JULY
I961
COMMUSICATIONB
2615
2-naphtho1,g m.p. 93-94'. The reation in ethylene was ring closed to 2Hj3H-thieno[3,2-b]pyrrole-3glycol gave the 91/10 seco compound (VI) as a sol- one (111) and desulfurized to authentic 2-acet INujol pyrrole (ITT) 815 em.-' vated crystal, m.p. 100-108', A,, KMR7 T, 7.69, 7.75 ppm. (two benzenoid methyl), ~ S C ; N ~S-(:H,COOH 9.13 ppm. triplet (methyl of ethyl group). The N substance after the crystal solvent is removed is H H ax 288 mp an oil, [a]: -25' (chloroform), ) mCHiOH ( E 3,000), (Anal. Calcd. for C15H2003: C, 72.55; H, 8.12. Found: C, 72.26; H, 8.12), acetate, m.p. polyphosphorict COCH3 145-150' (Anal. Calcd. for C17H2204:C, 70.32; acid H, 7.64. Found : C, 70.58; H, 7.61). H H
Ranez
In connection with our work on the S M R spectra of heteroaromatic c o m p ~ u n d s , ~ me - ~ have studied the S-MR spectrum of thiocyanopj rrole and some of its derivatives. The three bands in the aromatic region of thiocyanopyrrole display the shifts6 T, = 3.10, rB = 3.47, T , = 3.85 p.p.m. and Methane generated from all the reactions in- the coupling constants J,, = 1.5, J,, = 2.9, and volving the loss of the angular methyl group, was J p n = 3.6 c/s, which prove that the compound detected by gas chromatography. As only starting formed is the 2- isomer. This conclusion is based on material was recovered from the reaction in an- comparison of the above NMR parameters with hydrous pyridine, ethylene glycol or water in pyri- those observed in other pyrroles5 and in 2- and 3dine may be a hydrogen donor in the reaction. The thiocyanothiophenes. Furthermore, the S M R treatment of the above dienone (Ie) or trienone spectrum of the methylthiopyrrole obtained (Ia or Id) in acetic acid with zincQdoes not give through the reaction of the thiocyanopyrrole with an aromatic A-ring steroid, but a substance as- alkali and methyl iodide' is in agreement only sumed to be a bis compound showing polyene ab- with that expect,ed for the 2- isomer ( T , = 3.28, sorption. The detailed presentation of these re- T B = 3.77, rC = 3.90 p.p.m., J,, = 1.5, J,, = actions will be published in a forthcoming report. 2.8, and J B , = 3.4 c/s). The same methylthiopyrrole (b.p. 87-90°/17 mm., n? 1.5730. Anal. Calcd. INSTITUTE O F ilPPLIED MICROBIOLOGY K. TSCDA for C5H7XS:C, 53.06; H, 6.23; S,12.38; S, 28.33: THEUNIVERSITYOF TOKYO E. OHKI Found: C, 53.17; H, 6.42; h-, 12.35; S, 28.23) is obHOKGO, BUNKYO-KU, S. NOZOE TOKYO, JAPAN ?;. IKEKAWA tained by treating pyrrolemagnesium iodide with dimethyl disulfide and also by treating pyrrole with Received April 3, 1961 methylsulfenyl chloride. These results provide inde(8) T. Kariyone, T. Fukui, and T. Omoto, Yakugakuzassi pendent evidence that the methylthiopyrrole is (Tokyo), 78,710 (1958). (9) The reaction of a B-ring dienone to an aromatic com- the 2-isomer, as it is known that these types of repound and that of santonin to the bis compound with zinc actions lead to a-s~bstitution.7~~ in acetic acid has been reported. D. H. R. Barton and E. R. The XMR spectrum of the aldehyde obtained Thomas, J . Chem. Soc., 1842 (1953). K. Tsuda, E. Ohki, through Vilsmeier formylation of the methylthioand J. Suzuki, Chem. and Pharm. Bull. 9, 131 (1961). J. Simonsen, The Terpenes, Vol. 111, p. 273, Cambridge L'niv. pyrrole shows that the compound formed is 2Press (1952).
