MECHANISM O F A 4 L L E N I M I N E FORMATION
Jan. 20, 1962 157-158”; reported20 m.p. 97-99’, 152-1 54 .
m.p.
phenylurethan
O
(20) J. D. Roberts, E. R.Trumbull, W. Bennett and R , Armstrong, J . A m . Chem. SOC.,72, 3116 (1950).
[CONTRIBUTION FROM THE CHEMISTRY
195
Acknowledgment.-This investigation was made possible by the award of a visiting studentship (to K* from the Kao Soap CO’, Ltd’, Japan.
DEPARTMENT, UNIVERSITY OF CALIFORNIA, DAVIS,CALIFORNIA]
Amines Derived from Dlhalopropenes. I. The Mechanism of Allenimine BY ALBERTT. BOTTINIAND ROBERTE. OLSEN’~ RECEIVED JULY 31, 1961 Treatment of N-(2-bromoallyl)-n-propylamine (Ia) with sodium amide in tritium-labeled liquid ammonia has been found to yield radioactive N-n-propylallenimine (Ha). No significant exchange with solvent of the hydrogens bonded to carbon of Ia and IIa occurred under the reaction conditions. Compound IIa was degraded and determination of the radioactivity of the degradation products showed that 99.8 2~ 0,3’% of the tritium was incorporated in IIa a t the ring-methylene group. These results are consistent with the proposal that an N-alkylallenimine is formed v i a an elimination-addition mechanism involving an allenic amine intermediate (111).
When an N-(2-bromoallyl)-alkylamine (I) is treated with an alkali metal amide in liquid ammonia the principal product is the N-alkylallen( 1-alkyl-2-methyleneaziridine1 11). A imine small amount of the corresponding N-alkylpropargylamine (IV) also is formed. Several plausible mechanisms can be written to explain the formation of I1 from I.3 Internal displacement of bromide ion can be envisaged as occurring by an S ~ 2 - t y p reaction e (1) or an additionelimination reaction (2). Formation of I1 can also be the result of an elimination-addition reaction (3) involving an allenic amine intermediate (111). The possibility of an analogous elimination-addition mechanism (4) involving IV is precluded by the high yields of N-alkylpropargylamines obtained from reactions of N-(2-chloroallyl)-alkylamines with alkali metal amides in liquid ammonia. Ib, From consideration of these mechanisms, i t appeared likely that a choice between paths 1, 2 and 3 could be made by examination of I1 obtained from an appropriate N-(2-bromoallyl)-alkylamine (I) and amide ion in tritium-enriched ammonia. The absence of tritium in I1 would indicate that 1 is the reaction path since intermediates V in 2 and VI1 in 3 would be expected to abstract a proton from the solvent. However, the absence of tritium in I1 could not be used to rule out 2 as the reaction mechanism since i t is conceivable that loss of bromide ion from V to yield unlabeled I1 could occur a t a much faster rate than the reaction of V with ammonia to yield radioactive VI. If I and I1 (aside from the amine hydrogen) (1) (a) Acknowledgment is made to the donors of the Petroleum Research Fund, Administered by the American Chemical Society, for partial support of this research. (b) Presented in part at the 140th National meeting of the American Chemical Society, Chicago, Ill., September, 1961. (c) American Chemical Society-Petroleum Research Fund Fellow, 1961. (2) (a) C. B. Pollard and R . F. Parcell, J. A m . Chem. Soc., 73,2925 (1951); (b) A. T.Bottini and J. D. Roberts, ibid., 79, 1462 (1957); (c) unpublished work of A. T. Bottini, R . E. Olsen and B. J. King. (3) Cf.S. I. Miller and P. K . Yonan, J. A m . Cham. Soc., 79, 5931 (1957), and D. E. Jones, R . 0. Morris, C. A. Vernon and R . F. M. White, J . Chcm. Soc., 2349 (1960).
do not undergo exchange with the solvent under the reaction conditions, presence of tritium in I1 would definitely rule out 1 as the exclusive reaction mechanism. A choice between 2 and 3 could be made by degrading I1 and determining the location of the incorporated tritium. Path 2 would yield I1 labeled a t the exocyclic-methylene group; path 3 would yield I1 labeled a t the ringmethylene group. It was decided to treat N-(2-bromoallyl)-n propylamine (Ia, R = n-CsH,) with sodium amide in tritium-labeled liquid ammonia because it appeared likely that the N-n-propylallenimine (IIa, R = n-C3H7)obtained could be hydrogenated to di-n-propylamine in high yield2aand a convenient procedure for conversion of primary and secondary amines to the corresponding carbonyl compounds had been worked out in these laboratories. Compound I a was prepared in 81% yield by treatment of 2,3-dibromopropene with excess n-propylamine in water. Treatment of I a with sodium amide in liquid ammonia which had been equilibrated with a small amount of tritium oxide gave I I a in 39Qj, yield together with a small amount of N-n-propylpropargylamine (IVa, R = n-C3H7). Compound I I a was freed of the last traces of IVa by distillation from lithium aluminum hydride a t reduced pressure. Determination of the radioactivity of purified I I a indicated that i t had a specific activity equal to the calculated specific activity of the hydrogens of the ammonia, i.e., one-third the specific activity of the ammonia. The hydrogens bonded to carbon of I a and I I a did not undergo exchange with the solvent under the reaction conditions since treatment of excess I a with sodium amide in tritium-enriched liquid ammonia and treatment of radioactive I I a with sodium amide in ordinary liquid ammonia caused no significant change in the specific activity of either compound. These results ruled out 1 as the exclusive mode of formation of IIa. The equal specific activities of the product and the hydrogens of the solvent, which is to be expected if a carbanion intermediate such as VI1 gr, possibly, V is
196
ALBERTT. ROTTINI AND ROBERTE. OLSEN
VOl. 84
s-
Br
Br
I
/
II
BI
I
“s,
CH,--G--GH2 /-NH3 Rr
H
/
-
R
R
‘N’
I
L a
I
GH3-C-CH,
-NH3, Br-
N‘’
-NH2
I R
R
\
“3
-NH;
It
I
R
formed in the r e a ~ t i o nindicate ,~ further that path 1 accounts for little, if any, of the I I a f ~ r m e d . ~ Radioactive I I a was degraded to propionic acid in four steps. Liquid products were converted to colorless solid derivatives which were purified by several recrystallizations, and the specific activities of the derivatives were determined by the scintillation count technique. The activities of the degradation products are shown in 5 as percentages of the specific activity of TIa.
EH,EH,CO,H 27 5%!0 7 %
0 Z % t O 3%
NO2
I T (4) R. Wiberg, Chem. Ress., 65, 713 (19553. (5) Treatment of N-(2-bromoallyl)-methylamine (Ib. R = CsHv) with sodium amide in liquid ammonia containing 0.408 #c./mmole of tritium yielded N-methylallenimine (IIb, R = CHa) with a specific activity of 0.136 pc./mmole. Treatment of inactive IIb with sodium amide in liquid ammonia containing 1.0 /rc./mmole yielded IIb with a specific activity of