Nov. 5 , 1955
NOTES
5753
I t s melting point was unchanged after subjecting i t t o a ninetube countercurrent distribution using pH 3.5 phosphate buffer-isopropyl ether as the distribution phases. Anal. Calcd. for Cl0HI2O5:C, 56.60; H , 5.70. Found: C , 56.81; H, 6.07. Comparison of Natural a n d Synthetic Product.-The synthetic product was identical t o a sample of natural @-(4hydroxy-2-methy1phenoxy)-lactic acid. A mixed melting point with the natural product, m . p . 168-169'" gave no depression. Both compounds exhibited a n ultraviolet absorption maximum at 290 mp and a minimum at 251 mw and both gave a similar intensity of red color when their solutions were treated with Ehrlich reagent a s described b y Riley. Lf-hen subjected t o paper chromatography no difference could be detected between t h e natural product and the two synthetic isomers. In this study butanol saturated with 1.5 X ammonium hydroxide was used as the solvent. T h e location of the spots on the chromatogram was determined by spraying with brom thymol blue solution, and also by using a modification of Ehrlich reagent. T h e paper was sprayed with a freshly prepared mixture of one part of 0.5% NaN02 and 25 parts of solution containing 5 g. of sulfanilic acid and 50 ml. of hydrochloric acid per liter. Exposure of the sprayed paper to ammonia vapors located the positions of the compounds as well-defined intense red spots. 2-Benzyloxy-5-hydroxytoluene.-To confirm t h e location of the tetrahydropyranyl group in I (and t h e position of the hydroxyl group in I V ) compound I was converted t o the known toluhydroquinone monobenzyl ether (V). 1.04 g. (0.005 mole) of the monopyranylated ether was treated in the usual manner with one equivalent of benzyl chloride in a n aqueous solution of one equivalent of sodium hydroxide. Hydrolysis of the intermediate mixed diether and subsequent workup of the hydrolysis mixture yielded a small quantity of white crystals which on recrystallization from ligroin melted a t 69.570.5'. Baker and Brown* report a melting Doint of 69-70" for this comuound.
To gain insight as to whether the alicyclic double bond or the aromatic nucleus (or both) was being attacked, the rates of peracid consumption by A7desoxycodeine and dihydrodesoxycodeine were compared and found to be identical. From this it may be concluded that the aromatic nucleus was being oxidized a t least as rapidly as the alicyclic double bond, if the latter were being attacked a t all. A6-Desoxycodeine, A8-desoxycodeine and Gmethyl-A6-desoxycodeine were also subjected to perbenzoic acid under the same conditions to ascertain whether the position or degree of substitution of the alicyclic double bond might substantially affect its rate of oxidation. Oxidations were also conducted in benzene a t 8 and 23'. I n every case, irrespective of compound or conditions, the rate of perbenzoic acid consumption was identical with that found for dihydrodesoxycodeine. In a number of examples, a deactivation of the aromatic nucleus to peracid oxidation has been effected by substitution of bromine into the ring.5 This possibility was explored through the preparation of 1-bromodihydrodesoxycodeine, but again its rate of oxidation was identical with that of the unbrominated compound. As a final alternative, oxidation with monoperphthalic acid was examined, since the aromatic ring is relatively unreactive toward this oxidant.2 a Dihydrodesoxycodeine, A6-, A7-, A8- and 6-methylA6-desoxycodeinewere subjected to the action of monoperphthalic acid in acetone a t 0'. In every (11) Riley (ref 5, p 5713) reports m p 185-166'. case, there was a one-mole consumption of peracid \\'ALLACE LABORATORIES requiring about thirty minutes, and beyond that DIVISIONOF CARTER PRODUCTS, IAC. no further oxidation occurred. Although the aliSEW BRUNSWICK,S J . cyclic double bond again was inert, the clean-cut one-mole consumption and no further oxidation a t The Action of Peracids on the Desoxycodeinesl any other site in the molecule indicated that this might be a useful method for preparing amine oxBY HEXRYRAPOPORT AND ETHAN C. GALLOLVA'Y ides. RECEIVED JUSE 21, 1955 When the oxidations with monoperphthalic acid The fact that a number of various desoxycodeines were conducted with the objective of preparing are now easily available led us to examine their re- amine oxides, a good yield of the desired product action with peracids with the obiective of preparing was obtained with facility in each instance, as the corresponding epoxides. These in turn