Studies Directed Toward the Total Synthesis of Germanicol Citation Baldwin, Steven Worth (1969) Studies Directed Toward the Total Synthesis of Germanicol. Dissertation (Ph.D.), California Institute of Technology. doi:10.7907/R8K1-QY22. https://resolver.caltech.edu/CaltechTHESIS:01252016-111724271 Abstract An approach to the synthesis of the pentacyclic triterpene germanicol is discussed. Item Type: Thesis (Dissertation (Ph.D.)) Subject Keywords: (Chemistry) Degree Grantor: California Institute of Technology Division: Chemistry and Chemical Engineering Major Option: Chemistry Thesis Availability: Public (worldwide access) Research Advisor(s): Ireland, Robert E. Thesis Committee: Unknown, Unknown Defense Date: 26 September 1968 Funders: Funding Agency Grant Number NIH UNSPECIFIED Record Number: CaltechTHESIS:01252016-111724271 Persistent URL: https://resolver.caltech.edu/CaltechTHESIS:01252016-111724271 DOI: 10.7907/R8K1-QY22 Default Usage Policy: No commercial reproduction, distribution, display or performance rights in this work are provided. ID Code: 9542 Collection: CaltechTHESIS Deposited By: INVALID USER Deposited On: 25 Jan 2016 23:48 Last Modified: 25 Apr 2024 20:30 Thesis Files Preview PDF - Final Version See Usage Policy. 23MB Repository Staff Only: item control page

STUDIES DIRECTED TOW ARD THE TOT AL SYNTHESIS OF GERMANICOL Thesis by Steven Worth Baldwin Jn Partial Fulfillment of the Requirements For the Degree of Doctor of Philosophy California Institute of Technology Pasadena, California 1969 (Submitte d September 26, 1968) To Rita ii ACKNOWLEDGEMENTS I wish to express my appreciation to Professor Robert E. Ireland for his guidance during the course of this study. Thanks are also due to my colleagues for many exciting and fruitful discussions. I would like to thank the National Institutes of Health for financial support in the form of a trainee ship during the period of this research. iii ABSTRACT An approach to the syn.thesis of the pentacyclic trite rpene germanicol is discussed. iv TABLE OF CONTENTS Page i Title Page Acknowledgments ii Abstract iii Table of Contents iv Historical Introduction I Discussion 8 Experimental Section 51 Ethyl vinyl ketone 54 5, 8a13 -Dimethyl-3, 4, 8, 8a-tetrahydro-l, 6(2H, 7H)naphthalenedione i£;l.l 54 5, 8a13 .:.nimethyl-1, l-ethylenedioxy-1, 2, 3, 4, 8, 8ahexahydro-6(7H)-napththalenone (VIII) 55 I, l-Ethylenedioxy-1, 2, 3, 4b, 5, 9, 10, 10a-octahydro4b 13, 8, 10bf3 -trimethyl-7(6H)-phenanthrone W,.:;.il 56 I, 2, 3, 4b, 5 , 6, Saa 9, I 0, IO a-Dec ahydro -1, lethylenedioxy-4bf3 , 8, 8, 10bf3 -tetramethyl-7{8H)phenanthrone ~ 58 7 13-Acetoxy- 3, 4, 4aa, 4b, 5, 6, 7, 8, Baa, 9, IO, 10bdocedecahydro-4b13 , 8, 8, 10bl3 -tetramethyl-1(2H)phenanthrone ~ 59 ~ A. B. Reduction with lithium tri-tertbutoxyaluminum hydride 59 Acid hydrolysis of alcohol~ 59 v Page C. Acetylation of crude keto-alcohol D-5 ,,..,,.,,.... 60 D. Catalytic reduction of keto-acetate D-6 60 ~ 7 f3-Acetoxy-l-Carbethoxyethylidine-l, 2 , 3, 4, 4aa , 4b, 5, 6, 7, S, Saa, 9, 10, 10a-te tradecahydro-4bf3 , S, S,10af3 -tetramethylphenanthrene (E-1) ,,_,_ 61 1, 713-Dihydroxy-l, 4bf3, 8, 8, 10af3 -pentamethyl-1, 2, 3, 4, 4aa, 4b, 5, 6, 7, S, Saa, 9, 10, lOa-tetradecahydrophenanthrene {~--~l 62 1, 2, 3, 4, 4aa, 4b, 5, 6, Saa, 9, 10, lOa-Dodecahydro-1hydroxy-l, 4b f3, S, S, 10af3 -pentamethyl-7{SH)phenanthrone ~ 63 3, 4, 4aa, 4b, 5, 6, Saa, 9, 10, lOa-Decahydro-1, 4bj3 , S, S, 10 f3 -pentamethyl- 7 (SH) -phenanthrone (F - 4) 64 mm ~ A. Silica gel chromatography of olefins F-3 & F-4 ............... 64 B. Acid equilibration of olefin mixture _..,..,,.,. F-3 & F-4 __.... 65 C. Unsuccessful experiments 66 ~ 3, 4, 4aa, 4b, 5, 6, 7, S, Saa, 9, 10, lOb-Dodecahydro-7, 7ethylenedioxy-1, 4b f3, S, S, 10bf3 -pe ntamethylphenanthrene {~:..?) 67 ~· 7, 7-Ethylenedioxy-2-hydroxy-l-methylene -1, 2, 3, 4, 4aa, 4b, 5, 6, 7, S, Saa, 9, 10, 10a-tetradecahydro4b f3, S, 8,10 .f3 -tetramethylphenanthrene ~ 67 3, 4, 4a a, 4b, 5, 6, 7, S, Saa , 9, 10, lOa-Dodecahydro7, 7-ethylenedioxy-1-methylene -4b 13, 8, 8, 10af3 tetramethyl-Z(lH)-phenanthrone ~ 70 A. Chromium trioxide -dipyridine 70 B. Chromium trioxide in pyridine 70 C. Manganese dioxide 71 Vl Page 6, 6-Ethylenedioxy-8af3 -methyl- 3, 4, 4acx, 5, 6, 7, 8, 8aoctahydro-1(2H)-naphthalenone iS!;;.& 71 1, Sa (3-Dimethyl-6, 6-ethylenedioxy- 3, 4, 4aa, 5, 6, 7, 8, 8a-octohydronaphthalene S9..;;,ll 72 1, 2, 3, 4, 4aa, 5, 6, 7, 8, 8a-Decahydro-6, 6-ethylenedioxy2-hydroxy-8af3 -methyl-1-methylenonaphthalene ~) 7 3 6, 6-Ethylenedioxy-8af3 -methyl-1-methylene-3, 4, 4aa, 5, 6, 7, 8, 8a-octahydro-2(1H)-naphthaleneone ,,..,,.__ {G-7) 74 IE-Methoxybenzyl chloride 75 Benzyltriphenyltin 75 IE-Methoxybenzyltriphenyltin ~) 76 General procedure for the reaction of lithium organocopper and organolithium derivatives with model ~thylene ketone G-7 77 -- 6, 6-Ethylenedioxy-8af3-methyl-3, 4, 4aa, 5, 6, 7, 8, 8aoctahydro-l f3-{2' -phenylethyl)-2(1H)-naphthalenone (~)fP._C) 78 ·'*'~--- A. Without copper (I) iodide 78 B. With copper (I) iodide 79 1-Benzyl-6, 6-ethylenedioxy-8af3-methyl- 3, 4, 4aa, 5, 6, 7, 8, 8a-octahydro-2-naphthalylacetate {2S~2SP._C) ...wm 80 A. With copper {I) iodide 80 B. Without copper (I) iodide 80 6, 6-Ethylenedioxy-Sa (3-methyl- 3, 4, 4acx, 5, 6, 7, 8, Sa-octahydro-1-(2 '-phenylethyl)-2-naphthalylacetate ~) 81 vii Page 6, 6-Ethylenedioxy-8af3 -methyl- 3, 4, 4aa, 5, 6, 7, 8, Sa-octahydro-l-(2' -12:-methoxyphenylethyl}-2naphthalylacetate (XL} S2 6, 6-Ethylenedioxy-1(2 1 -m-methoxyphenylethyl}8a13-methyl- 3, 4, 4aa, 5, 6, 7, S, Sa-octahydro-2naphthalyl acetate ,.,._... (XL} S5 -- A. With copper (II) acetate SS B. With no copper (II} acetate S6 3,4,4aa,4b,5,6,7,S,Saa,9,10,10a-dodecahydro7, 7 -ethylene dioxy-1- (2 '-12:-me thoxyphenylethyl} 4a 13, 8, 8, 10a f3 -tetramethyl- 2-phenanthryl acetate (I-1} S6 --. A. With copper (II} acetate S6 B. With no copper additive S7 la, Saf3-Dimethyl-6, 6-ethylenedioxy-1 f3-(2 '-mmethoxyphenylethyl)- 3, 4, 4aa, 5, 6, 7, S, Sa- octahydro-2(1H)-naphthalenone ~ 90 la, Saf3 -Dimethyl- 3, 4, 4aa, 7, 8, Sa-hexahydro-lf3 (2' -m-methoxyphenylethyl} -2, 6(1H, 5H}naphthalenedione ~ 91 4a f3, 4ba-Dimethyl-8-methoxy-3, 4, 4a, 4b, 5, 6, 12, 12aa -octahydro-2{1H}-chrysenone ~ 91 3, 4, 4aa, 4b, 5, 6, 7, S, Saa, 9, 10 ,lOa-Dodecahydro7, 7-ethylenedioxy-1 f3-(2 '-!!1-methoxyphenylethyl}la, 4bf3, 8, 8, 10af3 -pentamethyl-2(1H}- phenanthrone 1tll 92 3, 4, 4aa, 4b, 5, 6, Saa, 9, 10, 10a-Decahydro-lf3 (tE-methoxyphenylethyl}-la, 4b f3, S, S, 10af3 -pentamethyl2, 7{1H, SH}-phenanthrenedione .il;.il 93 References 95 viii Page Proposition 1 102 Proposition 2 108 Proposition 3 119 Proposition 4 126 Proposition 5 132 xi TABLES Page Table I: Reactions of Organolithium Compounds with Methylene Ketone _.,..,... G-7 84 Table II: Reactions of Organomagnesium Compounds with Methylene Ketones .,..,.,,.. G-7 & F-7 .,..,.,,.. 89 Historical Introduction Triterpenes constitute a large and widely disp e rsed class of natural products. Their existence has been recognized at least since 1788 when betulin.,...,... i was first isolated (1), but the absence of tenable functionality in these molecules, coupled with their high degree of assymetry, shackled them as virtual unknowns until quite recently. It has also been suggested that a lack of physiological activity- -compared to the steroids and alkaloids - -temporarily rendered the trite rpenes as unattractive candidates for vigorous re search ( 2). That the triterpenes are as well characterized as they are today is due in large part to the considerable efforts of Ruzicka and his collaborators (3), who, relying heavily on information gained from dehydrogenation experiments, were finally able to determine the structure of oleanic acid_...,.. ii (4). i .,..,.. Subsequent progre ss in ii .,..,.. z Chart A ( -H-18 ) HO -H-lZ > -H-15 > HO HO -H-3 HO ) 0 - 10 3 establishing the relative ste re ochemistry of the trite rpene s has been thoroughly reviewed by several authors (5, 6, 7, 8). Recently the intermediacy of 2, 3-oxidosqualene A-1 in the """"""" biosynthe sis of f3 -amyrin A-6 has been established (9, 10, 11, 12) ~ and its involvement in trite rpene biosynthesis in general is thus implied. It is widely considered (13, 14, 15) that pentacyclic tri- terpenes are derivable by loss of hydrogen from the sequential carbonium ions that arise from the lupeol cation _.-..A-2 by trans-diaxial 1,2-methyl and hydrogen migrations. This process is illustrated in Chart A for the formation of germanicol A-4, 13-amyrin .....,..._... A-6, taraxerol ~ ~ and friedelin ~. To this date synthetic endeavors in the triterpene field have been sparce, due in large part to the same reasons that retarded structure elucidation. A competent survey of progress in this area through September, 1967 is available in the Ph.D. thesis of D. A. Evans (16). A noteworthy observation is that since the pioneering work of Halsall and Thomas (17), all reported pentacyclic trite rpenoid syntheses have involved an acid catalyzed formation of the 8,14 bond in ring C from a tetracyclic intermediate. This process is illustrated by the last step of the synthesis of a-onocerene ..._ iv reported by Corey and Sauers (18). 4 Patte rned aft e r Barton's cyc li zation of natural onoc e rin v (19), ""' this step i s repres e nta tive of the react ion in general, be ing both intractabl e a nd of low yi e ld (10%) . A close e xamination reveals that ) lV 111 """"" """"" ) AcO AcO v .,,..,,.,. vi ,,..,,,... a direct synthesis of a tritc rpene such as gerrnanicol A -4 or f3 - arnyrin ~ A-6 by a s imilar process would be unfeas i b l e . ~ Th e reaction condition s requir e d t o initia t e cycli zation would be suf fi cie nt a ls o to e q u ilibrate the doub le bond in the produc t. A loss of stere och e mi cal integr ity at the D /E ring fusion would result. This observation h a s b ee n con- firm e d by B row nl i~' ( 20) as i s s h own i n Lhc accon-ipanyi ng diagram. 5 viii .,..,..,,...,.. -ix ix/x .,..,.. .,..,..=4/3 Interestingly, Barton has been able to circumvent this problem in what constitutes the first total synthesis of 13-amyrin (21) outlined in Chart B. Starting with the previously synthesized 6 -amyrene B-1, ~ Barton proceeded to the 1$ isomer B-4 in a sequence of six reactions. ,,..,.,;.. The key step was the treatment of the unsaturated ketone .,..,..,,...,.. B-3 with boron trifluoride etherate to give a 1:1 mixture of the l&x and 1813 -- isomers of ketone B-4. A further sequence of fifteen reactions was required to functionalize position 3. Although this effort does 6 Chart B H + 3 steps > l """"' BF 3 .Et~O 3 steps ) 3 4 "V'. 3 steps 5 6 """"" 5 ste ps NC > HO 7 4 step ~ > 7 represent a successful synthesis of 13-amyrin B-8, it suffers from the ~ -- low efficiency accompanying the synthesis of starting mate rial B-1 and the awkward manipulation of functionality required to hydroxylate _....._ C-3. A recent attempt designed to circumvent the aforementioned problems of an acid catalyzed formation of ring C was reported by Beak and Monroe (22). It was hoped to be able to join fragments ,,..,,.. xi and -- xii by a Michael reaction and then selectively cyclize the intermediate -- tri-ketone in a basic medium to give pentacyclic compound xiii. These efforts were abandoned, however, when it was discovered that -- rmdel triketone xiv experienced only retro-aldolization on treatment with a variety of bases. > -- -- xii xi RO xiii ~ xiv ~ 8 Discussion Previous reported syntheses of pentacyclic trite rpene s have suffered from low efficiency plus a lack of stereochemical control and versatility. Correspondingly in the Ireland laboratories there have for some time been syntheses in progress based on the steroidal approach of progressive ring additions. To this end Dolphini (23) synthesized tricyclic ketone .,..,... I as a potential C/D/E unit of taraxerol . II but discontinued the project when the stereochemistry of the C-4a :;c .,..,... * The bicyclic compounds discussed herein are named and numbered as derivatives of naphthalene, no. 1754 in the Ring Index. Tricyclic compounds are considered as derivatives of phenanthrene, no. 3619 in the Ring Index. Tetracyclic compounds are treated as derivatives of chrysene, no. 5254 in the Ring Index. Pentacyclic compounds are considered as derivatives of picene, no. 6384 in the Ring Index (A. M. Patterson, L. T. Capell and D. F. Walker, "The Ring Index," American Chemical Society Special Issue Sales, 1960). 9 3 9 'I 2 9 3 8 ITEthyl group could not be rigorously assigned. In addition, Dolfini embarked upon a synthesis of tricyclic ketone III as a potentially ~ versatile intermediate possessing the A/B/C stereochemical features III ~ 10 Chart C 0 > > I """"' ) 0 3 """"' OH OH 0 5 """"' > 0 0 D -2 ..-_.., 7 """"' 11 pre sent in a variety of pentacyclic trite rpene s. These include germanicol A-4, ~ -amyrin .,..,..,,_ A-6, and taraxerol ,,..,,.. II. ~ The synthetic route employed by Dolfini is outlined in Chart C. The bicyclic tetrahydropyranyl ether _._ C-3, when treated with ethylvinyl ketone and sodium methoxide followed by acid hydrolysis, produced a 19% yield of tricyclic keto alcohol .,..,..,,_ C-4. Standard rrethylation procedures, however, afforded none of the desired rn:moalkylated product .-.,,..,, C-5; only dialkylation ,,....,_ C-6 was observed. As a result, progress toward keto alcohol....,._, III was temporarily halted • It is worthy of note that recently acceptable yields of ketone ,,..,..,_ C- 7 . have been obtained from unsaturated ketone .D-2 - - by prolonged treatrrent with potassium tert-butoxide in dimethyl sulfoxide followed by rrethyl iodide (24). Removal of the C-8a( 9) double bond, however, could not be effected. Simultaneously and independently, Brown (25) was able to obtain tricyclic dione ....,._, D-1 in 15% yield by direct annelation of diketone C-2 with ethyl vinyl ketone. ,..,...,.,.. Here the low yi e ld is ascribable to a cleavage reaction arising from competitive nucle ophilic attack of alkoxide on the vinylogous ~ -diketone C-2. ~ Precedent for such a cleavage has been documented by Wendler and co-workers (26), who reported the formation of keto acid,,..,,.. V from base treatment of 12 R OOR - OR> > --C-2 -- dione IV. --- Ketone ketal D-2 was obtained by selective ketalization of COOH > 0 0 v -.. --- --- D-1 and alkylation to give saturated ketone D-3 was accomplished by the reductive alkylation procedure of Stork (27). -- Reduction of D- 3 with lithium tri-~-butoxy aluminum hydride produced ketal alcohol ...__ D- 4 in 45"/o yield from...__ D-2. Subsequent conversions to tricyclic keto - acetate D-7 occurred smoothly as described in the Ph.D. thesis of K. Schmiege! (2). The research problem that logically arose from the successful synthesis of _........ D-7 was the development of a method for the addition of 13 Chart D > 0 C-2 1 """"""" """ OJ ___ > > 0 0 3 """' 0 > > HO 5 Ac > 7 """'" 14 --- rings D and E corresponding to a triterpene such as germanicol A-4. Several possibilities were suggested but the one initially decided upon, outlined in Chart E, appeared encouraging. The main difficulty encountered in all of the proposed seque nces was the gene ration of a quaternary carbon at C-1 of the starting material ,,.,,..,,_ D-7 , and thus C-14 of germanicol A-4. ~ HO AcO --- D-7 A-4 ,,.,,_... --- Precedent for the formation of keto ester E-2 was available from recent work of Sondheimer and co-workers (28) in their -- preparation of keto ester VII from keto acetate VI. Further, sele ctive -- Grignard addition to the dibromo tetrahydrofuran moiety of E- 3 was 0 ,,, \ > AcO -VI VII .,..,..,.,.. 15 Chart E Li C::C O CiH§ 5% HzS04 A c O AcO > D-7 ~ COOC 2 H 5 1) NaBH 2) dis am'hH Ac 2 Ac > 3) HCl 4) Br 2 3 .......... 16 anticipated on the basis of studies by Crombie and Harper (29, 30) concerning Grignard additions to 2, 3-dichloro tetrahydrofurans. Jn short, it app e ared that a route to pentacyclic compound E-7 was ~ available that had reasonable promise of success, if not compe lling beauty. In a trial reaction lithium ethoxy acetylide added smoothly to keto acetate D-7, and hydrolysis with dilute sulfuric acid followed .-- -- by reacetylation afforded a 56% yi e ld of the unsaturated ester E-1. At this time it became necessary to prepare a sufficie nt -- quantity of tricyclic keto acetate n.