(2) S. Gronowitz, and R. A. Hoffman, Arkiv Kemi, 16, 539 (1960). (3) R. A. Hoffman and S. Gronov.-itz, Arkiv Kemi. 16. 563 (1960). (4) S. Gronowitz and R. A. Hoffman, Arkiv Kemi, 16, 459 (1960). (5) S. Gronou-itz, A B . Hornfeldt, E. Gestblom, and R. A. Hoffman, Arkiv Kemi, in press. (6) The NMR spectra were obtained with a Varian Associates Model V-4300b spectrometer operating at 40 Mc/s. The chemical shifts [T-values, cf. G. V. D. Tiers, J . Phys. Chem., 62, 1151 (1958)l were obtained from dioxane solutions to which mere added traces of piperidine in order to eliminate the complicating effects of the couplings with the N-hydrogens. (7) M. S. Kharasch and 0. Reinmuth, Grignard Reactzon of A'onnzetallzc Substances, Prentice-Hall, Inc., Y e n . York, S . Y., 1954 pp. 75 ff. (8) A. H. Corwin in R. C. Elderfield, ed., Heterocyclic Compounds, Vol. I, John Wiley & Sons, Inc., Sew York, iY. Y., 1950, Chapter VI. I
On the Thiocyanation of Pyrrole Sir:
Matteson and Snyder1have recently claimed that thiocyanation of pyrrole with methanolic thiocyanogen a t -70" or with cupric thiocyanate a t 0' yields 3-thiocyanopyrrole (I) (m.p. 41.5-43 "). This is rather unexpected since most reagents attack pyrroles a t an unsubstituted a-position in preference to an unsubstituted /%position. They proved the structure of the thiocyaiiopyrrole by converting it to the (pyrroly1thio)acetic acid (11) which (1) D. S. Matteson and H. R. Snyder, J . Org. Chem., 22, 1500 (1957).
,
mcthylthio-.i-pSrrolealdehydc (m p 105-106O, ‘T(f10 = 0.69, 71 = 3.78, T4 3.12, T S C f I = 7.59 p p rn., .I7$ - 3 8 c / s ?no1 Cnlcd for CEW;?\‘OSC, ;il 0 4 ; TI, 1- X3; N.9 92; S,22.i1 : I’ound C, 41 ; TI, 3 13, h., 10.08; S,22.36), thus giving further ~r;dc!irc~ that 1he original rncthylthiopyrro!c is the 2- imvcr. Additioual evidence agaiiist preferential ,?-thioc.yaiiation in pyrrolcs is obtained from the flirt that 2-mcthylpj-rrole yields 5-thiocyano-2nicthplpyrrolc (m p G3 5-66’, r3 = 4.13, r4 = 3.36 p p.rn , J C H3 ~= 0 80, J c ‘ I 4~ ~= 0.33. = 3.55 c 9 Anal Cnlcd. for CJIJ2S: C, 32.13; $1, 4 38; S ,20.28; S, 23 19: Found: C, 52.22; H, 4 3 0 ; K, 20 21; S, 22 86.) upon thiocyanatiori with cupric thio-ynnatc, the XMR eT,idence for its structure bciiig based on the valucs of the chemical shifts and ring-coupling constants, iii addition t o the sidechain coiipliiigs Since thc qtructure of 111 is proved beyond any doubt by Snyder et al and since it is very improbable that any rearrangement occurred in the transformation of the thiocyanopyrrole to methylthiop>rrole or (prrro1ylthio)acetic acid, the discrepancy betwecii oiir rcsultq and those of Matteson and Snyder rcgarding the structure of the thiocyanopyrrole must hc ascribed to a rearrangement during the cychaat ion of (2-pyrroly1thio)acetic acid with pol>7p1.insphoricacid to 111. Rearrangements during treatment with poljyhosphoric acid, although not analogous to lhat found here, have been observed b,v others A detailed account of the observations reported herc n 4 l bc published in Brlziv Kemi. l o p ’
