COMMUNICATIONS TO TIIE EDITOR
2398
conjecture.' In particular it is not known how many types of non-classical ions might occur, i.e., honioallylic, bi- and tricyclobutonium, etc., nor what role classical ions might play in their reactions. \Ye wish to report results which demonstrate, in a striking manner, the absence of classical cations in some homoallylic solvolyses, and which also place some important restrictions on the type of bonding involved in these cations. \Ye have reported recently the synthesis of endo7-isopropylidenedehydronorborneol (Ia) . Basecatalyzed equilibration of Ia and gas chromatographic separation of the resultant mixture of dienols gave the exo-alcohol (IIa). Acetolysis of both endo- and exo-tosylates (Ib and IIb) proceeded readily a t :XIo (krb = 7 X IO+ set.-', k r r b = G X sec.-l) showing that both reactions are strongly anchimerically a ~ c e l e r a t e d . ~Despite ,~ the fg1-t that both solvolysr~s led t o positive charge u t
Vol. s2
sible that equilibration of the cations may become detectable in solvents in which they have longer lifetimes, and that cations from different sources (e.g., from the amine-nitrous acid reaction) may react differently. Experiments designed to test these points are in progress. DEPARTMENT OF CHEMISTRY C. 1% DEPUY IOWASTATEUXIVERSITY I. A. OCAWA J. C. MCDANIELS AYES, IOWA RECEIVED FEBRUARP 27, 1960
THE CORRELATION OF THE ATISINE AND VEATCHINE SERIES OF DITERPENE ALKALOIDS
Sir: The presently accepted structures? of atisine (I)3* [Aconitum heterophyllum] and veatchine (II)3J [Garrya veatchii] differ only in the terminus of the hydroxyl-bearing ring. Because of the failure to interrelate the two series, the possibility has been suggested recently that the allylic alcohol group in atisine may be on the cis- (rather than the trans-) branch (with respect to nitrogen) of the 2,2,2-bicyclooctane system.8 As subsequent evidence from this6 and another laboratoryh was not Ia R = F I Ira R = H compatible with this proposal, the direct correla11 I i O T W I3 13- OTOS tion of these two important classes of diterpene c R 0.k alkaloids became imperative. This paper reports carbon 2 , the reactions are completely stereospecific, the successful correlation of atisine and veatchine endo-tosylate (Ib) leading exclusively to endo- via the degradation product 111 and provides the acetate (IC),6 exo-tosylate giving only rearranged first unequivocal evidence for the common stereoacetate (111) as product. Clearly, the cationic chemistry of their carbon skeletons. center a t carbon 2 does not have classical character, since otherwise some mixing of products would have been observed. Llore importantly, equilibrium between the two hornoallylic ions (IV and L7) is also ruled out. Similar results were obtained 7
IY
Vr
in the solvolysis of the 5,6 dihydro compounds, endo-7-isopropylidene-norbornyl tosyla tes, where the same stereospecificity in solvolysis was observed, the endo tosylate ( k = 4 X sec.-l) giving only endo acetate and the ex0 ( k = 2 X sec.-l) giving only dihydro 111. The absence of common intermediates in the two solvolyses demands an activation energy for the interconversion of the homoallylic ions, and rules out a symmetrical structure in which both sides of the unsaturated center a t Cz are stabilized by interaction with the double bonds. It seems posexo- and
(1) For recent, comprehensive discussions of the problem see S. Winstein and E. M . Kosower, THISJOURNAL, 81, 4399 (1959), and R. H. Mazur, W. N. White, D. A. Semenow, C. C. Lee, M. S. Silver and J. D. Roberts, ibid., 81, 4390 (1959). (2) C. H. DePuy and P. R. Story, ibid., 88, 627 (1960). (3) S. Winstein, C. R. Lindegren, H . Marshall and L. L. lngraham, ibid., 7 6 , 147 (1953). (4) lnteraction between the 2 and 7 position of the type displayed in 1 V has been sought previously without success. See J. C. Martin and P. D. Bartlett, i b i d . , 79, 2533 (1957), and E. E . van Tamelen and C. I. Judd, ibid., 80, 6305 (19581. ( 5 ) A s little as 1 % of 111 could have been detected readily by gas phase chromatography.
Treatment of IVI9available from atisine hydrochloride in six steps,lo with sodium metaperiodate (1) This investigation was supported in part by Grants R G 5807 and RG 5807 (Cl) from the National lnstitutes of Health, United States Public Health Service. (2) The indicated structures assume the Irans-anti fusion of rings A and B which is common to most diterpenes. (3) K. Wiesner, et al , Chemistry a n d I n d u s f r y . 132, (1954); E x p e r i entia 11, 255 (1955). (4) S. W. Pelletier and W. A. Jacobs, TEISJOURNAL, 7 6 , 4496 (1954). ( 5 ) D. Dvornik and 0. E. Edwards, Chemistry a?td Industry, 633 (1958). (6) S. W. Pelletier, ibid., I l l 6 (1958). (7) K. Wiesner, et a!., THISJOURNAL, 76,6068 (1954). (8) K.Wiesner and 2. Valenta, Progress i n the Chemistry of Organic Natural Products, X V I , 52 (1958). (9) S. W. Pelletier and W. A. Jacobs, THISJOURNAL, 78, 4144 (1956). (IO) The intermediate azomethine acetate was prepared by the method of Dvornik and Edwards, C u n o d J . C h e ~ n, 35, 860 (1957).
hlay 5 , 1960
COMMUNICATIONS TO THE EDITOR
and potassium permanganatell gave directly the dicarboxylic acid V,l2 m.p. 285-289"; [a]"D +11.0"; vNujol (COaH) 1727, 1692 cm.-I; (N-Ac) 1610 crn.-l. Conversion of V to the dimethyl ester (VI), m.p. 196-198.5'; [a]2 9 ~ + 1 3 " ;vNujol (COzhle) 1724 cm.-'; (N-Ac) 1639 cm.-' and selective saponification afforded the monoesteracid, VII, m.p. 248-250"; [a]2 9 ~ + 1 5 0 vNujol ; (CO2H,COzh1e) 1712 cm.-l; (N-A4c) 1603 cm.-l. Rearrangement of the silver salt (VIII) with bro-
IV
2399
CH3
XI.R =4 c XII.R=H
. ---VI, R' = R" = CHa VII, R' = H ; R" = CHr VIII, R' = Ag; R" = CHj I
mine in carbon tetrachloride gave I X as an amorphous resin; vfiIm (C02hfe) 1730 crn.-'; (N-Ac) 1647 an.-'. Reductive debromination of I X with zinc in acetic acid and extensive chromatography, gave 111, m.p. 156-157"; [c~]~~D-18.9".
'
relation interrelates the three series of biogenetically important diterpene alkaloids, atisine, veatchine and garryfoline.
A
co
CH1
THEROCKEFELLER INSTITUTE N E W YORK21, NEWYORK S. W. PELLETIER RECEIVED MARCH 2, 1960
IX
THE TRANSITION OF CESIUM CHLORIDE
Sir:
Application of a parallel series of reactions to veatchine gave these compounds : veatchine diacetate chloride (X), m.p. 254-257'; vNUjo1 (0,4c)
+
1739 crn.-l, 1238 cm.-'; (>C=N