-7 to sustain a re search effort of some duration. This entire sequence will be briefly discussed ~ince several modifications of the old procedures significantly improved the efficiency of the synthesis. Following the procedure of Brown (31), base catalyzed condensation of ethyl vinyl ketone, prepared from 1-chloro- 3-pentanone by the procedure of Archer (32), with 2-methylcyclohexane -1, 3-dione afforded an 84% yie ld of unsaturated diketone _.,,.,,.. C-2. Jn an attempt to avoid the problems arising from bas e cleavage during the ensuing homoannelation, compound C-2 ,,..,.,.. was treated with ethylene glycol and a trace of acid in benzene to afford a 94% yield of crystalline unsaturated ketone ketal VIII. . - - This high yield is in contrast to the corresponding tricyclic unsaturate d diketone -- D-1 which underwent selective ketalization quite inefficiently (50%) (33) 17 -- in forming ketone kctal D-2. Fortunately when ketone ketal VIII, which ~ had previously been prepared by Kitahara and co-workers (81), was treated with ethyl vinyl ketone in the presence of sodium methoxide, > --- > --- C-2 VIII > --- D-2 --D-3 --- 52% of the starting material VIII could be recovered on high vacuum distillation. Chromatography of the distillation residue on alumina --- afforded a 52% yield of the desired product D-2 based on the unrecovered starting material. This corresponds to a yield of 25% based on the initial amount of starting material in the reaction. -- -- The combined yield for the conversion of C-2 to D-2 was 23% versus 7. 5% via the previous techniques. In addition, recycling the recovered ketone 18 ketal VIII twice increased this yield to 43%, r e presenting more than a --.... fivefold improvement on the previous procedure. Methylation at C-8 was accomplished as described by Brown (34), except that in this case an equivalent of tert-butyl alcohol was added to the reaction mixtu re. A 77% yield of ketone ketal D-3 was obtained after purification. --.... The remaining steps in the preparation of keto acetate ,,,._.,,... D-7 were per formed without intermediate purification. Ketone ketal D- 3 was ~ reduced with lithium tri-tert-butoxy aluminum hydride (35) in tetrahydrofuran to give crude alcohol ,,,__ D-4. The latter, after hydrolysis of the ketal in dilute mineral acid and acetylation with acetic anhydride in pyridine, was passed through an alumina column. Subsequent reduction in an hydrogen atmosphere with 10% palladium on carbon in ac e tic a cid afford e d, after recrystallization, an 81% yield (four steps) of pure tricyclic keto acetate D-7. --.. -- The combine d yie ld of D-7 from -- 2-methylcyclohexane-l, 3-dione C-1 was 22%. 8 steps 22% > Ac C-1 --.. D-7 ~ 19 While the preceding sequence was being compl eted, a second and more attractive route for the addition of rings D and E was conceived. If a compound containing a methylene ketone, as repre- sented in partial formula,,...,..,... IX, could be prepared, it seemed reasonable that it could be made to undergo conjugate addition with a suitable Grignard derivative RMgX. r.yo ~CHz Moreover, it should be possible to quench RMgX > IX -- -- -- """"""' x XII XI the intermediate magnesium enolate X with a variety of alkylating """'- agents, depending on the specific goal of the synthesis. It was anti- cipated that copper catalysis during the conjugate addition would be necessary. 20 To this end the methylene ketone derived from keto acetate D- 7 """"- was prepared as summarized in Chart F. Methyl lithium addition to keto acetate ,,..,..,,.,,... D-7 proceeded without incident to give a mixture of sparingly soluble diols _..,..,..,, F-1 as shown by the unresolved methyl region of the nmr spectrum. Without purification the C-7 hydroxyl group of _..,..,..,, F-1 was oxidized with an equivalent of Jones' reagent to give keto alcohol F-2, which the nmr spectrum indicated was still an epimeric ~ mixture at C-1. pyridine at -15 Dehydration of this mixture with thionyl chloride in ° occurred smoothly to give a mixture of olefins that consisted of 27% of the compound with the endocyclic double bond F-4 """"""" and 63% of its exocyclic isomer F- 3 as adduced from an integration of ~ the vinyl proton resonances in the nmr spectrum. The ratio of the amounts of the exocyclic isomer to the endocyclic isomer (2. 3:1) allows one to make some estimate of the relative populations of the C-1 alcohol epimers in .,..__ F-1 that arose during the methyl lithium addition • In studying the dehydration of the tertiary alcohols derived from cholestanone (36) with phosphorous oxychloride in pyridine, Barton showed that the equatorial 3 f3 -alcohol .,..,.,,..,,.. XIII gave predominantly the exocyclic methylene compound ,.,..,,.., XV, while the 3 a -alcohol -.,,..,... XIV afforded the endocyclic isomer XVI exclusively. ~ If the extrapolation as to reaction mechanism from the phosphorous oxychloride -pyridine reaction medium to that in the thionyl chloride -pyridine reaction medium is valid, then the ratio of the amounts of f3 -hydroxyl compound 21 Chart F D-7 1 .,,...,..,.,.. """"" CHz > 2 3 """"" """"" > > 4 """"" CHz 6 """"" > 7 """"" 22 XIII -- X IV ~ l -xv XVI ......._. CH2 -- F-3 F-4 ~ 23 f---~12 -.r~.. to a-hydroxyl compound ¥..--~g is at least 2. 3:1. ~· The direction of this ratio is in accord with that expected on the basis of steric ease of approach of the nucleophilic methyl lithium species to the carbonyl group. When purification of the olefin mixture F-3, F-4 was attempted ~~ by chromatography on silica gel, there was obtained a 2: 3 mixture of two products. The major product was the de sired endocyclic olefin ,!.:;! as shown by its ir and nmr spectra. The minor product, C19H 30o, had a strong, saturated carbonyl absorption in its infrared spectrum, but was devoid of any vinyl proton signals in its nmr spectrum. On this basis the new compound was tentatively as signed structure -~y_rg, ...-.wvm arising from a biogenetic-like migration of the C-lOa methyl group in ion XVII followed by loss of the C-4a proton. .-- Compound XIX, however, """"""" is also a possibility and cannot be ruled out on the basis of the available evidence. ) -- XIX XVII ---. XVIII -----.... ~ ~· 24 S e veral methods were tried to selectively secure endocyclic olefin F-4. ........,_ -- The first of these was the pyrolysis of the ester XX derived from diketene according to the procedure of Frisell and Lawesson (37). Although high yields of olefin F-4 _..,..,., were predicted, no recognizable products were obtained. Similarly, treatment of o~ Tl 0 0 -- xx F-4 _...._ the alcohol mixture~ , ~ with iodine at 140° according to the procedure of Church and Ireland (38) proved unsuccessful. An attempted equilibration of olefin mixture_...._ F-3, F-4 with anhydrous ~ formic acid led only to the incorporation of formate as shown by strong ester carbonyl absorption in the infrared spectrum. Fortunately it was discovered that the endocyclic olefin F-4 could be secured in ~ high yield by treatment of the mixture of olefins with E_-toluenesulfonic acid in benzene at reflux. ....... None of the rearranged product ~J:I_IJ was ~ detected. Protection of the C-7 carbonyl group by ketalization with 25 ethylene glycol proceeded without incident. The five-step trans- formation from keto acetate D-7 to endocyclic olefin F-5 could thus .-- ~ be accomplished in 86% yield if purification of the intermediates was avoided. The introduction of an oxygen function at C-2 was accomplished by photooxidation according to the general procedure of Nickon and Bagli (39), using hemataporphrin dihydrochloride as sensitizer. The yields of this reaction varied considerably from experiment to experiment, being within the range of 45-67%. -- Structure F-6 with an equatorial C-Za alcohol was assigned to the major product on the basis of the broadness of the C-2t3 axial hydrogen in the nmr spectrum. The minor product, although not isolated in a pure state, is considered to be the C-2 hydroxyl epimer since both compounds can be oxidized -- to the same ketone F-7. No products corresponding to tertiary alcohols such as ....,.,,..,... XXI were isolated, although the variety of products remaining in the reaction mixture does not preclude the possibility. An interesting side product isolated from the reaction mixture in 7% yield was the exocyclic olefin XXII. This product can be seen as ~ arising from allylic hydrogen abstraction from starting olefin F-5 ............... followed by rearrangement and hydrogen capture. The anticipated 1ag excited state of oxygen in sensitized photooxidations (40), which lies 22 kcal above the oxygen triplet ground state, would not be 26 H XXI XXII ~ ~ expected to be a good hydrogen abstractor on the basis of its electronic configuration. 1 The second excited state, however, E g+ has a singlet energy of 37 kcal. This species should resemble ground state oxygen in its electronic configuration (i.e., two electrons in different orbitals), and would be expected to be a good hydrogen abstractor. Perhaps significantly, the triplet energy of the hematoporphrin 1 sensitizer is 37 kcal {41), sufficient to produce some z;g+oxygen. It seems reasonable, therefore, to ascribe the formation of product ..... to hydrogen abstraction by this reactive oxyge n species • ~~g ~ The oxidation of allylic alcohol F-6 to methylene ketone F-7 ~ ~ proceeded smoothly with chromium trioxide-dipyridine in methylene chloride {42). This was fortunate, since the usual Sarett procedure (43) of chromium trioxide in pyridine as well as manganese dioxide had proved inferior. 27 At this point a model compound for the planned conjugate addition reaction was chosen such that a maximum of information could be obtained without exhausting the supply of tricyclic methylene ketone .,..,..,,.,.. F-7. As is desirable with all studies of this kind, the model employed should be available in sufficient quantity and should be as chemically similar to the authentic material as is conveniently possible. To this end was chosen the bicyclic methylene ketone .,..,..,,.,.. G-7, the eleven-step synthesis of which is outlined in Chart G (44). Keto alcohol G-3 was prepared as describe d by Dolfini (45). ~ 0 Ketalization of this material, followed by Jones' oxidation at o afforded an 86% yield of ketal ketone .,..,..,,.,.. G-4. , Treatment of this material with methyl lithium and subsequent dehydration of the intermediate tertiary alcohol with iodine at 140° (38) produced olefin G-5 in 70% ~ yield. Photooxygenation according to the procedure of Wharton (46} using ros e bengal as a sensitizer in isopropyl alochol gave, after reduction of the intermediate hydroperoxide, a 70% yield of allylic alcohol ,,__ G-6. This material was about a 9:1 mixture of the equatorial a-hydroxyl compound and its axial 13-isomer. No search for a compound corresponding to exocyclic methylene although it was very likely pre sent. m was conducted, Oxidation of the allylic alcohol mixture ,..,..,,_ G-6 with chromium trioxide in pyridine (43) afforded a 46% yield of model methylene ketone ,,.,,..,,._ G-7. 28 Ch a rt G > ) 1 C- 1 .....,...,.. """' OH > > 0 0 3 ,,....,... a 0 > > 4 """"' 0 > CH 2 29 Conjugate addition to a, f3 -unsaturated carbonyl compounds by organometallic reagents is a widely observed, yet little understood, process (47). By and large, the addition of a copper (I) or copper (II) salt to an organo magnesium or an organo lithium derivative facilitates conjugate (1, 4) addition relative to normal (1, 2) addition by enhancing the rate of the former and diminishing the rate of the latter . That a copper species is not always required, however, is confirmed by the observation (48) that methylmagnesium iodide adds to 3-methylene camphor -~i<}H to give the conjugate addition product ~~I_': exclusively. ~~m tr...-'-'•~ 0 > XX III ~ XXIV In a c ontinuing series of papers on the copper catalyzed conjugate additions of lithium and magnesium derivatives (49-54), House finds compelling evidence that the reactive species is an ethe r soluble organocopper complex such as XXV. Although a definitive ~ rrechanism is lacking, there are strong indications that conjugate addition does not involve a cyclic transition state (50, 55). House 30 CH 3 "-cu Li+ CH/ 3 xxv ~ interprets his data (49) as favoring a one electron transfer from copper (I) to the carbonyl group producing an intermediate radical anion followed by alkyl radical transfer to give the enolate of the observed product. R-CH'"-CH-C-R II ' f. 0 R' ,...cu(I) -M R-CH-CH:q-R . R' 0 M+ ":;£u(II) ) -. R-CH-CH=C-R I R' I O_M+ ,...cli<r> An interesting observation is that derivatives of metals less reactive than lithium and magnesium (Zn, Al, Cd, Be) exhibit a strong tendency toward conjugate addition in preference to 1, 2 addition even when no copper species is present (56-58). Although the effect of added copper salts would be interesting in these cases, no information could be found. 31 The 1E-methoxybenzyl group was chosen for the study of conjugate additions to methylene ketone G-7 since the product would ~ be adaptable to further transformations required in the synthesis of germanicol. Because the results from IE-methoxybenzyl derivatives closely paralleled those from the unsubstituted benzyl derivatives, the two were often used interchangably. 0 > > XXVI ........ G-7 _..,.,,,.. XXVII ______ XXVIII ... .. ... ~ ~ 32 When methylene ketone ....,,..,,_ G-7 was added to an ethereal solution of benzyl magnesium chloride containing 6. 5 mole % of copper (II) acetate and the reaction quenched with ammonium chloride, there was obtained a 41% yield of the de sired product XXIX. ~ In addition, unresolved mixtures of products containing both carbonyl and alcohol absorptions in their infrared spectra were found. Attempts at separation were unsuccessful, but it is thought that they represent 1, 2 ~as well as 1, 4 ~addition. Quenching with methyl iodide did not appreciably change the results, as shown by the persistence of considerable amounts of alcoholic material. XXIX xxx .,..,.._., Attention was next turned to the corre spending lithium de riv ative s. On the basis of the results of House (49-54) it was expected that the addition of one-half of an equivalent of copper (I) iodide to benzyllithium would produce a solution of lithiumdibenzylcoppe r. This material has the same stoichiometry as that of XXV, derived ~ 33 from m e thyllithium which gives consistently high conversions to 1 , 4-products. Prior to 1950 there were no convenient preparations of CH Li 2 6 + Cul > benzyllithium (59), but two potentially valuable preparations have recently been reported. The first {60) involves the cleavage of a benzyl ether with lithium metal. The anticipated metalation of an activated aromatic ring under these conditions, however, limits the value of this procedure. In the second method {61), the reaction of phenyllithium with benzyltriphenyltin in ether at - 35 benzyllithium and solid tetraphenyltin. ° produces The yields of benzyllithium in this reaction, as determined by double titration techniques {62), were about 800/o; the yield of precipitated tetraphenyltin was 90%. Parallel results were obtained for the synthesis and decomposition of ?E-methoxybenzyltriphenyltin ~~-2S!· ..-.r......... It appeared that a convenient preparation of benzyltin and IE-methoxybenzyltriphenyltin was in hand. 34 In practice the benzyllithium solution was separated from the tetraphenyltin by filtration, and copper (I) iodide added to the > -~1951 ......... 0 supernatant liquid at -20 • tially homongeneous. The resulting green solution was essen- --- When the methylene ketone G-7 was added and the reaction quenched with ammonium chloride, yields of 45% of the expected product ~~were obtained; alcoholic material was a considerably smaller portion of the reaction mixture than before. Interestingly, with no copper additive, 30% of compound~ was --- isolated in addition to 19% recovered starting material G-7 and 13% alcoholic material. An experiment in dimethoxy ethane with no copper additive yielded 42% product ~~Pf and 56% recovered starting --.r•~ nB.terial. Starting material recovery must arise from enolization by the benzyllithium, followed by protonation and recovery on workup. No starting material was recovered in reactions with added copper iodide. 35 I The results thus far indicate that enolization of starting rrnterial can be minimized using lithium dibenzylcopper, but that the biproducts in this reaction make purification of the product difficult. To resolve this difficulty, different quenching agents for lithium enolate _J5?9fg were sought. ..-.4¥!Wm The eventual necessity of a C-14a methyl group in germanicol A-4 required that the integrity of ~ the enolate be maintained for a future methylation step. bilities were considered. Chlorotrimethylsilane is an efficient ~-~iC_I! XXXIII XX XIV xxxv ....