dissolved in a minimum volume of coricd. ammonia and analyzed for cyanide. "Pyrrole black" was prepared b>-refluxing for 4 hr. 50 ml. Of a solution which contailled 0.43 g. of silver nitrate and 1 ml. of pyrrole. The heavy black precipitate (0.36 g.) that began t o appear after a few minutes \vas found t o be free of silver cyanide and silver acetylide; it reseiribled the mate-
rials obtained by earlier authors through t h e action of certain oxidizing agents and/or metal ions on pyrr01e.l~ _____ (13) Cf, E . & . , the chapter on "Pyrrolschwarz\~erbindungen" in H Fischer and A Stern. "Die Chemie des Pyrrols," Yo1 11, Akademische T'erlajisb.esellschaf1, I.eipzig, 1040, p p 4 3 7 ~ 4LO.
PXSADESA, CALIFORSIA
[FROM THE L)EPARTXENT O F ORGAXIC ~ F I E Y I S T R S ,HEDREW L'NIVERSIl'Y!
A Synthesis of 2,3,6,7-Dibenzophenanthrene B Y ERNSTD. BERGMASN.IND RAPKIEL1x.n RECEIVED JULS 15, 1957 Starting from d-metli?-l-3-decalorie, d,3,(j,i-dibenzophenanthrene has been s y n t h e h x l . synthesis of higher benzologs and of the analogous fluorene derivatives.
Among the dibeiizophenanthrenes and -fluerenes, the 2,3,6,7-isomers are probably the least accessible.' 2,3,6,7-Dibenzophenanthrene(I) and 2,3;6,7-dibenzofluorene (11)are interrelated through 3,3'-dicarboxy-2~2/-binaphthyl (111, R = COOH) which can be cyclized via its lead salt t o give 2,3,6,7-dibenzofluorenone.Z On the other hand, it should be possible in analogy with previous experience34 to obtain I by reduction of I11 (I< = COOH) t o the dihydroxymethyl compound I11 (R = CH,OH) from which the dibromo derivative I11 (R = CH,Br) would be accessible by con\entional means. Treatment of the latter with lithium phenyl, followed b y dehydrogenation, should gire I.
I
111
--
11
T h e iriethotl is applicable t o t h c
Ahalternative method has, therefore, been de-
veloped, which gives a lower over-all yield (lOyC) but has the advantage t h a t the starting material is easily accessible. -1s the above method has been described recently (after the completion of our experiments) by Badger and co-workersj6we refrain from a detailed discussion of this first part of our investigation. 2-Xethyl-3-decalone (T'), available from cyclohexanone and methyl isopropenyl ketone with subsequent hydrogenation (over-all yield 82%) ,i was converted into the pinacol VI and the latter dehydrated (to YII) and dehydrogenated t o 3,3'dimethyl-2,2'-binaphthyl(111, R = CH3). Upon treatment with KBS, this gave the same dibromo compound I11 (R = CH,Br) obtained by the previous route. By oxidation of 111>R = CH8,to 111,R = COOH, the two alternative routes have been further linked to each other. A s analogs and homologs of V are easily available by similar Xichael condensations,' this method appears applicable to the synthesis of even more highly condensed polycyclic systems.
IT.
\.\-e have carried out this synthesis in an over-all yield of 21c/'c, starting from 2-amino-3-naphthoic acid (IV) which was converted to I11 (R = COOH). However, this method suffers from the limitation that the preparation of TV is very c ~ m b e r s o n i e . ~ (1) E. Clar and I'. John, Bei., 6 4 , rj8l (1931); 62, 3021 (1.JZii); IC. Clar, F.John a n d B. Hawran, ibid., 62, 940 ( l ' J 2 9 ) ; E, Clar and 1).C . Stewart, J . C h e w . Soc., 3215 (1951); R. H . Martin, Helu. C h i m Acte, 30, 620 (1947); R. H . Martin a n d S. Vassart. Bzdl. SOL < / i i i i r . Belg., 61,234 (1952). (2) R. H. Martin, J. Chem. So:., 67Y (1041). (3) See, also for previous literature, E. D. Bergmann and Z. Pelchowicz, J . Org. Chent., 19, 1383, 1387 (1954); G. Wittig and H. Zimmerman, Ber., 86, G29 (1053); D. hl. Hall, E. E. Turner a n d I(.E. H a m lett, J . C h e w Soc., 1242 (1955). (4) Interesting variations of this method are t h e preparation, from 2,?'-dibromomethylbiphenyl, of . Dhenanthrene with sodamide in liuuid . . . ammonia (I. A. Kaye, THIS J O U R N A L , 73, 5467 (1951)) and t h a t of 9,lO-dihydrophenanthrene with sodium in t h e presence of catalytic quantities of tetraphenylethylene (E. Mueller a n d G. Roescheisen, Ber.. 90. 643 11957)). T h e original'kethod of Xoehlau ( B e r . . 28, 3096 (1896)) could not be duplicated by H. A. Harrison and F. A. Royle (J.Chon. Sot.., 81 (1926)). T h e present methods are based on the prolonged reaction c > F 2-hydroxy-3.naphthoi~ acid with ammonia and zinc chloride or fcrrous sulfate a t high temperature and pressure; Or$. Syizlhescs. 2 2 , I!) (lU.42); German Patent 488,91G ("Beilstein," 2nd Suppl,, Vol. 14. p :;23).
