Feb., 1951
THEDEHALOGENATION OF PROPARGYLIC BROMIDES
state in th.e conifers as was demonstrated by comparative studies on the native, and the enzymatically liberated lignins of white Scots pine. Should this condition prevail for the hardwoods studied then i t must be concluded that the high methoxyl contents of the “Klason lignins” obtained from the woods are artifacts resulting from the vigorous action of the reagents applied during the isolation. A study of the lignin liberated by means of fungal decay of native lignin-free hardwood should reveal which one of the above two possibilities actually prevails. If the lignin so liberated possesses a characteristically high methoxyl content and contains syringyl as well as guaiacyl nuclei, the
[CONTRIBUTION NO.768 FROM THE
first condition should represent the true state of the lignin. If, on the other hand, the enzymatically liberated lignin resembles the native lignin portion and contains only guaiacyl groups, then the conclusion must be drawn that there exists no essential difference between these hardwood lignins and softwood lignin. Such enzymatic studies are a t present in progress. Acknowledgment.-This work was carried out with the aid of a grant from the Office of Naval Research. NEWYORK58, N. Y. RECEIVED JUNE 19, 1950
DEPARTMENT OF
The Propargylic Rearrangement.
693
CHEMISTRY, UNIVERSITY OF PITTSBURGH]
11. The Dehalogenation of Propargylic Bromides
BY JOHN H. WOTIZ Propargylic bromides were dehalogenated in three different ways : by treating their Grignard reagent with water, by means of the zinc-copper alcohol reagent, or by the action of lithium aluminum hydride. The product was always a mixture of acetylenic and allenic hydrocarbons.
Possible reaction mechanisms are discussed.
I n a recent paper’ i t was shown that primary pro- and 1,2-heptadiene (IV). The yields were found to pargylic bromides, R 4 k C - C H 2 B r , can be con- depend on the method of preparation. The findverted in high yields v i a nitriles or by the malonic ings of Hennion3that 3-halo-1-alkynes yield mainly ester synthesis into acetylenic acids. However, 1,2-dienes when dehalogenated with zinc-copperthe same bromides when converted into Grignard alcohol were substantiated. However, the isomeric reagents on carbonation2 yielded a mixture of acet- 1-halo-2-alkyne with the reagent yielded only one ylenic and allenic acids. The present study was half as much allenic as acetylenic product. Individundertaken t o determine whether the formation of ual experiments are listed in Table I, and the yields an organometallic compound is necessary for the are calculated from the amount of hydrocarbons formation of rearranged (allenic) products. Such a distilling over the listed boiling range. Despite the possible explanation was suggested by Hennion and use of high efficiency fractionating columns, imperSheehan3 who dehalogenated n-propylethynylcar- fect separation took place. This was apparent binyl chloride, C3H,--CHC1-C=CH, and ob- from the refractive index and infrared spectrum tained 1,2-hexadiene in 71y0 yields. A small determinations of individual distillate fractions. amount of the isomeric 1-hexyne was found. The The product ratios I11 to I V and V to IV in Table I authors pictured the reaction mechanism as going are estimated from the refractive index and infrared through an intermediate organozinc compound, spectral data. The tabulation of constants of posC3H-CH-C=CH, which in the presence of alco- sible products is given in Table 11. I n Fig. 1 are I given the infrared spectra of two fractions of reacZnCl tion products in the dehalogenation of I1 by method hol yielded the allenic product. B, to illustrate how the products were qualitatively Two isomeric propargylic bromides, l-bromo-2- and quantitatively analyzed. A linear variation heptyne (I), C4H9-CrC-CH2Br, and 3-bromo-l- in refractive .indices of mixtures of isomeric hyheptyne (11), CdHgCH-CeCH, were dehalo- drocarbons was assumed in order to obtain a semiI quantitative estimation of composition. The semiBr genated by three different methods to yield mix- quantitative analysis on the basis of the infrared tures of acetylenic and allenic hydrocarbons. I n spectrum utilized the extinction coefficient for pure the first method (A) the bromides were converted IV, which was estimated for the typical allenic band into organomagnesium bromides2 which were then near 1950 cm.-l. This value allows an estimation hydrolyzed. I n the second method (B) the zinc- of this compound in mixtures by the use of Beer’s copperalcohol reagent described by Hennion3 was law. Usually the minor component was deterused. I n the third method (C) the halogen was mined spectroscopically and the other one by difremoved by the action of lithium aluminum hy- ference. Thus the fractions in Fig. 1 boiling a t 97 dride.4 In all three methods bromide I produced a and 103’, contain 10 and 0% of V, on the basis of mixture of 2-heptyne (111) and 1,2-heptadiene absorption, and 20 and 1%, respectively, on the (IV), and bromide 11, a mixture of 1-heptyne (V) basis of mixed refractive indices. (1) Newrnan and Wotiz, THIS JOURNAL, 71, 1292 (1949). (2) Wotiz, i b i d . , l a , 1639 (1950). (3) Hennion and Sheehan. ibid.. 71, 1964 (1949). (4) Trevoy and Brown, ibid., 71, 1676 (1949).
Discussion of Results The infrared spectrograms of samples in each individual experiment listed in Table I, show con-
JOHN
VOl. 73
H. WOTIZ
TABLE I ANALYSISOF PRODUCTS -,--Halide-
-Reagent-Quant., Quant., mole Methodd mole
Expt.
C41-I&!rCCHzBr C4HgCzCCHzBr C4HpC=CCHzBr C4H&-CCH2Br 5 CaH,CH--C=CH
1 2 3 4
A B C C A
0.6 .3 .3 .4 .2
Total yield,
% '
76 81 94' 47 50
a
0.4 0.125
'
Product
B.p. range, OC.
CSHIIC= CH
$ 1 ~ 5 ~
Ratio CcHpCH= C=CH2
...
C,HPC= CCHa
103-110 103-1 10 103-110 103--110 95-103
1.4315-1.4190 1.4280-1.4208 1.4156-1. 4192b 1.4182-1.4188 1.4153-1.4326
1.1 1.0 1.0
...
c
1.0
2.6
...
... ...
1.0 2.0 15.0b c
~
Br C&CH--C=CI€
(j
+3
B
D
-r
t i
95-103
1.4203-1.4330
1.0
22.0
...
,1