niight shown in Table I. Considering the vigor of the be convertible to codeine isomers with an oxygen alkaline hydrogen peroxide usually used for this purpose,6 and the selectivity of monoperphthalic function a t some position other than 6. For the three possible sites of oxidation, viz., acid in acetone a t Oo, the latter reagent appears to the tertiary amine, the alicyclic double bond, and be advantageous for amine oxide formation. the aromatic ring,2 the expectation was that the Experimental rates of oxidation would be appreciably different A6-Desoxycodeine was prepared by dehydrochlorination and would decrease in the above order.3 Initial of dihydrochlorocodide using a 20-hour reflux period in a experiments with perbenzoic acid and A'-desoxy- solution of sodium cyclohexylate in cyclohexanol instead of codeine in chloroform a t 0" did show an almost in- sodium methylate in methanol a t 140°.? A?- and A*-desoxycodeine were prepared by lithiun~ stantaneous consumption of one mole of peracid. hydride reduction of p-toluenesulfonylcodeine That this was due to amine oxide formation was aluminum a n d p-toluenesulfonylneopinc, respectivcly.8 Hydrogenaestablished by hydrogenation of the product to tion of A7-desoxycodeine gave dihydrodesoxycodeine.* dihydrodesoxycodeine. However, after this very (4) This oxidation of the aromatic ring most probably occurs a t rapid one-mole uptake (less than 15 minutes) con- Cs-Cc, since this is t h e position most susceptible t o ozonolysis [E. sumption of perbenzoic acid continued a t a mark- Speyer and A. Popp, Ber., 69, 390 (1926); E. Speyer, ibid.. 62, 200 edly decreased rate and did not reach two moles (1928); H.Rapoport and G. B. Payne, J . O r g . Chem., 15, 1093 (1950)I even when a n alicyclic double bond is present [E. Speyer and L. F. until after about forty hours. (1) Supported b y a grant from t h e National Institutes of Health, Bethesda, M d . (2) (a) H. Fernholz, Bey., 84, 110 (1951); (b) S. L. Friess, A. H. Soloway, B. K. Morse and W. C. Ingersoll, THISJ O U R N A L , 74, 1305
(1952). (3) U. Swrrn, Chein. Reus., 46, 1 (19.19).
Roell, Ber., 68,539 (1930)l. (5) J. Bdeseken and C. F. Metz, Rec. Irou. chiin., 64, 345 (193;); W. F. Short and H. Wang, J . Chem. Soc., 2979 (1951). ( 8 ) K . W. Bentley, "The Chemistry of the Morphine Alkaluids," Clarendon Press, Oxford, England, 1954. (7) L. F. Small and F . L. Cohen, THISJ O U R N A L , 63, 2214 (1931). (8) H.Rapoport and R . M. Bonner, iW., 73, 2872 (1951).
Vol. i i
3751
-&MINE
OXIDES
TABLEI SOME DESOXYCODEINES
OF
Yield, Compound
(
. T h e gcaneral proccdurc for- preparing the amine oxides cori- phoric acid gave improved yields. IVhen S5(% orsisted in treating the codeine derivative with approximatel>- thophosphoric acid was used i t was necessary to 3110 mole \ ( of ~nonoperplithalicacid as a 0.1 :lf solution in eliminate water by introduction of a calculated acetone aiid allowing t h e solution to stand a t 0" for 30 amount of acetic anhydride to the reaction mixture miiiutes. Excess saturatetl sodium bicarbonatc solution \\;is then added, the 11 irv was filtered, a n d thc filtrate prior to addition of the amine and olefin. Exccss TI as evaporated t o dr>-n acetic acid acted as solvent in the reaction which tween chloroform and 1 was carried out using N, N ',N"-trimethyltri methylof chloroform. Drying atid evaporating the chloroform enetriamine and a-methylstyrene on an equimolar left a residue \rliicli was either crystallized from benzene or chromatographed on alumina ( l l e r c k ) . I n the lattcr case, basis. I ~ n z e n ewas u x t l for dcvelopinetit and 5( isopropyl alcohol TABLE I i n benzene for elution. Evaporation of tlie combined isoPREPARATIOV O F 1-hlETHYL-4-PHENVL-I-ACETOXVPIPeRI1)ropyl alcohol-heiizene fractions g;i\.c DISE crh-stalline it1 some instances or were c Alole catalyst per Time, Tern),., Yiclil. zenc in others. mole a-methylstyrene h r . O C . i ;Imine oside picrate5 were prepared in a n d were recrystal ized from methanol. 4 llc5 17.2 Xone -1 115 *'HI. 1 1 . 0 &PO? 1 ) E P A R T M E X T O F CHEMISTRY A S D CHEMICAL EKGINEERITG I-YIVERSITY OF CALIFORNIA 5 115 ? ! I fi 0.8H2SO4 BERKELEY 4, CALIFOKSI.4
-
, IOi
c,.
I % r a u l l ,' ' o ~ ~ a l l i cSyntlx ! rib
Y m k , S.T , 1011,
Coll. \ - < , I . 11.
431
( 1 I ) 11. Hiihrnr. (11: , Y \ n , 20, 70 ( l ! l $ O )
I , Jclhn U'iley a n d
(1) R . H. K. Foster and A . J. C a r m a n , J . P l i o r i ~ a c o l E x f i f i . 7 ' h ~ + ' n f i . , 91, 195 ( 1 9 4 7 ) . ( 2 ) C. J . Schrnidlr anrl I