-m ~ Two possi- 36 oxygen scavenger as demonstrated by recent studies of Bloomfield (63) in trapping the intermediate enediolates of the acyloin condensation. ............. It was thus anticipated that the silyl ether ~lC~!l_l could be easily formed by adding chlorotrimethylsilane to enolate ~~lCJ!, and mw.,. could also be easily alkylated by successive treatment with methyllithium and methyl iodide to give the necessary C-1 methyl group corresponding to C-14 of germanicol. On the other hand, the susceptibility of the silyl ether ~~~~IJ ·~~ to water hydrolysis promised to make its isolation difficult. That the alkylation step was conceptually sound has recently been shown by Stork (64, 65) in several efficient alkylations of the silyl enol ethers of cyclohexanone systems. The quenching agent finally employed was acetic anhydride. The efficacy of this choice is demonstrated in the following experiments. When methylene ketone G-7 was treated as before with lithium ~· dibenzylcopper at -20° and the resulting reaction mixture quenched with acetic anhydride, two products were isolated, the anticipated enol acetate ~~~J:~I] in 46% yield and the unexpected enol acetate . . . .~~44"+~ X~XI~ in 26% yield. .-...c.v•~..., Product lC~lC~ is the result of conjugate phenyl ~mm addition and must be ascribed to the presence of either phenyllithium or lithium diphenylcoppe r at the time of reaction, even though an excess of benzyltriphenyltin relative to phenyllithium had been 37 XXXVI XXXVII XXXVIII XXXIX -------- r.-..wv...-m employed. Recalling that the yields of solid tetraphenyltin during the benzyllithium preparation were only 90%, it becomes obvious that there must be an equilibrium between the remaining tetraphenyltin and the benzyllithium. If competition for 38 conjugate addition favors lithium diphenylcopper, a relatively small proportion of phenyllithium would afford a significant amount of ...... Some indication as to the reactivity of lithium ... __ _ diphenylcopper was gained when an 89o/o yield of enol acetate ___ XXXIX .... .... enol acetate-~~-~!~· ~ ~ occurred on addition of methylene ketone G-7 to an ethereal solution ~ of two parts of phenyllithium and one part of copper {I) iodide. These results were paralleled with m-methoxybenzyltriphen~ltin and phenyllithium. The expected product .,..,..,... XL was formed in 51% yield, enol acetate -~~~~ in 25%. ·~~ XL ~ XXXIX ..------~ Running the reaction at lower temperatures (-60°) did not noticably affect the ratio of benzyl to phenyl addition. On the other hand, when the reaction was run at 0°, the yield of the expected benzyl product XL .,..,..,... decreased to 31%, while the corresponding phenyl addition product inc re as ed to 5 0%. One convenient interpretation is that the increased solubility of tetraphenyltin at the higher 39 temperature led to a corresponding increase in the phenyllithium concentration and thus the change in product distribution. The lack of sensitivity to lower temperatures, however, casts some doubt on this interpretation. Some ancillary notes concerning the reaction of methylene --- ketone G-7 with the lithium organocopper derivatives are in order. 0 When the reactions were run at o , there was often no recognizable product obtained whatsoever. The exact reason for this is uncertain, but probably stems from the reactivity of the methylene ketone itself. The ease of dimerization of cyclic methylene ketones relative t o their endocyclic counterparts is well documented (66); the added presenc e of a copper species may be sufficient to make precarious the balance between polymerization and the desired addition. Similar effects were noted when tetrahydrofuran was employed as a cosolvent. Again no useful products were found. It is reasoned that the superior ligand properties of tetrahydrofuran relative to ether result in an effective removal of copper from its anticipated role as a conjugate addition catalyst. A similar affect of uncatalyzed Grignard rate retardation by tetrahydrofuran has been noted by Holm (67) and by House (68). The results just discussed were discouraging. Although good yields of conjugate addition were obtained (70-80%), the mixture of 40 -- ... products, -~~-~Y} or XL and_~.?S~RC-• greatly reduced the effectiveness ~ of the method. __._ It was decided to return to the organomagnesium compounds and investigate the product distribution using the acetic anhydride quenching technique. The re was some concern that significant amounts of C-acylation would occur from the magnesium - enolate s in view of House's report ( 4 9) that e nolate s of type XLI and XLII give nearly equal amounts of dike tone XLIII and enol acetates ~ ~-1=!25 ..,_... - ~ and XLV when Mis magnesium, but only the enol acetates when Mis lithium. However, the observation of Malhotra and Johnson (55} - XLI fOCH 3 (CH 3} zCH-CH-COCH 3 XLIV XL III - XLV that only 0-acylation was observed from ketone XLVI was encouraging • . . .'+'*7. . 41 When methylene ketone G-7 with a trace of copper (II) acetate ~ and either benzyl magnesium chloride or IE-methoxy benzyl magnesium chloride were allowed to react, the product isolated after quenching with acetic anhydride was the expected enol acetate ~2S?C_Y!~l or XL ~--~ .-- MgBr > > XLVII XLVI + XLIX XLVIII -------- ·~4'47· XLVIII/XLIX ~ 1 in 71% and 68% yields respectively. found in either case. No product of C-acylation was Similarly, when the Grignard additions were performed with no added copper salt, the yields of ~~-~YJ!! and XL ~.,..,..,... were 73% and 71%. These conversions represent acceptable procedures. A summary of the results of the conjugate addition experiments is 42 -XL XXXVIII given in Table I for the organolithium compounds and in Table II for the organomagnesium derivatives (pages 84 and 89). Some subsequent conversions anticipated for the germanicol -- synthesis were carried out on model enol acetate XL and are outlined in Chart H. very useful. The methylation procedure of House (69) proved -- Treatment of XL with two equivalents of methyl lithium in dimethoxy ethane gave lithium enolate H-1 which, on alkylation with ~ an excess of methyl iodide, afforded a 78% yield of desired ketone H-2. ~ It was imperative that the newly introduced C-1 methyl group have the alpha configuration shown in structure~· Confirmation of this stereochemical assignment was obtained using some recent nmr studies (70-73) of the magnitude and direction of the shifts of the quaternary methyl resonances in aromatic solvents relative to inert solvents (CDCl3, CCl4). In general, these studies determined that 43 Chart H > > XL _...,... 1 ~ OCH I I Tl > 3 """"' 0 0 -4 3 44 in ketones the resonances of methyl groups lying in or in front of a plane perpendicular to the C-0 axis and passing through the carbonyl carbon will experience downfield (negative) shifts in benzene; methyl signals behind the plane will undergo upfield (positive) shifts. following example is illustrative. The C-4 methyl resonances of compound Lin deuteriochloroform both occur at 63 cps. ~ The In benzene solution the C-4a methyl group has a signal at 68 cps, a negative shift of 5 cps, while the C-4f3 methyl resonance occurs at 56 cps, a positive shift of 7 cps. Clearly the C-1 methyl group resonance in L ~ ketone H~Z should undergo a positive shift in benzene if the stereo- .-- chemistry is as indicated. The observed shift of +18 cps (from 78 cps to 60 cps) clearly demonstrates that the desired stereochemi str y has been achieved. 45 H-2 .............. An attempted cyclization of k e tone H-2 in E_-toluenesulfonic ac id ~ and toluene at reflux resulted in destruction of starting material. This r e sult is in contrast to ketone LI w hich unde r the sam e conditions .............. smoothly cyclize d to styrene LII (74). """"' Fortunate ly di one H- 4, the ~ product of mild acid hydrolysis of ketone ketal H-2, underwent smooth """"""" cyclodehydration in polyphosphoric acid at 60- 70° (75) to give the expected c h r ysene derivative H-5. ~ CH 3 LI """""""' LII """"" 46 • The success within the model series indicated that similar results in a synthesis of germanicol A-4 could be expected. ~ investigation is found in Chart I. A summary of the results of this -- Enol acetate I-1 was secured from ?E-methoxybenzylmagnesium chloride and copper (II) acetate in 71"/o yield. -- ' With no copper (II) acetate a 74% conversion to I-1 resulted. Methylation was accomplished as before with two equivalents of methyl -- lithium to give lithium enolate I-2 which was then quenched with methyl iodide, affording a 76"/o yield of ketone ketal I- 3. .-... The orientation of the C-1 methyl group is assigned the alpha axial configuration on the basis of its +17 cps shift in benzene relative to deuteirochloroform (from 77 cps to 60 cps). Hydrolysis of the C-7 ketal group to give dione ,.,_... 1-4 was routine. - Dione I-4 represents the extent to which the synthesis has been carried. A preliminary cyclodehydration with polyphosphoric acid at 70° was inconclusive. impure. Some product (=10"/o) was formed but was The indication is that although these reaction conditions were satisfactory in the model series, they are too severe in the present case. It is not expected that this will be a major obstacle. -- Dione I-4 has subsequently been prepared in the laboratories of Professor W. s. Johnson of Stanford University, using a natural relay, 47 by the degradative-synthetic sequence summarized in Chart J (76). A projected route for the completion of the synthesis of germanicol is outlined in Chart K; present any major difficulties. to the task is anticipated. as such it does not appear to An early and successful conclusion 48 Chart I > F-7 .,...,..,_ I ....... CH 3 ) 2 ....... 0 4 ....... .......3 49 Chart J --~:: -- - ~ Hz/Pt HOAc/HCl;) >AcO Ac 100° z ,,...,.. 1 (Bu0} 3 CrO ) Ac AcO 4 .....,... 3 ,_ H COO C H 3 ) NaBH 4 Z) di s a mBE· ) AcO AcO 5 ,,...,.. 1) Hz/Pt> HO Z} CrO 7 ...._ 3 0 8 .....,... 50 Chart K 1) Li/N.H 3 ,.,.....,,.,,,........,~----~> 2 ) H30+ AcO . 0 - I-5 1 ~ 1) A1Et 2 c~ 2) LAH 3) W-K 4) Ac 0/Pyr 2 0 1) Br' 2)-HBr AcO ) ) AcO - - -- -- 2 3 AcO 4 5 51 Experimental Section (a) All compounds described herein containing an assymetric carbon atom are .racemic; the prefix "dl 11 has been omitted. (b) Melting points (mp) were determined on a Kofler Micro Hot Stage melting point apparatus and are uncorrected. (c) hifrared (ir) spectra were determined on a Perkin - Elmer 237B grating infrared spectrophotometer. Solution spectra wer e observed in 0.1 mm cavity cells using chloroform as solvent and polystyrene calibration bands (3027 .1 and 1601. 4 cm -l). (d) Nuclear magnetic resonance (nmr) spectra were determined on a Varian spectrometer using Silinar C [Merck, Sharp, and Dahme trade natne for deuterochloroform (CDCI ) containing 1% tetra3 tnethylsilane {tms) as an internal standard] or benzene (C6H6) as solvent. (e) Resonances are recorded in ppm (6) downfield from tms. Vapor phase chromatographic (vpc) analyses were performed on an F & M Model 810 Research Chromatograph equipped with hydrogen flame detectors. Columns used were generally 6 1 x 1/8" 5-10% SE-30 or SE-52 on diatoport S. The carrier gas {helium) flow rate was maintained at 60 ml/min. Retention times (rt) were recorded in minutes. 52 (£) Analytical thin layer chromatography (tlc) was performed on l" x 3" microscope slides covered with silica gel G, Brinkmann Instrument Co. Components were detected by spraying with a 5% solution of phosphomolybdic acid in ethanol followed by heating at 0 100-15 o for several minutes. Preparative thick layer chromatography (ptlc) was performed on 20 x 20 cm plates coated with a 1 mm layer of silica gel PF Brinkmann Instruments Co. the aid of uv light. 2 54 +366 Bands were generally observed with Compounds were removed from specific bands by washing with ether-chloroform. (g) Elemental analyses were performed by Spang Microanalytical Laboratories, Ann Arbor, Michigan. (h) Anhydrous solvents were dried immediately prior to use. Ether, benzene, tetrohydrofuran (THF), and dimethoxyethane (glyme) we re distilled from lithium aluminum hydride (LAH); was distilled from calcium hydride; tert-butyl alcohol pyridine was distilled from either barium oxide or calcium hydride. Petroleum ether refers to the fraction of boiling point (bp) 30-60° as supplied by J. T. Baker Chemical Co. , 53 (i) The i solation procedure ge ne r a lly employed involved dilution and extraction with the solvent indicate d, separation of the aqueous and organic phases, and washing of the organic layers with saturate d sodium bicarbonate solution (if the reaction medium had been acidic), water until neutral (litmus), and brine. Concentration generally refers to evaporation to constant weight on a rotary evaporator. 54 ~.--The general procedure of Archer and co- workers (32) was followed. To a 3 1. flask equipped with a mechanical stirrer and a Claisen distilling head was added one -half of 5 35 g (4. 45 moles) 1-chloro-3-pentanone (77) and 1 g dihydroquinone. To this with moderate heating was slowly added 715 g (4. 97 moles) ovendried sodium benzoate. When the slurry became difficult to stir the remaining l-chloro-3-pentanone was added, followed by the remaining sodium benzoate. Continued heating and vigorous stirring resulted in a clear distillate (hp 95-123°) which was dried (CaCI 2 ) to afford 321 g clear liquid containing less than 1% 1-chloro- 3-pentanone by vpc (oven temperature 80°, rt 2. 5 min). Re distillation produced 30 9 g (82%) clear liquid, bp 100-103°; [lit (78) hp 101.1-101. 3° (732. 8 mm)]. 5, Sa t3 -Dimethyl-3, 4, 8, Sa-tetrahydro-1, 6 (2H , 7H)-naphthalenedione -....~ ,r;w ~·--The procedure of Brown ( 31) was followed in the preparation of this diketone. A mixture of 98.0 g (0.776 mole) 2-methylcyclohexane - 1, 3-dione (79),101.0 g (1.19 moles) freshly distilled ethyl vinyl ketone, and 5 pellets of potassium hydroxide in 300 ml of reagent methanol we re heated under reflux in a nitrogen atinosphe re for 4 hr. The excess ethyl vinyl ketone and methanol were removed on a rotary evaporator using benzene as an azeotroping agent . To this dark oil 55 was added 500 ml reagent benzene and 7. 5 ml pyrrolidine (Eastman Kodak, practical grade) and the resulting solution heated under reflux in a nitrogen atmosphere for 16 hr. The water formed during this period (13 ml, 93% of theoretical) was removed by means of a Dean-Stark water separator. The reaction mixture was cooled, diluted with 600 ml ether, and washed successively with 2% hydrochloric acid, water, and brine. Drying (MgS04) and concentration produced 153 g red oil which was vacuum distilled to afford 124 g (84%) light yellow oil, bp 123-149° (0.2-0.4 mm) that was homogeneous by vpc (oven temperature 200°, rt 3.1 min~. A small portion of this oil was crystallized by trituration with ether to give colorless crystals, mp 36-40 ° J2.it (80) mp 45-46J . ~~ ~·--To a ~ 5 1. flask equipped with an overhead stirrer was added 52.0 g (0.272 mole) dione _..,_ C-2 described above, 250 ml ethylene glycol, 686 mg £_-toluenesulfonic acid monohydrate, and 3500 ml reagent benzene. The solution was heated under reflux in an atmosphere of nitrogen for 4.5 hr. The water formed during the reaction was removed by means of a Dean-Stark water separator. Upon cooling, the reaction mixture was poured into a 5 1. separatory funnel, the glycol phase (lower) removed, and the benzene phase 56 worked up as usual. Afte r drying (MgS04) and concentration there r e maine d 64 g pale yellow oil which, on th e a ddition of 15 0 ml p e troleum e the r and cooling in a dry ice -acetone bath, deposited 0 53.1 g pale yellow powder, mp 48-50 • A second crop of crystals obtained in the same manner weighed 6. 7 g , mp 45-48°, making the combined yield 93. 50/o. Two recrystallizations from pentane and one from ether-pentane afforded the analytical sample, mp 56-56.5°. ir (CHCl3) 1655 cm 1180-1050 cm -1 -1 (C=O), 1605 cm (ketal); -1 (C=C), 1375 cm -1 (-CH 3), nmr (CDC1 3 ) :5 1.33 (s,3, C-8a -CH 3 ), O 3.95 (s , 4 , ketal). Anal. Calcd for c 14H 20o 3 : C, 71.16; H, 8.53. Found : C, 71.16; H, 8. 38. Subsequently an alternate report of ketal ketone VIII ......_ was encountered in the literature (81) with a reported melting point of 0 55-56 • l, l-Ethtlenedio~-1 , 2, 3, 4b, 5, 9, 10, 10a-octah.idro-4b~, 8 lOb~ - ~~- l!2:?). - -A flask containing 14. 97 g t~~ip._e_t.!i_y}_-_~(_6_~ ~1!!:'!1!1_1:.0_:r.:_~~~ _ ,.~.-.v.wr.wr..