to have a ketone carbonyl group formed with elimination of one mole of water, and to absorb one mole of hydrogen without reduction of the carbonyl group.5 AIoreover, it has been shown that this dehydration takes place between two tertiary hydroxyl groups of compound I.5 On the other hand, on the basis of the biogenetical viewpoiii t as well as experimental results, Marion and his co-morkers6-8 have pointed out that delcosiiiel (lucaconiiie) (I) probably possesses the same carbon-nitrogen nucleus as lycoctonine, and also that this base is represented by the structure I shown below. In the belief that their conclusions
I
OCHB OCHS 11. R=Hz, R ’ = R ” = A c 111. R = O , R ‘ = R ” = H
are quite reasonable, the present authors nom mould like to propose structures I1 aiid I11 for anhydrodia~etyllucaconine~ and anliydrooxolucaconine (111),4respectively. Compound I11 has previously been obtained from both compound I1 and oxolucneonine through two steps.4The mechanism of the above dehydration is considered to be analogous t o that of the dehydration of oxolycoctonine or
demethyleneoxodelpheline.g
The ultraviolet absorption spectrum of compound 111 in methanol shows a maximum a t 301 ilcknoi~ledgment~Financial support from the mp (log t: 2.01) while compound 11 manifests a Swedish Nal ural Science Research Council and maximum a t 237 mp (log E 3.20). The limiting from the Swedish Technical Rcsearch Council is structure IIa of compound I1 seems to give a good explanation of the marked difference between these acknowlcdgcd. two absorption bands. h similar phenomenon I K S T I l I TTz OF (’(HEMISTRY SALOGROAOV ITZ was observed and interpreted ill the case of some 1IiYIiA-BRITT.4 HORUFCLIIT C V I T E R H r Y O F UPPS4TA Bo G E ~ T B L O Mdelphinine and rieoliiie derivatives. l o ITPPSAI4, SME U L K RAGUARA IIOFFMAX (1) It has been shown that lucaconine is identical with Received April 27, 1961. (9)H.R Sngder, L. A. Carpino, J. F. Zack, Jr.. and J. I‘ LIilL, J . h n .Chem. Soc., 79, 25156 (1957). (10) J. E. htnfield, ST’. Davies, N. IT., Gamble, and S. Rliddleton, J . Cham. Soc 4791 (1956). (11) ST. I):~vi(’sa t i d ( 1957 ).
S
Middleton, Chenz. & I n d , 599
delcosine,2 and therefore anhydrodiacetyllt,caconine is identical n i t h diacetyldelcosine (m.p. 159-161”) obtained by Marion et nL.8 The name “lucaconine” should br revised to “delcosine.” (2) T. -4miya and T. Shima, in preparation. (3) W.I. Taylor, W.E. Walles, and L. Marion, Can. J . Chem., 32,780 (1954). (4) S. Furusawa, Bull. Chem. SOC.Japan, 32, 399 (1059) f5) T. Amiva and T. Shima. Bulr Chern. Soc Javan. 31. 1083(1958). “ (6) V. Skaric and L. Marion, J . Am. Chem. Soc., 80, 4434 (1958): V. Skaric and L. Marion, Can. J . Chem., 38, 2433 (1960). (7) R. A n e t , D. W. Clayton, and 1,. Marion, Can J . Chem., 35,397 (1957). (8) R. Anet and 1,. Marion, Can. J . Chenz., 36, 766 (1958). (9) E. S.Stern, The Alkalozds, Chemistry and Physioloqv, Vol. VXI, R. H. F. Manske, ed., Academic Press In?., New York, 1960, p. 473. (10) K. Wiesner, H. W. Brexver, D. L. Simons, 11. R. Babin, F. Bickelhaupt, J. Kallos, and T. Rogri, Tetrahedron Letten, S o . 3, 17 (1960). .
On Anhy drodiacetyllucaconine (Diacetyldelcosine,’ M.P. 159-161’) and Its Derivatives
Sir : I’reriously, it was shown that ai1 aconite nlkagave anhydrodiloid, lucaconine (I) (Cn4H3,07S), acetyllucaconine (11) (C28H4,0sS)on treatment with acetyl chloride.4Compound 11has been found
,
I