(i)
\ d \'I11
studying the spcctvu of the compounds prepared in this investigation) Some interesting Observations have been made. The diene VI1 shows practically no selecti\re absorption, most probably due to steric hindrance. The Same effectwill ex(f;) (>. hI. 13adgcr. 1' 1t JelTcries a n d I-ellow needles of m.p. 140’. 3,3’-Dibromomethyl-2,2’-binaphthyl(III, R = CH?Br).( a ) A solution of 5 g. of I11 ( R = C H 2 0 H ) in 600 ml. of benzene was prepared a t 50’; after addition of 2 drops of pyridine, 10 g. of phosphorus tribromide was added with stirring and the mixture kept a t 50” for 2 111. Cold water was added and the benzene layer mashed with sodiuni bicarbonate solution and water, dried and concentrated; frotii benzene-petroleum ether colorless needles of m . p . 172’ Experimental (lit.G 165-166”), yield 6.1 g. (8756); ultraviolet spectrum 3,3‘-Dihydroxymethyl-2,2’-binaphthyl(III, R = CH20H), (in ethanol): 238 mp (5.01); inflection a t 300 m p (3.76); prepared from 3,3’-dicarbomethoxq.-2,2’-binaphthyl (111, 331 mp (2.98). R = COOCHI) and lithium aluminum hydride in 96% Anal. Calcd. for CzYHIGBr2: C, 60.0; H, 3.7. Found: yield, crystallized from isopropyl alcohol in colorless needles C, 60.3; H, 4.0. of m.p. 230’ (literatureG228-229’); ultraviolet spectrum ( b ) The solution of 2.8 g. of 3,3’-dimethyl-2,2’-dinapli(in alcohol): 231 mp (log e 5.16); inflections at 280 mp thy1 (111, R = CHI) in 100 ml. of carbon tetrachloride was (4.19) and 293 mp (3.76); 320 m p (2.75). refluxed for 6 hr. with 3.6 g. of NBS and a trace of benzoyl Pinacol (VI) of 3-Methyl-2-decalone (V).-The mixture peroxide. The product was filtered and the solution conof 198 g. of 3-methyl-2-decalone ( V ) , 7 21 g. of aluminum centrated; from benzene-petroleum ether, m.p. 1 7 2 O , turnings, 8.5 g. of mercuric chloride and 180 ml. of dry yield 3.5 g. (7570). benzene was refluxed for 1 hr. with stirring. -it room 9,10-Dihydro-2,3,6,7-dibenzophenanthrene (VIII).-The temperature 60 ml. of water and 180 ml. of benzene were reaction between lithium phenyl and the dibromo compound added gradually, and the content of the flask was refluxed I11 ( R = CHpBr) was carried out in the same manner as for 1 hr. more. The product was filtered while still hot, described by Badger, et a1.6 VI11 was obtained in 3 i and the solid phase washed with 200 nil. of hot benzene. from benzene, m.p. 137-138’ (lit.6 140-141’); ultr Evaporation of the benzene in vacuo gave 130 g. (65C;,) of a spectrum (inalcohol): 231 m p (4.63); 264 (4.68); 276 (4.48); viscous oil, which was used directly for the next step. 279 (4.49); 319 (3.82). 3,3’-Dimethy1-2,2’-di-3,4,5,6,7,8,9,10-octahydronaphthy1 Anal. Calcd. for CazHis: C, 94.7; H, 5.3. Found: (VII).-The mixture of 100 g. of \-Iand 20 g. of dehydrated C, 94.3; H, 6.0. alum was heated a t 160-170” for 4 hr. and extracted with 2,3,6,7-Dibenzophenanthrene(I),-In an atmosphere of hot benzene. The solution was dried and distilled, b.p. nitrogen, 0.5 g. of VI11 and 0.5 g. of palladium-charcoal 170-175O (0.2 m n . ) , yield 60 g. ( 6 i 5 ~ ; ) ; ultraviolet spec(lOyc)were heated a t 350” for 4 hr. T h e product was extrum (in alcohol): inflection a t 250 m p (log e 3.16). Anal. Calcd. for Ce2HZ4:C, 88.5; H, 11.5. .Found: C, tracted with hot benzene and the residue of the solution recrystallized from the same solvent; yellow needles of m.p. 87.8; H, 11.3. 