--~ ~ (63.5 mmoles) ketone ketal VIII, mp 48-50.5 0 and 3.43 g (63.5 mmoles) ~ of fresh sodium methoxide in 50 ml reagent methanol under an atmosphere of nitrogen was heated (bath 68°) for 0.5 hr. To this was added dropwise over 0.5 hr 5.35 g (63.5 mmoles) freshly distilled 57 ethyl vinyl ketone and the resulting dark solution heated at reflux for 10 hr. After cooling, the reaction mixture was worked up as usual. Drying (MgS04) and concentration produced 18. 9 g light yellow oil which was vacuum distilled to give 7. 8 g (52%) of colorless oil, bp 105-111 0 (0.07 mm). Crystallization of this material from pentane in a dry ice-acetone bath gave 6.9 g white solid, mp 42-46 0 that was identical by a comparison of ir and nmr spectra to the starting bicyclic ketal ketone ,,..,.,,....,.. VIII. The pot residue from the distillation was chromatographed on 300 g Woelm activity III neutral alumina. Elution with 10 1. 7% ether-petroleum ether produced residual starting mate rial, several components of unknown structure, and a small amount of the desired tricyclic product. Further elution with 8 1. 18% ether-petroleum ether afforded 5.12 g pale yellow oil that deposited 4. 21 g white crystals, mp 62-66 °, on trituration with cold pentane. These crystals did not depress the melting point {mp 62- 65 °) of an authentic sample of tricyclic ketone ketal D-2 {82). In addition, the ir and nmr spectra of the two samples were identical. Rechroma- """"""" tography of the mother liquors on 40 g Woelm activity III neutral alumina yielded an additional 0. 61 g tricyclic ketone ketal D-2, mp 61~ 65°. The yield of_....,,._ D-2 in this experiment was thus 25% based on the amount of starting bicyclic ketone ketal and 52% based on unrecovered starting material. The range of yields for this reaction over 12 58 experiments was 31-52% based on unrecovered starting material. In addition, two recyclizations of recovered starting material raised the yield in this step to 43% based on starting material used in the first cycle. 10af3 -tetrameth l-7(8H)-phenanthrone (D-3). - -The ketone ketal D- 3 -----. was prepared by the general procedure of Brown {34). Thus a solution of 12.56 g {41.5 mmoles) tricyclic enone D-2 and 3.01 g ~ {40. 7 mmoles) dry tert-butyl alcohol in 250 ml anhydrous ether was added over five minutes to a flask containing a vigorously stirred solution of 1.10 g (159 mmoles) lithium wire in 1 1. dry {freshly distilled from sodium) ammonia and 300 ml anhydrous ether. After O. 25 hr a solution of 100 g dry {freshly distilled from calcium chloride) methyl iodide in 100 ml anhydrous ether was added in one lot. The ammonia was allowed to evaporate over 10 hr and the n the reaction mixture was diluted with 400 ml ether and washed with water and brine. Drying (MgS04) and concentration afforded 12. 6 g yellow oil. The addition of 25 ml ether and a seed crystal of ketone ketal ,...,..,,,...,. D- 3 pro0 duced after 12 hr at o 9. 30 g white solid, mp 94-98°. Recrystal- lization from petroleum ether gave 8.17 g (62%) colorless crystals, 0 Jl1l 97-99 • The combined mother liquors (4. 3 g) were chromatographe d 59 on 400 g alumina (Woelm III, neutral) and eluted with 7 1. 15% etherpetroleum ether to give 1. 74 g white prisms, mp 98-103°. The total crystalline material (9. 91 g, 77%) was identical by a comparison of ir and nmr spectra to authentic ketone ketal (D3). ,,.,.....,_ 713 -Acetoxx:- 3, 4, 4aa, 4b 5 6, 7, 8 Saa 9 10 lOa-docedecah~dro­ • . A. Reduction with lithium tri-.!!:.!.!,-butoxyaluminum hydride. - - The procedure of Brown {83) was modified for this reaction. To a solution of 2.41 g (7.58 mmoles) ketone ketal D-2, _...,...,.... mp 97-99°, in 75 ml anhydrous tetrahydrofuran was added a solution of 8. 00 g (31. 4 mmoles) lithium tri-tert-butoxyaluminum hydride (35) in one lot and the combined material heated at reflux under a nitrogen atmosphere for 1 hr. After cooling, 6. 5 ml 5% sodiwn hydroxide solution was added slowly and the resulting mixture stirred for 5 hr. The precipitate was removed by filtration and carefully washed with several portions of ether. Concentration of the organic material afforded 2.46 g white powder possessing strong alcohol absorption at 3600 and 3440 cm -I in its ir spectrum and no significant carbonyl band. This material was carried on without further purification. B. ~.--A solution of 2.46 g crude alcohol from above in 150 ml acetone and 50 ml 10% hydrochloric acid 60 was stirr e d at room temperature for 1 hr. Most of th e ace tone was re move d on a rotary e vaporator and the r es ulting aqueous phase worked up as usua l with e thel;'. Drying (MgS04) and c oncentration affo rded 2.12 g oily solid that wa s imme diately acetylated. alcohol D-5 was acetylated as described by Rasmusson (84). Thus ~ 2 .12 g of oily solid from above in 50 ml of 1:1 anhydrous pyridine ace tic anhy dride were allowed to stand at room temperature under a nitrogen atmosphere for 16 hr. The exce ss pyridine and ace t ic anhydride were rem o ved on a rotary e v aporator with several toluene a zeotropic d istillations to give 2. 59 g yellow solid. This crud e acetate D- 6 was passed through a col umn of 100 g Woe l m activity III n e utral ~ a lumina, e l utin g w ith 25% eth er-pe troleum eth er , to give 2 . 06 g white solid, mp 121-128 °. D. procedure of R asmus so n of ke to -acetate D-6. --Following the ( 84), 2. 06 g crude keto acetate P-6 in ~ 100 ml g l ac i a l aceti c acid wa s exposed to a n atmosphere of hydroge n in the presence of 3 50 m g of 10% palladium on carbon. After 205 ml of hydrogen h a d been consum ed (2 0 hr) the catalyst was removed by filtra tion and th e filtrate concentrated a t r e duce d pres sure. The 0 residue was dissolved in 7 0 m l acetone, c ooled to 0 , and tre ated with 1 ml of J ones ' reagent . The mixtur e was stir red for 0 . 5 hr and 61 then 2 ml iso-propyl alcohol added and the mixture stirred an additional 0. 25 hr. After 25 ml water were added, the acetone was removed on a rotary evaporator and resulting aqueous mixture worked up as usual with ether. The ethereal extract was dried (MgS0 ) and 4 concentrated to a crystalline solid. Recrystallization from ~-heptane afforded 1.96 g (81%) white plates, mp 153-156°; (reported (84) 111> 161-162.5°]. 713 -Acetoxy-1- c arbethoxyethylidine -1, 2, 3, 4 , 4aa, 4b, 5, 6, 7, 8, Saa, ~ 9, 10, 10a-tetradecahxdro-4bl3 , 8, 8, 10al3 -tetrameth,il:ehenanthrene ~· .QS.:.U. --Following the procedure of Sondheimer and co-workers (28), to a cooled (0°), stirred solution of 1. 25 ml methyl lithium solution (2.13 Min ether; 3. 25 mmoles) in 1 ml dry ether under a nitrogen atmosphere was added dropwise 326 mg (4. 60 mmoles) ethoxy acetylene in 2 ml dry ether. After stirring o. 5 hr at room temperature, the solution was cooled to -15 0 (ice-methanol) and 104 mg (0. 325 mmole ) keto acetate,..,..,...,.,.. D-7 in 3 ml dry ether was added dropwise over 15 min. After stirring at -10° for 1 hr and at room temperature for 18 hr, the reaction mixture was worked up with ether as usual. Drying (MgS04) and concentration produced 170 mg yellow oil which showed a strong acetylene absorpticm in the ir spectrum at 2250 cm -1. The crude oil was dissolved in 10 ml methanol and 1 ml 5% sulfuric acid 62 and stirred at room temperature for 1. 5 hr, whereupon a normal ether workup afforded, after drying (MgS04) and concentration, 140 mg yellow oil. This oil was re -acetylated in 10 ml 1:1 pyridine-acetic anhydride for 18 hr to give after evaporation of the solvent 129 (101%) yellow oil. Purification by ptlc (60% ether-petroleum ether) afforded 71 mg which crystallized from cold ether-hexane to give a white solid, 0 mp 121-123.5 • The analytical sample was recrystallized from ether- hexane to give white crystals, mp 122-123.5°. ir (CHC1 3 ) 1740 cm -1 {acetate C=O), 1710 cm -1 {a,13-unsatd ester), 16 27 cm -1 (C=C). C, 73.96; H, 9.94. 6,7,8,8aa,9,10,10a-tetradecahydrophenanthrene {F-1).--To a solution of 2.260 g (7.07 mmoles) keto-acetate D-7, mp 153-156°, in 150 ml """"""" anhydrous ether cooled to -15° (ice-methanol bath) under a nitrogen atmosphere was added 21 ml of a solution of methyl lithium (2. 0 M in ether; 42 mmoles) in 25 ml anhydrous ether over 0.5 hr. After stirring for 2 hr at room temperature, the reaction mixture was poured onto 20 g ice and worked up as usual with ether. Drying {MgS04) and concentration afforded 2. 04 g (98%) white solid which was a mixture of C-8 epime rs as evidenced by its nmr spectrum. A small sample recrystallized 3 times from acetone -n-hexane served as the 63 analytical sample , mp 118-121° J.385, 1375 cm -1 ir (CHCI 3 ) 3600, 345 0 cm -l (-OH), (-CH 3 ); nmr (CDC1 3 ) 6 0.67, o 0.75, o 0.85, 6 0.90, 6 0.95 (angular methyl signals ofF-lepimericmixture). """""" Anal. Calcd for c 19H 3 40z: C, 77.41; C, 77.50; H, 11.64. Found: H, 11.71. 1-7(8H)-phenanthrone (F-2 • --To a stirred solution of 1. 96 g ( 6. 68 mmoles) diol F-1in150 ml reagent acetone ~ cooled to 0 0 was added Jones' reagent dropwise until after one more drop a red-orange color persisted in the supernatant liquid for O. 25 hr. To this was added 6 drops iso-propyl alcohol and the resulting green 0 solution stirred for 0.25 hr at o • After 40 ml water were added and most of the ace ton e r em oved on a rotary evaporator, the resulting aqueous solution "vas worked up as usual with ether. The ethereal phase was dried (M gS04) and concentrated to afford 1. 89 g (97%) white solid which was homogeneous (althou gh epimeric at C-1) by analytical tlc (ether, rf 0.55). A small sample r ecrystallized twice from acetone- ~-heptane afforded the analytical sample, mp 96-100°. ir (CHC1 3 ) 3600, 3460 cm-I ( -OH), 1697 cm -l (C =0), 1385, 1375 cm -l (-CH ); 3 nmr ( CDCl3) 0 0.95, 0 1.00, 0 1.05, 0 1.08, 6 1.10 (angula r methyl signals of F-1 epimeric mixture ). ~ 64 Anal. Calcd for c 1 9H 3 202: 78.06; C, 78.03; H, 11.03. Found: C, 78.00, H, 10.94, 10.86. 3, 4, 4aa:, 4b, 5, 6, Baa:, 9, 10, lOa-Decahrdro-1, 4bi3 , 8, 8, 10ai3- • • ~.--To a stirred solution of 1. 70 g (5. 82 mmoles) keto alcohol F-2 in 20 ml dry pyridine under a ~ nitrogen atmosphere at -15 ° (ice -methanol bath) was added by syringe 1.23 ml (2.06 g, 17.3 mmoles) thionyl chloride. stirred at -15 The solution was ° for 0. 25 hr and then poured onto ice and worked up in the usual manner with ether-benzene. The combined organic layers were dried (MgS04) and concentrated to yield 1. 59 g (100%) light yellow oil that was shown to be 27% of the endocyclic isomer F-4 and ~ 6 3% of the exocyclic isomer F- 3 by integration of the vinyl protons in ~ the nmr [the endocyclic vinyl proton occurs as a broad absorption at j 5. 24 and the exocyclic vinyl protons as a doublet ( .:!_= 1. 5 Hz) at O 4.51]. These olefins were not resolvable by vpc (oven temperature 0 200 , rt 8 min). A. 25 O mg of the above olefin mixture was chromatographed on 35 g silica gel eluting with 200 ml 10% ether-petroleum ether, there was obtained 241 mg of white solid, the nmr of which showed no absorption due to the vinyl protons of the exocyclic olefin F- 3. ~ Vapor phase chromatography indicated the presence of two components in a 41:59 65 ratio (oven temperature 210°, rt 3. 2 and 4.1 min respectively}. Careful chromatography of this new mixture on 50 g Woelm activity III neutral alumina eluting with 200 ml 2% ether-petroleum ether effected their separation. The first component eluted was the minor isomer 0 Two recrystallizations from ~-heptane afforded XVIII, mp 56-59 • ~ the analytical sample, mp 67-69°. ir (CHCI 3 ) 1700 cm- 1 (C=O}, 1360, 1380cm-1 (-CH 3 ); nmr(CDCI 3 ) 60.97(s, 3, -CH 3}, 61.00 (s, 3, -CH 3 }, o 1.07 (s, 6, 2-CH3's}, o 1.10 (s, 3, -CH 3 ). Anal. Calcd for c 19H 30o: 82.94; C, 83.15; H, 11.02. Found: C , 82.86, H, 10.92, 10.96. The second component eluted was the endocyclic olefin F-4, ~ 0 rr.p 97-100 • Two recrystallizations from ~-hexane and one from 20% ether-hexane produced the analytical sample, mp 102-104°. 1700 cm-l (C=O), 1380 cm-l (-CH 3); nmr (CDCl3} ir (CHC1 3 ) o 0.93 (s, 3, -CH3), o 1.03 (s, 6, 2-CH 3 s), O 1.07 (s, 3, -CH3}, O 1.60 (s, 3, C-1 vinyl1 CH }, 3 o 5.24 (m, l, C-2 vinyl H). Anal. Calcd for c 19H 30 o: 83.19; C, 83.15; H, 11.02. Found: C, 83.28, H, 10.89, 10.94. of the remaining 1. 34 g of olefin mixture F-3 and F-4 and 121 mg ~ """"""' £_-toluenesulfonic acid monohydrate in 200 ml reagent benzene was 66 heated at reflux i.mdc r a nitrogen atmosphere for 4 hr. The reaction rrii.xturc was cooled, worked up with ether, dried (MgSO 4 ), and concentrated to afford 1. 27 g (94%) white solid, mp 93-97°. Recrystal- lization from a small amount of ~-hexane gave 1. 09 g (81%) crystals, mp 98-101°, that was identical with the endocyclic olefin ..__ F-4 from the previous experiment by a comparison of their ir and nmr spectra. C. ~--Several additional experiments were performed in an attempt to obtain endocyclic olefin F-4. ~ (i) Treatment of keto-alcohol ..__ F-2 with iodine at 140° as de scribed by Ireland {38) resulted in a gross mixture of products and significant loss of carbonyl absorption in the infrared. (ii) Attempted dehydration of keto-alcohol ..........._ F-2 by pyrolysis of the 13-keto ester derived from diketene (37) similarly met with little success. (iii) Treatment of olefin mixture ..........,_ F-3 and..__ F-4 with sodium acetate -acetic acid at room temperature for 16 hr resulted in no change in the isomer ratio. (iv) Equilibration of olefin mixture ..........,_ F- 3 and ..__ F-4 with pyridinium hydrochloride in pyridine at room temperature for 4 hr was unsuccessful. (v) Anhydrous formic acid treatment of olefin mixture..__ F-3 and ..__ F-4 at room temperature for 7 min resulted in the addition of formic acid to the olefin as evidenced by the appearance of an ester carbonyl at 1735 cm -l in the ir spectrum. 67 1, 4b(3, 8, 8, lOa (3-pentamethylphenanthrene (F-5). - -A stirred mixture ·~~~ of 3. 28 g (11. 9 mmole s) keto-olefin F-4, mp 94-97°, and 101 mg ~ .E,-toluenesulfonic acid in 100 ml ethylene glycol and 500 ml reagent benzene were heated at reflux in a nitrogen atmosphere for 10 hr. The water formed during the reaction was removed by means of a Dean-Stark water separator. The cooled reaction mixture was diluted with 150 ml ether, the ethylene glycol phase removed, and the organic layer worked up as usual. Drying (MgS0 4 ) and concentration afforded 3. 69 g (97%) white solid, mp 158-163°. Recrystallization of a portion 0 from ~-heptane produced the analytical sample, mp 168-169 • ir (CHCI 3 ) 1380 cm -1 (-CH3), 1180-1050 cm -1 (ketal); nmr {CDC1 ) 3 6 0.83 (s, 3, -CH 3 ), 5 0.90 (s, 3, -CH3), O 0.90 (s, 3, -CH3 ), O I.OO (s, 3, -CH 3 ), 0 1.55 (s, 3, C-1 vinyl -CH3 ), O 3.94 (s, 4, ketal), O 5.15 (m, 1, C-2 vinyl H). Anal. Calcd for c 21 H 3 40 2 : C, 79.19; H, IO. 76. Found: C, 79.29; H, 10.77. 5!...:l2.· --In a modification of the method of Nickon and Bagli (39), a solution of 1. 37 g (4. 31 mmoles) olefin F-5, mp 161-165~ and 100 mg ~ hematoporphrin dihydrochloride in 100 ml reagent pyridine was placed 68 in a 36 cm x 2 cm pyrex tube. A glass frit was suspended in the solution and a slow stream of oxygen gas that had been passed through a gas washing bottle containing pyridine was bubbled into the reaction mixture. Three 15 watt fluorescent desk lamps were placed 2-3 cm from, and parallel to, the reaction tube and irradiation begun. After 17.5 hr an additional 100 mg sensitizer were added and the irradiation continued for a total of 34. 5 hr. The pyridine was re- moved on a rotary evaporator by codistillation with benzene, and then 0 300 ml ether were added and the solution cooled to o and stirred 4 hr with excess lithium aluminum hydride. After quenching with 10% sodium hydroxide solution, the mixture was filtered and concentrated to afford 1. 2 g light yellow oil. The vpc indicated a complex mixture with 2 minor and 2 major components (oven temperature 290°, rt 1. 8, 1. 9, 2. 7, and 3 .1 min). Chromatography on 140 g Florisil and elution with 400 ml 10% ether-petroleum e ther produced 180 mg white solid that was a 2:1 mixture of two components by vpc (oven ternpe rature 240 °, rt 2. 5 and 2. 7 min). Separation of these components by ptlc (2% ether-petroleum ether) afforded 61 mg starting material and 95 mg (7%) exocyclic olefin 295!!.• mp 162-165°. Two recrystal- lizations from ether-~-hexane afforded the analytical sample, mp 167- o 167. 5 • ir (CHC1 3 ) 1380 cm -1 (-CH 3 ), 1180-1060 cm -1 (ke tal), 890 cm - (C= CH2 ); nmr (CDCI 3) o 0.85 (s, 3, -CH 3 ), 6 0.90 (s, 3, -CH3 ), 1 69 o 0.93 (s, 3, -CH 3), O 1.05 (s, 3, -CH 3 ), 0 3.95 (s, 4, ketal), O 4.50 (d, .'.!.