatment with phosphorus oxychloride the hydroxyl group in the 

 4 position is replaced by chlorine (yield 70 %), which is exchanged with 

 ethanolic ammonia under pressure by the amino groups (yield 70 %) 

 finally the ethoxy group in the 7 position is replaced by bromine (hj'-drogen 

 bromide in glacial acetic acid, yield 90%). Starting from acetonitrile and 

 ethyl acrylate, this synthesis of the pyrimidine moiety recjuires eight steps. 

 With the exception of formylation and ring closure the yields are good and 

 the reactions show no special difficulty. 



When Todd and BergeP^ reacted fi-ee acetamidine \\ith ethyl ethoxy- 

 methylene cyanoacetate, they obtained colorless needles of ethyl a-cyano- 

 /3-acetamidinoacrylate (yield 46%), and all efforts to close the ring directly 

 at various temperatures and various amounts of sodium ethj'late gave no 

 satisfactory results. Treatment of the intermediate by boiling with aqueous 

 2.5% sodium hydroxide closed the ring (yield 37%), giving a total yield 

 for the ring closure of only 17%. 



By replacement of the hydroxyl group with chlorine (60 to 70% yield) 

 and the amino group, 2-methyl-4-amino-5-cyanopyrimidine was obtained 

 with a yield of 40 %. Finally the cyano compound was catalytically re- 



" B. C. Ilesse, Am. J. Cfiem. 18, 72.3 (1896). 