2 5 7 O , vield 0.45 g. (9OC,c); ultraviolet spectrum (in alcohol): 3,3’-DimethyL2,2’-binaphthyl (111, R = CHB).-In a n 260 mp (4.83); 305 (4.70); 318 (4.90); 333 (4.57); 343 atmosphere of nitrogen, 10 g. of VI1 was heated at 300-330” (4.34);- 349 (4.42); 360 (4.28); 380 (2.70); 392 (2.57); with 7 g. of loc,%palladium-charcoal for 12 hr. The product 403 (2.r8); 415 (2.45); 427 (2.76). was extracted with a mixture of benzene and ether and the Anal. Calcd. for CaYHia: C, 95.0; H, 5.0. Found: residue of the solution triturated with petroleum ether and recrystallized from butyl alcohol; prisms of m.p. 144O, C, 94.8; H, 5.4. T h e dipicrate, prepared in benzene solution and recrysyield 6.5 g. (68Yc);ultraviolet spectrum (in alcohol): 233 tallized from benzene, formed orange-red needles of m.p. m p (loge 5.10); 290 m p (4.13); 321 m p (2.98). 184O, as indicated in the l i t e r a t ~ r e . ~
plain the absence of much fine structure in the compounds 111, as compared to 2,2’-binaphthyls: All of them show an intense absorption a t about 240 mp (log E 5.00) and after some inflections a much lower second maximum (log E 2.78 to 3.00) a t approximately 320 mp. Compound I has a very pronounced fine structure; its spectrum consists of two absorption regions : six high-intensity bands between 290 and 350 m p and five low-intensity bands between 390 and 430 mp.
(8) R. A . Friedel and hl. Orchin, “Ultraviolet Spectra of Aromatic Compounds,” S o . 309, John Wiley and Sons, New York, S . Y., 1951. T h e main absorption bands of 2,2’-binaphthyl are 254 mfi (log L 4.95) and 305 mfi (4.25).
[ COSTRIBUTIOX FROM
THE
(9) E. Clar and F. John, ref. 1
JERCSALEM, ISRAEL
COSVERSE LIEMORIAL LABORSTORY OF HARVARD UNIVERSITY]
cis-Hydroxylation of a Synthetic Steroid Intermediate with Iodine, Silver Acetate and Wet Acetic Acid BY R. B. WOODWARD AND F. V. BRUTCHER, JR.] RECEIVED AUGUST23, 1957 Iodine, silver acetate and wet acetic acid oxidize the 6,7-double bond of d2-anti-trans-4,4a,1b,5,8,8a-hexahydro-l,8adirnethyl-2(3H)phenanthrone ( I ) t o give, after hydrolysis, a 0-cis-glycol IVc in 71 % yield and a n a-cis-glycol in 2.5% yield. This reagent offers a method for producing in quantity the opposite cis-glycol stereoisomer compared with that derived from osmium tetroxide hydroxylation.
During recent steroid total synthesis work,2 os- nique which has the unique feature of producing the mium tetroxide was used with success in cis-hy- opposite cis-glycol stereoisomer compared with that derived from osmium tetroxide. This paper dedroxylating dl-anti-trans-4,4a,4b,5,8,8a-hexahydro1,8a-dimethyl-2(3H)phenanthrone(I). Since this scribes our reagent and discusses its mechanism of reagent possesses certain undesirable characteristics, hydroxylation. we later developed a new cis-hydroxylation techWe have found that when I is treated in acetic acid with iodine (1.05 equivalents), silver acetate (1) Dept of Chemistry, University of Pennsylvania, Phila. 4, Pa. (2.25 equivalents) and water (1.0 equivalent) and (2) R. B. Woodward, F. Sondheimer, D. T a u b , K. Heusler and finally heated at 90-95’ for three hours, oxidation W. XI. I f c l a m o r e , THISJ O U R N A L , 7 4 , 4223 (1952).