=1.5 Hz, 2, C=CH 2 ). Anal. Calcd for c 21 H 34o 2 : C, 79.24; C, 79.19; H, IO. 76. Found: H, 10.64. Further elution of the Florisil column with 400 ml 50% etherpetroleum ether produced 788 mg (62%) of a 1:9 mixture of allylic alcohols F-6, ~-OH and F-6,a-OH, that could be carried on as such """"""' """"""' without further purification. Two recrystallizations of a sample from ethanol-petroleum ether effected purification of the major F-6, a-OH """"""' isomer, mp 164. 5-165. 5 °. (This sample was placed on the hot stage pre-heated to 150° whereupon it melted and re-solidified as white needles to give the indicated mp). ir (CHCI3) 3600, 3450 cm -I (-OH), 1635 cm-I (C=C), 1380 cm-I (-CH 3), 1140-1060 cm-l {ketal), 890 cm- 1 (C=CH 2 ); nmr (CDCI 3 ) 6 0.85 (s, 3, -CH 3 ), 6 0.88 (s, 3, -CH3 ), 0 0.93 (s, 3, -CH3}, O 1.05 (s, 3, -CH 3 ), o 3.94 (s, 4, ketal), O 4.25 (m, 1, C-2 H), o 4.69 (m, l, C=CJJ, o 4.88 (m, 1, C=C Anal. Calcd for c 21 H 34o 3 : H, 10.17. C, 75. 41; H, 10. 25. ...,...H ). Found: C, 75. 26 ; 70 l-methylene-4b(3 , 8, 8, lOa (3-tetramethyl-2(1H)-phenanthrone (F-7). ............~ Collins and co-workers (42) a solution of 192 mg (0. 57 mmole) of 0 allylic alcohol .-... F-6, mp 152-157 , and 892 mg (3.42 mmoles) solid chromium trioxide -dipyridine in 20 ml reagent methylene chloride were stirred for 5 min. Filtration of the reaction mixture through 12 g Woelm activity III neutral alumina eluting with 25 ml methylene 0 chloride afforded 190 mg white solid, mp 154-157 • Recrystallization from acetone-~-hexane produced 172 mg (91%) white crystals, mp 159162 o that was identical with material prepared in part B below. procedure of Sarett (43), 306 mg {O. 92 mmole) allylic alcohol F-6 .-... in 5 ml dry pyridine was adde d to 360 mg (3. 60 mmoles) chromium 0 trioxide in 3 ml dry pyridine at o and the resulting brown mixture stirred at room temperature for 23 hr under a nitrogen atmosphere. After 50 ml reagent ether were added and the resulting mixture filtered through celite and worked up as usual, there remained upon drying {MgS0 4 ) and concentration 281 mg pale yellow solid, mp 1471530. Recrystallization from acetone-petroleum ether resulted in 198 mg {64%) white solid, mp 154.5-157 .5°. A small portion was recrystallized from acetone-~- hexane to give the analytical sample, 71 np 160-163°. 1380 cm -1 ir {CHC1 3 ) 1680 cm -l (unsatd C =O), 1610 cm- 1 (C= C), (-CH 3), 1180-1050 (ketal); nmr (CDC1 3 ) O 0.88 (s, 3, -CH ), 3 6 0. 97 (s, 6, 2-CH3'~, 6 1. 0 3 (s, 3, -CH 3 ), 5 3. 95 (s, 4, ketal), 6 5.01 (d, 1, !!_=1.5 Hz, C=C,H), Anal. Calcd for c 21H 32o 3 : o 5.54 (d, 1, J=l.5 Hz, C=C/H ). C, 75.86; H, 9.70. Found: C, 75.69; H, 9.57. C. Manganese dioxide. - -A slurry of 130 mg (0. 39 mmole) ~~ allylic alcohol mixture F-6 and 2. 03 g manganese dioxide in 20 ml """""' reagent chloroform was stirred at room temperature for 20 hr. Filtration and concentration yielded 128 mg white solid. spectrum indicated some residual alcohol absorption; [oven temperature 270°; The ir vpc analysis rt 4.1 min {ketone) and 5.4 min {alcoholD showed that only the major alcohol isomer (a-OH) had reacted to any appreciable extent. The remainder of the spectrum was similarly unencouraging. 6, 6-Ethylenedioxy-8a(3 -methyl-3, 4, 4aa, 5, 6, 7, 8, 8a-octah~~•• l_(.?!!l-:1!~~~!~!15>n_e_j9..:'!l· --A stirred solution of 79. 4 g (0. 44 mole) ~~-~-~~~ keto alcohol G-3 (45), 150 ml ethylene glycol, and 1.0 g £_-toluenesulfonic ~ acid in 2. 5 1. reagent benzene was heated at reflux under a nitrogen atmosphere for 3 hr. alcohol. 0 to o , Normal workup afforded 99.1 g crude ketal This material was dissolved in 11. reagent acetone, cooled and treated with 122 ml Jones' reagent (2 equiv) keeping the 72 temperature below 3°. Afte r diluting with 200 ml water, the bulk of the acetone was removed on a rotary evaporator, followed by a normal ether workup. Drying (MgS0 4 ) and concentration yielded 83. 6 g (86%) ketal ketone £.:!! mp 45-47°; [ut (38) mp 47-48J. 1, Saa -Dimethyl-6, 6-ethrlenedioxy- 3, 4, 4aa, 5, 6, 7, 8, 8a~··~ octohxdronaphthalene (G-5). --To 98. 7 g (0.44 mole) ketal ketone G-4 ~.-.~~ ~ in 1 1. dry ether in a nitrogen atmosphere was added dropwise 400 ml of 2 M methyl lithium (0.8 mole) in ether and the resulting solution stirred at room temperature for 1 hr. After quenching with saturated ammonium chloride solution followed by a normal ether workup, the ethereal layer was dried (MgS04) and concentrated to yield material that still possessed a medium (<15%) carbonyl absorption in the ir spectrum at 1710 cm -l. The material was then again treated as above with 400 ml of the methyl lithium solution (O. 8 mole) to afford after similar workup 100. 8 g (95o/o) of the crude tertiary alcohol. Then a ~lt of 99.9 g (0.41 mole) of the crude material was heated to 140° (38) and 1. 9 g iodine added in 3 portions over 1 hr. The dark nB.terial was cooled, diluted with ether, washed with 10% sodium thiosulfate, dried (MgS04), and concentrated, yielding 94. 0 g crude olefin. Vacuum distillation afforded 69. 0 g (70% from ..........._ G-4) of clear oil, bp 99-100° (O. 35 mm). A small sample was evaporatively 73 distilled (90-95 1375 cm -1 0 ir (CHC1 ) 3 at 0. 04 mm} to give the analytical sample . (-CH }, 1130-105 0 c m 3 -1 (ketal}; nmr (CDC1 } 0 0. 97 (s, 3, 3 C-Sa -CH 3 ), O 3.94 (s, 4, ketal}, 6 8.59 (m, l, C-2 vinyl H}. Anal. Calcdforc14H 22o 2 : C, 76.22; H, 9.74.Found: C, 76.04; H, 9. 76. ~~G~.--In an adaptation of the procedure of Wharton (46}, a solution of 15. 0 g (67. 5 mmole s} olefin G-5, bp 99-100 0 ,...,.-. (O. 35 mm} and 3.0 g rose bengal in 200 ml iso-propyl alcohol was placed in an immersion-type photoreaction vessel with a stream of oxygen bubbling through the solution from a glass frit in the bottom. Irradiation with a Hanovia 450 watt medium pressure 0 light source through a pyrex filter sleeve ( >.>3000A} was carried out over a 7 hr period. After the solvent was , removed on a rotary l evaporator, 500 ml ether was added and the solution, with excess LAH, stirred for 6 hr. Filtration and concentration afforded 16. 5 g yellow oil that was chromatographed on 300 g Florisil. Elution with 2 1. 15% ether-petroleum ether gave residual starting material plus other low boiling components. Further elution with 3 1. 60% ether-petroleum ether yielded 11. 8 g (72%) of solid material composed mainly of a 1:10 mixture of the epimeric alcohols~ (3 -OH; ~a-OH. Crystal- lization of a sample from ether-n-hexane afforded white needles, 74 0 mp 93-94.5 • Further recrystallization from ether-~ -h exane produced pure a -hydroxy compound as the analytical sample, mp 93-94. 5°. ir (CHCI 3 ) 3 6 00, 3450 cm (kf'tal); nmr (CDCl 3) (-OH), 1640 cm -1 (C =C), 11 30-1050 cm -1 o 0.99 (s, 3, C-8a -CH3), O 3.94 (s , 4, ketal), 6 4. 33 (m-broad, 1 , C-2 C =C -1 H), 6 4. 78 (m, 1 , C=C, ), o 5.07 (m , 1, H ..)-I ). Anal. Calcdfor c 14H 180 3 ; C, 70.53; H, 9.30. Found: C, 70.40; H, 9.22. octah dro-2 lH) -na hthaleneone (G-7). --As des cribed in the general procedure of Sarett (43), a mixture of 11. 6 g (49. 2 mmoles) of the allylic alcohol mixture~and~ (1:10) from the previous experime nt in 170 ml dry pyridine and 15. 5 g (155 mmole s) chromium trioxide in 160 ml dry pyridine was stirre d at room temperatur e for 12 hr. After 500 ml ether were added and the resulting solution filtered through super-eel; the filtrate was worked up as usual to produce 9. 4 g crude ketone G- 7. Recrystallization from ether gave ~ 0 5. 3 g (46%) pale yellow crystals, mp 92-94 • Recrystallization from ether afforded the analytical sample, mp 94-96°. ir (CHC1 3 ) 1685 cm -l (C=O), 1610 cm-l (C=C), 11 30-1050 cm-l (ketal); nmr (CDC1 3 ) .S 0 .95 (s, 3, C-8a -CH 3 ), O 3 .96 (s, 4, ketal), 0 5.12 (d, 1, J=l. 5 Hz, C=C,H), o 5.68 (d, 1, J=l.5 Hz, C =C ,..H ). 75 Anal. Calcd for c H o : 14 20 3 C , 71.16; H, 8.53. Found: C, 71.22; H, 8.51. JE-Methoxybenzyl chloride. - -Following a procedure referred to by • Streitweiser (85), 9 .11 g (65. 5 mmoles) IE-methoxybenzyl alcohol (Aldrich Chemical Co.) in 10 ml reagent benzene was cooled to o0 and treated with 20 ml thionyl chloride. After 10 min the excess thionyl chloride was removed on a rotary evaporator and the dark residual oil distilled to give 5. 7 g (55%) clear mobil oil, hp 152-155° (25 mm); flit (86) hp 124° (13 mmQ; ir (CHCl3) 2850 cm -l {-OCH 3) 1605, 1590, 1495 cm -1 {aromatic), 1040 cm {s, 3, -OCH 3 ), -1 (m ·~ thoxyl); nmr (CDC1 ) O 3. 72 3 o 4.48 {s, 2, -CH2 Cl), o 6.62-7.40 {m, 4, aromatic hydrogens). In a separate experiment conducted by Joseph Ham in these laboratories, the same chloride was obtained in 72% yield by refluxing a 1:1 mixture of IE-methoxybenzyl alcohol and triphenylphosphine in carbon tetrachloride for 3 hr {87). ~--In a variation of the "Organic Syntheses" {88) preparation of allyltriphenyltin, a solution of 10. 0 g (25. 9 mmoles) triphenyltinchloride {Matheson, Coleman, and Bell} and 5.06 g freshly distilled benzyl chloride (40. 0 mmole) in 30 ml dry THF was slowly added to a stirred, vigorously refluxing suspension of 2. 0 g 76 (82 mmoles) magnesium turnings in 40 ml dry ether. When the addition was completed (1 hr), 20 ml dry benzene was added and the mixture heated at reflux for 5 hr. The residual Grignard was quenched by th e addition of 7 ml saturated ammonium chloride solution. After diluting the decanted organic phase with ether followed by a normal workup, drying (Na 2 so 4 ) and concentration afforded 10. 4 g (93%) white solid, mp 88-90°; [lit (89) 90-91°]. J:E-Methoxybenzyltriphenyltin (XXXI).--A solution of 5.0 g (12.9 mmoles) • triphenyltinchloride (Matheson, Coleman, and Bell) and 3.15 g (20. 2 mmoles) ?E-methoxybenzyl chloride in 15 ml dry THF was slowly added to a stirred, vigorously refluxing {bath 61°) suspension of 1.0 g (41 mmoles) magnesium turnings in 20 ml dry ether under a nitrogen atmosphere. When the addition was completed (0. 5 hr) 10 ml dry benzene were added and the reaction mixture heated at reflux for 10. 5 hr. After quenching with 3 ml saturated ammonium chloride solution, the organic layer was decanted and worked up as usual with ether. Drying (MgS0 4 ) and concentration yielded a clear oil that 0 deposited 5. 75 g (94%) white solid, mp 57-591 on trituration with cold petroleum ether. . Recrystallization from petroleum ether gave 0 5.41 g (88%) white cubes, mp 60-61 • A further recrystallization of a portion from !;_-hexane gave the analytical sample, mp 61-61. 5°. 77 nmr (CDC1 3 ) o 2.92 (s, 2, -CHz-), o 3.48 (s, 3, -OCH 3 ), o 6.4-7.8 (m, 19, aromatic hydrogens ). Anal. Calcd for c 26 H 24 osn: Found: C, 66.11; H, 5.19; C, 66.28; H, 5.14; Sn, 25.19. Sn, 25.07. G e neral procedure for the reaction of lithium organocopper and or anolithium derivative s with model meth lene ketone G-7. Benzyllithium and IE-methoxybenzyllithium were prepared from the c orresponding triphenyltin compounds by reaction with phenyllithium (61). A 100 ml 3 neck flask (flask A) was fitted with a serum cap and a nitrogen inle t. A piece of glass tubing (5 mm od) passing through a rubber stopper (00) was placed so that the end of the tube came to within 5 mm of the bottom of flask A. Several glass wool plugs were placed in the tube to filter out the solid tetraphenyltin. The other end of the tube passed through a stop c ock and another rubber stopper {00) into a se c ond 100 ml 3 neck flask (flask B) similarly e quipped with a serum cap and nitrogen inlet . The desired triphenyltin derivative in 75% of the total solvent volume indicate d in T a ble I was placed in flask A and cool ed to th e d es ired temperatur e (usually - 35°). A calculated amount of standardized (62) phenyllithium solution (1. 65 Min benzene) was added by syringe to flask A. Within two minutes a heavy white precipitate of tetraphenyltin formed . After O. 5 hr at the indicated temperatur e , the bright yellow supernatant liquid 78 was forced through the connecting tube by a positive nitrogen pres sure 2 (5 lbs /in ) into flask B. The residual solution in flask A was w ashed into flask B with one portion of 10% of the total solvent volume. Flask B was brought to the desired t emperature and a weighed amount of solid copper {I) iodide added if required. After O. 5 hr the methylene ketone G- 7 in the remaining 15% of the solvent was added by syringe """""" and the solution kept at the indicated temperature for 1 hr before quenching with the appropriate reagent. Isolation by normal ether workup followed by drying and concentration afforded mate rial that was purified by ptlc, one plate for each 100 mg of methylene ketone G-7. """"""'" -8a~-m eth l-3,4,4aa,5,6,7,8,8a-octahydro-l~- (2 '-phe nylethyl)-2{1H)-naphthale none {XXIX). ~""'· 1. 69 g {3. 84 mmoles) benzyltriphenyltin and 3 .12 mmoles phenyllithium in 30 ml dry ether was prepared and trans ferred to flask Bas described previously. ---- At -50° a solution of 310 mg {l. 31 mmoles) k eton e G-7 in 6 ml dry ether was added by syringe. The solution was stirred for 1 hr, then quenched with solid ammonium c hloride and worked up with ether as usual. Drying (MgS0 4 ) and concentration afforded 745 mg oil that was purified by ptlc {25% ether-petroleum ether , double development). A minor {slow-moving) band weighing 61 mg {19%) was 79 0 crystalline starting material, mp 90-93 • The major (fast-moving) band crystallized from cold ~-hexan e to give 131 mg (30%) white solid, 0 mp 73-75 • Recrystallization from ~-hexane gave the analytical 0 sample, mp 75 - 76. 5 • ir (CHC1 3 ) 1710 cm -1 (C =0), 1600 and 1495 cm - 1 (aromatic), 1160-1030 cm- 1 (ketal}; nmr (CDC1 3 ) O 0.82 (s, 3, C-8a -CH 3 ), O 3.95 (s, 4, ketal), O 7.25 (s, 5, aromatic hydrogens). Anal. Calcd for c 21 H 28o 3 : C, 76. 79; H, 8.59. Found: C, 76. 71; H, 8. 61. Other material from the ptlc plates appeared to be a mixture of carbonyl and alcohol containing compounds as evidenced by ir spectra. B. prepared from 2.36 g (5.36 mmoles) benzyltriphenyltin and 4.37 mmoles phenyllithium in 55 ml dry ethe r and transferred as described above was added 504 mg (2. 65 mmoles) copper (I) iodide. The reaction mixtur e became a very dark gre e n and appear e d n e arly homoge neous. After 0.5 hr 400 mg {1.69 mmoles) ketone G-7 in 10 ml e ther was added """"""' by syringe. After 1 hr at -20° the r e action mixtur e was quenched with solid ammonium chloride and worked up, isolated, and purified as before to give 249 mg (45%) ketone XXIX. ~ observed. No starting material was 80 A. (I) iodide. --A solution of 1.54 ml phenyllithium ............ ,,....,..,,....,......,..,,....,.."""'""'~~............~ ~ solution (1. 65 Min benzene, 2. 47 mmoles) in 35 ml dry ether under a nitrogen atmosphere was cooled to -20° and 242 mg (1. 27 mmoles) copper (I) iodide added. The green solution was stirred for O. 5 hr and then 201 mg (O. 85 mmoles) ketone G-7 in 5 ml dry ether added """""' by syringe. After 1 hr 4 ml (43 mmoles) acetic anhydride wer e added and the solution stirred at room temperature for 1 hr. A normal ether workup, followed by drying (MgS04) and concentration yielded a light oil that was purified by ptlc (25% ether-petroleum ether, double development) to give 268 mg (890/o) homogeneous oil that crystallized on trituration with ether-~-hexane (1:5) to give white crystals, mp 6970°. ir (CHCI 3 ) 1740 cm-l (acetate C=O), 1650 cm-l (enol C=C), 16 00 and 1495 cm -1 (aromatic), 1140-1030 cm -1 (ke tal); nmr (CDC1 ) 3 CH 2 -) 6 7.05 (s, 5, aromatic hydrogens). Anal. Calcd for c 22 H 28o 4 : C , 7 4. 16; B. C, 74.13; H, 7. 92. Folllld: H, 7. 85 • ~--A solution of 1.03 ml phenyl- lithium solution under a nitrogen atmosphere (1. 65 Min b en zene, 1. 70 mmoles) in 40 ml dry ether was cooled to -15° and then 200 mg 81 (0. 85 mmoles) ketone G-7 in 8 ml dry ether added by syringe. ~ After 0 O. 7 hr at -15 , the reaction was quenched with 4 ml (43 mmoles) acetic anhydride and stirred at room temperature for 0. 5 hr. A normal ethereal workup and purification as in part A afforded 123 mg (32%) of enol acetate XXXIX, mp 64-67°,identical to that from part A ~ by ir and nmr spectral comparison. 6, 6-Ethylenedioxy-8af3 -meth 1- 3, 4, 4aa, 5, 6, 7, 8, 8a-octah dro-1~""""""'""""""""""""""'""""""'""'"""'""""""""'"""'"""'""'""'"""'""""""""""""'"""'""'"""'""""""''""""..,....,.~""' ~ - -A solution of benzyl- lithium from O. 842 g (1. 90 mmoles) benzyltriphenyltin and 1. 58 mmoles phenyllithium in 40 ml dry ether was prepared and transferred to flask B as described above. At -20°, 183 mg (O. 96 mmole) solid copper (I) iodide was added. After stirring this solution for 0. 