J. E. COLERIAN, C. KICCIUTIA N D DANIELSWEKN
5342
[ C O N T R I B U T I O N FROM THE
VOl.
7s
EASTERN REGIONAL KESEAKCEI LABOKAFORY'J
Improved Preparation of 9(10),10(9)-Ketohydroxystearic Acids by Oxidation of Oleic Acid with Potassium Permanganate in Neutral Solution2 BY JOSEPH E. COLEMAN, C. RICCIUTIAND DANIELSWERN RECEIVED MARCII19, 1956 The mild oxidation of oleic acid, as the potassium soap, with dilute aqueous potassium perinanganate has been reinvestigated and the marked effect of pH in determining thc course of the reaction has been interpreted. When the pH is maintained in the range of 9.0-9.5 by neutralizing the liberated alkali as i t is formed (equation l), 65-75% conversions to 9( l o ) , 10(9)-ketohydroxystearic acids are obtained. If excess alkali is added before oxidation or is permitted to accumulate (pH 12), yields of ketol are reduced to O-30% and yields of 9,lO-dihydroxystearic acid jump to 60-96%. It has been concluded that the "neutral" oxidation of potassium oleate by potassium permanganate was never actually studied by previous investigators. -4 mechanism based on stepwise (PH controlled) or complete (pH not controlled) hydrolysis of the intcrmediate cyclic coiriplex between oleic acid and potilssium permanganate is proposed to account for this striking reversal of products. Also, with pH coiitrol the concentration of the reaction mixture is increased 20-fold over the best values reported in the literature. The oxidation of claidic acid, however, is only slightly pH-dependent; this is not understood. Octadccadienoic acids are converted prcdoiniuantly to polymeric products.
To obtain a substantial quantity of 0(10),10(0)ketohydroxystearic acids for study, two preparative methods seemed to offer the most promise. The first involved the sequence of steps3s4 oleic acid
Br2
+9,10-dibromostearic acid
stearolic acid
KMnOd ____)
?;aSH2-
iYH3
Zn 9,lO-diketostearic acid ----f 9( 10),10(9)-ketohydroxystearic acids
melting 9,lO-dihydroxystearic acid, 1n.p. 131", is obtained.1° Fourth, if the following balanced equation for the ketol-forming reaction is correct, it is evident that although the system may have been neutral a t the start it should rapidly become strongly alkaline because of the formation of potassium hydroxide 3 -CH=CH-
+ 4KMn04 + 2Hz0 -+ 3 -C-CH + 4M110g + 4KOH 0 /I
OH I
(1)
The second was the so-called ''neutral]' oxidation of oleic acid with dilute potassium permanganate, I n our first experiment we observed that the pH first reported by Holde and Marcussonj and reinvestigated by later workers.6-8 Both methods of a neutral aqueous solution of potassium oleate, leave much to be desired, the first because of the which is about 10 a t 15', rose rapidly t o a niaximuni number of steps and the second because of the poor value of almost 12 by the time only 257, of the calyields (usually below 30%) and the high dilution culated aniount of potassium permanganate had (one gram of acid per liter of reaction mixture7) been added. In view of this result, it can be concluded that except for the initial stages of the reacrequired to attain even these yields. The second method was selected for study for tion, a truly neutral oxidation of potassium oleate several reasons. First, the direct, single-step p r e p by aqueous potassium permanganate has never ration of a-ketols in good yield from oleic acid, bee11 carried out. After about 25% of the caland other unsaturated compounds, was a desirable culated quantity of potassium permanganate has objective and had not yet been accomplished.!' been added, conditions are ideal for the formation Second, the inode of formation of a-ketols from oleic o f 9,lO-dihydroxystearic acid. We have shown acid had not been explained. 9,lO-Dihydroxy- d s o thLt in the absence of pH control the bulk of stearic acid is not an intermediate since it is uii- the ketol is formed in the initial stages of the reacaffected by potassium permanganate under the tion as we had anticipated. Without pH control during the reaction, it is eviconditions of the ketol-forming r e a ~ t i o n . ~Third, the striking effect of :.lksli on the products obtained dent why extremely high dilutions are required for warranted further study and was, indeed, easily best yields of ketohydroxystearic acids. At high amenable to experimental attlck. Thus, it was dilutions the concentration of hydroxide ions is rewell known that if one mole or more of potassium duced and their undesirable effect, as far as forhydroxide is present in excess per mole of potas- mation of ketols is concerned, is diminished. The reaction of potassium permanganate with sium oleate, no ketohydroxystearic wids are formed but an alniost quantitative yield of high- oleate, being too rapid for a kinetic study, was investigated by product analysis. The content of (1) A laboratory 01 t h e Eaatern UtiliLatiuri Ileseuch IJr;inch, A y r i unoxidized oleic acid in the crude oxidation product cultural Research Service, U. S . Department uf Agriculture. Article can be estimated from the iodine number. 9,lOnot copyrighted. (2) Presented a t the Fall hleeting of t h e Arnerican Chernical Society, Diketostearic acid can be determined polarographAtlantic City, S . J., September lti-21, 195Ci. ically," using the pure compound, m.p. 85", as a (3) J. I?. XcGhie, Chemislry & l i i d i r s l r y , 131 (1054). reference standard. Ketohydroxystearic acids are (4) N. A . Khan, I;. E. Deatherage and J . B. Bro%n, J . A m . Oil not reducible under these conditions. DeterChemisls' Soc., 28, 27 (1951). (5) D. Holde a n d J. Marcusson, Ber., 36, 2ti.57 (1'303). mination of carbonyl and hydroxyl oxygen permits (6) R. S . hlorrell and E. 0. Phillips, J . S o c . C h e m . I d . . 57, 245 calculation of the content of 9(10),10(9)-keto(1938). (7) G . King, J . C h e m . S O L . ,1788 (1936). (8) T. P. Hilditch and H . Plimmer, ibid., 204 (1942). (4) S . M. McElvain, T h e Acyloins in "Organic Reactions," Vol. 4, Chapter 4, John Wiley and Sons, Inc., Piew Uork, X. Y.,IS3S.
(10) A . 1,apworth and E . N. I f o t t r a m , J . C h e m So
