the inanufacturer as DC 200 and has a viscosity of 1000 centistokes a t 25". The peduorodimethylcydohexane was obtained from Halogen Chemicals, Inc., Columbia, S. C. Its isomeric composition was not specified. MARQUETTE UNIVERSITY SCOTTL. KITTSLEY
I t is apparent that the organolithium compound retained its configuration a t - 70°, but racemized when the temperature was raised to 0". The pphenylphenacyl ester derivative (from acid A) melted a t 59-60' after five recrystallizations and showed a rotation of [ C Y ] * ~ D-2.8" (0.0515 g./cc. MILWATJFCEE 3, WISCONSIN HERBERT A GOEDRY benzene, 1 = 1). A mixed melting point with an RECEIVED JTJLY 21, 1930 authentic sample showed no depression. ___ ____ From reported rotations for 2-iodoOctane3"and FORMATION OF OPTICALLY ACTIVE 2-methyoctanoic it is estimated that about 1-METHYLHEPTYLLITHIU'M eighty per cent. racemization occurred in the conSir: version of the iodide to the acid. The configuraThe carbonation of the products from a reac- tional relationships are such that either predomition of s-butyllithium and (-)2-iodooetane has nate inversion or retention must be assumed for led to the isolation of (-1.2-methyloctanoic acid. both the interchange and carbonation reactions We can interpret this fact only on the basis that s-Butyllithium and 2-iodooctane in pentane in (1) optically active I-methylheptyllithium was the absence of ether did not react significantly formed by a metal-halogen interchange reaction over a period of forty-two hours a t -77". The and (2) asymmetry was retained during the car- fact that ether so facilitates the interconversion bonation reaction. Previous attempts t o prepare suggests that the reaction proceeds by an ionic asymmetric organometallic compounds of this rather than by a free radical mechanism. type have been unsuccessful.' (3a) Pickard and Kenyon, J Chem S O L 99, , 69 ( 1 Y 1 1 ) I n a typical experiment, a solution of 18.4 g. of {b) Kothen and Levme, J Ckevs Phys , 7, 975 (1939) 2-iodooctane ( ~ ~ ~ ~ -45.86', 3 1 ) d26r 1.3181, W2'D THECHEMISTRY DEPARTMEVP 1.4861) in 100 cc. of petroleum ether (b. p. 28-38') was added over a two-hour period t o a petroleum NORTHWESTERN UNIVERSITY ROBERT L. LETSIYGER ether solution of 320 cc. of 1.04 M s-butyllithium Evilvsro~,ILLINOIS RECEIVED AUGUST21, 1950 and 25 cc. of peroxide free ether a t a temperature of -70'. Two minutes after completion of this addition the mixture was carbonated on Dry Ice From the acid fraction was isolated 14.84 g. of 2- MECHANISM OF THE OXO AND RELATED REACTIONS. 111. EVIDENCE FOR HOMOGENEOUS methylbutyric acid ( a 2 l O.OO', ~ = 2) and 0.66 HYDROGENATION g. of 2-methyloctanoic acid (b. p. 133' a t 11 mm., Sir: CY~%D -0.75 * 0.01', 1 = 0.5, n Z 51.4270). ~ It has been shown that the hydroformylation The p-chlorophenylisothiuronium salt of 2-methyloctanoic acid melted a t 143.5' after recrystalliza- (oxo) reaction is often accompanied by hydrogenation from dioxane, and the mixed melting point tion; indeed, in some instances hydrogenation with a pure sample of the racemic salt was the proceeds to the exclusion of hydroformylation.1,2 same. Regeneration of the acid from this salt We wish to report evidence strongly supporting yielded a material with undiminished rotation and the view that the hydrogenation is homogeneous and, in this respect, unique. ?Z=D 1.4280.2 The organic portion from the carThree experiments were performed a t 185" bonated products was halogen free, and the highest rotation of any fraction was CY% -0.19, E = with reduced cobalt as catalyst and butyraldehyde as substrate. The catalyst was prepared by 0.5 (for the portion boiling 125127' at 10 mm.). Three subsequent reactions differed in that the treating 9.3 g. of cobaltous formate and 50 ml. iodide was added over a period of twenty minutes of cyclohexane with 2000 p.s.i. of hydrogen a t was carbonated (-4) after one 185' for 2 hours in a stainless-steel autoclave of t -70°, (B) after sixty minutes 500-ml. capacity. An initial partial pressure of ) after warming to 0" over a 2000 p.s.i. of hydrogen was used in each of the twenty minute period. The results are indi- three experiments and the partial pressure of carbon monoxide was varied. cated in the table. a* iodide &D acid In the first experiment, no carbon monoxide (1 = l i (1 = 0 5 ) Reaction was present. The reduction proceeded smoothly, -44.36 -1 18 A and the calculated pressure drop was observed. -41.00 -0 96 B Distillation yielded 57 g. (77%) of butanol-1, -39 8.5 0 00 c b. p. 115.5-117°. I n a second experiment, 1000 (1) Tarbell and Weiss, Tms JOURNAL, 61, 1203 (1939); Porter, p.s.i. of carbon monoxide was added (total presibid.. M, 1436 (1936); Schwa- and Johnson, ibid., 118, 1063 ( l 9 3 l ) , sure 3000 p.s.i.). Approximately one mole of gas Pickard and Kenyon, J . Cbrm. SOC.,SS, 86 (1911): Zicglerand Wenz, was again absorbed and 49 g. (66%) of butanol-1, BEY..88,364 (1960); Wittig, Vldal and Bdnwt, Bid., 88,369 (1950); Wallis and Adams, TH18 JOWBNAL, I S , a836 (1933). The evidence b. p. 116-117.5', isolated. The reaction was preof Wallis and Adams fm an asymmetric carbanion was shown to be sumably homogeneous with the soluble dicobalt aroneoua by Wittig, Vidal and Bohnert. I
(2) Tha rcfmctive index of a pursed sample of 2-methylactPnoic add prplvcd from a orl-d reagent WM i 1 8 b 1.4280.
(1) Wender. Levine and Orchin, TEW JOURNAL,72, 4876 (1950) ( 2 ) R, Adkina and 0. h k , ibid., 70,888 (1949).