5 hr, 150 mg (O. 63 mmole) enone G-7 in 8 ml dry ether were added by ~ syringe and the solution stirred at -15 to -20 0 for 1 hr, whereupon 4 ml (43 mmol c s) acetic anhydride were added at room temperature. After 0.5 hr a normal ether workup with drying (MgS04) and concentration produced a clear oil that was purified by ptlc (25% ether-petroleum ether, double development). The major band, 126 mg, consisted of two compounds in a 2.4:1 molar ratio as determined by an integration of the C- 8a methyl and ketal nmr signals. mp 73-75 0 The major compound, of this mixture could be separated by crystallization from cold n-hexane. Two recrystallizations from the same solvent afforded 82 0 the analytical sample of enol acetate XXXVIII, mp 77-79 . ir ~ 1735 cm -l (acetate C=O), 1675 cm -l (enol C=C), 1600 and 1495 (aromatic), 1140-1030 cm-l (ketal) ; nmr (CDC1 3 ) 5 1.02 (s, 3, C-8a -CH3), o 2.13 (s, 3, -COCH3}, o 3.95 (s, 4, ketal), O 7.28 (s, 5, aromatic hydrogens). Anal. Calcd for c 23 H o : 30 4 C, 74.56; H, 8.16. Found: C, 74.57; H, 8.23. The minor product was shown to be the enol acetate XXXIX. ~ A second ptlc band, 19 mg, was pure enol acetate XXXIX. The yield ~ of 1,4-bcnzyl addition product XXXVIII was thus 47%; ~ that of 1,4- phenyl addition (XXXIX), 13%. ~ 6, 6-Ethylenedioxy-8at3 -methyl- 3, 4, 4aa, 5, 6, 7, 8, 8a-octahydro-l(2 '-IE--methoxyphenylethyl) -2-naphthalylacetate (XL). - -A solution of IE--me thoxybenzyllithium from 1.21 g (2.57 mmoles) of IE-methoxybenzyltriphenyltin and 2. 09 mmoles phenyllithium in 36 ml e th er was prepared and transfe rr e d to flask B as described above (8 ml ether for wash). iodide was added at -20 Then 2 42 mg (1.28 mmole s) solid copper (I) 0 and the solution stirre d for 0. 5 hr, where - upon 200 mg (O. 85 mmoles) ketone G-7 in 7 ml dry ether was add ed """""" by syringe. After 0.9 hr a t -20 to -15°, the solution was brought to reflux and quenche d with 4 ml (4 3 mmoles) ace tic anhydride. After o. 3 hr the reaction mixture was coole d, worked up with ether as usual, 83 dried (MgS0 4 ) a nd c oncentr a t ed . Purification by ptlc (2 5% ether - p etroleum e ther , double development) afforded 82 mg (25 % ) pure benzyl e nol acetat e XXXIX. The m ajor (slowe r ) band, 17 3 mg (51% ), ~ which cry s talli ze d from cold -n-hexane ,afforded the analyt i cal sample ,......_ XL, 0 rrp 72-74 • . ir {CHC1 3 ) 1740 cm 1600 1580, 1490 cm -1 1 -1 (enol C= C), (ke tal) ; nmr (CD C 1 ) 3 o 2.10 (s, 3, -COCH3), 6 3 .7 6 ( s, 3 , -OCH 3 ), 6 3.89(s, 4, ketal),o H, 8.02. (acetate c= 0), 16 75 cm (aromatic ) , 1140-1070 cm - 61.02 (s, 3, C-8a - C H 3 ), C, 71.99; -1 6.6-7.4(m, 4, aromatic hydrogens). Table I: Reactions of Organolithium Compounds with Methylene Ketone G-7 ~ mmoles mmoles mmole s mmole s G-7 Sn cmpd C6H 5 Li CuI _...,_ Total Solvent Volume {ml) T C6H5Li & Yield {T CuI&G- 7 Quenching 1, 4 Benzyl Addn~ Agent Addn (%) Yield 1 , 4 Phenyl Addn {%) -~ 1. 31 3.84a 3.12 o.o 36c -35° (-50°) o.63 l.92a 1. 5 7 o.o 3ld -35 1. 69 5.36a ·4.37 2,65 65c -30 1. 06 a 3.18 2.62 1. 5 9 54c -35° {-20°) Ac. 20 o.63 1.90 1. 5 8 0.96 54c -35° (-20°) 0.93 2.78a 2.29 1. 39 57c -60 a b ~ 0 0 0 0 0 {-35 ) 0 (-25 ) NH 4 c1f 30 {19)i not detd NH4 Ctf 42 {56)i not detd NH4Clf 45 not detd f 46k 26k Ac og 2 47k 13k h 5lk 2lk 0 Ac 2o 0 h (-20 ) Ac 2 o 51 25 of 31 50 36 {2l)j 12 2.09 1. 28 57c -30 2. 54 1. 27 74c 00 (00) Ac 0.82 o.o 30c -35° (-20°) Ac 2 o f o.o 2.47 1. 27 40c -20 Ac 2o f 0.85 o.o 1. 70 o.o 48c -15 Ac 2 o f 0.85 2. 5 3 2.04 1.27 54e -35° (-20°) Ac 2o f 0. 85 2.57 0.84 3.40b 0.40 1.20 0.85 b b 0 (-20 ) 0 {-20 ) 0 00 2 89 32 (23)j 0 0 benzyltriphenyltin. b) IE-methoxybenzyltriphenyltin. c) anhydrous ether. d) anhydrous glyme. e) anhydrous THF. f) quenching agent added at temperature of £,;J.. addition. g) quenching agent added at room temperature. h) quencher added at reaction mixture reflux and stirred 1 hr. i) recovered starting material. j) enol acetate of......_..,. G-7 . k) value obtained from integration of ketal and C-8a methyl nmr resonances of mixture. a) *'" 85 6, (J-Ethyl e n c dioxy-1-(2 '-:IE-1ne thoxyphenylethyl)-8a13-methyl3,4,4aa ,5,6,7,8,8a-octahydro-2-na A. A solution of 0. 668 g (4. 25 mmole s) :IE-methoxybenzyl chloride in 4 ml dry ether was added to a stirred suspension in a nitrogen atmosphere of 0 .124 g (5 .10 mmole s) rragne sium turnings in 10 ml ether over O. 25 hr. Afte r stirring at room temperature for 1 hr, the solution was cooled to o0 and 52 mg (O. 26 mmoles) cupric acetate monohydrate (recrystallized from acetic acid) added in one lot, whereupon the mixture was stirred for 6 min. Then 200 mg (O. 85 mmoles) ketone G-7 in 9 ml dry ether """""" were added over 10 min. A transient orange color was visible as a drop of the enone solution came in contact with the Grignard surface. After the addition was comple te, the reaction mixture was stirred at room temperature for 1 hr and then quenched with 4 ml (43 mmoles) acetic anhydride. The quenching was exothermic and produced a heavy precipitate. After 45 min the reaction was diluted with water and worked up as usual with ether. Drying and concentration afforded a clear oil that on purification by ptlc {25% ether-petroleum ether, double development) afforded 231 mg (68o/o) pure enol acetate ,,..,,...,... XL, mp 69-72°, as shown by a comparison of ir and nmr spectra with rraterial described in an earlier experiment. 86 B. With no copper {II) acetate. --As described in part A, a ~~ solution of 12:-methoxybenzyl m agnesium chloride was prepared from 1. 336 g (8. 50 mmoles) 12:-methoxybenzyl chloride in 8 ml dry ethe r and 0.248 g (10.20 mmoles) magnesium 0 The solution was cooled to o ether . turnings in 20 ml dry and 400 mg (1. 70 mmoles) ketone G-7in12 ml dry ether added over 20 min. After 1 hr at room """""""' temperature, 4 ml (43 mmoles) acetic anhydride were added and the mixture stirred for 45 min. A normal ether workup followed by drying (MgS04) and concentration afforded a light yellow oil that produced on purification as in part A 497 mg (7 3%) enol acetate .,...,..,.... XL, 0 mp 69- 72 • 3, 4, 4acx , 4b, 5, 6, 7, 8 , Saa , 9, 10, lOa-Dodecah:>::_dro -7, 7-ethx_l ened~ ~~· l-(2 '-m-methoxyphenylethyl)-4b(3 , 8, 8, 10at3 -tetramethyl- 2-phenanthryl ~~· ~A. ~--A solution of 12:-methoxybenzyl- magnesium chloride from 272 mg (1. 75 mmoles) 12:-methoxybenzyl chloride in 4 ml dry ether and 50.5 mg (2.08 mmoles) magnesium turnings in 10 ml dry ether was prepared as in the previous experiment. 0 After cooling the stirred solution to o , 20. 8 mg (0.104 mmoles) cupric acetate monohydrate {recrystallized from acetic acid) was added and in 7 min 171 mg (0.52 mmoles) tricyclic enone F-7 in 10 ml dry ~ ether added over 11 min. The solution was stirred at room temperature 87 for 1 hr a nd then quenched with 4 ml (43 mmoles) acetic anhydride and the n stirred at room temperatur e for O. 75 hr. A normal ether workup followed by drying (MgS04) and concentration produced 613 mg oil. Purification of this material by ptlc (20% ether-petroleum ether, double development) afforded 189 mg oil that deposited 181 mg (71%) white solid, mp 104-106°, on trituration with cold !:-hexane. Recrystal- li zation from :i:__-hexane yielded th e analytical sample, mp 109-lll 0 • ir (CHCI 3 ) 1740 cm -I (acetate C= 0), 1675 cm -l (enol C =C), 1600, 1580, 1485 cm -l (aromatic), 1380 cm -l (methyl), 1180-1040 cm -l {ketal) ; nmr (CDCI 3 ) {) 0.86 (s, C-4b -CH ), o 0.90 (s, C-8 -CH 3 ), 3 o0.94(s, C-8-CH 3 ),o l.03{s, C-10b-CH 3 ), 6 2.IO(s, 3, -COCH 3 ), o 3.78 (s, 3, -OCH 3 ), O 3.93 (s, 4, ketal), 6 6.6-7.4 (m, 4, aromatic hydrogens). Anal. Calcd for c 31H 44o 5 : C, 74.96; H, 8.94. Found: C, 75.ll; H, 9.03. benzylmagne sium chloride was prepared as in part A above from 441 mg (2. 81 mmoles) JE-methoxybenzyl chloride in 5 ml dry ether and 82 mg (3.36 mmoles) magnesium turnings in 10 ml dry ether. After cooling to 0°, 187 mg (0.56 mmole } tricyclic enone F-7 in 15 ml ~ dry ether was added over 0.25 hr. A transient yellow color was noted when a drop of enone solution hit the Grignard surface. After 1 hr 88 at room ten1pcraturc, the r eac tion was quenc h e d by the addition of 5 ml (54 n\mole s) acetic anhydride and stirring r es ulted in he a vy pr ec ipitate for 0.5 hr. A normal ether workup followed by drying {MgSO 4 ) and concentration produced a yellow oil which on purification by ptlc as in part A gave 209 mg {74%) of white tricyclic enol acetate 0 I-1, mp 104-106 • ,,...,..,..,... --- Table II: Reactions of Organomagnesiuin Compounds with Methylene Ketones G-7 &: F-7 """"""' mmoles Ketone 0.85 o. 85 a a mmoles Chloride mmoles Magnesium mmoles Cu(OAc)2·H20 Temp of Ketone Addn 4.25c 5.10 0.26 5.06 4.25 a 0.85 0.85 1. 70 4.25 a 4.25 a 0.52 o.56 b b Total Solvent Vol. (ml) Yieldl,4 Product (%) 00 NH4 Cl 25 4le 0.26 00 CH 3 I 27 37e 5 .10 0.26 00 Ac 2 o 26 70.5 5 .10 o.o 00 Ac 0 2 25 71 5 .10 0.26 00 Ac20 23 68 10. 20 o.o 00 Ac20 40 73 2.08 0 .104 00 Ac 20 30 71 3.36 o.o 00 Ac 2o 30 74 c 4.25 a 0.85 c Quenching Agent c d 8. sod 1. 75 2. 81 d d a) G-7. ,,...,,.,...,.. b) F-7. c) benzyl chloride d) m- methoxybenzylchloride e) product had alcohol ir absorption after extensive purification. ~ 00 '° 90 dro-2 lH)-na hthalenone H-2). - -In an adaptation of th e procedure of House and Trost (69), 93 mg (O. 23 mmole ) enol acetate XL in O. 8 ml glyme {doubly distilled from LAH) """""' was added to a solution under a nitrogen atmosphere of 0. 27 ml methyl lithium solution (1. 7 N in ether, 0. 46 mmole ) in 1. 2 ml anhydrous glyme. After stirring 0.5 hr at room tempe rature, 1.0 ml {16.1 rnnoles) methyl iodide, which had been freshly distilled from calcium chloride, was added and the solution stirred for 5 min before working up with ether. Drying (MgS0 4 ) and concentration afforded 85 mg (99%) yellow oil which deposited 66 mg (78o/o) white crystals, mp 630 68 , from cold !:-hexane. Recrystallization from ether-!:-hexane 0 gave the analytical sample, mp 69-70 • ir (CHC1 ) 1700 cm 3 -1 (C= 0), 1600, 1580, 1490 cm-l (aromatic ), 1380 cm -l (-CH 3 ) 1150-1040 cm-l (keta l); nmr (CDC1 3 ) 0 0.75 (s, 3, C-8a -CH3), 0 1.30 (s, 3, C-1 -CH 3 ), O 3.81 (s, 3, -OCH3), 0 3.95 (s, 4, ketal), O 6.6-7.3 {m, 4, aromatic hydrogens); nmr {C6H6) O 0.50 {C-8a -CH 3 ), 6 1.00 {C-1 -CH3). Anal. Calcd for c 23 H 32 04: C, 74.26; H, 8.67. C, 74.16; H, 8.66. Found: 91 (2'-IE-methoxyphenylethyl)-2, 6(1H, 5H)-naphthalenedione (H-4). -A solution of 56 mg {0.15 mmoles) ketone H-2 in 5 ml reagent acetone ~ and 1 ml 10% hydrochloric acid was allowed to stand O. 75 hr at room temperature and then worked up as usual with ether. Drying (Na 2 so 4 ) and concentration yielded 47 mg (96%) of material that crystallized, mp 111-114°, on trituration with ether. ~-hexa ne 1710 cm Recrystallization from acetone0 afforded the analytical sample, mp 127. 5 -128. 5 • -1 {C==O), 1600, 1585, 1490 cm 1040 cm-l (me thoxyl); -1 ir (CHC1 ) 3 {aromatics), 1380 cm- 1 (-CH 3 ), nmr (CDCl3) 0 0.95 (s, 3 , C-8a -CH 3), 0 1. 28 (s, 3, C-1 -CH3}, O 4.63 (s, 3, -OCH 3 ), 0 6.6-7.3 {m, 4, aromatic hydrogens}. Anal. Calcd for c 21 H 28o 3 : C, 76. 77; C, 76. 79; H, 8. 59. Found: H, 8.48. 2{1H}-chrysenone (H-5). - -Into a test tube reaction flask equipped • with a nitrogen inlet and a s c r e w-type m echanical stirrer blade was placed 28 mg (0. 085 mmoles) dione H-4. To this was added 15 ml ~ of a solution of polyphosphoric acid from 30 ml 85% phosphoric acid and 34 g phosphorous pentoxide. The thick solution was vigorously sti rre d a t 65-70° for 40 min and the n the mixture poured onto i ce and worked up with ether. Drying (MgSO 4 ) and conce ntr ati on yielded 92 28 mg (105%) light brown solid, mp 185-189°, that was >97% pure by vpc (oven temperature 290°, rt 2.4 min). Two recrystallizations from acetone produced colorless cubes as the analytical sample, mp 192-194°. ir (CHCI 3 ) 1710 cm- 1 (G:0), 1640 cm- 1 (styrene C=C), 1610, 1570, 1495 cm- 1 (aromatics), 1375 cm -I (methyls), 1035 cm -I (methoxyl); nmr (CDCI 3 ) {j 1. 00 (s, 3, C-4a -CH ), {j 1. 08 (s, 3, 3 C-4b -CH3), o 3. 78 (s, 3, -OCH 3), 6 6.15 (t, I, ~-5 Hz, C-11 vinyl hydrogen), {j 6.61 (s, l, C-7 aromatic hydrogen), O 6.73 (dd, 1, J~ 9 and 3 Hz, C-9 aromatic hydrogen), {j 7. 56 (d, I, ~=9Hz, C-10 aromatic hydrogen). Anal. Calcd for c 21 H 26 o 2 : C, 81. 31; C, 81.25; H, 8.44. Found: H, 8.45. When the cyclodehydration of ketone H-2 was attempted in toluene """""""' with £_-toluenesulfonic acid, exte nsive decomposition occurred. phenanthrone (I-3). --Following the procedure for ketone H-2, 71 mg ~ ~ {O .14 mmole) tricyclic enol acetate I-1 in I. 0 ml dry glyme (doubly """""""' distilled from LAH) was added to 0 .17 ml methyl lithium solution (1. 7 Min ether, 0. 29 mmoles) in 1. 0 ml dry glyme at room temperature under a nitrogen atmosphere. After stirring for 0. 5 hr, 1. 0 ml (16.1 mmoles) methyl iodide freshly distilled from calcium chloride 93 was added and the solution stirred for 5 min. A normal ether workup followed by drying (MgS04) and concentration afforded 66 mg oil which crystallized on trituration with cold ~-hexane to give 51 mg 0 (76%) white solid, mp 116-120 • Recrystallization from acetone-~0 hexane produced the analytical sample, mp 124-124. 5 • ir (CHC1 3 ) 1698 cm -l (C =0), 1600, 1585, 1485 cm -l (aromatic), 1380 cm -l (-CH ), 3 -1 1150-1050 cm (ketal); nmr (CDC1 3 ) o 0.83 (s, 3, -CH 3 ), 6 0.87 (s, 3, -CH3) , 0 0.92 (s, 6,2-CH3's), o 1.28 (s, 3, C-1 -CH3), 6 3.82 (s, 3, -OCH 3 ), O 3.98 (s, 4, ketal), o 6.6-7.3 (m, 4, aromatic hydrogens); nmr (C6H 6 ) o O. 72 (s, 6, C-1 and additional -CH 3 s), 1 o 0.92 (s, 3, - CH 3), o 1.01 (s, 6, 2-CH3 s). 1 Anal Calcd for C30H3404: C, 76. 87; H, 9. 4. Found: C, 76.98; H, 9.52. 3 , 4, 4aa, 4b, 5, 6, Baa, 9, 10, lOa-Decahydro-1(3 -(m-methoxyphenylethyl)la, 4b f3, 8, 8, lOa (3-pentamethyl-2, 7(1H, 8H) -phenanthrenedione (I-4). - A solution of 30 mg (0. 065 mmole) tricyclic k etone _....,.... I- 3 in 3 ml reagent acetone and o. 8 ml 10% hydrochloric acid was allowed to stand for O. 75 hr. An ether workup followed by drying (MgS0 4 ) and concentration yielded 27 mg (100%) white solid. Recrystallization from a cet one-~0 hexane gave 25 mg (89°/o) white needles, mp 158-159 • ir {CHC1 ) 3 1700 cm-l (C=O), 1600, 1580, 1485 cm-l (aromatics), 1380 cm-l (methyls), 1040 cm- 1 (methoxyl); nmr (CDC1 3 ) <Y 0.87 (s, 3, C-lOa -CH 3 ) o 0.97 (s, 3, C-4b -CH3), o 1.03 (s, 3, C-8 -CH3 ), o 1.10 (s, 3, C-8 -CH 3 ), o 1.30 (s, 3, C - 1 -CH 3 ), o 3.82 (s, 3, -OCH 3 ), 94 o 6 .6-7.4 (m, 4, aromatic hydrogens); nmr (C6H6 ) O 0.60 (s, 3 , - C H 3 ), O 0. 63 (s, 3 , -CH 3 ), O 0.90 (s, 3, -CH 3 ), (s, 3, -CH ), 3 o 0.97 o 1.05 (s, 3 , C-1 -CH 3 ). Ana l. Calcd for C 28 H 30 o 3 : C, 79.10 ; H, 9.43. C, 79.19; H, 9. 50. Found: 95 References 1. J. Simonsen and W. C. J. Ross, "The Terpenes," University Press, Cambridge, Vol. IV, 1957, p. 287. 2. K. Schmiege!, Ph.D. Thesis, Stanford University, 1962, p. 61. 3. A. R. Pinder, "The Chemistry of the Terpenes, 11 John Wiley and Sons, Inc., New York, 1960, p. 176. 4. B. Bischof, O. Jeger 1911 (1949). 5. T. G. Halsall and R. Alpin, Forschr. Chern. Org. Naturstoffe, 22, 153 (1964) . 6. -- and L. Ruzicka, Helv. Chim. Acta, 32, -- D. E. White, Rev. Pure~· Chem.,~, 191 {1966). 7. P. de Mayo, "The Higher Terpenoids, 11 Interscience, New York, 1959. 8. J. A. Barltrop and N. A. J. Rogers, Prog. in Org. Chem., 5, .96 (1961). 9. E. J. Corey and W. E. Russey, J. Am. Chern. Soc., 88, 4751 .(19 66). 10. E. J. Corey, W. E. Russey and P.R. Ortiz de Montellano, J. Am. Chem. Soc., 88, 4751 {1966). -- -- .- 11. E. J. Corey and P.R. Ortiz de Montellano, J. Am. Chem. Soc., 89, 3362 (1967) • .,..,..,,... 12. E. E. van Tamelen, J. D. Willet, R. B. Clayton J. Am. Chem. Soc., 88, 4752 (1966). -- -- and K. E. Lord, ,,_ 13. J. H. Richards and J. B. Hendrickson, "The Biosynthesis of Steroids, Terpenes, and Acetogenins," W. A. Benjamin, Inc., New York, 1964, pp. 257-288. 14. J. B. Pridham, flTerpenoids in Plants," Academic Press, London, (1967). 15. J. B. Hendrickson, "The Molecules of Nature," W. A. Benjamin, Inc., New York, 1965 , pp. 103-142. 16. D. Evans, Ph.D. Thesis, California Institute of Technology, 1967, pp. 6-17. 17 . T. G. Halsall and D. B. 18. E. J. Corey and R.R. Sauers, J. Am. Chem. Soc., 81, 1739 .(195 9). 19. D. H . R . Barton and K. H. Overton,~· Chem . Soc., 2639 (19 55). 20. G. Brownlie, M. B. E. Fayez, F. S. Spring, R. Stevenson and W. S. Strachan, ~· Chem. Soc., 1377 {195 6). 21. D. H. R. Barton, E. F. Lier 1031 (1968). 22. P. Beak and B. M. Monroe,~· Org. 23. J.E. Dolfini, Ph.D. Thesis, University of Michigan, 1961. 24. Schmiege!, p. 69. 25. M. Brown, Ph.D. Thesis, Stanford University, 1962. 26. N. L. Wendler, H. L. Slates Soc., 73, 3816 (1951). -- Thomas,~· Chem. Soc., 2431 (1956). andJ. F. McGhie, J. Chem. Soc., Chem.,~ 2778 {19 67) . and M. Tishler, J. Am. Chem. ,,..,,...,.. 27. G. Stork, P. Rosen and N. L. Goldman, J. Am. Chem. Soc., 83, 2 965 (1961). .- 28. N. Danieli, Y. Mazur and F. Sondheimer, Tetrahedron, 23, .509 (1967). 29. L. Crombie and S. H. Harper, ~· Chem. Soc., 1707 (1950) . 30. L. Crombie and S. H. Harper, ibid. , 1714 (1950). 31. M. Brown, p. 54. 97 32. 33. S. Archer, W. B. Dickinson and M. J. Unser, J. Org. Chem., 22, 92 (1957). -- R. E. Ireland, unpublished results. 34. M. Brown, p. 23. 35. H. c. BrownandR. F. McFarlin, J. Am. Chem. Soc., (1956). 36. D. H. R. Barton, A. Da. S. Campus-Neves and R. C. Cookson, J. Chem. Soc. , 3500 (195 6). 37. C. Frisell and S. -0. Lawesson, Arkiv. 401 (1961). 38. R. F. Church, R. E. Ireland and D.R. Shridhar, J. Org. Chem., 27, 707 (1962). 39. ll• 252 for Kemi, Band 17, -- A,. NickonandJ. F. Bagli, J. Am. Chem. Soc.,JU., 1498 (1961). 40. C. S. Foote, ACR, 1, 104 (1968). 41. D.R. Kearns, R. A. Hollins, A. U. Khan, R. W. Chambers and P. Radlick, J. Am. Chem. Soc., 5455 (1967). 42. J.C. Collins, W.W. Hess and F. J. Frank, Tetrahedron Letters, 3 363 (1968). 43. G. I. Foos, G. E. Arth, R. E. Beyler and L. H. Sarett, J. Am. Chem. Soc., 75, 421 (1955). ---... !tt• -- .....,.., 44. The synthesis of G-7 was completed jointly with, and using the procedure of, Dr~ R. Marshall. 45. J. E. Dolfini, p. 41. 46. -- P. S. Wharton, G. A. Hiegel and R. V. Coombs, J. Org. Chem., 28, 3217 (1963). 98 47. M. s. Kharasch and O. Reinmuth, "Grignard Reactions of Nonmetallic Substances, 11 Pr entice Hall, Inc., 1954, pp. 196-239. 48 . A. W. Burgstahler and M. Struble, !!_. Chem. Soc., 896 {1960). 49. H. O. House and W. F. Fischer, Jr.,!!_. Org. Chem., {19 68). 50. H. O. House, W. L. Respess and G. M. Whitesides, J. Org. Chem., 31, 3128 {1966) • ll· 949 .- 51. 52. H. O. House, D. D. Traficante and R. A. Evans, J. Org. Chem., 28' 348 {1963). -- ---- H. O. House and D. D. Traficante, ibid., 28, 355 (1963). -- 5 3. H. 0. House and H. W. Thompson, ibid. , 28, 360 (1963). 54. H. O. House a nd W. L. Respess, ibid. , 30, 301 (1965). 55. S. K. Malhotra and F. Johnson, J. Am. Chem. Soc., ,,..,,....,.. 87, 5493 (1965). 56. H. GilmanandR. H. Kirby,!!_. Am. Chem. Soc., 63, 2046 (1940) • .......... -- -- -- _ -- 57. G. Wittig, F. J. Meyer and G. Lange, Ann., 571, 167 (1951). 5 8. A. Luttringhaus, Jr., Ber. , 67, 1602 (19 34). 59. H. Gilman, H. A. Pacevitz and O. Baine, J. Am. Chem. Soc., 62, 1514 (1940) • .......... 60. H. Gilman, H. McNinchandD. Wittenberg, J. Org. Chem., 23' 2044 (1958). 61. -- """"- -- H. Gilman and S. Rosenberg, .:!_. Org. Chem. , ,,3.i• 2063 (1959). 62. H. Gilman and A.H. Haubein, J. Am. Chem. Soc., 66, 1515 """""" (1944). 63. a. J. J. Bloomfield, Tetrahedron Lette rs, 587 {1968). b. J. J. Bloomfie ld , ibid. , 5 91 (1968). 99 -- 64. G. StorkandP. F. Hudrlik, J. Am. Chem. Soc., 90, 4462 (1968). 65. G. Stork and P. F. Hudrlik, ibid., 90, 4464 (1968). 66. A study of the dimerization of a-methylene cyclohexanone is reported in C. Mannich, Ber., 74, 557 (1941). -- - - ,,...,.. - - .,_ 67. T. Holm, Tetrahedron L e tters, 3329 (1966). 68. H. O. House and J.E. Oliver , ~· Org. Chem.,~ 929 {1968). 69. H. O. House and B. Trost, ~· Org. Chem.,~ 2502 (1965). 70. N. S. Bhacca and D. H. Williams, 11 Applications of NMR Spectroscopy in Organic Chemistry, 11 Holden-Day, Inc., 1964, pp. 159-176. 71. P. C. Cherry, W. R. T. Cottrel, G. D. Meakins and E. E. Richards, J. Chem. Soc., 181 (1967). 72. J. Ronayne and D. H. Williams, Chem. Commun., 712 (1966). 73. R. F. ZurcheJ; Helv. Chim. Acta , ~ 2054 (1963). 74. D. A. Evans, p. 122. 75. G. H. Douglas, J.M. H. Grave s, D. Hartley, G. A. Hughes, B. J. McLoughlin, J. Siddall, H. Smith , J. Chem. Soc., 5072 {1963). 76. The degradation procedure to keto ester J-5 was established by R. J. Crawford; the synthetic sequence ~one J -8 was worked out by W. S. Johnson and coworkers. ~ 77. This material was used as supplied by Hi-Labs, Whitmore L ake, Michigan, without purification. 78. E. M. McMahon, J. N. Roper, Jr., W. P. Utermohlen, J r . , R.H. Hasek, R. C. Harris andJ. H. Brant, J. Am. Chem. Soc. , 70, 2971 {1948). -- 100 79. a. This di one was purchas ed fro1n Hi-Labs, Whitmore Lake, Michigan. Recrystalli7.ation from 95% ethanol produced y e llow solid, mp 205-208° [tit (79b) mp 208-210°]. b. S. Hamac handran and M. S. Newman, Org. Syn., 41, 38 (1961). ,,........... 80. V. F. Kucherov and A. Gurvi.ch, J. Gen. Chem. USSR, 31, 731 """""' (1961). 81. Y. Kitahara, A. Yoshikoshi and S. Oida, Tetrahedron Letters, 1763 (1964). 82. This sample, prepared by K. Bell and R. E. Ireland, had a melting point of 63-64°. 83. M. Brown, p. 7 3. 84. G. Rasmusson, Research Report to W. S. Johnson, Stanford University, 1964, p. IO. 85. A. Streitweiser, Jr. andW. C. Langworthy, J. Am. Chem. Soc., 8 5 , l 7 5 7 (196 3) • """"""' 86. R. Pschorr, Ann. , ,,.......... 391, 43 (J 912). 87. I. M. Downie, J. B. Holmes and J. B. L ee , Chem. and Ind., 900 (1966). 88. D. Seyferth and M.A. W e iner, Org. Syn., 41, 30 (1961). 89. H. GilmanandS. D. Rosenberg, J. Am. Chem. Soc., 74, 531 - -"""""' (1952). """"""' -- 101 Propositions 102 Proposition 1 A synthesis of the sesquiterpene cyclosativene from (-)- carvomenthone is proposed. Cyclosativene i is a recently characterized tetracyclic sesqui~ -- terpene (1) related to sativene ii in the same way that longicyclene iii .......... is related to longifolene (2). Despite its complex molecular architecture , -iii iv .......... several synthetic routes to i are available . """' One of these, the decom- -- position of either tosyl hydrazone v or diazo ketone vi, followed .......... by intramolecular addition of the resulting carbene to the double bond, 103 v ,,..,.. Vl ,,..,.. has the advantage of easy accessibility of the interme diates ,,..,.. v and/or -- vi and considerable literature precedent for the carbene addition step. Intramolecular carbene additions are well documented reactions (3), although the yields are often variable. Buchi and co-workers -- {3b) a cc omplishe d the total synthesis of thujopsine vii by photolysis of the sodium salt of tosyl hydrazone viii, although in this case th e major ,,..,,..,,... product was cyclopropene ,,..,.. ix. ) -vii viii ~ -ix Relative to the present problem , Doering {3c) has prepared nortricyclonone ,,..,.. xi by the copper catalyzed d ecompo s ition of diazo ketone,,..,.. x. This last reaction provides strong indication that diazo 104 0 ~ Cu/ A 65% > x Q Xl -.. -.. k e tone vi will react as planne d to produce cyclosative non e x ii. -.. ~ In fact, an examination of molecular models reveals that vi is better -.. > vi xii ~ ~ situate d for reaction than is x in that the carbe n e from th e former is -.. 0 a full 0. 8 A closer to its double bond than is the c a rbe ne from x 0 0 (2.5 A versus 3.3 A). In addition the carben e from v i i s better -.. alligne d to interact with the 1l cloud of its double bond . Proposed syn theses of tosyl hydra zone v and diazo ketone vi are outlined in -.. Chart A. -.. The alkylation of (-)-carvomenthone A-1 (4 , 5) with 2~ bromopropionaldehyde (6a) or 2-chloropropionalde hyd e ( 6 b) is 105 expected to giv e th e axially alkyl a t e d product _..._ A-2 (7). -- Acid catalyzed --- a Idol condensation of A- 2 to afford the bi eye lie keto ole fin A- 3 is anticipate d on the basis of work by Corey (8) and Julia (9) . The alternative cyclization l e ading to a cyclopropane d e rivative is not expec ted to be competitive. Reduction of the carbonyl group of A- 3 ~ with lithium aluminum hydride giving the equatorial a-alcohol (10) followed by bromination with triphe nyl phosphine dibromide (ll) w ill give the axial t3 -bromide A-4. ~ Displac e m e nt of th e bromine with --- sodium malonalde hyde should afford dialdehyde A-5, which on mono . d cca rbonylation and tr e atm ent with tosyl hydrazine will give the tosyl hydrazon e ,,,_ v. Alte rn.atively, treatment of ketone A- 3 with methoxymethylene ~ triphenylphosphorane (12) and hydrolysis of the resulting e nol ether -- should produce the equatorial aldehyde A-6. -- Oxidation to acid A- 7 followed by normal diazotization of the corresponding acid chloride will afford diazo ketone vi • ......... The decomposition of ,,,_ v either thermally or photolyti cally (13) is e xpected to produce cyclosativene ,...,. i directly. The therm a l decom- position of diazo ketone ,,,_ vi should produce cyclo s a tivenone ~which can be converted to cyclosativene by normal reductive procedures. 106 OH~CH 3 ,....~ 1 NaCH(CHO)z HOz -CO NHzNHTs vi ..-- > 107 References I. L. Smedman and E. Zavarin, Tetrahedron Letters, 3833 (1968) . 2. U. R. Nayak and S. Dev, ibid., 243 (1963). 3. a. G. Stork and J. Ficini, .:!_.Am. Chem . Soc., 83, 4678 (19 61). b. G. Buchi and J. D. White, ibid., 86, 2884 (19 64). c. W. von E. Doering, E.T. Fossel and R. L. Kaye, Tetrahedron, 21, 2S {1965). -- -- -- """" 4. The absolute configuration of cyclosativene i has not been estab lished. Longicyclene .-.... iii and longifolene --.. iv have the same absolute configuration (Sa). Sativene ii has the opposite configuration (Sb) and in all probability cyciosativene i does also. Therefore a synthesis of cyclosativene starting with T-)- carvomenthone A-1 should afford product of natural c onfiguration. .........,_, S. a. U. R. Nayak and S. D ev, Tetrahedron, 24, 4099 (19 68). b. P. De M ayo and R. E. Williams, J. Am. Chem. Soc., 87, -3275 (196S). a. A. M. Pujo and J. Boileu, Mem. poudres, 37, 3S (19SS) . b. C. R. Dick, J. Org. Chem ., ]:1 6. -- 1 -- -- 272 (1962). 7. R. Howe and F. J. Mc Quillin, .:!_. Chem . Soc. , 1194 (l 9S 8). 8. E. J. Corey and S. Nozoe, J. Am. Chem. Soc., 87, S728 (19 65). 9. S. Julia, Bull. soc. chim. France, 21, 780 (19 54). 10. E. L. Eliel and M. N. Rerick, J. Am. Chem. Soc., 82, 1367 (1960). 11. J.P. Schaefer and J. Higgins , !!_. Org. Chem., 32, S53 (19 67). 12. R. Greenwald, M. Chaykovsky, and E. J. Corey, ibid., 28, 1128 (196 3). 13. W. Kirmse, ''Carbene Chemistry," Academic Press, New York (1964) pp. 47-Sl. ---- --- - - -- --- """" ---- -- 108 Proposition 2 Expe riments to unambiguously determine the biosynthetic origin of the non-tryptophan portion of strychnine are proposed. Many theories have been advanced to explain the origin of the non-tryptophan derived fragment of the in.dole alkaloids (heavy lines in formulae for vindoline.,,...,.. i, catharanthine .,,...,.. ii, and strychnine .,,...,.. iii) • Woodward's proposal (1), based on the Barger-Hahn scheme (2), i .,,...,.. ii .,,...,.. iii _...,..., 109 that strychnine is derivable from cleaved dihydroxyphenylalanine has been discredited because of its inability to explain pertinent data. Wenke rt and Bringi 's theory implicating prephenic acid ( 3), and thus shikimic acid, has been proven incorrect by the sole incorporation of shikimic acid into the tryptophan portion of several indole alkaloids (4). A third theory that the C9_ 10 residue is the combination of three acetate units, a malonate unit, and formate received early experimental support (5) but has since been shown to be in error (5). The suggestion, however, that a terpenoid precursor is involved (7) has r eceived experimental verification in all of the alkaloids studied (8). In general this route involves the formation of a cyclo- pentanoid precursor through the successive inte rmediacy of acetate , tn:!valonate, and geranyl alcohol. Subsequent fragmentation and rearrangement in one of several ways leads to the pieces necessary for a lkaloid incorporation. Recently the intermediacy of loganin j_)( . OH 3CH3COO > \COH > CtzO-H-->~ ~OH--)_,,. for ii """' 110 ,,,OGluc -iv in alkaloid biosynthesis has been demonstrated (9), and other natural iridoids may also be involved. Although loganin has been shown to be -- derivable from ge raniol, the possibility that iv may be a de gradation product of higher terpenoids has been suggested (10). Several possibilities exist for the biosynthesis of strychnine. The non-tryptophan portion of this molecule consists of eleven carbons. Th is is from one to two more carbons than the other alkaloids considered and presents the problem of from where the extra carbon derived. The most logical answer, a priori, involves a one carbon residue, perhaps formate, which joins tryptamine and the terpenoid group in one of several possible ways (A- C). A. ~ ---iii 111 - B. c. - - -7 .iii - - - - - 7 .111 - The ten c arbon r e sidues might b e de r ive d from a cyc lopentanoid inte rm e diate as follows: A. > B. > c. > 112 Some expe rimental evidence for the first of these paths has been obtained by the re search group of Dr. Richards he re at Caltech, who noted an 18 ± 4% P cJ 4 -formate incorporation at C-21 in strychnine isolated from the seeds of~ vomica (11). It is proposed, however, that none of paths A-C is the origin of the eleven carbons, but rather that they arise from a c one acetate combined in the following manner. 9 unit plus The circled carbon is lost at some unspecified point during the biosynthesis. D. ---~ i ii ~ There are several reasons for preferring route D. The first is that it would be consistent with studies on other alkaloids in excluding a formate pool (5). economy. The second and most convincing reason is based on The existence of such alkaloids (12) as compactinervine v, .- echitamidine .vi, stemmaderine vii, .- and aspidospermatine viii ......_ is strong evidence that strychnine is derived as indicated in path D . These compounds have a terpenoid fragment with none, one, or both isopropyl carbons of the monoterpene missing, and also show both 113 OH C02CH3 v vi .- """" --- vii viii """" acetylated and unacctylated indole nitrogen. For strychnine this is best explained by the process shown with the circled methyl lost, probably as carbon dioxide. A recent report (13) that C-22 and C-23 of strychnine are derived from acetate of course does not differentiate between paths A, B, and D. D. > 114 If one is to accept this proposal, he must be able to expl ain the incorporation of formate at C-21 • One possibility is that a small portion of formate is converted to acetate in the plant. It is this In support de rived acetate that gives rise to radioactive strychnine. of this note the ready conversion of formaldehyde to glycolic aldehyde (1 4). Loss and recombination of water results in acetate. Also of interest is the report that acetic acid is produced directly by the action of heat and pressure on fo rmaldehyde (15). It is, therefore, not unreasonable to expect that a biological medium could produce acetate from fo rmate or formaldehyde. If this is the case, the expected activity at each carbon in the non-tryptophan portion of strychnine based on complete randomization (9%) is close to the lower limit of that observed. Furthermore , feeding experiments using 1- [ 14 c] - acetate pre dict 0% activity at C-17 for schemes A, B, and D, and 20% for scheme C. The experimental value of 0% eliminates path C from consideration. In addition, the Richards 1 group discovered 19 ::!:: 4% activity at C-22, a result in accord with only path D. Two experiments are proposed to unambiguously demonstrate the validity of path D. 3 -methyl- The first one invol ves feeding experiments with Q4 c] -geraniol ix or 4-methyl- °Q 4 c] -loganin x. The """"' """"' conversion of radioac tive geraniol to loganin in useable quantities 115 has been reported {16). Activity in strychnine at C-18 is predicted for paths C and D, and at C-22 for paths A and B. <?,H ,.,cH 3 '··---" ..OGlu ix ,,,.,_,. x """' A second experiment employs the feeding of 8labled loganin ,,,.,_,. xii. xii ,,,.,_,. Xl ,,,.,_,. [1 4 c] -geraniol~ or Path C predicts one half of the activity at C-23 and path D all of the activity at C-17. All of these carbons are accessible by degradative means, utilizing the information gained during the structure elucidation. C-22, C-23, C-17 (17) amyl NO iii """' SOC1 2 NaOCz~ ) N C-22 C-23 H 0 2 ) \ - - - - ) + C02 HCN OH ..__ C-17 116 C-18, C-19 (18) iii .,._.. KMn04 > 18 1N Na°> C-19 19 HOCHzCOzH ) 104 > HCOOH C-18 J COz 117 References -- 1. R. B. Woodward, Nature, 162, 155 (194 8). 2. a. G. Barger and C. Scholz, Helv. Chim. Acta.,~ 1343 (19 33). b. G. Hahn and H. Werner, Liebigs Ann., ~· 123 (19 31) . 3. E. Wenkert and N. V. Bringi, ~·Am. Chem. Soc., 81, 1474 (19 59) . 4. D. Groger, K. Stolle, K. Mathe s, Zeits. Fur Naturforsch, 3, 206. ~ 5. a. P. N. Edwards and E. Leete, Chem. and Ind., 1666 b. E. Leete, S. Ghosal, P. N. Edwards, J . Am. Chem. Soc., 84, 1068 (1962). (1961). ~ 6. 7. 8. a. A. R. Battersby, R. Binks, W. Lawrie, G. V. Parry, B. R. Webster, Proc. Chem. Soc., 369 (196 3). b. H. Goeggel and D. Arigoni, Experientia,~ 369 (1965 ). a. R. Thomas, Tet. L et ., 544 (1961). b. E. Wenkert, ~·Am. Chem. Soc., 84, 98 (1962) • a. A. R. Battersby, R. T. Brown, J. A. Knight, J. A. Martin, and A. O. Plunkett, Chem. Comm., 346 (19 66). b. P. Loew, H. Goeggel, and D. Arigoni, Chem. Comm., 347 (1966). c. E. S. Hall, F. McCapra, T. Money, K. Fukumoto, J. R. Hanson, B. S. Mootoo, G. T. Philips, and A. I. Scott , Chem. Comm., 348 (1966). d. T. Money, I. G. Wright, F. McCapra, and A. I. Scott, Proc. Nat. Ac ad. Sci. USA. , 5 3, 901 (1965). .......... - - -- --- - - - - e. -- F. McCapra, T. Money, A. I. Scott , and I. G. Wright , Chem . Comm., 537 (19 65 ). 118 9. £. H. G oeggel and D . Arigoni, Chem . Comm . , 538 (1965) . g. A . R. Batt ersby, R. T . B rown, R. S . Kupil , A. 0. Plunkett , and J. B . Taylor, Chem . Comm., 538 (1965 ) . h. E. L eete a n dS. U eda, T et. L et., 4915 (1966) . a. A . R . Battersby, R. S . Kapil , J . A. Martin and L . Mo, Chem . C omm . , 133 (19 68). b. P . L oew a n d D. Ar igoni, ibid ., 1 37 (1968). 10 . S. T . Murayama, P h . D . T h es i s , Cal ifornia Institute of Technology (19 69) p . 114. 11. Ch emist r y and Chemi cal E ngineering, 1966 Report, p . 107. 12. R . H . F . Manske, The A lkal oids, Vol. VIII, (Acade m ic Press, New York, 1965 ) pp . 457-68 . 13 . C h. Sch latter, E . E. Waldner, H. Schmid, D. Groger, K. Stolle a n d K . M oth es, Helv . Chim . Acta, 49, 1714 (1 966). - - -- - - .,._... 14 . J. F. Wal ker, F ormaldehyd e, (Reinhold, New York, 3rd ed., 1 9 6 4 ) p • 21 7 • 15 . S. J . Green and R . Handley, U.S . Patent# 1 ,950,027 (1934). 16. a. A . R. Batte r sby, R . S . Kapil and R . Southgate, Chem . Comm., 131 (1968) . b. S . B r echbu hler- B ader, C. J . Coscia , P . L oew, Ch. von Szczepanski a n d D. Arigoni, ibid., 136 (1968). a. H . Wie l and and W. Gum lichi , Ann., 494 , 191 (1932) . 17 . ~ 18. b. H. Wieland and K . Kaziro, Ann., ;&R, 60 (19 33) . a. H. L euchs and G. Schwaebel, Ber ., 46, 3693 {1913). b. H . L euchs and F. Krohnke , Ber ., 42 , 2498 {1909) . .- """' 119 Proposition 3 A study of th e synthetic utility of N-alkyl a lleneimines i is ......... proposed. A continuously e ncountered problem in organic chemistry is the formation of carbon-carbon bonds. Accordingly it is one of the mos t actively pursued areas of r e s earch. Re ce nt reports of new alkylating agents such as substituted chloroisoxazoles (1) as well as the nove l coupling r eactions between organic halides and '11-allyl ni c kel halide s (2) are illustrative. It is felt that the methylene aziridine nucleus i possesses ......... characteristics that would make valuabl e a study of its behavior under a vari e ty of carbon-carbon bond forming reactions. The syntheses CH~ z i,,N-R of i where R is a variety of alkyl groups i s conveniently acc omplished """' by means of the Gabriel aziridine synthesis ( 3). Consider first the r eaction of .....,... i (R= C H 3 ) with an a , t3 -unsaturated carbonyl c ompound under normal (4) e n amine alkylation c onditions. The expected imine ketone iv constitutes a one step formation of a ......... 120 functiona li ze d five -membered ring . Q The alternative produc t ,,_ v is not CH~ + ~N - CH 3 - CH 3 0 11 11 1 """"' """"""' - CH 3 N - CH 0 iv 3 v .......... """"' a nticip a ted o n the basis of it s hi ghly strained nature. note d that the ruptu re of a n a ziridine rirl g It should be to form the fiv e -me mbered rin g of an indole system has recently been r eported (5 ) . Na I In ad dition, > vi vii -- """"""' n uc leophi lic attack by a n enamine viii on anaziridine carboxyli c este r 121 ix leading to rupture of the aziridine ring has been claimed (6 ). ,,.,,.._ > ix ,,.,,.._ Vlll ~ Of perhaps g reater interest would be the reaction of N-alkyl aziridine s .,...,... i with e l ec trophilic olefins such as maleic anhydride or di butyl vinylb or onate xiii (7). The possibility of forming five - ~ rn=mbered ri ngs in such a fashion is interesting in that it is similar 0 0 Q 2'l: -R CH . + xi ,,.,,.._ cH + 2 ..........,... N-R i xii """" ~ D N- R > (Bu0 ) (BuO)z) Xlll > i """" µ N-R 2 XlV ,,.,,.._ 122 to the 1, 3-dipolar a dditions so well characterized by Huisgen (8). Although many dipoles containing N, 0, and C, or variations thereon, hav e been found effective in adding to double bonds, as yet no instance of an all carbon 1, 3-dipole has been reported. This probably arises because such a species would be electronically equivalent to an allylic carbene. Allylic carbene s are well known intermediates (9) , but the orbital arrangement in such a species precludes the concert ed one-step addition process that has been shown to prevail in successful 1, 3 -dipolar additions {8). Thus in the strictest sense the formation of ....._ xii or ,,_,.,..... xiv might not constitute a 1 , 3 -dipolar addition, although the end result would be equivalent. Alternative products in the formation of xii and xiv are the ,,_,.,..... spiro compounds xv and xvi. """" """"""' """"""' Although they are not expected, the recent report of the addition of tetrac yanoethyl e n e to N-ethylallene 123 0 (BuO)z~ ~ xvi ~ imine i (R = CzH 5 ) to give cyclobutane ~is not encouraging (1 0 ) • .......... TCNE ) xvii .;vw-. It is hoped that this r eac tion is ascribable to the special reactivity of TCNE (11) and not to the allene imine . Should the formation of xii and xiv proceed as intended, sim ilar .-..- ~ reactions of i with dienes such as cyc l opentadiene might provide a """" new route to functionali zed bicyclic derivatives such as xix . --...- > -R xviii xix ~ ~ 124 Ht~ fe rence s 1. a. G. Stork, S. Danishefsky and M. Ohashi, J . Arn. Chem. Soc., 89, 5459 (1967). """""' b. M. Ohashi, H. Kamachi, H. KakisawaandG. Stork, ibid ., 8 9 ' 5 4 6 0 (196 7) • ~ c. G. StorkandJ. E. McMurry, ibid., 89, 5461 {1967). d. G. StorkandJ. E. McMurry, ibid., 89, 5463 {1967). e. G. Stork a nd J.E. McMurry , ibid., 89, 5464 (1967) . a. E. J. Corey, M. F. Se mmelhack and L. S. Hegedus, ibid., 90, 2416 {1968). """"'" """"'" """' 2. ~ h. E. J. Corey, L. s. H egedus and M. F. Semmelhack , ibid., 90, 2417 (1968). ~ -- c. E. J. Corey and M. F. Semmelhack, ibid., 89, 2755 (19 6 7). d. E. J. Corey and E. K. W. Wat, ibid ., 89, 2757 (19 6 7). e. E. J. CoreyandE. Hamanaka, ibid., 89, 2758 (1967). -- -- -- --- -- 3. A. T. Bottini and H. E. Olsen, Org. Syn., 44, 53 (1964). 4. J. Szmuszkovicz, Adv. Org. Chem.,~ (196 3). 5. H. W. Whitlock and G. L. Smith, J. Arn. Chem. Soc. , 89, 3600 (1967). 6. J.E. Dolfini andJ. D. Simpson, ibi d., 87, 4381 {1965). 7. D.S. Mattes on, Organometal. Chem. Rev.,~ 1 {19 66). 8. a. R. Huis ge n, 11 Proceedings of the Rob ert A. Welch Foundation on Chemical Research, 11 Houston, T exas (1961) pp. 61-95. b. R. Huisgen, Angew. Chem. int. ed., 2, 565 {1963). -- -- -- """"" -- -- """' 125 9. W. Kirmse, ''Carbene Che mistry, 11 Academic Press, New York (1964) ' pp. 67-69. 10. R. C. Cookson, B. Halton , J. D.R. Stevens, C. T. Watts, J. Chem. Soc., 928(1967). 11. a. D. N. Dhar, Chem. Rev., ,,_ 67, 611 (1967). b. R. Winkler, Chimia, ,,..,,... 16, 360 (1962). Proposition 4 A study of the species derived from isothiazolium salts i and ........ benzisothiozolium salts ii is proposed. """"""' ~s~+ v-)-R ii ........ A carbene, or divalent uncharged carbon, behaves as an electrophile in most of its reactions. Insertion into single bonds and addition to multiple bonds are typical examples. It would seem possible, however, to construct a carbene which, because of unusual electronic contributions from neighboring substituents, would behave largely as a nucleophile. A likely candidate is a diamino carbene, but tetrakis{dimethylamino)ethylene does not appear to dissociate to give the bis {dimethylamino) carbene (1). More recently the olefin iii has been prepared by the action of ......... heat on compound iv, perhaps through the intermediacy of carbene v (2). ......... ......... The thermal reactions of iii shown below are illustrative of its .......- chemistry. 127 Ph I Ph I C7)-cc1 3~ c:> N 150 I Ph Ph iv ,,..,.. -iii CH3N0 2 > -iii v ,,..,.. c lh N 'cHCH2N02 N/ 72% I Ph Ph c I -iii Ph CHO > -- 90% ~h 0 iii 1'!" ~CHCOPh 6 Ph I ) o c:f--6 71% I Ph Ph I -iii PhCH2C02C2jj5 C>HcHco cH 2 2 5 Ph 89% 128 -- It is interesting that iii is unreactive to ole fins except tetra- chloroethylcne with whi ch the cyclopropyl addition product is formed in 6 7% yield. L emal, however, has demonstrated that free carbene s are not in fact involved in the thermal reactions of the dimer ( 3) and suggests a mechanism for its nucleophilic reactivity. Of more interest are the varied reports concerning the species derived from the benzthiozolium s .a lt vi. _..,.... When vi is reacted at 25 """' with a tertiary amine, the dimer is produced (4 ). °C Rapid quenching ~s ~ V-11>=<~ CH3 CH 3 viii Vl ~ ~ of the reaction mixture with a diazo compound gives the corresponding azine before the dimer is formed. If, however, the tetraazo compound ix is placed in THF at -10°, _..,... two moles of nitrogen are lost, no dim er being formed at all (5). The resulting solution is stable for THF -lg; ix ,.,,.... x ,.,,.... 129 short periods. A tetracya.noethylene adduct is formed readily and the species present will add to electrophilic azo compounds. Kinetic studies confirm the existence of a short-lived intermediate such as x. """"' No other report of the chemical reactivity has been made. A logical extension of the above observations is the investigation of the synthetic utility of similar nucleophilic intermediates. A large carbon fragment is not present in x but might be available through """"' the corresponding isothiozolium and benzisothiozolium compounds, -- xi and xii respectively (7). """"' If generated by the tetraazo derivatives, the neutral solution would be of great value when dealing with sensitive systems. xi """"' -xii 130 -- Whether xi and xii would behave as the carbene, the 1, 3-dipole ,,..,,... (8), or any true resonance hybrid is not readily apparent. The s ame question remains unanswered, however, even in the extensively studied thiamine chemistry (6) . But the fact that the initial step in the thiamine catalyze d dime rization of benzaldehyde is nucleophilic ---- carbonyl attack lends credence to our theory that xi and xii will ......... behave similarly. Speculation as to what reactions xi and xii might ,,..,,... actually undergo is largely unwarranted. Considering only carbonyl systems, however, three such reactions might be as shown below. xii 1. 2. 06 M' 0 CHR xi > --~ R ---··~> > xi """" 131 References I. N. Wiberg, J. W. Buchler, J. Am. Chem. Soc., 85, 243 (1963). 2. a. - -- -- -- """"" H. W. Wanzlick, E. Schikora, Ber., 94, 2389 (1961). -b. H. W. Wanzlick, H. Ahrens, Ber., 97, 2447 (196 4). -- -- 3. D. L emal, R. A. Louald, K. I. Kawano, J. Am. Chem. Soc., 86, 2518 (1964). 4. -- 5. H. Balli, Ang. Chem. intern. ed., 3, 809 (1964). 6. R. Breslow, J. Am. Chem. Soc., 80, 3719 (1958). 7. A good review of the chemistry of the isothiazole system is found in H. J. D en Hertog and H. c. Van der Plas, Advances in Heterocyclic Chemistry, _... 4, 107 (1965). 8. a. R. Hu is gen, "Proceedings of the Robert A. Welch Foundation on Chemical Research," Houston, Texas (1961) pp. 61-95. b. R. Huisgen, Angew. Chem . int. ed. 1 1;,.; 565 {1963). H. W. Wanzlick, H. J. Kleine, Ang. Chem. intern. ed . , 5, 126 ,,..,.. (19 66). - - -- -- ,,..,.. ,,..,.. 132 Proposition 5 The solvolysis of several 6-exo -alkenyl-5- endo-norbornenyl brosylate s is proposed. While considering synthetic routes to th e tetracyclic sesquiterpenes cyclosativene i and longicyclene ii {I), the possibility that such ,,.,,.... """"' ii ,,.,,.... interesting carbocyclic ar rangements might be attainable through solvolysis of a homoallylic norbornyl derivative such as ,,.,,... iii became apparent. The uncertainty surrounding the chemistry of the > iii .,..,,.,.., iv ,,.,,... 133 norbornenyl-nortricyclyl system, coupled with the uncertainty as to geometric requirements of the side chain and its double bond for cyclization, suggested that a broader consideration of similar cycl izations would be in order. The acetolysis of~- or endo-5-norbornenyl E_-bromobenzene - -- sulfonate v produces nortricyclyl acetate vi (R = Ac) """ and exo - Ac + -- -vi vii -- norbornenyl acetate vii in the approximate ratio of 9:1 (2). -- The olefinic product vii experiences partial carbon scrambling at positions 4 and 5 as determined by 14 c labeling experiments (2c ) • The solvolysis of nortricyclyl E_-bromobenzenesulfonate vi (R= Bs) .,.,,.... affords a smaller proportion of olefinic material than does the solvolysis of the homoallylically isomeric norbornenyl E_-bromobenzene sulfonate .,.,,.... v (3). This indicates that the solvolysis products from the two starting materials do not arise from a common intermediate. However, the product composition in both instances suggests that a major portion of the charge in each intermediate r e sides on the carbon corresponding to C-2 in structure v (4) . _... 134 It would thus appear possible for an intermediate homoallylic carbonium ion to be captured at C-2 by the double bond in a C-6 ~ side chain, provided that sufficient overlap between the double bond and C-2 vorbital can be attained. The possibility also exists for solvolytic participation of the side chain double bond at either C-2 or C-5, although it is not expected that C -5 participation will be significant. For the problem at hand, an examination of molecular models indicates that substantial overlap between the side chain double bond 0 and the C -2 1T orbital can exit in ,,,.,,... iii when n = 1 at a distance of 2. I A 0 and when n =2 at a distance of 2. 2 A. When n= 0 the side chain olefinic 1T orbitals are perpendicular to the C-2 orbital axis at a closest 0 approach distance of 2.4 A, a fact which may preclude cyclization m that compound. It is suggested that the solvolyses of compounds viii-xi ...,.._,,,.,,... be studied • Of primary interest will be the nature of the products formed, that is, the presence or absence of tetracyclic material. Because such solvolyses are best conducted in a medium such as formic acid which favors anchiome ric assistance (5) and is a lso relatively non-nucleophilic, the tetracyclic material is expected to be olefinic since elimination in the tertiary carbonium ions should predominate over solvent capture. Thus the tetracyclic material will be easily distinguishable from any 135 OBs ---~ -- viii xii ~ OBs ---~ Bs ·+ -- -- x xiv Bs ---~ COOH -xi xv ~ 136 uncyclizccl products. The attainment of a tertiary carbonium ion in .xii-xiv - . - should provide added driving force for th e cyclization step • The solvolysis of exo-acid xv will provide an interesting comparison -- -- """"'" with acid xvi which undergoes ready lactonization in sulfuric acid (6). In addition, the presence or absence of participation of the side chain COOH -xvi O---~o xvii _..,.,,..., olefinic bond can be determined by a measure of the solvolytic rate enhancement of viii-xi relative to a similar compound with a totally --..- saturated side chain. Compounds viii-xi can be prepared from aldehyde xviii, the --""""'" _..,.,,..., expected product (6) of the Diels-Alder reaction between cyclopentadiene and trans -acetoxyacolein (7) in straightforward manners as outlined in Chart A. 137 Chart A Ac CrO 3 > xviii _..,...,,.. OAc COOH 1) NaBH 2) (C H ) P•Br 6 5 3 2 Ac I) SOCl 2) (CH Cd 3) (C ~ 1 P CH 2 r 6 53 Bs Ac COOH Mg Br xi ............. =< 1) LAH 2) BsCl OBs viii ,,..,..,,..., OBs ix ,,...,.., Bs x rVV 138 References 1. See Proposition 1 this thesis. 2. a. S. Winstein, H. M. Walborsky and K. C. Schreiber, J. Arn. Chem. Soc., 72, 5795 (1950). -- b. ~ J. D. Roberts, W. Bennettand R. Armstrong, Jr., ibid., 72, 3329 (1950) • .,....,.... c. J. D. Roberts , C. C. Lee and W. H. Saunders, Jr., ibid., 7 7 ' 3 0 3 4 (195 5) • """" 3. :a. Schmid, unpublished results, quoted by S. Winstein and E. Kosower, ibid., 81, 4399 (1959). -- ~ 4. P. D. Bartlett and G.D. Sargent, ibid., 87, .,....,.... 1297 (1965) • 5. S. Winstein, C. R. Lindegren, H. Marshall and L. L. Ingraham, ibid.' 75' 147 (195 3). -- """" 6. K. Alder and G. Stein, Ann., 514, 197 (1934). 7. a. T. V. Protopopoua and A. P. Skoldinov, Zhur. Obshchei Khim., 28' 240 (195 8) • ....,_ b. The trans configuration of related compounds is discussed by A. A. Bothne r-By and R. K. Harris, Org. Chem., 30, 254 (1965). """" -- .,.,,.,,.. r.